)]}' { "log": [ { "commit": "8619f9b5bbcc5d1b1f9ce44d1beeb6c991782ea6", "tree": "e41ba74faf1740a4a1dfcf01d1f23b5479172845", "parents": [ "28be787b59403ca877de2d4b55cc8487d67d4b0e" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Sep 04 05:02:55 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:55 2019 +0000" }, "message": "pci: test for unexpectedly disabled bridges\n\nThe all-ones value is not just a \"device didn\u0027t exist\" case, it\u0027s also\npotentially a quite valid value, so not restoring it would be wrong.\n\nWhat *would* be interesting is to hear where the bad values came from in\nthe first place. It sounds like the device state is saved after the PCI\nbus controller in front of the device has been crapped on, resulting in the\nPCI config cycles never reaching the device at all.\n\nSomething along this patch (together with suspend/resume debugging output)\nmigth help pinpoint it. But it really sounds like something totally\nbrokenly turned off the PCI bridge (some ACPI shutdown crud? I wouldn\u0027t be\nentirely surprised)\n\nCc: Greg KH \u003cgreg@kroah.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "28be787b59403ca877de2d4b55cc8487d67d4b0e", "tree": "d081c7f24b8faa12083dca654ac78c5d4381f754", "parents": [ "e8ebf0f6aeab8426e00d0c9665530057d9ee8379" ], "author": { "name": "Andrew Morton", "email": "akpm@linux-foundation.org", "time": "Wed Sep 04 05:02:55 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:55 2019 +0000" }, "message": "kernel/fork.c: export kernel_thread() to modules\n\nmutex-subsystem-synchro-test-module.patch needs this\n\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "e8ebf0f6aeab8426e00d0c9665530057d9ee8379", "tree": "4e53014771c4c47f0c1ae6959baed5517d27da04", "parents": [ "33cca7cda77eb64aa5334eaebcab0d47370671f0" ], "author": { "name": "David Howells", "email": "dhowells@redhat.com", "time": "Wed Sep 04 05:02:55 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:55 2019 +0000" }, "message": "mutex subsystem, synchro-test module\n\nThe attached patch adds a module for testing and benchmarking mutexes,\nsemaphores and R/W semaphores.\n\nUsing it is simple:\n\n\tinsmod synchro-test.ko \u003cargs\u003e\n\nIt will exit with error ENOANO after running the tests and printing the\nresults to the kernel console log.\n\nThe available arguments are:\n\n (*) mx\u003dN\n\n\tStart up to N mutex thrashing threads, where N is at most 20. All will\n\ttry and thrash the same mutex.\n\n (*) sm\u003dN\n\n\tStart up to N counting semaphore thrashing threads, where N is at most\n\t20. All will try and thrash the same semaphore.\n\n (*) ism\u003dM\n\n\tInitialise the counting semaphore with M, where M is any positive\n\tinteger greater than zero. The default is 4.\n\n (*) rd\u003dN\n (*) wr\u003dO\n (*) dg\u003dP\n\n\tStart up to N reader thrashing threads, O writer thrashing threads and\n\tP downgrader thrashing threads, where N, O and P are at most 20\n\tapiece. All will try and thrash the same read/write semaphore.\n\n (*) elapse\u003dN\n\n\tRun the tests for N seconds. The default is 5.\n\n (*) load\u003dN\n\n\tEach thread delays for N uS whilst holding the lock. The dfault is 0.\n\n (*) interval\u003dN\n\n\tEach thread delays for N uS whilst not holding the lock. The default\n\tis 0.\n\n (*) do_sched\u003d1\n\n\tEach thread will call schedule if required after each iteration.\n\n (*) v\u003d1\n\n\tPrint more verbose information, including a thread iteration\n\tdistribution list.\n\nThe module should be enabled by turning on CONFIG_DEBUG_SYNCHRO_TEST to \"m\".\n\n[randy.dunlap@oracle.com: fix build errors, add \u003csched.h\u003e header file]\n[akpm@linux-foundation.org: remove smp_lock.h inclusion]\n[viro@ZenIV.linux.org.uk: kill daemonize() calls]\n[rdunlap@xenotime.net: fix printk format warrnings]\n[walken@google.com: add spinlock test]\n[walken@google.com: document default load and interval values]\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Adrian Bunk \u003cbunk@stusta.de\u003e\nSigned-off-by: Randy Dunlap \u003crdunlap@xenotime.net\u003e\nSigned-off-by: Michel Lespinasse \u003cwalken@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "33cca7cda77eb64aa5334eaebcab0d47370671f0", "tree": "a1d74faf424928db57cd61117d6f03fd79ffa58e", "parents": [ "5ad44286142c893253a666e1100947437c28aaf7" ], "author": { "name": "Matthew Wilcox", "email": "willy@parisc-linux.org", "time": "Wed Sep 04 05:02:55 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:55 2019 +0000" }, "message": "Releasing resources with children\n\nWhat does it mean to release a resource with children? Should the children\nbecome children of the released resource\u0027s parent? Should they be released\ntoo? Should we fail the release?\n\nI bet we have no callers who expect this right now, but with\ninsert_resource() we may get some. At the point where someone hits this\nBUG we can figure out what semantics we want.\n\nSigned-off-by: Matthew Wilcox \u003cwilly@parisc-linux.org\u003e\nCc: Matthew Wilcox \u003cwilly@infradead.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "5ad44286142c893253a666e1100947437c28aaf7", "tree": "0e5d92a7fc005a17174bb02c768512164efb01dd", "parents": [ "c103ed09b4a174bdeba2de646bdd1b0545b1bd89" ], "author": { "name": "Matthew Wilcox", "email": "matthew@wil.cx", "time": "Wed Sep 04 05:02:54 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:54 2019 +0000" }, "message": "Make sure nobody\u0027s leaking resources\n\nCurrently, releasing a resource also releases all of its children. That\nmade sense when request_resource was the main method of dividing up the\nmemory map. With the increased use of insert_resource, it seems to me that\nwe should instead reparent the newly orphaned resources. Before we do\nthat, let\u0027s make sure that nobody\u0027s actually relying on the current\nsemantics.\n\nSigned-off-by: Matthew Wilcox \u003cmatthew@wil.cx\u003e\nCc: Greg KH \u003cgreg@kroah.com\u003e\nCc: Ivan Kokshaysky \u003cink@jurassic.park.msu.ru\u003e\nCc: Dominik Brodowski \u003clinux@dominikbrodowski.net\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "c103ed09b4a174bdeba2de646bdd1b0545b1bd89", "tree": "2e51ea385b9a383cb48b55337654004508c0fcd8", "parents": [ "1da66c96c94d71ede2f477992b5ec688e8880433" ], "author": { "name": "Andi Kleen", "email": "ak@linux.intel.com", "time": "Wed Sep 04 05:02:54 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:54 2019 +0000" }, "message": "drivers/media/platform/sti/delta/delta-ipc.c: fix read buffer overflow\n\nThe single caller passes a string to delta_ipc_open, which copies with a\nfixed size larger than the string. So it copies some random data after\nthe original string the ro segment.\n\nIf the string was at the end of a page it may fault.\n\nJust copy the string with a normal strcpy after clearing the field.\n\nFound by a LTO build (which errors out)\nbecause the compiler inlines the functions and can resolve\nthe string sizes and triggers the compile time checks in memcpy.\n\nIn function `memcpy\u0027,\n inlined from `delta_ipc_open.constprop\u0027 at linux/drivers/media/platform/sti/delta/delta-ipc.c:178:0,\n inlined from `delta_mjpeg_ipc_open\u0027 at linux/drivers/media/platform/sti/delta/delta-mjpeg-dec.c:227:0,\n inlined from `delta_mjpeg_decode\u0027 at linux/drivers/media/platform/sti/delta/delta-mjpeg-dec.c:403:0:\n/home/andi/lsrc/linux/include/linux/string.h:337:0: error: call to `__read_overflow2\u0027 declared with attribute error: detected read beyond size of object passed as 2nd parameter\n __read_overflow2();\n\nLink: http://lkml.kernel.org/r/20171222001212.1850-1-andi@firstfloor.org\nSigned-off-by: Andi Kleen \u003cak@linux.intel.com\u003e\nCc: Hugues FRUCHET \u003chugues.fruchet@st.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "1da66c96c94d71ede2f477992b5ec688e8880433", "tree": "4861b0c68744fd61a7e60ea023df66ada57f04cc", "parents": [ "ece4ca7759771c232cc58d0f82bb8fdfc4ae9654" ], "author": { "name": "Andrew Morton", "email": "akpm@linux-foundation.org", "time": "Wed Sep 04 05:02:54 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:54 2019 +0000" }, "message": "drivers/tty/serial/sh-sci.c: suppress warning\n\ndrivers/tty/serial/sh-sci.c: In function sci_dma_rx_submit:\n./include/linux/spinlock.h:288:3: warning: flags may be used uninitialized in this function [-Wmaybe-uninitialized]\n _raw_spin_unlock_irqrestore(lock, flags); \\\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~\ndrivers/tty/serial/sh-sci.c:1353:16: note: flags was declared here\n unsigned long flags;\n\nCc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nCc: Jiri Slaby \u003cjslaby@suse.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "ece4ca7759771c232cc58d0f82bb8fdfc4ae9654", "tree": "3b5dea65d5946f3b35ae5b61c0240fe421c2dc61", "parents": [ "4aff5391e7f12c03ee1ae23b65499fc5759708d5" ], "author": { "name": "Joel Fernandes (Google)", "email": "joel@joelfernandes.org", "time": "Wed Sep 04 05:02:53 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:53 2019 +0000" }, "message": "ipc/sem.c: convert to use built-in RCU list checking\n\nCONFIG_PROVE_RCU_LIST requires list_for_each_entry_rcu() to pass a lockdep\nexpression if using srcu or locking for protection. It can only check\nregular RCU protection, all other protection needs to be passed as lockdep\nexpression.\n\nLink: http://lkml.kernel.org/r/20190830231817.76862-2-joel@joelfernandes.org\nSigned-off-by: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\nCc: Arnd Bergmann \u003carnd@arndb.de\u003e\nCc: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\nCc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nCc: \"Gustavo A. R. Silva\" \u003cgustavo@embeddedor.com\u003e\nCc: Jonathan Derrick \u003cjonathan.derrick@intel.com\u003e\nCc: Keith Busch \u003ckeith.busch@intel.com\u003e\nCc: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\nCc: \"Paul E. McKenney\" \u003cpaulmck@linux.vnet.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "4aff5391e7f12c03ee1ae23b65499fc5759708d5", "tree": "9eb9e2d55fe0da022bbccd54942353aa95f90093", "parents": [ "dd75303183483c711a534ac4aad756393161262b" ], "author": { "name": "Mark Rutland", "email": "mark.rutland@arm.com", "time": "Wed Sep 04 05:02:53 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:53 2019 +0000" }, "message": "mm: treewide: clarify pgtable_page_{ctor,dtor}() naming\n\nThe naming of pgtable_page_{ctor,dtor}() seems to have confused a few\npeople, and until recently arm64 used these erroneously/pointlessly for\nother levels of page table.\n\nTo make it incredibly clear that these only apply to the PTE level, and to\nalign with the naming of pgtable_pmd_page_{ctor,dtor}(), let\u0027s rename them\nto pgtable_pte_page_{ctor,dtor}().\n\nThese changes were generated with the following shell script:\n\n----\ngit grep -lw \u0027pgtable_page_.tor\u0027 | while read FILE; do\n sed -i \u0027{s/pgtable_page_ctor/pgtable_pte_page_ctor/}\u0027 $FILE;\n sed -i \u0027{s/pgtable_page_dtor/pgtable_pte_page_dtor/}\u0027 $FILE;\ndone\n----\n\n... with the documentation re-flowed to remain under 80 columns, and\nwhitespace fixed up in macros to keep backslashes aligned.\n\nThere should be no functional change as a result of this patch.\n\nLink: http://lkml.kernel.org/r/20190722141133.3116-1-mark.rutland@arm.com\nSigned-off-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\nReviewed-by: Mike Rapoport \u003crppt@linux.ibm.com\u003e\nAcked-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\t[m68k]\nCc: Anshuman Khandual \u003canshuman.khandual@arm.com\u003e\nCc: Matthew Wilcox \u003cwilly@infradead.org\u003e\nCc: Michal Hocko \u003cmhocko@suse.com\u003e\nCc: Yu Zhao \u003cyuzhao@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "dd75303183483c711a534ac4aad756393161262b", "tree": "f1b2d90e38507dac7306442aedc392512271887a", "parents": [ "2343c06764fea5085f5e4925b9fafe5764a461ab" ], "author": { "name": "Denis Efremov", "email": "efremov@linux.com", "time": "Wed Sep 04 05:02:53 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:53 2019 +0000" }, "message": "ntfs: remove (un)?likely() from IS_ERR() conditions\n\n\"likely(!IS_ERR(x))\" is excessive. IS_ERR() already uses\nunlikely() internally.\n\nLink: http://lkml.kernel.org/r/20190829165025.15750-11-efremov@linux.com\nSigned-off-by: Denis Efremov \u003cefremov@linux.com\u003e\nCc: Anton Altaparmakov \u003canton@tuxera.com\u003e\nCc: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "2343c06764fea5085f5e4925b9fafe5764a461ab", "tree": "003bbea1c5692d6d8a87701a054592c9458f168a", "parents": [ "c7ae7da308d48edf2272fa16db3f33f32715d91d" ], "author": { "name": "Denis Efremov", "email": "efremov@linux.com", "time": "Wed Sep 04 05:02:52 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:52 2019 +0000" }, "message": "IB/hfi1: remove unlikely() from IS_ERR*() condition\n\n\"unlikely(IS_ERR_OR_NULL(x))\" is excessive. IS_ERR_OR_NULL() already uses\nunlikely() internally.\n\nLink: http://lkml.kernel.org/r/20190829165025.15750-8-efremov@linux.com\nSigned-off-by: Denis Efremov \u003cefremov@linux.com\u003e\nCc: Mike Marciniszyn \u003cmike.marciniszyn@intel.com\u003e\nCc: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nCc: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "c7ae7da308d48edf2272fa16db3f33f32715d91d", "tree": "8c3a45ed93b96b62fefaeb2fabc446bafde719b2", "parents": [ "4c7fbf972fd620a7e8beda5d9fd280f2f6ac72fc" ], "author": { "name": "Denis Efremov", "email": "efremov@linux.com", "time": "Wed Sep 04 05:02:52 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:52 2019 +0000" }, "message": "xfs: remove unlikely() from WARN_ON() condition\n\n\"unlikely(WARN_ON(x))\" is excessive. WARN_ON() already uses unlikely()\ninternally.\n\nLink: http://lkml.kernel.org/r/20190829165025.15750-7-efremov@linux.com\nSigned-off-by: Denis Efremov \u003cefremov@linux.com\u003e\nReviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nCc: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "4c7fbf972fd620a7e8beda5d9fd280f2f6ac72fc", "tree": "97672c235262b55584bcbb43803e18a901b6bffc", "parents": [ "ee5ea1c51d72ece3c52946ae9a03f4e37f16cea6" ], "author": { "name": "Denis Efremov", "email": "efremov@linux.com", "time": "Wed Sep 04 05:02:52 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:52 2019 +0000" }, "message": "wimax/i2400m: remove unlikely() from WARN*() condition\n\n\"unlikely(WARN_ON(x))\" is excessive. WARN_ON() already uses unlikely()\ninternally.\n\nLink: http://lkml.kernel.org/r/20190829165025.15750-6-efremov@linux.com\nSigned-off-by: Denis Efremov \u003cefremov@linux.com\u003e\nCc: Inaky Perez-Gonzalez \u003cinaky.perez-gonzalez@intel.com\u003e\nCc: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "ee5ea1c51d72ece3c52946ae9a03f4e37f16cea6", "tree": "40a6b8bd2602f010153bb92d74f2294926aa0e51", "parents": [ "873a690015a30bb0f217b792b35aa9f3c741b352" ], "author": { "name": "Denis Efremov", "email": "efremov@linux.com", "time": "Wed Sep 04 05:02:51 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:51 2019 +0000" }, "message": "fs: remove unlikely() from WARN_ON() condition\n\n\"unlikely(WARN_ON(x))\" is excessive. WARN_ON() already uses unlikely()\ninternally.\n\nLink: http://lkml.kernel.org/r/20190829165025.15750-5-efremov@linux.com\nSigned-off-by: Denis Efremov \u003cefremov@linux.com\u003e\nCc: Alexander Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "873a690015a30bb0f217b792b35aa9f3c741b352", "tree": "3c9ae250d8d01a32e1bd6412d6b96a644081d426", "parents": [ "7e72857d5881b21eee5a002023ad4a8e5fa07d2f" ], "author": { "name": "Denis Efremov", "email": "efremov@linux.com", "time": "Wed Sep 04 05:02:51 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:51 2019 +0000" }, "message": "xen/events: remove unlikely() from WARN() condition\n\n\"unlikely(WARN(x))\" is excessive. WARN() already uses unlikely()\ninternally.\n\nLink: http://lkml.kernel.org/r/20190829165025.15750-4-efremov@linux.com\nSigned-off-by: Denis Efremov \u003cefremov@linux.com\u003e\nCc: Boris Ostrovsky \u003cboris.ostrovsky@oracle.com\u003e\nCc: Juergen Gross \u003cjgross@suse.com\u003e\nCc: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "7e72857d5881b21eee5a002023ad4a8e5fa07d2f", "tree": "439c3a872c40a24afc8e4400edf6ef67bcdb0535", "parents": [ "f2fb0129eb88d4ea3025673f3823b4ee6ff859ba" ], "author": { "name": "Denis Efremov", "email": "efremov@linux.com", "time": "Wed Sep 04 05:02:51 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:51 2019 +0000" }, "message": "drm/msm: remove unlikely() from WARN_ON() conditions\n\n\"unlikely(WARN_ON(x))\" is excessive. WARN_ON() already uses unlikely()\ninternally.\n\nLink: http://lkml.kernel.org/r/20190829165025.15750-2-efremov@linux.com\nSigned-off-by: Denis Efremov \u003cefremov@linux.com\u003e\nCc: Rob Clark \u003crobdclark@gmail.com\u003e\nCc: Sean Paul \u003csean@poorly.run\u003e\nCc: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "f2fb0129eb88d4ea3025673f3823b4ee6ff859ba", "tree": "6597151e5cf976c95b847ec64bf386f54ea97ba7", "parents": [ "baccab9b89f25cd6405d72b79a297981b2241ef2" ], "author": { "name": "Denis Efremov", "email": "efremov@linux.com", "time": "Wed Sep 04 05:02:51 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:51 2019 +0000" }, "message": "checkpatch: check for nested (un)?likely() calls\n\nIS_ERR(), IS_ERR_OR_NULL(), IS_ERR_VALUE() and WARN*() already contain\nunlikely() optimization internally. Thus, there is no point in calling\nthese functions and defines under likely()/unlikely().\n\nThis check is based on the coccinelle rule developed by Enrico Weigelt\nhttps://lore.kernel.org/lkml/1559767582-11081-1-git-send-email-info@metux.net/\n\nLink: http://lkml.kernel.org/r/20190829165025.15750-1-efremov@linux.com\nSigned-off-by: Denis Efremov \u003cefremov@linux.com\u003e\nCc: Joe Perches \u003cjoe@perches.com\u003e\nCc: Alexander Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Anton Altaparmakov \u003canton@tuxera.com\u003e\nCc: Boris Ostrovsky \u003cboris.ostrovsky@oracle.com\u003e\nCc: Boris Pismenny \u003cborisp@mellanox.com\u003e\nCc: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nCc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\nCc: Denis Efremov \u003cefremov@linux.com\u003e\nCc: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\nCc: Inaky Perez-Gonzalez \u003cinaky.perez-gonzalez@intel.com\u003e\nCc: Juergen Gross \u003cjgross@suse.com\u003e\nCc: Leon Romanovsky \u003cleon@kernel.org\u003e\nCc: Mike Marciniszyn \u003cmike.marciniszyn@intel.com\u003e\nCc: Rob Clark \u003crobdclark@gmail.com\u003e\nCc: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\nCc: Sean Paul \u003csean@poorly.run\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "baccab9b89f25cd6405d72b79a297981b2241ef2", "tree": "14d3be443f47041a9830d791e1eae4b62a0e384b", "parents": [ "9ae8652c0151f5c7e933564292be991a410e06c4" ], "author": { "name": "Mike Rapoport", "email": "rppt@linux.ibm.com", "time": "Wed Sep 04 05:02:50 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:50 2019 +0000" }, "message": "hexagon: drop empty and unused free_initrd_mem\n\nhexagon never reserves or initializes initrd and the only mention of it is\nthe empty free_initrd_mem() function.\n\nAs we have a generic implementation of free_initrd_mem(), there is no need\nto define an empty stub for the hexagon implementation and it can be\ndropped.\n\nLink: http://lkml.kernel.org/r/1565858133-25852-1-git-send-email-rppt@linux.ibm.com\nSigned-off-by: Mike Rapoport \u003crppt@linux.ibm.com\u003e\nReviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\nCc: Richard Kuo \u003crkuo@codeaurora.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "9ae8652c0151f5c7e933564292be991a410e06c4", "tree": "85b5c6d8524a21ff03553cf2b16c636bf7fe65ed", "parents": [ "88361c8bd4bc015996cca4ce3f90be773d71f354" ], "author": { "name": "Minchan Kim", "email": "minchan@kernel.org", "time": "Wed Sep 04 05:02:50 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:50 2019 +0000" }, "message": "mm: factor out common parts between MADV_COLD and MADV_PAGEOUT\n\nThere are many common parts between MADV_COLD and MADV_PAGEOUT.\nThis patch factor them out to save code duplication.\n\nLink: http://lkml.kernel.org/r/20190726023435.214162-6-minchan@kernel.org\nSigned-off-by: Minchan Kim \u003cminchan@kernel.org\u003e\nSuggested-by: Johannes Weiner \u003channes@cmpxchg.org\u003e\nAcked-by: Michal Hocko \u003cmhocko@suse.com\u003e\nCc: Chris Zankel \u003cchris@zankel.net\u003e\nCc: Daniel Colascione \u003cdancol@google.com\u003e\nCc: Dave Hansen \u003cdave.hansen@intel.com\u003e\nCc: Hillf Danton \u003chdanton@sina.com\u003e\nCc: James E.J. Bottomley \u003cJames.Bottomley@HansenPartnership.com\u003e\nCc: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\nCc: kbuild test robot \u003clkp@intel.com\u003e\nCc: Kirill A. Shutemov \u003ckirill.shutemov@linux.intel.com\u003e\nCc: Oleksandr Natalenko \u003coleksandr@redhat.com\u003e\nCc: Ralf Baechle \u003cralf@linux-mips.org\u003e\nCc: Richard Henderson \u003crth@twiddle.net\u003e\nCc: Shakeel Butt \u003cshakeelb@google.com\u003e\nCc: Sonny Rao \u003csonnyrao@google.com\u003e\nCc: Suren Baghdasaryan \u003csurenb@google.com\u003e\nCc: Tim Murray \u003ctimmurray@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "88361c8bd4bc015996cca4ce3f90be773d71f354", "tree": "d69bea5a97c194dcbd8b26110ab3ea1532618d0d", "parents": [ "264fefb7e312aa9536d45f249d86bb536755ee21" ], "author": { "name": "Andrew Morton", "email": "akpm@linux-foundation.org", "time": "Wed Sep 04 05:02:50 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:50 2019 +0000" }, "message": "mm-introduce-madv_pageout-fix-fix\n\nresolve conflicts with hmm.git\n\nCc: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nCc: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nCc: Christoph Hellwig \u003chch@lst.de\u003e\nCc: Minchan Kim \u003cminchan@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "264fefb7e312aa9536d45f249d86bb536755ee21", "tree": "cbff438e90466fa69b7c6f4208af64c3f566e3fe", "parents": [ "e15a4aac7db00045e9f394cfb7cd85ae6c368f40" ], "author": { "name": "Minchan Kim", "email": "minchan@kernel.org", "time": "Wed Sep 04 05:02:50 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:50 2019 +0000" }, "message": "mm: clear PG_active on MADV_PAGEOUT\n\nshrink_page_list expects every pages as argument should be no active LRU\npages so we need to clear PG_active.\n\nLink: http://lkml.kernel.org/r/20190802200643.GA181880@google.com\nFixes: 06a833a1167e (\"mm: introduce MADV_PAGEOUT\")\nSigned-off-by: Minchan Kim \u003cminchan@kernel.org\u003e\nReported-by: syzbot+8e6326965378936537c3@syzkaller.appspotmail.com\nAcked-by: Michal Hocko \u003cmhocko@suse.com\u003e\nCc: James E.J. Bottomley \u003cJames.Bottomley@HansenPartnership.com\u003e\nCc: Richard Henderson \u003crth@twiddle.net\u003e\nCc: Ralf Baechle \u003cralf@linux-mips.org\u003e\nCc: Chris Zankel \u003cchris@zankel.net\u003e\nCc: Daniel Colascione \u003cdancol@google.com\u003e\nCc: Dave Hansen \u003cdave.hansen@intel.com\u003e\nCc: Hillf Danton \u003chdanton@sina.com\u003e\nCc: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\nCc: Johannes Weiner \u003channes@cmpxchg.org\u003e\nCc: Kirill A. Shutemov \u003ckirill.shutemov@linux.intel.com\u003e\nCc: Oleksandr Natalenko \u003coleksandr@redhat.com\u003e\nCc: Shakeel Butt \u003cshakeelb@google.com\u003e\nCc: Sonny Rao \u003csonnyrao@google.com\u003e\nCc: Suren Baghdasaryan \u003csurenb@google.com\u003e\nCc: Tim Murray \u003ctimmurray@google.com\u003e\nCc: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "e15a4aac7db00045e9f394cfb7cd85ae6c368f40", "tree": "77c33cd87ee93cf28a6cb2946065095ecfebdd40", "parents": [ "cf9dfe6f5fca61324088c86042586da8e09fbd1c" ], "author": { "name": "Minchan Kim", "email": "minchan@kernel.org", "time": "Wed Sep 04 05:02:49 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:49 2019 +0000" }, "message": "mm: introduce MADV_PAGEOUT\n\nWhen a process expects no accesses to a certain memory range for a long\ntime, it could hint kernel that the pages can be reclaimed instantly but\ndata should be preserved for future use. This could reduce workingset\neviction so it ends up increasing performance.\n\nThis patch introduces the new MADV_PAGEOUT hint to madvise(2) syscall. \nMADV_PAGEOUT can be used by a process to mark a memory range as not\nexpected to be used for a long time so that kernel reclaims *any LRU*\npages instantly. The hint can help kernel in deciding which pages to\nevict proactively.\n\nA note: It doesn\u0027t apply SWAP_CLUSTER_MAX LRU page isolation limit\nintentionally because it\u0027s automatically bounded by PMD size. If PMD\nsize(e.g., 256) makes some trouble, we could fix it later by limit it to\nSWAP_CLUSTER_MAX[1].\n\n- man-page material\n\nMADV_PAGEOUT (since Linux x.x)\n\nDo not expect access in the near future so pages in the specified\nregions could be reclaimed instantly regardless of memory pressure.\nThus, access in the range after successful operation could cause\nmajor page fault but never lose the up-to-date contents unlike\nMADV_DONTNEED. Pages belonging to a shared mapping are only processed\nif a write access is allowed for the calling process.\n\nMADV_PAGEOUT cannot be applied to locked pages, Huge TLB pages, or\nVM_PFNMAP pages.\n\n[1] https://lore.kernel.org/lkml/20190710194719.GS29695@dhcp22.suse.cz/\n\nLink: http://lkml.kernel.org/r/20190726023435.214162-5-minchan@kernel.org\nSigned-off-by: Minchan Kim \u003cminchan@kernel.org\u003e\nReported-by: kbuild test robot \u003clkp@intel.com\u003e\nAcked-by: Michal Hocko \u003cmhocko@suse.com\u003e\nCc: James E.J. Bottomley \u003cJames.Bottomley@HansenPartnership.com\u003e\nCc: Richard Henderson \u003crth@twiddle.net\u003e\nCc: Ralf Baechle \u003cralf@linux-mips.org\u003e\nCc: Chris Zankel \u003cchris@zankel.net\u003e\nCc: Daniel Colascione \u003cdancol@google.com\u003e\nCc: Dave Hansen \u003cdave.hansen@intel.com\u003e\nCc: Hillf Danton \u003chdanton@sina.com\u003e\nCc: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\nCc: Johannes Weiner \u003channes@cmpxchg.org\u003e\nCc: Kirill A. Shutemov \u003ckirill.shutemov@linux.intel.com\u003e\nCc: Oleksandr Natalenko \u003coleksandr@redhat.com\u003e\nCc: Shakeel Butt \u003cshakeelb@google.com\u003e\nCc: Sonny Rao \u003csonnyrao@google.com\u003e\nCc: Suren Baghdasaryan \u003csurenb@google.com\u003e\nCc: Tim Murray \u003ctimmurray@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "cf9dfe6f5fca61324088c86042586da8e09fbd1c", "tree": "fbb72365b791272260c6cf4c418b3e5854f1f91d", "parents": [ "b92062956fdfcf73470febaa20ca0f8b80b61dc4" ], "author": { "name": "Minchan Kim", "email": "minchan@kernel.org", "time": "Wed Sep 04 05:02:49 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:49 2019 +0000" }, "message": "mm: change PAGEREF_RECLAIM_CLEAN with PAGE_REFRECLAIM\n\nThe local variable references in shrink_page_list is PAGEREF_RECLAIM_CLEAN\nas default. It is for preventing to reclaim dirty pages when CMA try to\nmigrate pages. Strictly speaking, we don\u0027t need it because CMA didn\u0027t\nallow to write out by .may_writepage \u003d 0 in reclaim_clean_pages_from_list.\n\nMoreover, it has a problem to prevent anonymous pages\u0027s swap out even\nthough force_reclaim \u003d true in shrink_page_list on upcoming patch. So\nthis patch makes references\u0027s default value to PAGEREF_RECLAIM and rename\nforce_reclaim with ignore_references to make it more clear.\n\nThis is a preparatory work for next patch.\n\nLink: http://lkml.kernel.org/r/20190726023435.214162-3-minchan@kernel.org\nSigned-off-by: Minchan Kim \u003cminchan@kernel.org\u003e\nAcked-by: Michal Hocko \u003cmhocko@suse.com\u003e\nAcked-by: Johannes Weiner \u003channes@cmpxchg.org\u003e\nCc: Chris Zankel \u003cchris@zankel.net\u003e\nCc: Daniel Colascione \u003cdancol@google.com\u003e\nCc: Dave Hansen \u003cdave.hansen@intel.com\u003e\nCc: Hillf Danton \u003chdanton@sina.com\u003e\nCc: James E.J. Bottomley \u003cJames.Bottomley@HansenPartnership.com\u003e\nCc: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\nCc: kbuild test robot \u003clkp@intel.com\u003e\nCc: Kirill A. Shutemov \u003ckirill.shutemov@linux.intel.com\u003e\nCc: Oleksandr Natalenko \u003coleksandr@redhat.com\u003e\nCc: Ralf Baechle \u003cralf@linux-mips.org\u003e\nCc: Richard Henderson \u003crth@twiddle.net\u003e\nCc: Shakeel Butt \u003cshakeelb@google.com\u003e\nCc: Sonny Rao \u003csonnyrao@google.com\u003e\nCc: Suren Baghdasaryan \u003csurenb@google.com\u003e\nCc: Tim Murray \u003ctimmurray@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "b92062956fdfcf73470febaa20ca0f8b80b61dc4", "tree": "1b27ed9a01bcf6e5ce490ac5c22940f50bac87fb", "parents": [ "7d041a311f9e734266373e7666723f8c24ea1011" ], "author": { "name": "Andrew Morton", "email": "akpm@linux-foundation.org", "time": "Wed Sep 04 05:02:49 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:49 2019 +0000" }, "message": "mm-introduce-madv_cold-fix\n\nresolve conflicts with hmm.git\n\nCc: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nCc: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nCc: Christoph Hellwig \u003chch@lst.de\u003e\nCc: Minchan Kim \u003cminchan@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "7d041a311f9e734266373e7666723f8c24ea1011", "tree": "33e9bcc966d07db9ce20db8ad0dfbab6bebb22bb", "parents": [ "289a83e83358048935fc416a3d19f8f4fc3be1b8" ], "author": { "name": "Minchan Kim", "email": "minchan@kernel.org", "time": "Wed Sep 04 05:02:48 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:48 2019 +0000" }, "message": "mm: introduce MADV_COLD\n\nPatch series \"Introduce MADV_COLD and MADV_PAGEOUT\", v7.\n\n- Background\n\nThe Android terminology used for forking a new process and starting an app\nfrom scratch is a cold start, while resuming an existing app is a hot\nstart. While we continually try to improve the performance of cold\nstarts, hot starts will always be significantly less power hungry as well\nas faster so we are trying to make hot start more likely than cold start.\n\nTo increase hot start, Android userspace manages the order that apps\nshould be killed in a process called ActivityManagerService. \nActivityManagerService tracks every Android app or service that the user\ncould be interacting with at any time and translates that into a ranked\nlist for lmkd(low memory killer daemon). They are likely to be killed by\nlmkd if the system has to reclaim memory. In that sense they are similar\nto entries in any other cache. Those apps are kept alive for\nopportunistic performance improvements but those performance improvements\nwill vary based on the memory requirements of individual workloads.\n\n- Problem\n\nNaturally, cached apps were dominant consumers of memory on the system. \nHowever, they were not significant consumers of swap even though they are\ngood candidate for swap. Under investigation, swapping out only begins\nonce the low zone watermark is hit and kswapd wakes up, but the overall\nallocation rate in the system might trip lmkd thresholds and cause a\ncached process to be killed(we measured performance swapping out vs. \nzapping the memory by killing a process. Unsurprisingly, zapping is 10x\ntimes faster even though we use zram which is much faster than real\nstorage) so kill from lmkd will often satisfy the high zone watermark,\nresulting in very few pages actually being moved to swap.\n\n- Approach\n\nThe approach we chose was to use a new interface to allow userspace to\nproactively reclaim entire processes by leveraging platform information. \nThis allowed us to bypass the inaccuracy of the kernel’s LRUs for pages\nthat are known to be cold from userspace and to avoid races with lmkd by\nreclaiming apps as soon as they entered the cached state. Additionally,\nit could provide many chances for platform to use much information to\noptimize memory efficiency.\n\nTo achieve the goal, the patchset introduce two new options for madvise. \nOne is MADV_COLD which will deactivate activated pages and the other is\nMADV_PAGEOUT which will reclaim private pages instantly. These new\noptions complement MADV_DONTNEED and MADV_FREE by adding non-destructive\nways to gain some free memory space. MADV_PAGEOUT is similar to\nMADV_DONTNEED in a way that it hints the kernel that memory region is not\ncurrently needed and should be reclaimed immediately; MADV_COLD is similar\nto MADV_FREE in a way that it hints the kernel that memory region is not\ncurrently needed and should be reclaimed when memory pressure rises.\n\n\nThis patch (of 5):\n\nWhen a process expects no accesses to a certain memory range, it could\ngive a hint to kernel that the pages can be reclaimed when memory pressure\nhappens but data should be preserved for future use. This could reduce\nworkingset eviction so it ends up increasing performance.\n\nThis patch introduces the new MADV_COLD hint to madvise(2) syscall. \nMADV_COLD can be used by a process to mark a memory range as not expected\nto be used in the near future. The hint can help kernel in deciding which\npages to evict early during memory pressure.\n\nIt works for every LRU pages like MADV_[DONTNEED|FREE]. IOW, It moves\n\n\tactive file page -\u003e inactive file LRU\n\tactive anon page -\u003e inacdtive anon LRU\n\nUnlike MADV_FREE, it doesn\u0027t move active anonymous pages to inactive file\nLRU\u0027s head because MADV_COLD is a little bit different symantic. \nMADV_FREE means it\u0027s okay to discard when the memory pressure because the\ncontent of the page is *garbage* so freeing such pages is almost zero\noverhead since we don\u0027t need to swap out and access afterward causes just\nminor fault. Thus, it would make sense to put those freeable pages in\ninactive file LRU to compete other used-once pages. It makes sense for\nimplmentaion point of view, too because it\u0027s not swapbacked memory any\nlonger until it would be re-dirtied. Even, it could give a bonus to make\nthem be reclaimed on swapless system. However, MADV_COLD doesn\u0027t mean\ngarbage so reclaiming them requires swap-out/in in the end so it\u0027s bigger\ncost. Since we have designed VM LRU aging based on cost-model, anonymous\ncold pages would be better to position inactive anon\u0027s LRU list, not file\nLRU. Furthermore, it would help to avoid unnecessary scanning if system\ndoesn\u0027t have a swap device. Let\u0027s start simpler way without adding\ncomplexity at this moment. However, keep in mind, too that it\u0027s a caveat\nthat workloads with a lot of pages cache are likely to ignore MADV_COLD on\nanonymous memory because we rarely age anonymous LRU lists.\n\n* man-page material\n\nMADV_COLD (since Linux x.x)\n\nPages in the specified regions will be treated as less-recently-accessed\ncompared to pages in the system with similar access frequencies. In\ncontrast to MADV_FREE, the contents of the region are preserved regardless\nof subsequent writes to pages.\n\nMADV_COLD cannot be applied to locked pages, Huge TLB pages, or VM_PFNMAP\npages.\n\nLink: http://lkml.kernel.org/r/20190726023435.214162-2-minchan@kernel.org\nSigned-off-by: Minchan Kim \u003cminchan@kernel.org\u003e\nReported-by: kbuild test robot \u003clkp@intel.com\u003e\nAcked-by: Michal Hocko \u003cmhocko@suse.com\u003e\nAcked-by: Johannes Weiner \u003channes@cmpxchg.org\u003e\nCc: James E.J. Bottomley \u003cJames.Bottomley@HansenPartnership.com\u003e\nCc: Richard Henderson \u003crth@twiddle.net\u003e\nCc: Ralf Baechle \u003cralf@linux-mips.org\u003e\nCc: Chris Zankel \u003cchris@zankel.net\u003e\nCc: Johannes Weiner \u003channes@cmpxchg.org\u003e\nCc: Daniel Colascione \u003cdancol@google.com\u003e\nCc: Dave Hansen \u003cdave.hansen@intel.com\u003e\nCc: Hillf Danton \u003chdanton@sina.com\u003e\nCc: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\nCc: Kirill A. Shutemov \u003ckirill.shutemov@linux.intel.com\u003e\nCc: Oleksandr Natalenko \u003coleksandr@redhat.com\u003e\nCc: Shakeel Butt \u003cshakeelb@google.com\u003e\nCc: Sonny Rao \u003csonnyrao@google.com\u003e\nCc: Suren Baghdasaryan \u003csurenb@google.com\u003e\nCc: Tim Murray \u003ctimmurray@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "289a83e83358048935fc416a3d19f8f4fc3be1b8", "tree": "c6a5e444ed0f7f3de34b7a380e4c799c78587871", "parents": [ "0669ab4a7a93e1b0749cc6f04caa2fc92790ff4a" ], "author": { "name": "Catalin Marinas", "email": "catalin.marinas@arm.com", "time": "Wed Sep 04 05:02:48 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:48 2019 +0000" }, "message": "mm: untag user pointers in mmap/munmap/mremap/brk\n\nThere isn\u0027t a good reason to differentiate between the user address space\nlayout modification syscalls and the other memory permission/attributes\nones (e.g. mprotect, madvise) w.r.t. the tagged address ABI. Untag the\nuser addresses on entry to these functions.\n\nLink: http://lkml.kernel.org/r/20190821164730.47450-2-catalin.marinas@arm.com\nSigned-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nAcked-by: Will Deacon \u003cwill@kernel.org\u003e\nAcked-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nCc: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nCc: Szabolcs Nagy \u003cszabolcs.nagy@arm.com\u003e\nCc: Kevin Brodsky \u003ckevin.brodsky@arm.com\u003e\nCc: Dave P Martin \u003cDave.Martin@arm.com\u003e\nCc: Dave Hansen \u003cdave.hansen@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "0669ab4a7a93e1b0749cc6f04caa2fc92790ff4a", "tree": "5a40f00130bed7be00bf7b0d400a2e37a18a700c", "parents": [ "e179ac35acd0456b46380e1dc8db15fef1c034f0" ], "author": { "name": "Andrey Konovalov", "email": "andreyknvl@google.com", "time": "Wed Sep 04 05:02:48 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:48 2019 +0000" }, "message": "vfio/type1: untag user pointers in vaddr_get_pfn\n\nThis patch is a part of a series that extends kernel ABI to allow to pass\ntagged user pointers (with the top byte set to something else other than\n0x00) as syscall arguments.\n\nvaddr_get_pfn() uses provided user pointers for vma lookups, which can\nonly by done with untagged pointers.\n\nUntag user pointers in this function.\n\nLink: http://lkml.kernel.org/r/87422b4d72116a975896f2b19b00f38acbd28f33.1563904656.git.andreyknvl@google.com\nSigned-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nReviewed-by: Eric Auger \u003ceric.auger@redhat.com\u003e\nReviewed-by: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nReviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nReviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: Dave Hansen \u003cdave.hansen@intel.com\u003e\nCc: Will Deacon \u003cwill@kernel.org\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\nCc: Jens Wiklander \u003cjens.wiklander@linaro.org\u003e\nCc: Khalid Aziz \u003ckhalid.aziz@oracle.com\u003e\nCc: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nCc: Mike Rapoport \u003crppt@linux.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "e179ac35acd0456b46380e1dc8db15fef1c034f0", "tree": "d3bf5e56cfb688ddc963383d093881e94b307640", "parents": [ "10783642bf75fad61e38b9714bcd353cb1b9dd12" ], "author": { "name": "Andrey Konovalov", "email": "andreyknvl@google.com", "time": "Wed Sep 04 05:02:48 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:48 2019 +0000" }, "message": "tee/shm: untag user pointers in tee_shm_register\n\nThis patch is a part of a series that extends kernel ABI to allow to pass\ntagged user pointers (with the top byte set to something else other than\n0x00) as syscall arguments.\n\ntee_shm_register()-\u003eoptee_shm_unregister()-\u003echeck_mem_type() uses provided\nuser pointers for vma lookups (via __check_mem_type()), which can only by\ndone with untagged pointers.\n\nUntag user pointers in this function.\n\nLink: http://lkml.kernel.org/r/4b993f33196b3566ac81285ff8453219e2079b45.1563904656.git.andreyknvl@google.com\nSigned-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nReviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\nAcked-by: Jens Wiklander \u003cjens.wiklander@linaro.org\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nCc: Dave Hansen \u003cdave.hansen@intel.com\u003e\nCc: Eric Auger \u003ceric.auger@redhat.com\u003e\nCc: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\nCc: Khalid Aziz \u003ckhalid.aziz@oracle.com\u003e\nCc: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nCc: Mike Rapoport \u003crppt@linux.ibm.com\u003e\nCc: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nCc: Will Deacon \u003cwill@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "10783642bf75fad61e38b9714bcd353cb1b9dd12", "tree": "f5414d19460e426ba735090bdb1b6ca39cd6f005", "parents": [ "bc6d3756dfa8eaf0ed09c0989f198ac526f5d266" ], "author": { "name": "Andrey Konovalov", "email": "andreyknvl@google.com", "time": "Wed Sep 04 05:02:47 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:47 2019 +0000" }, "message": "media/v4l2-core: untag user pointers in videobuf_dma_contig_user_get\n\nThis patch is a part of a series that extends kernel ABI to allow to pass\ntagged user pointers (with the top byte set to something else other than\n0x00) as syscall arguments.\n\nvideobuf_dma_contig_user_get() uses provided user pointers for vma\nlookups, which can only by done with untagged pointers.\n\nUntag the pointers in this function.\n\nLink: http://lkml.kernel.org/r/100436d5f8e4349a78f27b0bbb27e4801fcb946b.1563904656.git.andreyknvl@google.com\nSigned-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nReviewed-by: Khalid Aziz \u003ckhalid.aziz@oracle.com\u003e\nReviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\nAcked-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nCc: Dave Hansen \u003cdave.hansen@intel.com\u003e\nCc: Eric Auger \u003ceric.auger@redhat.com\u003e\nCc: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\nCc: Jens Wiklander \u003cjens.wiklander@linaro.org\u003e\nCc: Mike Rapoport \u003crppt@linux.ibm.com\u003e\nCc: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nCc: Will Deacon \u003cwill@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "bc6d3756dfa8eaf0ed09c0989f198ac526f5d266", "tree": "0a785d744eb4f246fdbfcefdc363014c46f4345d", "parents": [ "f1c200bb8e94d62726b2c7d04c0a70a7ad422201" ], "author": { "name": "Andrey Konovalov", "email": "andreyknvl@google.com", "time": "Wed Sep 04 05:02:47 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:47 2019 +0000" }, "message": "drm/radeon: untag user pointers in radeon_gem_userptr_ioctl\n\nThis patch is a part of a series that extends kernel ABI to allow to pass\ntagged user pointers (with the top byte set to something else other than\n0x00) as syscall arguments.\n\nIn radeon_gem_userptr_ioctl() an MMU notifier is set up with a (tagged)\nuserspace pointer. The untagged address should be used so that MMU\nnotifiers for the untagged address get correctly matched up with the right\nBO. This funcation also calls radeon_ttm_tt_pin_userptr(), which uses\nprovided user pointers for vma lookups, which can only by done with\nuntagged pointers.\n\nThis patch untags user pointers in radeon_gem_userptr_ioctl().\n\nLink: http://lkml.kernel.org/r/c856babeb67195b35603b8d5ba386a2819cec5ff.1563904656.git.andreyknvl@google.com\nSigned-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nReviewed-by: Khalid Aziz \u003ckhalid.aziz@oracle.com\u003e\nReviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\nSuggested-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\nAcked-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nCc: Dave Hansen \u003cdave.hansen@intel.com\u003e\nCc: Eric Auger \u003ceric.auger@redhat.com\u003e\nCc: Jens Wiklander \u003cjens.wiklander@linaro.org\u003e\nCc: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nCc: Mike Rapoport \u003crppt@linux.ibm.com\u003e\nCc: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nCc: Will Deacon \u003cwill@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "f1c200bb8e94d62726b2c7d04c0a70a7ad422201", "tree": "a82901b2a749697605f101be95888b5bb7c8a05e", "parents": [ "93040f1e8486b021d1c4e08dcf817286c9649c23" ], "author": { "name": "Andrey Konovalov", "email": "andreyknvl@google.com", "time": "Wed Sep 04 05:02:47 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:47 2019 +0000" }, "message": "drm/amdgpu: untag user pointers\n\nThis patch is a part of a series that extends kernel ABI to allow to pass\ntagged user pointers (with the top byte set to something else other than\n0x00) as syscall arguments.\n\nIn amdgpu_gem_userptr_ioctl() and amdgpu_amdkfd_gpuvm.c/init_user_pages()\nan MMU notifier is set up with a (tagged) userspace pointer. The untagged\naddress should be used so that MMU notifiers for the untagged address get\ncorrectly matched up with the right BO. This patch untag user pointers in\namdgpu_gem_userptr_ioctl() for the GEM case and in amdgpu_amdkfd_gpuvm_\nalloc_memory_of_gpu() for the KFD case. This also makes sure that an\nuntagged pointer is passed to amdgpu_ttm_tt_get_user_pages(), which uses\nit for vma lookups.\n\nLink: http://lkml.kernel.org/r/d684e1df08f2ecb6bc292e222b64fa9efbc26e69.1563904656.git.andreyknvl@google.com\nSigned-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nReviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\nSuggested-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\nAcked-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nCc: Dave Hansen \u003cdave.hansen@intel.com\u003e\nCc: Eric Auger \u003ceric.auger@redhat.com\u003e\nCc: Jens Wiklander \u003cjens.wiklander@linaro.org\u003e\nCc: Khalid Aziz \u003ckhalid.aziz@oracle.com\u003e\nCc: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nCc: Mike Rapoport \u003crppt@linux.ibm.com\u003e\nCc: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nCc: Will Deacon \u003cwill@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "93040f1e8486b021d1c4e08dcf817286c9649c23", "tree": "97df795f9cf748ddfca81de9c49415a641b1e79e", "parents": [ "c3569fb4eb121a241f03bb068b257fa30117c36c" ], "author": { "name": "Andrey Konovalov", "email": "andreyknvl@google.com", "time": "Wed Sep 04 05:02:46 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:46 2019 +0000" }, "message": "userfaultfd: untag user pointers\n\nThis patch is a part of a series that extends kernel ABI to allow to pass\ntagged user pointers (with the top byte set to something else other than\n0x00) as syscall arguments.\n\nuserfaultfd code use provided user pointers for vma lookups, which can\nonly by done with untagged pointers.\n\nUntag user pointers in validate_range().\n\nLink: http://lkml.kernel.org/r/cdc59ddd7011012ca2e689bc88c3b65b1ea7e413.1563904656.git.andreyknvl@google.com\nSigned-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nReviewed-by: Mike Rapoport \u003crppt@linux.ibm.com\u003e\nReviewed-by: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nReviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nReviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Dave Hansen \u003cdave.hansen@intel.com\u003e\nCc: Eric Auger \u003ceric.auger@redhat.com\u003e\nCc: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\nCc: Jens Wiklander \u003cjens.wiklander@linaro.org\u003e\nCc: Khalid Aziz \u003ckhalid.aziz@oracle.com\u003e\nCc: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nCc: Will Deacon \u003cwill@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "c3569fb4eb121a241f03bb068b257fa30117c36c", "tree": "fe34c32eddc915aad6e483381724dfd8bb1480e1", "parents": [ "2cb3b141f6d01a6ac91a988c75a99171fab90939" ], "author": { "name": "Andrey Konovalov", "email": "andreyknvl@google.com", "time": "Wed Sep 04 05:02:46 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:46 2019 +0000" }, "message": "fs/namespace: untag user pointers in copy_mount_options\n\nThis patch is a part of a series that extends kernel ABI to allow to pass\ntagged user pointers (with the top byte set to something else other than\n0x00) as syscall arguments.\n\nIn copy_mount_options a user address is being subtracted from TASK_SIZE. \nIf the address is lower than TASK_SIZE, the size is calculated to not\nallow the exact_copy_from_user() call to cross TASK_SIZE boundary. \nHowever if the address is tagged, then the size will be calculated\nincorrectly.\n\nUntag the address before subtracting.\n\nLink: http://lkml.kernel.org/r/1de225e4a54204bfd7f25dac2635e31aa4aa1d90.1563904656.git.andreyknvl@google.com\nSigned-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nReviewed-by: Khalid Aziz \u003ckhalid.aziz@oracle.com\u003e\nReviewed-by: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nReviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\nReviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Dave Hansen \u003cdave.hansen@intel.com\u003e\nCc: Eric Auger \u003ceric.auger@redhat.com\u003e\nCc: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\nCc: Jens Wiklander \u003cjens.wiklander@linaro.org\u003e\nCc: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nCc: Mike Rapoport \u003crppt@linux.ibm.com\u003e\nCc: Will Deacon \u003cwill@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "2cb3b141f6d01a6ac91a988c75a99171fab90939", "tree": "4370483a6752f8d94ce2dd2dc365c1ee72e998cb", "parents": [ "34927f2d8d29d85d9bca8a21bf4e8557676ded53" ], "author": { "name": "Andrey Konovalov", "email": "andreyknvl@google.com", "time": "Wed Sep 04 05:02:46 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:46 2019 +0000" }, "message": "mm: untag user pointers in get_vaddr_frames\n\nThis patch is a part of a series that extends kernel ABI to allow to pass\ntagged user pointers (with the top byte set to something else other than\n0x00) as syscall arguments.\n\nget_vaddr_frames uses provided user pointers for vma lookups, which can\nonly by done with untagged pointers. Instead of locating and changing all\ncallers of this function, perform untagging in it.\n\nLink: http://lkml.kernel.org/r/28f05e49c92b2a69c4703323d6c12208f3d881fe.1563904656.git.andreyknvl@google.com\nSigned-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nReviewed-by: Khalid Aziz \u003ckhalid.aziz@oracle.com\u003e\nReviewed-by: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nAcked-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nReviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Dave Hansen \u003cdave.hansen@intel.com\u003e\nCc: Eric Auger \u003ceric.auger@redhat.com\u003e\nCc: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\nCc: Jens Wiklander \u003cjens.wiklander@linaro.org\u003e\nCc: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nCc: Mike Rapoport \u003crppt@linux.ibm.com\u003e\nCc: Will Deacon \u003cwill@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "34927f2d8d29d85d9bca8a21bf4e8557676ded53", "tree": "b29a74ac9b7bfa24e8102e00256e42c5a60b5859", "parents": [ "b878b7d5f0dd7170fd9ff89825081880f4b1d5cc" ], "author": { "name": "Andrey Konovalov", "email": "andreyknvl@google.com", "time": "Wed Sep 04 05:02:46 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:46 2019 +0000" }, "message": "mm: untag user pointers in mm/gup.c\n\nThis patch is a part of a series that extends kernel ABI to allow to pass\ntagged user pointers (with the top byte set to something else other than\n0x00) as syscall arguments.\n\nmm/gup.c provides a kernel interface that accepts user addresses and\nmanipulates user pages directly (for example get_user_pages, that is used\nby the futex syscall). Since a user can provided tagged addresses, we\nneed to handle this case.\n\nAdd untagging to gup.c functions that use user addresses for vma lookups.\n\nLink: http://lkml.kernel.org/r/4731bddba3c938658c10ff4ed55cc01c60f4c8f8.1563904656.git.andreyknvl@google.com\nSigned-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nReviewed-by: Khalid Aziz \u003ckhalid.aziz@oracle.com\u003e\nReviewed-by: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nReviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\nReviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Dave Hansen \u003cdave.hansen@intel.com\u003e\nCc: Eric Auger \u003ceric.auger@redhat.com\u003e\nCc: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\nCc: Jens Wiklander \u003cjens.wiklander@linaro.org\u003e\nCc: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nCc: Mike Rapoport \u003crppt@linux.ibm.com\u003e\nCc: Will Deacon \u003cwill@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "b878b7d5f0dd7170fd9ff89825081880f4b1d5cc", "tree": "8064f55e9b70c692190a3efb4c9fa1afbda07b94", "parents": [ "4d258969e634eac030c4f4921a1562d895c55702" ], "author": { "name": "Andrey Konovalov", "email": "andreyknvl@google.com", "time": "Wed Sep 04 05:02:45 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:45 2019 +0000" }, "message": "mm: untag user pointers passed to memory syscalls\n\nThis patch is a part of a series that extends kernel ABI to allow to pass\ntagged user pointers (with the top byte set to something else other than\n0x00) as syscall arguments.\n\nThis patch allows tagged pointers to be passed to the following memory\nsyscalls: get_mempolicy, madvise, mbind, mincore, mlock, mlock2, mprotect,\nmremap, msync, munlock, move_pages.\n\nThe mmap and mremap syscalls do not currently accept tagged addresses. \nArchitectures may interpret the tag as a background colour for the\ncorresponding vma.\n\nLink: http://lkml.kernel.org/r/aaf0c0969d46b2feb9017f3e1b3ef3970b633d91.1563904656.git.andreyknvl@google.com\nSigned-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nReviewed-by: Khalid Aziz \u003ckhalid.aziz@oracle.com\u003e\nReviewed-by: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nReviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nReviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Dave Hansen \u003cdave.hansen@intel.com\u003e\nCc: Eric Auger \u003ceric.auger@redhat.com\u003e\nCc: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\nCc: Jens Wiklander \u003cjens.wiklander@linaro.org\u003e\nCc: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nCc: Mike Rapoport \u003crppt@linux.ibm.com\u003e\nCc: Will Deacon \u003cwill@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "4d258969e634eac030c4f4921a1562d895c55702", "tree": "fbdd5de30f0f3e1f31a13aa7ce04a39b92fb31fb", "parents": [ "7d96515bf5e38d9ff8da85e689e57729035527ef" ], "author": { "name": "Andrey Konovalov", "email": "andreyknvl@google.com", "time": "Wed Sep 04 05:02:45 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:45 2019 +0000" }, "message": "lib: untag user pointers in strn*_user\n\nPatch series \"arm64: untag user pointers passed to the kernel\", v19.\n\n\u003d\u003d\u003d Overview\n\narm64 has a feature called Top Byte Ignore, which allows to embed pointer\ntags into the top byte of each pointer. Userspace programs (such as\nHWASan, a memory debugging tool [1]) might use this feature and pass\ntagged user pointers to the kernel through syscalls or other interfaces.\n\nRight now the kernel is already able to handle user faults with tagged\npointers, due to these patches:\n\n1. 81cddd65 (\"arm64: traps: fix userspace cache maintenance emulation on a\n tagged pointer\")\n2. 7dcd9dd8 (\"arm64: hw_breakpoint: fix watchpoint matching for tagged\n\t pointers\")\n3. 276e9327 (\"arm64: entry: improve data abort handling of tagged\n\t pointers\")\n\nThis patchset extends tagged pointer support to syscall arguments.\n\nAs per the proposed ABI change [3], tagged pointers are only allowed to be\npassed to syscalls when they point to memory ranges obtained by anonymous\nmmap() or sbrk() (see the patchset [3] for more details).\n\nFor non-memory syscalls this is done by untaging user pointers when the\nkernel performs pointer checking to find out whether the pointer comes\nfrom userspace (most notably in access_ok). The untagging is done only\nwhen the pointer is being checked, the tag is preserved as the pointer\nmakes its way through the kernel and stays tagged when the kernel\ndereferences the pointer when perfoming user memory accesses.\n\nThe mmap and mremap (only new_addr) syscalls do not currently accept\ntagged addresses. Architectures may interpret the tag as a background\ncolour for the corresponding vma.\n\nOther memory syscalls (mprotect, etc.) don\u0027t do user memory accesses but\nrather deal with memory ranges, and untagged pointers are better suited to\ndescribe memory ranges internally. Thus for memory syscalls we untag\npointers completely when they enter the kernel.\n\n\u003d\u003d\u003d Other approaches\n\nOne of the alternative approaches to untagging that was considered is to\ncompletely strip the pointer tag as the pointer enters the kernel with\nsome kind of a syscall wrapper, but that won\u0027t work with the countless\nnumber of different ioctl calls. With this approach we would need a\ncustom wrapper for each ioctl variation, which doesn\u0027t seem practical.\n\nAn alternative approach to untagging pointers in memory syscalls prologues\nis to inspead allow tagged pointers to be passed to find_vma() (and other\nvma related functions) and untag them there. Unfortunately, a lot of\nfind_vma() callers then compare or subtract the returned vma start and end\nfields against the pointer that was being searched. Thus this approach\nwould still require changing all find_vma() callers.\n\n\u003d\u003d\u003d Testing\n\nThe following testing approaches has been taken to find potential issues\nwith user pointer untagging:\n\n1. Static testing (with sparse [2] and separately with a custom static\n analyzer based on Clang) to track casts of __user pointers to integer\n types to find places where untagging needs to be done.\n\n2. Static testing with grep to find parts of the kernel that call\n find_vma() (and other similar functions) or directly compare against\n vm_start/vm_end fields of vma.\n\n3. Static testing with grep to find parts of the kernel that compare\n user pointers with TASK_SIZE or other similar consts and macros.\n\n4. Dynamic testing: adding BUG_ON(has_tag(addr)) to find_vma() and running\n a modified syzkaller version that passes tagged pointers to the kernel.\n\nBased on the results of the testing the requried patches have been added\nto the patchset.\n\n\u003d\u003d\u003d Notes\n\nThis patchset is meant to be merged together with \"arm64 relaxed ABI\" [3].\n\nThis patchset is a prerequisite for ARM\u0027s memory tagging hardware feature\nsupport [4].\n\nThis patchset has been merged into the Pixel 2 \u0026 3 kernel trees and is\nnow being used to enable testing of Pixel phones with HWASan.\n\nThanks!\n\n[1] http://clang.llvm.org/docs/HardwareAssistedAddressSanitizerDesign.html\n\n[2] https://github.com/lucvoo/sparse-dev/commit/5f960cb10f56ec2017c128ef9d16060e0145f292\n\n[3] https://lkml.org/lkml/2019/6/12/745\n\n[4] https://community.arm.com/processors/b/blog/posts/arm-a-profile-architecture-2018-developments-armv85a\n\n\nThis patch (of 11)\n\nThis patch is a part of a series that extends kernel ABI to allow to pass\ntagged user pointers (with the top byte set to something else other than\n0x00) as syscall arguments.\n\nstrncpy_from_user and strnlen_user accept user addresses as arguments, and\ndo not go through the same path as copy_from_user and others, so here we\nneed to handle the case of tagged user addresses separately.\n\nUntag user pointers passed to these functions.\n\nNote, that this patch only temporarily untags the pointers to perform\nvalidity checks, but then uses them as is to perform user memory accesses.\n\nLink: http://lkml.kernel.org/r/c5a78bcad3e94d6cda71fcaa60a423231ae71e4c.1563904656.git.andreyknvl@google.com\nSigned-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\nReviewed-by: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nReviewed-by: Khalid Aziz \u003ckhalid.aziz@oracle.com\u003e\nAcked-by: Kees Cook \u003ckeescook@chromium.org\u003e\nReviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Dave Hansen \u003cdave.hansen@intel.com\u003e\nCc: Eric Auger \u003ceric.auger@redhat.com\u003e\nCc: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\nCc: Jens Wiklander \u003cjens.wiklander@linaro.org\u003e\nCc: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nCc: Mike Rapoport \u003crppt@linux.ibm.com\u003e\nCc: Will Deacon \u003cwill@kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "7d96515bf5e38d9ff8da85e689e57729035527ef", "tree": "f302ad5fa44cefd08c1380e491c183eb194636e0", "parents": [ "da26adbf34821aeb7942cc432b18b0132aabfd4a" ], "author": { "name": "Randy Dunlap", "email": "rdunlap@infradead.org", "time": "Wed Sep 04 05:02:45 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:45 2019 +0000" }, "message": "pinctrl: fix pxa2xx.c build warnings\n\nAdd #include of \u003clinux/pinctrl/machine.h\u003e to fix build\nwarnings in pinctrl-pxa2xx.c. Fixes these warnings:\n\nIn file included from ../drivers/pinctrl/pxa/pinctrl-pxa2xx.c:24:0:\n../drivers/pinctrl/pxa/../pinctrl-utils.h:36:8: warning: `enum pinctrl_map_type\u0027 declared inside parameter list [enabled by default]\n enum pinctrl_map_type type);\n ^\n../drivers/pinctrl/pxa/../pinctrl-utils.h:36:8: warning: its scope is only this definition or declaration, which is probably not what you want [enabled by default]\n\nLink: http://lkml.kernel.org/r/0024542e-cba9-8f13-6c18-32d0050a6007@infradead.org\nSigned-off-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\nCc: Robert Jarzmik \u003crobert.jarzmik@free.fr\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "da26adbf34821aeb7942cc432b18b0132aabfd4a", "tree": "f136e83fba4235d3e7a4201c01d8b5e796953a50", "parents": [ "6898c89d551f0d10fbdb68b9a57a71e4a2804428" ], "author": { "name": "Hugh Dickins", "email": "hughd@google.com", "time": "Wed Sep 04 05:02:44 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:44 2019 +0000" }, "message": "tmpfs: fixups to use of the new mount API\n\nSeveral fixups to shmem_parse_param() and tmpfs use of new mount API:\n\nmm/shmem.c manages filesystem named \"tmpfs\": revert \"shmem\" to \"tmpfs\"\nin its mount error messages.\n\n/sys/kernel/mm/transparent_hugepage/shmem_enabled has valid options\n\"deny\" and \"force\", but they are not valid as tmpfs \"huge\" options.\n\nThe \"size\" param is an alternative to \"nr_blocks\", and needs to be\nrecognized as changing max_blocks. And where there\u0027s ambiguity, it\u0027s\nbetter to mention \"size\" than \"nr_blocks\" in messages, since \"size\" is\nthe variant shown in /proc/mounts.\n\nshmem_apply_options() left ctx-\u003empol as the new mpol, so then it was\nfreed in shmem_free_fc(), and the filesystem went on to use-after-free.\n\nshmem_parse_param() issue \"tmpfs: Bad value for \u0027%s\u0027\" messages just\nlike fs_parse() would, instead of a different wording. Where config\ndisables \"mpol\" or \"huge\", say \"tmpfs: Unsupported parameter \u0027%s\u0027\".\n\nLink: http://lkml.kernel.org/r/alpine.LSU.2.11.1908191503290.1253@eggly.anvils\nFixes: 144df3b288c41 (\"vfs: Convert ramfs, shmem, tmpfs, devtmpfs, rootfs to use the new mount API\")\nSigned-off-by: Hugh Dickins \u003chughd@google.com\u003e\nCc: David Howells \u003cdhowells@redhat.com\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "6898c89d551f0d10fbdb68b9a57a71e4a2804428", "tree": "fd1b3f15d98d19f67268d6ec14db4116fbb28f2f", "parents": [ "b1437d148bfea97f4867d9bb17e3bf37d642a69a" ], "author": { "name": "Andrew Morton", "email": "akpm@linux-foundation.org", "time": "Wed Sep 04 05:02:44 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:44 2019 +0000" }, "message": "diff sucks\n\nCc: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nCc: Randy Dunlap \u003crdunlap@infradead.org\u003e\nCc: Mark Brown \u003cbroonie@kernel.org\u003e\nCc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nCc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nCc: Jani Nikula \u003cjani.nikula@linux.intel.com\u003e\nCc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\nCc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\nCc: Intel Graphics \u003cintel-gfx@lists.freedesktop.org\u003e\nCc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nCc: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "b1437d148bfea97f4867d9bb17e3bf37d642a69a", "tree": "5dad8d5606468656c8285919dfd555e5c01ba098", "parents": [ "a75ed18e283a90be0cfaf49291961d1cc55a2c81" ], "author": { "name": "Andrew Morton", "email": "akpm@linux-foundation.org", "time": "Wed Sep 04 05:02:44 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:44 2019 +0000" }, "message": "linux-next-rejects\n\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "a75ed18e283a90be0cfaf49291961d1cc55a2c81", "tree": "2e28ab916907d526876966a743acdb24ff09e4dd", "parents": [ "3ca91a98962397cad578261559e2f1fdccdadfa9" ], "author": { "name": "Andrew Morton", "email": "akpm@linux-foundation.org", "time": "Wed Sep 04 05:02:43 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:43 2019 +0000" }, "message": "linux-next\n\nGIT 5319f37db5f63a4d94f48a17a26573a22ff0dfd1\n\ncommit ecd55e367f3d706788632e176ec6b94e1a72a07c\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Mon Sep 2 09:43:23 2019 +0800\n\n usb: chipidea: msm: Use device-managed registration API\n \n Use devm_reset_controller_register to get rid\n of manual unregistration.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Peter Chen \u003cpeter.chen@nxp.com\u003e\n\ncommit 345f0254e5b2f4090e4a00ebc996e07e9bdcd070\nAuthor: Maya Nakamura \u003cm.maya.nakamura@gmail.com\u003e\nDate: Fri Jul 12 08:27:38 2019 +0000\n\n HID: hv: Remove dependencies on PAGE_SIZE for ring buffer\n \n Define the ring buffer size as a constant expression because it should\n not depend on the guest page size.\n \n Signed-off-by: Maya Nakamura \u003cm.maya.nakamura@gmail.com\u003e\n Reviewed-by: Michael Kelley \u003cmikelley@microsoft.com\u003e\n Acked-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n Signed-off-by: Sasha Levin \u003csashal@kernel.org\u003e\n\ncommit 039b1f4f24ecc8493b6bb9d70b4b78750d1b35c2\nAuthor: Fernando Fernandez Mancera \u003cffmancera@riseup.net\u003e\nDate: Sun Sep 1 11:48:08 2019 +0200\n\n netfilter: nft_socket: fix erroneous socket assignment\n \n The socket assignment is wrong, see skb_orphan():\n When skb-\u003edestructor callback is not set, but skb-\u003esk is set, this hits BUG().\n \n Link: https://bugzilla.redhat.com/show_bug.cgi?id\u003d1651813\n Fixes: 554ced0a6e29 (\"netfilter: nf_tables: add support for native socket matching\")\n Signed-off-by: Fernando Fernandez Mancera \u003cffmancera@riseup.net\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit 48bd0d68cd4b849c0bb99dc1fb01b013ad4aa942\nAuthor: Leonardo Bras \u003cleonardo@linux.ibm.com\u003e\nDate: Sat Aug 31 01:40:33 2019 -0300\n\n netfilter: bridge: Drops IPv6 packets if IPv6 module is not loaded\n \n A kernel panic can happen if a host has disabled IPv6 on boot and have to\n process guest packets (coming from a bridge) using it\u0027s ip6tables.\n \n IPv6 packets need to be dropped if the IPv6 module is not loaded, and the\n host ip6tables will be used.\n \n Signed-off-by: Leonardo Bras \u003cleonardo@linux.ibm.com\u003e\n Acked-by: Florian Westphal \u003cfw@strlen.de\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit 7f7378618b4103c083db7de5017df958f8ada070\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Fri Aug 30 16:23:16 2019 -0700\n\n power: supply: cpcap-charger: Enable vbus boost voltage\n \n We are currently not enabling VBUS boost for cpcap when in host mode.\n This means the VBUS is fed at the battery voltage level, which can cause\n flakeyness enumerating devices.\n \n Looks like the boost control for VBUS is CPCAP_BIT_VBUS_SWITCH that we\n must enable in the charger for nice 4.92 V VBUS output. And looks like\n we must not use the STBY pin enabling but must instead use manual VBUS\n control in phy-cpcap-usb.\n \n We want to do this in cpcap_charger_vbus_work() and also set a flag for\n feeding_vbus to avoid races between USB detection and charger detection,\n and disable charging if feeding_vbus is set.\n \n Cc: Jacopo Mondi \u003cjacopo@jmondi.org\u003e\n Cc: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n Cc: Marcel Partap \u003cmpartap@gmx.net\u003e\n Cc: Merlijn Wajer \u003cmerlijn@wizzup.org\u003e\n Cc: Michael Scott \u003chashcode0f@gmail.com\u003e\n Cc: NeKit \u003cnekit1000@gmail.com\u003e\n Cc: Pavel Machek \u003cpavel@ucw.cz\u003e\n Cc: Sebastian Reichel \u003csre@kernel.org\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit 7cfd33d997a4c320b6bbce5e9592230dae1e73d2\nAuthor: Yuanjiang Yu \u003cyuanjiang.yu@unisoc.com\u003e\nDate: Wed Jul 31 18:00:28 2019 +0800\n\n power: supply: sc27xx: Add POWER_SUPPLY_PROP_CALIBRATE attribute\n \n Add the \u0027POWER_SUPPLY_PROP_CALIBRATE\u0027 attribute to allow chareger manager\n to calibrate the battery capacity.\n \n Signed-off-by: Yuanjiang Yu \u003cyuanjiang.yu@unisoc.com\u003e\n Signed-off-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit 580665279fb6a16eb119ec80a5be1b3ec7a641a0\nAuthor: Yuanjiang Yu \u003cyuanjiang.yu@unisoc.com\u003e\nDate: Wed Jul 31 18:00:26 2019 +0800\n\n power: supply: sc27xx: Optimize the battery capacity calibration\n \n This patch factors out the capacity calibration into one single function\n to calibrate the battery capacity, and adding more abnormal cases to\n calibrate the capacity when the OCV value is not matchable with current\n capacity.\n \n Moreover we also allow to calibrate the capacity when charger magager\n tries to get current capacity to make sure we give a correct capacity\n for userspace.\n \n Signed-off-by: Yuanjiang Yu \u003cyuanjiang.yu@unisoc.com\u003e\n Signed-off-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit ff062d06948c012cbfd6f5947b599be5ea1c3799\nAuthor: Yuanjiang Yu \u003cyuanjiang.yu@unisoc.com\u003e\nDate: Wed Jul 31 18:00:27 2019 +0800\n\n power: supply: sc27xx: Make sure the alarm capacity is larger than 0\n \n We must make sure the alarm capacity is larger than 0, to help to\n calibrate the low battery capacity.\n \n Signed-off-by: Yuanjiang Yu \u003cyuanjiang.yu@unisoc.com\u003e\n Signed-off-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit 7384b0e7668f722fcf7abe56f61a7212debd0180\nAuthor: Yuanjiang Yu \u003cyuanjiang.yu@unisoc.com\u003e\nDate: Wed Jul 31 18:00:25 2019 +0800\n\n power: supply: sc27xx: Fix the the accuracy issue of coulomb calculation\n \n The Spreadtrum fuel gauge will multiply by 2 for counting the coulomb\n counter to improve the accuracy, which means the value saved in fuel\n gauge is: coulomb counter * 2 * 1000ma_adc. Thus fix the conversion\n formular to improve the accuracy of calculating the battery capacity.\n \n Signed-off-by: Yuanjiang Yu \u003cyuanjiang.yu@unisoc.com\u003e\n Signed-off-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit 168e68d072756d631a210342d0429c969501b7cf\nAuthor: Yuanjiang Yu \u003cyuanjiang.yu@unisoc.com\u003e\nDate: Wed Jul 31 18:00:24 2019 +0800\n\n power: supply: sc27xx: Fix conditon to enable the FGU interrupt\n \n We should allow to enable FGU interrupt to adjust the battery capacity,\n when charging status is POWER_SUPPLY_STATUS_DISCHARGING.\n \n Signed-off-by: Yuanjiang Yu \u003cyuanjiang.yu@unisoc.com\u003e\n Signed-off-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit 7cff19b9a8ff1fdc61655dad4e2009bb44099d70\nAuthor: Yuanjiang Yu \u003cyuanjiang.yu@unisoc.com\u003e\nDate: Wed Jul 31 18:00:23 2019 +0800\n\n power: supply: sc27xx: Add POWER_SUPPLY_PROP_ENERGY_FULL_DESIGN attribute\n \n Add POWER_SUPPLY_PROP_ENERGY_FULL_DESIGN attribute to provide the battery\u0027s\n design capacity for charger manager to calculate the charging counter.\n \n Signed-off-by: Yuanjiang Yu \u003cyuanjiang.yu@unisoc.com\u003e\n Signed-off-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit 46058b2f01c2d274d7d315acf994e8cc93d6a3f9\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Sat Aug 31 09:43:43 2019 +0200\n\n configfs: provide exclusion between IO and removals\n \n Make sure that attribute methods are not called after the item\n has been removed from the tree. To do so, we\n * at the point of no return in removals, grab -\u003efrag_sem\n exclusive and mark the fragment dead.\n * call the methods of attributes with -\u003efrag_sem taken\n shared and only after having verified that the fragment is still\n alive.\n \n The main benefit is for method instances - they are\n guaranteed that the objects they are accessing *and* all ancestors\n are still there. Another win is that we don\u0027t need to bother\n with extra refcount on config_item when opening a file -\n the item will be alive for as long as it stays in the tree, and\n we won\u0027t touch it/attributes/any associated data after it\u0027s\n been removed from the tree.\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit 47320fbe11a6059ae502c9c16b668022fdb4cf76\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Sun Aug 25 19:56:13 2019 -0400\n\n configfs: new object reprsenting tree fragments\n \n Refcounted, hangs of configfs_dirent, created by operations that add\n fragments to configfs tree (mkdir and configfs_register_{subsystem,group}).\n Will be used in the next commit to provide exclusion between fragment\n removal and -\u003eshow/-\u003estore calls.\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit f19e4ed1e1edbfa3c9ccb9fed17759b7d6db24c6\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Thu Aug 29 23:13:30 2019 -0400\n\n configfs_register_group() shouldn\u0027t be (and isn\u0027t) called in rmdirable parts\n \n revert cc57c07343bd \"configfs: fix registered group removal\"\n It was an attempt to handle something that fundamentally doesn\u0027t\n work - configfs_register_group() should never be done in a part\n of tree that can be rmdir\u0027ed. And in mainline it never had been,\n so let\u0027s not borrow trouble; the fix was racy anyway, it would take\n a lot more to make that work and desired semantics is not clear.\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit ff4dd081977da56566a848f071aed8fa92d604a1\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Fri Aug 30 11:30:03 2019 -0400\n\n configfs: stash the data we need into configfs_buffer at open time\n \n simplifies the -\u003eread()/-\u003ewrite()/-\u003erelease() instances nicely\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit f25a9f638c3f9694cfb47d6e561cd6c712149230\nAuthor: Steve French \u003cstfrench@microsoft.com\u003e\nDate: Fri Aug 30 02:12:41 2019 -0500\n\n smb3: add mount option to allow RW caching of share accessed by only 1 client\n \n If a share is known to be only to be accessed by one client, we\n can aggressively cache writes not just reads to it.\n \n Add \"cache\u003d\" option (cache\u003dsingleclient) for mounting read write shares\n (that will not be read or written to from other clients while we have\n it mounted) in order to improve performance.\n \n Signed-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n\ncommit 79436eb4eb39fbb7e40fbd4dfb5a7edfd6280146\nAuthor: Steve French \u003cstfrench@microsoft.com\u003e\nDate: Thu Aug 29 22:33:38 2019 -0500\n\n smb3: add some more descriptive messages about share when mounting cache\u003dro\n \n Add some additional logging so the user can see if the share they\n mounted with cache\u003dro is considered read only by the server\n \n CIFS: Attempting to mount //localhost/test\n CIFS VFS: mounting share with read only caching. Ensure that the share will not be modified while in use.\n CIFS VFS: read only mount of RW share\n \n CIFS: Attempting to mount //localhost/test-ro\n CIFS VFS: mounting share with read only caching. Ensure that the share will not be modified while in use.\n CIFS VFS: mounted to read only share\n \n Signed-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n Reviewed-by: Ronnie Sahlberg \u003clsahlber@redhat.com\u003e\n\ncommit 42297d0a58d178b704a987ba065c9e5a1c39a6e1\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Mon Sep 2 16:10:59 2019 +0100\n\n cifs: fix dereference on ses before it is null checked\n \n The assignment of pointer server dereferences pointer ses, however,\n this dereference occurs before ses is null checked and hence we\n have a potential null pointer dereference. Fix this by only\n dereferencing ses after it has been null checked.\n \n Addresses-Coverity: (\"Dereference before null check\")\n Fixes: 2808c6639104 (\"cifs: add new debugging macro cifs_server_dbg\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n\ncommit 0c588ab0b883ddfe2dfe1074069e861c16ad0f15\nAuthor: Ronnie Sahlberg \u003clsahlber@redhat.com\u003e\nDate: Wed Aug 28 17:15:35 2019 +1000\n\n cifs: add new debugging macro cifs_server_dbg\n \n which can be used from contexts where we have a TCP_Server_Info *server.\n This new macro will prepend the debugging string with \"Server:\u003cservername\u003e \"\n which will help when debugging issues on hosts with many cifs connections\n to several different servers.\n \n Convert a bunch of cifs_dbg(VFS) calls to cifs_server_dbg(VFS)\n \n Signed-off-by: Ronnie Sahlberg \u003clsahlber@redhat.com\u003e\n Signed-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n\ncommit d0e5899489d83463c2d5229b26336058c435e745\nAuthor: Ronnie Sahlberg \u003clsahlber@redhat.com\u003e\nDate: Fri Aug 30 09:53:56 2019 +1000\n\n cifs: use existing handle for compound_op(OP_SET_INFO) when possible\n \n If we already have a writable handle for a path we want to set the\n attributes for then use that instead of a create/set-info/close compound.\n \n Signed-off-by: Ronnie Sahlberg \u003clsahlber@redhat.com\u003e\n Signed-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n\ncommit d93cbd0d4430aed9066b9ebedb5840075c53f1f8\nAuthor: Ronnie Sahlberg \u003clsahlber@redhat.com\u003e\nDate: Fri Aug 30 08:25:46 2019 +1000\n\n cifs: create a helper to find a writeable handle by path name\n \n rename() takes a path for old_file and in SMB2 we used to just create\n a compound for create(old_path)/rename/close().\n If we already have a writable handle we can avoid the create() and close()\n altogether and just use the existing handle.\n \n For this situation, as we avoid doing the create()\n we also avoid triggering an oplock break for the existing handle.\n \n Signed-off-by: Ronnie Sahlberg \u003clsahlber@redhat.com\u003e\n Signed-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n\ncommit e8870aa07894300ec91e0f99dea738e90fbc0df8\nAuthor: Steve French \u003cstfrench@microsoft.com\u003e\nDate: Tue Aug 27 23:58:54 2019 -0500\n\n smb3: add mount option to allow forced caching of read only share\n \n If a share is immutable (at least for the period that it will\n be mounted) it would be helpful to not have to revalidate\n dentries repeatedly that we know can not be changed remotely.\n \n Add \"cache\u003d\" option (cache\u003dro) for mounting read only shares\n in order to improve performance in cases in which we know that\n the share will not be changing while it is in use.\n \n Signed-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n\ncommit 436be5557dcb06bfad91f8456450a5e371001da8\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 23 20:15:35 2019 +0800\n\n cifs: remove set but not used variables\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n fs/cifs/file.c: In function cifs_lock:\n fs/cifs/file.c:1696:24: warning: variable cinode set but not used [-Wunused-but-set-variable]\n fs/cifs/file.c: In function cifs_write:\n fs/cifs/file.c:1765:23: warning: variable cifs_sb set but not used [-Wunused-but-set-variable]\n fs/cifs/file.c: In function collect_uncached_read_data:\n fs/cifs/file.c:3578:20: warning: variable tcon set but not used [-Wunused-but-set-variable]\n \n \u0027cinode\u0027 is never used since introduced by\n commit 03776f4516bc (\"CIFS: Simplify byte range locking code\")\n \u0027cifs_sb\u0027 is not used since commit cb7e9eabb2b5 (\"CIFS: Use\n multicredits for SMB 2.1/3 writes\").\n \u0027tcon\u0027 is not used since commit d26e2903fc10 (\"smb3: fix bytes_read statistics\")\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n\ncommit ba940bacb93e9e5c9068c0e57f4fd9c4c6615fba\nAuthor: Steve French \u003cstfrench@microsoft.com\u003e\nDate: Mon Aug 5 17:07:26 2019 -0500\n\n smb3: Incorrect size for netname negotiate context\n \n It is not null terminated (length was off by two).\n \n Also see similar change to Samba:\n \n https://gitlab.com/samba-team/samba/merge_requests/666\n \n Reported-by: Stefan Metzmacher \u003cmetze@samba.org\u003e\n Signed-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n\ncommit 442606bd677b35af6b1a3d42dca360040353456d\nAuthor: zhengbin \u003czhengbin13@huawei.com\u003e\nDate: Tue Aug 20 22:00:47 2019 +0800\n\n cifs: remove unused variable\n \n In smb3_punch_hole, variable cifsi set but not used, remove it.\n In cifs_lock, variable netfid set but not used, remove it.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: zhengbin \u003czhengbin13@huawei.com\u003e\n Signed-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n\ncommit c12189c51e2664dc6df334d1d021df7d72614329\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Wed Jul 31 10:05:26 2019 +0100\n\n cifs: remove redundant assignment to variable rc\n \n Variable rc is being initialized with a value that is never read\n and rc is being re-assigned a little later on. The assignment is\n redundant and hence can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n\ncommit 7b88cd70b0d0f26f1246eff3e56b50ee691459e0\nAuthor: Steve French \u003cstfrench@microsoft.com\u003e\nDate: Thu Jul 25 18:19:42 2019 -0500\n\n smb3: add missing flag definitions\n \n SMB3 and 3.1.1 added two additional flags including\n the priority mask. Add them to our protocol definitions\n in smb2pdu.h. See MS-SMB2 2.2.1.2\n \n Signed-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n Signed-off-by: Ronnie Sahlberg \u003clsahlber@redhat.com\u003e\n Reviewed-by: Pavel Shilovsky \u003cpshilov@microsoft.com\u003e\n\ncommit b52cb401d6444a5c562fe1d4ec8cc969693937e0\nAuthor: Ronnie Sahlberg \u003clsahlber@redhat.com\u003e\nDate: Thu Jul 25 13:08:43 2019 +1000\n\n cifs: add passthrough for smb2 setinfo\n \n Add support to send smb2 set-info commands from userspace.\n \n Signed-off-by: Ronnie Sahlberg \u003clsahlber@redhat.com\u003e\n Signed-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n Reviewed-by: Paulo Alcantara \u003cpalcantara@suse.de\u003e\n\ncommit ba44b62016c8527d8ad4524c1104ee654eee9e7b\nAuthor: Ronnie Sahlberg \u003clsahlber@redhat.com\u003e\nDate: Tue Jul 16 15:07:08 2019 +1000\n\n cifs: prepare SMB2_Flush to be usable in compounds\n \n Create smb2_flush_init() and smb2_flush_free() so we can use the flush command\n in compounds.\n \n Signed-off-by: Ronnie Sahlberg \u003clsahlber@redhat.com\u003e\n Signed-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n\ncommit f5229c319282c26007632d6ce11c479462a62cba\nAuthor: Steve French \u003cstfrench@microsoft.com\u003e\nDate: Fri Jul 19 08:15:55 2019 +0000\n\n cifs: allow chmod to set mode bits using special sid\n \n When mounting with \"modefromsid\" set mode bits (chmod) by\n adding ACE with special SID (S-1-5-88-3-\u003cmode\u003e) to the ACL.\n Subsequent patch will fix setting default mode on file\n create and mkdir.\n \n See See e.g.\n https://docs.microsoft.com/en-us/previous-versions/windows/it-pro/windows-server-2008-R2-and-2008/hh509017(v\u003dws.10)\n \n Signed-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n\ncommit a2b8f333a3fc6f1b7c59fc907ad35ce713b73ed3\nAuthor: Steve French \u003cstfrench@microsoft.com\u003e\nDate: Fri Jul 19 06:30:07 2019 +0000\n\n cifs: get mode bits from special sid on stat\n \n When mounting with \"modefromsid\" retrieve mode bits from\n special SID (S-1-5-88-3) on stat. Subsequent patch will fix\n setattr (chmod) to save mode bits in S-1-5-88-3-\u003cmode\u003e\n \n Note that when an ACE matching S-1-5-88-3 is not found, we\n default the mode to an approximation based on the owner, group\n and everyone permissions (as with the \"cifsacl\" mount option).\n \n See See e.g.\n https://docs.microsoft.com/en-us/previous-versions/windows/it-pro/windows-server-2008-R2-and-2008/hh509017(v\u003dws.10)\n \n Signed-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n\ncommit 0387ad783b17c70145cff7f9a1a78e786506fdd2\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Tue Jul 23 16:09:19 2019 +0100\n\n fs: cifs: cifsssmb: remove redundant assignment to variable ret\n \n The variable ret is being initialized however this is never read\n and later it is being reassigned to a new value. The initialization\n is redundant and hence can be removed.\n \n Addresses-Coverity: (\"Unused Value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n\ncommit a48ffa319dd172a39f5c5f804e71126077941948\nAuthor: Ronnie Sahlberg \u003clsahlber@redhat.com\u003e\nDate: Wed Jul 24 11:43:49 2019 +1000\n\n cifs: fix a comment for the timeouts when sending echos\n \n Clarify a trivial comment\n \n Signed-off-by: Ronnie Sahlberg \u003clsahlber@redhat.com\u003e\n Signed-off-by: Steve French \u003cstfrench@microsoft.com\u003e\n\ncommit 343160e98cff359d9fc96f3efc95d90f5a9c2d35\nAuthor: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\nDate: Wed Jul 3 10:48:10 2019 +0200\n\n power: supply: max77650: add MODULE_ALIAS()\n \n Define a MODULE_ALIAS() in the charger sub-driver for max77650 so that\n the appropriate module gets loaded together with the core mfd driver.\n \n Signed-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit a2ea3cd78abd0dfdc0b39decb4a74870a2a47dbf\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Mon Sep 2 03:01:03 2019 -0300\n\n drm/radeon: guard against calling an unpaired radeon_mn_unregister()\n \n This check was accidently deleted in the below commit. There are cases\n where the driver will call unregister even though it hasn\u0027t registered\n anything.\n \n CPU 0 Unable to handle kernel paging request at virtual address 0000001c, epc \u003d\u003d 808de6d4, ra \u003d\u003d 804d32ec\n Call Trace:\n [\u003c808de6d4\u003e] mutex_lock+0x8/0x44\n [\u003c804d32ec\u003e] radeon_mn_unregister+0x3c/0xb0\n [\u003c8041583c\u003e] radeon_gem_object_free+0x18/0x2c\n [\u003c803a451c\u003e] drm_gem_object_release_handle+0x74/0xac\n [\u003c803a45d0\u003e] drm_gem_handle_delete+0x7c/0x128\n [\u003c803a5bf4\u003e] drm_ioctl_kernel+0xb0/0x108\n [\u003c803a5e74\u003e] drm_ioctl+0x200/0x3a8\n [\u003c803e07b4\u003e] radeon_drm_ioctl+0x54/0xc0\n [\u003c801214dc\u003e] do_vfs_ioctl+0x4e8/0x81c\n [\u003c80121864\u003e] ksys_ioctl+0x54/0xb0\n [\u003c8001100c\u003e] syscall_common+0x34/0x58\n \n Link: https://lore.kernel.org/r/2fc7ef14-e89a-1f2d-381d-1c9b05da02d3@gmail.com\n Fixes: 534e5f84b7a9 (\"drm/radeon: use mmu_notifier_get/put for struct radeon_mn\")\n Reported-by: Petr Cvek \u003cpetrcvekcz@gmail.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 3ad1aab70acb410e711b3917bba422e4c1d7343b\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Mon Sep 2 02:47:05 2019 -0300\n\n csky: add missing brackets in a macro for tlb.h\n \n As an earlier patch made the macro argument more complicated, compilation\n now fails with:\n \n In file included from mm/madvise.c:30:\n mm/madvise.c: In function \u0027madvise_free_single_vma\u0027:\n arch/csky/include/asm/tlb.h:11:11: error:\n invalid type argument of \u0027-\u003e\u0027 (have \u0027struct mmu_gather\u0027)\n \n Link: https://lore.kernel.org/r/20190901193601.GB5208@mellanox.com\n Fixes: 923bfc561e75 (\"pagewalk: separate function pointers from iterator data\")\n Reported-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 72e538f6ee883dde0925c89d288d16070f3286e4\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Jul 5 10:36:12 2019 +0100\n\n power: supply: isp1704: remove redundant assignment to variable ret\n \n The variable ret is being assigned with a value that is never\n read and it is being updated later with a new value. The\n assignment is redundant and can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit 5c35ba9b4743ac82613e8ff306287c8837f21f24\nAuthor: Angus Ainslie (Purism) \u003cangus@akkea.ca\u003e\nDate: Fri Jul 5 05:37:51 2019 -0600\n\n power: supply: bq25890_charger: Add the BQ25895 part\n \n The BQ25895 is almost identical to the BQ25890.\n \n Signed-off-by: Angus Ainslie (Purism) \u003cangus@akkea.ca\u003e\n Reviewed-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit 204205836f4d9a44c3e8825a9c14704fc5b2cdfd\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Mon Jul 8 20:32:59 2019 +0800\n\n power: supply: sc27xx: Replace devm_add_action() followed by failure action with devm_add_action_or_reset()\n \n devm_add_action_or_reset() is introduced as a helper function which\n internally calls devm_add_action(). If devm_add_action() fails\n then it will execute the action mentioned and return the error code.\n This reduce source code size (avoid writing the action twice)\n and reduce the likelyhood of bugs.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Reviewed-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit 08614b407d8dffd9136c79ce6fd579be42173cc3\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Mon Jul 8 20:32:51 2019 +0800\n\n power: supply: sc27xx: Introduce local variable \u0027struct device *dev\u0027\n \n Introduce local variable \u0027struct device *dev\u0027 and use it instead of\n dereferencing it repeatly.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Reviewed-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit 7d04b0b13b1175ce0c4bdc77f1278c1f120f874f\nAuthor: Matteo Croce \u003cmcroce@redhat.com\u003e\nDate: Mon Sep 2 12:21:37 2019 +0200\n\n mvpp2: percpu buffers\n \n Every mvpp2 unit can use up to 8 buffers mapped by the BM (the HW buffer\n manager). The HW will place the frames in the buffer pool depending on the\n frame size: short (\u003c 128 bytes), long (\u003c 1664) or jumbo (up to 9856).\n \n As any unit can have up to 4 ports, the driver allocates only 2 pools,\n one for small and one long frames, and share them between ports.\n When the first port MTU is set higher than 1664 bytes, a third pool is\n allocated for jumbo frames.\n \n This shared allocation makes impossible to use percpu allocators,\n and creates contention between HW queues.\n \n If possible, i.e. if the number of possible CPU are less than 8 and jumbo\n frames are not used, switch to a new scheme: allocate 8 per-cpu pools for\n short and long frames and bind every pool to an RXQ.\n \n When the first port MTU is set higher than 1664 bytes, the allocation\n scheme is reverted to the old behaviour (3 shared pools), and when all\n ports MTU are lowered, the per-cpu buffers are allocated again.\n \n Signed-off-by: Matteo Croce \u003cmcroce@redhat.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 136163618e3a9db403ef832ac32ab0b173ef503e\nAuthor: Matteo Croce \u003cmcroce@redhat.com\u003e\nDate: Mon Sep 2 12:21:36 2019 +0200\n\n mvpp2: refactor BM pool functions\n \n Refactor mvpp2_bm_pool_create(), mvpp2_bm_pool_destroy() and\n mvpp2_bm_pools_init() so that they accept a struct device instead\n of a struct platform_device, as they just need platform_device-\u003edev.\n \n Removing such dependency makes the BM code more reusable in context\n where we don\u0027t have a pointer to the platform_device.\n \n Signed-off-by: Matteo Croce \u003cmcroce@redhat.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 59857e9eadceb4002b2bd601d526091e4524d610\nAuthor: Matwey V. Kornilov \u003cmatwey@sai.msu.ru\u003e\nDate: Sat Jul 13 18:42:48 2019 +0300\n\n power: reset: reboot-mode: Fix author email format\n \n Closing angle bracket was missing.\n \n Signed-off-by: Matwey V. Kornilov \u003cmatwey@sai.msu.ru\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit 9eab9a5b4c53df850c5a7f4f48b783fac3bbffe8\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Jul 17 22:18:48 2019 +0800\n\n power: supply: ab8500: remove set but not used variables \u0027vbup33_vrtcn\u0027 and \u0027bup_vch_range\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warnings:\n \n drivers/power/supply/ab8500_charger.c:\n In function ab8500_charger_init_hw_registers:\n drivers/power/supply/ab8500_charger.c:3013:24: warning:\n variable vbup33_vrtcn set but not used [-Wunused-but-set-variable]\n drivers/power/supply/ab8500_charger.c:3013:5: warning:\n variable bup_vch_range set but not used [-Wunused-but-set-variable]\n \n Fixes: 4c4268dc97c4 (\"power: supply: ab8500: Drop AB8540/9540 support\")\n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit 40badfa396093fe85d83a5a57711b4830e85a9e0\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Sun Jul 21 18:03:28 2019 +0200\n\n power: supply: max17042_battery: Fix a typo in function names\n \n It is likely that \u0027max10742_[un]lock_model()\u0027 functions should be\n \u0027max17042_[un]lock_model()\u0027\n (0 and 7 switched in 10742)\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit 4ba0ebbc6cdecb9fad7c551a3d97b172ebc7b2fa\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Sat Aug 31 15:46:19 2019 +0300\n\n net: dsa: Fix off-by-one number of calls to devlink_port_unregister\n \n When a function such as dsa_slave_create fails, currently the following\n stack trace can be seen:\n \n [ 2.038342] sja1105 spi0.1: Probed switch chip: SJA1105T\n [ 2.054556] sja1105 spi0.1: Reset switch and programmed static config\n [ 2.063837] sja1105 spi0.1: Enabled switch tagging\n [ 2.068706] fsl-gianfar soc:ethernet@2d90000 eth2: error -19 setting up slave phy\n [ 2.076371] ------------[ cut here ]------------\n [ 2.080973] WARNING: CPU: 1 PID: 21 at net/core/devlink.c:6184 devlink_free+0x1b4/0x1c0\n [ 2.088954] Modules linked in:\n [ 2.092005] CPU: 1 PID: 21 Comm: kworker/1:1 Not tainted 5.3.0-rc6-01360-g41b52e38d2b6-dirty #1746\n [ 2.100912] Hardware name: Freescale LS1021A\n [ 2.105162] Workqueue: events deferred_probe_work_func\n [ 2.110287] [\u003cc03133a4\u003e] (unwind_backtrace) from [\u003cc030d8cc\u003e] (show_stack+0x10/0x14)\n [ 2.117992] [\u003cc030d8cc\u003e] (show_stack) from [\u003cc10b08d8\u003e] (dump_stack+0xb4/0xc8)\n [ 2.125180] [\u003cc10b08d8\u003e] (dump_stack) from [\u003cc0349d04\u003e] (__warn+0xe0/0xf8)\n [ 2.132018] [\u003cc0349d04\u003e] (__warn) from [\u003cc0349e34\u003e] (warn_slowpath_null+0x40/0x48)\n [ 2.139549] [\u003cc0349e34\u003e] (warn_slowpath_null) from [\u003cc0f19d74\u003e] (devlink_free+0x1b4/0x1c0)\n [ 2.147772] [\u003cc0f19d74\u003e] (devlink_free) from [\u003cc1064fc0\u003e] (dsa_switch_teardown+0x60/0x6c)\n [ 2.155907] [\u003cc1064fc0\u003e] (dsa_switch_teardown) from [\u003cc1065950\u003e] (dsa_register_switch+0x8e4/0xaa8)\n [ 2.164821] [\u003cc1065950\u003e] (dsa_register_switch) from [\u003cc0ba7fe4\u003e] (sja1105_probe+0x21c/0x2ec)\n [ 2.173216] [\u003cc0ba7fe4\u003e] (sja1105_probe) from [\u003cc0b35948\u003e] (spi_drv_probe+0x80/0xa4)\n [ 2.180920] [\u003cc0b35948\u003e] (spi_drv_probe) from [\u003cc0a4c1cc\u003e] (really_probe+0x108/0x400)\n [ 2.188711] [\u003cc0a4c1cc\u003e] (really_probe) from [\u003cc0a4c694\u003e] (driver_probe_device+0x78/0x1bc)\n [ 2.196933] [\u003cc0a4c694\u003e] (driver_probe_device) from [\u003cc0a4a3dc\u003e] (bus_for_each_drv+0x58/0xb8)\n [ 2.205414] [\u003cc0a4a3dc\u003e] (bus_for_each_drv) from [\u003cc0a4c024\u003e] (__device_attach+0xd0/0x168)\n [ 2.213637] [\u003cc0a4c024\u003e] (__device_attach) from [\u003cc0a4b1d0\u003e] (bus_probe_device+0x84/0x8c)\n [ 2.221772] [\u003cc0a4b1d0\u003e] (bus_probe_device) from [\u003cc0a4b72c\u003e] (deferred_probe_work_func+0x84/0xc4)\n [ 2.230686] [\u003cc0a4b72c\u003e] (deferred_probe_work_func) from [\u003cc03650a4\u003e] (process_one_work+0x218/0x510)\n [ 2.239772] [\u003cc03650a4\u003e] (process_one_work) from [\u003cc03660d8\u003e] (worker_thread+0x2a8/0x5c0)\n [ 2.247908] [\u003cc03660d8\u003e] (worker_thread) from [\u003cc036b348\u003e] (kthread+0x148/0x150)\n [ 2.255265] [\u003cc036b348\u003e] (kthread) from [\u003cc03010e8\u003e] (ret_from_fork+0x14/0x2c)\n [ 2.262444] Exception stack(0xea965fb0 to 0xea965ff8)\n [ 2.267466] 5fa0: 00000000 00000000 00000000 00000000\n [ 2.275598] 5fc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000\n [ 2.283729] 5fe0: 00000000 00000000 00000000 00000000 00000013 00000000\n [ 2.290333] ---[ end trace ca5d506728a0581a ]---\n \n devlink_free is complaining right here:\n \n WARN_ON(!list_empty(\u0026devlink-\u003eport_list));\n \n This happens because devlink_port_unregister is no longer done right\n away in dsa_port_setup when a DSA_PORT_TYPE_USER has failed.\n Vivien said about this change that:\n \n Also no need to call devlink_port_unregister from within dsa_port_setup\n as this step is inconditionally handled by dsa_port_teardown on error.\n \n which is not really true. The devlink_port_unregister function _is_\n being called unconditionally from within dsa_port_setup, but not for\n this port that just failed, just for the previous ones which were set\n up.\n \n ports_teardown:\n for (i \u003d 0; i \u003c port; i++)\n dsa_port_teardown(\u0026ds-\u003eports[i]);\n \n Initially I was tempted to fix this by extending the \"for\" loop to also\n cover the port that failed during setup. But this could have potentially\n unforeseen consequences unrelated to devlink_port or even other types of\n ports than user ports, which I can\u0027t really test for. For example, if\n for some reason devlink_port_register itself would fail, then\n unconditionally unregistering it in dsa_port_teardown would not be a\n smart idea. The list might go on.\n \n So just make dsa_port_setup undo the setup it had done upon failure, and\n let the for loop undo the work of setting up the previous ports, which\n are guaranteed to be brought up to a consistent state.\n \n Fixes: 955222ca5281 (\"net: dsa: use a single switch statement for port setup\")\n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Reviewed-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit e33b4325e60e146c2317a8b548cbd633239ff83b\nAuthor: Yizhuo \u003cyzhai003@ucr.edu\u003e\nDate: Fri Aug 30 19:00:48 2019 -0700\n\n net: stmmac: dwmac-sun8i: Variable \"val\" in function sun8i_dwmac_set_syscon() could be uninitialized\n \n In function sun8i_dwmac_set_syscon(), local variable \"val\" could\n be uninitialized if function regmap_field_read() returns -EINVAL.\n However, it will be used directly in the if statement, which\n is potentially unsafe.\n \n Signed-off-by: Yizhuo \u003cyzhai003@ucr.edu\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a21cf11bc57fca136f98e81ebc6a6eaf52952b8f\nAuthor: Jiri Pirko \u003cjiri@mellanox.com\u003e\nDate: Fri Aug 30 10:25:30 2019 +0200\n\n mlx5: Add missing init_net check in FIB notifier\n \n Take only FIB events that are happening in init_net into account. No other\n namespaces are supported.\n \n Signed-off-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Acked-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 0e063317d579d0a7fca957461332b6832da69850\nAuthor: Michal Simek \u003cmichal.simek@xilinx.com\u003e\nDate: Mon Sep 2 15:00:40 2019 +0200\n\n power: reset: gpio-restart: Fix typo when gpio reset is not found\n \n Trivial patch which just corrects error message.\n \n Fixes: 371bb20d6927 (\"power: Add simple gpio-restart driver\")\n Signed-off-by: Michal Simek \u003cmichal.simek@xilinx.com\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit 544b3e7fd2d39b571298c64000423df21b912a3b\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Mon Sep 2 19:45:06 2019 +0200\n\n staging: exfat: make exfat depend on BLOCK\n \n This should fix a build error in some configurations when CONFIG_BLOCK\n is not selected. Also properly set the dependancy for no FAT support at\n the same time.\n \n Reported-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\n Cc: Valdis Kletnieks \u003cvaldis.kletnieks@vt.edu\u003e\n Link: https://lore.kernel.org/r/20190902174631.GB31445@kroah.com\n Acked-by: Randy Dunlap \u003crdunlap@infradead.org\u003e # build-tested\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 0cd80c707791f26f413c8464fe66f2606c430352\nAuthor: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\nDate: Thu Aug 29 14:41:04 2019 +0100\n\n virtio-fs: add Documentation/filesystems/virtiofs.rst\n \n Add information about the new \"virtiofs\" file system.\n \n Signed-off-by: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\n Signed-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n\ncommit 26125580e9223c9cf8d1ad6523abe3ffd0d7a024\nAuthor: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\nDate: Tue Jun 12 09:41:17 2018 +0100\n\n virtio-fs: add virtiofs filesystem\n \n Add a basic file system module for virtio-fs. This does not yet contain\n shared data support between host and guest or metadata coherency speedups.\n However it is already significantly faster than virtio-9p.\n \n Design Overview\n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n \n With the goal of designing something with better performance and local file\n system semantics, a bunch of ideas were proposed.\n \n - Use fuse protocol (instead of 9p) for communication between guest and\n host. Guest kernel will be fuse client and a fuse server will run on\n host to serve the requests.\n \n - For data access inside guest, mmap portion of file in QEMU address space\n and guest accesses this memory using dax. That way guest page cache is\n bypassed and there is only one copy of data (on host). This will also\n enable mmap(MAP_SHARED) between guests.\n \n - For metadata coherency, there is a shared memory region which contains\n version number associated with metadata and any guest changing metadata\n updates version number and other guests refresh metadata on next access.\n This is yet to be implemented.\n \n How virtio-fs differs from existing approaches\n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n \n The unique idea behind virtio-fs is to take advantage of the co-location of\n the virtual machine and hypervisor to avoid communication (vmexits).\n \n DAX allows file contents to be accessed without communication with the\n hypervisor. The shared memory region for metadata avoids communication in\n the common case where metadata is unchanged.\n \n By replacing expensive communication with cheaper shared memory accesses,\n we expect to achieve better performance than approaches based on network\n file system protocols. In addition, this also makes it easier to achieve\n local file system semantics (coherency).\n \n These techniques are not applicable to network file system protocols since\n the communications channel is bypassed by taking advantage of shared memory\n on a local machine. This is why we decided to build virtio-fs rather than\n focus on 9P or NFS.\n \n Caching Modes\n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n \n Like virtio-9p, different caching modes are supported which determine the\n coherency level as well. The “cache\u003dFOO” and “writeback” options control\n the level of coherence between the guest and host filesystems.\n \n - cache\u003dnone\n metadata, data and pathname lookup are not cached in guest. They are\n always fetched from host and any changes are immediately pushed to host.\n \n - cache\u003dalways\n metadata, data and pathname lookup are cached in guest and never expire.\n \n - cache\u003dauto\n metadata and pathname lookup cache expires after a configured amount of\n time (default is 1 second). Data is cached while the file is open\n (close to open consistency).\n \n - writeback/no_writeback\n These options control the writeback strategy. If writeback is disabled,\n then normal writes will immediately be synchronized with the host fs.\n If writeback is enabled, then writes may be cached in the guest until\n the file is closed or an fsync(2) performed. This option has no effect\n on mmap-ed writes or writes going through the DAX mechanism.\n \n Signed-off-by: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\n Signed-off-by: Vivek Goyal \u003cvgoyal@redhat.com\u003e\n Signed-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n\ncommit eb3d8f42231aec65b64b079dd17bd6c008a3fe29\nAuthor: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\nDate: Wed Aug 28 11:26:13 2019 -0400\n\n NFS: Fix inode fileid checks in attribute revalidation code\n \n We want to throw out the attrbute if it refers to the mounted on fileid,\n and not the real fileid. However we do not want to block cache consistency\n updates from NFSv4 writes.\n \n Reported-by: Murphy Zhou \u003cjencce.kernel@gmail.com\u003e\n Fixes: 7e10cc25bfa0 (\"NFS: Don\u0027t refresh attributes with mounted-on-file...\")\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n\ncommit 9c551fac9ea9629cdc6deb205292c2bbe7b38917\nAuthor: amy.shih \u003camy.shih@advantech.com.tw\u003e\nDate: Mon Jun 18 15:57:19 2085 +0000\n\n hwmon: (nct7904) Fix incorrect temperature limitation register setting of LTD.\n \n According to kernel hwmon sysfs-interface documentation, temperature\n critical max value, typically greater than corresponding temp_max values.\n Thus, reads the LTD_HV_HL (LTD HIGH VALUE HIGH LIMITATION) and LTD_LV_HL\n (LTD LOW VALUE HIGH LIMITATION) for case hwmon_temp_crit and\n hwmon_temp_crit_hyst. Reads the LTD_HV_LL (HIGH VALUE LOW LIMITATION)\n and LTD_LV_LL (LOW VALUE LOW LIMITATION) for case hwmon_temp_max\n and hwmon_temp_max_hyst.\n \n Signed-off-by: amy.shih \u003camy.shih@advantech.com.tw\u003e\n Link: https://lore.kernel.org/r/20850618155720.24857-1-Amy.Shih@advantech.com.tw\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 4b6253fa736494346bab381114611e358afae545\nAuthor: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nDate: Fri Aug 9 12:50:15 2019 -0700\n\n Input: bu21013_ts - switch to using standard touchscreen properties\n \n This switches the driver over to the standard touchscreen properties for\n coordinate transformation, while keeping old bindings working as well.\n \n Tested-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 307ec663f6c3e9a0a80c7092a1dea6e09debc127\nAuthor: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nDate: Fri Aug 9 12:08:14 2019 -0700\n\n Input: bu21013_ts - switch to using MT-B (slotted) protocol\n \n MT-B protocol is more efficient and everyone expects it. We use in-kernel\n tracking to identify contacts.\n \n Tested-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 18aa4d6b67483b4dac08a537f6c4fc943eea5386\nAuthor: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nDate: Thu Aug 8 17:41:12 2019 -0700\n\n Input: bu21013_ts - fix suspend when wake source\n \n If the touchscreen is configured as wakeup source we should not be cutting\n off power to it.\n \n Also, now that the driver relies on I2C client to supply IRQ, we do not\n need to explicitly enable and disable IRQ for wakeup: if device is created\n as wakeup source, I2C core will mark interrupt as wakeup one.\n \n Tested-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit db3e34d4036db59cbacc1bff1c622b23d85b766b\nAuthor: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nDate: Fri Aug 9 10:07:06 2019 -0700\n\n Input: bu21013_ts - use interrupt from I2C client\n \n Instead of trying to map INT GPIO to interrupt, let\u0027s use one supplied by\n I2C client. If there is none - bail. This will also allow us to treat INT\n GPIO as optional, as per the binding.\n \n Tested-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 1eb7b4cacc01771ae42fcbc5ae9a4bc1d13c1dbc\nAuthor: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nDate: Thu Aug 8 12:09:43 2019 -0700\n\n Input: bu21013_ts - remove support for platform data\n \n There are no current users of the platform data in the tree, and\n any new users should either use device tree, or static device\n properties to describe the device.\n \n This change drop the platform data definition and handling and moves the\n driver over to generic device properties API. We also drop support for the\n external clock. If it is needed we will have to extend the bindings to\n supply the clock reference and handle it properly in the driver.\n \n Also, wakeup setting should be coming from I2C client.\n \n Tested-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit caeb1aae283f9b93d92400b1cb0395bdfd0d586c\nAuthor: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nDate: Thu Aug 8 14:28:48 2019 -0700\n\n Input: bu21013_ts - convert to using managed resources\n \n This allows trimming error unwinding and device removal handling.\n \n Tested-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit a5ee351d5bc7c9e0abdbc3d345a503cc07d0623f\nAuthor: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nDate: Thu Aug 8 16:02:43 2019 -0700\n\n Input: bu21013_ts - remove useless comments\n \n The comments for individual functions in the driver do not provide any\n additional information beyond what function names indicate.\n \n Tested-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 0570bab054106f68533dcb0de2e64c48b15d7b9f\nAuthor: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nDate: Thu Aug 8 15:53:43 2019 -0700\n\n Input: bu21013_ts - annotate supend/resume methods as __maybe_unused\n \n Instead if #ifdef-ing out suspend and resume methods, let\u0027s mark\n them as __maybe_unused to get better compile time coverage.\n \n Tested-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit a866781e84dd022dde5d307093555e65c31277de\nAuthor: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nDate: Thu Aug 8 14:19:36 2019 -0700\n\n Input: bu21013_ts - rename some variables\n \n \"bu21013_data\" and \"struct bu21013_ts_data\" are a tad long, let\u0027s call them\n \"ts\" and \"struct bu21013_ts\".\n \n Also rename retval to error in bu21013_init_chip() and adjust formatting;\n i2c_smbus_write_byte_data() returns negative on error and 0 on success, so\n we simply test if whether erro is 0 or not.\n \n Tested-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit a1b70a44b80af641a441937803cb8251e8e6d8e3\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Thu Aug 8 11:03:59 2019 -0700\n\n Input: bu21013_ts - convert to use GPIO descriptors\n \n This driver can use GPIO descriptors rather than GPIO numbers\n without any problems, convert it. Name the field variables after\n the actual pins on the chip rather than the \"reset\" and \"touch\"\n names from the devicetree bindings that are vaguely inaccurate.\n \n No in-tree users pass GPIO numbers in platform data so drop\n this. Descriptor tables can be used to get these GPIOs from a board\n file if need be.\n \n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 5dc15dcf39c40fcc420e41b43e31ab260c3feec3\nAuthor: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nDate: Fri Aug 9 09:48:51 2019 -0700\n\n ARM: ux500: improve BU21013 touchpad bindings\n \n In preparation to update to bu21013_tp driver properly annotate GPIOs\n property (the INT GPIOs are active low, not open drain), and also define\n interrupt lines so we do not have to have special conversion in the driver.\n \n Tested-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 27c8bac145ef7f2662d8696e49d627b6b4ca256a\nAuthor: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\nDate: Tue Aug 27 03:44:57 2019 +0000\n\n hwmon: (as370-hwmon) Add DT bindings for Synaptics AS370 PVT\n \n Add device tree bindings for Synaptics AS370 PVT sensors.\n \n Signed-off-by: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190827113337.384457f6@xhacker.debian\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 658e687b4218b42d03a5b0b79a292c9c48f1ddaf\nAuthor: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\nDate: Tue Aug 27 03:44:15 2019 +0000\n\n hwmon: Add Synaptics AS370 PVT sensor driver\n \n Add a new driver for Synaptics AS370 PVT sensors. Currently, only\n temperature is supported.\n \n Signed-off-by: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Link: https://lore.kernel.org/r/20190827113259.4fb64a17@xhacker.debian\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 0f459556cd18db192a2a5dd6e5fd9d545f058bb6\nAuthor: Eddie James \u003ceajames@linux.ibm.com\u003e\nDate: Fri Aug 30 14:11:03 2019 -0500\n\n pmbus: (ibm-cffps) Add support for version 2 of the PSU\n \n Version 2 of the PSU supports a second page of data and changes the\n format of the FW version. Use the devicetree binding to differentiate\n between the version the driver should use.\n \n Signed-off-by: Eddie James \u003ceajames@linux.ibm.com\u003e\n Link: https://lore.kernel.org/r/1567192263-15065-4-git-send-email-eajames@linux.ibm.com\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 2e091fac4871d9e6a22895cef507850b90749257\nAuthor: Eddie James \u003ceajames@linux.ibm.com\u003e\nDate: Fri Aug 30 14:11:01 2019 -0500\n\n dt-bindings: hwmon: Document ibm,cffps2 compatible string\n \n Document the compatible string for version 2 of the IBM CFFPS PSU.\n \n Signed-off-by: Eddie James \u003ceajames@linux.ibm.com\u003e\n Link: https://lore.kernel.org/r/1567192263-15065-2-git-send-email-eajames@linux.ibm.com\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 9e362992ff34cab695e669ef1e3f4f65463087e5\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Aug 28 21:34:24 2019 +0300\n\n drm/i915: Clean up HDMI deep color handling a bit\n \n Reogranize the HDMI deep color state computation to just\n loop over possible bpc values. Avoids having to maintain\n so many variants of the clock etc.\n \n The current code also looks confused w.r.t. port_clock vs.\n bw_constrained. It would happily update port_clock for\n deep color but then not actually enable deep color due to\n bw_constrained being set. The new logic handles that case\n correctly.\n \n v2: Pull stuff into separate funcs (Jani)\n \n Reviewed-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190828183424.7856-1-ville.syrjala@linux.intel.com\n\ncommit 66a990dd0c49b534b2396934c1659b3fef34233d\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Fri Aug 30 21:27:19 2019 +0300\n\n drm/i915: Prefer encoder-\u003ename over port_name()\n \n enum port is a mess now because it no longer matches the spec\n at all. Let\u0027s start to dig ourselves out of this hole by\n reducing our reliance on port_name(). This should at least make\n a bunch of debug messages a bit more sensible while we think how\n to fill the the hole properly.\n \n Based on the following cocci script with a lot of manual cleanup\n (all the format strings etc.):\n @@\n expression E;\n @@\n (\n - port_name(E-\u003eport)\n + E-\u003ebase.base.id, E-\u003ebase.name\n |\n - port_name(E.port)\n + E.base.base.id, E.base.name\n )\n \n @@\n enum port P;\n expression E;\n @@\n P \u003d E-\u003eport\n \u003c...\n - port_name(P)\n + E-\u003ebase.base.id, E-\u003ebase.name\n ...\u003e\n \n @@\n enum port P;\n expression E;\n @@\n P \u003d E.port\n \u003c...\n - port_name(P)\n + E.base.base.id, E.base.name\n ...\u003e\n \n @@\n expression E;\n @@\n {\n - enum port P \u003d E;\n ... when !\u003d P\n }\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190830182719.32608-1-ville.syrjala@linux.intel.com\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n\ncommit 75427b2a2bffc083d51dec389c235722a9c69b05\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Aug 28 13:20:59 2019 +0300\n\n drm/i915: Limit MST to \u003c\u003d 8bpc once again\n \n My attempt at allowing MST to use the higher color depths has\n regressed some configurations. Apparently people have setups\n where all MST streams will fit into the DP link with 8bpc but\n won\u0027t fit with higher color depths.\n \n What we really should be doing is reducing the bpc for all the\n streams on the same link until they start to fit. But that requires\n a bit more work, so in the meantime let\u0027s revert back closer to\n the old behavior and limit MST to at most 8bpc.\n \n Cc: stable@vger.kernel.org\n Cc: Lyude Paul \u003clyude@redhat.com\u003e\n Tested-by: Geoffrey Bennett \u003cgmux22@gmail.com\u003e\n Fixes: f1477219869c (\"drm/i915: Remove the 8bpc shackles from DP MST\")\n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111505\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190828102059.2512-1-ville.syrjala@linux.intel.com\n Reviewed-by: Lyude Paul \u003clyude@redhat.com\u003e\n\ncommit 86759eeb32f70193ca7ebd50cc9b70ccb73d3d89\nAuthor: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\nDate: Fri Aug 30 15:04:16 2019 +0200\n\n ARM: multi_v7_defconfig: Make MAX77802 regulator driver built-in\n \n Maxim 77802 PMIC is a main PMIC for the following Exynos5 based boards:\n Odroid XU, Chromebook Pit and Chromebook Pi. Driver for its voltage\n regulator is needed very early during boot to properly instantiate SD/MMC\n devices and mount rootfs, so change that driver to be compiled-in.\n \n Signed-off-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 60cb036dbda7b37d52e1803ac3c5127e3d741313\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Mar 25 16:38:32 2019 +0000\n\n vfs: Convert ceph to use the new mount API\n \n Convert the ceph filesystem to the new internal mount API as the old\n one will be obsoleted and removed. This allows greater flexibility in\n communication of mount parameters between userspace, the VFS and the\n filesystem.\n \n See Documentation/filesystems/mount_api.txt for more information.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Reviewed-by: \"Yan, Zheng\" \u003czyan@redhat.com\u003e\n cc: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n cc: Sage Weil \u003csage@redhat.com\u003e\n cc: ceph-devel@vger.kernel.org\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 348140124f0949d773e13ee9a191243f4e9abf00\nAuthor: Andrew Price \u003canprice@redhat.com\u003e\nDate: Wed Mar 27 14:46:00 2019 +0000\n\n gfs2: Convert gfs2 to fs_context\n \n Convert gfs2 and gfs2meta to fs_context. Removes the duplicated vfs code\n from gfs2_mount and instead uses the new vfs_get_block_super() before\n switching the -\u003eroot to the appropriate dentry.\n \n The mount option parsing has been converted to the new API and error\n reporting for invalid options has been made more precise at the same\n time.\n \n All of the mount/remount code has been moved into ops_fstype.c\n \n Signed-off-by: Andrew Price \u003canprice@redhat.com\u003e\n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n cc: cluster-devel@redhat.com\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 529803dce39b44d3772bb2049f06edcde55376aa\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Thu Mar 21 11:09:39 2019 +0000\n\n vfs: Kill mount_single()\n \n Kill mount_single() as nothing now uses it. It has been replaced by\n vfs_get_super() keyed with vfs_get_single_super or\n vfs_get_single_reconf_super.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 514b423fe4e80aaddd58d21e8652122e8db191d5\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Thu Mar 21 10:34:14 2019 +0000\n\n vfs: Convert spufs to use the new mount API\n \n Convert the spufs filesystem to the new internal mount API as the old\n one will be obsoleted and removed. This allows greater flexibility in\n communication of mount parameters between userspace, the VFS and the\n filesystem.\n \n See Documentation/filesystems/mount_api.txt for more information.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n cc: Jeremy Kerr \u003cjk@ozlabs.org\u003e\n cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n cc: linuxppc-dev@lists.ozlabs.org\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 75231fb6e1965a8a148a297e882afd71d2420606\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Thu Mar 21 09:35:44 2019 +0000\n\n vfs: Convert hypfs to use the new mount API\n \n Convert the hypfs filesystem to the new internal mount API as the old\n one will be obsoleted and removed. This allows greater flexibility in\n communication of mount parameters between userspace, the VFS and the\n filesystem.\n \n See Documentation/filesystems/mount_api.txt for more information.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n cc: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\n cc: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\n cc: linux-s390@vger.kernel.org\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 0a421fe37c0a4b35d62cffb0b29ff5ec1f206079\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Thu Mar 21 10:08:08 2019 +0000\n\n hypfs: Fix error number left in struct pointer member\n \n In hypfs_fill_super(), if hypfs_create_update_file() fails,\n sbi-\u003eupdate_file is left holding an error number. This is passed to\n hypfs_kill_super() which doesn\u0027t check for this.\n \n Fix this by not setting sbi-\u003eupdate_value until after we\u0027ve checked for\n error.\n \n Fixes: 24bbb1faf3f0 (\"[PATCH] s390_hypfs filesystem\")\n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n cc: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\n cc: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\n cc: linux-s390@vger.kernel.org\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit fab3d3fd23ce74e0eb6271d375348db25f7c276d\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Thu Mar 21 08:42:49 2019 +0000\n\n vfs: Convert pstore to use the new mount API\n \n Convert the pstore filesystem to the new internal mount API as the old\n one will be obsoleted and removed. This allows greater flexibility in\n communication of mount parameters between userspace, the VFS and the\n filesystem.\n \n See Documentation/filesystems/mount_api.txt for more information.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Acked-by: Kees Cook \u003ckeescook@chromium.org\u003e\n cc: Anton Vorontsov \u003canton@enomsg.org\u003e\n cc: Colin Cross \u003cccross@android.com\u003e\n cc: Tony Luck \u003ctony.luck@intel.com\u003e\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 97dae379269bb26bc28833578b6f979a7aa4b91a\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Thu Mar 21 08:27:55 2019 +0000\n\n vfs: Convert tracefs to use the new mount API\n \n Convert the tracefs filesystem to the new internal mount API as the old\n one will be obsoleted and removed. This allows greater flexibility in\n communication of mount parameters between userspace, the VFS and the\n filesystem.\n \n See Documentation/filesystems/mount_api.txt for more information.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n cc: Steven Rostedt \u003crostedt@goodmis.org\u003e\n cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 2fa9d3c7b604e580aab3ecafd9241fc7c66aa758\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Jan 21 14:04:22 2019 +0000\n\n vfs: Convert debugfs to use the new mount API\n \n Convert the debugfs filesystem to the new internal mount API as the old\n one will be obsoleted and removed. This allows greater flexibility in\n communication of mount parameters between userspace, the VFS and the\n filesystem.\n \n See Documentation/filesystems/mount_api.txt for more information.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n cc: \"Rafael J. Wysocki\" \u003crafael@kernel.org\u003e\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit d3c317e89f57ca9a719bc7a60e5cfbb08b6120bd\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Thu Mar 21 09:22:36 2019 +0000\n\n vfs: Add a single-or-reconfig keying to vfs_get_super()\n \n Add an additional keying mode to vfs_get_super() to indicate that only a\n single superblock should exist in the system, and that, if it does, further\n mounts should invoke reconfiguration upon it.\n \n This allows mount_single() to be replaced.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 03a09e7dcee9f232362ce1448fedecb31c000290\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Mar 25 16:38:32 2019 +0000\n\n vfs: Convert squashfs to use the new mount API\n \n Convert the squashfs filesystem to the new internal mount API as the old\n one will be obsoleted and removed. This allows greater flexibility in\n communication of mount parameters between userspace, the VFS and the\n filesystem.\n \n See Documentation/filesystems/mount_api.txt for more information.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n cc: Phillip Lougher \u003cphillip@squashfs.org.uk\u003e\n cc: squashfs-devel@lists.sourceforge.net\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit f5299d87ac2afac82ee0763c8997b6291a70d35e\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Mar 25 16:38:32 2019 +0000\n\n mtd: Kill mount_mtd()\n \n Kill mount_mtd() as it has now been replaced by vfs_get_mtd_super() in the\n new mount API and nothing now uses it.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n cc: David Woodhouse \u003cdwmw2@infradead.org\u003e\n cc: Brian Norris \u003ccomputersforpeace@gmail.com\u003e\n cc: Boris Brezillon \u003cbbrezillon@kernel.org\u003e\n cc: Marek Vasut \u003cmarek.vasut@gmail.com\u003e\n cc: Richard Weinberger \u003crichard@nod.at\u003e\n cc: linux-mtd@lists.infradead.org\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit d1f8b99efc6cd71125f716f10771169c0abb891b\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Mar 25 16:38:32 2019 +0000\n\n vfs: Convert jffs2 to use the new mount API\n \n Convert the jffs2 filesystem to the new internal mount API as the old\n one will be obsoleted and removed. This allows greater flexibility in\n communication of mount parameters between userspace, the VFS and the\n filesystem.\n \n See Documentation/filesystems/mount_api.txt for more information.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n cc: David Woodhouse \u003cdwmw2@infradead.org\u003e\n cc: linux-mtd@lists.infradead.org\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit d74503eed1b5d422979eff2aaa6270853d4cc666\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Mar 25 16:38:32 2019 +0000\n\n vfs: Convert cramfs to use the new mount API\n \n Convert the cramfs filesystem to the new internal mount API as the old\n one will be obsoleted and removed. This allows greater flexibility in\n communication of mount parameters between userspace, the VFS and the\n filesystem.\n \n See Documentation/filesystems/mount_api.txt for more information.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Tested-by: Nicolas Pitre \u003cnico@fluxnic.net\u003e\n Acked-by: Nicolas Pitre \u003cnico@fluxnic.net\u003e\n cc: linux-mtd@lists.infradead.org\n cc: linux-block@vger.kernel.org\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit f6dc88cfeb11499f2259b5477bb4e965e1525c9b\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Mar 25 16:38:31 2019 +0000\n\n vfs: Convert romfs to use the new mount API\n \n Convert the romfs filesystem to the new internal mount API as the old\n one will be obsoleted and removed. This allows greater flexibility in\n communication of mount parameters between userspace, the VFS and the\n filesystem.\n \n See Documentation/filesystems/mount_api.txt for more information.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n cc: linux-mtd@lists.infradead.org\n cc: linux-block@vger.kernel.org\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 51eab38291868a7360a67a9ed95588e4a0c8e8a1\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Mar 25 16:38:31 2019 +0000\n\n mtd: Provide fs_context-aware mount_mtd() replacement\n \n Provide a function, vfs_get_mtd_super(), to replace mount_mtd(), using an\n fs_context struct to hold the parameters. The mtd device pointer is cached\n in the struct, along with a destructor pointer as the mtd core may be in a\n module and, as such, unreachable by the core code that would release the\n fs_context.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n cc: David Woodhouse \u003cdwmw2@infradead.org\u003e\n cc: Brian Norris \u003ccomputersforpeace@gmail.com\u003e\n cc: Boris Brezillon \u003cbbrezillon@kernel.org\u003e\n cc: Marek Vasut \u003cmarek.vasut@gmail.com\u003e\n cc: Richard Weinberger \u003crichard@nod.at\u003e\n cc: linux-mtd@lists.infradead.org\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 8965e019cc96e89a0080b170885bba469a6761e9\nAuthor: Guillaume Gardet \u003cguillaume.gardet@arm.com\u003e\nDate: Fri Aug 30 12:45:01 2019 +0200\n\n ARM: dts: exynos: Enable GPU/Mali T604 on Arndale board\n \n Tested with kmscube and some glmark2* tests on arndale board.\n \n Signed-off-by: Guillaume Gardet \u003cguillaume.gardet@arm.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 124e0f4a39f7218f858586e06d9ece3dff757662\nAuthor: Guillaume Gardet \u003cguillaume.gardet@arm.com\u003e\nDate: Fri Aug 30 12:45:00 2019 +0200\n\n ARM: dts: exynos: Enable GPU/Mali T604 on Chromebook Snow\n \n Tested with kmscube and some glmark2* tests on Chromebook snow.\n Frequency adapts with load.\n \n Signed-off-by: Guillaume Gardet \u003cguillaume.gardet@arm.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 137cae0b7af72bda284a58ab69dbef0a0245579f\nAuthor: Guillaume Gardet \u003cguillaume.gardet@arm.com\u003e\nDate: Fri Aug 30 12:44:59 2019 +0200\n\n ARM: dts: exynos: Add GPU/Mali T604 node to Exynos5250\n \n Add nodes for GPU (Mali T604) to Exynos5250.\n \n Signed-off-by: Guillaume Gardet \u003cguillaume.gardet@arm.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 4c43ad9e9d41d2bf9f98ee45a2ad134d4edd9ea9\nAuthor: Guillaume Gardet \u003cguillaume.gardet@arm.com\u003e\nDate: Fri Aug 30 12:45:02 2019 +0200\n\n ARM: dts: exynos: Fix min/max buck4 for GPU on Arndale board\n \n Required to have GPU voltage scaling working properly.\n \n Signed-off-by: Guillaume Gardet \u003cguillaume.gardet@arm.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit d33cd42d86671bed870827aa399aeb9f1da74119\nAuthor: Sam Bazley \u003csambazley@fastmail.com\u003e\nDate: Sun Sep 1 03:31:30 2019 +0100\n\n ALSA: hda/realtek - Add quirk for HP Pavilion 15\n \n HP Pavilion 15 (AMD Ryzen-based model) with 103c:84e7 needs the same\n quirk like HP Envy/Spectre x360 for enabling the mute LED over Mic3 pin.\n \n [ rearranged in the SSID number order by tiwai ]\n \n Signed-off-by: Sam Bazley \u003csambazley@fastmail.com\u003e\n Cc: \u003cstable@vger.kernel.org\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 5833f5a5daf37f54ff7cc15f03ac0df201ccff92\nAuthor: Guillaume Gardet \u003cguillaume.gardet@arm.com\u003e\nDate: Fri Aug 30 12:44:58 2019 +0200\n\n dt-bindings: gpu: mali: Add Samsung exynos5250 compatible\n \n Add \"samsung,exynos5250-mali\" binding.\n \n Signed-off-by: Guillaume Gardet \u003cguillaume.gardet@arm.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 33385da8d2189e9e43f8c32bfdbf6972b4eea699\nAuthor: Sibi Sankar \u003csibis@codeaurora.org\u003e\nDate: Sun Sep 1 23:14:07 2019 +0530\n\n dt-bindings: reset: pdc: Convert PDC Global bindings to yaml\n \n Convert PDC Global bindings to yaml and add SC7180 PDC global to the list\n of possible bindings.\n \n Signed-off-by: Sibi Sankar \u003csibis@codeaurora.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit 68d5ecf727051605577b4c7a10b133687e7facf1\nAuthor: Sibi Sankar \u003csibis@codeaurora.org\u003e\nDate: Sun Sep 1 23:14:06 2019 +0530\n\n dt-bindings: reset: aoss: Convert AOSS reset bindings to yaml\n \n Convert AOSS reset bindings to yaml and add SC7180 AOSS reset to the list\n of possible bindings.\n \n Signed-off-by: Sibi Sankar \u003csibis@codeaurora.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit 5b0eeeaa37615df37a9a30929b73e9defe61ca84\nAuthor: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\nDate: Fri Aug 30 14:52:42 2019 +0200\n\n ARM: dts: exynos: Mark LDO10 as always-on on Peach Pit/Pi Chromebooks\n \n Commit aff138bf8e37 (\"ARM: dts: exynos: Add TMU nodes regulator supply\n for Peach boards\") assigned LDO10 to Exynos Thermal Measurement Unit,\n but it turned out that it supplies also some other critical parts and\n board freezes/crashes when it is turned off.\n \n The mentioned commit made Exynos TMU a consumer of that regulator and in\n typical case Exynos TMU driver keeps it enabled from early boot. However\n there are such configurations (example is multi_v7_defconfig), in which\n some of the regulators are compiled as modules and are not available\n from early boot. In such case it may happen that LDO10 is turned off by\n regulator core, because it has no consumers yet (in this case consumer\n drivers cannot get it, because the supply regulators for it are not yet\n available). This in turn causes the board to crash. This patch restores\n \u0027always-on\u0027 property for the LDO10 regulator.\n \n Fixes: aff138bf8e37 (\"ARM: dts: exynos: Add TMU nodes regulator supply for Peach boards\")\n Signed-off-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 2cf842ba2f4172a47c03c24d1c450f2c4fc95c19\nAuthor: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\nDate: Fri Aug 23 16:53:56 2019 +0200\n\n ARM: dts: exynos: Remove not accurate secondary ADC compatible\n \n The Exynos3250 ADC has its own compatible because of differences from\n other Exynos SoCs. Therefore it is not entirely compatible with\n samsung,exynos-adc-v2. Remove the samsung,exynos-adc-v2.\n \n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 6369512541e7be367cb6a8fb73336e55bea26138\nAuthor: Vivek Goyal \u003cvgoyal@redhat.com\u003e\nDate: Mon May 6 15:35:43 2019 -0400\n\n fuse: allow skipping control interface and forced unmount\n \n virtio-fs does not support aborting requests which are being\n processed. That is requests which have been sent to fuse daemon on host.\n \n Signed-off-by: Vivek Goyal \u003cvgoyal@redhat.com\u003e\n Signed-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n\ncommit aeec95daeac1f97570ac881a707938667b861af5\nAuthor: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\nDate: Thu Aug 29 11:01:20 2019 +0200\n\n fuse: dissociate DESTROY from fuseblk\n \n Allow virtio-fs to also send DESTROY request.\n \n Signed-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n\ncommit bf518e8f8083487eab75a34dc585fc1ab4565cf9\nAuthor: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\nDate: Wed Aug 15 17:42:34 2018 +0200\n\n fuse: delete dentry if timeout is zero\n \n Don\u0027t hold onto dentry in lru list if need to re-lookup it anyway at next\n access. Only do this if explicitly enabled, otherwise it could result in\n performance regression.\n \n More advanced version of this patch would periodically flush out dentries\n from the lru which have gone stale.\n \n Signed-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n\ncommit 86a303ea9bdee647700ea81689cfffa21d349e4d\nAuthor: Vivek Goyal \u003cvgoyal@redhat.com\u003e\nDate: Wed Mar 6 16:51:40 2019 -0500\n\n fuse: separate fuse device allocation and installation in fuse_conn\n \n As of now fuse_dev_alloc() both allocates a fuse device and installs it in\n fuse_conn list. fuse_dev_alloc() can fail if fuse_device allocation fails.\n \n virtio-fs needs to initialize multiple fuse devices (one per virtio queue).\n It initializes one fuse device as part of call to fuse_fill_super_common()\n and rest of the devices are allocated and installed after that.\n \n But, we can\u0027t afford to fail after calling fuse_fill_super_common() as we\n don\u0027t have a way to undo all the actions done by fuse_fill_super_common().\n So to avoid failures after the call to fuse_fill_super_common(),\n pre-allocate all fuse devices early and install them into fuse connection\n later.\n \n This patch provides two separate helpers for fuse device allocation and\n fuse device installation in fuse_conn.\n \n Signed-off-by: Vivek Goyal \u003cvgoyal@redhat.com\u003e\n Signed-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n\ncommit c86d5d8e82e08fada8a8fbddbcc2b46543a126da\nAuthor: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\nDate: Mon Jun 18 15:53:19 2018 +0100\n\n fuse: add fuse_iqueue_ops callbacks\n \n The /dev/fuse device uses fiq-\u003ewaitq and fasync to signal that requests are\n available. These mechanisms do not apply to virtio-fs. This patch\n introduces callbacks so alternative behavior can be used.\n \n Note that queue_interrupt() changes along these lines:\n \n spin_lock(\u0026fiq-\u003ewaitq.lock);\n wake_up_locked(\u0026fiq-\u003ewaitq);\n + kill_fasync(\u0026fiq-\u003efasync, SIGIO, POLL_IN);\n spin_unlock(\u0026fiq-\u003ewaitq.lock);\n - kill_fasync(\u0026fiq-\u003efasync, SIGIO, POLL_IN);\n \n Since queue_request() and queue_forget() also call kill_fasync() inside\n the spinlock this should be safe.\n \n Signed-off-by: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\n Signed-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n\ncommit 226a46262cdb1d0e7a1af0aab5e51bf01ef9b592\nAuthor: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\nDate: Wed Jun 13 10:23:04 2018 +0100\n\n fuse: extract fuse_fill_super_common()\n \n fuse_fill_super() includes code to process the fd\u003d option and link the\n struct fuse_dev to the fd\u0027s struct file. In virtio-fs there is no file\n descriptor because /dev/fuse is not used.\n \n This patch extracts fuse_fill_super_common() so that both classic fuse and\n virtio-fs can share the code to initialize a mount.\n \n Signed-off-by: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\n Signed-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n\ncommit c0b64faf0fe6ca2574a00faed1ae833130db4e08\nAuthor: Matthias Kaehlcke \u003cmka@chromium.org\u003e\nDate: Tue Jul 9 12:00:07 2019 -0700\n\n backlight: pwm_bl: Set scale type for brightness curves specified in the DT\n \n Check if a brightness curve specified in the device tree is linear or\n not and set the corresponding property accordingly. This makes the\n scale type available to userspace via the \u0027scale\u0027 sysfs attribute.\n \n To determine if a curve is linear it is compared to a interpolated linear\n curve between min and max brightness. The curve is considered linear if\n no value deviates more than +/-5% of ${brightness_range} from their\n interpolated value.\n \n Signed-off-by: Matthias Kaehlcke \u003cmka@chromium.org\u003e\n Acked-by: Daniel Thompson \u003cdaniel.thompson@linaro.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 511a204638d7d750f859c332635d09f38273b4f0\nAuthor: Matthias Kaehlcke \u003cmka@chromium.org\u003e\nDate: Tue Jul 9 12:00:06 2019 -0700\n\n backlight: pwm_bl: Set scale type for CIE 1931 curves\n \n For backlight curves calculated with the CIE 1931 algorithm set\n the brightness scale type to non-linear. This makes the scale type\n available to userspace via the \u0027scale\u0027 sysfs attribute.\n \n Signed-off-by: Matthias Kaehlcke \u003cmka@chromium.org\u003e\n Tested-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Acked-by: Daniel Thompson \u003cdaniel.thompson@linaro.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit d55c028f8b25bdaaba9ae08026052b5b44d031b0\nAuthor: Matthias Kaehlcke \u003cmka@chromium.org\u003e\nDate: Tue Jul 9 12:00:05 2019 -0700\n\n backlight: Expose brightness curve type through sysfs\n \n Backlight brightness curves can have different shapes. The two main\n types are linear and non-linear curves. The human eye doesn\u0027t\n perceive linearly increasing/decreasing brightness as linear (see\n also 88ba95bedb79 \"backlight: pwm_bl: Compute brightness of LED\n linearly to human eye\"), hence many backlights use non-linear (often\n logarithmic) brightness curves. The type of curve currently is opaque\n to userspace, so userspace often uses more or less reliable heuristics\n (like the number of brightness levels) to decide whether to treat a\n backlight device as linear or non-linear.\n \n Export the type of the brightness curve via the new sysfs attribute\n \u0027scale\u0027. The value of the attribute can be \u0027linear\u0027, \u0027non-linear\u0027 or\n \u0027unknown\u0027. For devices that don\u0027t provide information about the scale\n of their brightness curve the value of the \u0027scale\u0027 attribute is \u0027unknown\u0027.\n \n Signed-off-by: Matthias Kaehlcke \u003cmka@chromium.org\u003e\n Reviewed-by: Daniel Thompson \u003cdaniel.thompson@linaro.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 6451e123dec34e295cb0affcd48f671275bdb671\nAuthor: Matthias Kaehlcke \u003cmka@chromium.org\u003e\nDate: Tue Jul 9 12:00:04 2019 -0700\n\n MAINTAINERS: Add entry for stable backlight sysfs ABI documentation\n \n Add an entry for the stable backlight sysfs ABI to the MAINTAINERS\n file.\n \n Signed-off-by: Matthias Kaehlcke \u003cmka@chromium.org\u003e\n Acked-by: Daniel Thompson \u003cdaniel.thompson@linaro.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit ec665b756e6f79c60078b00dbdabea3aa8a4b787\nAuthor: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\nDate: Wed Jul 31 11:40:18 2019 +0300\n\n backlight: gpio-backlight: Correct initial power state handling\n \n The default-on property - or the def_value via legacy pdata) should be\n handled as:\n if it is 1, the backlight must be enabled (kept enabled)\n if it is 0, the backlight must be disabled (kept disabled)\n \n This only works for the case when default-on is set. If it is not set then\n the brightness of the backlight is set to 0. Now if the backlight is\n enabled by external driver (graphics) the backlight will stay disabled since\n the brightness is configured as 0. The backlight will not turn on.\n \n In order to minimize screen flickering during device boot:\n \n The initial brightness should be set to 1.\n \n If booted in non DT mode or no phandle link to the backlight node:\n follow the def_value/default-on to select UNBLANK or POWERDOWN\n \n If in DT boot we have phandle link then leave the GPIO in a state which the\n bootloader left it and let the user of the backlight to configure it\n further.\n \n Signed-off-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Reviewed-by: Daniel Thompson \u003cdaniel.thompson@linaro.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 28a1d72a221ecdf8f63205d40cb654c81a2b2da7\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Tue Aug 20 17:34:39 2019 +0200\n\n video: backlight: tosa_lcd: drop check because i2c_unregister_device() is NULL safe\n \n No need to check the argument of i2c_unregister_device() because the\n function itself does it.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Reviewed-by: Daniel Thompson \u003cdaniel.thompson@linaro.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit bcd69da98e36afccb4cb5fbdbd2b7bb78c07184d\nAuthor: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\nDate: Tue Aug 13 13:58:53 2019 +0200\n\n video: backlight: Drop default m for {LCD,BACKLIGHT_CLASS_DEVICE}\n \n When running \"make oldconfig\" on a .config where\n CONFIG_BACKLIGHT_LCD_SUPPORT is not set, two new config options\n (\"Lowlevel LCD controls\" and \"Lowlevel Backlight controls\") appear, both\n defaulting to \"m\".\n \n Drop the \"default m\", as options should default to disabled, and because\n several driver config options already select LCD_CLASS_DEVICE or\n BACKLIGHT_CLASS_DEVICE when needed.\n \n Fixes: 8c5dc8d9f19c7992 (\"video: backlight: Remove useless BACKLIGHT_LCD_SUPPORT kernel symbol\")\n Signed-off-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n Reviewed-by: Daniel Thompson \u003cdaniel.thompson@linaro.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit b0f76676ed4145ad94f518789710e0b1c5b12208\nAuthor: Vivek Goyal \u003cvgoyal@redhat.com\u003e\nDate: Wed Jun 5 15:50:43 2019 -0400\n\n fuse: export fuse_dequeue_forget() function\n \n File systems like virtio-fs need to do not have to play directly with\n forget list data structures. There is a helper function use that instead.\n \n Rename dequeue_forget() to fuse_dequeue_forget() and export it so that\n stacked filesystems can use it.\n \n Signed-off-by: Vivek Goyal \u003cvgoyal@redhat.com\u003e\n Signed-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n\ncommit ad69ac5bc8157188faf225898ef4ce8e160a87ed\nAuthor: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\nDate: Fri Jun 22 13:48:30 2018 +0100\n\n fuse: export fuse_get_unique()\n \n virtio-fs will need unique IDs for FORGET requests from outside\n fs/fuse/dev.c. Make the symbol visible.\n \n Signed-off-by: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\n Signed-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n\ncommit 294acb6399dea6c5afeab6fb60c3f518a792f036\nAuthor: Vivek Goyal \u003cvgoyal@redhat.com\u003e\nDate: Wed Mar 6 16:51:39 2019 -0500\n\n fuse: export fuse_send_init_request()\n \n This will be used by virtio-fs to send init request to fuse server after\n initialization of virt queues.\n \n Signed-off-by: Vivek Goyal \u003cvgoyal@redhat.com\u003e\n Signed-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n\ncommit 6f25ee765c7cfa03152cd99384ed4731bdba0136\nAuthor: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\nDate: Thu Jun 21 09:34:25 2018 +0100\n\n fuse: export fuse_len_args()\n \n virtio-fs will need to query the length of fuse_arg lists. Make the symbol\n visible.\n \n Signed-off-by: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\n Signed-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n\ncommit b4c19acbe00715b01d6d9a89c2c36008152be8c3\nAuthor: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\nDate: Thu Jun 21 09:33:40 2018 +0100\n\n fuse: export fuse_end_request()\n \n virtio-fs will need to complete requests from outside fs/fuse/dev.c. Make\n the symbol visible.\n \n Signed-off-by: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\n Signed-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n\ncommit e4ad40d4dd0bb0592889ccb02546652edd3edade\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Mar 25 16:38:31 2019 +0000\n\n vfs: subtype handling moved to fuse\n \n The unused vfs code can be removed. Don\u0027t pass empty subtype (same as if\n -\u003eparse callback isn\u0027t called).\n \n The bits that are left involve determining whether it\u0027s permitted to split the\n filesystem type string passed in to mount(2). Consequently, this means that we\n cannot get rid of the FS_HAS_SUBTYPE flag unless we define that a type string\n with a dot in it always indicates a subtype specification.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n Signed-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n\ncommit 5eae593422daee92b354bb36685018e16d0da7f6\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Mar 25 16:38:31 2019 +0000\n\n vfs: Convert fuse to use the new mount API\n \n Convert the fuse filesystem to the new internal mount API as the old\n one will be obsoleted and removed. This allows greater flexibility in\n communication of mount parameters between userspace, the VFS and the\n filesystem.\n \n See Documentation/filesystems/mount_api.txt for more information.\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n Signed-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n\ncommit c709bf455d60662377a47686180c0ac5c65f4d07\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Fri Aug 30 18:12:36 2019 +0200\n\n microblaze: remove ioremap_fullcache\n \n No callers of this function.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Michal Simek \u003cmichal.simek@xilinx.com\u003e\n\ncommit e8bd417aab0c72bfb54465596b16085702ba0405\nAuthor: Sasha Levin \u003csashal@kernel.org\u003e\nDate: Fri Jul 5 16:47:46 2019 -0400\n\n tpm/tpm_ftpm_tee: Document fTPM TEE driver\n \n Documentation briefly the new fTPM driver running inside TEE.\n \n Signed-off-by: Sasha Levin \u003csashal@kernel.org\u003e\n Reviewed-by: Jarkko Sakkinen \u003cjarkko.sakkinen@linux.intel.com\u003e\n Signed-off-by: Jarkko Sakkinen \u003cjarkko.sakkinen@linux.intel.com\u003e\n\ncommit 09e574831b277a3f77d78ceadd08a3859a84fdb3\nAuthor: Sasha Levin \u003csashal@kernel.org\u003e\nDate: Fri Jul 5 16:47:45 2019 -0400\n\n tpm/tpm_ftpm_tee: A driver for firmware TPM running inside TEE\n \n Add a driver for a firmware TPM running inside TEE.\n \n Documentation of the firmware TPM:\n https://www.microsoft.com/en-us/research/publication/ftpm-software-implementation-tpm-chip/ .\n \n Implementation of the firmware TPM:\n https://github.com/Microsoft/ms-tpm-20-ref/tree/master/Samples/ARM32-FirmwareTPM\n \n Tested-by: Ilias Apalodimas \u003cilias.apalodimas@linaro.org\u003e\n Tested-by: Thirupathaiah Annapureddy \u003cthiruan@microsoft.com\u003e\n Signed-off-by: Thirupathaiah Annapureddy \u003cthiruan@microsoft.com\u003e\n Co-authored-by: Sasha Levin \u003csashal@kernel.org\u003e\n Signed-off-by: Sasha Levin \u003csashal@kernel.org\u003e\n Reviewed-by: Jarkko Sakkinen \u003cjarkko.sakkinen@linux.intel.com\u003e\n Signed-off-by: Jarkko Sakkinen \u003cjarkko.sakkinen@linux.intel.com\u003e\n\ncommit 28eba2fddd5a4292bf9e3dc55193728f3055b782\nAuthor: Jarkko Sakkinen \u003cjarkko.sakkinen@linux.intel.com\u003e\nDate: Mon Jul 8 17:56:39 2019 +0300\n\n tpm: Remove a deprecated comments about implicit sysfs locking\n \n Remove all comments about implicit locking tpm-sysfs.c as the file was\n updated in Linux v5.1 to use explicit locking.\n \n Signed-off-by: Jarkko Sakkinen \u003cjarkko.sakkinen@linux.intel.com\u003e\n\ncommit 1ea32c83c699df32689d329b2415796b7bfc2f6e\nAuthor: Stefan Berger \u003cstefanb@linux.ibm.com\u003e\nDate: Thu Aug 29 20:09:06 2019 -0400\n\n tpm_tis_core: Set TPM_CHIP_FLAG_IRQ before probing for interrupts\n \n The tpm_tis_core has to set the TPM_CHIP_FLAG_IRQ before probing for\n interrupts since there is no other place in the code that would set\n it.\n \n Cc: linux-stable@vger.kernel.org\n Fixes: 570a36097f30 (\"tpm: drop \u0027irq\u0027 from struct tpm_vendor_specific\")\n Signed-off-by: Stefan Berger \u003cstefanb@linux.ibm.com\u003e\n Signed-off-by: Jarkko Sakkinen \u003cjarkko.sakkinen@linux.intel.com\u003e\n\ncommit 5b359c7c43727e624eac3efc7ad21bd2defea161\nAuthor: Stefan Berger \u003cstefanb@linux.ibm.com\u003e\nDate: Tue Aug 20 08:25:17 2019 -0400\n\n tpm_tis_core: Turn on the TPM before probing IRQ\u0027s\n \n The interrupt probing sequence in tpm_tis_core cannot obviously run with\n the TPM power gated. Power on the TPM with tpm_chip_start() before\n probing IRQ\u0027s. Turn it off once the probing is complete.\n \n Cc: linux-stable@vger.kernel.org\n Fixes: a3fbfae82b4c (\"tpm: take TPM chip power gating out of tpm_transmit()\")\n Signed-off-by: Stefan Berger \u003cstefanb@linux.ibm.com\u003e\n Reviewed-by: Jarkko Sakkinen \u003cjarkko.sakkinen@linux.intel.com\u003e\n Signed-off-by: Jarkko Sakkinen \u003cjarkko.sakkinen@linux.intel.com\u003e\n\ncommit 14591e2351ef36a5b7509ad19043bd8c4b9f241f\nAuthor: Lukas Bulwahn \u003clukas.bulwahn@gmail.com\u003e\nDate: Sun Aug 25 19:00:15 2019 +0200\n\n MAINTAINERS: fix style in KEYS-TRUSTED entry\n \n Mimi Zohar used spaces instead of a tab when adding Jarkko Sakkinen as\n further maintainer to the KEYS-TRUSTED section entry.\n \n In fact, ./scripts/checkpatch.pl -f MAINTAINERS complains:\n \n WARNING: MAINTAINERS entries use one tab after TYPE:\n #8581: FILE: MAINTAINERS:8581:\n +M: Jarkko Sakkinen \u003cjarkko.sakkinen@linux.intel.com\u003e\n \n The issue was detected when writing a script that parses MAINTAINERS.\n \n Fixes: 34bccd61b139 (\"MAINTAINERS: add Jarkko as maintainer for trusted keys\")\n Signed-off-by: Lukas Bulwahn \u003clukas.bulwahn@gmail.com\u003e\n Acked-by: Jarkko Sakkinen \u003cjarkko.sakkinen@linux.intel.com\u003e\n Signed-off-by: Jarkko Sakkinen \u003cjarkko.sakkinen@linux.intel.com\u003e\n\ncommit 9a8892922547da193d0cf81fc8053c642f90d2b4\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Aug 19 13:52:52 2019 +0300\n\n reset: Remove copy\u0027n\u0027paste redundancy in the comments\n \n It seems the commit bb475230b8e5\n (\"reset: make optional functions really optional\")\n brought couple of redundant lines in the comments.\n \n Drop them here.\n \n Cc: Ramiro Oliveira \u003cRamiro.Oliveira@synopsys.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit faf08e4cb7eaa6bc82a3877e4a1f873c51de295c\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Thu Aug 29 17:52:24 2019 +0100\n\n smack: Implement the watch_key and post_notification hooks [untested]\n \n Implement the watch_key security hook in Smack to make sure that a key\n grants the caller Read permission in order to set a watch on a key.\n \n Also implement the post_notification security hook to make sure that the\n notification source is granted Write permission by the watch queue.\n \n For the moment, the watch_devices security hook is left unimplemented as\n it\u0027s not obvious what the object should be since the queue is global and\n didn\u0027t previously exist.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n\ncommit 6fb035502956ad1f338ca61c057653d5372ffd8c\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Aug 23 09:58:58 2019 +0900\n\n ASoC: soc-core: move soc_probe_link_components() position\n \n It is easy to read code if it is cleanly using paired function/naming,\n like start \u003c-\u003e stop, register \u003c-\u003e unregister, etc, etc.\n But, current ALSA SoC code is very random, unbalance, not paired, etc.\n It is easy to create bug at the such code, and it will be difficult to\n debug.\n \n soc_probe_link_components() has paired soc_remove_link_components(),\n but, these are implemented at different place.\n So it is difficult to confirm code.\n This patch moves soc_probe_link_components() next to\n soc_remove_link_components().\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87o90g7lbd.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 95c267dd20431f0eb54ca204bd73a7d85c532a37\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Aug 23 09:58:52 2019 +0900\n\n ASoC: soc-core: add snd_soc_dapm_init()\n \n It is easy to read code if it is cleanly using paired function/naming,\n like start \u003c-\u003e stop, register \u003c-\u003e unregister, etc, etc.\n But, current ALSA SoC code is very random, unbalance, not paired, etc.\n It is easy to create bug at the such code, and it will be difficult to\n debug.\n \n soc-dapm has snd_soc_dapm_free() which cleanups debugfs, widgets, list.\n But, there is no paired initialize function.\n This patch adds snd_soc_dapm_init() and initilaizing dapm\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87pnkw7lbj.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit b614beafa495c7f6fbc15cb6977e3fe48beea1e5\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Aug 23 09:58:47 2019 +0900\n\n ASoC: soc-core: dapm related setup at one place\n \n Current ASoC setups some dapm related member at\n snd_soc_component_initialize() which is called when component was\n registered, and setups remaining member at soc_probe_component()\n which is called when component was probed.\n This kind of setup separation is no meanings, and it is very\n difficult to read and confusable.\n This patch setups all dapm settings at one place.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87r25c7lbo.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit ffd60fba19d9752f553aac367cd40362011ab6c9\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Aug 23 09:58:42 2019 +0900\n\n ASoC: soc-core: move soc_probe_component() position\n \n It is easy to read code if it is cleanly using paired function/naming,\n like start \u003c-\u003e stop, register \u003c-\u003e unregister, etc, etc.\n But, current ALSA SoC code is very random, unbalance, not paired, etc.\n It is easy to create bug at the such code, and it will be difficult to\n debug.\n \n soc_probe_comonent() has paired soc_remove_comonent(),\n but, these are implemented at different place.\n So it is difficult to confirm code.\n This patch moves soc_probe_component() next to\n soc_remove_component().\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87sgps7lbt.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9a7c9fe1203eb360b3a01b65a18bcd0de6670c53\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Aug 23 09:58:37 2019 +0900\n\n ASoC: soc-core: add soc_rtd_free()\n \n It is easy to read code if it is cleanly using paired function/naming,\n like start \u003c-\u003e stop, register \u003c-\u003e unregister, etc, etc.\n But, current ALSA SoC code is very random, unbalance, not paired, etc.\n It is easy to create bug at the such code, and it will be difficult to\n debug.\n \n soc_rtd_init() was soc_post_component_init(), but there was no\n its paired soc_post_component_free(), but it is done at\n soc_remove_link_dais().\n This means it is difficult to find related code.\n \n This patch adds soc_rtd_free() which is paired soc_rtd_init().\n soc_rtd_xxx() will be more cleanuped in the future.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87tva87lby.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 4fbc485324d2975c54201091dfad0a7dd4902324\nAuthor: Alexander Sverdlin \u003calexander.sverdlin@gmail.com\u003e\nDate: Sat Aug 31 20:04:02 2019 +0200\n\n spi: ep93xx: Repair SPI CS lookup tables\n \n The actual device name of the SPI controller being registered on EP93xx is\n \"spi0\" (as seen by gpiod_find_lookup_table()). This patch fixes all\n relevant lookup tables and the following failure (seen on EDB9302):\n \n ep93xx-spi ep93xx-spi.0: failed to register SPI master\n ep93xx-spi: probe of ep93xx-spi.0 failed with error -22\n \n Fixes: 1dfbf334f1236 (\"spi: ep93xx: Convert to use CS GPIO descriptors\")\n Cc: stable@vger.kernel.org\n Signed-off-by: Alexander Sverdlin \u003calexander.sverdlin@gmail.com\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Reviewed-by: Lukasz Majewski \u003clukma@denx.de\u003e\n Link: https://lore.kernel.org/r/20190831180402.10008-1-alexander.sverdlin@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit d0087e72710ca7d1b309bf427286da58418ea89e\nAuthor: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\nDate: Fri Aug 30 09:17:37 2019 +0200\n\n regulator: provide regulator_bulk_set_supply_names()\n \n There are many regulator consumers who - before using the regulator\n bulk functions - set the supply names in regulator_bulk_data using\n a for loop.\n \n Let\u0027s provide a simple helper in the consumer API that allows users\n to do the same with a single function call.\n \n Signed-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n Link: https://lore.kernel.org/r/20190830071740.4267-2-brgl@bgdev.pl\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit dbcc574a4bfad810ce6e118f3cf327c7b6e7c897\nAuthor: Wei Yongjun \u003cweiyongjun1@huawei.com\u003e\nDate: Mon Aug 19 06:17:33 2019 +0000\n\n drm/etnaviv: fix missing unlock on error in etnaviv_iommuv1_context_alloc()\n \n Add the missing unlock before return from function etnaviv_iommuv1_context_alloc()\n in the error handling case.\n \n Fixes: 27b67278e007 (\"drm/etnaviv: rework MMU handling\")\n Signed-off-by: Wei Yongjun \u003cweiyongjun1@huawei.com\u003e\n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n\ncommit a0753c29004f4983e303abce019f29e183b1ee48\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon May 20 08:48:46 2019 +0100\n\n afs: Support RCU pathwalk\n \n Make afs_permission() and afs_d_revalidate() do initial checks in RCU-mode\n pathwalk to reduce latency in pathwalk elements that get done multiple\n times. We don\u0027t need to query the server unless we\u0027ve received a\n notification from it that something has changed or the callback has\n expired.\n \n This requires that we can request a key and check permits under RCU\n conditions if we need to.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n\ncommit 8b6a666a97544bf307190a05947742b8357aa962\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon May 20 08:48:46 2019 +0100\n\n afs: Provide an RCU-capable key lookup\n \n Provide an RCU-capable key lookup function. We don\u0027t want to call\n afs_request_key() in RCU-mode pathwalk as request_key() might sleep, even if\n we don\u0027t ask it to construct anything as it might find a key that is currently\n undergoing construction.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n\ncommit 23a289137ab82daeea826eeb9556c6f89b1fcd67\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Tue Aug 20 09:22:38 2019 +0100\n\n afs: Use afs_extract_discard() rather than iov_iter_discard()\n \n Use afs_extract_discard() rather than iov_iter_discard() as the former is a\n wrapper for the latter, providing a place to put tracepoints.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n\ncommit 52c9c130780095fdf72fbf6920a6a1c936fb5e6f\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 9 16:42:31 2019 +0800\n\n afs: remove unused variable \u0027afs_zero_fid\u0027\n \n fs/afs/fsclient.c:18:29: warning:\n afs_zero_fid defined but not used [-Wunused-const-variable\u003d]\n \n It is never used since commit 025db80c9e42 (\"afs: Trace\n the initiation and completion of client calls\")\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n\ncommit cacf2d7dcf966a57e30f66c0598c88e2b64865a3\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 9 16:40:37 2019 +0800\n\n afs: remove unused variable \u0027afs_voltypes\u0027\n \n fs/afs/volume.c:15:26: warning:\n afs_voltypes defined but not used [-Wunused-const-variable\u003d]\n \n It is not used since commit d2ddc776a458 (\"afs: Overhaul\n volume and server record caching and fileserver rotation\")\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n\ncommit 28e6fcc871bcff640c8960448034ea3a7c7fdfa3\nAuthor: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\nDate: Mon Sep 2 11:53:09 2019 +0200\n\n mfd: cros_ec: Use mfd_add_hotplug_devices() helper\n \n Use mfd_add_hotplug_devices() helper to register the subdevices. The\n helper allows us to reduce the boiler plate and also registers the\n subdevices in the same way as used in other functions used in this\n files.\n \n Signed-off-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Reviewed-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Tested-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 832a636f6afe16aa95cb7281c4eb2194b402515d\nAuthor: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\nDate: Mon Sep 2 11:53:08 2019 +0200\n\n mfd: cros_ec: Add convenience struct to define autodetectable CrOS EC subdevices\n \n The CrOS EC is gaining lots of subdevices that are autodetectable by\n sending the EC_FEATURE_GET_CMD, it takes fair amount of boiler plate\n code to add those devices. So, add a struct that can be used to quickly\n add new subdevices without having to duplicate code.\n \n Signed-off-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Acked-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Tested-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit b027dcf7c4c3e0d22b9ce84011eaf02cf0adf559\nAuthor: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\nDate: Mon Sep 2 11:53:07 2019 +0200\n\n mfd: cros_ec: Add convenience struct to define dedicated CrOS EC MCUs\n \n With the increasing use of dedicated CrOS EC MCUs, it takes a fair amount\n of boiler plate code to add those devices, add a struct that can be used\n to specify a dedicated CrOS EC MCU so we can just add a new item to it to\n define a new dedicated MCU.\n \n Signed-off-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Acked-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Tested-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 5156fb75ead1dad9cdd250db898011e92c6ecb22\nAuthor: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\nDate: Mon Sep 2 11:53:06 2019 +0200\n\n mfd: cros_ec: Use kzalloc and cros_ec_cmd_xfer_status helper\n \n This patch makes use of cros_ec_cmd_xfer_status() instead of\n cros_ec_cmd_xfer() so we can remove some redundant code. It also uses\n kzalloc instead of kmalloc so we can remove more redundant code.\n \n Signed-off-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Acked-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Tested-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 840d9f131f65b021e0a73f3371f3194897dba6ad\nAuthor: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\nDate: Mon Sep 2 11:53:05 2019 +0200\n\n mfd / platform: cros_ec: Reorganize platform and mfd includes\n \n There is a bit of mess between cros-ec mfd includes and platform\n includes. For example, we have a linux/mfd/cros_ec.h include that\n exports the interface implemented in platform/chrome/cros_ec_proto.c. Or\n we have a linux/mfd/cros_ec_commands.h file that is non related to the\n multifunction device (in the sense that is not exporting any function of\n the mfd device). This causes crossed includes between mfd and\n platform/chrome subsystems and makes the code difficult to read, apart\n from creating \u0027curious\u0027 situations where a platform/chrome driver includes\n a linux/mfd/cros_ec.h file just to get the exported functions that are\n implemented in another platform/chrome driver.\n \n In order to have a better separation on what the cros-ec multifunction\n driver does and what the cros-ec core provides move and rework the\n affected includes doing:\n \n - Move cros_ec_commands.h to include/linux/platform_data/cros_ec_commands.h\n - Get rid of the parts that are implemented in the platform/chrome/cros_ec_proto.c\n driver from include/linux/mfd/cros_ec.h to a new file\n include/linux/platform_data/cros_ec_proto.h\n - Update all the drivers with the new includes, so\n - Drivers that only need to know about the protocol include\n - linux/platform_data/cros_ec_proto.h\n - linux/platform_data/cros_ec_commands.h\n - Drivers that need to know about the cros-ec mfd device also include\n - linux/mfd/cros_ec.h\n \n Signed-off-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Acked-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Mark Brown \u003cbroonie@kernel.org\u003e\n Acked-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n Acked-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Acked-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Acked-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n Acked-by: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\n Acked-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n Acked-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n Acked-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Reviewed-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Tested-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Series changes: 3\n - Fix dereferencing pointer to incomplete type \u0027struct cros_ec_dev\u0027 (lkp)\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 2fa2b980e3fe187a0b916bfe6bd5822889b44d5e\nAuthor: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\nDate: Mon Sep 2 11:53:04 2019 +0200\n\n mfd / platform: cros_ec: Rename config to a better name\n \n The cros-ec-dev is a multifunction device that now doesn\u0027t implement any\n chardev communication interface. MFD_CROS_EC_CHARDEV doesn\u0027t look\n a good name to describe that device and can cause confusion. Hence\n rename it to CROS_EC_DEV.\n \n Signed-off-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Acked-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Tested-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 459aedb9a5d4b92e4aee343ee8ee5aeca8e1d93a\nAuthor: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\nDate: Mon Sep 2 11:53:03 2019 +0200\n\n mfd: cros_ec: Switch to use the new cros-ec-chardev driver\n \n With the purpose of remove the things that far extends the bounds of\n what a MFD was designed to do, instantiate the new platform misc\n cros-ec-chardev driver and get rid of all the unneeded code. After this\n patch the misc chardev driver is a sub-device of the MFD, and all the\n new file operations should be implemented there.\n \n Signed-off-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Acked-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Tested-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit eda2e30c6684d67288edb841c6125d48c608a242\nAuthor: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\nDate: Mon Sep 2 11:53:02 2019 +0200\n\n mfd / platform: cros_ec: Miscellaneous character device to talk with the EC\n \n That\u0027s a driver to talk with the ChromeOS Embedded Controller via a\n miscellaneous character device, it creates an entry in /dev for every\n instance and implements basic file operations for communicating with the\n Embedded Controller with an userspace application. The API is moved to\n the uapi folder, which is supposed to contain the user space API of the\n kernel.\n \n Note that this will replace current character device interface\n implemented in the cros-ec-dev driver in the MFD subsystem. The idea is\n to move all the functionality that extends the bounds of what MFD was\n designed to platform/chrome subsystem.\n \n Signed-off-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Acked-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Tested-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 47f11e0b40e97f373da4efbacee0a9526c816ed5\nAuthor: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\nDate: Mon Sep 2 11:53:01 2019 +0200\n\n mfd / platform: cros_ec: Move cros-ec core driver out from MFD\n \n Now, the ChromeOS EC core driver has nothing related to an MFD device, so\n move that driver from the MFD subsystem to the platform/chrome subsystem.\n \n Signed-off-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Acked-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Acked-by: Mark Brown \u003cbroonie@kernel.org\u003e\n Acked-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n Acked-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Acked-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Acked-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n Acked-by: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\n Acked-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n Acked-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n Acked-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Reviewed-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Tested-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 7aa703bb8824384baad732043a925b46a4f3efa8\nAuthor: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\nDate: Mon Sep 2 11:53:00 2019 +0200\n\n mfd / platform: cros_ec: Handle chained ECs as platform devices\n \n An MFD is a device that contains several sub-devices (cells). For instance,\n the ChromeOS EC fits in this description as usually contains a charger and\n can have other devices with different functions like a Real-Time Clock,\n an Audio codec, a Real-Time Clock, ...\n \n If you look at the driver, though, we\u0027re doing something odd. We have\n two MFD cros-ec drivers where one of them (cros-ec-core) instantiates\n another MFD driver as sub-driver (cros-ec-dev), and the latest\n instantiates the different sub-devices (Real-Time Clock, Audio codec,\n etc).\n \n MFD\n ------------------------------------------\n cros-ec-core\n |___ mfd-cellA (cros-ec-dev)\n | |__ mfd-cell0\n | |__ mfd-cell1\n | |__ ...\n |\n |___ mfd-cellB (cros-ec-dev)\n |__ mfd-cell0\n |__ mfd-cell1\n |__ ...\n \n The problem that was trying to solve is to describe some kind of topology for\n the case where we have an EC (cros-ec) chained with another EC\n (cros-pd). Apart from that this extends the bounds of what MFD was\n designed to do we might be interested on have other kinds of topology that\n can\u0027t be implemented in that way.\n \n Let\u0027s prepare the code to move the cros-ec-core part from MFD to\n platform/chrome as this is clearly a platform specific thing non-related\n to a MFD device.\n \n platform/chrome | MFD\n ------------------------------------------\n |\n cros-ec ________|___ cros-ec-dev\n | |__ mfd-cell0\n | |__ mfd-cell1\n | |__ ...\n |\n cros-pd ________|___ cros-ec-dev\n | |__ mfd-cell0\n | |__ mfd-cell1\n | |__ ...\n \n Signed-off-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Acked-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Tested-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 38f6a046f02331e7efd513c73dbfe9d7eb9cd0a4\nAuthor: Lukasz Luba \u003cl.luba@partner.samsung.com\u003e\nDate: Wed Jun 5 11:12:32 2019 +0200\n\n include: dt-bindings: add Performance Monitoring Unit for Exynos\n \n This patch add support of a new feature which can be used in DT:\n Performance Monitoring Unit with defined event data type.\n In this patch the event data types are defined for Exynos PPMU.\n The patch also updates the MAINTAINERS file accordingly and\n adds the header file to devfreq event subsystem.\n \n Acked-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Lukasz Luba \u003cl.luba@partner.samsung.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 8391c6cb2414d9a75bbe247a838d28cb0cee77ee\nAuthor: Josef Friedl \u003cjosef.friedl@speed.at\u003e\nDate: Sun Aug 18 15:56:08 2019 +0200\n\n mfd: mt6323: Add MT6323 RTC and PWRC\n \n Add entry for RTC and Power Controller to MT6323.\n \n Signed-off-by: Josef Friedl \u003cjosef.friedl@speed.at\u003e\n Signed-off-by: Frank Wunderlich \u003cfrank-w@public-files.de\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 7c3f7cd5a0c96ab40baaaf08968249fcb96c4057\nAuthor: Josef Friedl \u003cjosef.friedl@speed.at\u003e\nDate: Sun Aug 18 15:56:06 2019 +0200\n\n mfd: mt6323: Replace boilerplate resource code with DEFINE_RES_* macros\n \n Simplifies and reduces LoC.\n \n Signed-off-by: Josef Friedl \u003cjosef.friedl@speed.at\u003e\n Signed-off-by: Frank Wunderlich \u003cfrank-w@public-files.de\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 09c1dec470029f812e86aa92908abfd9132c2fab\nAuthor: Josef Friedl \u003cjosef.friedl@speed.at\u003e\nDate: Sun Aug 18 15:56:04 2019 +0200\n\n mfd: mt6397: Add mutex include\n \n Add missing mutex.h.\n \n Signed-off-by: Josef Friedl \u003cjosef.friedl@speed.at\u003e\n Signed-off-by: Frank Wunderlich \u003cfrank-w@public-files.de\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 8ab1267ff73dc372d63dffc7736058d456b8c75b\nAuthor: Josef Friedl \u003cjosef.friedl@speed.at\u003e\nDate: Sun Aug 18 15:56:01 2019 +0200\n\n dt-bindings: mfd: mediatek: Add MT6323 Power Controller\n \n - Add Power Controller section to existing binding document.\n - Add MT6323 PWRC bindings document with example.\n \n Suggested-by: Frank Wunderlich \u003cfrank-w@public-files.de\u003e\n Signed-off-by: Josef Friedl \u003cjosef.friedl@speed.at\u003e\n Signed-off-by: Frank Wunderlich \u003cfrank-w@public-files.de\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 7051919d17512f51e21d91b6564882b0476f2a2c\nAuthor: Josef Friedl \u003cjosef.friedl@speed.at\u003e\nDate: Sun Aug 18 15:56:00 2019 +0200\n\n dt-bindings: mfd: mediatek: Update RTC to include MT6323\n \n Add MT6323 to RTC bindings.\n \n Signed-off-by: Josef Friedl \u003cjosef.friedl@speed.at\u003e\n Signed-off-by: Frank Wunderlich \u003cfrank-w@public-files.de\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit b788d111e6760602a097cd774e09d0f427de707c\nAuthor: Frank Wunderlich \u003cfrank-w@public-files.de\u003e\nDate: Sun Aug 18 15:55:59 2019 +0200\n\n dt-bindings: mfd: mediatek: mt6397: Change to relative paths\n \n Paths in dt-bindings should be relative as suggested by Lee Jones.\n \n Suggested-By: Lee Jones \u003clee.jones@linaro.org\u003e\n Signed-off-by: Frank Wunderlich \u003cfrank-w@public-files.de\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit fea3ac55e112ee52feba03c48c182ab6d0ad8e92\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Thu Aug 29 13:25:01 2019 +0200\n\n mfd: db8500-prcmu: Support the higher DB8520 ARMSS\n \n The DB8520 used in a lot of Samsung phones has a slightly higher\n maximum ARMSS frequency than the DB8500. In order to not confuse\n the OPP framework and cpufreq, make sure the PRCMU driver\n returns the correct frequency.\n \n Cc: Stephan Gerhold \u003cstephan@gerhold.net\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 569fac74627cc332a2097a7a4bfdc654b8e7f273\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Wed Aug 21 11:37:12 2019 +0300\n\n mfd: intel-lpss: Use MODULE_SOFTDEP() instead of implicit request\n \n There is no need to handle optional module request in the driver\n when user space tools has that feature for ages.\n \n Replace custom code by MODULE_SOFTDEP() macro to let user space know\n that we would like to have the DMA driver loaded first, if any.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 1094422253db01ac3a12bde010c75e5135021d2d\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Tue Aug 20 17:34:43 2019 +0200\n\n mfd: htc-i2cpld: Drop check because i2c_unregister_device() is NULL safe\n \n No need to check the argument of i2c_unregister_device() because the\n function itself does it.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit f68c0a873ef28637a201ec37e1bafdf040813454\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Tue Aug 20 12:37:15 2019 +0200\n\n mfd: sm501: Include the GPIO driver header\n \n This driver creates a gpio chip so it needs to include the\n appropriate header \u003clinux/gpio/driver.h\u003e explicitly rather\n than implicitly.\n \n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit c5b90cb26e83ad89cb18e5ec97a992f02d8f750d\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Fri Aug 16 20:56:02 2019 +0300\n\n mfd: intel-lpss: Add Intel Skylake ACPI IDs\n \n Some of the laptops, like ASUS U306UA, may expose LPSS devices via ACPI.\n \n Add their IDs to the list.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit cbd1c5c4d443c4a83c9c7eecc0561b9aeafddf50\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Fri Aug 16 20:33:42 2019 +0300\n\n mfd: intel-lpss: Consistently use GENMASK()\n \n Since we already are using BIT() macro, use GENMASK() as well for sake of\n consistency.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit b9a801dfa59163dc2db8147a98af406eb79e51de\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Thu Aug 1 22:03:35 2019 +0300\n\n mfd: Add support for Merrifield Basin Cove PMIC\n \n Add an MFD driver for Intel Merrifield Basin Cove PMIC.\n \n Firmware on the platforms which are using Basin Cove PMIC is \"smarter\"\n than on the rest supported by vanilla kernel. It handles first level\n of interrupt itself, while others do it on OS level.\n \n The driver is done in the same way as the rest of Intel PMIC MFD drivers\n in the kernel to support the initial design. The design allows to use\n one driver among few PMICs without knowing implementation details of\n the each hardware version or generation.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit b65dc4f6b339ff57321fd95f2f7b6197a3c24ba4\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Tue Aug 13 18:31:33 2019 +0800\n\n mfd: ezx-pcap: Replace mutex_lock with spin_lock\n \n As mutex_lock might sleep. Function pcap_adc_irq is an interrupt handler.\n The use of mutex_lock in pcap_adc_irq may cause sleep in IRQ context.\n Replace mutex_lock with spin_lock to avoid this.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit c08428c528bc7827b29e112ed5a2c5efe053a1e9\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Wed Mar 27 14:15:16 2019 +0000\n\n vfs: Create fs_context-aware mount_bdev() replacement\n \n Create a function, vfs_get_block_super(), that is fs_context-aware and a\n replacement for mount_bdev(). It caches the block device pointer and file\n open mode in the fs_context struct so that this information can be passed\n into sget_fc()\u0027s test and set functions.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n cc: Jens Axboe \u003caxboe@kernel.dk\u003e\n cc: linux-block@vger.kernel.org\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 56d250ef9650edce600c96e2f918b9b9bafda85e\nAuthor: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\nDate: Thu Aug 29 11:01:18 2019 +0200\n\n cuse: fix broken release\n \n The inode parameter in cuse_release() is likely *not* a fuse inode. It\u0027s a\n small wonder it didn\u0027t blow up until now.\n \n Signed-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n\ncommit 17b2cbe294922ec2b78e29b93ed602b56cd75a4e\nAuthor: Maxim Patlasov \u003cmpatlasov@virtuozzo.com\u003e\nDate: Mon Jul 22 10:17:17 2019 +0300\n\n fuse: cleanup fuse_wait_on_page_writeback\n \n fuse_wait_on_page_writeback() always returns zero and nobody cares.\n Let\u0027s make it void.\n \n Signed-off-by: Maxim Patlasov \u003cmpatlasov@virtuozzo.com\u003e\n Signed-off-by: Vasily Averin \u003cvvs@virtuozzo.com\u003e\n Signed-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n\ncommit 1fb027d7596464d3fad3ed59f70f43807ef926c6\nAuthor: Kirill Smelkov \u003ckirr@nexedi.com\u003e\nDate: Mon Jul 8 17:03:31 2019 +0000\n\n fuse: require /dev/fuse reads to have enough buffer capacity (take 2)\n \n [ This retries commit d4b13963f217 (\"fuse: require /dev/fuse reads to have\n enough buffer capacity\"), which was reverted. In this version we require\n only `sizeof(fuse_in_header) + sizeof(fuse_write_in)` instead of 4K for\n FUSE request header room, because, contrary to libfuse and kernel client\n behaviour, GlusterFS actually provides only so much room for request\n header. ]\n \n A FUSE filesystem server queues /dev/fuse sys_read calls to get filesystem\n requests to handle. It does not know in advance what would be that request\n as it can be anything that client issues - LOOKUP, READ, WRITE, ... Many\n requests are short and retrieve data from the filesystem. However WRITE and\n NOTIFY_REPLY write data into filesystem.\n \n Before getting into operation phase, FUSE filesystem server and kernel\n client negotiate what should be the maximum write size the client will ever\n issue. After negotiation the contract in between server/client is that the\n filesystem server then should queue /dev/fuse sys_read calls with enough\n buffer capacity to receive any client request - WRITE in particular, while\n FUSE client should not, in particular, send WRITE requests with \u003e\n negotiated max_write payload. FUSE client in kernel and libfuse\n historically reserve 4K for request header. However an existing filesystem\n server - GlusterFS - was found which reserves only 80 bytes for header room\n (\u003d `sizeof(fuse_in_header) + sizeof(fuse_write_in)`).\n \n Since\n \n `sizeof(fuse_in_header) + sizeof(fuse_write_in)` \u003d\u003d\n `sizeof(fuse_in_header) + sizeof(fuse_read_in)` \u003d\u003d\n `sizeof(fuse_in_header) + sizeof(fuse_notify_retrieve_in)`\n \n is the absolute minimum any sane filesystem should be using for header\n room, the contract is that filesystem server should queue sys_reads with\n `sizeof(fuse_in_header) + sizeof(fuse_write_in)` + max_write buffer.\n \n If the filesystem server does not follow this contract, what can happen\n is that fuse_dev_do_read will see that request size is \u003e buffer size,\n and then it will return EIO to client who issued the request but won\u0027t\n indicate in any way that there is a problem to filesystem server.\n This can be hard to diagnose because for some requests, e.g. for\n NOTIFY_REPLY which mimics WRITE, there is no client thread that is\n waiting for request completion and that EIO goes nowhere, while on\n filesystem server side things look like the kernel is not replying back\n after successful NOTIFY_RETRIEVE request made by the server.\n \n We can make the problem easy to diagnose if we indicate via error return to\n filesystem server when it is violating the contract. This should not\n practically cause problems because if a filesystem server is using shorter\n buffer, writes to it were already very likely to cause EIO, and if the\n filesystem is read-only it should be too following FUSE_MIN_READ_BUFFER\n minimum buffer size.\n \n Please see [1] for context where the problem of stuck filesystem was hit\n for real (because kernel client was incorrectly sending more than\n max_write data with NOTIFY_REPLY; see also previous patch), how the\n situation was traced and for more involving patch that did not make it\n into the tree.\n \n [1] https://marc.info/?l\u003dlinux-fsdevel\u0026m\u003d155057023600853\u0026w\u003d2\n \n Signed-off-by: Kirill Smelkov \u003ckirr@nexedi.com\u003e\n Tested-by: Sander Eikelenboom \u003clinux@eikelenboom.it\u003e\n Cc: Han-Wen Nienhuys \u003chanwen@google.com\u003e\n Cc: Jakob Unterwurzacher \u003cjakobunt@gmail.com\u003e\n Signed-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n\ncommit 58985a9d2d03e977db93bf574a16162766a318fe\nAuthor: Valdis Klētnieks \u003cvaldis.kletnieks@vt.edu\u003e\nDate: Fri Aug 30 12:42:39 2019 -0400\n\n drivers/staging/exfat - by default, prohibit mount of fat/vfat\n \n Concerns have been raised about the exfat driver accidentally mounting\n fat/vfat file systems. Add an extra configure option to help prevent that.\n \n Signed-off-by: Valdis Kletnieks \u003cvaldis.kletnieks@vt.edu\u003e\n Link: https://lore.kernel.org/r/245727.1567183359@turing-police\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 8288022284859acbcc3cf1a073a1e2692d6c2543\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Thu Aug 1 14:33:30 2019 -0700\n\n power: supply: Init device wakeup after device_add()\n \n We may want to use the device pointer in device_init_wakeup() with\n functions that expect the device to already be added with device_add().\n For example, if we were to link the device initializing wakeup to\n something in sysfs such as a class for wakeups we\u0027ll run into an error.\n It looks like this code was written with the assumption that the device\n would be added before initializing wakeup due to the order of operations\n in power_supply_unregister().\n \n Let\u0027s change the order of operations so we don\u0027t run into problems here.\n \n Fixes: 948dcf966228 (\"power_supply: Prevent suspend until power supply events are processed\")\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Tri Vo \u003ctrong@android.com\u003e\n Cc: Kalesh Singh \u003ckaleshsingh@google.com\u003e\n Cc: Ravi Chandra Sadineni \u003cravisadineni@chromium.org\u003e\n Cc: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Acked-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit 10948061162a4bd9c54086e04dc8cd11faaf5bb5\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Aug 5 14:44:18 2019 -0500\n\n power: supply: ab8500_charger: Mark expected switch fall-through\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warning (Building: allmodconfig arm):\n \n drivers/power/supply/ab8500_charger.c:738:6: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit fe55e770327363304c4111423e6f7ff3c650136d\nAuthor: Michael Nosthoff \u003ccommitted@heine.so\u003e\nDate: Fri Aug 16 09:58:42 2019 +0200\n\n power: supply: sbs-battery: only return health when battery present\n \n when the battery is set to sbs-mode and no gpio detection is enabled\n \"health\" is always returning a value even when the battery is not present.\n All other fields return \"not present\".\n This leads to a scenario where the driver is constantly switching between\n \"present\" and \"not present\" state. This generates a lot of constant\n traffic on the i2c.\n \n This commit changes the response of \"health\" to an error when the battery\n is not responding leading to a consistent \"not present\" state.\n \n Fixes: 76b16f4cdfb8 (\"power: supply: sbs-battery: don\u0027t assume MANUFACTURER_DATA formats\")\n Cc: \u003cstable@vger.kernel.org\u003e\n Signed-off-by: Michael Nosthoff \u003ccommitted@heine.so\u003e\n Reviewed-by: Brian Norris \u003cbriannorris@chromium.org\u003e\n Tested-by: Brian Norris \u003cbriannorris@chromium.org\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit 47c169e17bbd97a0dce44676ac10cbd317c35a11\nAuthor: Denis Efremov \u003cefremov@linux.com\u003e\nDate: Tue Aug 13 09:13:58 2019 +0300\n\n MAINTAINERS: N900: Remove isp1704_charger.h record\n \n Update MAINTAINERS to reflect that isp1704_charger.h file was removed.\n \n Cc: Pavel Machek \u003cpavel@ucw.cz\u003e\n Cc: linux-pm@vger.kernel.org\n Fixes: f5d782d46aa5 (\"power: supply: isp1704: switch to gpiod API\")\n Signed-off-by: Denis Efremov \u003cefremov@linux.com\u003e\n Reviewed-by: Pali Rohár \u003cpali.rohar@gmail.com\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit 99956a9e08251a1234434b492875b1eaff502a12\nAuthor: Michael Nosthoff \u003ccommitted@heine.so\u003e\nDate: Fri Aug 16 09:37:42 2019 +0200\n\n power: supply: sbs-battery: use correct flags field\n \n the type flag is stored in the chip-\u003eflags field not in the\n client-\u003eflags field. This currently leads to never using the ti\n specific health function as client-\u003eflags doesn\u0027t use that bit.\n So it\u0027s always falling back to the general one.\n \n Fixes: 76b16f4cdfb8 (\"power: supply: sbs-battery: don\u0027t assume MANUFACTURER_DATA formats\")\n Cc: \u003cstable@vger.kernel.org\u003e\n Signed-off-by: Michael Nosthoff \u003ccommitted@heine.so\u003e\n Reviewed-by: Brian Norris \u003cbriannorris@chromium.org\u003e\n Reviewed-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit 0f1979b402df5f0dd86425830ddaa191d70f3655\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Fri Aug 30 18:12:35 2019 +0200\n\n m68k: Remove ioremap_fullcache()\n \n No callers of this function.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lore.kernel.org/r/20190830161237.23033-2-hch@lst.de\n Signed-off-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n\ncommit 2cecd1f11ca4a7d402bee4bb12198e5283a4fb81\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Sat Aug 17 09:32:35 2019 +0200\n\n m68k: Simplify ioremap_nocache()\n \n Just define ioremap_nocache to ioremap instead of duplicating the\n inline. Also define ioremap_uc in terms of ioremap instead of\n using a double indirection.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lore.kernel.org/r/20190817073253.27819-9-hch@lst.de\n Signed-off-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n\ncommit 29d9a0b50736768f042752070e5cdf4e4d4c00df\nAuthor: John S. Gruber \u003cJohnSGruber@gmail.com\u003e\nDate: Mon Sep 2 00:00:54 2019 +0200\n\n x86/boot: Preserve boot_params.secure_boot from sanitizing\n \n Commit\n \n a90118c445cc (\"x86/boot: Save fields explicitly, zero out everything else\")\n \n now zeroes the secure boot setting information (enabled/disabled/...)\n passed by the boot loader or by the kernel\u0027s EFI handover mechanism.\n \n The problem manifests itself with signed kernels using the EFI handoff\n protocol with grub and the kernel loses the information whether secure\n boot is enabled in the firmware, i.e., the log message \"Secure boot\n enabled\" becomes \"Secure boot could not be determined\".\n \n efi_main() arch/x86/boot/compressed/eboot.c sets this field early but it\n is subsequently zeroed by the above referenced commit.\n \n Include boot_params.secure_boot in the preserve field list.\n \n [ bp: restructure commit message and massage. ]\n \n Fixes: a90118c445cc (\"x86/boot: Save fields explicitly, zero out everything else\")\n Signed-off-by: John S. Gruber \u003cJohnSGruber@gmail.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Reviewed-by: John Hubbard \u003cjhubbard@nvidia.com\u003e\n Cc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: Juergen Gross \u003cjgross@suse.com\u003e\n Cc: Mark Brown \u003cbroonie@kernel.org\u003e\n Cc: stable \u003cstable@vger.kernel.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: x86-ml \u003cx86@kernel.org\u003e\n Link: https://lkml.kernel.org/r/CAPotdmSPExAuQcy9iAHqX3js_fc4mMLQOTr5RBGvizyCOPcTQQ@mail.gmail.com\n\ncommit fc603294267f890b34684588f6980ea8c7b72ecf\nAuthor: Mark Bloch \u003cmarkb@mellanox.com\u003e\nDate: Thu Aug 29 23:42:38 2019 +0000\n\n net/mlx5: Set only stag for match untagged packets\n \n cvlan_tag enabled in match criteria and disabled in\n match value means both S \u0026 C tags don\u0027t exist (untagged of both).\n \n Signed-off-by: Mark Bloch \u003cmarkb@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit f813cb506b8c9abc106314e1f95c9f2ebf260988\nAuthor: Maor Gottlieb \u003cmaorg@mellanox.com\u003e\nDate: Thu Aug 29 23:42:36 2019 +0000\n\n net/mlx5: Add stub for mlx5_eswitch_mode\n \n Return MLX5_ESWITCH_NONE when CONFIG_MLX5_ESWITCH\n is not selected.\n \n Signed-off-by: Maor Gottlieb \u003cmaorg@mellanox.com\u003e\n Reviewed-by: Mark Bloch \u003cmarkb@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 3a6ef5158d4a01a2ddd88c3c978e6de0d10ee759\nAuthor: Maor Gottlieb \u003cmaorg@mellanox.com\u003e\nDate: Thu Aug 29 23:42:34 2019 +0000\n\n net/mlx5: Avoid disabling RoCE when uninitialized\n \n Move the check if RoCE steering is initialized to the\n disable RoCE function, it will ensure that we disable\n RoCE only if we succeeded in enabling it before.\n \n Fixes: 80f09dfc237f (\"net/mlx5: Eswitch, enable RoCE loopback traffic\")\n Signed-off-by: Maor Gottlieb \u003cmaorg@mellanox.com\u003e\n Reviewed-by: Mark Bloch \u003cmarkb@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 97b5484ed608605dca9e461c0289d1195ba9608d\nAuthor: Alex Vesker \u003cvalex@mellanox.com\u003e\nDate: Thu Aug 29 23:42:32 2019 +0000\n\n net/mlx5: Add HW bits and definitions required for SW steering\n \n Add the required Software Steering hardware definitions and\n bits to mlx5_ifc.\n \n Signed-off-by: Alex Vesker \u003cvalex@mellanox.com\u003e\n Signed-off-by: Yevgeny Klitenik \u003ckliten@mellanox.com\u003e\n Reviewed-by: Mark Bloch \u003cmarkb@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit c9b9dcb430b3cd0ad2b04c360c4e528d73430481\nAuthor: Ariel Levkovich \u003clariel@mellanox.com\u003e\nDate: Thu Aug 29 23:42:30 2019 +0000\n\n net/mlx5: Move device memory management to mlx5_core\n \n Move the device memory allocation and deallocation commands\n SW ICM memory to mlx5_core to expose this API for all\n mlx5_core users.\n \n This comes as preparation for supporting SW steering in kernel\n where it will be required to allocate and register device\n memory for direct rule insertion.\n \n In addition, an API to register this device memory for future\n remote access operations is introduced using the create_mkey\n commands.\n \n Signed-off-by: Ariel Levkovich \u003clariel@mellanox.com\u003e\n Reviewed-by: Mark Bloch \u003cmarkb@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 65efb20db59e797d7401f8adcb6d34ff06c5b7cd\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Sat Jul 27 16:29:22 2019 -0400\n\n devpts_pty_kill(): don\u0027t bother with d_delete()\n \n we are not retaining dentries there anyway (simple_dentry_operations),\n so d_delete()+dput() \u003d\u003d d_drop()+dput()\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 28cd4af104d6a9ff7ad380b79a8e39e165a25398\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Sat Jul 27 16:14:37 2019 -0400\n\n infiniband: don\u0027t bother with d_delete()\n \n Dentries are never retained there; d_delete() + dput() is no\n different from d_drop() + dput().\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 1916c61779fb05aaac92261863136ce6c20d16a0\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Sat Jul 27 16:10:56 2019 -0400\n\n hypfs: don\u0027t bother with d_delete()\n \n If that\u0027s not the last reference, d_delete() will do d_drop().\n If it is, dput() immediately after it will unhash the sucker\n anyway, since -\u003ed_delete() the method is always_delete_dentry().\n \n IOW, there\u0027s no point trying to turn it into a negative hashed\n dentry - it won\u0027t stick around anyway. Just d_drop() it and be\n done with that.\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit dc7857ad37119fd3b4e50e6b371bc856557976f0\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Tue Jul 16 21:20:17 2019 -0400\n\n fs/namei.c: keep track of nd-\u003eroot refcount status\n \n The rules for nd-\u003eroot are messy:\n * if we have LOOKUP_ROOT, it doesn\u0027t contribute to refcounts\n * if we have LOOKUP_RCU, it doesn\u0027t contribute to refcounts\n * if nd-\u003eroot.mnt is NULL, it doesn\u0027t contribute to refcounts\n * otherwise it does contribute\n \n terminate_walk() needs to drop the references if they are contributing.\n So everything else should be careful not to confuse it, leading to\n rather convoluted code.\n \n It\u0027s easier to keep track of whether we\u0027d grabbed the reference(s)\n explicitly. Use a new flag for that. Don\u0027t bother with zeroing\n nd-\u003eroot.mnt on unlazy failures and in terminate_walk - it\u0027s not\n needed anymore (terminate_walk() won\u0027t care and the next path_init()\n will zero nd-\u003eroot in !LOOKUP_ROOT case anyway).\n \n Resulting rules for nd-\u003eroot refcounts are much simpler: they are\n contributing iff LOOKUP_ROOT_GRABBED is set in nd-\u003eflags.\n\ncommit 9b87647c665dbf93173ca2f43986902b59dfbbba\nAuthor: Denis Efremov \u003cefremov@linux.com\u003e\nDate: Thu Aug 29 00:09:34 2019 +0300\n\n asm-generic: add unlikely to default BUG_ON(x)\n \n Add unlikely to default BUG_ON(x) in !CONFIG_BUG. It makes\n the define consistent with BUG_ON(x) in CONFIG_BUG.\n \n Signed-off-by: Denis Efremov \u003cefremov@linux.com\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: \u003clinux-arch@vger.kernel.org\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 602828c1aade576ac5f3fbd59b4eb014c5fc2414\nAuthor: Nicolas Pitre \u003cnico@fluxnic.net\u003e\nDate: Tue Aug 20 23:05:42 2019 -0400\n\n __div64_const32(): improve the generic C version\n \n Let\u0027s rework that code to avoid large immediate values and convert some\n 64-bit variables to 32-bit ones when possible. This allows gcc to\n produce smaller and better code. This even produces optimal code on\n RISC-V.\n \n Signed-off-by: Nicolas Pitre \u003cnico@fluxnic.net\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit fa7da7449eb3172e9ef56ae69a1db237cb26c8df\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Sun Aug 25 17:41:52 2019 +0200\n\n power: supply: axp288_fuel_gauge: Add Minix Neo Z83-4 to the blacklist\n \n The Minix Neo Z83-4 is another mini PC using the AXP288 PMIC where the\n EFI code does not disable the charger part of the PMIC causing us to report\n battery readings (of always 100%) to userspace even though there is no\n battery in this wall-outlet powered device.\n \n Add it to the blacklist to avoid the bogus battery status reporting.\n \n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit 6f3ed834717b9fcfb4993afea231937422b8ad10\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Sun Aug 25 17:41:51 2019 +0200\n\n power: supply: axp288_fuel_gauge: Sort the DMI blacklist alphabetically\n \n The blacklist is getting big enough that it is good to have some sort\n of fixed order for it, sort it alphabetically.\n \n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit f1b937cc86bedf94dbc3478c2c0dc79471081fff\nAuthor: Romain Izard \u003cromain.izard.pro@gmail.com\u003e\nDate: Fri Aug 30 15:15:56 2019 +0200\n\n power: supply: register HWMON devices with valid names\n \n With the introduction of the HWMON compatibility layer to the power\n supply framework in Linux 5.3, all power supply devices\u0027 names can be\n used directly to create HWMON devices with the same names.\n \n But HWMON has rules on allowable names that are different from those\n used in the power supply framework. The dash character is forbidden, as\n it is used by the libsensors library in userspace as a separator,\n whereas this character is used in the device names in more than half of\n the existing power supply drivers. This last case is consistent with the\n typical naming usage with MFD and Device Tree.\n \n This leads to warnings in the kernel log, with the format:\n \n power_supply gpio-charger: hwmon: \\\n \u0027gpio-charger\u0027 is not a valid name attribute, please fix\n \n Add a protection to power_supply_add_hwmon_sysfs() that replaces any\n dash in the device name with an underscore when registering with the\n HWMON framework. Other forbidden characters (star, slash, space, tab,\n newline) are not replaced, as they are not in common use.\n \n Fixes: e67d4dfc9ff1 (\"power: supply: Add HWMON compatibility layer\")\n Signed-off-by: Romain Izard \u003cromain.izard.pro@gmail.com\u003e\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit 45de77ff8286156f78ac2beef03f6088bb3e451b\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Sat Aug 31 16:18:36 2019 -0400\n\n net: dsa: mv88e6xxx: centralize SERDES IRQ handling\n \n The .serdes_irq_setup are all following the same steps: get the SERDES\n lane, get the IRQ mapping, request the IRQ, then enable it. So do\n the .serdes_irq_free implementations: get the SERDES lane, disable\n the IRQ, then free it.\n \n This patch removes these operations in favor of generic functions.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 907b9b9fcaef7fb193105d9d61cd795b38ee2530\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Sat Aug 31 16:18:35 2019 -0400\n\n net: dsa: mv88e6xxx: introduce .serdes_irq_status\n \n Introduce a new .serdes_irq_status operation to prepare the abstraction\n of IRQ thread from the SERDES IRQ setup code.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 61a46b4147b2767728eb0486be31e2e4b65dd3de\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Sat Aug 31 16:18:34 2019 -0400\n\n net: dsa: mv88e6xxx: introduce .serdes_irq_enable\n \n Introduce a new .serdes_irq_enable operation to prepare the abstraction\n of IRQ enabling from the SERDES IRQ setup code.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit dc272f600eb3c3453c1289b095c6749d3241ae81\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Sat Aug 31 16:18:33 2019 -0400\n\n net: dsa: mv88e6xxx: pass lane to .serdes_power\n \n Now the first step of all .serdes_power implementations is getting\n the lane mapping. Since we have an operation for that, call it in\n the wrapper and pass the lane down to the .serdes_power operation.\n \n This also allows to avoid querying the SERDES lane twice in\n mv88e6xxx_port_set_cmode.\n \n At the same time provide mv88e6xxx_serdes_power_{up,down} helpers\n and prefer up/down instead of on/off as in the documentation.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6600d8e5822d9b3f98c82db3e96df6895b1e87d8\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Sat Aug 31 16:18:32 2019 -0400\n\n net: dsa: mv88e6xxx: merge mv88e6352_serdes_power_set\n \n The mv88e6352_serdes_power_set helper is only used at one place, in\n mv88e6352_serdes_power. Keep it simple and merge the two functions\n together.\n \n Use mv88e6xxx_serdes_get_lane instead of mv88e6352_port_has_serdes\n to avoid moving code. No functional changes.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9db4a7257e601491cf8e0b7e7e1461f7cde294bb\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Sat Aug 31 16:18:31 2019 -0400\n\n net: dsa: mv88e6xxx: implement mv88e6352_serdes_get_lane\n \n Even though 88E6352 has no dedicated lane for SERDES interfaces, it\n uses a similar code as the other .serdes_get_lane implementations to\n check the port\u0027s CMODE and ensure that SERDES operations are doable.\n \n For consistency, implement mv88e6352_serdes_get_lane for the 88E6352\n and similar switches which simply returns an unused 0xff lane address.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 5122d4ec9e8053a5944bf77db6bd6c89143531d7\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Sat Aug 31 16:18:30 2019 -0400\n\n net: dsa: mv88e6xxx: simplify .serdes_get_lane\n \n Because the mapping between a SERDES interface and its lane is static,\n we don\u0027t need to stick with negative error codes actually and we can\n simply return 0 if there is no lane, just like the IRQ mapping.\n \n This way we can keep a simple and intuitive API using unsigned lane\n numbers while simplifying the implementations with single return\n statements. Last but not least, fix the reverse chrismas tree in\n mv88e6390x_serdes_get_lane.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4241ef52372ebee93f4ed67e08c9316ccb20bdd7\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Sat Aug 31 16:18:29 2019 -0400\n\n net: dsa: mv88e6xxx: introduce .serdes_irq_mapping\n \n Introduce a new .serdes_irq_mapping operation to prepare the\n abstraction of IRQ mapping from the SERDES IRQ setup code.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f441ed0f4c89c14dd68ce477f513bae43509237e\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Sat Aug 31 16:18:28 2019 -0400\n\n net: dsa: mv88e6xxx: fix SERDES IRQ mapping\n \n The current mv88e6xxx SERDES code checks for negative error code from\n irq_find_mapping, while this function returns an unsigned integer. This\n patch removes this dead code and simply returns 0 is no IRQ is found.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 946bc2509eb88ff0c64dd727a11c5a29b6bc2574\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Sat Aug 31 16:18:27 2019 -0400\n\n net: dsa: mv88e6xxx: check errors in mv88e6352_serdes_irq_link\n \n The mv88e6352_serdes_irq_link helper is not checking for any error that\n may occur during hardware accesses. Worst, the \"up\" boolean is set from\n the potentially unused \"status\" variable, if read operations failed.\n \n As done in mv88e6390_serdes_irq_link_sgmii, return right away and do\n not call dsa_port_phylink_mac_change if an error occurred.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b943e03341e69cfd384ceabbb4efc28a82f69e0c\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Aug 31 12:29:11 2019 +0000\n\n net: hns3: remove set but not used variable \u0027qos\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c: In function \u0027hclge_restore_vlan_table\u0027:\n drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c:8016:18: warning:\n variable \u0027qos\u0027 set but not used [-Wunused-but-set-variable]\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Fixes: 70a214903da9 (\"net: hns3: reduce the parameters of some functions\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit bdad7529ee4eca42c7034f34d2fe96de121aae34\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Sat Aug 31 08:29:49 2019 +0100\n\n net: hns3: remove redundant assignment to pointer reg_info\n \n Pointer reg_info is being initialized with a value that is never read and\n is being re-assigned a little later on. The assignment is redundant\n and hence can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 56fcd40f8a51db475389856f31685a1b87a3cc9c\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Sun Sep 1 16:52:05 2019 +0100\n\n netlabel: remove redundant assignment to pointer iter\n \n Pointer iter is being initialized with a value that is never read and\n is being re-assigned a little later on. The assignment is redundant\n and hence can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Acked-by: Paul Moore \u003cpaul@paul-moore.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit dc161162e42ca51daff50e86a4a4bb8395d60501\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Sun Sep 1 10:42:44 2019 +0200\n\n r8169: don\u0027t set bit RxVlan on RTL8125\n \n RTL8125 uses a different register for VLAN offloading config,\n therefore don\u0027t set bit RxVlan.\n \n Fixes: f1bce4ad2f1c (\"r8169: add support for RTL8125\")\n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7981dc1262f5e74cd0a5d5fe1abea3facff984e0\nAuthor: Jordan Crouse \u003cjcrouse@codeaurora.org\u003e\nDate: Tue Aug 13 08:46:24 2019 -0600\n\n drm/msm: Remove Kconfig default\n \n Remove the default for CONFIG_DRM_MSM and let the user select the driver\n manually as one does.\n \n Additionally select QCOM_COMMAND_DB for ARCH_QCOM targets to make sure\n it doesn\u0027t get missed when we need it for a6xx targets.\n \n v2: Move from default \u0027m\u0027 to no default\n \n Signed-off-by: Jordan Crouse \u003cjcrouse@codeaurora.org\u003e\n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n\ncommit f49475f07a067f8ddd6e744c9bb9953e53ce6c6b\nAuthor: Rob Clark \u003crobdclark@chromium.org\u003e\nDate: Tue Aug 27 14:10:09 2019 -0700\n\n drm/msm/dpu: remove stray \"\\n\"\n \n The extra line-break in traces was annoying me.\n \n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n Reviewed-by: Sean Paul \u003csean@poorly.run\u003e\n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n\ncommit 846e5a322bfbc106b4425e5baa9ca11ae90eb907\nAuthor: Rob Clark \u003crobdclark@chromium.org\u003e\nDate: Tue Aug 20 16:12:28 2019 -0700\n\n drm/msm/dpu: fix \"frame done\" timeouts\n \n Previously, dpu_crtc_frame_event_work() would try to aquire all the\n modeset locks in order to check whether it can release bandwidth. (If\n we only have cmd-mode display, bandwidth can be released at frame-done\n time.)\n \n The problem with this is that it is also responsible for signalling\n frame_done_comp, which dpu_crtc_commit_kickoff() waits on if there is\n already a frame pending. This is called in the msm_atomic_commit_tail()\n path.. which means that for non-nonblock commits, at least some of the\n modeset locks are already held.\n \n Re-work this scheme to use a reference count to track our need to have\n clocks enabled. It is incremented for each atomic commit, and\n decremented in the corresponding frame-done. Additionally, any crtc\n used in video mode hold an extra reference while they are enabled. The\n net effect is that we can determine in frame-done whether it is safe to\n drop bandwidth without needing to aquire any modeset locks.\n \n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n Reviewed-by: Sean Paul \u003csean@chromium.org\u003e\n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n\ncommit fb0f209861652f9fc687a055a695e349bd40cd57\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Aug 5 18:49:28 2019 -0500\n\n drm/msm: Use struct_size() helper\n \n One of the more common cases of allocation size calculations is finding\n the size of a structure that has a zero-sized array at the end, along\n with memory for some number of elements for that array. For example:\n \n struct msm_gem_submit {\n ...\n struct {\n ...\n } bos[0];\n };\n \n Make use of the struct_size() helper instead of an open-coded version\n in order to avoid any potential type mistakes.\n \n So, replace the following form:\n \n sizeof(*submit) + ((u64)nr_bos * sizeof(submit-\u003ebos[0]))\n \n with:\n \n struct_size(submit, bos, nr_bos)\n \n This code was detected with the help of Coccinelle.\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n\ncommit 653c3c73ce6445822b09a2db1981410a64c1d57c\nAuthor: Jordan Crouse \u003cjcrouse@codeaurora.org\u003e\nDate: Thu Jul 25 10:53:55 2019 -0600\n\n drm/msm: Use generic bulk clock function\n \n Remove the homebrewed bulk clock get function and replace it with\n devm_clk_bulk_get_all().\n \n Signed-off-by: Jordan Crouse \u003cjcrouse@codeaurora.org\u003e\n Reviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Reviewed-by: Rob Clark \u003crobdclark@gmail.com\u003e\n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n\ncommit 27cfa9de20d93ae3fff8985c3e6aca38f9bf092d\nAuthor: Jordan Crouse \u003cjcrouse@codeaurora.org\u003e\nDate: Thu Jul 25 10:53:08 2019 -0600\n\n drm/msm: Annotate intentional switch statement fall throughs\n \n Explicitly mark intentional fall throughs in switch statements to keep\n -Wimplicit-fallthrough from complaining.\n \n Signed-off-by: Jordan Crouse \u003cjcrouse@codeaurora.org\u003e\n Reviewed-by: Rob Clark \u003crobdclark@gmail.com\u003e\n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n\ncommit 6a3549ac2abed878f718e02066d5da46b9eaf3ff\nAuthor: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\nDate: Mon Jul 8 08:12:24 2019 -0700\n\n drm/msm/mdp5: Find correct node for creating gem address space\n \n Creating the msm gem address space requires a reference to the dev where\n the iommu is located. The driver currently assumes this is the same as\n the platform device, which breaks when the iommu is outside of the\n platform device (ie in the parent). Default to using the platform device,\n but check to see if that has an iommu reference, and if not, use the parent\n device instead. This should handle all the various iommu designs for\n mdp5 supported systems.\n \n Signed-off-by: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\n Tested-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n\ncommit ef7d1ded86c51d078cc222c7eb55eb64396f14bf\nAuthor: Brian Masney \u003cmasneyb@onstation.org\u003e\nDate: Sat Jul 6 07:11:38 2019 -0400\n\n drm/msm/phy/dsi_phy: silence -EPROBE_DEFER warnings\n \n The following errors show up when booting the Nexus 5:\n \n msm_dsi_phy fd922a00.dsi-phy: [drm:dsi_phy_driver_probe] *ERROR*\n dsi_phy_regulator_init: failed to init regulator, ret\u003d-517\n msm_dsi_phy fd922a00.dsi-phy: [drm:dsi_phy_driver_probe] *ERROR*\n dsi_phy_driver_probe: failed to init regulator\n \n dsi_phy_regulator_init() already logs the error, so no need to log\n the same error a second time in dsi_phy_driver_probe(). This patch\n also changes dsi_phy_regulator_init() to not log the error if the\n error code is -EPROBE_DEFER to reduce noise in dmesg.\n \n Signed-off-by: Brian Masney \u003cmasneyb@onstation.org\u003e\n Reviewed-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n [add some {}\u0027s]\n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n\ncommit 3fa63b8e6f5364fa7c4765fc0b3ded8eeb73ee26\nAuthor: Rob Clark \u003crobdclark@chromium.org\u003e\nDate: Wed Jul 3 07:00:35 2019 -0700\n\n drm/msm/a6xx: add missing MODULE_FIRMWARE()\n \n For platforms that require the \"zap shader\" to take the GPU out of\n secure mode at boot, we also need the zap fw to end up in the initrd.\n \n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n Reviewed-by: Jordan Crouse \u003cjcrouse@codeaurora.org\u003e\n Reviewed-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n\ncommit 6b0c4d2539da4f87004073c3bddb843d6334570e\nAuthor: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\nDate: Mon Jul 1 10:45:06 2019 -0700\n\n drm/msm/mdp5: Add msm8998 support\n \n Add support for MDP5 version v3.0 found on msm8998.\n \n Signed-off-by: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\n Reviewed-by: Rob Clark \u003crobdclark@gmail.com\u003e\n [silence unitialized variable warnings]\n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n\ncommit 57bd739ffe7a89aaaf1e33e01e2e9f0f559e8679\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Tue Aug 20 20:33:37 2019 -0500\n\n usb: udc: lpc32xx: Silence fall-through warning\n \n Silence the following fall-through warning by adding a break statement:\n \n drivers/usb/gadget/udc/lpc32xx_udc.c:2230:3: warning: this statement may\n fall through [-Wimplicit-fallthrough\u003d]\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n\ncommit 35de8ef771145e06118299ee311793fd02ebc404\nAuthor: Rob Clark \u003crobdclark@chromium.org\u003e\nDate: Sun Jun 30 06:14:42 2019 -0700\n\n drm/msm/dpu: remove dpu_mdss:hwversion\n \n Unused and the extra rpm get/put interferes with handover from\n bootloader (ie. happens before we have a chance to check if\n things are already enabled).\n \n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n Reviewed-by: Jordan Crouse \u003cjcrouse@codeaurora.org\u003e\n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n\ncommit 82e969d12c61995374ff753924aee50e698f3a2a\nAuthor: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\nDate: Wed Jun 26 11:00:15 2019 -0700\n\n drm: msm: Fix add_gpu_components\n \n add_gpu_components() adds found GPU nodes from the DT to the match list,\n regardless of the status of the nodes. This is a problem, because if the\n nodes are disabled, they should not be on the match list because they will\n not be matched. This prevents display from initing if a GPU node is\n defined, but it\u0027s status is disabled.\n \n Fix this by checking the node\u0027s status before adding it to the match list.\n \n Fixes: dc3ea265b856 (\"drm/msm: Drop the gpu binding\")\n Signed-off-by: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\n Reviewed-by: Rob Clark \u003crobdclark@gmail.com\u003e\n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n\ncommit 82237df9cb40db4dd5bd6c549c48cce8862827d1\nAuthor: Brian Masney \u003cmasneyb@onstation.org\u003e\nDate: Wed Jun 26 22:05:15 2019 -0400\n\n drm/msm: correct NULL pointer dereference in context_init\n \n Correct attempted NULL pointer dereference in context_init() when\n running without an IOMMU.\n \n Signed-off-by: Brian Masney \u003cmasneyb@onstation.org\u003e\n Fixes: 295b22ae596c (\"drm/msm: Pass the MMU domain index in struct msm_file_private\")\n Reviewed-by: Rob Clark \u003crobdclark@gmail.com\u003e\n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n\ncommit 1d1ad57393cdb58a0f960fcc53170bfe23fdb643\nAuthor: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\nDate: Fri Jun 28 09:28:31 2019 -0700\n\n drm/msm: Transition console to msm framebuffer\n \n If booting a device using EFI, efifb will likely come up and claim the\n console. When the msm display stack finally comes up, we want the\n console to move over to the msm fb, so add support to kick out any\n firmware based framebuffers to accomplish the console transition.\n \n Suggested-by: Rob Clark \u003crobdclark@gmail.com\u003e\n Signed-off-by: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\n Reviewed-by: Rob Clark \u003crobdclark@gmail.com\u003e\n Tested-by: Rob Clark \u003crobdclark@gmail.com\u003e\n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n\ncommit 219310a50a2690df96f4823bff2957a1c7e24385\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Sat Jun 29 14:59:31 2019 +0200\n\n drm/msm/hdmi: Convert to use GPIO descriptors\n \n This switches the MSM HDMI code to use GPIO descriptors.\n Normally we would fetch the GPIOs from the device with the\n flags GPIOD_IN or GPIOD_OUT_[LOW|HIGH] to set up the lines\n immediately, but since the code seems eager to actively\n drive the lines high/low when turning HDMI on and off, we\n just fetch the GPIOs as-is and keep the code explicitly\n driving them.\n \n The old code would try legacy bindings (GPIOs without any\n \"-gpios\" suffix) but this has been moved to the gpiolib\n as a quirk by the previous patch.\n \n Cc: Rob Clark \u003crobdclark@gmail.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: linux-arm-msm@vger.kernel.org\n Cc: freedreno@lists.freedesktop.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n\ncommit 9a31616d8e43e617ec77830a88b026a471cd0fbe\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Sat Jun 29 14:59:29 2019 +0200\n\n drm/msm/dpu: Drop unused GPIO code\n \n The DPU has some kind of idea that it wants to be able to\n bring up power using GPIO lines. The struct dss_gpio is however\n completely unused and should this be done, it should be done\n using the GPIO descriptor framework rather than this API\n which relies on the global GPIO numberspace. Delete this\n code before anyone hurt themselves.\n \n The inclusion of \u003clinux/gpio.h\u003e was abused to get some OF\n and IRQ headers implicitly included into the DPU utilities,\n make these includes explicit and push them down into the actual\n implementation.\n \n Cc: Rob Clark \u003crobdclark@gmail.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: linux-arm-msm@vger.kernel.org\n Cc: freedreno@lists.freedesktop.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n\ncommit 8c0a2af8d564f36fd2964998962963bb0261f421\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Sat Jun 29 14:59:28 2019 +0200\n\n drm/msm/dsi: Drop unused GPIO includes\n \n This DSI driver uses the new descriptor API so these old\n GPIO API includes are surplus.\n \n Cc: Rob Clark \u003crobdclark@gmail.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: linux-arm-msm@vger.kernel.org\n Cc: freedreno@lists.freedesktop.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n\ncommit f10bbfe6a470f74593ebdcb4008cb7d32b5394b4\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Sat Jun 29 14:59:27 2019 +0200\n\n drm/msm/mdp4: Drop unused GPIO include\n \n This file is not using any symbols from \u003clinux/gpio.h\u003e so just\n drop this include.\n \n Cc: Rob Clark \u003crobdclark@gmail.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: linux-arm-msm@vger.kernel.org\n Cc: freedreno@lists.freedesktop.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Reviewed-by: Brian Masney \u003cmasneyb@onstation.org\u003e\n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n\ncommit 079674a0a34502f96cd0c26f9439ecdd8474a78c\nAuthor: Shubhashree Dhar \u003cdhar@codeaurora.org\u003e\nDate: Mon Jun 24 11:57:12 2019 +0530\n\n drm/msm/dpu: Correct dpu encoder spinlock initialization\n \n dpu encoder spinlock should be initialized during dpu encoder\n init instead of dpu encoder setup which is part of commit.\n There are chances that vblank control uses the uninitialized\n spinlock if not initialized during encoder init.\n \n Change-Id: I5a18b95fa47397c834a266b22abf33a517b03a4e\n Signed-off-by: Shubhashree Dhar \u003cdhar@codeaurora.org\u003e\n Reviewed-by: Jeykumar Sankaran \u003cjsanka@codeaurora.org\u003e\n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n\ncommit 4e4210402639f16dfbfa5ba277aeb88fff401cec\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Fri Aug 30 18:08:19 2019 +0300\n\n leds: trigger: gpio: GPIO 0 is valid\n \n Allow all valid GPIOs to be used in the driver.\n \n Fixes: 17354bfe8527 (\"leds: Add gpio-led trigger\")\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit d4b02200ae7cc5886be01c0d7f1d335f4b6966a7\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Thu Aug 29 19:53:20 2019 -0500\n\n leds: pwm: Use struct_size() helper\n \n One of the more common cases of allocation size calculations is finding\n the size of a structure that has a zero-sized array at the end, along\n with memory for some number of elements for that array. For example:\n \n struct led_pwm_priv {\n ...\n struct led_pwm_data leds[0];\n };\n \n Make use of the struct_size() helper instead of an open-coded version\n in order to avoid any potential type mistakes.\n \n So, replace the following function:\n \n static inline size_t sizeof_pwm_leds_priv(int num_leds)\n {\n return sizeof(struct led_pwm_priv) +\n (sizeof(struct led_pwm_data) * num_leds);\n }\n \n with:\n \n struct_size(priv, leds, count)\n \n This code was detected with the help of Coccinelle.\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 1669ec78b87c0f04c963a5fc0885505d97468c7a\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Fri Aug 30 13:14:48 2019 -0500\n\n leds: is31fl32xx: Use struct_size() helper\n \n One of the more common cases of allocation size calculations is finding\n the size of a structure that has a zero-sized array at the end, along\n with memory for some number of elements for that array. For example:\n \n struct is31fl32xx_priv {\n ...\n struct is31fl32xx_led_data leds[0];\n };\n \n Make use of the struct_size() helper instead of an open-coded version\n in order to avoid any potential type mistakes.\n \n So, replace the following function:\n \n static inline size_t sizeof_is31fl32xx_priv(int num_leds)\n {\n return sizeof(struct is31fl32xx_priv) +\n (sizeof(struct is31fl32xx_led_data) * num_leds);\n }\n \n with:\n \n struct_size(priv, leds, count)\n \n This code was detected with the help of Coccinelle.\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 2637fd436c60c599e0967faee1903cce58c4a7f9\nAuthor: Dan Murphy \u003cdmurphy@ti.com\u003e\nDate: Wed Aug 28 10:22:19 2019 -0500\n\n leds: ti-lmu-common: Fix coccinelle issue in TI LMU\n \n Fix the coccinelle issues found in the TI LMU common code\n \n drivers/leds/leds-ti-lmu-common.c:97:20-29: WARNING: Unsigned expression compared with zero: ramp_down \u003c 0\n drivers/leds/leds-ti-lmu-common.c:97:5-12: WARNING: Unsigned expression compared with zero: ramp_up \u003c 0\n \n Fixes: 3fce8e1eb994 (\"leds: TI LMU: Add common code for TI LMU devices\")\n Signed-off-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 070a0eed7580a976b397dce981db9746df714327\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Tue Aug 27 14:52:05 2019 -0700\n\n leds: lm3532: Avoid potentially unpaired regulator calls\n \n We may currently get unpaired regulator calls when configuring the LED\n brightness via sysfs in case of regulator calls producing errors. Let\u0027s\n fix this by maintaining local state for enabled.\n \n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 68bfcf6debd72121758d6e0fbf75f506126d7b9a\nAuthor: Alexander Dahl \u003cada@thorsis.com\u003e\nDate: Tue Aug 27 15:00:27 2019 +0200\n\n leds: syscon: Use resource managed variant of device register\n \n We have a MFD driver compiled as module instantiating this driver. When\n unloading that module, those LED devices are not removed, which produces\n conflicts, when that module is inserted again.\n \n Signed-off-by: Alexander Dahl \u003cada@thorsis.com\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 7c322056e3564da1b5bdc3f3cb79229582955eb2\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Mon Aug 26 22:34:42 2019 +0200\n\n leds: Replace {devm_}led_classdev_register() macros with inlines\n \n Replace preprocessor macro aliases for legacy LED registration helpers\n with inline functions. It will allow to avoid misleading compiler error\n messages about missing symbol that actually wasn\u0027t explicitly used\n in the code. It used to occur when CONFIG_LEDS_CLASS was undefined\n and legacy (non-ext) function had been used in the code.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n\ncommit 4a9eb8154ffd26a56121a4a531b04cf32942a49b\nAuthor: Bastian Krause \u003cbst@pengutronix.de\u003e\nDate: Fri Aug 30 15:47:55 2019 +0200\n\n dt-bindings: rtc: ds1307: add rx8130 compatible\n \n The required OF entries were added with commit 47dd47292828\n (\"rtc: ds1307: add OF and ACPI entries for Epson RX8130\").\n \n Cc: devicetree@vger.kernel.org\n Signed-off-by: Bastian Krause \u003cbst@pengutronix.de\u003e\n Link: https://lore.kernel.org/r/20190830134755.1121-1-bst@pengutronix.de\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit d76a81d0c262154ed5ec0307bdd195f2758ccc70\nAuthor: Alejandro González \u003calejandro.gonzalez.correo@gmail.com\u003e\nDate: Wed Aug 21 23:00:56 2019 +0200\n\n rtc: sun6i: Allow using as wakeup source from suspend\n \n This patch allows userspace to set up wakeup alarms on any RTC handled by the\n sun6i driver, and adds the necessary PM operations to allow resuming from\n suspend when the configured wakeup alarm fires a IRQ. Of course, that the\n device actually resumes depends on the suspend state and how a particular\n hardware reacts to it, but that is out of scope for this patch.\n \n I\u0027ve tested these changes on a Pine H64 model B, which contains a\n Allwinner H6 SoC, with the help of CONFIG_PM_TEST_SUSPEND kernel option.\n These are the interesting outputs from the kernel and commands which\n show that it works. As every RTC handled by this driver is largely the\n same, I think that it shouldn\u0027t introduce any regression on other SoCs,\n but I may be wrong.\n \n [ 1.092705] PM: test RTC wakeup from \u0027freeze\u0027 suspend\n [ 1.098230] PM: suspend entry (s2idle)\n [ 1.212907] PM: suspend devices took 0.080 seconds\n (The SoC freezes for some seconds)\n [ 3.197604] PM: resume devices took 0.104 seconds\n [ 3.215937] PM: suspend exit\n \n [ 1.092812] PM: test RTC wakeup from \u0027mem\u0027 suspend\n [ 1.098089] PM: suspend entry (deep)\n [ 1.102033] PM: suspend exit\n [ 1.105205] PM: suspend test failed, error -22\n \n In any case, the RTC alarm interrupt gets fired as exptected:\n \n $ echo +5 \u003e /sys/class/rtc/rtc0/wakealarm \u0026\u0026 sleep 5 \u0026\u0026 grep rtc /proc/interrupts\n 29: 1 0 0 0 GICv2 133 Level 7000000.rtc\n \n Signed-off-by: Alejandro González \u003calejandro.gonzalez.correo@gmail.com\u003e\n Acked-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\n Link: https://lore.kernel.org/r/20190821210056.11995-1-alejandro.gonzalez.correo@gmail.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit f648d40b99ba7363bf4a1bbca5cc93904725dceb\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Thu Aug 29 23:25:47 2019 +0200\n\n rtc: pcf8563: let the core handle range offsetting\n \n Set the RTC range properly and use the core windowing and offsetting to\n (unfortunately) map back to a 1970-2069 range.\n \n Link: https://lore.kernel.org/r/20190829212547.19185-5-alexandre.belloni@bootlin.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 497dd8f203e6d5384f5d2bab0508db112f7e3665\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Tue Aug 27 18:17:55 2019 +0800\n\n f2fs: add missing documents of reserve_root/resuid/resgid\n \n Add missing documents.\n \n Fixes: 7e65be49ed94f (\"f2fs: add reserved blocks for root user\")\n Fixes: 7c2e59632b846 (\"f2fs: add resgid and resuid to reserve root blocks\")\n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 48aeaa459579d28ac4791a55c31ea1c8477d9259\nAuthor: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\nDate: Tue Aug 27 17:02:03 2019 -0700\n\n f2fs: convert inline_data in prior to i_size_write\n \n This can guarantee inline_data has smaller i_size.\n \n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 9530ba093d8de200e1edbd09a436e8f1b17639b2\nAuthor: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\nDate: Wed Aug 28 19:58:26 2019 -0700\n\n f2fs: fix flushing node pages when checkpoint is disabled\n \n This patch fixes skipping node page writes when checkpoint is disabled.\n In this period, we can\u0027t rely on checkpoint to flush node pages.\n \n Fixes: fd8c8caf7e7c (\"f2fs: let checkpoint flush dnode page of regular\")\n Fixes: 4354994f097d (\"f2fs: checkpoint disabling\")\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit ba4aa6103c448bedab3eb061d788106f9d8e713b\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Tue Aug 20 22:34:22 2019 +0800\n\n f2fs: fix to avoid corruption during inline conversion\n \n - f2fs_setattr\n - truncate_setsize (expand i_size)\n - f2fs_convert_inline_inode\n - f2fs_convert_inline_page\n - f2fs_reserve_block\n - f2fs_get_node_info failed\n \n Once we fail in above path, inline flag will remain, however\n - we\u0027ve reserved one block at inode.i_addr[0]\n - i_size has expanded\n \n Fix error path to avoid inode corruption.\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 0e762d7154bd903a0a97b20e56d1d582f76eb76c\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Fri Aug 23 17:58:36 2019 +0800\n\n f2fs: enhance f2fs_is_checkpoint_ready()\u0027s readability\n \n This patch changes sematics of f2fs_is_checkpoint_ready()\u0027s return\n value as: return true when checkpoint is ready, other return false,\n it can improve readability of below conditions.\n \n f2fs_submit_page_write()\n ...\n if (is_sbi_flag_set(sbi, SBI_IS_SHUTDOWN) ||\n !f2fs_is_checkpoint_ready(sbi))\n __submit_merged_bio(io);\n \n f2fs_balance_fs()\n ...\n if (!f2fs_is_checkpoint_ready(sbi))\n return;\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 56bc729da35441569664be89b8f5cdf732f23db0\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Fri Aug 23 17:58:35 2019 +0800\n\n f2fs: clean up __bio_alloc()\u0027s parameter\n \n Just cleanup, no logic change.\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit f88c760df45a9ae5466b4fb2b9aaf9be4444be65\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Fri Aug 23 17:58:34 2019 +0800\n\n f2fs: fix wrong error injection path in inc_valid_block_count()\n \n If FAULT_BLOCK type error injection is on, in inc_valid_block_count()\n we may decrease sbi-\u003ealloc_valid_block_count percpu stat count\n incorrectly, fix it.\n \n Fixes: 36b877af7992 (\"f2fs: Keep alloc_valid_block_count in sync\")\n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 5ba6c6f19f6efd349145199950b68ea154cf37a1\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Thu Aug 22 20:17:56 2019 +0800\n\n f2fs: fix to writeout dirty inode during node flush\n \n As Eric reported:\n \n On xfstest generic/204 on f2fs, I\u0027m getting a kernel BUG.\n \n allocate_segment_by_default+0x9d/0x100 [f2fs]\n f2fs_allocate_data_block+0x3c0/0x5c0 [f2fs]\n do_write_page+0x62/0x110 [f2fs]\n f2fs_do_write_node_page+0x2b/0xa0 [f2fs]\n __write_node_page+0x2ec/0x590 [f2fs]\n f2fs_sync_node_pages+0x756/0x7e0 [f2fs]\n block_operations+0x25b/0x350 [f2fs]\n f2fs_write_checkpoint+0x104/0x1150 [f2fs]\n f2fs_sync_fs+0xa2/0x120 [f2fs]\n f2fs_balance_fs_bg+0x33c/0x390 [f2fs]\n f2fs_write_node_pages+0x4c/0x1f0 [f2fs]\n do_writepages+0x1c/0x70\n __writeback_single_inode+0x45/0x320\n writeback_sb_inodes+0x273/0x5c0\n wb_writeback+0xff/0x2e0\n wb_workfn+0xa1/0x370\n process_one_work+0x138/0x350\n worker_thread+0x4d/0x3d0\n kthread+0x109/0x140\n \n The root cause of this issue is, in a very small partition, e.g.\n in generic/204 testcase of fstest suit, filesystem\u0027s free space\n is 50MB, so at most we can write 12800 inline inode with command:\n `echo XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX \u003e $SCRATCH_MNT/$i`,\n then filesystem will have:\n - 12800 dirty inline data page\n - 12800 dirty inode page\n - and 12800 dirty imeta (dirty inode)\n \n When we flush node-inode\u0027s page cache, we can also flush inline\n data with each inode page, however it will run out-of-free-space\n in device, then once it triggers checkpoint, there is no room for\n huge number of imeta, at this time, GC is useless, as there is no\n dirty segment at all.\n \n In order to fix this, we try to recognize inode page during\n node_inode\u0027s page flushing, and update inode page from dirty inode,\n so that later another imeta (dirty inode) flush can be avoided.\n \n Reported-and-tested-by: Eric Biggers \u003cebiggers@kernel.org\u003e\n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit fbce5d4ab3abc8033ea869771ce9306b144e7c3d\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Wed Aug 21 23:13:35 2019 +0800\n\n f2fs: optimize case-insensitive lookups\n \n This patch ports below casefold enhancement patch from ext4 to f2fs\n \n commit 3ae72562ad91 (\"ext4: optimize case-insensitive lookups\")\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit c7d5f6dbd9f9a68374f3d24c63a99691d792fd29\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Thu Aug 29 23:25:46 2019 +0200\n\n rtc: pcf8563: remove useless indirection\n \n pcf8563_rtc_read_time and pcf8563_set_datetime are only used after casting\n dev to an i2c_client. Remove that useless indirection.\n \n Link: https://lore.kernel.org/r/20190829212547.19185-4-alexandre.belloni@bootlin.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 8d3f805e6896798a1a40173d3734463363431b8e\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Thu Aug 29 23:25:45 2019 +0200\n\n rtc: pcf8563: convert to devm_rtc_allocate_device\n \n This allows further improvement of the driver.\n \n Link: https://lore.kernel.org/r/20190829212547.19185-3-alexandre.belloni@bootlin.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit deaa3ff4984f85fd8957841c314b94bbf81c9cd7\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Thu Aug 29 23:25:44 2019 +0200\n\n rtc: pcf8563: add Microcrystal RV8564 compatible\n \n Add a compatible string for the Microcrystal RV8564.\n \n Link: https://lore.kernel.org/r/20190829212547.19185-2-alexandre.belloni@bootlin.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit cd646ec003c5a24b1d86d1ba032d2f4eb3aa8e49\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Thu Aug 29 23:25:43 2019 +0200\n\n rtc: pcf8563: add Epson RTC8564 compatible\n \n Add a compatible string for the Epson RTC8564.\n \n Link: https://lore.kernel.org/r/20190829212547.19185-1-alexandre.belloni@bootlin.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 4053e74996b81902b2dd538717b4fd2fc1426d8b\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Fri Aug 30 15:31:24 2019 +0200\n\n rtc: s35390a: convert to devm_i2c_new_dummy_device()\n \n I was about to simplify the call to i2c_unregister_device() when I\n realized that converting to devm_i2c_new_dummy_device() will simplify\n the driver a lot. So I took this approach.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Link: https://lore.kernel.org/r/20190830133124.21633-3-wsa+renesas@sang-engineering.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 59a7f24fceb3aef6bd16f5aeb852c823dddab04d\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Fri Aug 30 15:31:23 2019 +0200\n\n rtc: max77686: convert to devm_i2c_new_dummy_device()\n \n I was about to simplify the call to i2c_unregister_device() when I\n realized that converting to devm_i2c_new_dummy_device() will simplify\n the driver a lot. So I took this approach.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Link: https://lore.kernel.org/r/20190830133124.21633-2-wsa+renesas@sang-engineering.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 6f671045b642b726f50124439235b6870c2d13ae\nAuthor: Ben Wei \u003cbenwei@fb.com\u003e\nDate: Fri Aug 30 20:50:51 2019 +0000\n\n net/ncsi: add response handlers for PLDM over NC-SI\n \n This patch adds handlers for PLDM over NC-SI command response.\n \n This enables NC-SI driver recognizes the packet type so the responses\n don\u0027t get dropped as unknown packet type.\n \n PLDM over NC-SI are not handled in kernel driver for now, but can be\n passed back to user space via Netlink for further handling.\n \n Signed-off-by: Ben Wei \u003cbenwei@fb.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7ef66122bdb3b839e9f51b76d7e600b6e21ef648\nAuthor: Biwen Li \u003cbiwen.li@nxp.com\u003e\nDate: Thu Aug 29 10:14:18 2019 +0800\n\n rtc: pcf85363/pcf85263: fix regmap error in set_time\n \n Issue:\n - # hwclock -w\n hwclock: RTC_SET_TIME: Invalid argument\n \n Why:\n - Relative commit: 8b9f9d4dc511 (\"regmap: verify if register is\n writeable before writing operations\"), this patch\n will always check for unwritable registers, it will compare reg\n with max_register in regmap_writeable.\n \n - The pcf85363/pcf85263 has the capability of address wrapping\n which means if you access an address outside the allowed range\n (0x00-0x2f) hardware actually wraps the access to a lower address.\n The rtc-pcf85363 driver will use this feature to configure the time\n and execute 2 actions in the same i2c write operation (stopping the\n clock and configure the time). However the driver has also\n configured the `regmap maxregister` protection mechanism that will\n block accessing addresses outside valid range (0x00-0x2f).\n \n How:\n - Split of writing regs to two parts, first part writes control\n registers about stop_enable and resets, second part writes\n RTC time and date registers.\n \n Signed-off-by: Biwen Li \u003cbiwen.li@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190829021418.4607-1-biwen.li@nxp.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit c59a9fc7272e56ba0575b9fcc346c5e4bf16b39d\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Wed Aug 28 22:50:56 2019 +0200\n\n rtc: snvs: switch to rtc_time64_to_tm/rtc_tm_to_time64\n \n Call the 64bit versions of rtc_tm time conversion now that the range is\n enforced by the core.\n \n Link: https://lore.kernel.org/r/20190828205056.31321-2-alexandre.belloni@bootlin.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 79610340cac8840ad67572ca13afae03265edae6\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Wed Aug 28 22:50:55 2019 +0200\n\n rtc: snvs: set range\n \n The SNVS is a 47-bit counter incremented at 32768Hz, it is then a 32-bit\n second counter.\n \n Link: https://lore.kernel.org/r/20190828205056.31321-1-alexandre.belloni@bootlin.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 58b6be4175a4b4a58210bf171ac90886244da98c\nAuthor: Parav Pandit \u003cparav@mellanox.com\u003e\nDate: Fri Aug 30 05:39:45 2019 -0500\n\n devlink: Use switch-case instead of if-else\n \n Make core more readable with switch-case for various port flavours.\n \n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Parav Pandit \u003cparav@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c7282b501f22a21ac8dfb6d3856aa1a92a7df5d5\nAuthor: Parav Pandit \u003cparav@mellanox.com\u003e\nDate: Fri Aug 30 05:39:44 2019 -0500\n\n devlink: Make port index data type as unsigned int\n \n Devlink port index attribute is returned to users as u32 through\n netlink response.\n Change index data type from \u0027unsigned\u0027 to \u0027unsigned int\u0027 to avoid\n below checkpatch.pl warning.\n \n WARNING: Prefer \u0027unsigned int\u0027 to bare use of \u0027unsigned\u0027\n 81: FILE: include/net/devlink.h:81:\n + unsigned index;\n \n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Parav Pandit \u003cparav@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 26811cc9f55acf835f7fdadc5ff2bbd6f06bc3ac\nAuthor: Davide Caratti \u003cdcaratti@redhat.com\u003e\nDate: Fri Aug 30 12:25:49 2019 +0200\n\n net: tls: export protocol version, cipher, tx_conf/rx_conf to socket diag\n \n When an application configures kernel TLS on top of a TCP socket, it\u0027s\n now possible for inet_diag_handler() to collect information regarding the\n protocol version, the cipher type and TX / RX configuration, in case\n INET_DIAG_INFO is requested.\n \n Signed-off-by: Davide Caratti \u003cdcaratti@redhat.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 61723b393292f1e4ea27f8d123384d50b176c29d\nAuthor: Davide Caratti \u003cdcaratti@redhat.com\u003e\nDate: Fri Aug 30 12:25:48 2019 +0200\n\n tcp: ulp: add functions to dump ulp-specific information\n \n currently, only getsockopt(TCP_ULP) can be invoked to know if a ULP is on\n top of a TCP socket. Extend idiag_get_aux() and idiag_get_aux_size(),\n introduced by commit b37e88407c1d (\"inet_diag: allow protocols to provide\n additional data\"), to report the ULP name and other information that can\n be made available by the ULP through optional functions.\n \n Users having CAP_NET_ADMIN privileges will then be able to retrieve this\n information through inet_diag_handler, if they specify INET_DIAG_INFO in\n the request.\n \n Signed-off-by: Davide Caratti \u003cdcaratti@redhat.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 15a7dea750e0162f273c6e61a94f96944b75b31e\nAuthor: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\nDate: Fri Aug 30 12:25:47 2019 +0200\n\n net/tls: use RCU protection on icsk-\u003eicsk_ulp_data\n \n We need to make sure context does not get freed while diag\n code is interrogating it. Free struct tls_context with\n kfree_rcu().\n \n We add the __rcu annotation directly in icsk, and cast it\n away in the datapath accessor. Presumably all ULPs will\n do a similar thing.\n \n Signed-off-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit fcee7de608a2e4cb3481fd224b11242427d8ec28\nAuthor: Marc Gonzalez \u003cmarc.w.gonzalez@free.fr\u003e\nDate: Mon Jun 24 15:26:05 2019 +0200\n\n arm64: defconfig: Enable CPU clock drivers for Qualcomm msm8916\n \n QCOM_A53PLL and QCOM_CLK_APCS_MSM8916 used to be enabled by default in\n drivers/clk/qcom/Kconfig. A recent patch changed that by dropping the\n \u0027default ARCH_QCOM\u0027 directive.\n \n Add the two options explicitly in the arm64 defconfig, to avoid\n functional regressions.\n \n Acked-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Signed-off-by: Marc Gonzalez \u003cmarc.w.gonzalez@free.fr\u003e\n [bjorn: Rewrote subject]\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 6377910c5c149f12874a0489963e7158c378917b\nAuthor: Jordan Crouse \u003cjcrouse@codeaurora.org\u003e\nDate: Tue Aug 13 08:46:25 2019 -0600\n\n arm64: defconfig: Add DRM_MSM to defconfigs with ARCH_QCOM\n \n Now that CONFIG_DRM_MSM is no longer default \u0027y\u0027 add it as a module to all\n ARCH_QCOM enabled defconfigs to restore the previous expected build\n behavior.\n \n I split this off from the original patch to separate out the ARM64 portions.\n \n Signed-off-by: Jordan Crouse \u003cjcrouse@codeaurora.org\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Signed-off-by: Andy Gross \u003cagross@kernel.org\u003e\n\ncommit 1cd3e52a07b86cf03079e05968d1a2641b6261c4\nAuthor: Andy Gross \u003cagross@kernel.org\u003e\nDate: Sun Sep 1 00:37:54 2019 -0500\n\n ARM: multi_v7_defconfig: Add DRM_MSM to defconfigs with ARCH_QCOM\n \n Now that CONFIG_DRM_MSM is no longer default \u0027y\u0027 add it as a module to all\n ARCH_QCOM enabled defconfigs to restore the previous expected build\n behavior.\n \n I split this off from the original patch to make this change unique to\n the multi_v7_defconfig\n \n Signed-off-by: Jordan Crouse \u003cjcrouse@codeaurora.org\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Signed-off-by: Andy Gross \u003cagross@kernel.org\u003e\n\ncommit 162eeae200e02ce8cc76a4d3cc8928da8145b20a\nAuthor: Jordan Crouse \u003cjcrouse@codeaurora.org\u003e\nDate: Tue Aug 13 08:46:25 2019 -0600\n\n ARM: qcom_defconfig: Add DRM_MSM to defconfigs with ARCH_QCOM\n \n Now that CONFIG_DRM_MSM is no longer default \u0027y\u0027 add it as a module to all\n ARCH_QCOM enabled defconfigs to restore the previous expected build\n behavior.\n \n Signed-off-by: Jordan Crouse \u003cjcrouse@codeaurora.org\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit ae31a514a134d9e4ca1d7b0f0a19b5934747d79f\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Sat Aug 31 17:35:53 2019 -0300\n\n objtool: Ignore intentional differences for the x86 insn decoder\n \n Since we need to build this in !x86, we need to explicitely use the x86\n files, not things like asm/insn.h, so we intentionally differ from the\n master copy in the kernel sources, add -I diff directives to ignore just\n these differences when checking for drift.\n \n Acked-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Link: http://lore.kernel.org/lkml/20190830193109.p7jagidsrahoa4pn@treble\n Acked-by: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/n/tip-j965m9b7xtdc83em3twfkh9o@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 2ffd84ae973b5ad16be96840574bb1142fda268a\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Sat Aug 31 17:29:47 2019 -0300\n\n objtool: Update sync-check.sh from perf\u0027s check-headers.sh\n \n To allow using the -I trick that will be needed for checking the x86\n insn decoder files.\n \n Without the specific -I lines we still get the same warnings as before:\n \n $ make -C tools/objtool/ clean ; make -C tools/objtool/\n make: Entering directory \u0027/home/acme/git/perf/tools/objtool\u0027\n CLEAN objtool\n find -name \u0027*.o\u0027 -delete -o -name \u0027\\.*.cmd\u0027 -delete -o -name \u0027\\.*.d\u0027 -delete\n rm -f arch/x86/inat-tables.c fixdep\n \u003cSNIP\u003e\n LD objtool-in.o\n make[1]: Leaving directory \u0027/home/acme/git/perf/tools/objtool\u0027\n Warning: Kernel ABI header at \u0027tools/arch/x86/include/asm/inat.h\u0027 differs from latest version at \u0027arch/x86/include/asm/inat.h\u0027\n diff -u tools/arch/x86/include/asm/inat.h arch/x86/include/asm/inat.h\n Warning: Kernel ABI header at \u0027tools/arch/x86/include/asm/insn.h\u0027 differs from latest version at \u0027arch/x86/include/asm/insn.h\u0027\n diff -u tools/arch/x86/include/asm/insn.h arch/x86/include/asm/insn.h\n Warning: Kernel ABI header at \u0027tools/arch/x86/lib/inat.c\u0027 differs from latest version at \u0027arch/x86/lib/inat.c\u0027\n diff -u tools/arch/x86/lib/inat.c arch/x86/lib/inat.c\n Warning: Kernel ABI header at \u0027tools/arch/x86/lib/insn.c\u0027 differs from latest version at \u0027arch/x86/lib/insn.c\u0027\n diff -u tools/arch/x86/lib/insn.c arch/x86/lib/insn.c\n /home/acme/git/perf/tools/objtool\n LINK objtool\n make: Leaving directory \u0027/home/acme/git/perf/tools/objtool\u0027\n $\n \n The next patch will add the -I lines for those files.\n \n Acked-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Link: http://lore.kernel.org/lkml/20190830193109.p7jagidsrahoa4pn@treble\n Acked-by: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/n/tip-vu3p38mnxlwd80rlsnjkqcf2@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 87a682a7c4e719d238d1839098375470b55e2097\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Sat Aug 31 17:14:19 2019 -0300\n\n perf build: Ignore intentional differences for the x86 insn decoder\n \n Since we need to build this in !x86, we need to explicitely use the x86\n files, not things like asm/insn.h, so we intentionally differ from the\n master copy in the kernel sources, add -I diff directives to ignore just\n these differences when checking for drift.\n \n Acked-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Acked-by: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/n/tip-9qziqjjt120mmz6kyepka9p7@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 00a263902ac3da886cf275663a938f503a853f68\nAuthor: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\nDate: Thu Aug 29 17:41:21 2019 -0500\n\n perf intel-pt: Use shared x86 insn decoder\n \n Now that there\u0027s a common version of the decoder for all tools, use it\n instead of the local copy.\n \n Also use perf\u0027s check-headers.sh script to diff the decoder files to\n make sure they remain in sync with the kernel version. Objtool has a\n similar check.\n \n Committer notes:\n \n Had to keep this all pointing explicitely to x86 headers/files, i.e.\n instead of asm/isnn.h we had to use ../include/asm/insn.h when the files\n were in differemt dirs, or just replace \"\u003casm/foo.h\u003e\" with \"foo.h\".\n \n This way we continue to be able to process perf.data files with Intel PT\n traces in distros other than x86.\n \n Also fixed up the awk script paths to use $(srcdir)/tools/arch instead\n or relative directories so that we keep detached tarballs (make help |\n grep perf) working.\n \n For now the include lines in these headers are being ignored so as not\n to flag false reports of kernel/tools out of sync.\n \n Signed-off-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Reviewed-by: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: x86@kernel.org\n Link: http://lore.kernel.org/lkml/8a37e615d2880f039505d693d1e068a009358a2b.1567118001.git.jpoimboe@redhat.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit f1da0a6c136542b9571b30af27bc1e84254f7a47\nAuthor: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\nDate: Thu Aug 29 17:41:20 2019 -0500\n\n perf intel-pt: Remove inat.c from build dependency list\n \n intel-pt-insn-decoder.c includes inat.c directly, so it already has an\n implicit dependency on inat.c. The Build file dependency is redundant.\n \n Signed-off-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Reviewed-by: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\n Acked-by: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: x86@kernel.org\n Link: http://lore.kernel.org/lkml/53776d6d29bc9eceb571d52df8fa32250c58a0f3.1567118001.git.jpoimboe@redhat.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 58993fb2c5115e93c52058e25f9b2ff289374870\nAuthor: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\nDate: Thu Aug 29 17:41:19 2019 -0500\n\n perf: Update .gitignore file\n \n After a \"make tools/perf\", git reports the following untracked files:\n \n tools/perf/feature/\n tools/perf/fixdep\n tools/perf/libtraceevent-dynamic-list\n \n Add these generated files to perf\u0027s .gitignore file.\n \n Signed-off-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Reviewed-by: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\n Acked-by: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: x86@kernel.org\n Link: http://lore.kernel.org/lkml/03acbc6c2fbc72054861f6c301875db75db33030.1567118001.git.jpoimboe@redhat.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit d046b725487a97a3a3b35a00e84ca093963b8b4e\nAuthor: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\nDate: Thu Aug 29 17:41:18 2019 -0500\n\n objtool: Move x86 insn decoder to a common location\n \n The kernel tree has three identical copies of the x86 instruction\n decoder. Two of them are in the tools subdir.\n \n The tools subdir is supposed to be completely standalone and separate\n from the kernel. So having at least one copy of the kernel decoder in\n the tools subdir is unavoidable. However, we don\u0027t need *two* of them.\n \n Move objtool\u0027s copy of the decoder to a shared location, so that perf\n will also be able to use it.\n \n Signed-off-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Reviewed-by: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: x86@kernel.org\n Link: http://lore.kernel.org/lkml/55b486b88f6bcd0c9a2a04b34f964860c8390ca8.1567118001.git.jpoimboe@redhat.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit f01642e4912bb80a01d693f4cc6fb0897207a090\nAuthor: Jin Yao \u003cyao.jin@linux.intel.com\u003e\nDate: Wed Aug 28 13:59:32 2019 +0800\n\n perf metricgroup: Support multiple events for metricgroup\n \n Some uncore metrics don\u0027t work as expected. For example, on\n cascadelakex:\n \n root@lkp-csl-2sp2:~# perf stat -M UNC_M_PMM_BANDWIDTH.TOTAL -a -- sleep 1\n \n Performance counter stats for \u0027system wide\u0027:\n \n 1841092 unc_m_pmm_rpq_inserts\n 3680816 unc_m_pmm_wpq_inserts\n \n 1.001775055 seconds time elapsed\n \n root@lkp-csl-2sp2:~# perf stat -M UNC_M_PMM_READ_LATENCY -a -- sleep 1\n \n Performance counter stats for \u0027system wide\u0027:\n \n 860649746 unc_m_pmm_rpq_occupancy.all\n 1840557 unc_m_pmm_rpq_inserts\n 12790627455 unc_m_clockticks\n \n 1.001773348 seconds time elapsed\n \n No metrics \u0027UNC_M_PMM_BANDWIDTH.TOTAL\u0027 or \u0027UNC_M_PMM_READ_LATENCY\u0027 are\n reported.\n \n The issue is, the case of an alias expanding to mulitple events is not\n supported, typically the uncore events. (see comments in\n find_evsel_group()).\n \n For UNC_M_PMM_BANDWIDTH.TOTAL in above example, the expanded event group\n is \u0027{unc_m_pmm_rpq_inserts,unc_m_pmm_wpq_inserts}:W\u0027, but the actual\n events passed to find_evsel_group are:\n \n unc_m_pmm_rpq_inserts\n unc_m_pmm_rpq_inserts\n unc_m_pmm_rpq_inserts\n unc_m_pmm_rpq_inserts\n unc_m_pmm_rpq_inserts\n unc_m_pmm_rpq_inserts\n unc_m_pmm_wpq_inserts\n unc_m_pmm_wpq_inserts\n unc_m_pmm_wpq_inserts\n unc_m_pmm_wpq_inserts\n unc_m_pmm_wpq_inserts\n unc_m_pmm_wpq_inserts\n \n For this multiple events case, it\u0027s not supported well.\n \n This patch introduces a new field \u0027metric_leader\u0027 in struct evsel. The\n first event is considered as a metric leader. For the rest of same\n events, they point to the first event via it\u0027s metric_leader field in\n struct evsel.\n \n This design is for adding the counting results of all same events to the\n first event in group (the metric_leader).\n \n With this patch,\n \n root@lkp-csl-2sp2:~# perf stat -M UNC_M_PMM_BANDWIDTH.TOTAL -a -- sleep 1\n \n Performance counter stats for \u0027system wide\u0027:\n \n 1842108 unc_m_pmm_rpq_inserts # 337.2 MB/sec UNC_M_PMM_BANDWIDTH.TOTAL\n 3682209 unc_m_pmm_wpq_inserts\n \n 1.001819706 seconds time elapsed\n \n root@lkp-csl-2sp2:~# perf stat -M UNC_M_PMM_READ_LATENCY -a -- sleep 1\n \n Performance counter stats for \u0027system wide\u0027:\n \n 861970685 unc_m_pmm_rpq_occupancy.all # 219.4 ns UNC_M_PMM_READ_LATENCY\n 1842772 unc_m_pmm_rpq_inserts\n 12790196356 unc_m_clockticks\n \n 1.001749103 seconds time elapsed\n \n Now we can see the correct metrics \u0027UNC_M_PMM_BANDWIDTH.TOTAL\u0027 and\n \u0027UNC_M_PMM_READ_LATENCY\u0027.\n \n Signed-off-by: Jin Yao \u003cyao.jin@linux.intel.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Kan Liang \u003ckan.liang@linux.intel.com\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lore.kernel.org/lkml/20190828055932.8269-5-yao.jin@linux.intel.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 287f2649f791819dd2d8f32f0213c8c521d6dfa0\nAuthor: Jin Yao \u003cyao.jin@linux.intel.com\u003e\nDate: Wed Aug 28 13:59:31 2019 +0800\n\n perf metricgroup: Scale the metric result\n \n Some metrics define the scale unit, such as\n \n {\n \"BriefDescription\": \"Intel Optane DC persistent memory read latency (ns). Derived from unc_m_pmm_rpq_occupancy.all\",\n \"Counter\": \"0,1,2,3\",\n \"EventCode\": \"0xE0\",\n \"EventName\": \"UNC_M_PMM_READ_LATENCY\",\n \"MetricExpr\": \"UNC_M_PMM_RPQ_OCCUPANCY.ALL / UNC_M_PMM_RPQ_INSERTS / UNC_M_CLOCKTICKS\",\n \"MetricName\": \"UNC_M_PMM_READ_LATENCY\",\n \"PerPkg\": \"1\",\n \"ScaleUnit\": \"6000000000ns\",\n \"UMask\": \"0x1\",\n \"Unit\": \"iMC\"\n },\n \n For above example, the ratio should be,\n \n ratio \u003d (UNC_M_PMM_RPQ_OCCUPANCY.ALL / UNC_M_PMM_RPQ_INSERTS / UNC_M_CLOCKTICKS) * 6000000000\n \n But in current code, the ratio is not scaled ( * 6000000000)\n \n With this patch, the ratio is scaled and the unit (ns) is printed.\n \n For example,\n # 219.4 ns UNC_M_PMM_READ_LATENCY\n \n Signed-off-by: Jin Yao \u003cyao.jin@linux.intel.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Kan Liang \u003ckan.liang@linux.intel.com\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lore.kernel.org/lkml/20190828055932.8269-4-yao.jin@linux.intel.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit a55ab7c4ca6986a542d313b02043a39ebf712a39\nAuthor: Jin Yao \u003cyao.jin@linux.intel.com\u003e\nDate: Wed Aug 28 13:59:29 2019 +0800\n\n perf pmu: Change convert_scale from static to global\n \n The function convert_scale() can be used to convert string to unit and\n scale. For example,\n \n s \u003d \"6000000000ns\";\n convert_scale(s, \u0026unit, \u0026scale);\n \n unit \u003d \"ns\", scale \u003d 6000000000.\n \n Currently this function is static. This patch renames the function to\n perf_pmu__convert_scale and changes the function to global. No\n functional change.\n \n Signed-off-by: Jin Yao \u003cyao.jin@linux.intel.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Kan Liang \u003ckan.liang@linux.intel.com\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lore.kernel.org/lkml/20190828055932.8269-2-yao.jin@linux.intel.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit d3300a3c4e76ccecf4daa889327e340a870c550b\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 30 15:09:54 2019 -0300\n\n perf symbols: Move mem_info and branch_info out of symbol.h\n \n The mem_info struct goes to mem-events.h and branch_info goes to\n branch.h, where they belong, this way we can remove several headers from\n symbols.h and trim the include dependency tree more.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-aupw71xnravcsu2xoabfmhpc@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit f2a39fe84901df2b3d1bec3459b65cee3e8db57c\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 30 14:45:20 2019 -0300\n\n perf auxtrace: Uninline functions that touch perf_session\n \n So that we don\u0027t carry the session.h include directive in auxtrace.h,\n which in turn opens a can of worms of files that were getting all sorts\n of things via that include, fix them all.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-d2d83aovpgri2z75wlitquni@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit fa0d98462fae5d4951f22f3ac1090d48c53396d1\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 30 12:52:25 2019 -0300\n\n perf tools: Remove needless evlist.h include directives\n \n Remove the last unneeded use of cache.h in a header, we can check where\n it is really needed, i.e. we can remove it and be sure that it isn\u0027t\n being obtained indirectly.\n \n This is an old file, used by now incorrectly in many places, so it was\n providing includes needed indirectly, fixup this fallout.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-3x3l8gihoaeh7714os861ia7@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 7ae811b12e419fd70b7d7159f20ed8519bbe18cc\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 30 12:29:03 2019 -0300\n\n perf tools: Remove needless evlist.h include directives\n \n Now that evlist.h isn\u0027t included by any other header, we can check where\n it is really needed, i.e. we can remove it and be sure that it isn\u0027t\n being obtained indirectly.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-6d7kape36m94a266md0d3xbh@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit ef7d95661d046eddf2cf33847278781404679a2f\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 30 12:18:50 2019 -0300\n\n perf tools: Remove needless thread_map.h include directives\n \n Now that thread_map.h isn\u0027t included by any other header, we can check where\n it is really needed, i.e. we can remove it and be sure that it isn\u0027t\n being obtained indirectly.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-fyzvg64cz1ikvyxp8d6nrhz1@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 4becb2395f9166b11d68817ed4af8fc06b840908\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 30 12:13:45 2019 -0300\n\n perf tools: Remove needless thread.h include directives\n \n Now that thread.h isn\u0027t included by any other header, we can check where\n it is really needed, i.e. we can remove it and be sure that it isn\u0027t\n being obtained indirectly.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-kh333ivjbw05wsggckpziu86@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit df1a0a110c2c0138665f6d8ec96812ea14c2d818\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 30 12:07:23 2019 -0300\n\n perf tools: Remove needless map.h include directives\n \n Now that map.h isn\u0027t included by any other header, we can check where\n it is really needed, i.e. we can remove it and be sure that it isn\u0027t\n being obtained indirectly.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-iu8ylqky7g1i9i54v3y7qovw@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit afce8c482c48e2c42c155eeae4cd048c2b5fbb99\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 30 11:56:47 2019 -0300\n\n perf probe: No need for symbol.h, symbol_conf is enough\n \n Remove one more unneeded use of symbol.h\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-vrda1tuem1o8pk82t2kfjtun@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 5c9dbe6da13398d09efc9ec479194afa6d9ec9e6\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 30 11:54:00 2019 -0300\n\n perf tools: Remove needless sort.h include directives\n \n Now that sort.h isn\u0027t included by any other header, we can check where\n it is really needed, i.e. we can remove it and be sure that it isn\u0027t\n being obtained indirectly.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-tom8k0lbsxd9joprr8zpu6w1@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 4772925885dac93aa5f00d1c1da93277778099cd\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 30 11:44:32 2019 -0300\n\n perf tools: Move \u0027struct events_stats\u0027 and prototypes to separate header\n \n This will allow us to untangle the header dependency a bit more, as some\n places will not need event.h anymore.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-enqncj29ovzaat3cd9203rwl@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 171f7474b6bb6c7074431f76c28ea87d625c68fd\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 30 11:28:14 2019 -0300\n\n perf hist: Remove needless ui/progress.h from hist.h\n \n We only need a forward declaration, add it and fixup all the files that\n need ui_progress definitions but were wrongly getting it from hist.h.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-84a90o9jdxybffxo9jmouokw@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 4a3cec84949d14dc3ef7fb8a51b8949af93cac13\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 30 11:11:01 2019 -0300\n\n perf dsos: Move the dsos struct and its methods to separate source files\n \n So that we can reduce the header dependency tree further, in the process\n noticed that lots of places were getting even things like build-id\n routines and \u0027struct perf_tool\u0027 definition indirectly, so fix all those\n too.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-ti0btma9ow5ndrytyoqdk62j@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit b1d1b094f7570a13dd7c9b995209baacc8aa6273\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 30 10:26:37 2019 -0300\n\n perf symbols: Move symsrc prototypes to a separate header\n \n So that we can remove dso.h from symbol.h and reduce the header\n dependency tree.\n \n Fixup cases where struct dso guts are needed but were obtained via\n symbol.h, indirectly.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-ip683cegt306ncu3gsz7ii21@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit c38fa94d188234f40fe3911b1a7650e056ef42ea\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 30 10:19:19 2019 -0300\n\n perf symbols: Add missing linux/refcount.h to symbol.h\n \n We use refcount_t there, so we need that header or else it\u0027ll break when\n we remove dso.h, that is from where it is getting that definition now...\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-5albrk0uve6x9cf6x3ebwpae@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 9bea81b36a8d7d23c9bb7f4d63a9a842eec134a0\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 30 10:01:50 2019 -0300\n\n perf symbol: Move C++ demangle defines to the only file using it\n \n No need to have it generally available in such a critical header as\n symbol.h.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-es1ufxv7bihiumytn5dm3drn@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit fac583fdb6741bf4850928b2a5bb8b0118b5879c\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 30 09:43:25 2019 -0300\n\n perf dso: Adopt DSO related macros from symbol.h\n \n Reducing the size of symbol.h by removing things that are better placed\n somewhere else.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-edenkmjt1oe5fks2s6umd30b@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit e97fd1383cd77c467d2aed7fa4e596789df83977\nAuthor: Tzvetomir Stoyanov \u003ctstoyanov@vmware.com\u003e\nDate: Mon Aug 5 16:43:15 2019 -0400\n\n libtraceevent: Change users plugin directory\n \n To be compliant with XDG user directory layout, the user\u0027s plugin\n directory is changed from ~/.traceevent/plugins to\n ~/.local/lib/traceevent/plugins/\n \n Suggested-by: Patrick McLean \u003cchutzpah@gentoo.org\u003e\n Signed-off-by: Tzvetomir Stoyanov \u003ctstoyanov@vmware.com\u003e\n Cc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Patrick McLean \u003cchutzpah@gentoo.org\u003e\n Cc: linux-trace-devel@vger.kernel.org\n Link: https://lore.kernel.org/linux-trace-devel/20190313144206.41e75cf8@patrickm/\n Link: http://lore.kernel.org/linux-trace-devel/20190801074959.22023-4-tz.stoyanov@gmail.com\n Link: http://lore.kernel.org/lkml/20190805204355.344622683@goodmis.org\n Signed-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 5d6552ab3b71fac48a9c7452c7014d37ca80b17f\nAuthor: Tzvetomir Stoyanov \u003ctstoyanov@vmware.com\u003e\nDate: Mon Aug 5 16:43:14 2019 -0400\n\n libtraceevent: Remove tep_register_trace_clock()\n \n The tep_register_trace_clock() API is used to instruct the traceevent\n library how to print the event time stamps. As event print interface if\n redesigned, this API is not needed any more. The new event print API is\n flexible and the user can specify how the time stamps are printed.\n \n Signed-off-by: Tzvetomir Stoyanov \u003ctstoyanov@vmware.com\u003e\n Cc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Patrick McLean \u003cchutzpah@gentoo.org\u003e\n Cc: linux-trace-devel@vger.kernel.org\n Link: http://lore.kernel.org/linux-trace-devel/20190801074959.22023-3-tz.stoyanov@gmail.com\n Link: http://lore.kernel.org/lkml/20190805204355.195042846@goodmis.org\n Signed-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 38847db9740a984e8538ce2573cbc0fc7edf22b3\nAuthor: Tzvetomir Stoyanov \u003ctstoyanov@vmware.com\u003e\nDate: Mon Aug 5 16:43:13 2019 -0400\n\n libtraceevent, perf tools: Changes in tep_print_event_* APIs\n \n Libtraceevent APIs for printing various trace events information are\n complicated, there are complex extra parameters. To control the way\n event information is printed, the user should call a set of functions in\n a specific sequence.\n \n These APIs are reimplemented to provide a more simple interface for\n printing event information.\n \n Removed APIs:\n \n tep_print_event_task()\n tep_print_event_time()\n tep_print_event_data()\n tep_event_info()\n tep_is_latency_format()\n tep_set_latency_format()\n tep_data_latency_format()\n tep_set_print_raw()\n \n A new API for printing event information is introduced:\n void tep_print_event(struct tep_handle *tep, struct trace_seq *s,\n struct tep_record *record, const char *fmt, ...);\n where \"fmt\" is a printf-like format string, followed by the event\n fields to be printed. Supported fields:\n TEP_PRINT_PID, \"%d\" - event PID\n TEP_PRINT_CPU, \"%d\" - event CPU\n TEP_PRINT_COMM, \"%s\" - event command string\n TEP_PRINT_NAME, \"%s\" - event name\n TEP_PRINT_LATENCY, \"%s\" - event latency\n TEP_PRINT_TIME, %d - event time stamp. A divisor and precision\n can be specified as part of this format string:\n \"%precision.divisord\". Example:\n \"%3.1000d\" - divide the time by 1000 and print the first 3 digits\n before the dot. Thus, the time stamp \"123456000\" will be printed as\n \"123.456\"\n TEP_PRINT_INFO, \"%s\" - event information.\n TEP_PRINT_INFO_RAW, \"%s\" - event information, in raw format.\n \n Example:\n tep_print_event(tep, s, record, \"%16s-%-5d [%03d] %s %6.1000d %s %s\",\n TEP_PRINT_COMM, TEP_PRINT_PID, TEP_PRINT_CPU,\n TEP_PRINT_LATENCY, TEP_PRINT_TIME, TEP_PRINT_NAME, TEP_PRINT_INFO);\n Output:\n ls-11314 [005] d.h. 185207.366383 function __wake_up\n \n Signed-off-by: Tzvetomir Stoyanov \u003ctstoyanov@vmware.com\u003e\n Cc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: linux-trace-devel@vger.kernel.org\n Cc: Patrick McLean \u003cchutzpah@gentoo.org\u003e\n Link: http://lore.kernel.org/linux-trace-devel/20190801074959.22023-2-tz.stoyanov@gmail.com\n Link: http://lore.kernel.org/lkml/20190805204355.041132030@goodmis.org\n Signed-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 4cb3c6d546aa5493a960d05eb73bad8e69a58574\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 29 17:19:02 2019 -0300\n\n perf event: Remove needless include directives from event.h\n \n bpf.h and build-id.h are not needed at all in event.h, remove them.\n \n And fixup the fallout of files that were getting needed stuff from this\n now pruned include.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-rdm3dgtlrndmmnlc4bafsg3b@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 03e61929c0d227ed3e1c322fc3804216ea298b7e\nAuthor: Shawn Lin \u003cshawn.lin@rock-chips.com\u003e\nDate: Fri Aug 30 08:26:47 2019 +0800\n\n arm64: dts: rockchip: limit clock rate of MMC controllers for RK3328\n \n 150MHz is a fundamental limitation of RK3328 Soc, w/o this limitation,\n eMMC, for instance, will run into 200MHz clock rate in HS200 mode, which\n makes the RK3328 boards not always boot properly. By adding it in\n rk3328.dtsi would also obviate the worry of missing it when adding new\n boards.\n \n Fixes: 52e02d377a72 (\"arm64: dts: rockchip: add core dtsi file for RK3328 SoCs\")\n Cc: stable@vger.kernel.org\n Cc: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Cc: Liang Chen \u003ccl@rock-chips.com\u003e\n Signed-off-by: Shawn Lin \u003cshawn.lin@rock-chips.com\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit b6b5574b80d6ce6ca87ae3ea1e97cff1bf730f2e\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 29 17:10:59 2019 -0300\n\n perf env: Remove env.h from other headers where just a fwd decl is needed\n \n And fixup the fallout of c files not building due to now missing\n headers.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-sw8k3kpla98pr3rqypbjk9hf@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 8520a98dbab61e9e340cdfb72dd17ccc8a98961e\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 29 16:18:59 2019 -0300\n\n perf debug: Remove needless include directives from debug.h\n \n All we need there is a forward declaration for \u0027union perf_event\u0027, so\n remove it from there and add missing header directives in places using\n things from this indirect include.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-7ftk0ztstqub1tirjj8o8xbl@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit e56476897448a0cc6f748535a199d2b68ad47d55\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:14 2019 -0700\n\n fpga: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Moritz Fischer \u003cmdf@kernel.org\u003e\n Cc: linux-fpga@vger.kernel.org\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Moritz Fischer \u003cmdf@kernel.org\u003e\n\ncommit 849dbf09234a432795e2e210de16a51ddb2228f5\nAuthor: Sudarsana Reddy Kalluru \u003cskalluru@marvell.com\u003e\nDate: Fri Aug 30 00:42:06 2019 -0700\n\n qede: Add support for dumping the grc data.\n \n This patch adds driver support for configuring grc dump config flags, and\n dumping the grc data via ethtool get/set-dump interfaces.\n \n Signed-off-by: Sudarsana Reddy Kalluru \u003cskalluru@marvell.com\u003e\n Signed-off-by: Ariel Elior \u003caelior@marvell.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 3b86bd076284724489381e391f84a34078b3d5bc\nAuthor: Sudarsana Reddy Kalluru \u003cskalluru@marvell.com\u003e\nDate: Fri Aug 30 00:42:05 2019 -0700\n\n qed: Add APIs for configuring grc dump config flags.\n \n The patch adds driver support for configuring the grc dump config flags.\n \n Signed-off-by: Sudarsana Reddy Kalluru \u003cskalluru@marvell.com\u003e\n Signed-off-by: Ariel Elior \u003caelior@marvell.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d44a3ced7023d3ff05e3be947b094e1d96562584\nAuthor: Sudarsana Reddy Kalluru \u003cskalluru@marvell.com\u003e\nDate: Fri Aug 30 00:42:04 2019 -0700\n\n qede: Add support for reading the config id attributes.\n \n Add driver support for dumping the config id attributes via ethtool dump\n interfaces.\n \n Signed-off-by: Sudarsana Reddy Kalluru \u003cskalluru@marvell.com\u003e\n Signed-off-by: Ariel Elior \u003caelior@marvell.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 2d4c849530a9771ed38d4046d2b631c1a4e2f9a6\nAuthor: Sudarsana Reddy Kalluru \u003cskalluru@marvell.com\u003e\nDate: Fri Aug 30 00:42:03 2019 -0700\n\n qed: Add APIs for reading config id attributes.\n \n The patch adds driver support for reading the config id attributes from NVM\n flash partition.\n \n Signed-off-by: Sudarsana Reddy Kalluru \u003cskalluru@marvell.com\u003e\n Signed-off-by: Ariel Elior \u003caelior@marvell.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 5f33183b7fdfeba98d02b099c9de887378d47943\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Fri Aug 30 03:53:25 2019 +0300\n\n net: dsa: tag_8021q: Restore bridge VLANs when enabling vlan_filtering\n \n The bridge core assumes that enabling/disabling vlan_filtering will\n translate into the simple toggling of a flag for switchdev drivers.\n \n That is clearly not the case for sja1105, which alters the VLAN table\n and the pvids in order to obtain port separation in standalone mode.\n \n There are 2 parts to the issue.\n \n First, tag_8021q changes the pvid to a unique per-port rx_vid for frame\n identification. But we need to disable tag_8021q when vlan_filtering\n kicks in, and at that point, the VLAN configured as pvid will have to be\n removed from the filtering table of the ports. With an invalid pvid, the\n ports will drop all traffic. Since the bridge will not call any vlan\n operation through switchdev after enabling vlan_filtering, we need to\n ensure we\u0027re in a functional state ourselves. Hence read the pvid that\n the bridge is aware of, and program that into our ports.\n \n Secondly, tag_8021q uses the 1024-3071 range privately in\n vlan_filtering\u003d0 mode. Had the user installed one of these VLANs during\n a previous vlan_filtering\u003d1 session, then upon the next tag_8021q\n cleanup for vlan_filtering to kick in again, VLANs in that range will\n get deleted unconditionally, hence breaking user expectation. So when\n deleting the VLANs, check if the bridge had knowledge about them, and if\n it did, re-apply the settings. Wrap this logic inside a\n dsa_8021q_vid_apply helper function to reduce code duplication.\n \n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Reviewed-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f40d9b20864ae571c31f57989fd869958ce675c2\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Fri Aug 30 03:53:24 2019 +0300\n\n net: bridge: Populate the pvid flag in br_vlan_get_info\n \n Currently this simplified code snippet fails:\n \n br_vlan_get_pvid(netdev, \u0026pvid);\n br_vlan_get_info(netdev, pvid, \u0026vinfo);\n ASSERT(!(vinfo.flags \u0026 BRIDGE_VLAN_INFO_PVID));\n \n It is intuitive that the pvid of a netdevice should have the\n BRIDGE_VLAN_INFO_PVID flag set.\n \n However I can\u0027t seem to pinpoint a commit where this behavior was\n introduced. It seems like it\u0027s been like that since forever.\n \n At a first glance it would make more sense to just handle the\n BRIDGE_VLAN_INFO_PVID flag in __vlan_add_flags. However, as Nikolay\n explains:\n \n There are a few reasons why we don\u0027t do it, most importantly because\n we need to have only one visible pvid at any single time, even if it\u0027s\n stale - it must be just one. Right now that rule will not be violated\n by this change, but people will try using this flag and could see two\n pvids simultaneously. You can see that the pvid code is even using\n memory barriers to propagate the new value faster and everywhere the\n pvid is read only once. That is the reason the flag is set\n dynamically when dumping entries, too. A second (weaker) argument\n against would be given the above we don\u0027t want another way to do the\n same thing, specifically if it can provide us with two pvids (e.g. if\n walking the vlan list) or if it can provide us with a pvid different\n from the one set in the vg. [Obviously, I\u0027m talking about RCU\n pvid/vlan use cases similar to the dumps. The locked cases are fine.\n I would like to avoid explaining why this shouldn\u0027t be relied upon\n without locking]\n \n So instead of introducing the above change and making sure of the pvid\n uniqueness under RCU, simply dynamically populate the pvid flag in\n br_vlan_get_info().\n \n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Acked-by: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 94044dfb0bcf270ee5cebd590cfa0c2a1a979687\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Fri Aug 30 22:24:46 2019 +0900\n\n MAINTAINERS: update entry of firewire audio drivers\n \n This commit adds myself as one of maintainers for firewire audio\n drivers and IEC 61883-1/6 packet streaming engine. I call them ALSA\n firewire stack as a whole.\n \n 6 years ago I joined in development for this category of drivers with\n heavy reverse-engineering tasks and over 100 models are now available\n from ALSA applications. IEEE 1394 bus itself and units on the bus are\n enough legacy but the development still continues.\n \n I have a plan to add drastic enhancement in kernel v5.5 and v5.6 period.\n This commit adds myself into MAINTAINERS so that developers and users\n can easily find active developer to post their issues, especially for\n regression.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit dffa8feb308455f9b3ce0eeb55a4eac3afc0786b\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 30 19:19:29 2019 +0100\n\n drm/i915/perf: Assert locking for i915_init_oa_perf_state()\n \n We use the context-\u003epin_mutex to serialise updates to the OA config and\n the registers values written into each new context. Document this\n relationship and assert we do hold the context-\u003epin_mutex as used by\n gen8_configure_all_contexts() to serialise updates to the OA config\n itself.\n \n v2: Add a white-lie for when we call intel_gt_resume() from init.\n v3: Lie while we have the context pinned inside atomic reset.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Reviewed-by: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e #v1\n Link: https://patchwork.freedesktop.org/patch/msgid/20190830181929.18663-1-chris@chris-wilson.co.uk\n\ncommit 72b8cba90957157610c689de23f35b635262212b\nAuthor: Michal Simek \u003cmichal.simek@xilinx.com\u003e\nDate: Thu Aug 22 16:22:24 2019 +0200\n\n hwmon: (iio_hwmon) Enable power exporting from IIO\n \n There is no reason why power channel shouldn\u0027t be exported as is done for\n voltage, current, temperature and humidity.\n \n Power channel is available on iio ina226 driver.\n \n Sysfs IIO documentation for power attribute added by commit 7c6d5c7ee883\n (\"iio: Documentation: Add missing documentation for power attribute\")\n is declaring that value is in mili-Watts but hwmon interface is expecting\n value in micro-Watts that\u0027s why there is a need for mili-Watts to\n micro-Watts conversion.\n \n Tested on Xilinx ZCU102 board.\n \n Signed-off-by: Michal Simek \u003cmichal.simek@xilinx.com\u003e\n Link: https://lore.kernel.org/r/db71f5ae87e4521a2856a1be5544de0b6cede575.1566483741.git.michal.simek@xilinx.com\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 75d5d9d74538d6291b4294bf794aeec55e913ed2\nAuthor: John Wang \u003cwangzqbj@inspur.com\u003e\nDate: Mon Aug 19 17:14:25 2019 +0800\n\n dt-bindings: Add ipsps1 as a trivial device\n \n The ipsps1 is an Inspur Power System power supply unit\n \n Signed-off-by: John Wang \u003cwangzqbj@inspur.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 2f7fed7e04655782e504c403d5a2b56f0ec25601\nAuthor: John Wang \u003cwangzqbj@inspur.com\u003e\nDate: Mon Aug 19 17:15:09 2019 +0800\n\n hwmon: pmbus: Add Inspur Power System power supply driver\n \n Add the driver to monitor Inspur Power System power supplies\n with hwmon over pmbus.\n \n This driver adds sysfs attributes for additional power supply data,\n including vendor, model, part_number, serial number,\n firmware revision, hardware revision, and psu mode(active/standby).\n \n Signed-off-by: John Wang \u003cwangzqbj@inspur.com\u003e\n Reviewed-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n Link: https://lore.kernel.org/r/20190819091509.29276-1-wangzqbj@inspur.com\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 31a2a068d36bcd85bcf0d72a397de504370b9f59\nAuthor: Max Staudt \u003cmax@enpas.org\u003e\nDate: Mon Aug 19 14:16:17 2019 +0200\n\n hwmon/ltc2990: Generalise DT to fwnode support\n \n ltc2990 will now use device_property_read_u32_array() instead of\n of_property_read_u32_array() - allowing the use of software nodes\n via fwnode_create_software_node().\n \n This allows code using i2c_new_device() to specify a default\n measurement mode for the LTC2990 via fwnode_create_software_node().\n \n Signed-off-by: Max Staudt \u003cmax@enpas.org\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n Link: https://lore.kernel.org/r/20190819121618.16557-2-max@enpas.org\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit df90ff256339e3508f3a829fc097b1f054c3d27b\nAuthor: Stefan Wahren \u003cwahrenst@gmx.net\u003e\nDate: Tue Aug 13 20:17:28 2019 +0200\n\n hwmon: (raspberrypi) update MODULE_AUTHOR() email address\n \n The email address listed in MODULE_AUTHOR() will be disabled in the\n near future. Replace it with my private one.\n \n Signed-off-by: Stefan Wahren \u003cwahrenst@gmx.net\u003e\n Link: https://lore.kernel.org/r/1565720249-6549-2-git-send-email-wahrenst@gmx.net\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 40760e7d1f40988d70dc48a7d73da5a0e214bf3a\nAuthor: Iker Perez del Palomar Sustatxa \u003ciker.perez@codethink.co.uk\u003e\nDate: Thu Aug 8 09:02:46 2019 +0100\n\n hwmon: (lm75) Modularize lm75_write and make hwmon_chip writable\n \n * Create two separate functions to write into hwmon_temp and hwmon_chip.\n * Call the functions from lm75_write.\n * Make hwm_chip writable if the chip supports more than one sample time.\n \n Signed-off-by: Iker Perez del Palomar Sustatxa \u003ciker.perez@codethink.co.uk\u003e\n Link: https://lore.kernel.org/r/20190808080246.8371-5-iker.perez@codethink.co.uk\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 3a3aceaa65b7801c426da688d925aacd454b45e5\nAuthor: Iker Perez del Palomar Sustatxa \u003ciker.perez@codethink.co.uk\u003e\nDate: Thu Aug 8 09:02:45 2019 +0100\n\n hwmon: (lm75) Add new fields into lm75_params_\n \n The new fields are included to prepare the driver for next patch. The\n fields are:\n \n * *resolutions: Stores all the supported resolutions by the device.\n * num_sample_times: Stores the number of possible sample times.\n * *sample_times: Stores all the possible sample times to be set.\n * sample_set_masks: The set_masks for the possible sample times\n * sample_clr_mask: Clear mask to set the default sample time.\n \n Signed-off-by: Iker Perez del Palomar Sustatxa \u003ciker.perez@codethink.co.uk\u003e\n Link: https://lore.kernel.org/r/20190808080246.8371-4-iker.perez@codethink.co.uk\n [groeck: Minor structure documentation fixes]\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 3c8e3f749de45f76f9e584da517f7b31a5ac64b8\nAuthor: Iker Perez del Palomar Sustatxa \u003ciker.perez@codethink.co.uk\u003e\nDate: Thu Aug 8 09:02:44 2019 +0100\n\n hwmon: (lm75) Create function from code to write into registers\n \n Wrap the existing code to write configurations into registers in\n a function.\n \n Added error handling to the function.\n \n Signed-off-by: Iker Perez del Palomar Sustatxa \u003ciker.perez@codethink.co.uk\u003e\n Link: https://lore.kernel.org/r/20190808080246.8371-3-iker.perez@codethink.co.uk\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit dfaf09845ca76854aa69590df2601399d8691d23\nAuthor: Iker Perez del Palomar Sustatxa \u003ciker.perez@codethink.co.uk\u003e\nDate: Thu Aug 8 09:02:43 2019 +0100\n\n hwmon: (lm75) Create structure to save all the configuration parameters.\n \n * Add to lm75_data kind field to store the kind of device the driver is\n working with.\n * Add an structure to store the configuration parameters of all the\n supported devices.\n * Delete resolution_limits from lm75_data and include them in the structure\n described above.\n * Add a pointer to the configuration parameters structure to be used as a\n reference to obtain the parameters.\n * Delete switch-case approach to get the device configuration parameters.\n * The structure is cleaner and easier to maintain.\n \n Signed-off-by: Iker Perez del Palomar Sustatxa \u003ciker.perez@codethink.co.uk\u003e\n Link: https://lore.kernel.org/r/20190808080246.8371-2-iker.perez@codethink.co.uk\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit a6b16141ac7690faebe15d047703f06da3fbb1b3\nAuthor: amy.shih \u003camy.shih@advantech.com.tw\u003e\nDate: Wed Aug 7 01:38:41 2019 +0000\n\n hwmon: (nct7904) Add extra sysfs support for fan, voltage and temperature.\n \n NCT-7904D also supports reading of channel limitation registers as well\n as SMI status registers for fan, voltage and temperature monitoring.\n It also supports reading the temperature sensor type (thermal diode,\n thermistor, AMD SB-TSI or Intel PECI).\n \n Add the following sysfs nodes:\n \n -fan[1-*]_min\n -fan[1-*]_alarm\n -in[1-*]_min\n -in[1-*]_max\n -in[1-*]_alarm\n -temp[1-*]_max\n -temp[1-*]_max_hyst\n -temp[1-*]_emergency\n -temp[1-*]_emergency_hyst\n -temp[1-*]_alarm\n -temp[1-*]_type\n \n Signed-off-by: Amy Shih \u003camy.shih@advantech.com.tw\u003e\n Link: https://lore.kernel.org/r/20190807013842.24451-1-Amy.Shih@advantech.com.tw\n [groeck: Clarified description]\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 174a283e0f09cb98ce8a6aaa51d89d5f603c48a4\nAuthor: Jean Delvare \u003cjdelvare@suse.de\u003e\nDate: Tue Aug 6 10:21:23 2019 +0200\n\n hwmon: w83795: Fan control option isn\u0027t that dangerous\n \n I have been using SENSORS_W83795_FANCTRL for several years and never\n had any problem. When the driver was added, I had not tested that\n part of the driver yet so I wanted to be super cautious, but time has\n shown that it works just fine.\n \n In the long run I even believe that we should drop the option and\n enable the feature unconditionally. It doesn\u0027t do anything until the\n user explicitly starts twiddling with sysfs attributes anyway.\n \n Signed-off-by: Jean Delvare \u003cjdelvare@suse.de\u003e\n Link: https://lore.kernel.org/r/20190806102123.3118bcc5@endymion\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 02f59d1f5acbe57713755bb7c0c4b04c2b6e99d6\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:17 2019 -0700\n\n hwmon: (npcm750-pwm-fan) Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Jean Delvare \u003cjdelvare@suse.com\u003e\n Cc: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Cc: linux-hwmon@vger.kernel.org\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n [groeck: Dropped jz4740-hwmon.c (driver is being removed)]\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 262ce4af99e1d04a866ac8bc88a68ff28acf9cc1\nAuthor: Wang Shenran \u003cshenran268@gmail.com\u003e\nDate: Wed Jul 24 11:01:10 2019 +0300\n\n hwmon: (acpi_power_meter) Change log level for \u0027unsafe software power cap\u0027\n \n At boot time, the acpi_power_meter driver logs the following error level\n message: \"Ignoring unsafe software power cap\". Having read about it from\n a few sources, it seems that the error message can be quite misleading.\n \n While the message can imply that Linux is ignoring the fact that the\n system is operating in potentially dangerous conditions, the truth is\n the driver found an ACPI_PMC object that supports software power\n capping. The driver simply decides not to use it, perhaps because it\n doesn\u0027t support the object.\n \n The best solution is probably changing the log level from error to warning.\n All sources I have found, regarding the error, have downplayed its\n significance. There is not much of a reason for it to be on error level,\n while causing potential confusions or misinterpretations.\n \n Signed-off-by: Wang Shenran \u003cshenran268@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190724080110.6952-1-shenran268@gmail.com\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit f15f5263b728ef559dcdea95f5fc6983191c003f\nAuthor: Björn Gerhart \u003cgerhart@posteo.de\u003e\nDate: Tue Jul 23 18:06:46 2019 +0200\n\n hwmon: (nct6775) Integrate new model nct6116\n \n Add support for NCT6116D to nct6775 driver.\n \n Signed-off-by: Bjoern Gerhart \u003cgerhart@posteo.de\u003e\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 3a1eec1726964fa0f4170ce272331934cd73ed93\nAuthor: Grant McEwan \u003cgrant.mcewan@alliedtelesis.co.nz\u003e\nDate: Mon Jul 22 10:55:30 2019 +1200\n\n hwmon: (adt7475) Convert to use hwmon_device_register_with_groups()\n \n hwmon_device_register() is a deprecated function and produces a warning.\n \n Converting the driver to use the hwmon_device_register_with_groups()\n instead.\n \n Signed-off-by: Grant McEwan \u003cgrant.mcewan@alliedtelesis.co.nz\u003e\n Link: https://lore.kernel.org/r/20190721225530.28799-2-grant.mcewan@alliedtelesis.co.nz\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit df625020092dbbe7977bf9e4569a2a3e9ee73ce1\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:10 2019 +0200\n\n hwmon: (w83781d) convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Link: https://lore.kernel.org/r/20190722172611.3797-4-wsa+renesas@sang-engineering.com\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 195efc81b3af43361e6cf9654deef465710739e3\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:09 2019 +0200\n\n hwmon: (smm665) convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Link: https://lore.kernel.org/r/20190722172611.3797-3-wsa+renesas@sang-engineering.com\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 8342d5e69957b86cd90b4f89617109b6b906018b\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:08 2019 +0200\n\n hwmon: (asb100) convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Link: https://lore.kernel.org/r/20190722172611.3797-2-wsa+renesas@sang-engineering.com\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit b59ee840b02851704c8d0f521142343219c5e086\nAuthor: Marcel Bocu \u003cmarcel.p.bocu@gmail.com\u003e\nDate: Mon Jul 22 20:46:53 2019 +0300\n\n hwmon: (k10temp) Add support for AMD family 17h, model 70h CPUs\n \n It would seem like model 70h is behaving in the same way as model 30h,\n so let\u0027s just add the new F3 PCI ID to the list of compatible devices.\n \n Unlike previous Ryzen/Threadripper, Ryzen gen 3 processors do not need\n temperature offsets anymore. This has been reported in the press and\n verified on my Ryzen 3700X by checking that the idle temperature\n reported by k10temp is matching the temperature reported by the\n firmware.\n \n Vicki Pfau sent an identical patch after I checked that no-one had\n written this patch. I would have been happy about dropping my patch but\n unlike for his patch series, I had already Cc:ed the x86 people and\n they already reviewed the changes. Since Vicki has not answered to\n any email after his initial series, let\u0027s assume she is on vacation\n and let\u0027s avoid duplication of reviews from the maintainers and merge\n my series. To acknowledge Vicki\u0027s anteriority, I added her S-o-b to\n the patch.\n \n v2, suggested by Guenter Roeck and Brian Woods:\n - rename from 71h to 70h\n \n Signed-off-by: Vicki Pfau \u003cvi@endrift.com\u003e\n Signed-off-by: Marcel Bocu \u003cmarcel.p.bocu@gmail.com\u003e\n Tested-by: Marcel Bocu \u003cmarcel.p.bocu@gmail.com\u003e\n \n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: Borislav Petkov \u003cbp@alien8.de\u003e\n Cc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\n Cc: x86@kernel.org\n Cc: \"Woods, Brian\" \u003cBrian.Woods@amd.com\u003e\n Cc: Clemens Ladisch \u003cclemens@ladisch.de\u003e\n Cc: Jean Delvare \u003cjdelvare@suse.com\u003e\n Cc: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Cc: linux-hwmon@vger.kernel.org\n Link: https://lore.kernel.org/r/20190722174653.2391-1-marcel.p.bocu@gmail.com\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 483e79a703a0f34d8807fb5d333688538b343e4c\nAuthor: Marcel Bocu \u003cmarcel.p.bocu@gmail.com\u003e\nDate: Mon Jul 22 20:45:10 2019 +0300\n\n x86/amd_nb: Add PCI device IDs for family 17h, model 70h\n \n The AMD Ryzen gen 3 processors came with a different PCI IDs for the\n function 3 \u0026 4 which are used to access the SMN interface. The root\n PCI address however remained at the same address as the model 30h.\n \n Adding the F3/F4 PCI IDs respectively to the misc and link ids appear\n to be sufficient for k10temp, so let\u0027s add them and follow up on the\n patch if other functions need more tweaking.\n \n Vicki Pfau sent an identical patch after I checked that no-one had\n written this patch. I would have been happy about dropping my patch but\n unlike for his patch series, I had already Cc:ed the x86 people and\n they already reviewed the changes. Since Vicki has not answered to\n any email after his initial series, let\u0027s assume she is on vacation\n and let\u0027s avoid duplication of reviews from the maintainers and merge\n my series. To acknowledge Vicki\u0027s anteriority, I added her S-o-b to\n the patch.\n \n v2, suggested by Guenter Roeck and Brian Woods:\n - rename from 71h to 70h\n \n Signed-off-by: Vicki Pfau \u003cvi@endrift.com\u003e\n Signed-off-by: Marcel Bocu \u003cmarcel.p.bocu@gmail.com\u003e\n Tested-by: Marcel Bocu \u003cmarcel.p.bocu@gmail.com\u003e\n Acked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Brian Woods \u003cbrian.woods@amd.com\u003e\n Acked-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e # pci_ids.h\n \n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: Borislav Petkov \u003cbp@alien8.de\u003e\n Cc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\n Cc: x86@kernel.org\n Cc: \"Woods, Brian\" \u003cBrian.Woods@amd.com\u003e\n Cc: Clemens Ladisch \u003cclemens@ladisch.de\u003e\n Cc: Jean Delvare \u003cjdelvare@suse.com\u003e\n Cc: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Cc: linux-hwmon@vger.kernel.org\n Link: https://lore.kernel.org/r/20190722174510.2179-1-marcel.p.bocu@gmail.com\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit ca6d3c53be00bea7117ff68807a01021a84b185a\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Mon Jul 22 08:07:45 2019 -0300\n\n docs: hwmon: pxe1610: convert to ReST format and add to the index\n \n This document was recently introduced. Convert it to ReST\n just like the other hwmon documents, adding it to the hwmon index.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Link: https://lore.kernel.org/r/657bf022625e0888d3becf10c78d162eeb864608.1563792334.git.mchehab+samsung@kernel.org\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 9c26024e71ebbe97244c2aad93762e3e820271f7\nAuthor: Robert Karszniewicz \u003cavoidr@riseup.net\u003e\nDate: Sun Jul 21 14:00:51 2019 +0200\n\n hwmon: (k8temp) update to use new hwmon registration API\n \n Removes:\n - hwmon_dev from k8temp_data struct, as that is now passed\n to callbacks, anyway.\n - other k8temp_data struct fields, too.\n - k8temp_update_device()\n \n Also reduces binary size:\n text data bss dec hex filename\n 4139 1448 0 5587 15d3 drivers/hwmon/k8temp.ko.bak\n 3103 1220 0 4323 10e3 drivers/hwmon/k8temp.ko\n \n Signed-off-by: Robert Karszniewicz \u003cavoidr@firemail.cc\u003e\n Signed-off-by: Robert Karszniewicz \u003cavoidr@riseup.net\u003e\n Link: https://lore.kernel.org/r/20190721120051.28064-1-avoidr@riseup.net\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 24a39fcd95289d6d462e24aa5c98203211a7bd94\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Sun Jul 21 12:15:53 2019 +0200\n\n hwmon: (pmbus/max31785) Remove a useless #define\n \n There is a typo in MAX37185_NUM_FAN_PAGES. To be consistent, it should be\n MAX31785_NUM_FAN_PAGES (1 and 7 switched).\n \n At line 24, we already have:\n #define MAX31785_NR_FAN_PAGES 6\n and MAX37185_NUM_FAN_PAGES seems to be unused.\n \n It is likely that it is only a typo and/or a left-over.\n So, axe it.\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Link: https://lore.kernel.org/r/20190721101553.20911-1-christophe.jaillet@wanadoo.fr\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 7621c3ccffc5fac698ded69d1b1c7ca799903afb\nAuthor: Daniel Mack \u003cdaniel@zonque.org\u003e\nDate: Thu Jul 11 14:45:04 2019 +0200\n\n hwmon: (lm75) add support for PCT2075\n \n The NXP PCT2075 is largely compatible with other chips already supported\n by the LM75 driver. It uses an 11-bit resolution and defaults to 100 ms\n sampling period. The datasheet is here:\n \n https://www.nxp.com/docs/en/data-sheet/PCT2075.pdf\n \n Signed-off-by: Daniel Mack \u003cdaniel@zonque.org\u003e\n Link: https://lore.kernel.org/r/20190711124504.7580-2-daniel@zonque.org\n [groeck: Documentation update]\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 98cd95a59980e4c579f694d05844acac7cada4af\nAuthor: Daniel Mack \u003cdaniel@zonque.org\u003e\nDate: Thu Jul 11 14:45:03 2019 +0200\n\n device-tree: bindinds: add NXP PCT2075 as compatible device to LM75\n \n The PCT2075 is compatible to other chips that are already handled by\n the LM75 driver.\n \n Signed-off-by: Daniel Mack \u003cdaniel@zonque.org\u003e\n Link: https://lore.kernel.org/r/20190711124504.7580-1-daniel@zonque.org\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 5dffa3c05e9315d4633d343875314575ed719476\nAuthor: Guenter Roeck \u003clinux@roeck-us.net\u003e\nDate: Mon Jul 1 11:12:38 2019 -0700\n\n hwmon: Remove ads1015 driver\n \n A driver for ADS1015 with more functionality is available in the iio\n subsystem.\n \n Remove the hwmon driver as duplicate. If the chip is used for hardware\n monitoring, the iio-\u003ehwmon bridge should be used.\n \n Cc: Dirk Eibach \u003ceibach@gdsys.de\u003e\n Link: https://lore.kernel.org/r/1562004758-13025-1-git-send-email-linux@roeck-us.net\n Acked-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n Acked-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit e027a2dea5fdde05740a98f571fc2903060559ce\nAuthor: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nDate: Mon Aug 19 16:00:02 2019 -0500\n\n hwmon (coretemp) Fix a memory leak bug\n \n In coretemp_init(), \u0027zone_devices\u0027 is allocated through kcalloc().\n However, it is not deallocated in the following execution if\n platform_driver_register() fails, leading to a memory leak. To fix this\n issue, introduce the \u0027outzone\u0027 label to free \u0027zone_devices\u0027 before\n returning the error.\n \n Signed-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\n Link: https://lore.kernel.org/r/1566248402-6538-1-git-send-email-wenwen@cs.uga.edu\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 7d82fcc9d9e81241778aaa22fda7be753e237d86\nAuthor: Guenter Roeck \u003clinux@roeck-us.net\u003e\nDate: Thu Aug 8 12:00:18 2019 -0700\n\n hwmon: (lm75) Fix write operations for negative temperatures\n \n Writes into limit registers fail if the temperature written is negative.\n The regmap write operation checks the value range, regmap_write accepts\n an unsigned int as parameter, and the temperature value passed to\n regmap_write is kept in a variable declared as long. Negative values\n are converted large unsigned integers, which fails the range check.\n Fix by type casting the temperature to u16 when calling regmap_write().\n \n Cc: Iker Perez del Palomar Sustatxa \u003ciker.perez@codethink.co.uk\u003e\n Fixes: e65365fed87f (\"hwmon: (lm75) Convert to use regmap\")\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 6f8c8f3c31015808100ee54fc471ff5dffdf1734\nAuthor: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\nDate: Thu Aug 8 10:01:44 2019 +0200\n\n hwmon: pmbus: ucd9000: remove unneeded include\n \n Build bot reports the following build issue after commit 9091373ab7ea\n (\"gpio: remove less important #ifdef around declarations):\n \n In file included from drivers/hwmon/pmbus/ucd9000.c:19:0:\n \u003e\u003e include/linux/gpio/driver.h:576:1: error: redefinition of \u0027gpiochip_add_pin_range\u0027\n gpiochip_add_pin_range(struct gpio_chip *chip, const char *pinctl_name,\n ^~~~~~~~~~~~~~~~~~~~~~\n In file included from drivers/hwmon/pmbus/ucd9000.c:18:0:\n include/linux/gpio.h:245:1: note: previous definition of \u0027gpiochip_add_pin_range\u0027 was here\n gpiochip_add_pin_range(struct gpio_chip *chip, const char *pinctl_name,\n ^~~~~~~~~~~~~~~~~~~~~~\n In file included from drivers/hwmon/pmbus/ucd9000.c:19:0:\n \u003e\u003e include/linux/gpio/driver.h:583:1: error: redefinition of \u0027gpiochip_add_pingroup_range\u0027\n gpiochip_add_pingroup_range(struct gpio_chip *chip,\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~\n In file included from drivers/hwmon/pmbus/ucd9000.c:18:0:\n include/linux/gpio.h:254:1: note: previous definition of \u0027gpiochip_add_pingroup_range\u0027 was here\n gpiochip_add_pingroup_range(struct gpio_chip *chip,\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~\n In file included from drivers/hwmon/pmbus/ucd9000.c:19:0:\n \u003e\u003e include/linux/gpio/driver.h:591:1: error: redefinition of \u0027gpiochip_remove_pin_ranges\u0027\n gpiochip_remove_pin_ranges(struct gpio_chip *chip)\n ^~~~~~~~~~~~~~~~~~~~~~~~~~\n In file included from drivers/hwmon/pmbus/ucd9000.c:18:0:\n include/linux/gpio.h:263:1: note: previous definition of \u0027gpiochip_remove_pin_ranges\u0027 was here\n gpiochip_remove_pin_ranges(struct gpio_chip *chip)\n \n This is caused by conflicting defines from linux/gpio.h and\n linux/gpio/driver.h. Drivers should not include both the legacy and\n the new API headers. This driver doesn\u0027t even use linux/gpio.h so\n remove it.\n \n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Cc: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n Link: https://lore.kernel.org/r/20190808080144.6183-1-brgl@bgdev.pl\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 0642ea2409f3bfa105570e12854b8e2628db6835\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Sun Aug 4 17:56:43 2019 +0800\n\n ext4 crypto: fix to check feature status before get policy\n \n When getting fscrypt policy via EXT4_IOC_GET_ENCRYPTION_POLICY, if\n encryption feature is off, it\u0027s better to return EOPNOTSUPP instead of\n ENODATA, so let\u0027s add ext4_has_feature_encrypt() to do the check for\n that.\n \n This makes it so that all fscrypt ioctls consistently check for the\n encryption feature, and makes ext4 consistent with f2fs in this regard.\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n [EB - removed unneeded braces, updated the documentation, and\n added more explanation to commit message]\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 657fb9c742664d7a0dc1e7012cd6cbf7f67523ee\nAuthor: Alex Lu \u003calex_lu@realsil.com.cn\u003e\nDate: Sat Aug 31 16:46:11 2019 +0800\n\n Bluetooth: btrtl: Remove trailing newline from calls to rtl_dev macros\n \n These printing macros already add a trailing newline, so drop these\n unnecessary additional newlines.\n \n Signed-off-by: Alex Lu \u003calex_lu@realsil.com.cn\u003e\n Signed-off-by: Marcel Holtmann \u003cmarcel@holtmann.org\u003e\n\ncommit 51b4c991d0f63dbf596944739901339c347c2e30\nAuthor: Alex Lu \u003calex_lu@realsil.com.cn\u003e\nDate: Sat Aug 31 16:41:13 2019 +0800\n\n Bluetooth: btrtl: Remove redundant prefix from calls to rtl_dev macros\n \n the rtl: or RTL: prefix in the string is pointless. The rtl_dev_* macros\n already does that.\n \n Signed-off-by: Alex Lu \u003calex_lu@realsil.com.cn\u003e\n Signed-off-by: Marcel Holtmann \u003cmarcel@holtmann.org\u003e\n\ncommit 663f3ae2e512b50eb7a73dfaa5786009ad34af87\nAuthor: Alex Lu \u003calex_lu@realsil.com.cn\u003e\nDate: Sat Aug 31 16:36:02 2019 +0800\n\n Bluetooth: btrtl: Add firmware version print\n \n This patch is used to print fw version for debug convenience\n \n Signed-off-by: Alex Lu \u003calex_lu@realsil.com.cn\u003e\n Signed-off-by: Marcel Holtmann \u003cmarcel@holtmann.org\u003e\n\ncommit 42344113ba7a1ed7b5654cd5270af0d5698d8521\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Thu Jun 13 15:43:20 2019 +0200\n\n mips/atomic: Fix smp_mb__{before,after}_atomic()\n \n Recent probing at the Linux Kernel Memory Model uncovered a\n \u0027surprise\u0027. Strongly ordered architectures where the atomic RmW\n primitive implies full memory ordering and\n smp_mb__{before,after}_atomic() are a simple barrier() (such as MIPS\n without WEAK_REORDERING_BEYOND_LLSC) fail for:\n \n *x \u003d 1;\n atomic_inc(u);\n smp_mb__after_atomic();\n r0 \u003d *y;\n \n Because, while the atomic_inc() implies memory order, it\n (surprisingly) does not provide a compiler barrier. This then allows\n the compiler to re-order like so:\n \n atomic_inc(u);\n *x \u003d 1;\n smp_mb__after_atomic();\n r0 \u003d *y;\n \n Which the CPU is then allowed to re-order (under TSO rules) like:\n \n atomic_inc(u);\n r0 \u003d *y;\n *x \u003d 1;\n \n And this very much was not intended. Therefore strengthen the atomic\n RmW ops to include a compiler barrier.\n \n Reported-by: Andrea Parri \u003candrea.parri@amarulasolutions.com\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit 1c6c1ca318585f1096d4d04bc722297c85e9fb8a\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Thu Jun 13 15:43:19 2019 +0200\n\n mips/atomic: Fix loongson_llsc_mb() wreckage\n \n The comment describing the loongson_llsc_mb() reorder case doesn\u0027t\n make any sense what so ever. Instruction re-ordering is not an SMP\n artifact, but rather a CPU local phenomenon. Clarify the comment by\n explaining that these issue cause a coherence fail.\n \n For the branch speculation case; if futex_atomic_cmpxchg_inatomic()\n needs one at the bne branch target, then surely the normal\n __cmpxch_asm() implementation does too. We cannot rely on the\n barriers from cmpxchg() because cmpxchg_local() is implemented with\n the same macro, and branch prediction and speculation are, too, CPU\n local.\n \n Fixes: e02e07e3127d (\"MIPS: Loongson: Introduce and use loongson_llsc_mb()\")\n Cc: Huacai Chen \u003cchenhc@lemote.com\u003e\n Cc: Huang Pei \u003chuangpei@loongson.cn\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit dfc8d8de855d566eb83a27e58a69741de42a90da\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Thu Jun 13 15:43:18 2019 +0200\n\n mips/atomic: Fix cmpxchg64 barriers\n \n There were no memory barriers on the 32bit implementation of\n cmpxchg64(). Fix this.\n \n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit ae688e1720fd387de34f2140a735917411672bf1\nAuthor: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\nDate: Fri Aug 30 14:15:35 2019 +0100\n\n arm64: dts: renesas: hihope-common: Fix eMMC status\n \n SDHI3 got accidentally disabled while adding USB 2.0 support,\n this patch fixes it.\n \n Fixes: 734d277f412a (\"arm64: dts: renesas: hihope-common: Add USB 2.0 support\")\n Signed-off-by: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n\ncommit c563034309ada118606e8e76762562c37414abc0\nAuthor: Vidya Sagar \u003cvidyas@nvidia.com\u003e\nDate: Tue Aug 13 17:06:27 2019 +0530\n\n PCI: tegra: Add Tegra194 PCIe support\n \n Add support for Synopsys DesignWare core IP based PCIe host controller\n present in the Tegra194 SoC.\n \n Signed-off-by: Vidya Sagar \u003cvidyas@nvidia.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n\ncommit 7109d817db2e020379d0f245300b8ffe651c5c04\nAuthor: Marek Behún \u003cmarek.behun@nic.cz\u003e\nDate: Wed Aug 28 17:12:43 2019 +0200\n\n arm64: dts: marvell: add DTS for Turris Mox\n \n This adds support for the Turris Mox board from CZ.NIC.\n \n Turris Mox is as modular router based on the Armada 3720 SOC (same as\n EspressoBin).\n \n The basic board can be extended by different modules.\n If those are connected, U-Boot lets the kernel know via device-tree.\n \n Since modules can be connected in different order and some modules can\n be connected multiple times (up to three modules containing 8-port\n ethernet switch in DSA configuration can be connected) we decided\n against using device-tree overlays, because it got complicated rather\n quickly. (For example the SFP module can be connected directly to the\n CPU, or after a switch module. There are four cases and all would need\n different SFP overlay. There are two types of switch modules (8-port\n with pass-through and 4-port with no pass-through). For those we would\n again need at least 6 more overlays.)\n \n We therefore decided to put all the possibly connected devices in one\n device-tree and disable them by default. When U-Boot finds out which\n modules are connected, it fixes the loaded device-tree accordingly just\n before boot. By Rob Herring\u0027s suggestion we also made it so that U-Boot\n completely removes nodes which are disabled after this fixup.\n \n Signed-off-by: Marek Behún \u003cmarek.behun@nic.cz\u003e\n Cc: Rob Herring \u003crobh@kernel.org\u003e\n Cc: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n Cc: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n\ncommit 3e8e28f28e400818f23292ec1a9d3ce8035bee2d\nAuthor: Marek Behún \u003cmarek.behun@nic.cz\u003e\nDate: Wed Aug 28 17:12:42 2019 +0200\n\n dt-bindings: marvell: document Turris Mox compatible\n \n This adds the documentation for the Turris Mox compatible in armada-37xx\n device-tree binding documentation.\n \n Signed-off-by: Marek Behún \u003cmarek.behun@nic.cz\u003e\n Cc: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n Cc: Rob Herring \u003crobh@kernel.org\u003e\n Cc: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Signed-off-by: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n\ncommit 8ef75105b5b18d9f957c37cd736fee490a913ace\nAuthor: Marek Behún \u003cmarek.behun@nic.cz\u003e\nDate: Wed Aug 28 17:12:41 2019 +0200\n\n arm64: dts: marvell: armada-37xx: add SPI CS1 pinctrl\n \n This adds pinctrl node for the GPIO to be used as SPI chip select 1.\n \n Signed-off-by: Marek Behún \u003cmarek.behun@nic.cz\u003e\n Cc: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n Signed-off-by: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n\ncommit 1baa2800e62d7d58c04cb871a6e4189fc3e1ccf1\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Fri Aug 30 08:56:56 2019 -0700\n\n xfs: remove the unused XFS_ALLOC_USERDATA flag\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit ecfc28a41cf108c112ba5a06a83b52fd6deb3129\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Fri Aug 30 08:56:55 2019 -0700\n\n xfs: cleanup xfs_fsb_to_db\n \n This function isn\u0027t a macro anymore, so remove various superflous braces,\n and explicit cast that is done implicitly due to the return value, use\n a normal if statement instead of trying to squeeze everything together.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit adcb0ca2330b1d88533e70026ce3f9cc5e2253b6\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Fri Aug 30 08:56:55 2019 -0700\n\n xfs: fix the dax supported check in xfs_ioctl_setattr_dax_invalidate\n \n Setting the DAX flag on the directory of a file system that is not on a\n DAX capable device makes as little sense as setting it on a regular file\n on the same file system.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit 40144e49ff84c3bd6bd091b58115257670be8803\nAuthor: Jan Kara \u003cjack@suse.cz\u003e\nDate: Thu Aug 29 09:04:12 2019 -0700\n\n xfs: Fix stale data exposure when readahead races with hole punch\n \n Hole puching currently evicts pages from page cache and then goes on to\n remove blocks from the inode. This happens under both XFS_IOLOCK_EXCL\n and XFS_MMAPLOCK_EXCL which provides appropriate serialization with\n racing reads or page faults. However there is currently nothing that\n prevents readahead triggered by fadvise() or madvise() from racing with\n the hole punch and instantiating page cache page after hole punching has\n evicted page cache in xfs_flush_unmap_range() but before it has removed\n blocks from the inode. This page cache page will be mapping soon to be\n freed block and that can lead to returning stale data to userspace or\n even filesystem corruption.\n \n Fix the problem by protecting handling of readahead requests by\n XFS_IOLOCK_SHARED similarly as we protect reads.\n \n CC: stable@vger.kernel.org\n Link: https://lore.kernel.org/linux-fsdevel/CAOQ4uxjQNmxqmtA_VbYW0Su9rKRk2zobJmahcyeaEVOFKVQ5dw@mail.gmail.com/\n Reported-by: Amir Goldstein \u003camir73il@gmail.com\u003e\n Signed-off-by: Jan Kara \u003cjack@suse.cz\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit cf1ea0592dbf109e7e7935b7d5b1a47a1ba04174\nAuthor: Jan Kara \u003cjack@suse.cz\u003e\nDate: Thu Aug 29 09:04:11 2019 -0700\n\n fs: Export generic_fadvise()\n \n Filesystems will need to call this function from their fadvise handlers.\n \n CC: stable@vger.kernel.org\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Jan Kara \u003cjack@suse.cz\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit 692fe62433d4ca47605b39f7c416efd6679ba694\nAuthor: Jan Kara \u003cjack@suse.cz\u003e\nDate: Thu Aug 29 09:04:11 2019 -0700\n\n mm: Handle MADV_WILLNEED through vfs_fadvise()\n \n Currently handling of MADV_WILLNEED hint calls directly into readahead\n code. Handle it by calling vfs_fadvise() instead so that filesystem can\n use its -\u003efadvise() callback to acquire necessary locks or otherwise\n prepare for the request.\n \n Suggested-by: Amir Goldstein \u003camir73il@gmail.com\u003e\n Reviewed-by: Boaz Harrosh \u003cboazh@netapp.com\u003e\n CC: stable@vger.kernel.org\n Signed-off-by: Jan Kara \u003cjack@suse.cz\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit ddbca70cc45c0ac97ff6d9529e45f10b8ae73ad4\nAuthor: Dave Chinner \u003cdchinner@redhat.com\u003e\nDate: Thu Aug 29 09:04:10 2019 -0700\n\n xfs: allocate xattr buffer on demand\n \n When doing file lookups and checking for permissions, we end up in\n xfs_get_acl() to see if there are any ACLs on the inode. This\n requires and xattr lookup, and to do that we have to supply a buffer\n large enough to hold an maximum sized xattr.\n \n On workloads were we are accessing a wide range of cache cold files\n under memory pressure (e.g. NFS fileservers) we end up spending a\n lot of time allocating the buffer. The buffer is 64k in length, so\n is a contiguous multi-page allocation, and if that then fails we\n fall back to vmalloc(). Hence the allocation here is /expensive/\n when we are looking up hundreds of thousands of files a second.\n \n Initial numbers from a bpf trace show average time in xfs_get_acl()\n is ~32us, with ~19us of that in the memory allocation. Note these\n are average times, so there are going to be affected by the worst\n case allocations more than the common fast case...\n \n To avoid this, we could just do a \"null\" lookup to see if the ACL\n xattr exists and then only do the allocation if it exists. This,\n however, optimises the path for the \"no ACL present\" case at the\n expense of the \"acl present\" case. i.e. we can halve the time in\n xfs_get_acl() for the no acl case (i.e down to ~10-15us), but that\n then increases the ACL case by 30% (i.e. up to 40-45us).\n \n To solve this and speed up both cases, drive the xattr buffer\n allocation into the attribute code once we know what the actual\n xattr length is. For the no-xattr case, we avoid the allocation\n completely, speeding up that case. For the common ACL case, we\u0027ll\n end up with a fast heap allocation (because it\u0027ll be smaller than a\n page), and only for the rarer \"we have a remote xattr\" will we have\n a multi-page allocation occur. Hence the common ACL case will be\n much faster, too.\n \n Signed-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit 9df243a1a9e607e7cf5d20ee46edd5ec84b7e400\nAuthor: Dave Chinner \u003cdchinner@redhat.com\u003e\nDate: Thu Aug 29 09:04:10 2019 -0700\n\n xfs: consolidate attribute value copying\n \n The same code is used to copy do the attribute copying in three\n different places. Consolidate them into a single function in\n preparation from on-demand buffer allocation.\n \n Signed-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit e3cc4554ce1b2cc43262c93ebe6d8cde780d29ee\nAuthor: Dave Chinner \u003cdchinner@redhat.com\u003e\nDate: Thu Aug 29 09:04:09 2019 -0700\n\n xfs: move remote attr retrieval into xfs_attr3_leaf_getvalue\n \n Because we repeat exactly the same code to get the remote attribute\n value after both calls to xfs_attr3_leaf_getvalue() if it\u0027s a remote\n attr. Just do it in xfs_attr3_leaf_getvalue() so the callers don\u0027t\n have to care about it.\n \n Signed-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit a0e959d3c9d52a83e01a97ef352a1465550f9329\nAuthor: Dave Chinner \u003cdchinner@redhat.com\u003e\nDate: Thu Aug 29 09:04:09 2019 -0700\n\n xfs: remove unnecessary indenting from xfs_attr3_leaf_getvalue\n \n Signed-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit 728bcaa3e0f9820e923dabe5e4c34e87100bb995\nAuthor: Dave Chinner \u003cdchinner@redhat.com\u003e\nDate: Thu Aug 29 09:04:08 2019 -0700\n\n xfs: make attr lookup returns consistent\n \n Shortform, leaf and remote value attr value retrieval return\n different values for success. This makes it more complex to handle\n actual errors xfs_attr_get() as some errors mean success and some\n mean failure. Make the return values consistent for success and\n failure consistent for all attribute formats.\n \n Signed-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit 756c6f0f7efe8759ff6dda35d220e2e753e2b0e3\nAuthor: Dave Chinner \u003cdchinner@redhat.com\u003e\nDate: Thu Aug 29 09:04:08 2019 -0700\n\n xfs: reverse search directory freespace indexes\n \n When a directory is growing rapidly, new blocks tend to get added at\n the end of the directory. These end up at the end of the freespace\n index, and when the directory gets large finding these new\n freespaces gets expensive. The code does a linear search across the\n frespace index from the first block in the directory to the last,\n hence meaning the newly added space is the last index searched.\n \n Instead, do a reverse order index search, starting from the last\n block and index in the freespace index. This makes most lookups for\n free space on rapidly growing directories O(1) instead of O(N), but\n should not have any impact on random insert workloads because the\n average search length is the same regardless of which end of the\n array we start at.\n \n The result is a major improvement in large directory grow rates:\n \n create time(sec) / rate (files/s)\n File count vanilla Prev commit Patched\n 10k 0.41 / 24.3k 0.42 / 23.8k 0.41 / 24.3k\n 20k 0.74 / 27.0k 0.76 / 26.3k 0.75 / 26.7k\n 100k 3.81 / 26.4k 3.47 / 28.8k 3.27 / 30.6k\n 200k 8.58 / 23.3k 7.19 / 27.8k 6.71 / 29.8k\n 1M 85.69 / 11.7k 48.53 / 20.6k 37.67 / 26.5k\n 2M 280.31 / 7.1k 130.14 / 15.3k 79.55 / 25.2k\n 10M 3913.26 / 2.5k 552.89 / 18.1k\n \n Signed-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit 610125ab1e4b1b48dcffe74d9d82b0606bf1b923\nAuthor: Dave Chinner \u003cdchinner@redhat.com\u003e\nDate: Thu Aug 29 09:04:07 2019 -0700\n\n xfs: speed up directory bestfree block scanning\n \n When running a \"create millions inodes in a directory\" test\n recently, I noticed we were spending a huge amount of time\n converting freespace block headers from disk format to in-memory\n format:\n \n 31.47% [kernel] [k] xfs_dir2_node_addname\n 17.86% [kernel] [k] xfs_dir3_free_hdr_from_disk\n 3.55% [kernel] [k] xfs_dir3_free_bests_p\n \n We shouldn\u0027t be hitting the best free block scanning code so hard\n when doing sequential directory creates, and it turns out there\u0027s\n a highly suboptimal loop searching the the best free array in\n the freespace block - it decodes the block header before checking\n each entry inside a loop, instead of decoding the header once before\n running the entry search loop.\n \n This makes a massive difference to create rates. Profile now looks\n like this:\n \n 13.15% [kernel] [k] xfs_dir2_node_addname\n 3.52% [kernel] [k] xfs_dir3_leaf_check_int\n 3.11% [kernel] [k] xfs_log_commit_cil\n \n And the wall time/average file create rate differences are\n just as stark:\n \n create time(sec) / rate (files/s)\n File count vanilla patched\n 10k 0.41 / 24.3k 0.42 / 23.8k\n 20k 0.74 / 27.0k 0.76 / 26.3k\n 100k 3.81 / 26.4k 3.47 / 28.8k\n 200k 8.58 / 23.3k 7.19 / 27.8k\n 1M 85.69 / 11.7k 48.53 / 20.6k\n 2M 280.31 / 7.1k 130.14 / 15.3k\n \n The larger the directory, the bigger the performance improvement.\n \n Signed-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit 0e822255f95db400e56e19773e04755d79b50bda\nAuthor: Dave Chinner \u003cdchinner@redhat.com\u003e\nDate: Thu Aug 29 09:04:07 2019 -0700\n\n xfs: factor free block index lookup from xfs_dir2_node_addname_int()\n \n Simplify the logic in xfs_dir2_node_addname_int() by factoring out\n the free block index lookup code that finds a block with enough free\n space for the entry to be added. The code that is moved gets a major\n cleanup at the same time, but there is no algorithm change here.\n \n Signed-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit a07258a695281109422011182db069a0e6f8855e\nAuthor: Dave Chinner \u003cdchinner@redhat.com\u003e\nDate: Thu Aug 29 09:04:06 2019 -0700\n\n xfs: factor data block addition from xfs_dir2_node_addname_int()\n \n Factor out the code that adds a data block to a directory from\n xfs_dir2_node_addname_int(). This makes the code flow cleaner and\n more obvious and provides clear isolation of upcoming optimsations.\n \n Signed-off-By: Dave Chinner \u003cdchinner@redhat.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit aee7754bbeb1000992ed7e6945418a145d181074\nAuthor: Dave Chinner \u003cdchinner@redhat.com\u003e\nDate: Thu Aug 29 09:04:06 2019 -0700\n\n xfs: move xfs_dir2_addname()\n \n This gets rid of the need for a forward declaration of the static\n function xfs_dir2_addname_int() and readies the code for factoring\n of xfs_dir2_addname_int().\n \n Signed-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit 39ee2239a5a212cbba9d96050bd11cfd0f26634d\nAuthor: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nDate: Wed Aug 28 14:39:46 2019 -0700\n\n xfs: remove all *_ITER_CONTINUE values\n \n Iterator functions already use 0 to signal \"continue iterating\", so get\n rid of the #defines and just do it directly.\n \n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Reviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n\ncommit f3b17320db25b4cdd50f0396b096644455357dac\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Sat Aug 17 09:32:32 2019 +0200\n\n openrisc: map as uncached in ioremap\n \n Openrisc is the only architecture not mapping ioremap as uncached,\n which has been the default since the Linux 2.6.x days. Switch it\n over to implement uncached semantics by default.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Stafford Horne \u003cshorne@gmail.com\u003e\n\ncommit eabe7e9a21a27c6124967169e5bfea5678af493f\nAuthor: Stafford Horne \u003cshorne@gmail.com\u003e\nDate: Sat Aug 31 10:02:10 2019 +0900\n\n or1k: dts: Add ethoc device to SMP devicetree\n \n This patch adds the ethoc device configuration to the OpenRISC basic SMP\n device tree config. This was tested with qemu.\n \n Signed-off-by: Stafford Horne \u003cshorne@gmail.com\u003e\n\ncommit c312ef176399e04fc5f7f2809d9a589751fbf6d9\nAuthor: Dan Williams \u003cdan.j.williams@intel.com\u003e\nDate: Thu Aug 29 16:30:34 2019 -0700\n\n libata/ahci: Drop PCS quirk for Denverton and beyond\n \n The Linux ahci driver has historically implemented a configuration fixup\n for platforms / platform-firmware that fails to enable the ports prior\n to OS hand-off at boot. The fixup was originally implemented way back\n before ahci moved from drivers/scsi/ to drivers/ata/, and was updated in\n 2007 via commit 49f290903935 \"ahci: update PCS programming\". The quirk\n sets a port-enable bitmap in the PCS register at offset 0x92.\n \n This quirk could be applied generically up until the arrival of the\n Denverton (DNV) platform. The DNV AHCI controller architecture supports\n more than 6 ports and along with that the PCS register location and\n format were updated to allow for more possible ports in the bitmap. DNV\n AHCI expands the register to 32-bits and moves it to offset 0x94.\n \n As it stands there are no known problem reports with existing Linux\n trying to set bits at offset 0x92 which indicates that the quirk is not\n applicable. Likely it is not applicable on a wider range of platforms,\n but it is difficult to discern which platforms if any still depend on\n the quirk.\n \n Rather than try to fix the PCS quirk to consider the DNV register layout\n instead require explicit opt-in. The assumption is that the OS driver\n need not touch this register, and platforms can be added with a new\n boad_ahci_pcs7 board-id when / if problematic platforms are found in the\n future. The logic in ahci_intel_pcs_quirk() looks for all Intel AHCI\n instances with \"legacy\" board-ids and otherwise skips the quirk if the\n board was matched by class-code.\n \n Reported-by: Stephen Douthit \u003cstephend@silicom-usa.com\u003e\n Cc: Christoph Hellwig \u003chch@infradead.org\u003e\n Reviewed-by: Stephen Douthit \u003cstephend@silicom-usa.com\u003e\n Signed-off-by: Dan Williams \u003cdan.j.williams@intel.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 974ceb21fcf9c0345570ea194e799a4ee7842f33\nAuthor: Denis Efremov \u003cefremov@linux.com\u003e\nDate: Thu Aug 29 19:50:24 2019 +0300\n\n udp: Remove unlikely() from IS_ERR*() condition\n \n \"unlikely(IS_ERR_OR_NULL(x))\" is excessive. IS_ERR_OR_NULL() already uses\n unlikely() internally.\n \n Signed-off-by: Denis Efremov \u003cefremov@linux.com\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: Joe Perches \u003cjoe@perches.com\u003e\n Cc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n Cc: netdev@vger.kernel.org\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7cf92ccb85554c9550bc0a8e892f68f92985024c\nAuthor: Denis Efremov \u003cefremov@linux.com\u003e\nDate: Thu Aug 29 19:50:17 2019 +0300\n\n net/mlx5e: Remove unlikely() from WARN*() condition\n \n \"unlikely(WARN_ON_ONCE(x))\" is excessive. WARN_ON_ONCE() already uses\n unlikely() internally.\n \n Signed-off-by: Denis Efremov \u003cefremov@linux.com\u003e\n Cc: Boris Pismenny \u003cborisp@mellanox.com\u003e\n Cc: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n Cc: Leon Romanovsky \u003cleon@kernel.org\u003e\n Cc: Joe Perches \u003cjoe@perches.com\u003e\n Cc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n Cc: netdev@vger.kernel.org\n Acked-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ae2930583c7ae2159e26a54b15e52b1d4d5782a0\nAuthor: Stafford Horne \u003cshorne@gmail.com\u003e\nDate: Sat Aug 31 10:00:34 2019 +0900\n\n or1k: dts: Fix ethoc network configuration in or1ksim devicetree\n \n This fixes several issues with the ethoc network device config.\n \n Fisrt off, the compatible property used an obsolete compatibility\n string; this caused the initialization to be skipped. Next, the\n register map was not given enough space to allocate ring descriptors,\n this caused module initialization to abort. Finally, we need to mark\n this device as big-endian as needed by openrisc.\n \n This was tested by me in qemu, the setup is documented on the qemu wiki:\n \n https://wiki.qemu.org/Documentation/Platforms/OpenRISC\n \n Signed-off-by: Stafford Horne \u003cshorne@gmail.com\u003e\n\ncommit 0feacaa21634014148068035b02eade71f853496\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Fri Aug 30 16:39:54 2019 -0700\n\n writeback: don\u0027t access page-\u003emapping directly in track_foreign_dirty TP\n \n page-\u003emapping may encode different values in it and page_mapping()\n should always be used to access the mapping pointer.\n track_foreign_dirty tracepoint was incorrectly accessing page-\u003emapping\n directly. Use page_mapping() instead. Also, add NULL checks while at\n it.\n \n Fixes: 3a8e9ac89e6a (\"writeback: add tracepoints for cgroup foreign writebacks\")\n Reported-by: Jan Kara \u003cjack@suse.cz\u003e\n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit ee594bfff389aa9105f713135211c0da736e5698\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Tue Jul 16 21:05:36 2019 -0400\n\n fs/namei.c: new helper - legitimize_root()\n \n identical logics in unlazy_walk() and unlazy_child()\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit ce6595a28a15c874aee374757dcd08f537d7b24d\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Sun Jul 14 16:42:44 2019 -0400\n\n kill the last users of user_{path,lpath,path_dir}()\n \n old wrappers with few callers remaining; put them out of their misery...\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 6b61aed06a3b0eb8e8d49142ddd1e101a064444d\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Sun Jul 14 14:00:23 2019 -0400\n\n namei.h: get the comments on LOOKUP_... in sync with reality\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit fbb7d9d56d167247f2eb7261038385cab1073c37\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Sun Jul 14 13:24:54 2019 -0400\n\n kill LOOKUP_NO_EVAL, don\u0027t bother including namei.h from audit.h\n \n The former has no users left; the latter was only to get LOOKUP_...\n values to remapper in audit_inode() and that\u0027s an ex-parrot now.\n \n All places that use symbols from namei.h include it either directly\n or (in a few cases) via a local header, like fs/autofs/autofs_i.h\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit e72cb7d6245380acc11a24b75a865f7104ac8b33\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Fri Aug 30 19:10:38 2019 -0400\n\n bnxt_en: Fix compile error regression with CONFIG_BNXT_SRIOV not set.\n \n Add a new function bnxt_get_registered_vfs() to handle the work\n of getting the number of registered VFs under #ifdef CONFIG_BNXT_SRIOV.\n The main code will call this function and will always work correctly\n whether CONFIG_BNXT_SRIOV is set or not.\n \n Fixes: 230d1f0de754 (\"bnxt_en: Handle firmware reset.\")\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 909548d6c5789dbd6cd4c7a6b32fcf0ae974f5ee\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Wed Aug 21 18:26:58 2019 +0900\n\n riscv: add arch/riscv/Kbuild\n \n Use the standard obj-y form to specify the sub-directories under\n arch/riscv/. No functional change intended.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Paul Walmsley \u003cpaul.walmsley@sifive.com\u003e\n\ncommit 9659ef791c3f17f24631ee9c8808f6af181c9ef5\nAuthor: Anders Roxell \u003canders.roxell@linaro.org\u003e\nDate: Wed Aug 14 13:14:35 2019 +0200\n\n selftests: tpm2: install python files\n \n Both test_smoke.sh and test_space.sh require tpm2.py and tpm2_test.py.\n \n Rework so that tpm2.py and tpm2_test.py gets installed, added them to\n the variable TEST_PROGS_EXTENDED.\n \n Signed-off-by: Anders Roxell \u003canders.roxell@linaro.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit 0aa3769d0a99eaee85ed89972e5c560edf61e6f4\nAuthor: Anders Roxell \u003canders.roxell@linaro.org\u003e\nDate: Wed Aug 14 13:16:51 2019 +0200\n\n selftests: livepatch: add missing fragments to config\n \n When generating config with \u0027make defconfig kselftest-merge\u0027 fragment\n CONFIG_TEST_LIVEPATCH\u003dm isn\u0027t set.\n \n Rework to enable CONFIG_LIVEPATCH and CONFIG_DYNAMIC_DEBUG as well.\n \n Signed-off-by: Anders Roxell \u003canders.roxell@linaro.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit 037caee7779f830e2a133a70dfbe4e8868377723\nAuthor: George G. Davis \u003cgeorge_davis@mentor.com\u003e\nDate: Fri Aug 30 18:31:33 2019 -0400\n\n selftests: watchdog: cleanup whitespace in usage options\n \n Convert hard spaces to tabs in usage options.\n \n Suggested-by: Shuah Khan \u003cshuah@kernel.org\u003e\n Signed-off-by: George G. Davis \u003cgeorge_davis@mentor.com\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit f2683bd8d5bdebb929f05ae26ce6d9b578927ce5\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Fri Aug 30 19:31:09 2019 -0400\n\n [PATCH] fix d_absolute_path() interplay with fsmount()\n \n stuff in anon namespace should be treated as unattached.\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit d57f172c99bd252fbfc93a9b38d8d42642a4b377\nAuthor: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\nDate: Tue Aug 27 02:25:31 2019 +0000\n\n doc/af_xdp: include unaligned chunk case\n \n The addition of unaligned chunks mode, the documentation needs to be\n updated to indicate that the incoming addr to the fill ring will only be\n masked if the user application is run in the aligned chunk mode. This patch\n also adds a line to explicitly indicate that the incoming addr will not be\n masked if running the user application in the unaligned chunk mode.\n \n Signed-off-by: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 3945b37a975d22bed33b525f2933bac82424793d\nAuthor: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\nDate: Tue Aug 27 02:25:30 2019 +0000\n\n samples/bpf: use hugepages in xdpsock app\n \n This patch modifies xdpsock to use mmap instead of posix_memalign. With\n this change, we can use hugepages when running the application in unaligned\n chunks mode. Using hugepages makes it more likely that we have physically\n contiguous memory, which supports the unaligned chunk mode better.\n \n Signed-off-by: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 03895e63ff97447e5eab1b71ed46e3662a45bfe8\nAuthor: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\nDate: Tue Aug 27 02:25:29 2019 +0000\n\n samples/bpf: add buffer recycling for unaligned chunks to xdpsock\n \n This patch adds buffer recycling support for unaligned buffers. Since we\n don\u0027t mask the addr to 2k at umem_reg in unaligned mode, we need to make\n sure we give back the correct (original) addr to the fill queue. We achieve\n this using the new descriptor format and associated masks. The new format\n uses the upper 16-bits for the offset and the lower 48-bits for the addr.\n Since we have a field for the offset, we no longer need to modify the\n actual address. As such, all we have to do to get back the original address\n is mask for the lower 48 bits (i.e. strip the offset and we get the address\n on it\u0027s own).\n \n Signed-off-by: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\n Signed-off-by: Bruce Richardson \u003cbruce.richardson@intel.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit c543f54698229b50078b72760b9e8a4731cffa03\nAuthor: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\nDate: Tue Aug 27 02:25:28 2019 +0000\n\n samples/bpf: add unaligned chunks mode support to xdpsock\n \n This patch adds support for the unaligned chunks mode. The addition of the\n unaligned chunks option will allow users to run the application with more\n relaxed chunk placement in the XDP umem.\n \n Unaligned chunks mode can be used with the \u0027-u\u0027 or \u0027--unaligned\u0027 command\n line options.\n \n Signed-off-by: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\n Signed-off-by: Ciara Loftus \u003cciara.loftus@intel.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 10d30e301732636d93d7dcd2e0e6cd34d0454509\nAuthor: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\nDate: Tue Aug 27 02:25:27 2019 +0000\n\n libbpf: add flags to umem config\n \n This patch adds a \u0027flags\u0027 field to the umem_config and umem_reg structs.\n This will allow for more options to be added for configuring umems.\n \n The first use for the flags field is to add a flag for unaligned chunks\n mode. These flags can either be user-provided or filled with a default.\n \n Since we change the size of the xsk_umem_config struct, we need to version\n the ABI. This patch includes the ABI versioning for xsk_umem__create. The\n Makefile was also updated to handle multiple function versions in\n check-abi.\n \n Signed-off-by: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\n Signed-off-by: Ciara Loftus \u003cciara.loftus@intel.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 282c0c798f8ec883c2ac2f1ce2dc06ef9421731c\nAuthor: Maxim Mikityanskiy \u003cmaximmi@mellanox.com\u003e\nDate: Tue Aug 27 02:25:26 2019 +0000\n\n net/mlx5e: Allow XSK frames smaller than a page\n \n Relax the requirements to the XSK frame size to allow it to be smaller\n than a page and even not a power of two. The current implementation can\n work in this mode, both with Striding RQ and without it.\n \n The code that checks `mtu + headroom \u003c\u003d XSK frame size` is modified\n accordingly. Any frame size between 2048 and PAGE_SIZE is accepted.\n \n Functions that worked with pages only now work with XSK frames, even if\n their size is different from PAGE_SIZE.\n \n With XSK queues, regardless of the frame size, Striding RQ uses the\n stride size of PAGE_SIZE, and UMR MTTs are posted using starting\n addresses of frames, but PAGE_SIZE as page size. MTU guarantees that no\n packet data will overlap with other frames. UMR MTT size is made equal\n to the stride size of the RQ, because UMEM frames may come in random\n order, and we need to handle them one by one. PAGE_SIZE is just a power\n of two that is bigger than any allowed XSK frame size, and also it\n doesn\u0027t require making additional changes to the code.\n \n Signed-off-by: Maxim Mikityanskiy \u003cmaximmi@mellanox.com\u003e\n Reviewed-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit beb3e4b29530d54f2ef956acfb4999c775f10519\nAuthor: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\nDate: Tue Aug 27 02:25:25 2019 +0000\n\n mlx5e: modify driver for handling offsets\n \n With the addition of the unaligned chunks option, we need to make sure we\n handle the offsets accordingly based on the mode we are currently running\n in. This patch modifies the driver to appropriately mask the address for\n each case.\n \n Signed-off-by: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit d8c3061e5edd3b1f32c32fb9ebe11fad3c1bce82\nAuthor: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\nDate: Tue Aug 27 02:25:24 2019 +0000\n\n ixgbe: modify driver for handling offsets\n \n With the addition of the unaligned chunks option, we need to make sure we\n handle the offsets accordingly based on the mode we are currently running\n in. This patch modifies the driver to appropriately mask the address for\n each case.\n \n Signed-off-by: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 2f86c806a8a89f34c463705d00420acf49302162\nAuthor: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\nDate: Tue Aug 27 02:25:23 2019 +0000\n\n i40e: modify driver for handling offsets\n \n With the addition of the unaligned chunks option, we need to make sure we\n handle the offsets accordingly based on the mode we are currently running\n in. This patch modifies the driver to appropriately mask the address for\n each case.\n \n Signed-off-by: Bruce Richardson \u003cbruce.richardson@intel.com\u003e\n Signed-off-by: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit c05cd3645814724bdeb32a2b4d953b12bdea5f8c\nAuthor: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\nDate: Tue Aug 27 02:25:22 2019 +0000\n\n xsk: add support to allow unaligned chunk placement\n \n Currently, addresses are chunk size aligned. This means, we are very\n restricted in terms of where we can place chunk within the umem. For\n example, if we have a chunk size of 2k, then our chunks can only be placed\n at 0,2k,4k,6k,8k... and so on (ie. every 2k starting from 0).\n \n This patch introduces the ability to use unaligned chunks. With these\n changes, we are no longer bound to having to place chunks at a 2k (or\n whatever your chunk size is) interval. Since we are no longer dealing with\n aligned chunks, they can now cross page boundaries. Checks for page\n contiguity have been added in order to keep track of which pages are\n followed by a physically contiguous page.\n \n Signed-off-by: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\n Signed-off-by: Ciara Loftus \u003cciara.loftus@intel.com\u003e\n Signed-off-by: Bruce Richardson \u003cbruce.richardson@intel.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit b35a2d3e895600d8a1273b44f68128491faab3b8\nAuthor: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\nDate: Tue Aug 27 02:25:21 2019 +0000\n\n ixgbe: simplify Rx buffer recycle\n \n Currently, the dma, addr and handle are modified when we reuse Rx buffers\n in zero-copy mode. However, this is not required as the inputs to the\n function are copies, not the original values themselves. As we use the\n copies within the function, we can use the original \u0027obi\u0027 values\n directly without having to mask and add the headroom.\n \n Signed-off-by: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 10912fc9faa1da6cb49996810d55242e82b46896\nAuthor: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\nDate: Tue Aug 27 02:25:20 2019 +0000\n\n i40e: simplify Rx buffer recycle\n \n Currently, the dma, addr and handle are modified when we reuse Rx buffers\n in zero-copy mode. However, this is not required as the inputs to the\n function are copies, not the original values themselves. As we use the\n copies within the function, we can use the original \u0027old_bi\u0027 values\n directly without having to mask and add the headroom.\n \n Signed-off-by: Kevin Laatz \u003ckevin.laatz@intel.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 1c6d6e021c452ac15ce034e6762f0477f1cf7f19\nAuthor: Masanari Iida \u003cstandby24x7@gmail.com\u003e\nDate: Thu Aug 29 09:01:30 2019 +0900\n\n selftests/bpf: Fix a typo in test_offload.py\n \n This patch fix a spelling typo in test_offload.py\n \n Signed-off-by: Masanari Iida \u003cstandby24x7@gmail.com\u003e\n Acked-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 0741be358d5adc266244f909e4434fe17fe1a109\nAuthor: Petar Penkov \u003cppenkov@google.com\u003e\nDate: Tue Aug 27 16:46:22 2019 -0700\n\n bpf: fix error check in bpf_tcp_gen_syncookie\n \n If a SYN cookie is not issued by tcp_v#_gen_syncookie, then the return\n value will be exactly 0, rather than \u003c\u003d 0. Let\u0027s change the check to\n reflect that, especially since mss is an unsigned value and cannot be\n negative.\n \n Fixes: 70d66244317e (\"bpf: add bpf_tcp_gen_syncookie helper\")\n Reported-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Petar Penkov \u003cppenkov@google.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit f24e29099f4faf1969151fb4b857fda16b7c1a72\nAuthor: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\nDate: Tue Aug 27 22:36:29 2019 -0700\n\n nfp: bpf: add simple map op cache\n \n Each get_next and lookup call requires a round trip to the device.\n However, the device is capable of giving us a few entries back,\n instead of just one.\n \n In this patch we ask for a small yet reasonable number of entries\n (4) on every get_next call, and on subsequent get_next/lookup calls\n check this little cache for a hit. The cache is only kept for 250us,\n and is invalidated on every operation which may modify the map\n (e.g. delete or update call). Note that operations may be performed\n simultaneously, so we have to keep track of operations in flight.\n \n Signed-off-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Reviewed-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit bc2796db5a0246acc998cc5ab2e7f8d6e4e4c146\nAuthor: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\nDate: Tue Aug 27 22:36:28 2019 -0700\n\n nfp: bpf: rework MTU checking\n \n If control channel MTU is too low to support map operations a warning\n will be printed. This is not enough, we want to make sure probe fails\n in such scenario, as this would clearly be a faulty configuration.\n \n Signed-off-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Reviewed-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 3d1da92baffe69cf847699ceccf4297356da58fa\nAuthor: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\nDate: Mon Aug 26 15:55:40 2019 -0700\n\n drm/i915: Add 324mhz and 326.4mhz cdclks for gen11+\n \n The bspec was recently updated with these new cdclk values for ICL, EHL,\n and TGL.\n \n Bspec: 20598\n Bspec: 49201\n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Signed-off-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190826225540.11987-3-matthew.d.roper@intel.com\n\ncommit 385ba629aa1cdca5373c1fdbf6693ade5062ad27\nAuthor: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\nDate: Thu Aug 29 17:48:28 2019 -0700\n\n drm/i915: Allow /2 CD2X divider on gen11+\n \n The bspec has just recently been updated with new cdclk values that\n require the use of a /2 CD2X divider rather than a /1 divider. Once we\n add the divider selection logic to ICL+ cdclk programming, we have\n pretty much the same logic we were already using on CNL, so it\u0027s simpler\n to drop icl_set_cdclk() completely and reuse cnl_set_cdclk() on gen11+\n platforms as well.\n \n v2:\n - Using ICL_CDCLK_CD2X_PIPE_NONE + BXT_CDCLK_CD2X_PIPE(pipe) for TGL is\n correct, but looks really confusing. Add some TGL_ macros that alias\n these to avoid confusion. (Ville)\n - Use DIV_ROUND_CLOSEST rather than / when applying the divider. (Ville)\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190830004828.19359-1-matthew.d.roper@intel.com\n\ncommit 5b84ad2e89215835ea6476ba05295e849b30886b\nAuthor: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\nDate: Fri Aug 30 12:00:40 2019 +0100\n\n tools: bpftool: do not link twice against libbpf.a in Makefile\n \n In bpftool\u0027s Makefile, $(LIBS) includes $(LIBBPF), therefore the library\n is used twice in the linking command. No need to have $(LIBBPF) (from\n $^) on that command, let\u0027s do with \"$(OBJS) $(LIBS)\" (but move $(LIBBPF)\n _before_ the -l flags in $(LIBS)).\n \n Signed-off-by: Ilya Leoshkevich \u003ciii@linux.ibm.com\u003e\n Signed-off-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit fbdb620b7c674928a5713dbad9b07252ff929d0a\nAuthor: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\nDate: Fri Aug 30 12:00:39 2019 +0100\n\n tools: bpf: account for generated feature/ and libbpf/ directories\n \n When building \"tools/bpf\" from the top of the Linux repository, the\n build system passes a value for the $(OUTPUT) Makefile variable to\n tools/bpf/Makefile and tools/bpf/bpftool/Makefile, which results in\n generating \"libbpf/\" (for bpftool) and \"feature/\" (bpf and bpftool)\n directories inside the tree.\n \n This commit adds such directories to the relevant .gitignore files, and\n edits the Makefiles to ensure they are removed on \"make clean\". The use\n of \"rm\" is also made consistent throughout those Makefiles (relies on\n the $(RM) variable, use \"--\" to prevent interpreting\n $(OUTPUT)/$(DESTDIR) as options.\n \n v2:\n - New patch.\n \n Signed-off-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 45c5589d07158eb1d0b5b313314706024bf451c5\nAuthor: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\nDate: Fri Aug 30 12:00:38 2019 +0100\n\n tools: bpftool: improve and check builds for different make invocations\n \n There are a number of alternative \"make\" invocations that can be used to\n compile bpftool. The following invocations are expected to work:\n \n - through the kbuild system, from the top of the repository\n (make tools/bpf)\n - by telling make to change to the bpftool directory\n (make -C tools/bpf/bpftool)\n - by building the BPF tools from tools/\n (cd tools \u0026\u0026 make bpf)\n - by running make from bpftool directory\n (cd tools/bpf/bpftool \u0026\u0026 make)\n \n Additionally, setting the O or OUTPUT variables should tell the build\n system to use a custom output path, for each of these alternatives.\n \n The following patch fixes the following invocations:\n \n $ make tools/bpf\n $ make tools/bpf O\u003d\u003cdir\u003e\n $ make -C tools/bpf/bpftool OUTPUT\u003d\u003cdir\u003e\n $ make -C tools/bpf/bpftool O\u003d\u003cdir\u003e\n $ cd tools/ \u0026\u0026 make bpf O\u003d\u003cdir\u003e\n $ cd tools/bpf/bpftool \u0026\u0026 make OUTPUT\u003d\u003cdir\u003e\n $ cd tools/bpf/bpftool \u0026\u0026 make O\u003d\u003cdir\u003e\n \n After this commit, the build still fails for two variants when passing\n the OUTPUT variable:\n \n $ make tools/bpf OUTPUT\u003d\u003cdir\u003e\n $ cd tools/ \u0026\u0026 make bpf OUTPUT\u003d\u003cdir\u003e\n \n In order to remember and check what make invocations are supposed to\n work, and to document the ones which do not, a new script is added to\n the BPF selftests. Note that some invocations require the kernel to be\n configured, so the script skips them if no .config file is found.\n \n v2:\n - In make_and_clean(), set $ERROR to 1 when \"make\" returns non-zero,\n even if the binary was produced.\n - Run \"make clean\" from the correct directory (bpf/ instead of bpftool/,\n when relevant).\n \n Reported-by: Lorenz Bauer \u003clmb@cloudflare.com\u003e\n Signed-off-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit e0a43aa3e4a40fb88d3fa75397da0bf1e60a6c45\nAuthor: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\nDate: Fri Aug 30 12:00:37 2019 +0100\n\n tools: bpftool: ignore make built-in rules for getting kernel version\n \n Bpftool calls the toplevel Makefile to get the kernel version for the\n sources it is built from. But when the utility is built from the top of\n the kernel repository, it may dump the following error message for\n certain architectures (including x86):\n \n $ make tools/bpf\n [...]\n make[3]: *** [checkbin] Error 1\n [...]\n \n This does not prevent bpftool compilation, but may feel disconcerting.\n The \"checkbin\" arch-dependent target is not supposed to be called for\n target \"kernelversion\", which is a simple \"echo\" of the version number.\n \n It turns out this is caused by the make invocation in tools/bpf/bpftool,\n which attempts to find implicit rules to apply. Extract from debug\n output:\n \n Reading makefiles...\n Reading makefile \u0027Makefile\u0027...\n Reading makefile \u0027scripts/Kbuild.include\u0027 (search path) (no ~ expansion)...\n Reading makefile \u0027scripts/subarch.include\u0027 (search path) (no ~ expansion)...\n Reading makefile \u0027arch/x86/Makefile\u0027 (search path) (no ~ expansion)...\n Reading makefile \u0027scripts/Makefile.kcov\u0027 (search path) (no ~ expansion)...\n Reading makefile \u0027scripts/Makefile.gcc-plugins\u0027 (search path) (no ~ expansion)...\n Reading makefile \u0027scripts/Makefile.kasan\u0027 (search path) (no ~ expansion)...\n Reading makefile \u0027scripts/Makefile.extrawarn\u0027 (search path) (no ~ expansion)...\n Reading makefile \u0027scripts/Makefile.ubsan\u0027 (search path) (no ~ expansion)...\n Updating makefiles....\n Considering target file \u0027scripts/Makefile.ubsan\u0027.\n Looking for an implicit rule for \u0027scripts/Makefile.ubsan\u0027.\n Trying pattern rule with stem \u0027Makefile.ubsan\u0027.\n [...]\n Trying pattern rule with stem \u0027Makefile.ubsan\u0027.\n Trying implicit prerequisite \u0027scripts/Makefile.ubsan.o\u0027.\n Looking for a rule with intermediate file \u0027scripts/Makefile.ubsan.o\u0027.\n Avoiding implicit rule recursion.\n Trying pattern rule with stem \u0027Makefile.ubsan\u0027.\n Trying rule prerequisite \u0027prepare\u0027.\n Trying rule prerequisite \u0027FORCE\u0027.\n Found an implicit rule for \u0027scripts/Makefile.ubsan\u0027.\n Considering target file \u0027prepare\u0027.\n File \u0027prepare\u0027 does not exist.\n Considering target file \u0027prepare0\u0027.\n File \u0027prepare0\u0027 does not exist.\n Considering target file \u0027archprepare\u0027.\n File \u0027archprepare\u0027 does not exist.\n Considering target file \u0027archheaders\u0027.\n File \u0027archheaders\u0027 does not exist.\n Finished prerequisites of target file \u0027archheaders\u0027.\n Must remake target \u0027archheaders\u0027.\n Putting child 0x55976f4f6980 (archheaders) PID 31743 on the chain.\n \n To avoid that, pass the -r and -R flags to eliminate the use of make\n built-in rules (and while at it, built-in variables) when running\n command \"make kernelversion\" from bpftool\u0027s Makefile.\n \n Signed-off-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 1c8f9b91c456f5b47a377a0c8c5d4130fc39433a\nAuthor: Yauheni Kaliuta \u003cyauheni.kaliuta@redhat.com\u003e\nDate: Wed Aug 28 21:28:46 2019 +0300\n\n bpf: s390: add JIT support for multi-function programs\n \n This adds support for bpf-to-bpf function calls in the s390 JIT\n compiler. The JIT compiler converts the bpf call instructions to\n native branch instructions. After a round of the usual passes, the\n start addresses of the JITed images for the callee functions are\n known. Finally, to fixup the branch target addresses, we need to\n perform an extra pass.\n \n Because of the address range in which JITed images are allocated on\n s390, the offsets of the start addresses of these images from\n __bpf_call_base are as large as 64 bits. So, for a function call,\n the imm field of the instruction cannot be used to determine the\n callee\u0027s address. Use bpf_jit_get_func_addr() helper instead.\n \n The patch borrows a lot from:\n \n commit 8c11ea5ce13d (\"bpf, arm64: fix getting subprog addr from aux\n for calls\")\n \n commit e2c95a61656d (\"bpf, ppc64: generalize fetching subprog into\n bpf_jit_get_func_addr\")\n \n commit 8484ce8306f9 (\"bpf: powerpc64: add JIT support for\n multi-function programs\")\n \n (including the commit message).\n \n test_verifier (5.3-rc6 with CONFIG_BPF_JIT_ALWAYS_ON\u003dy):\n \n without patch:\n Summary: 1501 PASSED, 0 SKIPPED, 47 FAILED\n \n with patch:\n Summary: 1540 PASSED, 0 SKIPPED, 8 FAILED\n \n Signed-off-by: Yauheni Kaliuta \u003cyauheni.kaliuta@redhat.com\u003e\n Acked-by: Ilya Leoshkevich \u003ciii@linux.ibm.com\u003e\n Tested-by: Ilya Leoshkevich \u003ciii@linux.ibm.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit daa664a5cdd599583fceb9018c49d783bc5a0d5d\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Thu Aug 29 19:15:17 2019 +0300\n\n net/mlx5e: Move local var definition into ifdef block\n \n New local variable \"struct flow_block_offload *f\" was added to\n mlx5e_setup_tc() in recent rtnl lock removal patches. The variable is used\n in code that is only compiled when CONFIG_MLX5_ESWITCH is enabled. This\n results compilation warning about unused variable when CONFIG_MLX5_ESWITCH\n is not set. Move the variable definition into eswitch-specific code block\n from the beginning of mlx5e_setup_tc() function.\n \n Fixes: c9f14470d048 (\"net: sched: add API for registering unlocked offload block callbacks\")\n Reported-by: tanhuazhong \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f2b795ea025c743073933d5a121e22a2c3e66c99\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Thu Aug 29 19:15:16 2019 +0300\n\n net: sched: cls_matchall: cleanup flow_action before deallocating\n \n Recent rtnl lock removal patch changed flow_action infra to require proper\n cleanup besides simple memory deallocation. However, matchall classifier\n was not updated to call tc_cleanup_flow_action(). Add proper cleanup to\n mall_replace_hw_filter() and mall_reoffload().\n \n Fixes: 5a6ff4b13d59 (\"net: sched: take reference to action dev before calling offloads\")\n Reported-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Tested-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit de8e1beb191912f94bcab46e72f247daaf664d15\nAuthor: Luke Hsiao \u003clukehsiao@google.com\u003e\nDate: Thu Aug 29 10:02:44 2019 -0400\n\n tcp_bbr: clarify that bbr_bdp() rounds up in comments\n \n This explicitly clarifies that bbr_bdp() returns the rounded-up value of\n the bandwidth-delay product and why in the comments.\n \n Signed-off-by: Luke Hsiao \u003clukehsiao@google.com\u003e\n Acked-by: Soheil Hassas Yeganeh \u003csoheil@google.com\u003e\n Acked-by: Neal Cardwell \u003cncardwell@google.com\u003e\n Acked-by: Priyaranjan Jha \u003cpriyarjha@google.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit fa730a3bb9d35fa2289a2b6a0a4846154873f6f2\nAuthor: Jiri Pirko \u003cjiri@mellanox.com\u003e\nDate: Thu Aug 29 15:38:42 2019 +0200\n\n sched: act_vlan: implement stats_update callback\n \n Implement this callback in order to get the offloaded stats added to the\n kernel stats.\n \n Reported-by: Pengfei Liu \u003cpengfeil@mellanox.com\u003e\n Signed-off-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ebdd3c0b179e4cd7b2c53a2962e4cc36ec412f7f\nAuthor: Masanari Iida \u003cstandby24x7@gmail.com\u003e\nDate: Sat Aug 3 09:01:26 2019 +0900\n\n selftest/ftrace: Fix typo in trigger-snapshot.tc\n \n This patch fixes a spelling typo in trigger-snapshot.tc\n \n [skhan@linuxfoundation.org: Fix typo in commit log]\n Signed-off-by: Masanari Iida \u003cstandby24x7@gmail.com\u003e\n Acked-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit 27382472adcd62e2082942202ccfa321c9775769\nAuthor: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nDate: Fri Aug 30 14:34:05 2019 -0700\n\n net: stmmac: depend on COMMON_CLK\n \n Fixes: 190f73ab4c43 (\"net: stmmac: setup higher frequency clk support for EHL \u0026 TGL\")\n Signed-off-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7bff9779d76917adfa7c67712aa6ae3b7e5098ba\nAuthor: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\nDate: Fri Aug 30 12:16:44 2019 +0200\n\n drm/i915: Fix regression with crtc disable ordering\n \n When we moved the code to disable crtc\u0027s to a separate patch,\n we forgot to ensure that for_each_oldnew_intel_crtc_in_state_reverse()\n was moved as well.\n \n Signed-off-by: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Fixes: 66d9cec8a6c9 (\"drm/i915/display: Move the commit_tail() disable sequence to separate function\")\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Manasi Navare \u003cmanasi.d.navare@intel.com\u003e\n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Reviewed-by: Manasi Navare \u003cmanasi.d.navare@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Manasi Navare \u003cmanasi.d.navare@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190830101644.8740-1-maarten.lankhorst@linux.intel.com\n\ncommit 77afe3b82e97c2d1a066d038344bbce8bb8f7d85\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Thu Aug 29 00:14:50 2019 +0100\n\n arcnet: capmode: remove redundant assignment to pointer pkt\n \n Pointer pkt is being initialized with a value that is never read\n and pkt is being re-assigned a little later on. The assignment is\n redundant and hence can be removed.\n \n Addresses-Coverity: (\"Ununsed value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit acfb50e4e773c9a5755a3c265c7c20d37a8642e5\nAuthor: Vasundhara Volam \u003cvasundhara-v.volam@broadcom.com\u003e\nDate: Thu Aug 29 23:55:05 2019 -0400\n\n bnxt_en: Add FW fatal devlink_health_reporter.\n \n Health show command example and output:\n \n $ devlink health show pci/0000:af:00.0 reporter fw_fatal\n \n pci/0000:af:00.0:\n name fw_fatal\n state healthy error 1 recover 1 grace_period 0 auto_recover true\n \n Fatal events from firmware or missing periodic heartbeats will\n be reported and recovery will be handled.\n \n We also turn on the support flags when we register with the firmware to\n enable this health and recovery feature in the firmware.\n \n Cc: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Vasundhara Volam \u003cvasundhara-v.volam@broadcom.com\u003e\n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d1db9e166bf6a50e1e6713f3fd3b4de6007e3671\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Thu Aug 29 23:55:04 2019 -0400\n\n bnxt_en: Add bnxt_fw_exception() to handle fatal firmware errors.\n \n This call will handle fatal firmware errors by forcing a reset on the\n firmware. The master function driver will carry out the forced reset.\n The sequence will go through the same bnxt_fw_reset_task() workqueue.\n This fatal reset differs from the non-fatal reset at the beginning\n stages. From the BNXT_FW_RESET_STATE_ENABLE_DEV state onwards where\n the firmware is coming out of reset, it is practically identical to the\n non-fatal reset.\n \n The next patch will add the periodic heartbeat check and the devlink\n reporter to report the fatal event and to initiate the bnxt_fw_exception()\n call.\n \n Signed-off-by: Vasundhara Volam \u003cvasundhara-v.volam@broadcom.com\u003e\n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit cbb51067a5f5fbae733283b67fc8013881eb4bb1\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Thu Aug 29 23:55:03 2019 -0400\n\n bnxt_en: Add RESET_FW state logic to bnxt_fw_reset_task().\n \n This state handles driver initiated chip reset during error recovery.\n Only the master function will perform this step during error recovery.\n The next patch will add code to initiate this reset from the master\n function.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b4fff2079d1080af7dcad8ad0e80cc89e1ee000c\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Thu Aug 29 23:55:02 2019 -0400\n\n bnxt_en: Do not send firmware messages if firmware is in error state.\n \n Add a flag to mark that the firmware has encountered fatal condition.\n The driver will not send any more firmware messages and will return\n error to the caller. Fix up some clean up functions to continue\n and not abort when the firmware message function returns error.\n \n This is preparation work to fully handle firmware error recovery\n under fatal conditions.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 2cd8696850450b750f278be06ee56eb51d84621c\nAuthor: Vasundhara Volam \u003cvasundhara-v.volam@broadcom.com\u003e\nDate: Thu Aug 29 23:55:01 2019 -0400\n\n bnxt_en: Retain user settings on a VF after RESET_NOTIFY event.\n \n Retain the VF MAC address, default VLAN, TX rate control, trust settings\n of VFs after firmware reset.\n \n Signed-off-by: Vasundhara Volam \u003cvasundhara-v.volam@broadcom.com\u003e\n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 657a33c8a0a2342e91259b28356838dc89216b19\nAuthor: Vasundhara Volam \u003cvasundhara-v.volam@broadcom.com\u003e\nDate: Thu Aug 29 23:55:00 2019 -0400\n\n bnxt_en: Add devlink health reset reporter.\n \n Add devlink health reporter for the firmware reset event. Once we get\n the notification from firmware about the impending reset, the driver\n will report this to devlink and the call to bnxt_fw_reset() will be\n initiated to complete the reset sequence.\n \n Cc: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Vasundhara Volam \u003cvasundhara-v.volam@broadcom.com\u003e\n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 230d1f0de754b483ec6eefc1ca5aaeff2b6b9a4c\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Thu Aug 29 23:54:59 2019 -0400\n\n bnxt_en: Handle firmware reset.\n \n Add the bnxt_fw_reset() main function to handle firmware reset. This\n is triggered by firmware to initiate an orderly reset, for example\n when a non-fatal exception condition has been detected. bnxt_fw_reset()\n will first wait for all VFs to shutdown and then start the\n bnxt_fw_reset_task() work queue to go through the sequence of reset,\n re-probe, and re-initialization.\n \n The next patch will add the devlink reporter to start the sequence and\n call bnxt_fw_reset().\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 2151fe0830fdb951f8ecfcfe67306fdef2366aa0\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Thu Aug 29 23:54:58 2019 -0400\n\n bnxt_en: Handle RESET_NOTIFY async event from firmware.\n \n This event from firmware signals a coordinated reset initiated by the\n firmware. It may be triggered by some error conditions encountered\n in the firmware or other orderly reset conditions.\n \n We store the parameters from this event. Subsequent patches will\n add logic to handle reset itself using devlink reporters.\n \n Signed-off-by: Vasundhara Volam \u003cvasundhara-v.volam@broadcom.com\u003e\n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6763c779c2d8b568b2e174f3eeeaf644fa38b34d\nAuthor: Vasundhara Volam \u003cvasundhara-v.volam@broadcom.com\u003e\nDate: Thu Aug 29 23:54:57 2019 -0400\n\n bnxt_en: Add new FW devlink_health_reporter\n \n Create new FW devlink_health_reporter, to know the current health\n status of FW.\n \n Command example and output:\n $ devlink health show pci/0000:af:00.0 reporter fw\n \n pci/0000:af:00.0:\n name fw\n state healthy error 0 recover 0\n \n FW status: Healthy; Reset count: 1\n \n Cc: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Vasundhara Volam \u003cvasundhara-v.volam@broadcom.com\u003e\n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 3bc7d4a352efe5b596883ef16b769055320db1f6\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Thu Aug 29 23:54:56 2019 -0400\n\n bnxt_en: Add BNXT_STATE_IN_FW_RESET state.\n \n The new flag will be set in subsequent patches when firmware is\n going through reset. If bnxt_close() is called while the new flag\n is set, the FW reset sequence will have to be aborted because the\n NIC is prematurely closed before FW reset has completed. We also\n reject SRIOV configurations while FW reset is in progress.\n \n v2: No longer drop rtnl_lock() in close and wait for FW reset to complete.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7e914027f757b656cd681ba4fe75f3984531ee50\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Thu Aug 29 23:54:55 2019 -0400\n\n bnxt_en: Enable health monitoring.\n \n Handle the async event from the firmware that enables firmware health\n monitoring. Store initial health metrics.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9ffbd67734909ca8bb099e62f06387649b43d5a8\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Thu Aug 29 23:54:54 2019 -0400\n\n bnxt_en: Pre-map the firmware health monitoring registers.\n \n Pre-map the GRC registers for periodic firmware health monitoring.\n \n Signed-off-by: Vasundhara Volam \u003cvasundhara-v.volam@broadcom.com\u003e\n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 07f83d72d238f5d41b03d6142641129e8a7a0ec4\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Thu Aug 29 23:54:53 2019 -0400\n\n bnxt_en: Discover firmware error recovery capabilities.\n \n Call the new firmware API HWRM_ERROR_RECOVERY_QCFG if it is supported\n to discover the firmware health and recovery capabilities and settings.\n This feature allows the driver to reset the chip if firmware crashes and\n becomes unresponsive.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ec5d31e3c15d5233b491400133c67f78a320062c\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Thu Aug 29 23:54:52 2019 -0400\n\n bnxt_en: Handle firmware reset status during IF_UP.\n \n During IF_UP, newer firmware has a new status flag that indicates that\n firmware has reset. Add new function bnxt_fw_init_one() to re-probe the\n firmware and re-setup VF resources on the PF if necessary. If the\n re-probe fails, set a flag to prevent bnxt_open() from proceeding again.\n \n Signed-off-by: Vasundhara Volam \u003cvasundhara-v.volam@broadcom.com\u003e\n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 91b9be487001d344a39c453ade6cdbd125e06208\nAuthor: Vasundhara Volam \u003cvasundhara-v.volam@broadcom.com\u003e\nDate: Thu Aug 29 23:54:51 2019 -0400\n\n bnxt_en: Register buffers for VFs before reserving resources.\n \n When VFs need to be reconfigured dynamically after firmwware reset, the\n configuration sequence on the PF needs to be changed to register the VF\n buffers first. Otherwise, some VF firmware commands may not succeed as\n there may not be PF buffers ready for the re-directed firmware commands.\n \n This sequencing did not matter much before when we only supported\n the normal bring-up of VFs.\n \n Signed-off-by: Vasundhara Volam \u003cvasundhara-v.volam@broadcom.com\u003e\n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 702d5011ab5e7b9afe44058d33a89d1501645a10\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Thu Aug 29 23:54:50 2019 -0400\n\n bnxt_en: Refactor bnxt_sriov_enable().\n \n Refactor the hardware/firmware configuration portion in\n bnxt_sriov_enable() into a new function bnxt_cfg_hw_sriov(). This\n new function can be called after a firmware reset to reconfigure the\n VFs previously enabled.\n \n v2: straight refactor of the code. Reordering done in the next patch.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ba642ab773db97c32293547485f562d2dfc06666\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Thu Aug 29 23:54:49 2019 -0400\n\n bnxt_en: Prepare bnxt_init_one() to be called multiple times.\n \n In preparation for the new firmware reset feature, some of the logic\n in bnxt_init_one() and related functions will be called again after\n firmware has reset. Reset some of the flags and capabilities so that\n everything that can change can be re-initialized. Refactor some\n functions to probe firmware versions and capabilities. Check some\n buffers before allocating as they may have been allocated previously.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 5bedb5296e33e889818d77c8ec69040481fab157\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Thu Aug 29 23:54:48 2019 -0400\n\n bnxt_en: Suppress all error messages in hwrm_do_send_msg() in silent mode.\n \n If the silent parameter is set, suppress all messages when there is\n no response from firmware. When polling for firmware to come out of\n reset, no response may be normal and we want to suppress the error\n messages. Also, don\u0027t poll for the firmware DMA response if Bus Master\n is disabled. This is in preparation for error recovery when firmware\n may be in error or reset state or Bus Master is disabled.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a798302d56f56fb7ad6a01f64f495aeafeb6c0f0\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Thu Aug 29 23:54:47 2019 -0400\n\n bnxt_en: Simplify error checking in the SR-IOV message forwarding functions.\n \n There are 4 functions handling message forwarding for SR-IOV. They\n check for non-zero firmware response code and then return -1. There\n is no need to do this anymore. The main messaging function will\n now return standard error code. Since we don\u0027t need to examine the\n response, we can use the hwrm_send_message() variant which will\n take the mutex automatically.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d4f1420d365633490aa134abfe408453d1c7c238\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Thu Aug 29 23:54:46 2019 -0400\n\n bnxt_en: Convert error code in firmware message response to standard code.\n \n The main firmware messaging function returns the firmware defined error\n code and many callers have to convert to standard error code for proper\n propagation to userspace. Convert bnxt_hwrm_do_send_msg() to return\n standard error code so we can do away with all the special error code\n handling by the many callers.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a935cb7ec449bca1adf806d7fb00f5032b63c6e0\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Thu Aug 29 23:54:45 2019 -0400\n\n bnxt_en: Remove the -1 error return code from bnxt_hwrm_do_send_msg().\n \n Replace the non-standard -1 code with -EBUSY when there is no firmware\n response after waiting for the maximum timeout.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b3b0ddd07e63d564a3d5500938913805d06a1682\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Thu Aug 29 23:54:44 2019 -0400\n\n bnxt_en: Use a common function to print the same ethtool -f error message.\n \n The same message is printed 3 times in the code, so use a common function\n to do that.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 70359dbe2475d43a914b5259b870b45b15c8ecad\nAuthor: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\nDate: Fri Aug 30 11:25:38 2019 +0200\n\n net: sgi: ioc3-eth: no need to stop queue set_multicast_list\n \n netif_stop_queue()/netif_wake_qeue() aren\u0027t needed for changing\n multicast filters.\n \n Signed-off-by: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d1c9454274d84671737574ec21c69f26c0433307\nAuthor: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\nDate: Fri Aug 30 11:25:37 2019 +0200\n\n net: sgi: ioc3-eth: protect emcr in all cases\n \n emcr in private struct wasn\u0027t always protected by spinlock.\n \n Signed-off-by: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 3498cb272e87af8d3e7d6884e57cbe7689215672\nAuthor: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\nDate: Fri Aug 30 11:25:36 2019 +0200\n\n net: sgi: ioc3-eth: Fix IPG settings\n \n The half/full duplex settings for inter packet gap counters/timer were\n reversed.\n \n Signed-off-by: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 8dff19a6dcb8bcab8bed86e60c8033aaf0a13646\nAuthor: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\nDate: Fri Aug 30 11:25:35 2019 +0200\n\n net: sgi: ioc3-eth: use csum_fold\n \n replace open coded checksum folding by csum_fold.\n \n Signed-off-by: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ed870f6a7aa2120720e9cd463926969699036eaa\nAuthor: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\nDate: Fri Aug 30 11:25:34 2019 +0200\n\n net: sgi: ioc3-eth: use dma-direct for dma allocations\n \n Replace the homegrown DMA memory allocation, which only works on\n SGI-IP27 machines, with the generic dma allocations.\n \n Signed-off-by: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 850d2fed5bcebbcbb3f450ffad9bb4dd096ff829\nAuthor: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\nDate: Fri Aug 30 11:25:33 2019 +0200\n\n net: sgi: ioc3-eth: refactor rx buffer allocation\n \n Move common code for rx buffer setup into ioc3_alloc_skb and deal\n with allocation failures. Also clean up allocation size calculation.\n \n Signed-off-by: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 19a957b6b464ddabd32e5ebcc44df8eb6380881b\nAuthor: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\nDate: Fri Aug 30 11:25:32 2019 +0200\n\n net: sgi: ioc3-eth: split ring cleaning/freeing and allocation\n \n Do tx ring cleaning and freeing of rx buffers, when chip is shutdown and\n allocate buffers before bringing chip up.\n \n Signed-off-by: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit fcd0da5a6d8784aba38916cdba86abd83694bc73\nAuthor: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\nDate: Fri Aug 30 11:25:31 2019 +0200\n\n net: sgi: ioc3-eth: introduce chip start function\n \n ioc3_init did everything from reset to init rings to starting the chip.\n This change move out chip start into a new function as preparation\n for easier handling of receive buffer allocation failures.\n \n Signed-off-by: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9c328b05443a09bd62e4fcae284aeebbd13ca30e\nAuthor: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\nDate: Fri Aug 30 11:25:30 2019 +0200\n\n net: sgi: ioc3-eth: separate tx and rx ring handling\n \n After allocation of descriptor memory is now done once in probe\n handling of tx ring is completely done by ioc3_clean_tx_ring. So\n we remove the remaining tx ring actions out of ioc3_alloc_rings\n and ioc3_free_rings and rename it to ioc3_[alloc|free]_rx_bufs\n to better describe what they are doing.\n \n Signed-off-by: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 489467e52486d483550ac3e3ed33dceb93b1e4a9\nAuthor: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\nDate: Fri Aug 30 11:25:29 2019 +0200\n\n net: sgi: ioc3-eth: get rid of ioc3_clean_rx_ring()\n \n Move clearing of the descriptor valid bit into ioc3_alloc_rings. This\n makes ioc3_clean_rx_ring obsolete.\n \n Signed-off-by: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c7b57274754971cda56fdd935e503ee619419596\nAuthor: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\nDate: Fri Aug 30 11:25:28 2019 +0200\n\n net: sgi: ioc3-eth: allocate space for desc rings only once\n \n Memory for descriptor rings are allocated/freed, when interface is\n brought up/down. Since the size of the rings is not changeable by\n hardware, we now allocate rings now during probe and free it, when\n device is removed.\n \n Signed-off-by: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 141a7dbb886f39fae55815229dc4d91d56b36175\nAuthor: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\nDate: Fri Aug 30 11:25:27 2019 +0200\n\n net: sgi: ioc3-eth: use defines for constants dealing with desc rings\n \n Descriptor ring sizes of the IOC3 are more or less fixed size. To\n make clearer where there is a relation to ring sizes use defines.\n \n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c1b6a3d85d3f2a9500a9fddf6820a0fd1b1bc4e8\nAuthor: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\nDate: Fri Aug 30 11:25:26 2019 +0200\n\n net: sgi: ioc3-eth: remove checkpatch errors/warning\n \n Before massaging the driver further fix oddities found by checkpatch like\n - wrong indention\n - comment formatting\n - use of printk instead or netdev_xxx/pr_xxx\n \n Signed-off-by: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit cbe7d51745f9334d05bf2fdd915322e159bbcaa8\nAuthor: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\nDate: Fri Aug 30 11:25:25 2019 +0200\n\n MIPS: SGI-IP27: restructure ioc3 register access\n \n Break up the big ioc3 register struct into functional pieces to\n make use in sub-function drivers more straightforward. And while\n doing that get rid of all volatile access by using readX/writeX.\n \n Signed-off-by: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 688125a6e7872a8c827d7848ba8fbc41785fb9c6\nAuthor: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\nDate: Fri Aug 30 11:25:24 2019 +0200\n\n MIPS: SGI-IP27: remove ioc3 ethernet init\n \n Removed not needed disabling of ethernet interrupts in IP27 platform code.\n \n Acked-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Signed-off-by: Thomas Bogendoerfer \u003ctbogendoerfer@suse.de\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6ea425b50c4853eefad10d49de901b68d326319d\nAuthor: George G. Davis \u003cgeorge_davis@mentor.com\u003e\nDate: Fri Aug 30 15:32:23 2019 -0400\n\n selftests: watchdog: Add optional file argument\n \n Some systems have multiple watchdog devices where the first device\n registered is assigned to the /dev/watchdog device file. In order\n to test other watchdog devices, add an optional file argument for\n selecting non-default watchdog devices for testing.\n \n Tested-by: Eugeniu Rosca \u003cerosca@de.adit-jv.com\u003e\n Signed-off-by: George G. Davis \u003cgeorge_davis@mentor.com\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit 9c9284f9cee9052da4cad575da8dc5f2bbb31065\nAuthor: Austin Kim \u003caustindh.kim@gmail.com\u003e\nDate: Fri Aug 30 17:07:04 2019 +0900\n\n drm/amdgpu: Move null pointer dereference check\n \n Null pointer dereference check should have been checked,\n ahead of below routine.\n struct amdgpu_device *adev \u003d hwmgr-\u003eadev;\n \n With this commit, it could avoid potential NULL dereference.\n \n Signed-off-by: Austin Kim \u003caustindh.kim@gmail.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 20c14ee135011f2a0fb2f987f05fd9655e804a9d\nAuthor: Petr Cvek \u003cpetrcvekcz@gmail.com\u003e\nDate: Fri Aug 30 16:31:58 2019 +0200\n\n drm/amdgpu: Fix undefined dm_ip_block for navi12\n \n There is missing \"if defined\" CONFIG_DRM_AMD_DC block for non DC\n configurations. This will cause link error. The patch is fixing that.\n \n Bug: https://bugs.freedesktop.org/show_bug.cgi?id\u003d110979\n Signed-off-by: Petr Cvek \u003cpetrcvekcz@gmail.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 537e3bbfeea10c57479acd64b9387c20e555a958\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Fri Dec 14 11:21:41 2018 +0800\n\n drm/amdgpu: fix no interrupt issue for renoir emu (v2)\n \n In renoir\u0027s vega10_ih model, there\u0027s a security change in mmIH_CHICKEN\n register, that limits IH to use physical address (FBPA, GPA) directly.\n Those chicken bits need to be programmed first.\n \n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 75966255881127f4a2c0ab7e18d9224672bdeddb\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Fri Dec 14 11:16:36 2018 +0800\n\n drm/amdgpu: update IH_CHICKEN in oss 4.0 IP header for VG/RV series\n \n In Renoir\u0027s emulator, those chicken bits need to be programmed.\n \n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ea1fc5e1ff7106f39d63e4a944990674b27af40d\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Fri Aug 30 09:53:27 2019 +0800\n\n drm/amd/powerplay: SMU_MSG_OverridePcieParameters is unsupport for APU\n \n For apu, SMU_MSG_OverridePcieParameters is unsupport.\n So return directly in smu_override_pcie_parameters function.\n \n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 60fc35f327e0a9e60b955c0f3c3ed623608d1baa\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Fri Aug 30 22:42:55 2019 +0300\n\n ahci: Do not export local variable ahci_em_messages\n \n The commit ed08d40cdec4\n (\"ahci: Changing two module params with static and __read_mostly\")\n moved ahci_em_messages to be static while missing the fact of exporting it.\n \n WARNING: \"ahci_em_messages\" [vmlinux] is a static EXPORT_SYMBOL_GPL\n \n Drop export for the local variable ahci_em_messages.\n \n Fixes: ed08d40cdec4 (\"ahci: Changing two module params with static and __read_mostly\")\n Cc: Chuansheng Liu \u003cchuansheng.liu@intel.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit aabbe344dc3ca5f7d8263a02608ba6179e8a4499\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 30 19:03:25 2019 +0100\n\n drm/i915: Use RCU for unlocked vm_idr lookup\n \n Since i915_address_space is now RCU protected, we can do the vm_idr lookup\n without taking the vm_idr_mutex, just with the rcu_read_lock() instead.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190830180325.7755-1-chris@chris-wilson.co.uk\n\ncommit 0b2d2c2eecf27f403df60ebab06cad7b6bfe84f8\nAuthor: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\nDate: Tue Aug 27 12:14:47 2019 -0400\n\n drm/amdgpu: Handle job is NULL use case in amdgpu_device_gpu_recover\n \n This should be checked at all places job is accessed.\n \n Signed-off-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c1d143dd2ac8d481500fec4c0d715ff301d2766f\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 30 19:00:00 2019 +0100\n\n drm/i915: Remove ppgtt-\u003edirty_engines\n \n This is no longer used anywhere and so can be removed. However, tracking\n the dirty status on the ppgtt doesn\u0027t work very well if the ppgtt is\n shared, so perhaps for the best that it is no longer required.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190830180000.24608-3-chris@chris-wilson.co.uk\n\ncommit 0b718ba1e884f64dce27c19311dd2859b87e56b9\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 30 18:59:59 2019 +0100\n\n drm/i915/gtt: Downgrade Cherryview back to aliasing-ppgtt\n \n With the upcoming change in timing (dramatically reducing the latency\n between manipulating the ppGTT and execution), no amount of tweaking\n could save Cherryview, it would always fail to invalidate its TLB.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Acked-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190830180000.24608-2-chris@chris-wilson.co.uk\n\ncommit 3dc007fe9b2b256d2779bb9e9a0d372eb1d90643\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 30 18:59:58 2019 +0100\n\n drm/i915/gtt: Downgrade gen7 (ivb, byt, hsw) back to aliasing-ppgtt\n \n With the upcoming change in timing (dramatically reducing the latency\n between manipulating the ppGTT and execution), no amount of tweaking\n could save Baytrail, it would always fail to invalidate its TLB. Ville\n was right, Baytrail is beyond hope.\n \n v2: Rollback on all gen7; same timing instability on TLB invalidation.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Acked-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190830180000.24608-1-chris@chris-wilson.co.uk\n\ncommit 3940ba8eea8c6c89f547a1bd153977cece5fecd2\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Sat Aug 17 09:32:41 2019 +0200\n\n asm-generic: don\u0027t provide __ioremap\n \n __ioremap is not a kernel API, but used for helpers with differing\n semantics in arch code. We should not provide it in as-generic.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Paul Walmsley \u003cpaul.walmsley@sifive.com\u003e\n Tested-by: Paul Walmsley \u003cpaul.walmsley@sifive.com\u003e # rv32, rv64 boot\n Acked-by: Paul Walmsley \u003cpaul.walmsley@sifive.com\u003e # arch/riscv\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 1a914990ffe997090a516681ac97519e7c1b22a2\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Aug 30 10:07:11 2019 +0100\n\n wimax/i2400m: remove debug containing bogus calculation of index\n \n The subtraction of the two pointers is automatically scaled by the\n size of the size of the object the pointers point to, so the division\n by sizeof(*i2400m-\u003ebarker) is incorrect. This has been broken since\n day one of the driver and is only debug, so remove the debug completely.\n \n Also move \u0026\u0026 in condition to clean up a checkpatch warning.\n \n Addresses-Coverity: (\"Extra sizeof expression\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Reviewed-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 55507aea58824578610eb0cb5c250a0c997987c9\nAuthor: Krzysztof Wilczynski \u003ckw@linux.com\u003e\nDate: Mon Aug 26 00:10:39 2019 +0200\n\n PCI: Remove unnecessary returns\n \n Remove unnecessary \"return\" statements at the end of void functions. No\n functional change intended.\n \n Link: https://lore.kernel.org/r/20190825221039.6977-1-kw@linux.com\n Link: https://lore.kernel.org/r/20190826095143.21353-1-kw@linux.com\n Signed-off-by: Krzysztof Wilczynski \u003ckw@linux.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit b071c1fd7a8ad041759d9a9e87d1b4333417e36c\nAuthor: Lubomir Rintel \u003clkundrak@v3.sk\u003e\nDate: Wed Aug 7 15:20:49 2019 +0200\n\n PCI: OF: Correct of_irq_parse_pci() documentation\n \n 530210c7814e (\"of/irq: Replace of_irq with of_phandle_args\") changed the\n of_irq_parse_pci() parameter type but didn\u0027t change the corresponding\n documentation. Update the function doc to match.\n \n Link: https://lore.kernel.org/r/20190807132049.10304-1-lkundrak@v3.sk\n Signed-off-by: Lubomir Rintel \u003clkundrak@v3.sk\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit d1bbf38aaf882240cccca414e5a48db42e105da7\nAuthor: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\nDate: Tue Jul 30 08:04:00 2019 -0500\n\n PCI: Fix typos and whitespace errors\n \n Fix typos in drivers/pci. Comment and whitespace changes only.\n \n Link: https://lore.kernel.org/r/20190819115306.27338-1-kw@linux.com\n Signed-off-by: Krzysztof Wilczynski \u003ckw@linux.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Acked-by: Thomas Petazzoni \u003cthomas.petazzoni@bootlin.com\u003e # armada8k\n\ncommit 37b6b4dafa46ff9bfc96d26db0b71d04bce2b828\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Mon Jan 12 09:44:44 2015 +0300\n\n xtensa: add support for call0 ABI in userspace\n \n Provide a Kconfig choice to select whether only the default ABI, only\n call0 ABI or both are supported. The default for XEA2 is windowed, but\n it may change for XEA3. Call0 only runs userspace with PS.WOE disabled.\n Supporting both windowed and call0 ABIs is tricky, as there\u0027s no\n indication in the ELF binaries which ABI they use. So it is done by\n probing: each process is started with PS.WOE disabled, but the handler\n of an illegal instruction exception taken with PS.WOE retries faulting\n instruction after enabling PS.WOE. It must happen before any signal is\n delivered to the process, otherwise it may be delivered incorrectly.\n \n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit 474f249f130ecaef3c4a9d4cb29a692d8ec05b1a\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Fri Aug 30 11:16:06 2019 -0700\n\n xtensa: clean up PS_WOE_BIT usage\n \n PS_WOE_BIT is mainly used to generate PS.WOE mask in the code. Introduce\n PS_WOE_MASK macro and use it instead.\n \n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit d95f1a542c3df396137afa217ef9bd39cb8931ca\nAuthor: Logan Gunthorpe \u003clogang@deltatee.com\u003e\nDate: Wed Aug 28 15:40:54 2019 -0600\n\n RISC-V: Implement sparsemem\n \n Implement sparsemem support for Risc-v which helps pave the\n way for memory hotplug and eventually P2P support.\n \n Introduce Kconfig options for virtual and physical address bits which\n are used to calculate the size of the vmemmap and set the\n MAX_PHYSMEM_BITS.\n \n The vmemmap is located directly before the VMALLOC region and sized\n such that we can allocate enough pages to populate all the virtual\n address space in the system (similar to the way it\u0027s done in arm64).\n \n During initialization, call memblocks_present() and sparse_init(),\n and provide a stub for vmemmap_populate() (all of which is similar to\n arm64).\n \n [greentime.hu@sifive.com: fixed pfn_valid, FIXADDR_TOP and fixed a bug\n rebasing onto v5.3]\n Signed-off-by: Greentime Hu \u003cgreentime.hu@sifive.com\u003e\n Signed-off-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Reviewed-by: Palmer Dabbelt \u003cpalmer@sifive.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Cc: Albert Ou \u003caou@eecs.berkeley.edu\u003e\n Cc: Andrew Waterman \u003candrew@sifive.com\u003e\n Cc: Olof Johansson \u003colof@lixom.net\u003e\n Cc: Michael Clark \u003cmichaeljclark@mac.com\u003e\n Cc: Rob Herring \u003crobh@kernel.org\u003e\n Cc: Zong Li \u003czong@andestech.com\u003e\n Reviewed-by: Mike Rapoport \u003crppt@linux.ibm.com\u003e\n [paul.walmsley@sifive.com: updated to apply; minor commit message\n reformat]\n Signed-off-by: Paul Walmsley \u003cpaul.walmsley@sifive.com\u003e\n\ncommit 4f3f90084673fa1d7f8e81b9dbee652d2fae3092\nAuthor: Bin Meng \u003cbmeng.cn@gmail.com\u003e\nDate: Wed Aug 7 09:13:38 2019 -0700\n\n riscv: Using CSR numbers to access CSRs\n \n Since commit a3182c91ef4e (\"RISC-V: Access CSRs using CSR numbers\"),\n we should prefer accessing CSRs using their CSR numbers, but there\n are several leftovers like sstatus / sptbr we missed.\n \n Signed-off-by: Bin Meng \u003cbmeng.cn@gmail.com\u003e\n Reviewed-by: Anup Patel \u003canup@brainfault.org\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Paul Walmsley \u003cpaul.walmsley@sifive.com\u003e\n\ncommit 40e98130c32889e4ed128d129b824107b90501fe\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Thu Aug 29 14:15:26 2019 -0700\n\n drm/i915: unify icp, tgp and mcc irq setup\n \n Use a single function to setup the SDE irq and make MCC, ICP and TGP use\n it, just like was done for the irq handler.\n \n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190829211526.30525-4-jose.souza@intel.com\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n\ncommit b32821c036310c49cdc45639b1d451b21fc3f17b\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Thu Aug 29 14:15:25 2019 -0700\n\n drm/i915: parameterize SDE hotplug registers\n \n Ice Lake, Tiger Lake and Elkhart Lake all have different port\n configurations and all of them can be parameterized the same way to form\n the SDE hotplug bitmask. Avoid making them a special case an just use\n the parameterized macros.\n \n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190829211526.30525-3-jose.souza@intel.com\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n\ncommit 58676af69c2eb1e0202215fcfc1d421f186226ed\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Thu Aug 29 14:15:24 2019 -0700\n\n drm/i915: unify icp, tgp and mcc irq handling\n \n The differences are only on the pins, trigger and long_detect function.\n The MCC handling is already partially merged, so merge TGP as well.\n Remove the pins argument from icp_irq_handler() so we have all the\n differences between the 3 set in a common if ladder.\n \n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190829211526.30525-2-jose.souza@intel.com\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n\ncommit ed3126fa0d393825f2f134668b31363581035825\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Thu Aug 29 14:15:23 2019 -0700\n\n drm/i915: parameterize south hpd macros\n \n South, follow the north.\n \n Instead of defining separate macros for each port, make them take port\n as parameter as done for TC ports and for north engine. This will allow\n us to easily extend this as needed.\n \n tgp_ddi_port_hotplug_long_detect() is also removed as after the EHL\n introduction the tgp variant is an exact copy of icp.\n \n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190829211526.30525-1-jose.souza@intel.com\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n\ncommit 20a438d53fd9d12a894161bc56cbeab7a9993c39\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Fri Aug 23 09:26:32 2019 -0300\n\n media: sun4i: Make sun4i_csi_formats static\n \n The sun4i_csi_formats array is only used in sun4i_v4l2.c, so it doesn\u0027t\n make any sense to have it !static.\n \n [Sakari Ailus: const static -\u003e static const]\n \n Reported-by: Hans Verkuil \u003chverkuil@xs4all.nl\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 497144a5b7af12097c09b0ca30409ee7122499a0\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Wed Aug 28 14:52:09 2019 +0200\n\n ASoC: dt-bindings: Convert Allwinner A64 analog codec to a schema\n \n The Allwinner A64 SoC has an embedded audio codec that uses a separate\n controller to drive its analog part, which is supported in Linux, with a\n matching Device Tree binding.\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for that controller over to a YAML schemas.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190828125209.28173-5-mripard@kernel.org\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 8a99f76ac1a591f27f5216951bf8a8fdad58e987\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Wed Aug 28 14:52:07 2019 +0200\n\n ASoC: dt-bindings: Convert Allwinner A33 codec to a schema\n \n The Allwinner A33 SoC have an embedded audio codec that is supported in Linux,\n with a matching Device Tree binding.\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for that controller over to a YAML schemas.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190828125209.28173-3-mripard@kernel.org\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 1a8e7cdfa4f5872bf0c202d09bff6628aba6b9f6\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Wed Aug 28 14:52:05 2019 +0200\n\n ASoC: dt-bindings: sun4i-spdif: Fix dma-names warning\n \n Even though the H6 compatible has been properly added, the exeption for the\n number of DMA channels hasn\u0027t been updated, leading in a validation\n warning.\n \n Fix this.\n \n Fixes: b20453031472 (\"dt-bindings: sound: sun4i-spdif: Add Allwinner H6 compatible\")\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190828125209.28173-1-mripard@kernel.org\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c321d43b8da1b81a7902cfc9f6b93e75941c0066\nAuthor: Tycho Andersen \u003ctycho@tycho.ws\u003e\nDate: Mon Aug 26 08:43:02 2019 -0600\n\n selftests/seccomp: fix build on older kernels\n \n The seccomp selftest goes to some length to build against older kernel\n headers, viz. all the #ifdefs at the beginning of the file.\n \n Commit 201766a20e30 (\"ptrace: add PTRACE_GET_SYSCALL_INFO request\")\n introduces some additional macros, but doesn\u0027t do the #ifdef dance.\n Let\u0027s add that dance here to avoid:\n \n gcc -Wl,-no-as-needed -Wall seccomp_bpf.c -lpthread -o seccomp_bpf\n In file included from seccomp_bpf.c:51:\n seccomp_bpf.c: In function ‘tracer_ptrace’:\n seccomp_bpf.c:1787:20: error: ‘PTRACE_EVENTMSG_SYSCALL_ENTRY’ undeclared (first use in this function); did you mean ‘PTRACE_EVENT_CLONE’?\n EXPECT_EQ(entry ? PTRACE_EVENTMSG_SYSCALL_ENTRY\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n ../kselftest_harness.h:608:13: note: in definition of macro ‘__EXPECT’\n __typeof__(_expected) __exp \u003d (_expected); \\\n ^~~~~~~~~\n seccomp_bpf.c:1787:2: note: in expansion of macro ‘EXPECT_EQ’\n EXPECT_EQ(entry ? PTRACE_EVENTMSG_SYSCALL_ENTRY\n ^~~~~~~~~\n seccomp_bpf.c:1787:20: note: each undeclared identifier is reported only once for each function it appears in\n EXPECT_EQ(entry ? PTRACE_EVENTMSG_SYSCALL_ENTRY\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n ../kselftest_harness.h:608:13: note: in definition of macro ‘__EXPECT’\n __typeof__(_expected) __exp \u003d (_expected); \\\n ^~~~~~~~~\n seccomp_bpf.c:1787:2: note: in expansion of macro ‘EXPECT_EQ’\n EXPECT_EQ(entry ? PTRACE_EVENTMSG_SYSCALL_ENTRY\n ^~~~~~~~~\n seccomp_bpf.c:1788:6: error: ‘PTRACE_EVENTMSG_SYSCALL_EXIT’ undeclared (first use in this function); did you mean ‘PTRACE_EVENT_EXIT’?\n : PTRACE_EVENTMSG_SYSCALL_EXIT, msg);\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~\n ../kselftest_harness.h:608:13: note: in definition of macro ‘__EXPECT’\n __typeof__(_expected) __exp \u003d (_expected); \\\n ^~~~~~~~~\n seccomp_bpf.c:1787:2: note: in expansion of macro ‘EXPECT_EQ’\n EXPECT_EQ(entry ? PTRACE_EVENTMSG_SYSCALL_ENTRY\n ^~~~~~~~~\n make: *** [Makefile:12: seccomp_bpf] Error 1\n \n [skhan@linuxfoundation.org: Fix checkpatch error in commit log]\n Signed-off-by: Tycho Andersen \u003ctycho@tycho.ws\u003e\n Fixes: Commit 201766a20e30 (\"ptrace: add PTRACE_GET_SYSCALL_INFO request\")\n Acked-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit 5ad32b3acded06183f40806f76b030c3143017bb\nAuthor: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nDate: Sun Jun 23 16:17:10 2019 -0700\n\n isofs: Initialize filesystem timestamp ranges\n \n Fill in the appropriate limits to avoid inconsistencies\n in the vfs cached inode times when timestamps are\n outside the permitted range.\n \n Reference: http://www.ecma-international.org/publications/standards/Ecma-119.htm\n \n Signed-off-by: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\n Acked-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n\ncommit 83b8a3fbe3aa82ac3c253b698ae6a9be2dbdd5e0\nAuthor: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nDate: Sun Jun 23 16:00:01 2019 -0700\n\n pstore: fs superblock limits\n \n Leaving granularity at 1ns because it is dependent on the specific\n attached backing pstore module. ramoops has microsecond resolution.\n \n Fix the readback of ramoops fractional timestamp microseconds,\n which has incorrectly been reporting the value as nanoseconds.\n \n Fixes: 3f8f80f0cfeb (\"pstore/ram: Read and write to the \u0027compressed\u0027 flag of pstore\").\n \n Signed-off-by: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\n Acked-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Acked-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Cc: anton@enomsg.org\n Cc: ccross@android.com\n Cc: keescook@chromium.org\n Cc: tony.luck@intel.com\n\ncommit 8833293d0accf2b6a9ddaaafd198f7e1bf1d3dc6\nAuthor: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nDate: Sun Jun 23 15:10:51 2019 -0700\n\n fs: omfs: Initialize filesystem timestamp ranges\n \n Fill in the appropriate limits to avoid inconsistencies\n in the vfs cached inode times when timestamps are\n outside the permitted range.\n \n Signed-off-by: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\n Acked-by: Bob Copeland \u003cme@bobcopeland.com\u003e\n Acked-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Cc: me@bobcopeland.com\n Cc: linux-karma-devel@lists.sourceforge.net\n\ncommit cdd62b5b07e8d4e0a397254688b8178f3a7e03d1\nAuthor: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nDate: Mon Jul 1 17:09:38 2019 -0700\n\n fs: hpfs: Initialize filesystem timestamp ranges\n \n Fill in the appropriate limits to avoid inconsistencies\n in the vfs cached inode times when timestamps are\n outside the permitted range.\n \n Also change the local_to_gmt() to use time64_t instead\n of time32_t.\n \n Signed-off-by: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\n Acked-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Cc: mikulas@artax.karlin.mff.cuni.cz\n\ncommit 45d0dbd15a3b89ced53e058bc71b751e6473b664\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Mon Aug 26 17:33:17 2019 -0500\n\n drm/panfrost: Remove unnecessary hwaccess_lock spin_lock\n \n With the introduction of the as_lock to serialize address space registers,\n the hwaccess_lock is only used within the job code and is not protecting\n anything. panfrost_job_hw_submit() only accesses registers for 1 job slot\n and it\u0027s already serialized by drm_sched.\n \n Fixes: 7282f7645d06 (\"drm/panfrost: Implement per FD address spaces\")\n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190826223317.28509-9-robh@kernel.org\n\ncommit 62f1089f3cbe7d99ced92bf96a8158813b75e5e8\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Mon Aug 26 17:33:16 2019 -0500\n\n drm/panfrost: Flush and disable address space when freeing page tables\n \n Currently, page tables are freed without disabling the address space first.\n This probably is fine as we\u0027ll switch to new page tables when the address\n space is allocated again and runtime PM suspend will reset the GPU\n clearing the registers. However, it\u0027s better to clean up after ourselves.\n There is also a problem that we could be accessing the h/w in\n tlb_inv_context() when suspended.\n \n Rework the disable code to make sure we flush caches/TLBs and disable the\n address space before freeing the page tables if we are not suspended. As\n the tlb_inv_context() hook is only called when freeing the page tables and\n we do a flush before disabling the AS, lets remove the flush from\n tlb_inv_context and avoid any runtime PM issues.\n \n Fixes: 7282f7645d06 (\"drm/panfrost: Implement per FD address spaces\")\n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Reviewed-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190826223317.28509-8-robh@kernel.org\n\ncommit 5924d40958dfc2b8996fbf788a9d58e411a6db71\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Mon Aug 26 17:33:15 2019 -0500\n\n drm/panfrost: Add cache/TLB flush before switching address space\n \n It\u0027s not entirely clear if this is required, but add a flush of GPU caches\n and TLBs before we change an address space to new page tables.\n \n Fixes: 7282f7645d06 (\"drm/panfrost: Implement per FD address spaces\")\n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Reviewed-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190826223317.28509-7-robh@kernel.org\n\ncommit 86df65f39b009e00cf2826eac72c5a969b878065\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Mon Aug 26 17:33:14 2019 -0500\n\n drm/panfrost: Split mmu_hw_do_operation into locked and unlocked version\n \n In preparation to call mmu_hw_do_operation with the as_lock already held,\n Add a mmu_hw_do_operation_locked function.\n \n Fixes: 7282f7645d06 (\"drm/panfrost: Implement per FD address spaces\")\n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190826223317.28509-6-robh@kernel.org\n\ncommit ec7eba47da867e1dd0ff7d442a8717d70c064658\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Mon Aug 26 17:33:13 2019 -0500\n\n drm/panfrost: Rework page table flushing and runtime PM interaction\n \n There is no point in resuming the h/w just to do flush operations and\n doing so takes several locks which cause lockdep issues with the shrinker.\n Rework the flush operations to only happen when the h/w is already awake.\n This avoids taking any locks associated with resuming which trigger\n lockdep warnings.\n \n Fixes: 013b65101315 (\"drm/panfrost: Add madvise and shrinker support\")\n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Reviewed-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190826223317.28509-5-robh@kernel.org\n\ncommit e316f08f1abf5f1116118e25dce7bc3e9ab03246\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Mon Aug 26 17:33:12 2019 -0500\n\n drm/panfrost: Remove unnecessary mmu-\u003elock mutex\n \n There\u0027s no need to serialize io-pgtable calls and the as_lock is\n sufficient to serialize flush operations, so we can remove the per\n page table lock.\n \n Fixes: 7282f7645d06 (\"drm/panfrost: Implement per FD address spaces\")\n Suggested-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Reviewed-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190826223317.28509-4-robh@kernel.org\n\ncommit 330bec4b7ccf4a556469cfec2ad695a180e35954\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Mon Aug 26 17:33:11 2019 -0500\n\n drm/panfrost: Hold runtime PM reference until jobs complete\n \n Doing a pm_runtime_put as soon as a job is submitted is wrong as it should\n not happen until the job completes. It works currently because we are\n relying on the autosuspend timeout to keep the h/w enabled.\n \n Fixes: f3ba91228e8e (\"drm/panfrost: Add initial panfrost driver\")\n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: Steven Price \u003csteven.price@arm.com\u003e\n Cc: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190826223317.28509-3-robh@kernel.org\n\ncommit 635430797d3fccb958e929d3911e39cb9c1ea641\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Mon Aug 26 17:33:10 2019 -0500\n\n drm/panfrost: Rework runtime PM initialization\n \n There\u0027s a few issues with the runtime PM initialization.\n \n The documentation states pm_runtime_set_active() should be called before\n pm_runtime_enable(). The pm_runtime_put_autosuspend() could suspend the GPU\n before panfrost_perfcnt_init() is called which touches the h/w. The\n autosuspend delay keeps things from breaking. There\u0027s no need explicitly\n power off the GPU only to wake back up with pm_runtime_get_sync(). Just\n delaying pm_runtime_enable to the end of probe is sufficient.\n \n Lets move all the runtime PM calls into the probe() function so they are\n all in one place and are done after all initialization.\n \n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: Steven Price \u003csteven.price@arm.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190826223317.28509-2-robh@kernel.org\n\ncommit 8758553791dfb31ec2640f6b69df748720c0b1a4\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Thu Aug 29 23:03:27 2019 -0500\n\n iommu/qcom: Use struct_size() helper\n \n One of the more common cases of allocation size calculations is finding\n the size of a structure that has a zero-sized array at the end, along\n with memory for some number of elements for that array. For example:\n \n struct qcom_iommu_dev {\n ...\n struct qcom_iommu_ctx *ctxs[0]; /* indexed by asid-1 */\n };\n \n Make use of the struct_size() helper instead of an open-coded version\n in order to avoid any potential type mistakes.\n \n So, replace the following form:\n \n sizeof(*qcom_iommu) + (max_asid * sizeof(qcom_iommu-\u003ectxs[0]))\n \n with:\n \n struct_size(qcom_iommu, ctxs, max_asid)\n \n Also, notice that, in this case, variable sz is not necessary,\n hence it is removed.\n \n This code was detected with the help of Coccinelle.\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit d127bc9be856098cc2410c1266ed64e258bc5377\nAuthor: Tom Murphy \u003cmurphyt7@tcd.ie\u003e\nDate: Mon Aug 26 05:48:21 2019 +0100\n\n iommu: Remove wrong default domain comments\n \n These comments are wrong. request_default_domain_for_dev doesn\u0027t just\n handle direct mapped domains.\n \n Signed-off-by: Tom Murphy \u003cmurphyt7@tcd.ie\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 7893b59b1e2d3ff954ecc5a407f7e4fde078d6d2\nAuthor: Jacob Pan \u003cjacob.jun.pan@linux.intel.com\u003e\nDate: Mon Aug 26 08:53:29 2019 -0700\n\n iommu/vt-d: Remove global page flush support\n \n Global pages support is removed from VT-d spec 3.0. Since global pages G\n flag only affects first-level paging structures and because DMA request\n with PASID are only supported by VT-d spec. 3.0 and onward, we can\n safely remove global pages support.\n \n For kernel shared virtual address IOTLB invalidation, PASID\n granularity and page selective within PASID will be used. There is\n no global granularity supported. Without this fix, IOTLB invalidation\n will cause invalid descriptor error in the queued invalidation (QI)\n interface.\n \n Fixes: 1c4f88b7f1f9 (\"iommu/vt-d: Shared virtual address in scalable\n mode\")\n Reported-by: Sanjay K Kumar \u003csanjay.k.kumar@intel.com\u003e\n Signed-off-by: Jacob Pan \u003cjacob.jun.pan@linux.intel.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 028ca4db0a6e0be470ae866d1afd63e8740aa3e7\nAuthor: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nDate: Thu Mar 21 14:34:38 2019 -0700\n\n fs: ceph: Initialize filesystem timestamp ranges\n \n Fill in the appropriate limits to avoid inconsistencies\n in the vfs cached inode times when timestamps are\n outside the permitted range.\n \n According to the disscussion in\n https://patchwork.kernel.org/patch/8308691/ we agreed to use\n unsigned 32 bit timestamps on ceph.\n Update the limits accordingly.\n \n Signed-off-by: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\n Acked-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Cc: zyan@redhat.com\n Cc: sage@redhat.com\n Cc: idryomov@gmail.com\n Cc: ceph-devel@vger.kernel.org\n\ncommit 452c2779410a03ac0c6be0a8a91c83aa80bdd7e5\nAuthor: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nDate: Fri Mar 8 12:40:03 2019 -0800\n\n fs: sysv: Initialize filesystem timestamp ranges\n \n Fill in the appropriate limits to avoid inconsistencies\n in the vfs cached inode times when timestamps are\n outside the permitted range.\n \n Signed-off-by: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\n Acked-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Cc: hch@infradead.org\n\ncommit 487b25bc4be9c15a5edb81c4e52160b5e84d1c4f\nAuthor: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nDate: Thu Mar 21 14:05:12 2019 -0700\n\n fs: affs: Initialize filesystem timestamp ranges\n \n Fill in the appropriate limits to avoid inconsistencies\n in the vfs cached inode times when timestamps are\n outside the permitted range.\n \n Also fix timestamp calculation to avoid overflow\n while converting from days to seconds.\n \n Signed-off-by: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\n Acked-by: David Sterba \u003cdsterba@suse.com\u003e\n Acked-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Cc: dsterba@suse.com\n\ncommit c0da64f6bb674d20a8e84bcd30679c1483668fb2\nAuthor: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nDate: Fri Mar 22 16:42:54 2019 -0700\n\n fs: fat: Initialize filesystem timestamp ranges\n \n Fill in the appropriate limits to avoid inconsistencies\n in the vfs cached inode times when timestamps are\n outside the permitted range.\n \n Some FAT variants indicate that the years after 2099 are not supported.\n Since commit 7decd1cb0305 (\"fat: Fix and cleanup timestamp conversion\")\n we support the full range of years that can be represented, up to 2107.\n \n Signed-off-by: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\n Acked-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Cc: hirofumi@mail.parknet.co.jp\n\ncommit cb7a69e605908c34aad47644afeb26a765ade8d7\nAuthor: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nDate: Fri Mar 22 14:32:35 2019 -0700\n\n fs: cifs: Initialize filesystem timestamp ranges\n \n Fill in the appropriate limits to avoid inconsistencies\n in the vfs cached inode times when timestamps are\n outside the permitted range.\n \n Also fixed cnvrtDosUnixTm calculations to avoid int overflow\n while computing maximum date.\n \n References:\n \n http://cifs.com/\n \n https://docs.microsoft.com/en-us/openspecs/windows_protocols/ms-cifs/d416ff7c-c536-406e-a951-4f04b2fd1d2b\n \n Signed-off-by: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\n Acked-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Cc: sfrench@samba.org\n Cc: linux-cifs@vger.kernel.org\n\ncommit 1fcb79c1b218018b64919de8af20e64340850716\nAuthor: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nDate: Tue Mar 26 12:55:56 2019 -0700\n\n fs: nfs: Initialize filesystem timestamp ranges\n \n Fill in the appropriate limits to avoid inconsistencies\n in the vfs cached inode times when timestamps are\n outside the permitted range.\n \n The time formats for various verious is detailed in the\n RFCs as below:\n \n https://tools.ietf.org/html/rfc7862(time metadata)\n https://tools.ietf.org/html/rfc7530:\n \n nfstime4\n \n struct nfstime4 {\n int64_t seconds;\n uint32_t nseconds;\n };\n \n https://tools.ietf.org/html/rfc1094\n \n struct timeval {\n unsigned int seconds;\n unsigned int useconds;\n };\n \n https://tools.ietf.org/html/rfc1813\n \n struct nfstime3 {\n uint32 seconds;\n uint32 nseconds;\n };\n \n Use the limits as per the RFC.\n \n Signed-off-by: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\n Acked-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Cc: trond.myklebust@hammerspace.com\n Cc: anna.schumaker@netapp.com\n Cc: linux-nfs@vger.kernel.org\n\ncommit 4881c4971df04107b37c29bb6c719ec29ceb6571\nAuthor: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nDate: Sun Jan 21 18:04:24 2018 -0800\n\n ext4: Initialize timestamps limits\n \n ext4 has different overflow limits for max filesystem\n timestamps based on the extra bytes available.\n \n The timestamp limits are calculated according to the\n encoding table in\n a4dad1ae24f85i(ext4: Fix handling of extended tv_sec):\n \n * extra msb of adjust for signed\n * epoch 32-bit 32-bit tv_sec to\n * bits time decoded 64-bit tv_sec 64-bit tv_sec valid time range\n * 0 0 1 -0x80000000..-0x00000001 0x000000000 1901-12-13..1969-12-31\n * 0 0 0 0x000000000..0x07fffffff 0x000000000 1970-01-01..2038-01-19\n * 0 1 1 0x080000000..0x0ffffffff 0x100000000 2038-01-19..2106-02-07\n * 0 1 0 0x100000000..0x17fffffff 0x100000000 2106-02-07..2174-02-25\n * 1 0 1 0x180000000..0x1ffffffff 0x200000000 2174-02-25..2242-03-16\n * 1 0 0 0x200000000..0x27fffffff 0x200000000 2242-03-16..2310-04-04\n * 1 1 1 0x280000000..0x2ffffffff 0x300000000 2310-04-04..2378-04-22\n * 1 1 0 0x300000000..0x37fffffff 0x300000000 2378-04-22..2446-05-10\n \n Note that the time limits are not correct for deletion times.\n \n Added a warn when an inode cannot be extended to incorporate an\n extended timestamp.\n \n Signed-off-by: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\n Reviewed-by: Andreas Dilger \u003cadilger@dilger.ca\u003e\n Acked-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Cc: tytso@mit.edu\n Cc: adilger.kernel@dilger.ca\n Cc: linux-ext4@vger.kernel.org\n\ncommit d5c6e2d5188d9677462f3827034d060372a13b6e\nAuthor: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nDate: Wed Mar 6 14:09:42 2019 -0800\n\n 9p: Fill min and max timestamps in sb\n \n struct p9_wstat and struct p9_stat_dotl indicate that the\n wire transport uses u32 and u64 fields for timestamps.\n Fill in the appropriate limits to avoid inconsistencies in\n the vfs cached inode times when timestamps are outside the\n permitted range.\n \n Note that the upper bound for V9FS_PROTO_2000L is retained as S64_MAX.\n This is because that is the upper bound supported by vfs.\n \n Signed-off-by: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\n Acked-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Cc: ericvh@gmail.com\n Cc: lucho@ionkov.net\n Cc: asmadeus@codewreck.org\n Cc: v9fs-developer@lists.sourceforge.net\n\ncommit 22b139691f9eb8b9d0bfd7341fa7436cb7a9491d\nAuthor: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nDate: Tue Jul 30 08:22:29 2019 -0700\n\n fs: Fill in max and min timestamps in superblock\n \n Fill in the appropriate limits to avoid inconsistencies\n in the vfs cached inode times when timestamps are\n outside the permitted range.\n \n Even though some filesystems are read-only, fill in the\n timestamps to reflect the on-disk representation.\n \n Signed-off-by: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Acked-By: Tigran Aivazian \u003caivazian.tigran@gmail.com\u003e\n Acked-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Cc: aivazian.tigran@gmail.com\n Cc: al@alarsen.net\n Cc: coda@cs.cmu.edu\n Cc: darrick.wong@oracle.com\n Cc: dushistov@mail.ru\n Cc: dwmw2@infradead.org\n Cc: hch@infradead.org\n Cc: jack@suse.com\n Cc: jaharkes@cs.cmu.edu\n Cc: luisbg@kernel.org\n Cc: nico@fluxnic.net\n Cc: phillip@squashfs.org.uk\n Cc: richard@nod.at\n Cc: salah.triki@gmail.com\n Cc: shaggy@kernel.org\n Cc: linux-xfs@vger.kernel.org\n Cc: codalist@coda.cs.cmu.edu\n Cc: linux-ext4@vger.kernel.org\n Cc: linux-mtd@lists.infradead.org\n Cc: jfs-discussion@lists.sourceforge.net\n Cc: reiserfs-devel@vger.kernel.org\n\ncommit 42e729b9ddbbc40e5732f062ef2fa0554c652fb5\nAuthor: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nDate: Sun Jan 21 18:04:26 2018 -0800\n\n utimes: Clamp the timestamps before update\n \n POSIX is ambiguous on the behavior of timestamps for\n futimens, utimensat and utimes. Whether to return an\n error or silently clamp a timestamp beyond the range\n supported by the underlying filesystems is not clear.\n \n POSIX.1 section for futimens, utimensat and utimes says:\n (http://pubs.opengroup.org/onlinepubs/9699919799/functions/futimens.html)\n \n The file\u0027s relevant timestamp shall be set to the greatest\n value supported by the file system that is not greater\n than the specified time.\n \n If the tv_nsec field of a timespec structure has the special\n value UTIME_NOW, the file\u0027s relevant timestamp shall be set\n to the greatest value supported by the file system that is\n not greater than the current time.\n \n [EINVAL]\n A new file timestamp would be a value whose tv_sec\n component is not a value supported by the file system.\n \n The patch chooses to clamp the timestamps according to the\n filesystem timestamp ranges and does not return an error.\n This is in line with the behavior of utime syscall also\n since the POSIX page(http://pubs.opengroup.org/onlinepubs/009695399/functions/utime.html)\n for utime does not mention returning an error or clamping like above.\n \n Same for utimes http://pubs.opengroup.org/onlinepubs/009695399/functions/utimes.html\n \n Signed-off-by: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\n Acked-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n\ncommit f8b92ba67c5d3a9e9468320078a97d950a3e748b\nAuthor: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nDate: Mon Apr 15 14:17:12 2019 -0700\n\n mount: Add mount warning for impending timestamp expiry\n \n The warning reuses the uptime max of 30 years used by\n settimeofday().\n \n Note that the warning is only emitted for writable filesystem mounts\n through the mount syscall. Automounts do not have the same warning.\n \n Print out the warning in human readable format using the struct tm.\n After discussion with Arnd Bergmann, we chose to print only the year number.\n The raw s_time_max is also displayed, and the user can easily decode\n it e.g. \"date -u -d @$((0x7fffffff))\". We did not want to consolidate\n struct rtc_tm and struct tm just to print the date using a format specifier\n as part of this series.\n Given that the rtc_tm is not compiled on all architectures, this is not a\n trivial patch. This can be added in the future.\n \n Signed-off-by: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\n Acked-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n\ncommit 3818c1907a5e4e8fbd57fb14cea77de7c507111a\nAuthor: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nDate: Sat Jun 15 09:00:05 2019 -0700\n\n timestamp_truncate: Replace users of timespec64_trunc\n \n Update the inode timestamp updates to use timestamp_truncate()\n instead of timespec64_trunc().\n \n The change was mostly generated by the following coccinelle\n script.\n \n virtual context\n virtual patch\n \n @r1 depends on patch forall@\n struct inode *inode;\n identifier i_xtime \u003d~ \"^i_[acm]time$\";\n expression e;\n @@\n \n inode-\u003ei_xtime \u003d\n - timespec64_trunc(\n + timestamp_truncate(\n ...,\n - e);\n + inode);\n \n Signed-off-by: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\n Acked-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Acked-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Cc: adrian.hunter@intel.com\n Cc: dedekind1@gmail.com\n Cc: gregkh@linuxfoundation.org\n Cc: hch@lst.de\n Cc: jaegeuk@kernel.org\n Cc: jlbec@evilplan.org\n Cc: richard@nod.at\n Cc: tj@kernel.org\n Cc: yuchao0@huawei.com\n Cc: linux-f2fs-devel@lists.sourceforge.net\n Cc: linux-ntfs-dev@lists.sourceforge.net\n Cc: linux-mtd@lists.infradead.org\n\ncommit 50e17c000c467fbc927fc001df99beb4027a5323\nAuthor: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nDate: Sun Jan 21 18:04:25 2018 -0800\n\n vfs: Add timestamp_truncate() api\n \n timespec_trunc() function is used to truncate a\n filesystem timestamp to the right granularity.\n But, the function does not clamp tv_sec part of the\n timestamps according to the filesystem timestamp limits.\n \n The replacement api: timestamp_truncate() also alters the\n signature of the function to accommodate filesystem\n timestamp clamping according to flesystem limits.\n \n Note that the tv_nsec part is set to 0 if tv_sec is not within\n the range supported for the filesystem.\n \n Signed-off-by: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\n Acked-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n\ncommit 188d20bcd1ebd8277d9b8a79525bd66b66d40a2a\nAuthor: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nDate: Sun Jan 21 18:04:23 2018 -0800\n\n vfs: Add file timestamp range support\n \n Add fields to the superblock to track the min and max\n timestamps supported by filesystems.\n \n Initially, when a superblock is allocated, initialize\n it to the max and min values the fields can hold.\n Individual filesystems override these to match their\n actual limits.\n \n Pseudo filesystems are assumed to always support the\n min and max allowable values for the fields.\n \n Signed-off-by: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\n Acked-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n\ncommit 0ce4a85f4f015d13a6a6b4954f4e8e63cd83f7b1\nAuthor: Lu Baolu \u003cbaolu.lu@linux.intel.com\u003e\nDate: Mon Aug 26 16:50:56 2019 +0800\n\n Revert \"iommu/vt-d: Avoid duplicated pci dma alias consideration\"\n \n This reverts commit 557529494d79f3f1fadd486dd18d2de0b19be4da.\n \n Commit 557529494d79f (\"iommu/vt-d: Avoid duplicated pci dma alias\n consideration\") aimed to address a NULL pointer deference issue\n happened when a thunderbolt device driver returned unexpectedly.\n \n Unfortunately, this change breaks a previous pci quirk added by\n commit cc346a4714a59 (\"PCI: Add function 1 DMA alias quirk for\n Marvell devices\"), as the result, devices like Marvell 88SE9128\n SATA controller doesn\u0027t work anymore.\n \n We will continue to try to find the real culprit mentioned in\n 557529494d79f, but for now we should revert it to fix current\n breakage.\n \n Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id\u003d204627\n Cc: Stijn Tintel \u003cstijn@linux-ipv6.be\u003e\n Cc: Petr Vandrovec \u003cpetr@vandrovec.name\u003e\n Reported-by: Stijn Tintel \u003cstijn@linux-ipv6.be\u003e\n Reported-by: Petr Vandrovec \u003cpetr@vandrovec.name\u003e\n Signed-off-by: Lu Baolu \u003cbaolu.lu@linux.intel.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 2178add02238f8352f5b3294a79f4763183aade6\nAuthor: Lubomir Rintel \u003clkundrak@v3.sk\u003e\nDate: Thu Aug 22 11:26:32 2019 +0200\n\n irqchip/mmp: Coexist with GIC root IRQ controller\n \n On MMP3, the GIC can be set as a root IRQ interrupt controller. If the\n device tree indicated that GIC is enabled, avoid hooking up\n mmp2_handle_irq().\n \n The interrupt muxes are still being used.\n \n Signed-off-by: Lubomir Rintel \u003clkundrak@v3.sk\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190822092643.593488-10-lkundrak@v3.sk\n\ncommit 9e8e8912b05f276dd02d39cb596dc3cf03718377\nAuthor: Andres Salomon \u003cdilinger@queued.net\u003e\nDate: Thu Aug 22 11:26:31 2019 +0200\n\n irqchip/mmp: Mask off interrupts from other cores\n \n On mmp3, there\u0027s an extra set of ICU registers (ICU2) that handle\n interrupts on the extra cores. When masking off interrupts on MP1,\n these should be masked as well.\n \n We add a new interrupt controller via device tree to identify when we\u0027re\n looking at an mmp3 machine via compatible field of \"marvell,mmp3-intc\".\n \n [lkundrak@v3.sk: Changed \"mrvl,mmp3-intc\" compatible strings to\n \"marvell,mmp3-intc\". Tidied up the subject line a bit.]\n \n Signed-off-by: Andres Salomon \u003cdilinger@queued.net\u003e\n Signed-off-by: Lubomir Rintel \u003clkundrak@v3.sk\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190822092643.593488-9-lkundrak@v3.sk\n --\n Changes since v1:\n - Moved mmp3-specific mmp_icu2_base initialization from mmp_init_bases() to\n mmp3_of_init() so that we don\u0027t have to check for marvell,mmp3-intc\n compatibility twice.\n - Drop an superfluous call to irq_set_default_host()\n \n arch/arm/mach-mmp/regs-icu.h | 3 +++\n drivers/irqchip/irq-mmp.c | 48 ++++++++++++++++++++++++++++++++++++\n 2 files changed, 51 insertions(+)\n \n Signed-off-by: Andres Salomon \u003cdilinger@queued.net\u003e\n Signed-off-by: Lubomir Rintel \u003clkundrak@v3.sk\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190822092643.593488-9-lkundrak@v3.sk\n\ncommit a46bc5fd8b205050ebbdccc6d5ca4124edb8dc6c\nAuthor: Lubomir Rintel \u003clkundrak@v3.sk\u003e\nDate: Thu Aug 22 11:26:30 2019 +0200\n\n irqchip/mmp: Add missing chained_irq_{enter,exit}()\n \n The lack of chained_irq_exit() leaves the muxed interrupt masked on MMP3.\n For reasons unknown this is not a problem on MMP2.\n \n Signed-off-by: Lubomir Rintel \u003clkundrak@v3.sk\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190822092643.593488-8-lkundrak@v3.sk\n\ncommit d6a95280ba169c3a3d632d983cc6977c544a06e8\nAuthor: Lubomir Rintel \u003clkundrak@v3.sk\u003e\nDate: Thu Aug 22 11:26:29 2019 +0200\n\n irqchip/mmp: Do not use of_address_to_resource() to get mux regs\n \n The \"regs\" property of the \"mrvl,mmp2-mux-intc\" devices are silly. They\n are offsets from intc\u0027s base, not addresses on the parent bus. At this\n point it probably can\u0027t be fixed.\n \n On an OLPC XO-1.75 machine, the muxes are children of the intc, not the\n axi bus, and thus of_address_to_resource() won\u0027t work. We should treat\n the values as mere integers as opposed to bus addresses.\n \n Signed-off-by: Lubomir Rintel \u003clkundrak@v3.sk\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Link: https://lore.kernel.org/r/20190822092643.593488-7-lkundrak@v3.sk\n\ncommit 6b0c54e7f2715997c366e8374209bc74259b0a59\nAuthor: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\nDate: Sat Aug 24 09:47:12 2019 +0800\n\n iommu/dma: Fix for dereferencing before null checking\n \n The cookie is dereferenced before null checking in the function\n iommu_dma_init_domain.\n \n This patch moves the dereferencing after the null checking.\n \n Fixes: fdbe574eb693 (\"iommu/dma: Allow MSI-only cookies\")\n Signed-off-by: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit b2fb4b77994abc1107c35547f3e123dce8e9f67d\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Thu Aug 29 18:16:35 2019 +0200\n\n irqchip/meson-gpio: Add support for meson sm1 SoCs\n \n The meson sm1 SoCs uses the same type of GPIO interrupt controller IP\n block as the other meson SoCs, A total of 100 pins can be spied on:\n \n - 223:100 undefined (no interrupt)\n - 99:97 3 pins on bank GPIOE\n - 96:77 20 pins on bank GPIOX\n - 76:61 16 pins on bank GPIOA\n - 60:53 8 pins on bank GPIOC\n - 52:37 16 pins on bank BOOT\n - 36:28 9 pins on bank GPIOH\n - 27:12 16 pins on bank GPIOZ\n - 11:0 12 pins in the AO domain\n \n Mapping is the same as the g12a family but the sm1 controller\n allows to trig an irq on both edges of the input signal. This was\n not possible with the previous SoCs families\n \n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n Tested-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Link: https://lore.kernel.org/r/20190829161635.25067-3-jbrunet@baylibre.com\n\ncommit abc08aac82af0c71e30b446575f5810c9cc11640\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Thu Aug 29 18:16:34 2019 +0200\n\n dt-bindings: interrupt-controller: New binding for the meson sm1 SoCs\n \n Update the dt-binding to add support for the sm1 SoC family in the\n amlogic GPIO interrupt controller driver.\n \n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190829161635.25067-2-jbrunet@baylibre.com\n\ncommit 1ee9feb2c9f893b893c900d2492c6a01dca680f3\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:02:08 2019 +0800\n\n iommu/mediatek: Clean up struct mtk_smi_iommu\n \n Remove the \"struct mtk_smi_iommu\" to simplify the code since it has only\n one item in it right now.\n \n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Reviewed-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit ec2da07ca1202552d87fb01b238d46642817da2b\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:02:07 2019 +0800\n\n memory: mtk-smi: Get rid of need_larbid\n \n The \"mediatek,larb-id\" has already been parsed in MTK IOMMU driver.\n It\u0027s no need to parse it again in SMI driver. Only clean some codes.\n This patch is fit for all the current mt2701, mt2712, mt7623, mt8173\n and mt8183.\n \n After this patch, the \"mediatek,larb-id\" only be needed for mt2712\n which have 2 M4Us. In the other SoCs, we can get the larb-id from M4U\n in which the larbs in the \"mediatek,larbs\" always are ordered.\n \n Correspondingly, the larb_nr in the \"struct mtk_smi_iommu\" could also\n be deleted.\n \n CC: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Reviewed-by: Evan Green \u003cevgreen@chromium.org\u003e\n Reviewed-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit b9475b3471f81a60a727886230b0903be72dc96d\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:02:06 2019 +0800\n\n iommu/mediatek: Fix VLD_PA_RNG register backup when suspend\n \n The register VLD_PA_RNG(0x118) was forgot to backup while adding 4GB\n mode support for mt2712. this patch add it.\n \n Fixes: 30e2fccf9512 (\"iommu/mediatek: Enlarge the validate PA range\n for 4GB mode\")\n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Reviewed-by: Evan Green \u003cevgreen@chromium.org\u003e\n Reviewed-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 567e58cf96dda2cc1d8dbdafbc1c089b5ea2be25\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:02:05 2019 +0800\n\n memory: mtk-smi: Add bus_sel for mt8183\n \n There are 2 mmu cells in a M4U HW. we could adjust some larbs entering\n mmu0 or mmu1 to balance the bandwidth via the smi-common register\n SMI_BUS_SEL(0x220)(Each larb occupy 2 bits).\n \n In mt8183, For better performance, we switch larb1/2/5/7 to enter\n mmu1 while the others still keep enter mmu0.\n \n In mt8173 and mt2712, we don\u0027t get the performance issue,\n Keep its default value(0x0), that means all the larbs enter mmu0.\n \n Note: smi gen1(mt2701/mt7623) don\u0027t have this bus_sel.\n \n And, the base of smi-common is completely different with smi_ao_base\n of gen1, thus I add new variable for that.\n \n CC: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Reviewed-by: Evan Green \u003cevgreen@chromium.org\u003e\n Reviewed-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 4f0a1a1ae35197e9b7a6d11b4e881bcfe973ad8f\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:02:04 2019 +0800\n\n memory: mtk-smi: Invoke pm runtime_callback to enable clocks\n \n This patch only move the clk_prepare_enable and config_port into the\n runtime suspend/resume callback. It doesn\u0027t change the code content\n and sequence.\n \n This is a preparing patch for adjusting SMI_BUS_SEL for mt8183.\n (SMI_BUS_SEL need to be restored after smi-common resume every time.)\n Also it gives a chance to get rid of mtk_smi_larb_get/put which could\n be a next topic.\n \n CC: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Reviewed-by: Evan Green \u003cevgreen@chromium.org\u003e\n Reviewed-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 15a01f4c60607ad888faf0386070181f9c97577f\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:02:03 2019 +0800\n\n iommu/mediatek: Add mmu1 support\n \n Normally the M4U HW connect EMI with smi. the diagram is like below:\n EMI\n |\n M4U\n |\n smi-common\n |\n -----------------\n | | | | ...\n larb0 larb1 larb2 larb3\n \n Actually there are 2 mmu cells in the M4U HW, like this diagram:\n \n EMI\n ---------\n | |\n mmu0 mmu1 \u003c- M4U\n | |\n ---------\n |\n smi-common\n |\n -----------------\n | | | | ...\n larb0 larb1 larb2 larb3\n \n This patch add support for mmu1. In order to get better performance,\n we could adjust some larbs go to mmu1 while the others still go to\n mmu0. This is controlled by a SMI COMMON register SMI_BUS_SEL(0x220).\n \n mt2712, mt8173 and mt8183 M4U HW all have 2 mmu cells. the default\n value of that register is 0 which means all the larbs go to mmu0\n defaultly.\n \n This is a preparing patch for adjusting SMI_BUS_SEL for mt8183.\n \n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Reviewed-by: Evan Green \u003cevgreen@chromium.org\u003e\n Reviewed-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 907ba6a195991adb2a3edf4aff0d8dbb308d4d97\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:02:02 2019 +0800\n\n iommu/mediatek: Add mt8183 IOMMU support\n \n The M4U IP blocks in mt8183 is MediaTek\u0027s generation2 M4U which use\n the ARM Short-descriptor like mt8173, and most of the HW registers\n are the same.\n \n Here list main differences between mt8183 and mt8173/mt2712:\n 1) mt8183 has only one M4U HW like mt8173 while mt2712 has two.\n 2) mt8183 don\u0027t have the \"bclk\" clock, it use the EMI clock instead.\n 3) mt8183 can support the dram over 4GB, but it doesn\u0027t call this \"4GB\n mode\".\n 4) mt8183 pgtable base register(0x0) extend bit[1:0] which represent\n the bit[33:32] in the physical address of the pgtable base, But the\n standard ttbr0[1] means the S bit which is enabled defaultly, Hence,\n we add a mask.\n 5) mt8183 HW has a GALS modules, SMI should enable \"has_gals\" support.\n 6) mt8183 need reset_axi like mt8173.\n 7) the larb-id in smi-common is remapped. M4U should add its larbid_remap.\n \n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Reviewed-by: Evan Green \u003cevgreen@chromium.org\u003e\n Reviewed-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 64fea74a0d205ac9aff39997e5fe9d64115a4bd3\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:02:01 2019 +0800\n\n memory: mtk-smi: Add gals support\n \n In some SoCs like mt8183, SMI add GALS(Global Async Local Sync) module\n which can help synchronize for the modules in different clock frequency.\n It can be seen as a \"asynchronous fifo\". This is a example diagram:\n \n M4U\n |\n ----------\n | |\n gals0-rx gals1-rx\n | |\n | |\n gals0-tx gals1-tx\n | |\n ------------\n SMI Common\n ------------\n |\n +-----+--------+-----+- ...\n | | | |\n | gals-rx gals-rx |\n | | | |\n | | | |\n | gals-tx gals-tx |\n | | | |\n larb1 larb2 larb3 larb4\n \n GALS only help transfer the command/data while it doesn\u0027t have the\n configuring register, thus it has the special \"smi\" clock and doesn\u0027t\n have the \"apb\" clock. From the diagram above, we add \"gals0\" and\n \"gals1\" clocks for smi-common and add a \"gals\" clock for smi-larb.\n \n This patch adds gals clock supporting in the SMI. Note that some larbs\n may still don\u0027t have the \"gals\" clock like larb1 and larb4 above.\n \n This is also a preparing patch for mt8183 which has GALS.\n \n CC: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Reviewed-by: Evan Green \u003cevgreen@chromium.org\u003e\n Reviewed-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 2b326d8b1d2b443c3dad96f54c6b279559f87751\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:02:00 2019 +0800\n\n iommu/mediatek: Move vld_pa_rng into plat_data\n \n Both mt8173 and mt8183 don\u0027t have this vld_pa_rng(valid physical address\n range) register while mt2712 have. Move it into the plat_data.\n \n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Reviewed-by: Evan Green \u003cevgreen@chromium.org\u003e\n Reviewed-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 50822b0b948fabfc8bcee9a89d031c276b135506\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:01:59 2019 +0800\n\n iommu/mediatek: Move reset_axi into plat_data\n \n In mt8173 and mt8183, 0x48 is REG_MMU_STANDARD_AXI_MODE while it is\n REG_MMU_CTRL in the other SoCs, and the bits meaning is completely\n different with the REG_MMU_STANDARD_AXI_MODE.\n \n This patch moves this property to plat_data, it\u0027s also a preparing\n patch for mt8183.\n \n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Reviewed-by: Nicolas Boichat \u003cdrinkcat@chromium.org\u003e\n Reviewed-by: Evan Green \u003cevgreen@chromium.org\u003e\n Reviewed-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit acb3c92a61306a8bc3b6bb8ed72977201affdd9a\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:01:58 2019 +0800\n\n iommu/mediatek: Refine protect memory definition\n \n The protect memory setting is a little different in the different SoCs.\n In the register REG_MMU_CTRL_REG(0x110), the TF_PROT(translation fault\n protect) shift bit is normally 4 while it shift 5 bits only in the\n mt8173. This patch delete the complex MACRO and use a common if-else\n instead.\n \n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Reviewed-by: Evan Green \u003cevgreen@chromium.org\u003e\n Reviewed-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit b3e5eee76548f621bb1c3e9dab4a4ff98bebf0de\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:01:57 2019 +0800\n\n iommu/mediatek: Add larb-id remapped support\n \n The larb-id may be remapped in the smi-common, this means the\n larb-id reported in the mtk_iommu_isr isn\u0027t the real larb-id,\n \n Take mt8183 as a example:\n M4U\n |\n ---------------------------------------------\n | SMI common |\n -0-----7-----5-----6-----1-----2------3-----4- \u003c- Id remapped\n | | | | | | | |\n larb0 larb1 IPU0 IPU1 larb4 larb5 larb6 CCU\n disp vdec img cam venc img cam\n As above, larb0 connects with the id 0 in smi-common.\n larb1 connects with the id 7 in smi-common.\n ...\n If the larb-id reported in the isr is 7, actually it\u0027s larb1(vdec).\n In order to output the right larb-id in the isr, we add a larb-id\n remapping relationship in this patch.\n \n If there is no this larb-id remapping in some SoCs, use the linear\n mapping array instead.\n \n This also is a preparing patch for mt8183.\n \n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Reviewed-by: Nicolas Boichat \u003cdrinkcat@chromium.org\u003e\n Reviewed-by: Evan Green \u003cevgreen@chromium.org\u003e\n Reviewed-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n \n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 2aa4c2597cd67139697eb3095d5b6d61074c7ffc\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:01:56 2019 +0800\n\n iommu/mediatek: Add bclk can be supported optionally\n \n In some SoCs, M4U doesn\u0027t have its \"bclk\", it will use the EMI\n clock instead which has always been enabled when entering kernel.\n \n Currently mt2712 and mt8173 have this bclk while mt8183 doesn\u0027t.\n \n This also is a preparing patch for mt8183.\n \n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Reviewed-by: Evan Green \u003cevgreen@chromium.org\u003e\n Reviewed-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit b4dad40e4f35bbf2393f35f4492acf799eb8136d\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:01:55 2019 +0800\n\n iommu/mediatek: Adjust the PA for the 4GB Mode\n \n After extending the v7s support PA[33:32] for MediaTek, we have to adjust\n the PA ourself for the 4GB mode.\n \n In the 4GB Mode, the PA will remap like this:\n CPU PA -\u003e M4U output PA\n 0x4000_0000 0x1_4000_0000 (Add bit32)\n 0x8000_0000 0x1_8000_0000 ...\n 0xc000_0000 0x1_c000_0000 ...\n 0x1_0000_0000 0x1_0000_0000 (No change)\n \n 1) Always add bit32 for CPU PA in -\u003emap.\n 2) Discard the bit32 in iova_to_phys if PA \u003e 0x1_4000_0000 since the\n iommu consumer always use the CPU PA.\n \n Besides, the \"oas\" always is set to 34 since v7s has already supported our\n case.\n \n Both mt2712 and mt8173 support this \"4GB mode\" while the mt8183 don\u0027t.\n The PA in mt8183 won\u0027t remap.\n \n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 4c019de653237674d38cf2b3119153b144ffe173\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:01:54 2019 +0800\n\n iommu/io-pgtable-arm-v7s: Extend to support PA[33:32] for MediaTek\n \n MediaTek extend the arm v7s descriptor to support up to 34 bits PA where\n the bit32 and bit33 are encoded in the bit9 and bit4 of the PTE\n respectively. Meanwhile the iova still is 32bits.\n \n Regarding whether the pagetable address could be over 4GB, the mt8183\n support it while the previous mt8173 don\u0027t, thus keep it as is.\n \n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Acked-by: Will Deacon \u003cwill@kernel.org\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 73d50811bc91d2a173213a78b6b43ac762f6cc54\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:01:53 2019 +0800\n\n iommu/io-pgtable-arm-v7s: Rename the quirk from MTK_4GB to MTK_EXT\n \n In previous mt2712/mt8173, MediaTek extend the v7s to support 4GB dram.\n But in the latest mt8183, We extend it to support the PA up to 34bit.\n Then the \"MTK_4GB\" name is not so fit, This patch only change the quirk\n name to \"MTK_EXT\".\n \n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Acked-by: Will Deacon \u003cwill@kernel.org\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 7f315c9da9f5e5f9ba720ca4b5e04e5ee91833f9\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:01:52 2019 +0800\n\n iommu/io-pgtable-arm-v7s: Use ias/oas to check the valid iova/pa\n \n Use ias/oas to check the valid iova/pa. Synchronize this checking with\n io-pgtable-arm.c.\n \n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Acked-by: Will Deacon \u003cwill@kernel.org\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 5950b9541b509037aeb9e7b95f88a692bf3f46a3\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:01:51 2019 +0800\n\n iommu/io-pgtable-arm-v7s: Add paddr_to_iopte and iopte_to_paddr helpers\n \n Add two helper functions: paddr_to_iopte and iopte_to_paddr.\n \n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Reviewed-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Reviewed-by: Evan Green \u003cevgreen@chromium.org\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 76ce65464fcd2c21db84391572b7938b716aceb0\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:01:50 2019 +0800\n\n iommu/mediatek: Fix iova_to_phys PA start for 4GB mode\n \n In M4U 4GB mode, the physical address is remapped as below:\n \n CPU Physical address:\n \n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n \n 0 1G 2G 3G 4G 5G\n |---A---|---B---|---C---|---D---|---E---|\n +--I/O--+------------Memory-------------+\n \n IOMMU output physical address:\n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n \n 4G 5G 6G 7G 8G\n |---E---|---B---|---C---|---D---|\n +------------Memory-------------+\n \n The Region \u0027A\u0027(I/O) can not be mapped by M4U; For Region \u0027B\u0027/\u0027C\u0027/\u0027D\u0027, the\n bit32 of the CPU physical address always is needed to set, and for Region\n \u0027E\u0027, the CPU physical address keep as is. something looks like this:\n CPU PA -\u003e M4U OUTPUT PA\n 0x4000_0000 0x1_4000_0000 (Add bit32)\n 0x8000_0000 0x1_8000_0000 ...\n 0xc000_0000 0x1_c000_0000 ...\n 0x1_0000_0000 0x1_0000_0000 (No change)\n \n Additionally, the iommu consumers always use the CPU phyiscal address.\n \n The PA in the iova_to_phys that is got from v7s always is u32, But\n from the CPU point of view, PA only need add BIT(32) when PA \u003c 0x4000_0000.\n \n Fixes: 30e2fccf9512 (\"iommu/mediatek: Enlarge the validate PA range\n for 4GB mode\")\n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 42d42c76f8d2225aaed55997949d37424906e188\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:01:49 2019 +0800\n\n memory: mtk-smi: Use a struct for the platform data for smi-common\n \n Use a struct as the platform special data instead of the enumeration.\n \n Also there is a minor change that moving the position of\n \"enum mtk_smi_gen\" definition, this is because we expect define\n \"struct mtk_smi_common_plat\" before it is referred.\n \n This is a preparing patch for mt8183.\n \n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Reviewed-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Reviewed-by: Evan Green \u003cevgreen@chromium.org\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 2e9b0908f5aa26336b2a39091db1a10402054128\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:01:48 2019 +0800\n\n memory: mtk-smi: Use a general config_port interface\n \n The config_port of mt2712 and mt8183 are the same. Use a general\n config_port interface instead.\n \n In addition, in mt2712, larb8 and larb9 are the bdpsys larbs which\n are not the normal larb, their register space are different from the\n normal one. thus, we can not call the general config_port. In mt8183,\n IPU0/1 and CCU connect with smi-common directly, they also are not\n the normal larb. Hence, we add a \"larb_direct_to_common_mask\" for these\n larbs which connect to smi-commmon directly.\n \n This is also a preparing patch for adding mt8183 SMI support.\n \n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Reviewed-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Reviewed-by: Evan Green \u003cevgreen@chromium.org\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit cecdce9d7eed33ce24da8dd4a2151b68d5c17865\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:01:47 2019 +0800\n\n iommu/mediatek: Use a struct as the platform data\n \n Use a struct as the platform special data instead of the enumeration.\n This is a prepare patch for adding mt8183 iommu support.\n \n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Reviewed-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Reviewed-by: Evan Green \u003cevgreen@chromium.org\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 29746d012588f1de8517fc6921683c3844120989\nAuthor: Yong Wu \u003cyong.wu@mediatek.com\u003e\nDate: Sat Aug 24 11:01:46 2019 +0800\n\n dt-bindings: mediatek: Add binding for mt8183 IOMMU and SMI\n \n This patch adds decriptions for mt8183 IOMMU and SMI.\n \n mt8183 has only one M4U like mt8173 and is also MTK IOMMU gen2 which\n uses ARM Short-Descriptor translation table format.\n \n The mt8183 M4U-SMI HW diagram is as below:\n \n EMI\n |\n M4U\n |\n ----------\n | |\n gals0-rx gals1-rx\n | |\n | |\n gals0-tx gals1-tx\n | |\n ------------\n SMI Common\n ------------\n |\n +-----+-----+--------+-----+-----+-------+-------+\n | | | | | | | |\n | | gals-rx gals-rx | gals-rx gals-rx gals-rx\n | | | | | | | |\n | | | | | | | |\n | | gals-tx gals-tx | gals-tx gals-tx gals-tx\n | | | | | | | |\n larb0 larb1 IPU0 IPU1 larb4 larb5 larb6 CCU\n disp vdec img cam venc img cam\n \n All the connections are HW fixed, SW can NOT adjust it.\n \n Compared with mt8173, we add a GALS(Global Async Local Sync) module\n between SMI-common and M4U, and additional GALS between larb2/3/5/6\n and SMI-common. GALS can help synchronize for the modules in different\n clock frequency, it can be seen as a \"asynchronous fifo\".\n \n GALS can only help transfer the command/data while it doesn\u0027t have\n the configuring register, thus it has the special \"smi\" clock and it\n doesn\u0027t have the \"apb\" clock. From the diagram above, we add \"gals0\"\n and \"gals1\" clocks for smi-common and add a \"gals\" clock for smi-larb.\n \n \u003eFrom the diagram above, IPU0/IPU1(Image Processor Unit) and CCU(Camera\n Control Unit) is connected with smi-common directly, we can take them\n as \"larb2\", \"larb3\" and \"larb7\", and their register spaces are\n different with the normal larb.\n \n Signed-off-by: Yong Wu \u003cyong.wu@mediatek.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Evan Green \u003cevgreen@chromium.org\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 4dbd258ff63e0597ee8fb44d277c6c701f5019d9\nAuthor: Eric Auger \u003ceric.auger@redhat.com\u003e\nDate: Wed Aug 21 14:09:40 2019 +0200\n\n iommu: Revisit iommu_insert_resv_region() implementation\n \n Current implementation is recursive and in case of allocation\n failure the existing @regions list is altered. A non recursive\n version looks better for maintainability and simplifies the\n error handling. We use a separate stack for overlapping segment\n merging. The elements are sorted by start address and then by\n type, if their start address match.\n \n Note this new implementation may change the region order of\n appearance in /sys/kernel/iommu_groups/\u003cn\u003e/reserved_regions\n files but this order has never been documented, see\n commit bc7d12b91bd3 (\"iommu: Implement reserved_regions\n iommu-group sysfs file\").\n \n Signed-off-by: Eric Auger \u003ceric.auger@redhat.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 2c70010867f164d1b30e787e360e05d10cc40046\nAuthor: Nadav Amit \u003cnamit@vmware.com\u003e\nDate: Tue Aug 20 01:53:17 2019 -0700\n\n iommu/vt-d: Fix wrong analysis whether devices share the same bus\n \n set_msi_sid_cb() is used to determine whether device aliases share the\n same bus, but it can provide false indications that aliases use the same\n bus when in fact they do not. The reason is that set_msi_sid_cb()\n assumes that pdev is fixed, while actually pci_for_each_dma_alias() can\n call fn() when pdev is set to a subordinate device.\n \n As a result, running an VM on ESX with VT-d emulation enabled can\n results in the log warning such as:\n \n DMAR: [INTR-REMAP] Request device [00:11.0] fault index 3b [fault reason 38] Blocked an interrupt request due to source-id verification failure\n \n This seems to cause additional ata errors such as:\n ata3.00: qc timeout (cmd 0xa1)\n ata3.00: failed to IDENTIFY (I/O error, err_mask\u003d0x4)\n \n These timeouts also cause boot to be much longer and other errors.\n \n Fix it by checking comparing the alias with the previous one instead.\n \n Fixes: 3f0c625c6ae71 (\"iommu/vt-d: Allow interrupts from the entire bus for aliased devices\")\n Cc: stable@vger.kernel.org\n Cc: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Cc: David Woodhouse \u003cdwmw2@infradead.org\u003e\n Cc: Joerg Roedel \u003cjoro@8bytes.org\u003e\n Cc: Jacob Pan \u003cjacob.jun.pan@linux.intel.com\u003e\n Signed-off-by: Nadav Amit \u003cnamit@vmware.com\u003e\n Reviewed-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit f12dc871e3ec1269b65080853fc295a831b2fa5f\nAuthor: Nadav Amit \u003cnamit@vmware.com\u003e\nDate: Sun Oct 21 21:45:48 2018 -0700\n\n kgdb: fix comment regarding static function\n \n The comment that says that module_event() is not static is clearly\n wrong.\n \n Signed-off-by: Nadav Amit \u003cnamit@vmware.com\u003e\n\ncommit 3a8e9ac89e6a5106cfb6b85d4c9cf9bfa3519bc7\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Thu Aug 29 15:47:19 2019 -0700\n\n writeback: add tracepoints for cgroup foreign writebacks\n \n cgroup foreign inode handling has quite a bit of heuristics and\n internal states which sometimes makes it difficult to understand\n what\u0027s going on. Add tracepoints to improve visibility.\n \n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit c60d67aad7b7d5c2597113ddf346cf01ad0872c9\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Mon Jul 29 23:13:59 2019 +0800\n\n kdb: Replace strncmp with str_has_prefix\n \n strncmp(str, const, len) is error-prone.\n We had better use newly introduced\n str_has_prefix() instead of it.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n\ncommit 5c51630c7517c0ebdf52fff5a092c1c1b82c47e1\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Thu Aug 29 17:01:34 2019 +0100\n\n selinux: Implement the watch_key security hook\n \n Implement the watch_key security hook to make sure that a key grants the\n caller View permission in order to set a watch on a key.\n \n For the moment, the watch_devices security hook is left unimplemented as\n it\u0027s not obvious what the object should be since the queue is global and\n didn\u0027t previously exist.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Acked-by: Stephen Smalley \u003csds@tycho.nsa.gov\u003e\n\ncommit 0d87308cca2c124f9bce02383f1d9632c9be89c4\nAuthor: Eric Dumazet \u003cedumazet@google.com\u003e\nDate: Wed Aug 28 06:13:38 2019 -0700\n\n iommu/iova: Avoid false sharing on fq_timer_on\n \n In commit 14bd9a607f90 (\"iommu/iova: Separate atomic variables\n to improve performance\") Jinyu Qi identified that the atomic_cmpxchg()\n in queue_iova() was causing a performance loss and moved critical fields\n so that the false sharing would not impact them.\n \n However, avoiding the false sharing in the first place seems easy.\n We should attempt the atomic_cmpxchg() no more than 100 times\n per second. Adding an atomic_read() will keep the cache\n line mostly shared.\n \n This false sharing came with commit 9a005a800ae8\n (\"iommu/iova: Add flush timer\").\n \n Signed-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\n Fixes: 9a005a800ae8 (\u0027iommu/iova: Add flush timer\u0027)\n Cc: Jinyu Qi \u003cjinyuqi@huawei.com\u003e\n Cc: Joerg Roedel \u003cjroedel@suse.de\u003e\n Acked-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit f8f4b6ccd2adc1e3489d5b6c1b22876c4168e4d7\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Fri Jun 7 15:13:21 2019 +0100\n\n Add sample notification program\n \n This needs to be linked with -lkeyutils.\n \n It is run like:\n \n ./watch_test\n \n and watches \"/\" for mount changes and the current session keyring for key\n changes:\n \n # keyctl add user a a @s\n 1035096409\n # keyctl unlink 1035096409 @s\n \n producing:\n \n # ./watch_test\n ptrs h\u003d4 t\u003d2 m\u003d20003\n NOTIFY[00000004-00000002] ty\u003d0003 sy\u003d0002 i\u003d01000010\n KEY 2ffc2e5d change\u003d2[linked] aux\u003d1035096409\n ptrs h\u003d6 t\u003d4 m\u003d20003\n NOTIFY[00000006-00000004] ty\u003d0003 sy\u003d0003 i\u003d01000010\n KEY 2ffc2e5d change\u003d3[unlinked] aux\u003d1035096409\n \n Other events may be produced, such as with a failing disk:\n \n ptrs h\u003d5 t\u003d2 m\u003d6000004\n NOTIFY[00000005-00000002] ty\u003d0004 sy\u003d0006 i\u003d04000018\n BLOCK 00800050 e\u003d6[critical medium] s\u003d5be8\n \n This corresponds to:\n \n print_req_error: critical medium error, dev sdf, sector 23528 flags 0\n \n in dmesg.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n\ncommit cd51316a9ed21238a6db0e8bc287e8a5ae282af8\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Fri Jun 7 15:13:20 2019 +0100\n\n usb: Add USB subsystem notifications\n \n Add a USB subsystem notification mechanism whereby notifications about\n hardware events such as device connection, disconnection, reset and I/O\n errors, can be reported to a monitoring process asynchronously.\n \n Firstly, an event queue needs to be created:\n \n fd \u003d open(\"/dev/event_queue\", O_RDWR);\n ioctl(fd, IOC_WATCH_QUEUE_SET_SIZE, page_size \u003c\u003c n);\n \n then a notification can be set up to report USB notifications via that\n queue:\n \n struct watch_notification_filter filter \u003d {\n .nr_filters \u003d 1,\n .filters \u003d {\n [0] \u003d {\n .type \u003d WATCH_TYPE_USB_NOTIFY,\n .subtype_filter[0] \u003d UINT_MAX;\n },\n },\n };\n ioctl(fd, IOC_WATCH_QUEUE_SET_FILTER, \u0026filter);\n notify_devices(fd, 12);\n \n After that, records will be placed into the queue when events occur on a\n USB device or bus. Records are of the following format:\n \n struct usb_notification {\n struct watch_notification watch;\n __u32 error;\n __u32 reserved;\n __u8 name_len;\n __u8 name[0];\n } *n;\n \n Where:\n \n n-\u003ewatch.type will be WATCH_TYPE_USB_NOTIFY\n \n n-\u003ewatch.subtype will be the type of notification, such as\n NOTIFY_USB_DEVICE_ADD.\n \n n-\u003ewatch.info \u0026 WATCH_INFO_LENGTH will indicate the length of the\n record.\n \n n-\u003ewatch.info \u0026 WATCH_INFO_ID will be the second argument to\n device_notify(), shifted.\n \n n-\u003eerror and n-\u003ereserved are intended to convey information such as\n error codes, but are currently not used\n \n n-\u003ename_len and n-\u003ename convey the USB device name as an\n unterminated string. This may be truncated - it is currently\n limited to a maximum 63 chars.\n \n Note that it is permissible for event records to be of variable length -\n or, at least, the length may be dependent on the subtype.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n cc: linux-usb@vger.kernel.org\n\ncommit 648864496767b97727cb60494af80f705a93d6b2\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Fri Jun 7 15:13:20 2019 +0100\n\n block: Add block layer notifications\n \n Add a block layer notification mechanism whereby notifications about\n block-layer events such as I/O errors, can be reported to a monitoring\n process asynchronously.\n \n Firstly, an event queue needs to be created:\n \n fd \u003d open(\"/dev/event_queue\", O_RDWR);\n ioctl(fd, IOC_WATCH_QUEUE_SET_SIZE, page_size \u003c\u003c n);\n \n then a notification can be set up to report block notifications via that\n queue:\n \n struct watch_notification_filter filter \u003d {\n .nr_filters \u003d 1,\n .filters \u003d {\n [0] \u003d {\n .type \u003d WATCH_TYPE_BLOCK_NOTIFY,\n .subtype_filter[0] \u003d UINT_MAX;\n },\n },\n };\n ioctl(fd, IOC_WATCH_QUEUE_SET_FILTER, \u0026filter);\n watch_devices(fd, 12);\n \n After that, records will be placed into the queue when, for example, errors\n occur on a block device. Records are of the following format:\n \n struct block_notification {\n struct watch_notification watch;\n __u64 dev;\n __u64 sector;\n } *n;\n \n Where:\n \n n-\u003ewatch.type will be WATCH_TYPE_BLOCK_NOTIFY\n \n n-\u003ewatch.subtype will be the type of notification, such as\n NOTIFY_BLOCK_ERROR_CRITICAL_MEDIUM.\n \n n-\u003ewatch.info \u0026 WATCH_INFO_LENGTH will indicate the length of the\n record.\n \n n-\u003ewatch.info \u0026 WATCH_INFO_ID will be the second argument to\n watch_devices(), shifted.\n \n n-\u003edev will be the device numbers munged together.\n \n n-\u003esector will indicate the affected sector (if appropriate for the\n event).\n \n Note that it is permissible for event records to be of variable length -\n or, at least, the length may be dependent on the subtype.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n\ncommit e916ad29d96485e5aa3d3237bfeab1522c713d5e\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Fri Aug 30 06:10:58 2019 -0700\n\n blkcg: add missing NULL check in ioc_cpd_alloc()\n \n ioc_cpd_alloc() forgot to check NULL return from kzalloc(). Add it.\n \n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 4072761b981c630acbc6195391e5a22bae114e39\nAuthor: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\nDate: Fri Aug 30 11:58:49 2019 +0300\n\n drm/i915: Indent GuC/WOPCM documentation sections\n \n Indent GuC/WOPCM documentation correctly to reside under\n \"Memory Management and Command Submission\" section to avoid\n it escaping to the upper level navigation.\n \n Signed-off-by: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Acked-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190830085849.12519-2-joonas.lahtinen@linux.intel.com\n\ncommit b047463c852272ef9956ad3a4c706f78f8b06c17\nAuthor: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\nDate: Fri Aug 30 11:58:48 2019 +0300\n\n drm/i915: Remove link to missing \"Batchbuffer Pools\" documentation\n \n The referenced documentation section has been removed. Remove the\n link to avoid warning when building the documentation.\n \n Signed-off-by: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190830085849.12519-1-joonas.lahtinen@linux.intel.com\n\ncommit 3bbb79856f4d17487b4f007434892bea3895eed2\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Wed Aug 21 18:16:06 2019 +0100\n\n Add a general, global device notification watch list\n \n Create a general, global watch list that can be used for the posting of\n device notification events, for such things as device attachment,\n detachment and errors on sources such as block devices and USB devices.\n This can be enabled with:\n \n CONFIG_DEVICE_NOTIFICATIONS\n \n To add a watch on this list, an event queue must be created and configured:\n \n fd \u003d open(\"/dev/event_queue\", O_RDWR);\n ioctl(fd, IOC_WATCH_QUEUE_SET_SIZE, page_size \u003c\u003c n);\n \n and then a watch can be placed upon it using a system call:\n \n watch_devices(fd, 12, 0);\n \n Unless the application wants to receive all events, it should employ\n appropriate filters. For example, to receive just USB notifications, it\n could do:\n \n struct watch_notification_filter filter \u003d {\n .nr_filters \u003d 1,\n .filters \u003d {\n [0] \u003d {\n .type \u003d WATCH_TYPE_USB_NOTIFY,\n .subtype_filter[0] \u003d UINT_MAX;\n },\n },\n };\n ioctl(fd, IOC_WATCH_QUEUE_SET_FILTER, \u0026filter);\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n\ncommit 67a53081e655d41a77f510377364600e5e9bf89c\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 15 11:25:50 2019 +0530\n\n i2c: iproc: Make bcm_iproc_i2c_quirks constant\n \n Static structure bcm_iproc_i2c_quirks, of type i2c_adapter_quirks, is\n only used when being assigned to constant field quirks of a variable\n having type i2c_adapter. Hence make bcm_iproc_i2c_quirks constant as\n well to prevent it from unintended modification.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Reviewed-by: Ray Jui \u003cray.jui@broadcom.com\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit a22331076dba1453386a95d332e6f7f8917c9159\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Wed Aug 21 18:16:06 2019 +0100\n\n keys: Add a notification facility\n \n Add a key/keyring change notification facility whereby notifications about\n changes in key and keyring content and attributes can be received.\n \n Firstly, an event queue needs to be created:\n \n fd \u003d open(\"/dev/event_queue\", O_RDWR);\n ioctl(fd, IOC_WATCH_QUEUE_SET_SIZE, page_size \u003c\u003c n);\n \n then a notification can be set up to report notifications via that queue:\n \n struct watch_notification_filter filter \u003d {\n .nr_filters \u003d 1,\n .filters \u003d {\n [0] \u003d {\n .type \u003d WATCH_TYPE_KEY_NOTIFY,\n .subtype_filter[0] \u003d UINT_MAX,\n },\n },\n };\n ioctl(fd, IOC_WATCH_QUEUE_SET_FILTER, \u0026filter);\n keyctl_watch_key(KEY_SPEC_SESSION_KEYRING, fd, 0x01);\n \n After that, records will be placed into the queue when events occur in\n which keys are changed in some way. Records are of the following format:\n \n struct key_notification {\n struct watch_notification watch;\n __u32 key_id;\n __u32 aux;\n } *n;\n \n Where:\n \n n-\u003ewatch.type will be WATCH_TYPE_KEY_NOTIFY.\n \n n-\u003ewatch.subtype will indicate the type of event, such as\n NOTIFY_KEY_REVOKED.\n \n n-\u003ewatch.info \u0026 WATCH_INFO_LENGTH will indicate the length of the\n record.\n \n n-\u003ewatch.info \u0026 WATCH_INFO_ID will be the second argument to\n keyctl_watch_key(), shifted.\n \n n-\u003ekey will be the ID of the affected key.\n \n n-\u003eaux will hold subtype-dependent information, such as the key\n being linked into the keyring specified by n-\u003ekey in the case of\n NOTIFY_KEY_LINKED.\n \n Note that it is permissible for event records to be of variable length -\n or, at least, the length may be dependent on the subtype. Note also that\n the queue can be shared between multiple notifications of various types.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n\ncommit 539005ffc6260fcb3bb4171138be5f66a41185a9\nAuthor: Lori Hikichi \u003clori.hikichi@broadcom.com\u003e\nDate: Thu Aug 8 09:07:53 2019 +0530\n\n i2c: iproc: Add full name of devicetree node to adapter name\n \n Add the full name of the devicetree node to the adapter name.\n Without this change, all adapters have the same name making it difficult\n to distinguish between multiple instances.\n The most obvious way to see this is to use the utility i2c_detect.\n e.g. \"i2c-detect -l\"\n \n Before\n i2c-1 i2c Broadcom iProc I2C adapter I2C adapter\n i2c-0 i2c Broadcom iProc I2C adapter I2C adapter\n \n After\n i2c-1 i2c Broadcom iProc (i2c@e0000) I2C adapter\n i2c-0 i2c Broadcom iProc (i2c@b0000) I2C adapter\n \n Now it is easy to figure out which adapter maps to a which DT node.\n \n Signed-off-by: Lori Hikichi \u003clori.hikichi@broadcom.com\u003e\n Signed-off-by: Rayagonda Kokatanur \u003crayagonda.kokatanur@broadcom.com\u003e\n Reviewed-by: Ray Jui \u003cray.jui@broadcom.com\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit ce24d039a79de20baa8bbc6a1fbf939e18f2a23d\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Wed Aug 21 18:16:05 2019 +0100\n\n General notification queue with user mmap()\u0027able ring buffer\n \n Implement a misc device that implements a general notification queue as a\n ring buffer that can be mmap()\u0027d from userspace.\n \n The way this is done is:\n \n (1) An application opens the device and indicates the size of the ring\n buffer that it wants to reserve in pages (this can only be set once):\n \n fd \u003d open(\"/dev/watch_queue\", O_RDWR);\n ioctl(fd, IOC_WATCH_QUEUE_NR_PAGES, nr_of_pages);\n \n (2) The application should then map the pages that the device has\n reserved. Each instance of the device created by open() allocates\n separate pages so that maps of different fds don\u0027t interfere with one\n another. Multiple mmap() calls on the same fd, however, will all work\n together.\n \n page_size \u003d sysconf(_SC_PAGESIZE);\n mapping_size \u003d nr_of_pages * page_size;\n char *buf \u003d mmap(NULL, mapping_size, PROT_READ|PROT_WRITE,\n MAP_SHARED, fd, 0);\n \n The ring is divided into 8-byte slots. Entries written into the ring are\n variable size and can use between 1 and 63 slots. A special entry is\n maintained in the first two slots of the ring that contains the head and\n tail pointers. This is skipped when the ring wraps round. Note that\n multislot entries, therefore, aren\u0027t allowed to be broken over the end of\n the ring, but instead \"skip\" entries are inserted to pad out the buffer.\n \n Each entry has a 1-slot header that describes it:\n \n struct watch_notification {\n __u32 type:24;\n __u32 subtype:8;\n __u32 info;\n };\n \n The type indicates the source (eg. mount tree changes, superblock events,\n keyring changes, block layer events) and the subtype indicates the event\n type (eg. mount, unmount; EIO, EDQUOT; link, unlink). The info field\n indicates a number of things, including the entry length, an ID assigned to\n a watchpoint contributing to this buffer, type-specific flags and meta\n flags, such as an overrun indicator.\n \n Supplementary data, such as the key ID that generated an event, are\n attached in additional slots.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d1d2a2417438cc82c55db234bb2ec33456bcb805\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Wed Aug 21 18:16:05 2019 +0100\n\n security: Add a hook for the point of notification insertion\n \n Add a security hook that allows an LSM to rule on whether a notification\n message is allowed to be inserted into a particular watch queue.\n \n The hook is given the following information:\n \n (1) The credentials of the triggerer (which may be init_cred for a system\n notification, eg. a hardware error).\n \n (2) The credentials of the whoever set the watch.\n \n (3) The notification message.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n cc: Casey Schaufler \u003ccasey@schaufler-ca.com\u003e\n cc: Stephen Smalley \u003csds@tycho.nsa.gov\u003e\n cc: linux-security-module@vger.kernel.org\n\ncommit 48108451ade420704f75cc4ffb57d5f5255b8dd8\nAuthor: Alex Lu \u003calex_lu@realsil.com.cn\u003e\nDate: Fri Aug 30 20:02:14 2019 +0800\n\n Bluetooth: btrtl: Set HCI_QUIRK_SIMULTANEOUS_DISCOVERY\n \n Realtek Bluetooth controllers can do both LE scan and BR/EDR inquiry\n at once, need to set HCI_QUIRK_SIMULTANEOUS_DISCOVERY quirk.\n \n Signed-off-by: Alex Lu \u003calex_lu@realsil.com.cn\u003e\n Signed-off-by: Marcel Holtmann \u003cmarcel@holtmann.org\u003e\n\ncommit 6531deb8a79d21da3a46e01ab02efc293f0596e7\nAuthor: Harish Bandi \u003cc-hbandi@codeaurora.org\u003e\nDate: Fri Aug 30 17:58:36 2019 +0530\n\n Bluetooth: hci_qca: wait for Pre shutdown complete event before sending the Power off pulse\n \n When SoC receives pre shut down command, it share the same\n with other COEX shared clients. So SoC needs a short time\n after sending VS pre shutdown command before turning off\n the regulators and sending the power off pulse. Along with\n short delay, needs to wait for command complete event for\n Pre shutdown VS command\n \n Signed-off-by: Harish Bandi \u003cc-hbandi@codeaurora.org\u003e\n Reviewed-by: Balakrishna Godavarthi \u003cbgodavar@codeaurora.org\u003e\n Signed-off-by: Marcel Holtmann \u003cmarcel@holtmann.org\u003e\n\ncommit e5b3ae3e8c4772106713698075a87bff1e86e710\nAuthor: Tzung-Bi Shih \u003ctzungbi@google.com\u003e\nDate: Fri Aug 30 15:42:40 2019 +0800\n\n ASoC: mediatek: mt8183: fix space issues\n \n Fix space issues:\n - No space before \u0027;\u0027.\n - No trailing blank line.\n \n Signed-off-by: Tzung-Bi Shih \u003ctzungbi@google.com\u003e\n Link: https://lore.kernel.org/r/20190830074240.195166-5-tzungbi@google.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit fc347483b87f029c1bc18a8492623ca21dcd7607\nAuthor: Tzung-Bi Shih \u003ctzungbi@google.com\u003e\nDate: Fri Aug 30 15:42:39 2019 +0800\n\n ASoC: mediatek: mt8183: remove forward declaration of headset_init\n \n The forward declaration of mt8183_mt6358_ts3a227_max98357_headset_init\n is for cyclic dependency between card, headset_dev, and headset_init.\n \n It used to be:\n - card depends on headset_dev\n - headset_dev depends on headset_init\n - headset_init depends on card\n \n Commit a962a809e5e4 (\"ASoC: mediatek: mt8183: make headset codec\n optional\") removed the cyclic dependency.\n \n Thus, it is safe to remove the forward declaration.\n \n Signed-off-by: Tzung-Bi Shih \u003ctzungbi@google.com\u003e\n Link: https://lore.kernel.org/r/20190830074240.195166-4-tzungbi@google.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 95d779b4b8e7a62d53c7afeba45207089dcee4c8\nAuthor: Tzung-Bi Shih \u003ctzungbi@google.com\u003e\nDate: Fri Aug 30 15:42:38 2019 +0800\n\n ASoC: mediatek: mt8183: move headset jack to card-specific storage\n \n Move headset jack descriptor from module global scope to card-specific\n storage to make its ownership more explicit.\n \n Signed-off-by: Tzung-Bi Shih \u003ctzungbi@google.com\u003e\n Link: https://lore.kernel.org/r/20190830074240.195166-3-tzungbi@google.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 96ed76983307a8a24400495e4e6453607e765fcf\nAuthor: Tzung-Bi Shih \u003ctzungbi@google.com\u003e\nDate: Fri Aug 30 15:42:37 2019 +0800\n\n ASoC: mediatek: mt8183: move private structure\n \n Move private structure to the beginning of file to declare earlier\n so that most functions can see it.\n \n Signed-off-by: Tzung-Bi Shih \u003ctzungbi@google.com\u003e\n Link: https://lore.kernel.org/r/20190830074240.195166-2-tzungbi@google.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit ca105398430de8c11e5e56a7bbfb00746ac5a268\nAuthor: Rayagonda Kokatanur \u003crayagonda.kokatanur@broadcom.com\u003e\nDate: Fri Aug 30 09:58:45 2019 +0530\n\n spi: bcm-qspi: Make BSPI default mode\n \n The spi-nor controller defaults to BSPI mode, hence switch back\n to its default mode after MSPI operations (write or erase)\n are completed.\n \n Signed-off-by: Rayagonda Kokatanur \u003crayagonda.kokatanur@broadcom.com\u003e\n Reviewed-by: Mark Brown \u003cbroonie@kernel.org\u003e\n Reviewed-by: Kamal Dasu \u003ckdasu.kdev@gmail.com\u003e\n Link: https://lore.kernel.org/r/1567139325-7912-1-git-send-email-rayagonda.kokatanur@broadcom.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit ace55c411b11d9b12f500c7433bf469c26130182\nAuthor: Tomer Maimon \u003ctmaimon77@gmail.com\u003e\nDate: Wed Aug 28 17:25:13 2019 +0300\n\n spi: npcm-fiu: add NPCM FIU controller driver\n \n Add Nuvoton NPCM BMC Flash Interface Unit(FIU) SPI master\n controller driver using SPI-MEM interface.\n \n The FIU supports single, dual or quad communication interface.\n \n the FIU controller can operate in following modes:\n - User Mode Access(UMA): provides flash access by using an\n indirect address/data mechanism.\n - direct rd/wr mode: maps the flash memory into the core\n address space.\n - SPI-X mode: used for an expansion bus to an ASIC or CPLD.\n \n Signed-off-by: Tomer Maimon \u003ctmaimon77@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190828142513.228556-3-tmaimon77@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 91d0c59f460fd61db977d35b0d7c057128e7f7b7\nAuthor: Tomer Maimon \u003ctmaimon77@gmail.com\u003e\nDate: Wed Aug 28 17:25:12 2019 +0300\n\n dt-binding: spi: add NPCM FIU controller\n \n Added device tree binding documentation for Nuvoton BMC\n NPCM Flash Interface Unit(FIU) SPI master controller\n using SPI-MEM interface.\n \n Signed-off-by: Tomer Maimon \u003ctmaimon77@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190828142513.228556-2-tmaimon77@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 94e9c0f522142e47348486f98916cb0b16e72571\nAuthor: Jarkko Nikula \u003cjarkko.nikula@linux.intel.com\u003e\nDate: Thu Aug 29 15:50:00 2019 +0300\n\n spi: dw-pci: Add MODULE_DEVICE_TABLE\n \n MODULE_DEVICE_TABLE is missing from the PCI part of the driver. Add it\n so userspace can autoload the the driver when it is built as module.\n \n Signed-off-by: Jarkko Nikula \u003cjarkko.nikula@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190829125000.26303-1-jarkko.nikula@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit baedad4c35ce35973263f8991d288c280f4bd05d\nAuthor: Mark Brown \u003cbroonie@kernel.org\u003e\nDate: Thu Aug 29 13:54:35 2019 +0100\n\n MAINTAINERS: Add keyword pattern on regulator_get_optional()\n \n In an effort to try to contain abuses of regulator_get_optional() add a\n keyword entry to the MAINTAINERS stanza for the regulator API so that the\n regulator maintainers get CCed on new usages.\n \n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190829125435.48770-1-broonie@kernel.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit e02e23d5f01937cec2c76a8e61aea0b4e5f53626\nAuthor: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\nDate: Thu Aug 29 06:50:46 2019 +0000\n\n regulator: sy8824x: add prefixes to BUCK_EN and MODE macros\n \n Add prefixes to BUCK_EN and MODE macros to namespace them.\n \n Signed-off-by: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\n Link: https://lore.kernel.org/r/20190829143927.395d0385@xhacker.debian\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit dab106742318c833e1f97d0e6be86b82e33bc5a3\nAuthor: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\nDate: Thu Aug 29 06:49:18 2019 +0000\n\n regulator: sy8824x: use c++style for the comment block near SPDX\n \n Update the entire comment block to be C++ style so it looks consistent.\n \n Signed-off-by: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\n Link: https://lore.kernel.org/r/20190829143749.4b42bc65@xhacker.debian\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 50bc5731f7fc086693d78e42b7d252b97a35cda1\nAuthor: Mark Brown \u003cbroonie@kernel.org\u003e\nDate: Fri Aug 30 12:29:22 2019 +0100\n\n regulator: mt6358: Add BROKEN dependency while waiting for MFD to merge\n \n The mt6358 driver was merged in error, it depends on an existing MFD\n rather than a newly added one and needs updates to that driver. Disable\n the build until those are merged.\n \n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 37638af8135c44e9c9e4b2744eadfd3839efc023\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Thu Aug 29 16:42:13 2019 -0500\n\n ASoC: SOF/Intel: fix selection of SND_INTEL_NHTL\n \n We should only select SND_INTEL_NHLT when ACPI is defined. This was\n done for the legacy HDAudio driver but not for DSP-enabled cases,\n leading to compilation errors with randconfig.\n \n Fix by aligning on the same solution.\n \n For the Skylake driver this is overkill since there is a top-level\n dependency on ACPI, but it doesn\u0027t hurt and it\u0027s better to have\n consistency.\n \n Fixes: 68b953aeb50d9 (\u0027ASoC: SOF: Intel: hda: fixup HDaudio topology name with DMIC number\u0027)\n Reported-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190829214213.11653-1-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 54d037d5a466ce518ee4bb81f39b79ac7d843ba6\nAuthor: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\nDate: Thu Aug 22 13:36:16 2019 +0200\n\n ASoC: Intel: broadwell: Simplify device probe\n \n With legacy ADSP private context adjusted, there is no need for double\n safety.\n \n Signed-off-by: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190822113616.22702-5-cezary.rojewski@intel.com\n Tested-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c25e93bba90b3f194c43a37fe2fcdb0727c4ab84\nAuthor: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\nDate: Thu Aug 22 13:36:15 2019 +0200\n\n ASoC: Intel: bdw-rt5677: Simplify device probe\n \n With legacy ADSP private context adjusted, there is no need for double\n safety.\n \n Signed-off-by: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190822113616.22702-4-cezary.rojewski@intel.com\n Tested-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 1fc3e6b2ac3f8125ae54685c600d594d690a268a\nAuthor: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\nDate: Thu Aug 22 13:36:14 2019 +0200\n\n ASoC: Intel: haswell: Simplify device probe\n \n With legacy ADSP private context adjusted, there is no need for double\n safety.\n \n Signed-off-by: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190822113616.22702-3-cezary.rojewski@intel.com\n Tested-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit ca964edf0ddbfec2cb10b3d251d09598e7ca9b13\nAuthor: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\nDate: Thu Aug 22 13:36:13 2019 +0200\n\n ASoC: Intel: Haswell: Adjust machine device private context\n \n Apart from Haswell machines, all other devices have their private data\n set to snd_soc_acpi_mach instance.\n \n Changes for HSW/ BDW boards introduced with series:\n https://patchwork.kernel.org/cover/10782035/\n \n added support for dai_link platform_name adjustments within card probe\n routines. These take for granted private_data points to\n snd_soc_acpi_mach whereas for Haswell, it\u0027s sst_pdata instead. Change\n private context of platform_device - representing machine board - to\n address this.\n \n Fixes: e87055d732e3 (\"ASoC: Intel: haswell: platform name fixup support\")\n Fixes: 7e40ddcf974a (\"ASoC: Intel: bdw-rt5677: platform name fixup support\")\n Fixes: 2d067b2807f9 (\"ASoC: Intel: broadwell: platform name fixup support\")\n Signed-off-by: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190822113616.22702-2-cezary.rojewski@intel.com\n Tested-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 69f34053f883a53e86fa8ff7b99a81d8b5d50dae\nAuthor: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\nDate: Fri Aug 30 13:38:41 2019 +0300\n\n ASoC: ti: davinci-evm: Do not fail if the dai_set_sysclk returns -ENOTSUPP\n \n The davinci McBSP (davinci-i2s) driver does not implement the set_sysclk\n callback, which is fine and should not be treated as error.\n \n Signed-off-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Link: https://lore.kernel.org/r/20190830103841.25128-5-peter.ujfalusi@ti.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 27b7def93aaba0838eb90e6b3c3110a856bb09fc\nAuthor: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\nDate: Fri Aug 30 13:38:40 2019 +0300\n\n ASoC: ti: edma-pcm: Fix for legacy dma_slave_map based channel lookup\n \n Most of the daVinci devices does not boot with DT. In this case the DMA\n channel is looked up with dma_slave_map and for that the chan_names[]\n must be configured.\n \n Both McASP and ASP/McBSP uses \"tx\" and \"rx\" as channel names, so we can\n just do this when the dev-\u003eof_node is not valid.\n \n Signed-off-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Link: https://lore.kernel.org/r/20190830103841.25128-4-peter.ujfalusi@ti.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 2231b2c63f869528504195f202d9585a4f00f143\nAuthor: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\nDate: Fri Aug 30 13:38:38 2019 +0300\n\n ASoC: ti: davinci-i2s: Add S32_LE as support format\n \n ASP/McBSP can support 8/16/20/24/32 bits word in theory. I have only tested\n S16_LE and S32_LE, the other formats might not work so only extend the\n supported formats with S32_LE for now.\n \n Signed-off-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Link: https://lore.kernel.org/r/20190830103841.25128-2-peter.ujfalusi@ti.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit a2dc6f82fd86fa165222f6062e2478fd122f9f1c\nAuthor: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\nDate: Fri Aug 30 13:38:39 2019 +0300\n\n ASoC: ti: davinci-i2s: Move the XSYNCERR workaround to .prepare callback\n \n Currently the driver uses snd_soc_rtdcom_lookup() in it\u0027s mcbsp_start\n function to try to stop/restart the DMA as the initial XSYNCERR workaround\n need to be done before the DMA is armed.\n \n There are couple of things wrong with this:\n - the driver crashes with NULL pointer dereference as the\n component-\u003edriver-\u003eops is actually NULL\n - the driver should not use snd_soc_rtdcom_lookup() in the first place\n - Fiddling with DMA is never a good thing\n \n Move the workaround handling to .prepare which is called before the DMA is\n armed, so it complies with the requirements.\n \n Reported-by (usage of snd_soc_rtdcom_lookup): Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Signed-off-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Link: https://lore.kernel.org/r/20190830103841.25128-3-peter.ujfalusi@ti.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 3d708895325b78506e8daf00ef31549476e8586a\nAuthor: Qian Cai \u003ccai@lca.pw\u003e\nDate: Wed Aug 28 17:39:43 2019 -0400\n\n iommu/amd: Silence warnings under memory pressure\n \n When running heavy memory pressure workloads, the system is throwing\n endless warnings,\n \n smartpqi 0000:23:00.0: AMD-Vi: IOMMU mapping error in map_sg (io-pages:\n 5 reason: -12)\n Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385 Gen10, BIOS A40\n 07/10/2019\n swapper/10: page allocation failure: order:0, mode:0xa20(GFP_ATOMIC),\n nodemask\u003d(null),cpuset\u003d/,mems_allowed\u003d0,4\n Call Trace:\n \u003cIRQ\u003e\n dump_stack+0x62/0x9a\n warn_alloc.cold.43+0x8a/0x148\n __alloc_pages_nodemask+0x1a5c/0x1bb0\n get_zeroed_page+0x16/0x20\n iommu_map_page+0x477/0x540\n map_sg+0x1ce/0x2f0\n scsi_dma_map+0xc6/0x160\n pqi_raid_submit_scsi_cmd_with_io_request+0x1c3/0x470 [smartpqi]\n do_IRQ+0x81/0x170\n common_interrupt+0xf/0xf\n \u003c/IRQ\u003e\n \n because the allocation could fail from iommu_map_page(), and the volume\n of this call could be huge which may generate a lot of serial console\n output and cosumes all CPUs.\n \n Fix it by silencing the warning in this call site, and there is still a\n dev_err() later to notify the failure.\n \n Signed-off-by: Qian Cai \u003ccai@lca.pw\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit c8cd6e7f159e6f8d79a23df4aeaa7a540415951b\nAuthor: Felix Fietkau \u003cnbd@nbd.name\u003e\nDate: Wed Aug 28 12:20:42 2019 +0200\n\n cfg80211: add local BSS receive time to survey information\n \n This is useful for checking how much airtime is being used up by other\n transmissions on the channel, e.g. by calculating (time_rx - time_bss_rx)\n or (time_busy - time_bss_rx - time_tx)\n \n Signed-off-by: Felix Fietkau \u003cnbd@nbd.name\u003e\n Link: https://lore.kernel.org/r/20190828102042.58016-1-nbd@nbd.name\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 03adcbd996be7ce81cac793b1511406a7a4df117\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Thu Aug 29 14:34:42 2019 +0100\n\n arm64: atomics: Use K constraint when toolchain appears to support it\n \n The \u0027K\u0027 constraint is a documented AArch64 machine constraint supported\n by GCC for matching integer constants that can be used with a 32-bit\n logical instruction. Unfortunately, some released compilers erroneously\n accept the immediate \u00274294967295\u0027 for this constraint, which is later\n refused by GAS at assembly time. This had led us to avoid the use of\n the \u0027K\u0027 constraint altogether.\n \n Instead, detect whether the compiler is up to the job when building the\n kernel and pass the \u0027K\u0027 constraint to our 32-bit atomic macros when it\n appears to be supported.\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 5aad6cdabbf91fd330bd216fe3c93d90f78bc7e7\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Thu Aug 29 14:33:23 2019 +0100\n\n arm64: atomics: Undefine internal macros after use\n \n We use a bunch of internal macros when constructing our atomic and\n cmpxchg routines in order to save on boilerplate. Avoid exposing these\n directly to users of the header files.\n \n Reviewed-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit b32baf91f60fb9c7010bff87e68132f2ce31d9a8\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Thu Aug 29 11:52:47 2019 +0100\n\n arm64: lse: Make ARM64_LSE_ATOMICS depend on JUMP_LABEL\n \n Support for LSE atomic instructions (CONFIG_ARM64_LSE_ATOMICS) relies on\n a static key to select between the legacy LL/SC implementation which is\n available on all arm64 CPUs and the super-duper LSE implementation which\n is available on CPUs implementing v8.1 and later.\n \n Unfortunately, when building a kernel with CONFIG_JUMP_LABEL disabled\n (e.g. because the toolchain doesn\u0027t support \u0027asm goto\u0027), the static key\n inside the atomics code tries to use atomics itself. This results in a\n mess of circular includes and a build failure:\n \n In file included from ./arch/arm64/include/asm/lse.h:11,\n from ./arch/arm64/include/asm/atomic.h:16,\n from ./include/linux/atomic.h:7,\n from ./include/asm-generic/bitops/atomic.h:5,\n from ./arch/arm64/include/asm/bitops.h:26,\n from ./include/linux/bitops.h:19,\n from ./include/linux/kernel.h:12,\n from ./include/asm-generic/bug.h:18,\n from ./arch/arm64/include/asm/bug.h:26,\n from ./include/linux/bug.h:5,\n from ./include/linux/page-flags.h:10,\n from kernel/bounds.c:10:\n ./include/linux/jump_label.h: In function ‘static_key_count’:\n ./include/linux/jump_label.h:254:9: error: implicit declaration of function ‘atomic_read’ [-Werror\u003dimplicit-function-declaration]\n return atomic_read(\u0026key-\u003eenabled);\n ^~~~~~~~~~~\n \n [ ... more of the same ... ]\n \n Since LSE atomic instructions are not critical to the operation of the\n kernel, make them depend on JUMP_LABEL at compile time.\n \n Reviewed-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 0533f97b4356bfa8af5d4758c6c3fe703bb010d9\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Thu Aug 29 11:49:10 2019 +0100\n\n arm64: asm: Kill \u0027asm/atomic_arch.h\u0027\n \n The contents of \u0027asm/atomic_arch.h\u0027 can be split across some of our\n other \u0027asm/\u0027 headers. Remove it.\n \n Reviewed-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 0ca98b2456fbd8a465098fe3735ae2c7645a76e8\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Thu Aug 29 11:22:30 2019 +0100\n\n arm64: lse: Remove unused \u0027alt_lse\u0027 assembly macro\n \n The \u0027alt_lse\u0027 assembly macro has been unused since 7c8fc35dfc32\n (\"locking/atomics/arm64: Replace our atomic/lock bitop implementations\n with asm-generic\").\n \n Remove it.\n \n Reviewed-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 89781d0806c2c4f29072d3f00cb2dd4274aabc3d\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Fri Aug 30 12:03:38 2019 +0200\n\n ALSA: hda/realtek - Fix overridden device-specific initialization\n \n The recent change to shuffle the codec initialization procedure for\n Realtek via commit 607ca3bd220f (\"ALSA: hda/realtek - EAPD turn on\n later\") caused the silent output on some machines. This change was\n supposed to be safe, but it isn\u0027t actually; some devices have quirk\n setups to override the EAPD via COEF or BTL in the additional verb\n table, which is applied at the beginning of snd_hda_gen_init(). And\n this EAPD setup is again overridden in alc_auto_init_amp().\n \n For recovering from the regression, tell snd_hda_gen_init() not to\n apply the verbs there by a new flag, then apply the verbs in\n alc_init().\n \n BugLink: https://bugzilla.kernel.org/show_bug.cgi?id\u003d204727\n Fixes: 607ca3bd220f (\"ALSA: hda/realtek - EAPD turn on later\")\n Cc: \u003cstable@vger.kernel.org\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 04e3543e228fcbb000de922ec5d366f398bdd9ae\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Wed Aug 14 16:03:48 2019 +0200\n\n microblaze: use the generic dma coherent remap allocator\n \n This switches to using common code for the DMA allocations, including\n potential use of the CMA allocator if configured.\n \n Switching to the generic code enables DMA allocations from atomic\n context, which is required by the DMA API documentation, and also\n adds various other minor features drivers start relying upon. It\n also makes sure we have on tested code base for all architectures\n that require uncached pte bits for coherent DMA allocations.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Michal Simek \u003cmichal.simek@xilinx.com\u003e\n\ncommit d3b9f659fac6e5f74f857364c5ed08b94e7f94ae\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Wed Aug 14 16:03:47 2019 +0200\n\n microblaze/nommu: use the generic uncached segment support\n \n Stop providing our own arch alloc/free hooks for nommu platforms and\n just expose the segment offset and use the generic dma-direct\n allocator.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Michal Simek \u003cmichal.simek@xilinx.com\u003e\n\ncommit 692059318c0fc6c3584b861adc67abbb0c1598f0\nAuthor: Ramalingam C \u003cramalingam.c@intel.com\u003e\nDate: Wed Aug 28 22:12:16 2019 +0530\n\n drm/i915/hdcp: Enable HDCP 1.4 and 2.2 on Gen12+\n \n \u003eFrom Gen12 onwards, HDCP HW block is implemented within transcoders.\n Till Gen11 HDCP HW block was part of DDI.\n \n Hence required changes in HW programming is handled here.\n \n As ME FW needs the transcoder detail on which HDCP is enabled\n on Gen12+ platform, we are populating the detail in hdcp_port_data.\n \n v2:\n _MMIO_TRANS is used [Lucas and Daniel]\n platform check is moved into the caller [Lucas]\n v3:\n platform check is moved into a macro [Shashank]\n v4:\n Few optimizations in the coding [Shashank]\n v5:\n Fixed alignment in macro definition in i915_reg.h [Shashank]\n unused variables \"reg\" is removed.\n v6:\n Configuring the transcoder at compute_config.\n transcoder is used instead of pipe in macros.\n Rebased.\n v7:\n transcoder is cached at intel_hdcp\n hdcp_port_data is configured with transcoder index asper ME FW.\n v8:\n s/trans/cpu_transcoder\n s/tc/cpu_transcoder\n v9:\n rep_ctl is prepared for TCD too.\n return moved into deault of rep_ctl prepare function [Shashank]\n \n Signed-off-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Reviewed-by: Shashank Sharma \u003cshashank.sharma@intel.com\u003e\n Acked-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Signed-off-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190828164216.405-7-ramalingam.c@intel.com\n\ncommit 39e2df090c3cdce23a064929333b3be119111765\nAuthor: Ramalingam C \u003cramalingam.c@intel.com\u003e\nDate: Wed Aug 28 22:12:15 2019 +0530\n\n drm/i915/hdcp: update current transcoder into intel_hdcp\n \n On gen12+ platforms, HDCP HW is associated to the transcoder.\n Hence on every modeset update associated transcoder into the\n intel_hdcp of the port.\n \n v2:\n s/trans/cpu_transcoder [Jani]\n v3:\n comment is added for fw_ddi init for gen12+ [Shashank]\n only hdcp capable transcoder is translated into fw_tc [Shashank]\n v4:\n fw_tc initialization is kept for modeset. [Tomas]\n few extra doc is added at port_data init [Tomas]\n v5:\n Few comments are improvised [Tomas]\n \n Signed-off-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Acked-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Reviewed-by: Shashank Sharma \u003cshashank.sharma@intel.com\u003e\n Signed-off-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190828164216.405-6-ramalingam.c@intel.com\n\ncommit 2d15cf1b9ae4c623b1a05838317f4c18fd534c9a\nAuthor: Ramalingam C \u003cramalingam.c@intel.com\u003e\nDate: Wed Aug 28 22:12:14 2019 +0530\n\n misc/mei/hdcp: Fill transcoder index in port info\n \n For gen12+ platform we need to pass the transcoder info\n as part of the port info into ME FW.\n \n This change fills the payload for ME FW from hdcp_port_data.\n \n v2:\n Doc is enhanced for physical_port and attached_transcoder [Tomas]\n \n Signed-off-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Acked-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Reviewed-by: Shashank Sharma \u003cshashank.sharma@intel.com\u003e\n Acked-by: Tomas Winkler \u003ctomas.winkler@intel.com\u003e\n Signed-off-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190828164216.405-5-ramalingam.c@intel.com\n\ncommit 807c71d59a1063badc8bf60ee71470c2ff4be1cc\nAuthor: Ramalingam C \u003cramalingam.c@intel.com\u003e\nDate: Wed Aug 28 22:12:13 2019 +0530\n\n drm: Extend I915 mei interface for transcoder info\n \n I915 needs to send the index of the transcoder as per ME FW.\n \n To support this, define enum mei_fw_tc and add as a member into\n the struct hdcp_port_data.\n \n v2:\n Typo in commit msg is fixed [Shashank]\n v3:\n kdoc is added for mei_fw_tc [Tomas]\n s/MEI_TC_x/MEI_TRANSCODER_x\n \n Signed-off-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Acked-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Acked-by: Tomas Winkler \u003ctomas.winkler@intel.com\u003e\n Reviewed-by: Shashank Sharma \u003cshashank.sharma@intel.com\u003e\n Signed-off-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190828164216.405-4-ramalingam.c@intel.com\n\ncommit 5b6030da28cd6d6f5d129d7039ea0db83bcee6cf\nAuthor: Ramalingam C \u003cramalingam.c@intel.com\u003e\nDate: Wed Aug 28 22:12:12 2019 +0530\n\n drm: Move port definition back to i915 header\n \n We dont need the definition of the enum port outside I915, anymore.\n Hence move enum port definition into I915 driver itself.\n \n v2:\n intel_display.h is included in intel_hdcp.h\n v3:\n enum port is declared in headers.\n v4:\n commit msg is rephrased.\n v5:\n copyright year is updated [Tomas]\n \n Signed-off-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Reviewed-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Reviewed-by: Shashank Sharma \u003cshashank.sharma@intel.com\u003e\n Reviewed-by: Tomas Winkler \u003ctomas.winkler@intel.com\u003e\n Signed-off-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190828164216.405-3-ramalingam.c@intel.com\n\ncommit 0dcceb35a13de07d8c28305b69a33b339c7eba0a\nAuthor: Ramalingam C \u003cramalingam.c@intel.com\u003e\nDate: Wed Aug 28 22:12:11 2019 +0530\n\n drm/i915: mei_hdcp: I915 sends ddi index as per ME FW\n \n I915 converts it\u0027s port value into ddi index defiend by ME FW\n and pass it as a member of hdcp_port_data structure.\n \n Hence expose the enum mei_fw_ddi to I915 through\n i915_mei_interface.h.\n \n v2:\n Copyright years are bumped [Tomas]\n \n Signed-off-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Acked-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Reviewed-by: Shashank Sharma \u003cshashank.sharma@intel.com\u003e\n Acked-by: Tomas Winkler \u003ctomas.winkler@intel.com\u003e\n Signed-off-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190828164216.405-2-ramalingam.c@intel.com\n\ncommit 424c38a4e32509ae82dc9d7300432295806cb911\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Tue Aug 27 13:58:50 2019 +0200\n\n drm/selftests: modes: Add more unit tests for the cmdline parser\n \n Let\u0027s add some unit tests for the recent bugs we just fixed.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Tested-by: Thomas Graichen \u003cthomas.graichen@gmail.com\u003e\n Reviewed-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Signed-off-by: Maxime Ripard \u003cmripard@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190827115850.25731-4-mripard@kernel.org\n\ncommit 3764137906a5acece8b5546873a70b1e6263a1a6\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Tue Aug 27 13:58:49 2019 +0200\n\n drm/modes: Introduce a whitelist for the named modes\n \n The named modes support has introduced a number of glitches that were in\n part due to the fact that the parser will take any string as a named mode.\n \n Since we shouldn\u0027t have a lot of options there (and they should be pretty\n standard), let\u0027s introduce a whitelist of the available named modes so that\n the kernel can differentiate between a poorly formed command line and a\n named mode.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Tested-by: Thomas Graichen \u003cthomas.graichen@gmail.com\u003e\n Reviewed-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Signed-off-by: Maxime Ripard \u003cmripard@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190827115850.25731-3-mripard@kernel.org\n\ncommit 728a257f652aee5dd03ecde90b11f414a255e08b\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Tue Aug 27 13:58:48 2019 +0200\n\n drm/modes: Fix the command line parser to take force options into account\n \n The command line parser when it has been rewritten introduced a regression\n when the only thing on the command line is an option to force the detection\n of a connector (such as video\u003dHDMI-A-1:d), which are completely valid.\n \n It\u0027s been further broken by the support for the named modes which take\n anything that is not a resolution as a named mode.\n \n Let\u0027s fix this by running the extra command line option parser on the named\n modes if they only take a single character.\n \n Fixes: e08ab74bd4c7 (\"drm/modes: Rewrite the command line parser\")\n Reported-by: Jernej Škrabec \u003cjernej.skrabec@gmail.com\u003e\n Reported-by: Thomas Graichen \u003cthomas.graichen@googlemail.com\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Tested-by: Thomas Graichen \u003cthomas.graichen@gmail.com\u003e\n Reviewed-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Signed-off-by: Maxime Ripard \u003cmripard@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190827115850.25731-2-mripard@kernel.org\n\ncommit 325d0ab3a1d1ad8b8c4f9f03fd866e55b0e24254\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Tue Aug 27 13:58:47 2019 +0200\n\n drm/modes: Add a switch to differentiate free standing options\n \n Some extra command line options can be either specified without anything\n else on the command line (basically all the force connection options), but\n some other are only relevant when matched with a resolution (margin and\n interlace).\n \n Let\u0027s add a switch to restrict if needed the available option set.\n \n Fixes: e08ab74bd4c7 (\"drm/modes: Rewrite the command line parser\")\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Tested-by: Thomas Graichen \u003cthomas.graichen@gmail.com\u003e\n Reviewed-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Signed-off-by: Maxime Ripard \u003cmripard@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190827115850.25731-1-mripard@kernel.org\n\ncommit 1bbbbcfdc0f0fa7a98ba0d551fd03d2c45d5a318\nAuthor: Mao Wenan \u003cmaowenan@huawei.com\u003e\nDate: Wed Aug 28 16:07:40 2019 +0800\n\n crypto: hisilicon - select CRYPTO_LIB_DES while compiling SEC driver\n \n When CRYPTO_DEV_HISI_SEC\u003dy, below compilation error is found after\n \u0027commit 894b68d8be4b (\"crypto: hisilicon/des - switch to new verification routines\")\u0027:\n \n drivers/crypto/hisilicon/sec/sec_algs.o: In function `sec_alg_skcipher_setkey_des_cbc\u0027:\n sec_algs.c:(.text+0x11f0): undefined reference to `des_expand_key\u0027\n drivers/crypto/hisilicon/sec/sec_algs.o: In function `sec_alg_skcipher_setkey_des_ecb\u0027:\n sec_algs.c:(.text+0x1390): undefined reference to `des_expand_key\u0027\n make: *** [vmlinux] Error 1\n \n This because DES library has been moved to lib/crypto in this commit\n \u002704007b0e6cbb (\"crypto: des - split off DES library from generic DES cipher driver\")\u0027.\n Fix this by selecting CRYPTO_LIB_DES in CRYPTO_DEV_HISI_SEC.\n \n Fixes: 04007b0e6cbb (\"crypto: des - split off DES library from generic DES cipher driver\")\n Fixes: 894b68d8be4b (\"crypto: hisilicon/des - switch to new verification routines\")\n \n Signed-off-by: Mao Wenan \u003cmaowenan@huawei.com\u003e\n Acked-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 9ecf5ad522e09d6e11a7e0a0b1845622a480f478\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Sun Aug 25 20:18:41 2019 +0200\n\n crypto: sha256 - Add missing MODULE_LICENSE() to lib/crypto/sha256.c\n \n lib/crypto/sha256.c / lib/crypto/libsha256.o may end up being a module,\n so it needs a MODULE_LICENSE() line, add this.\n \n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit c03a509304954c5ed58ac9c607e20f1b55f88a28\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 22 18:47:31 2019 +0300\n\n crypto: ccp - invoke fallback for XTS ciphertext stealing\n \n For correctness and compliance with the XTS-AES specification, we are\n adding support for ciphertext stealing to XTS implementations, even\n though no use cases are known that will be enabled by this.\n \n Since the ccp driver already has a fallback skcipher standby for\n dealing with input sizes other than [16, 512, 1024, 2048, 4096],\n just drop the check against the block size.\n \n Cc: Tom Lendacky \u003cthomas.lendacky@amd.com\u003e\n Cc: Gary Hook \u003cgary.hook@amd.com\u003e\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 2be7f901610ca0d8588865b13ca9046ec32fb9bb\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 22 22:46:49 2019 +0800\n\n crypto: nx - remove unused variables \u0027nx_driver_string\u0027 and \u0027nx_driver_version\u0027\n \n drivers/crypto/nx/nx.h:12:19: warning:\n nx_driver_string defined but not used [-Wunused-const-variable\u003d]\n drivers/crypto/nx/nx.h:13:19: warning:\n nx_driver_version defined but not used [-Wunused-const-variable\u003d]\n \n They are never used, so just remove it.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 25e9960c370b23d6627a3424d0d9820e72152096\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 22 22:44:44 2019 +0800\n\n crypto: atmel - Fix -Wunused-const-variable warning\n \n drivers/crypto/atmel-i2c.h:68:3: warning:\n error_list defined but not used [-Wunused-const-variable\u003d]\n \n error_list is only used in atmel-i2c.c,\n so just move the definition over there.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit f1d087b912c61f23685fcfc1bf61dfce517d89f3\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 22 22:41:38 2019 +0800\n\n crypto: aegis128 - Fix -Wunused-const-variable warning\n \n crypto/aegis.h:27:32: warning:\n crypto_aegis_const defined but not used [-Wunused-const-variable\u003d]\n \n crypto_aegis_const is only used in aegis128-core.c,\n just move the definition over there.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 6796c02a4b7ef3761df040db4905227ddf8dac77\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Aug 20 13:24:01 2019 -0700\n\n crypto: caam - add clock entry for i.MX8MQ\n \n Add clock entry needed to support i.MX8MQ.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Chris Spencer \u003cchristopher.spencer@sea.co.uk\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Cc: Aymen Sghaier \u003caymen.sghaier@nxp.com\u003e\n Cc: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Cc: linux-crypto@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit a6727055971b4fa8d07db08088ff618c57f7d296\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Aug 20 13:24:00 2019 -0700\n\n crypto: caam - always select job ring via RSR on i.MX8MQ\n \n Per feedback from NXP tech support the way to use register based\n service interface on i.MX8MQ is to follow the same set of steps\n outlined for the case when virtualization is enabled, regardless if it\n is. Current version of SRM for i.MX8MQ speaks of DECO DID_MS and DECO\n DID_LS registers, but apparently those are not implemented, so the\n case when SCFGR[VIRT_EN]\u003d0 should be handled the same as the case when\n SCFGR[VIRT_EN]\u003d1\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Chris Spencer \u003cchristopher.spencer@sea.co.uk\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Cc: Aymen Sghaier \u003caymen.sghaier@nxp.com\u003e\n Cc: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Cc: linux-crypto@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit a1cf573ee95d5a15bdd1d33310d179d92b229dd1\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Aug 20 13:23:59 2019 -0700\n\n crypto: caam - select DMA address size at runtime\n \n i.MX8 mScale SoC still use 32-bit addresses in its CAAM implmentation,\n so we can\u0027t rely on sizeof(dma_addr_t) to detemine CAAM pointer\n size. Convert the code to query CTPR and MCFGR for that during driver\n probing.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Chris Spencer \u003cchristopher.spencer@sea.co.uk\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Cc: Aymen Sghaier \u003caymen.sghaier@nxp.com\u003e\n Cc: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Cc: linux-crypto@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit dff36801a9f0cca4f8b074b4c8229dcc0400acd1\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Aug 20 13:23:58 2019 -0700\n\n crypto: caam - don\u0027t hardcode inpentry size\n \n Using dma_addr_t for elements of JobR input ring is not appropriate on\n all 64-bit SoCs, some of which, like i.MX8MQ, use only 32-bit wide\n pointers there. Convert all of the code to use explicit helper\n function that can be later extended to support i.MX8MQ. No functional\n change intended.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Chris Spencer \u003cchristopher.spencer@sea.co.uk\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Cc: Aymen Sghaier \u003caymen.sghaier@nxp.com\u003e\n Cc: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Cc: linux-crypto@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 6c5f898f12a3d65f04a928ffd7e38ae397c9a0d4\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Aug 20 13:23:57 2019 -0700\n\n crypto: caam - drop explicit usage of struct jr_outentry\n \n Using struct jr_outentry to specify the layout of JobR output ring is\n not appropriate for all 64-bit SoC, since some of them, like i.MX8MQ,\n use 32-bit pointers there which doesn\u0027t match 64-bit\n dma_addr_t. Convert existing code to use explicit helper functions to\n access any of the JobR output ring elements, so that the support for\n i.MX8MQ can be added later. No functional change intended.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Chris Spencer \u003cchristopher.spencer@sea.co.uk\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Cc: Aymen Sghaier \u003caymen.sghaier@nxp.com\u003e\n Cc: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Cc: linux-crypto@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit e27d96298bb893aacfeced1045c757b51abf301b\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Aug 20 13:23:56 2019 -0700\n\n crypto: caam - move cpu_to_caam_dma() selection to runtime\n \n Instead of selecting the implementation of\n cpu_to_caam_dma()/caam_dma_to_cpu() at build time using the\n preprocessor, convert the code to do that at run-time using IS_ENABLED\n macro. This is needed to add support for i.MX8MQ. No functional change\n intended.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Chris Spencer \u003cchristopher.spencer@sea.co.uk\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Cc: Aymen Sghaier \u003caymen.sghaier@nxp.com\u003e\n Cc: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Cc: linux-crypto@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 1a3daadce955530df92b1bb22093618dd26a1717\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Aug 20 13:23:55 2019 -0700\n\n crypto: caam - make CAAM_PTR_SZ dynamic\n \n In order to be able to configure CAAM pointer size at run-time, which\n needed to support i.MX8MQ, which is 64-bit SoC with 32-bit pointer\n size, convert CAAM_PTR_SZ to refer to a global variable of the same\n name (\"caam_ptr_sz\") and adjust the rest of the code accordingly. No\n functional change intended.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Chris Spencer \u003cchristopher.spencer@sea.co.uk\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Cc: Aymen Sghaier \u003caymen.sghaier@nxp.com\u003e\n Cc: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Cc: linux-crypto@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 3a0944c5362654e13cde558ef81309a0fcf9112b\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Aug 20 13:23:54 2019 -0700\n\n crypto: caam - share definition for MAX_SDLEN\n \n Both qi.h and cammalg_qi2.h seem to define identical versions of\n MAX_SDLEN. Move it to desc_constr.h to avoid duplication.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Chris Spencer \u003cchristopher.spencer@sea.co.uk\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Cc: Aymen Sghaier \u003caymen.sghaier@nxp.com\u003e\n Cc: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Cc: linux-crypto@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 6e05542fd6061d3862c85fd049cb953f1cf5b264\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Aug 20 13:23:53 2019 -0700\n\n crypto: caam - drop 64-bit only wr/rd_reg64()\n \n Since 32-bit of both wr_reg64 and rd_reg64 now use 64-bit IO helpers,\n these functions should no longer be necessary. No functional change intended.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Cc: Chris Spencer \u003cchristopher.spencer@sea.co.uk\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Cc: Aymen Sghaier \u003caymen.sghaier@nxp.com\u003e\n Cc: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Cc: linux-crypto@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 393d2d0fe8e738521a73486e345531a715ad58b7\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Aug 20 13:23:52 2019 -0700\n\n crypto: caam - use ioread64*_hi_lo in rd_reg64\n \n Following the same transformation logic as outlined in previous commit\n converting wr_reg64, convert rd_reg64 to use helpers from\n \u003clinux/io-64-nonatomic-hi-lo.h\u003e first. No functional change intended.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Cc: Chris Spencer \u003cchristopher.spencer@sea.co.uk\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Cc: Aymen Sghaier \u003caymen.sghaier@nxp.com\u003e\n Cc: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Cc: linux-crypto@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 9f5db8b5ea275fe22613a08bc6e3d9dedcbbf40f\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Aug 20 13:23:51 2019 -0700\n\n crytpo: caam - make use of iowrite64*_hi_lo in wr_reg64\n \n In order to be able to unify 64 and 32 bit implementations of\n wr_reg64, let\u0027s convert it to use helpers from\n \u003clinux/io-64-nonatomic-hi-lo.h\u003e first. Here are the steps of the\n transformation:\n \n 1. Inline wr_reg32 helpers:\n \n if (!caam_imx \u0026\u0026 caam_little_end) {\n if (caam_little_end) {\n iowrite32(data \u003e\u003e 32, (u32 __iomem *)(reg) + 1);\n iowrite32(data, (u32 __iomem *)(reg));\n } else {\n iowrite32be(data \u003e\u003e 32, (u32 __iomem *)(reg) + 1);\n iowrite32be(data, (u32 __iomem *)(reg));\n }\n } else {\n if (caam_little_end) {\n iowrite32(data \u003e\u003e 32, (u32 __iomem *)(reg));\n iowrite32(data, (u32 __iomem *)(reg) + 1);\n } else {\n iowrite32be(data \u003e\u003e 32, (u32 __iomem *)(reg));\n iowrite32be(data, (u32 __iomem *)(reg) + 1);\n }\n }\n \n 2. Transfrom the conditionals such that the check for\n \u0027caam_little_end\u0027 is at the top level:\n \n if (caam_little_end) {\n if (!caam_imx) {\n iowrite32(data \u003e\u003e 32, (u32 __iomem *)(reg) + 1);\n iowrite32(data, (u32 __iomem *)(reg));\n } else {\n iowrite32(data \u003e\u003e 32, (u32 __iomem *)(reg));\n iowrite32(data, (u32 __iomem *)(reg) + 1);\n }\n } else {\n iowrite32be(data \u003e\u003e 32, (u32 __iomem *)(reg));\n iowrite32be(data, (u32 __iomem *)(reg) + 1);\n }\n \n 3. Invert the check for !caam_imx:\n \n if (caam_little_end) {\n if (caam_imx) {\n iowrite32(data \u003e\u003e 32, (u32 __iomem *)(reg));\n iowrite32(data, (u32 __iomem *)(reg) + 1);\n } else {\n iowrite32(data \u003e\u003e 32, (u32 __iomem *)(reg) + 1);\n iowrite32(data, (u32 __iomem *)(reg));\n }\n } else {\n iowrite32be(data \u003e\u003e 32, (u32 __iomem *)(reg));\n iowrite32be(data, (u32 __iomem *)(reg) + 1);\n }\n \n 4. Make use of iowrite64* helpers from \u003clinux/io-64-nonatomic-hi-lo.h\u003e\n \n if (caam_little_end) {\n if (caam_imx) {\n iowrite32(data \u003e\u003e 32, (u32 __iomem *)(reg));\n iowrite32(data, (u32 __iomem *)(reg) + 1);\n } else {\n iowrite64(data, reg);\n }\n } else {\n iowrite64be(data, reg);\n }\n \n No functional change intended.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Cc: Chris Spencer \u003cchristopher.spencer@sea.co.uk\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Cc: Aymen Sghaier \u003caymen.sghaier@nxp.com\u003e\n Cc: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Cc: linux-crypto@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit d488dfd9b210e6d393627b418b35c2e8e851b9d6\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Aug 20 13:23:50 2019 -0700\n\n crypto: caam - request JR IRQ as the last step\n \n In order to avoid any risk of JR IRQ request being handled while some\n of the resources used for that are not yet allocated move the code\n requesting said IRQ to the endo of caam_jr_init().\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Chris Spencer \u003cchristopher.spencer@sea.co.uk\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Cc: Aymen Sghaier \u003caymen.sghaier@nxp.com\u003e\n Cc: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Cc: linux-crypto@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit a6c4194ead005e83ca49226b9ac5fdcba7ff0a04\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Aug 20 13:23:49 2019 -0700\n\n crypto: caam - convert caam_jr_init() to use devres\n \n Use devres to allocate all of the resources in caam_jr_init() (DMA\n coherent and regular memory, IRQs) drop calls to corresponding\n deallocation routines. No functional change intended.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Cc: Aymen Sghaier \u003caymen.sghaier@nxp.com\u003e\n Cc: linux-crypto@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 51e002e949d45b1d645681c4c2b8612174537221\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Aug 20 13:23:48 2019 -0700\n\n crypto: caam - simplfy clock initialization\n \n Simplify clock initialization code by converting it to use clk-bulk,\n devres and soc_device_match() match table. No functional change\n intended.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Tested-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Cc: Chris Spencer \u003cchristopher.spencer@sea.co.uk\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Cc: Aymen Sghaier \u003caymen.sghaier@nxp.com\u003e\n Cc: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Cc: linux-crypto@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 70c0cda27a10df44e81887bc4cb10f587236a941\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Aug 20 13:23:47 2019 -0700\n\n crypto: caam - move DMA mask selection into a function\n \n Exactly the same code to figure out DMA mask is repeated twice in the\n driver code. To avoid repetition, move that logic into a standalone\n subroutine in intern.h. While at it re-shuffle the code to make it\n more readable with early returns.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Cc: Chris Spencer \u003cchristopher.spencer@sea.co.uk\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Cc: Aymen Sghaier \u003caymen.sghaier@nxp.com\u003e\n Cc: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Cc: linux-crypto@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 671e50384ee6267052c5d613dc84a72b7fe9ef4b\nAuthor: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\nDate: Tue Aug 20 14:26:39 2019 +0300\n\n crypto: caam/qi - use print_hex_dump_debug function to print debug messages\n \n Use print_hex_dump_debug function to print debug messages, instead of\n print_hex_dump inside #ifdef DEBUG.\n \n Fixes: 6e005503199b (\"crypto: caam - print debug messages at debug level\")\n Signed-off-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit f6cc45c854da0c964d1541712bb3326c72020a1c\nAuthor: Pascal van Leeuwen \u003cpvanleeuwen@insidesecure.com\u003e\nDate: Mon Aug 19 16:40:26 2019 +0200\n\n crypto: inside-secure - add support for using the EIP197 without vendor firmware\n \n Until now, the inside-secure driver required a set of firmware images\n supplied by the silicon vendor, typically under NDA, to be present in\n /lib/firmware/inside-secure in order to be able to function.\n This patch removes the dependence on this official vendor firmware by\n falling back to generic \"mini\" FW - developed specifically for this\n driver - that can be provided under GPL 2.0 through linux-firmwares.\n \n Signed-off-by: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Acked-by: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 625f269a5a7a3643771320387e474bd0a61d9654\nAuthor: Pascal van Leeuwen \u003cpvanleeuwen@insidesecure.com\u003e\nDate: Mon Aug 19 16:40:25 2019 +0200\n\n crypto: inside-secure - add support for PCI based FPGA development board\n \n This patch adds support for a PCIE development board with FPGA from Xilinx,\n to facilitate pre-silicon driver development by both Inside Secure and its\n IP customers. Since Inside Secure neither produces nor has access to actual\n silicon, this is required functionality to allow us to contribute.\n \n Signed-off-by: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Acked-by: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 70e39e2259a716cce1fecd9974942ee1ed58a9a5\nAuthor: Pascal van Leeuwen \u003cpascalvanl@gmail.com\u003e\nDate: Mon Aug 19 16:40:24 2019 +0200\n\n crypto: inside-secure - Remove redundant algo to engine mapping code\n \n This removes some code determine which engine has which algorithms which\n was effectively redundant (may have been forward-looking?) due to always\n enabling all algorithms for all currently supported engines.\n A future patch will use a different, more scalable approach to achieve\n this. This is removed now because otherwise the next patch will add new\n hardware which would otherwise have to be added to all algorithms, so\n now is a convenient time to just get rid of this.\n \n Signed-off-by: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Acked-by: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 0f6e5c8234778f7dadf5789ce85ac1e062ebed9f\nAuthor: Pascal van Leeuwen \u003cpvanleeuwen@insidesecure.com\u003e\nDate: Mon Aug 19 16:40:23 2019 +0200\n\n crypto: inside-secure - make driver selectable for non-Marvell hardware\n \n While being a generic EIP97/EIP197 driver, the driver was only selectable\n for Marvell Armada hardware. This fix makes the driver selectable for any\n Device Tree supporting kernel configuration, allowing it to be used for\n other compatible hardware by just adding the correct device tree entry.\n \n It also allows the driver to be selected for PCI(E) supporting kernel con-\n figurations, to be able to use it with PCIE based FPGA development boards\n for pre-silicon driver development by both Inside Secure and its IP custo-\n mers.\n \n Signed-off-by: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Acked-by: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit c4624488665fff8edddb00ed0fd00ea4c985d591\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Mon Aug 19 17:22:26 2019 +0300\n\n crypto: s5p - use correct block size of 1 for ctr(aes)\n \n Align the s5p ctr(aes) implementation with other implementations\n of the same mode, by setting the block size to 1.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Acked-by: Kamil Konieczny \u003ck.konieczny@partner.samsung.com\u003e\n Reviewed-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 84a0b00aa486c0ab0216771c5cb988e2277b7498\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Mon Aug 19 17:22:25 2019 +0300\n\n crypto: s5p - deal gracefully with bogus input sizes\n \n The s5p skcipher driver returns -EINVAL for zero length inputs, which\n deviates from the behavior of the generic ECB template, and causes fuzz\n tests to fail. In cases where the input is not a multiple of the AES\n block size (and the chaining mode is not CTR), it prints an error to\n the kernel log, which is a thing we usually try to avoid in response\n to situations that can be triggered by unprivileged users.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Acked-by: Kamil Konieczny \u003ck.konieczny@partner.samsung.com\u003e\n Reviewed-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 735177ca148af5049e9e98b38eaba0c769182655\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Mon Aug 19 17:17:36 2019 +0300\n\n crypto: arm64/aes - implement accelerated ESSIV/CBC mode\n \n Add an accelerated version of the \u0027essiv(cbc(aes),sha256)\u0027 skcipher,\n which is used by fscrypt or dm-crypt on systems where CBC mode is\n signficantly more performant than XTS mode (e.g., when using a h/w\n accelerator which supports the former but not the latter) This avoids\n a separate call into the AES cipher for every invocation.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 65d0042b529048b80dcb1b6b4a738a8f71828146\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Mon Aug 19 17:17:35 2019 +0300\n\n crypto: arm64/aes-cts-cbc - factor out CBC en/decryption of a walk\n \n The plain CBC driver and the CTS one share some code that iterates over\n a scatterwalk and invokes the CBC asm code to do the processing. The\n upcoming ESSIV/CBC mode will clone that pattern for the third time, so\n let\u0027s factor it out first.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit f975abb23c6f3e551b5b9ac991941756bb1c2f3d\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Mon Aug 19 17:17:34 2019 +0300\n\n crypto: essiv - add tests for essiv in cbc(aes)+sha256 mode\n \n Add a test vector for the ESSIV mode that is the most widely used,\n i.e., using cbc(aes) and sha256, in both skcipher and AEAD modes\n (the latter is used by tcrypt to encapsulate the authenc template\n or h/w instantiations of the same)\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 389139b34f407da7c09bc26c4d943f52742a6d42\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Mon Aug 19 17:15:00 2019 +0300\n\n crypto: arm64/aegis128 - use explicit vector load for permute vectors\n \n When building the new aegis128 NEON code in big endian mode, Clang\n complains about the const uint8x16_t permute vectors in the following\n way:\n \n crypto/aegis128-neon-inner.c:58:40: warning: vector initializers are not\n compatible with NEON intrinsics in big endian mode\n [-Wnonportable-vector-initialization]\n static const uint8x16_t shift_rows \u003d {\n ^\n crypto/aegis128-neon-inner.c:58:40: note: consider using vld1q_u8() to\n initialize a vector from memory, or vcombine_u8(vcreate_u8(), vcreate_u8())\n to initialize from integer constants\n \n Since the same issue applies to the uint8x16x4_t loads of the AES Sbox,\n update those references as well. However, since GCC does not implement\n the vld1q_u8_x4() intrinsic, switch from IS_ENABLED() to a preprocessor\n conditional to conditionally include this code.\n \n Reported-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Tested-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 309b77e0f8bfa3126ff12949173e6d45801a968b\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Mon Aug 19 07:18:33 2019 +0200\n\n crypto: picoxcell - Fix the name of the module in the description of CRYPTO_DEV_PICOXCELL\n \n The help section says that the module will be called \u0027pipcoxcell_crypto\u0027.\n This is likely a typo.\n Use \u0027picoxcell_crypto\u0027 instead\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 15f6eb5351a2487b47c2c53e8e0569aaf1e3d56e\nAuthor: Ondrej Jirman \u003cmegous@megous.com\u003e\nDate: Fri Aug 23 12:31:39 2019 +0200\n\n arm64: dts: allwinner: orange-pi-3: Enable UART1 / Bluetooth\n \n The board contains AP6256 WiFi/BT module that has its bluetooth part\n connected to SoC\u0027s UART1 port. Enable this port, and add node for the\n bluetooth device.\n \n Bluetooth part is named bcm4345c5.\n \n You\u0027ll need a BCM4345C5.hcd firmware file that can be found in the\n Xulongs\u0027s repository for H6:\n \n https://github.com/orangepi-xunlong/OrangePiH6_external/tree/master/ap6256\n \n The driver expects the firmware at the following path relative to the\n firmware directory:\n \n brcm/BCM4345C5.hcd\n \n Signed-off-by: Ondrej Jirman \u003cmegous@megous.com\u003e\n Signed-off-by: Marcel Holtmann \u003cmarcel@holtmann.org\u003e\n\ncommit 067fe418f5e1feef75b6312856e5d5267179de73\nAuthor: Ondrej Jirman \u003cmegous@megous.com\u003e\nDate: Fri Aug 23 12:31:38 2019 +0200\n\n arm64: dts: allwinner: h6: Add pin configs for uart1\n \n Orange Pi 3 uses UART1 for bluetooth. Add pinconfigs so that we can use\n them.\n \n Signed-off-by: Ondrej Jirman \u003cmegous@megous.com\u003e\n Signed-off-by: Marcel Holtmann \u003cmarcel@holtmann.org\u003e\n\ncommit 3105c48ee8b7306f12c6099286b583f359d5f5a3\nAuthor: Ondrej Jirman \u003cmegous@megous.com\u003e\nDate: Fri Aug 23 12:31:37 2019 +0200\n\n bluetooth: hci_bcm: Give more time to come out of reset\n \n Some supported devices need more time to come out of reset (eg.\n BCM4345C5 in AP6256).\n \n I don\u0027t have/found a datasheet, so the value was arrive at\n experimentally with the Oprange Pi 3 board. Without increased delay,\n I got intermittent failures during probe. This is a Bluetooth 5.0\n device, so maybe that\u0027s why it takes longer to initialize than the\n others.\n \n Signed-off-by: Ondrej Jirman \u003cmegous@megous.com\u003e\n Signed-off-by: Marcel Holtmann \u003cmarcel@holtmann.org\u003e\n\ncommit 8552444514531555620a920166027764e44d58d5\nAuthor: Ondrej Jirman \u003cmegous@megous.com\u003e\nDate: Fri Aug 23 12:31:36 2019 +0200\n\n bluetooth: bcm: Add support for loading firmware for BCM4345C5\n \n Detect BCM4345C5 and load a corresponding firmware file.\n \n Signed-off-by: Ondrej Jirman \u003cmegous@megous.com\u003e\n Signed-off-by: Marcel Holtmann \u003cmarcel@holtmann.org\u003e\n\ncommit 0af6f0a4748e72b9b59313b104466230cd25d9bd\nAuthor: Ondrej Jirman \u003cmegous@megous.com\u003e\nDate: Fri Aug 23 12:31:35 2019 +0200\n\n dt-bindings: net: Add compatible for BCM4345C5 bluetooth device\n \n This is present in the AP6526 WiFi/Bluetooth 5.0 module.\n \n Signed-off-by: Ondrej Jirman \u003cmegous@megous.com\u003e\n Acked-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Marcel Holtmann \u003cmarcel@holtmann.org\u003e\n\ncommit 048ea5526658012b88b024e840e60eb2cca1d602\nAuthor: Rocky Liao \u003crjliao@codeaurora.org\u003e\nDate: Wed Aug 21 14:23:39 2019 +0800\n\n Bluetooth: hci_qca: Set HCI_QUIRK_SIMULTANEOUS_DISCOVERY for QCA UART Radio\n \n QCA UART Bluetooth controllers can do both LE scan and BR/EDR inquiry\n at once, need to set HCI_QUIRK_SIMULTANEOUS_DISCOVERY quirk.\n \n Signed-off-by: Rocky Liao \u003crjliao@codeaurora.org\u003e\n Signed-off-by: Marcel Holtmann \u003cmarcel@holtmann.org\u003e\n\ncommit f52b5d058b317613037c8d2e2d9e0fa7350dae2d\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 15 11:21:49 2019 +0530\n\n Bluetooth: hci_qca: Make structure qca_proto constant\n \n Static structure qca_proto, of type hci_uart_proto, is used four times:\n as the last argument in function hci_uart_register_device(), and as the\n only argument to functions hci_uart_register_proto() and\n hci_uart_unregister_proto(). In all three of these functions, the\n parameter corresponding to qca_proto is declared as constant. Therefore,\n make qca_proto itself constant as well in order to protect it from\n unintended modification.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Marcel Holtmann \u003cmarcel@holtmann.org\u003e\n\ncommit c6ed0ba8b76ed57a346fddd1be8fc0a168190c9c\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 15 11:22:55 2019 +0530\n\n Bluetooth: 6lowpan: Make variable header_ops constant\n \n Static variable header_ops, of type header_ops, is used only once, when\n it is assigned to field header_ops of a variable having type net_device.\n This corresponding field is declared as const in the definition of\n net_device. Hence make header_ops constant as well to protect it from\n unnecessary modification.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Marcel Holtmann \u003cmarcel@holtmann.org\u003e\n\ncommit 8528d50827b79756d4c73a1a02ebdccc2e93e7f7\nAuthor: Alex Lu \u003calex_lu@realsil.com.cn\u003e\nDate: Wed Aug 14 20:02:52 2019 +0800\n\n Bluetooth: btusb: Fix suspend issue for Realtek devices\n \n From the perspective of controller, global suspend means there is no\n SET_FEATURE (DEVICE_REMOTE_WAKEUP) and controller would drop the\n firmware. It would consume less power. So we should not send this kind\n of SET_FEATURE when host goes to suspend state.\n Otherwise, when making device enter selective suspend, host should send\n SET_FEATURE to make sure the firmware remains.\n \n Signed-off-by: Alex Lu \u003calex_lu@realsil.com.cn\u003e\n Signed-off-by: Marcel Holtmann \u003cmarcel@holtmann.org\u003e\n\ncommit a4f077da6ed3c098175b40165b6863f927312c56\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Mon Aug 12 14:22:11 2019 +0800\n\n Bluetooth: hci_bcm: Fix -Wunused-const-variable warnings\n \n If CONFIG_ACPI is not set, gcc warn this:\n \n drivers/bluetooth/hci_bcm.c:831:39: warning:\n acpi_bcm_int_last_gpios defined but not used [-Wunused-const-variable\u003d]\n drivers/bluetooth/hci_bcm.c:838:39: warning:\n acpi_bcm_int_first_gpios defined but not used [-Wunused-const-variable\u003d]\n \n move them to #ifdef CONFIG_ACPI block.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Marcel Holtmann \u003cmarcel@holtmann.org\u003e\n\ncommit d2edeeec232dd53803e085c3245e088cd9795457\nAuthor: Matthias Kaehlcke \u003cmka@chromium.org\u003e\nDate: Wed Aug 7 11:58:49 2019 -0700\n\n Bluetooth: hci_qca: Remove redundant initializations to zero\n \n The qca_data structure is allocated with kzalloc() and hence\n zero-initialized. Remove a bunch of unnecessary explicit\n initializations of struct members to zero.\n \n Signed-off-by: Matthias Kaehlcke \u003cmka@chromium.org\u003e\n Reviewed-by: Balakrishna Godavarthi \u003cbgodavar@codeaurora.org\u003e\n Signed-off-by: Marcel Holtmann \u003cmarcel@holtmann.org\u003e\n\ncommit 623fc8e457f7365b34be4d4e89b2472fe2cec43a\nAuthor: Spoorthi Ravishankar Koppad \u003cspoorthix.k@intel.com\u003e\nDate: Mon Jul 15 17:05:22 2019 +0530\n\n Bluetooth: Add support for utilizing Fast Advertising Interval\n \n Changes made to add support for fast advertising interval\n as per core 4.1 specification, section 9.3.11.2.\n \n A peripheral device entering any of the following GAP modes and\n sending either non-connectable advertising events or scannable\n undirected advertising events should use adv_fast_interval2\n (100ms - 150ms) for adv_fast_period(30s).\n \n - Non-Discoverable Mode\n - Non-Connectable Mode\n - Limited Discoverable Mode\n - General Discoverable Mode\n \n Signed-off-by: Spoorthi Ravishankar Koppad \u003cspoorthix.k@intel.com\u003e\n Signed-off-by: Marcel Holtmann \u003cmarcel@holtmann.org\u003e\n\ncommit 6078f28cf9d1487c900964b8c3bf69253133ee42\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Thu Aug 29 20:22:06 2019 +0900\n\n mmc: sdhci: use lower/upper_32_bits() macros for DMA addresses\n \n Currently, the DMA addresses are casted to (u64) for the upper 32bits\n to avoid \"right shift count \u003e\u003d width of type\" warning.\n \n \u003clinux/kernel.h\u003e provides macros to address this, and I like the macro\n names are self-documenting.\n \n I introduced a new helper, sdhci_set_adma_addr() to avoid the code\n duplication.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Acked-by: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit d620d496ba9b752559c87b9b4e77e7a752da4d91\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Thu Aug 29 19:49:28 2019 +0900\n\n mmc: sdhci-cadence: override spec version\n \n The datasheet of the IP (sd4hc) says it is compiatible with SDHCI v4,\n but the spec version field in the version register is read as 2\n (i.e. SDHCI_SPEC_300) based on the RTL provided by Cadence.\n \n Socionext did not fix it up when it integrated the IP into the SoCs.\n So, it is working as SDHCI v3.\n \n It is not a real problem because there is no difference in the program\n flow in sdhci.c between SDHCI_SPEC_300/400, but set the real version\n just in case.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Acked-by: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit 355af48d4898fc4857ae894d11d396659b1724fe\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Thu Aug 29 19:49:27 2019 +0900\n\n mmc: sdhci: constify references of parameters to __sdhci_read_caps()\n \n __sdhci_read_caps() does not modify *ver, *caps, or *caps1.\n \n Probably, the caller of this function will want to constifythe\n parameters passed in.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Acked-by: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit 0b09197146bed1f38fbe2075b8f57caeaacef9f4\nAuthor: yong mao \u003cyong.mao@mediatek.com\u003e\nDate: Wed Aug 28 17:45:24 2019 +0800\n\n mmc: mediatek: enable SDIO IRQ low level trigger function\n \n SDIO IRQ is not defaultly triggered by low level,\n but by falling edge. It needs to set related register\n to enable SDIO IRQ low level trigger function.\n Otherwise the SDIO IRQ may be lost in some specail condition.\n \n Signed-off-by: Yong Mao \u003cyong.mao@mediatek.com\u003e\n Signed-off-by: Chaotian Jing \u003cchaotian.jing@mediatek.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit f6c6a6c7bd9fb8f78d67f9a5b3708a966e556831\nAuthor: Ben Chuang \u003cben.chuang@genesyslogic.com.tw\u003e\nDate: Tue Aug 27 08:33:22 2019 +0800\n\n mmc: sdhci: Export sdhci_abort_tuning function symbol\n \n Export sdhci_abort_tuning() function symbols which are used by other SD Host\n controller driver modules.\n \n Signed-off-by: Ben Chuang \u003cben.chuang@genesyslogic.com.tw\u003e\n Co-developed-by: Michael K Johnson \u003cjohnsonm@danlj.org\u003e\n Signed-off-by: Michael K Johnson \u003cjohnsonm@danlj.org\u003e\n Acked-by: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit 75f692537e2f813a5606f4d98e12f7d16cd8b08c\nAuthor: Ben Chuang \u003cben.chuang@genesyslogic.com.tw\u003e\nDate: Tue Aug 27 08:33:09 2019 +0800\n\n PCI: Add Genesys Logic, Inc. Vendor ID\n \n Add the Genesys Logic, Inc. vendor ID to pci_ids.h.\n \n Signed-off-by: Ben Chuang \u003cben.chuang@genesyslogic.com.tw\u003e\n Co-developed-by: Michael K Johnson \u003cjohnsonm@danlj.org\u003e\n Signed-off-by: Michael K Johnson \u003cjohnsonm@danlj.org\u003e\n Acked-by: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit d2e1a708b633d92f192d6160501c5892791dd0d7\nAuthor: Ben Chuang \u003cben.chuang@genesyslogic.com.tw\u003e\nDate: Tue Aug 27 08:32:55 2019 +0800\n\n mmc: sdhci: Add PLL Enable support to internal clock setup\n \n The GL9750 and GL9755 chipsets, and possibly others, require PLL Enable\n setup as part of the internal clock setup as described in 3.2.1 Internal\n Clock Setup Sequence of SD Host Controller Simplified Specification\n Version 4.20.\n \n Signed-off-by: Ben Chuang \u003cben.chuang@genesyslogic.com.tw\u003e\n Co-developed-by: Michael K Johnson \u003cjohnsonm@danlj.org\u003e\n Signed-off-by: Michael K Johnson \u003cjohnsonm@danlj.org\u003e\n Acked-by: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit dffa1c3825a05f7e257567331673018f952b478a\nAuthor: Ben Chuang \u003cben.chuang@genesyslogic.com.tw\u003e\nDate: Tue Aug 27 08:32:42 2019 +0800\n\n mmc: sdhci: Change timeout of loop for checking internal clock stable\n \n According to section 3.2.1 internal clock setup in SD Host Controller\n Simplified Specifications 4.20, the timeout of loop for checking\n internal clock stable is defined as 150ms.\n \n Signed-off-by: Ben Chuang \u003cben.chuang@genesyslogic.com.tw\u003e\n Co-developed-by: Michael K Johnson \u003cjohnsonm@danlj.org\u003e\n Signed-off-by: Michael K Johnson \u003cjohnsonm@danlj.org\u003e\n Acked-by: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit 9607af6f857ff062b29562fc1fe36d22b16b9d27\nAuthor: Ashish Kumar \u003cAshish.Kumar@nxp.com\u003e\nDate: Thu Aug 29 17:37:25 2019 +0530\n\n mtd: spi-nor: Rename \"n25q512a\" to \"mt25qu512a (n25q512a)\"\n \n n25q512a was rebranded to mt25qu512a after its spin off from STM.\n mt25qu512a is different only in terms of operating frequency, the\n JEDEC id is the same as in n25q512a.\n \n Dual reads are supported (0x3b, 0x3c), set the SPI_NOR_DUAL_READ flag.\n 4-byte opcodes are supported, set the SPI_NOR_4B_OPCODES flag.\n \n Tested Single I/O and QUAD I/O mode on LS1046FRWY.\n \n Signed-off-by: Kuldeep Singh \u003ckuldeep.singh@nxp.com\u003e\n Signed-off-by: Ashish Kumar \u003cAshish.Kumar@nxp.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n [tudor.ambarus@microchip.com: rename entry to \"mt25qu512a (n25q512a)\",\n reword commit message, order entry by size, drop comment as it looked\n redundant]\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n\ncommit 173c3d47736e31e388e3ff24cfd5d624f3d8d236\nAuthor: Ashish Kumar \u003cAshish.Kumar@nxp.com\u003e\nDate: Thu Aug 29 17:37:24 2019 +0530\n\n mtd: spi-nor: Add support for mt35xu02g\n \n mt35xu02g is an Octal flash supporting Single and OCTAL I/O.\n Tested on LS1028ARDB.\n \n Signed-off-by: Kuldeep Singh \u003ckuldeep.singh@nxp.com\u003e\n Signed-off-by: Ashish Kumar \u003cashish.kumar@nxp.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n [tudor.ambarus@microchip.com: reword commit message]\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n\ncommit 097a802ae15e2c6d17256a6fe0712a655ba10823\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Fri Aug 30 01:17:41 2019 +0800\n\n erofs: reduntant assignment in __erofs_get_meta_page()\n \n As Joe Perches suggested [1],\n err \u003d bio_add_page(bio, page, PAGE_SIZE, 0);\n - if (unlikely(err !\u003d PAGE_SIZE)) {\n + if (err !\u003d PAGE_SIZE) {\n err \u003d -EFAULT;\n goto err_out;\n }\n \n The initial assignment to err is odd as it\u0027s not\n actually an error value -E\u003cFOO\u003e but a int size\n from a unsigned int len.\n \n Here the return is either 0 or PAGE_SIZE.\n \n This would be more legible to me as:\n \n if (bio_add_page(bio, page, PAGE_SIZE, 0) !\u003d PAGE_SIZE) {\n err \u003d -EFAULT;\n goto err_out;\n }\n \n [1] https://lore.kernel.org/r/74c4784319b40deabfbaea92468f7e3ef44f1c96.camel@perches.com/\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190829171741.225219-1-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 8d8a09b093d7073465c824f74caf315c073d3875\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Fri Aug 30 00:38:27 2019 +0800\n\n erofs: remove all likely/unlikely annotations\n \n As Dan Carpenter suggested [1], I have to remove\n all erofs likely/unlikely annotations.\n \n [1] https://lore.kernel.org/linux-fsdevel/20190829154346.GK23584@kadam/\n Reported-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190829163827.203274-1-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 5cf89673645f7edfa075afc9e1d531db24706329\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Fri Aug 30 08:56:37 2019 +0200\n\n MAINTAINERS: fix up exfat directory location\n \n I messed up on the exfat MAINTAINER entry, the code is in\n drivers/staging/exfat/ not fs/exfat/\n \n Cc: Valdis Kletnieks \u003cvaldis.kletnieks@vt.edu\u003e\n Reported-by: Sasha Levin \u003calexander.levin@microsoft.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit c48c9f7ff32b8b3965a08e40eb6763682d905b5d\nAuthor: Valdis Klētnieks \u003cvaldis.kletnieks@vt.edu\u003e\nDate: Wed Aug 28 18:08:17 2019 +0200\n\n staging: exfat: add exfat filesystem code to staging\n \n The exfat code needs a lot of work to get it into \"real\" shape for\n the fs/ part of the kernel, so put it into drivers/staging/ for now so\n that it can be worked on by everyone in the community.\n \n The full specification of the filesystem can be found at:\n https://docs.microsoft.com/en-us/windows/win32/fileio/exfat-specification\n \n Signed-off-by: Valdis Kletnieks \u003cvaldis.kletnieks@vt.edu\u003e\n Signed-off-by: Sasha Levin \u003calexander.levin@microsoft.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Link: https://lore.kernel.org/r/20190828160817.6250-1-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 18a93cd38be3e69ac5b067c570a78a369b79e31d\nAuthor: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nDate: Fri Aug 30 13:56:14 2019 +1000\n\n usb: gadget: net2280: Add workaround for AB chip Errata 11\n \n The errata description is:\n \n Workaround for Default Duration of LFPS Handshake Signaling for\n Device-Initiated U1 Exit is too short.\n \n The default duration of the LFPS handshake generated by USB3380 for a device-initiated U1-exit may not be\n long enough for certain SuperSpeed downstream ports (SuperSpeed hubs/hosts) to recognize. This could lead\n to USB3380 entering the recovery state pre-maturely and ending up in the SS.Inactive state.\n \n I have observed various enumeration failures, seemingly related to\n lost transactions or SETUP status phases on modern hosts (typically\n thunderbolt capable systems) without this workaround.\n \n Signed-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit ec8ca8a3489cc87ca43a865af7e7393690841810\nAuthor: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nDate: Fri Aug 30 13:56:10 2019 +1000\n\n usb: gadget: net2280: Move all \"ll\" registers in one structure\n \n The split into multiple structures of the \"ll\" register bank is\n impractical. It makes it hard to add ll_lfps_timers_2 which is\n at offset 0x794, which is outside of the existing \"lfps\" structure\n and would require us to add yet another one.\n \n Instead, move all the \"ll\" registers into a single usb338x_ll_regs\n structure, and add ll_lfps_timers_2 while at it. It will be used\n in a subsequent patch.\n \n Signed-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 17b63704ec7cd4a37c513d964bc09c7e3fde8988\nAuthor: Thinh Nguyen \u003cThinh.Nguyen@synopsys.com\u003e\nDate: Thu Aug 29 18:00:16 2019 -0700\n\n usb: dwc3: gadget: Workaround Mirosoft\u0027s BESL check\n \n While testing our host system using Microsoft\u0027s usb stack against our\n gadget for various BESL values, we found an issue with their usb stack\n when the recommended baseline BESL value is 0 (125us) or when the deep\n BESL is 1 or less. The Windows host will issue a usb reset immediately\n after it receives the extended BOS descriptor and the enumeration will\n fail after a few attempts.\n \n To keep compatibility with Microsoft\u0027s host usb stack, let\u0027s workaround\n this issue by using the recommended baseline BESL of 1 (or 150us)\n and clamp the deep BESL value within 2 to 15.\n \n This was tested against Windows 10 build 18956.\n \n Signed-off-by: Thinh Nguyen \u003cthinhn@synopsys.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 6af86bdb8ad41f4cf1292d3b10857dc322758328\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Fri Aug 30 10:14:56 2019 +0900\n\n ALSA: firewire-motu: add support for MOTU 4pre\n \n MOTU 4pre was launched in 2012 by MOTU, Inc. This commit allows userspace\n applications can transmit and receive PCM frames and MIDI messages for\n this model via ALSA PCM interface and RawMidi/Sequencer interfaces.\n \n The device supports MOTU protocol version 3. Unlike the other devices, the\n device is simply designed. The size of data block is fixed to 10 quadlets\n during available sampling rates (44.1 - 96.0 kHz). Each data block\n includes 1 source packet header, 2 data chunks for messages, 8 data chunks\n for PCM samples and 2 data chunks for padding to quadlet alignment. The\n device has no MIDI, optical, BNC and AES/EBU interfaces.\n \n Like support for the other MOTU devices, the quality of playback sound\n is not enough good with periodical noise yet.\n \n $ python2 crpp \u003c ~/git/am-config-rom/motu/motu-4pre.img\n ROM header and bus information block\n -----------------------------------------------------------------\n 400 041078cc bus_info_length 4, crc_length 16, crc 30924\n 404 31333934 bus_name \"1394\"\n 408 20ff7000 irmc 0, cmc 0, isc 1, bmc 0, cyc_clk_acc 255, max_rec 7 (256)\n 40c 0001f200 company_id 0001f2 |\n 410 000a41c5 device_id 00000a41c5 | EUI-64 0001f200000a41c5\n \n root directory\n -----------------------------------------------------------------\n 414 0004ef04 directory_length 4, crc 61188\n 418 030001f2 vendor\n 41c 0c0083c0 node capabilities per IEEE 1394\n 420 d1000002 --\u003e unit directory at 428\n 424 8d000005 --\u003e eui-64 leaf at 438\n \n unit directory at 428\n -----------------------------------------------------------------\n 428 0003ceda directory_length 3, crc 52954\n 42c 120001f2 specifier id\n 430 13000045 version\n 434 17103800 model\n \n eui-64 leaf at 438\n -----------------------------------------------------------------\n 438 0002d248 leaf_length 2, crc 53832\n 43c 0001f200 company_id 0001f2 |\n 440 000a41c5 device_id 00000a41c5 | EUI-64 0001f200000a41c5\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit e7ee96dfb8c2687a29d2c5c3b06c967fa54b839c\nAuthor: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nDate: Wed Aug 28 14:37:57 2019 -0700\n\n xfs: remove all *_ITER_ABORT values\n \n Use -ECANCELED to signal \"stop iterating\" instead of these magical\n *_ITER_ABORT values, since it\u0027s duplicative.\n \n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Reviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n\ncommit 9b123d1ea23701bc00ebf712f1e03b25b8195eeb\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:57:01 2019 +1000\n\n powerpc/64s/exception: reduce page fault unnecessary loads\n \n This avoids 3 loads in the radix page fault case, 1 load in the\n hash fault case, and 2 loads in the hash miss page fault case.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-37-npiggin@gmail.com\n\ncommit 05f97d94dd0e2883bc7b2e6b7b5e4c088e0d1437\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:57:00 2019 +1000\n\n powerpc/64s/exception: Remove pointless KVM handler name bifurcation\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-36-npiggin@gmail.com\n\ncommit 1b3599829a2560fe6c9a5a4a6bb6b2bc4e5bdaee\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:59 2019 +1000\n\n powerpc/64s/exception: program check handler do not branch into a macro\n \n It is clever, but the small code saving is not worth the spaghetti of\n jumping to a label in an expanded macro, particularly when the label\n is just a number rather than a descriptive name.\n \n So expand the INT_COMMON macro twice, once for the stack and no stack\n cases, and branch to those. The slight code size increase is worth\n the improved clarity of branches for this non-performance critical\n code.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-35-npiggin@gmail.com\n\ncommit c7c5cbb42d6e207a059c64740b1654376619345e\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:58 2019 +1000\n\n powerpc/64s/exception: move interrupt entry code above the common handler\n \n This better reflects the order in which the code is executed.\n \n No generated code change except BUG line number constants.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-34-npiggin@gmail.com\n\ncommit d1a84718888e768296557b83f4c56cb1caef8fdd\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:57 2019 +1000\n\n powerpc/64s/exception: INT_COMMON add DAR, DSISR, reconcile options\n \n Move DAR and DSISR saving to pt_regs into INT_COMMON. Also add an\n option to expand RECONCILE_IRQ_STATE.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-33-npiggin@gmail.com\n\ncommit 8c9fb5d4f3ddf02fb0fa3dec2dffd6b007ac894c\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:56 2019 +1000\n\n powerpc/64s/exception: Expand EXCEPTION_PROLOG_COMMON_1 and 2 into caller\n \n No generated code change except BUG line number constants.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-32-npiggin@gmail.com\n\ncommit 5d5e0edfd5fa2a60d0f9b8d6ece1a5ce51aae3b5\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:55 2019 +1000\n\n powerpc/64s/exception: Expand EXCEPTION_COMMON macro into caller\n \n No generated code change except BUG line number constants.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-31-npiggin@gmail.com\n\ncommit bcbceed40a8c355b48678d90c5c407dfca811f0e\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:54 2019 +1000\n\n powerpc/64s/exception: Add INT_COMMON gas macro to generate common exception code\n \n No generated code change except BUG line number constants.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-30-npiggin@gmail.com\n\ncommit 9a9c739aa83d031da7468028de8a65608146eccc\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:53 2019 +1000\n\n powerpc/64s/exception: Merge EXCEPTION_PROLOG_COMMON_2/3\n \n Merge EXCEPTION_PROLOG_COMMON_3 into EXCEPTION_PROLOG_COMMON_2.\n \n No generated code change except BUG line number constants.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-29-npiggin@gmail.com\n\ncommit 7027d53d1ab17d28b65913148585d5a331446b8b\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:52 2019 +1000\n\n powerpc/64s/exception: KVM_HANDLER reorder arguments to match other macros\n \n Also change argument name (n -\u003e vec) to match others.\n \n No generated code change.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-28-npiggin@gmail.com\n\ncommit 141fed2669a93604bd5ce8b793d85f4798626ef5\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:51 2019 +1000\n\n powerpc/64s/exception: Add INT_KVM_HANDLER gas macro\n \n Replace the 4 variants of cpp macros with one gas macro.\n \n No generated code change except BUG line number constants.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-27-npiggin@gmail.com\n\ncommit 4515c5fa41936088a57efe0b64d1bb46a4943582\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:50 2019 +1000\n\n powerpc/64s/exception: INT_HANDLER support HDAR/HDSISR and use it in HDSI\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-26-npiggin@gmail.com\n\ncommit 52b989231c6fad42dff57b69a53f38756db48e06\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:49 2019 +1000\n\n powerpc/64s/exception: Add the virt variant of the denorm interrupt handler\n \n All other virt handlers have the prolog code in the virt vector rather\n than branch to the real vector. Follow this pattern in the denorm virt\n handler.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-25-npiggin@gmail.com\n\ncommit d29768e13cf6b50bd54a690e2ac52ab71465e2eb\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:48 2019 +1000\n\n powerpc/64s/exception: remove EXCEPTION_PROLOG_0/1, rename _2\n \n EXCEPTION_PROLOG_0 and _1 have only a single caller, so expand them\n into it.\n \n Rename EXCEPTION_PROLOG_2_REAL to INT_SAVE_SRR_AND_JUMP and\n EXCEPTION_PROLOG_2_VIRT to INT_VIRT_SAVE_SRR_AND_JUMP, which are\n more descriptive.\n \n No generated code change except BUG line number constants.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-24-npiggin@gmail.com\n\ncommit 9b40f62b8a49f0c922eb4b6cf6502f1307745c98\nAuthor: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\nDate: Thu Aug 29 23:36:08 2019 +1000\n\n powerpc/64s/exceptions: Use keyword params to shorten arg lists\n \n The argument lists for the INT_HANDLER macro are getting a bit\n unwieldy. Use keyword parameters with default values to shorten them.\n \n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190830011426.16810-1-mpe@ellerman.id.au\n\ncommit 3f1071ec39f79ad1e288d2f8e5cbe5697ced695b\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Wed Aug 28 15:21:08 2019 -0500\n\n net: spider_net: Use struct_size() helper\n \n One of the more common cases of allocation size calculations is finding\n the size of a structure that has a zero-sized array at the end, along\n with memory for some number of elements for that array. For example:\n \n struct spider_net_card {\n ...\n struct spider_net_descr darray[0];\n };\n \n Make use of the struct_size() helper instead of an open-coded version\n in order to avoid any potential type mistakes.\n \n So, replace the following form:\n \n sizeof(struct spider_net_card) + (tx_descriptors + rx_descriptors) * sizeof(struct spider_net_descr)\n \n with:\n \n struct_size(card, darray, tx_descriptors + rx_descriptors)\n \n Notice that, in this case, variable alloc_size is not necessary, hence it\n is removed.\n \n Building: allmodconfig powerpc.\n \n This code was detected with the help of Coccinelle.\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b3a42e3a78ce0599c700a8c89befc67a35d54c31\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Wed Aug 28 22:29:05 2019 +0200\n\n r8169: add support for EEE on RTL8125\n \n This adds EEE support for RTL8125 based on the vendor driver.\n Supported is EEE for 100Mbps and 1Gbps. Realtek recommended to not yet\n enable EEE for 2.5Gbps due to potential compatibility issues. Also\n ethtool doesn\u0027t support yet controlling EEE for 2.5Gbps and 5Gbps.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 02bf642b188a32b63e2406ed61e940dd80e5b1c7\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Wed Aug 28 22:28:32 2019 +0200\n\n r8169: add RTL8125 PHY initialization\n \n This patch adds PHY initialization magic copied from the r8125 vendor\n driver. In addition it supports loading the firmware for chip version\n RTL_GIGA_MAC_VER_61.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f1bce4ad2f1cee6759711904b9fffe4a3dd8af87\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Wed Aug 28 22:28:03 2019 +0200\n\n r8169: add support for RTL8125\n \n This adds support for 2.5Gbps chip RTL8125, it\u0027s partially based on the\n r8125 vendor driver. Tested with a Delock 89531 PCIe card against a\n Netgear GS110MX Multi-Gig switch. Firmware isn\u0027t strictly needed,\n but on some systems there may be compatibility issues w/o firmware.\n Firmware has been submitted to linux-firmware.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ae84bc18733752e9bf47227bd80b3c0f3649b8d0\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Wed Aug 28 22:27:30 2019 +0200\n\n r8169: don\u0027t use bit LastFrag in tx descriptor after send\n \n On RTL8125 this bit is always cleared after send. Therefore check for\n tx_skb-\u003eskb being set what is functionally equivalent.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7366016d2d4c7b2e5168db6fa7920fa094561db5\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Wed Aug 28 22:26:51 2019 +0200\n\n r8169: read common register for PCI commit\n \n RTL8125 uses a different register number for IntrMask.\n To net have side effects by reading a random register let\u0027s\n use a register that is the same on all supported chip families.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit bcf2b868a5ae8b9b332176b65247953176630990\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Wed Aug 28 22:26:13 2019 +0200\n\n r8169: move disabling interrupt coalescing to RTL8169/RTL8168 init\n \n RTL8125 doesn\u0027t support the same coalescing registers, therefore move\n this initialization to the 8168/6169-specific init.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ce37115e3a5741219ceb0bb26de23faba6b93881\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Wed Aug 28 22:25:32 2019 +0200\n\n r8169: factor out reading MAC address from registers\n \n For RTL8125 we will have to read the MAC address also from another\n register range, therefore create a small helper.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c623305bf465a77a4e34f46f48e27809df2f6cce\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Wed Aug 28 22:24:54 2019 +0200\n\n r8169: restrict rtl_is_8168evl_up to RTL8168 chip versions\n \n Extend helper rtl_is_8168evl_up to properly work once we add\n mac version numbers \u003e51 for RTL8125.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c1d532d268cb99c17f7094d26819a0a1142aee51\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Wed Aug 28 22:24:13 2019 +0200\n\n r8169: change interrupt mask type to u32\n \n RTL8125 uses a 32 bit interrupt mask even though only bits in the\n lower 16 bits are used. Change interrupt mask size to u32 to be\n prepared and reintroduce helper rtl_get_events.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7299417c8214ce09c0da01d7719c7b1c11503578\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:47 2019 +1000\n\n powerpc/64s/exception: Replace PROLOG macros and EXC helpers with a gas macro\n \n This creates a single macro that generates the exception prolog code,\n with variants specified by arguments, rather than assorted nested\n macros for different variants.\n \n The increasing length of macro argument list is not nice to read or\n modify, but this is a temporary condition that will be improved in\n later changes.\n \n No generated code change except BUG line number constants and label\n names.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-23-npiggin@gmail.com\n\ncommit 5ff79a5ea69f8f5d1131064af9f8c9a8b1bc8266\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:46 2019 +1000\n\n powerpc/64s/exception: remove 0xb00 handler\n \n This vector is not used by any supported processor, and has been\n implemented as an unknown exception going back to 2.6. There is\n nothing special about 0xb00, so remove it like other unused\n vectors.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-22-npiggin@gmail.com\n\ncommit 9a7a0773d7d2fd1e4c581f42ad75de7872a174dc\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:45 2019 +1000\n\n powerpc/64s/exception: Fix performance monitor virt handler\n \n The perf virt handler uses EXCEPTION_PROLOG_2_REAL rather than _VIRT.\n In practice this is okay because the _REAL variant is usable by virt\n mode interrupts, but should be fixed (and is a performance win).\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-21-npiggin@gmail.com\n\ncommit def0db4f9ddc24ad7b37735c34a78e9c3c7978ef\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:44 2019 +1000\n\n powerpc/64s/exception: Add EXC_HV_OR_STD, which selects HSRR if HVMODE\n \n Add EXC_HV_OR_STD and use it to consolidate the 0x500 external\n interrupt.\n \n Executed code is unchanged.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-20-npiggin@gmail.com\n\ncommit a243281195c338489ec5088380821f64340cb82f\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:43 2019 +1000\n\n powerpc/64s/exception: move head-64.h exception code to exception-64s.S\n \n The head-64.h code should deal only with the head code sections\n and offset calculations.\n \n No generated code change except BUG line number constants.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-19-npiggin@gmail.com\n\ncommit c31f7134dc53f7020b3d49e846d1b950a761e324\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:42 2019 +1000\n\n powerpc/64s/exception: Fix DAR load for handle_page_fault error case\n \n This buglet goes back to before the 64/32 arch merge, but it does not\n seem to have had practical consequences because bad_page_fault does\n not use the 2nd argument, but rather regs-\u003edar/nip.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-18-npiggin@gmail.com\n\ncommit b3fe35261e329e15736bc95630fd865df9896c66\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:41 2019 +1000\n\n powerpc/64s/exception: machine check improve labels and comments\n \n Short forward and backward branches can be given number labels,\n but larger significant divergences in code path a more readable\n if they\u0027re given descriptive names.\n \n Also adjusts a comment to account for guest delivery.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-17-npiggin@gmail.com\n\ncommit fce16d482276f059b08368c833a1188ac1f25e86\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:40 2019 +1000\n\n powerpc/64s/exception: untangle early machine check handler branch\n \n machine_check_early_common now branches to machine_check_handle_early\n which is its only caller.\n \n Move interleaving code out of the way, and remove the branch.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-16-npiggin@gmail.com\n\ncommit b7d9ccec3056913528690c5fae7cc86a5ea3dffc\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:39 2019 +1000\n\n powerpc/64s/exception: machine check move unrecoverable handling out of line\n \n Similarly to the previous change, all callers of the unrecoverable\n handler run relocated so can reach it with a direct branch. This makes\n it easy to move out of line, which makes the \"normal\" path less\n cluttered and easier to follow.\n \n MSR[ME] manipulation still requires the rfi, so that is moved out of\n line to its own function.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-15-npiggin@gmail.com\n\ncommit 296e753fb447eb14c8c9fd6a7c48e7ffab269343\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:38 2019 +1000\n\n powerpc/64s/exception: simplify machine check early path\n \n machine_check_handle_early_common can reach machine_check_handle_early\n directly now that it runs at the relocated address, so just branch\n directly.\n \n The rfi sequence is required to enable MSR[ME] but that step is moved\n into a helper function, making the code easier to follow.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-14-npiggin@gmail.com\n\ncommit abd1f4ca2b41ffba768c3baadc006a95d178fbf1\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:37 2019 +1000\n\n powerpc/64s/exception: machine check move tramp code\n \n Following convention, move the tramp code (unrelocated) above the\n common handlers (relocated).\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-13-npiggin@gmail.com\n\ncommit c8eb54dbc8087c3d114cb583925395e211bfffa4\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:36 2019 +1000\n\n powerpc/64s/exception: machine check restructure to reuse common macros\n \n Follow the pattern of sreset and HMI handlers more closely: use\n EXCEPTION_PROLOG_COMMON_1 rather than open-coding it, and run the\n handler at the relocated location.\n \n This helps later simplification and code sharing.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-12-npiggin@gmail.com\n\ncommit 272f636445cf556498c8840dc63ad1218e94391b\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:35 2019 +1000\n\n powerpc/64s/exception: machine check pseries should skip the late handler for kernel MCEs\n \n The powernv machine check handler copes with taking a MCE from one of\n three contexts, guest, kernel, and user. In each case the early\n handler runs first on a special stack, then:\n \n - The guest case branches to the KVM interrupt handler (via standard\n interrupt macros).\n - The user case will run the \"late\" handler which is like a normal\n interrupt that runs in virtual mode and uses the regular kernel\n stack.\n - The kernel case queues the event and schedules it for processing\n with irq work.\n \n The last case is important, it must not enable virtual memory because\n the MMU state may not be set up to deal with that (e.g., SLB might be\n clear), it must not use the regular kernel stack for similar reasons\n (e.g., might be in OPAL with OPAL stack in r1), and the kernel does\n not expect anything to touch its stack if interrupts are disabled.\n \n The pseries handler does not do this queueing, but instead it always\n runs the late handler for host MCEs, which has some of the same\n problems.\n \n Now that pseries is using machine_check_events, change it to do the\n same as powernv and queue events for kernel MCEs.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-11-npiggin@gmail.com\n\ncommit 9ca766f9891d23743b4e1a7b1cafdc63723cd6a7\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:34 2019 +1000\n\n powerpc/64s/pseries: machine check convert to use common event code\n \n The common machine_check_event data structures and queues are mostly\n platform independent, with powernv decoding SRR1/DSISR/etc., into\n machine_check_event objects.\n \n This patch converts pseries to use this infrastructure by decoding\n fwnmi/rtas data into machine_check_event objects.\n \n This allows queueing to be used by a subsequent change to delay the\n virtual mode handling of machine checks that occur in kernel space\n where it is unsafe to switch immediately to virtual mode, similarly\n to powernv.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n [mpe: Fix implicit fallthrough warnings in mce_handle_error()]\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-10-npiggin@gmail.com\n\ncommit 7290f3b3d3e66b54720f23079ffc60e0b7bbb0cc\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:33 2019 +1000\n\n powerpc/64s/powernv: machine check dump SLB contents\n \n Re-use the code introduced in pseries to save and dump the contents\n of the SLB in the case of an SLB involved machine check exception.\n \n This patch also avoids allocating the SLB save array on pseries radix.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-9-npiggin@gmail.com\n\ncommit 0b66370c61fcf5fcc1d6901013e110284da6e2bb\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:32 2019 +1000\n\n powerpc/64s/exception: machine check use correct cfar for late handler\n \n Bare metal machine checks run an \"early\" handler in real mode before\n running the main handler which reports the event.\n \n The main handler runs exactly as a normal interrupt handler, after the\n \"windup\" which sets registers back as they were at interrupt entry.\n CFAR does not get restored by the windup code, so that will be wrong\n when the handler is run.\n \n Restore the CFAR to the saved value before running the late handler.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-8-npiggin@gmail.com\n\ncommit fa2760eca504f554a5adb6cd2f576828933c4c7b\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:31 2019 +1000\n\n powerpc/64s/exception: machine check remove machine_check_pSeries_0 branch\n \n This label has only one caller, so unwind the branch and move it\n inline. The location of the comment is adjusted to match similar\n one in system reset.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-7-npiggin@gmail.com\n\ncommit b5c27f7c5679c3726148fd25ad220b4560d210cf\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:30 2019 +1000\n\n powerpc/64s/exception: machine check pseries should always run the early handler\n \n Now that pseries with fwnmi registered runs the early machine check\n handler, there is no good reason to special case the non-fwnmi case\n and skip the early handler. Reducing the code and number of paths is\n a top priority for asm code, it\u0027s better to handle this in C where\n possible (and the pseries early handler is a no-op if fwnmi is not\n registered).\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-6-npiggin@gmail.com\n\ncommit fe9d482b1d87c76441492e51d866cee652eee4d5\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:29 2019 +1000\n\n powerpc/64s/exception: machine check adjust RFI target\n \n The host kernel delivery case for powernv does RFI_TO_USER_OR_KERNEL,\n but should just use RFI_TO_KERNEL which makes it clear this is not a\n user case.\n \n This is not a bug because RFI_TO_USER_OR_KERNEL deals with kernel\n returns just fine.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-5-npiggin@gmail.com\n\ncommit 19dbe673e62b076f00ae2841fcf5898b4728d5ab\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:28 2019 +1000\n\n powerpc/64s/exception: machine check fix KVM guest test\n \n The machine_check_handle_early hypervisor guest test is skipped if\n !HVMODE or MSR[HV]\u003d0, which is wrong for PR or nested hypervisors\n that could be running a guest in this state.\n \n Test HSTATE_IN_GUEST up front and use that to branch out to the KVM\n handler, then MSR[PR] alone can test for this kernel\u0027s userspace.\n This matches all other interrupt handling.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-4-npiggin@gmail.com\n\ncommit 1039f62431e2aa16487cd6d64bc841d71f6465b8\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:27 2019 +1000\n\n powerpc/64s/exception: machine check remove bitrotted comment\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-3-npiggin@gmail.com\n\ncommit 0be9f7fd5d8fd984b34ad98838ef7cfd0079ddae\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Fri Aug 2 20:56:26 2019 +1000\n\n powerpc/64s/exception: machine check fwnmi remove HV case\n \n fwnmi does not trigger in HV mode, so remove always-true feature test.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802105709.27696-2-npiggin@gmail.com\n\ncommit 42aa15cf05c0a47cc5807c21c7ff471b80cad371\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Wed Aug 28 14:55:11 2019 -0400\n\n net: dsa: mv88e6xxx: fix freeing unused SERDES IRQ\n \n Now mv88e6xxx does not enable its ports at setup itself and let\n the DSA core handle this, unused ports are disabled without being\n powered on first. While that is expected, the SERDES powering code\n was assuming that a port was already set up before powering it down,\n resulting in freeing an unused IRQ. The patch fixes this assumption.\n \n Fixes: b759f528ca3d (\"net: dsa: mv88e6xxx: enable SERDES after setup\")\n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Reviewed-by: Marek Behún \u003cmarek.behun@nic.cz\u003e\n Tested-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 5d24da1e5ccb0ec3b1ec39e582d6cdc0806dbb39\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Wed Aug 28 12:26:59 2019 -0400\n\n net: dsa: mv88e6xxx: keep CMODE writable code private\n \n This is a follow-up patch for commit 7a3007d22e8d (\"net: dsa:\n mv88e6xxx: fully support SERDES on Topaz family\").\n \n Since .port_set_cmode is only called from mv88e6xxx_port_setup_mac and\n mv88e6xxx_phylink_mac_config, it is fine to keep this \"make writable\"\n code private to the mv88e6341_port_set_cmode implementation, instead\n of adding yet another operation to the switch info structure.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Reviewed-by: Marek Behún \u003cmarek.behun@nic.cz\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b98f0f530d7978157b553a97556c5cc6489b3108\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Wed Aug 28 12:26:11 2019 -0400\n\n net: dsa: mv88e6xxx: get serdes lane after lock\n \n This is a follow-up patch for commit 17deaf5cb37a (\"net: dsa:\n mv88e6xxx: create serdes_get_lane chip operation\").\n \n The .serdes_get_lane implementations access the CMODE of a port,\n even though it is cached at the moment, it is safer to call them\n after the mutex is locked, not before.\n \n At the same time, check for an eventual error and return IRQ_DONE,\n instead of blindly ignoring it.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Reviewed-by: Marek Behún \u003cmarek.behun@nic.cz\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit dd2956eab104185bda122af162f90b06322a05e6\nAuthor: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\nDate: Wed Aug 28 22:23:16 2019 +0800\n\n net: hns3: not allow SSU loopback while execute ethtool -t dev\n \n The current loopback mode is to add 0x1F to the SMAC address\n as the DMAC address and enable the promiscuous mode.\n However, if the VF address is the same as the DMAC address,\n the loopback test fails.\n \n Loopback can be enabled in three places: SSU, MAC, and serdes.\n By default, SSU loopback is enabled, so if the SMAC and the DMAC\n are the same, the packets are looped back in the SSU. If SSU loopback\n is disabled, packets can reach MAC even if SMAC is the same as DMAC.\n \n Therefore, this patch disables the SSU loopback before the loopback\n test. In this way, the SMAC and DMAC can be the same, and the\n promiscuous mode does not need to be enabled. And this is not\n valid in version 0x20.\n \n This patch also uses a macro to replace 0x1F.\n \n Fixes: c39c4d98dc65 (\"net: hns3: Add mac loopback selftest support in hns3 driver\")\n Signed-off-by: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 2336f19d789223b9f42f111aab8de6ad66d12c28\nAuthor: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\nDate: Wed Aug 28 22:23:15 2019 +0800\n\n net: hns3: check reset interrupt status when reset fails\n \n Currently, the reset interrupt will be cleared firstly, so when\n reset fails, if interrupt status register has reset interrupt,\n it means there is a new coming reset.\n \n Fixes: 72e2fb07997c (\"net: hns3: clear reset interrupt status in hclge_irq_handle()\")\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c9765a89d142f614b917ead75e23f1915f7e748d\nAuthor: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\nDate: Wed Aug 28 22:23:14 2019 +0800\n\n net: hns3: add phy selftest function\n \n Currently, the loopback test supports only mac selftest and serdes\n selftest. This patch adds phy selftest.\n \n Signed-off-by: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a83d29618b1cd9176ad33fc415f5e033fd6a45a2\nAuthor: Weihang Li \u003cliweihang@hisilicon.com\u003e\nDate: Wed Aug 28 22:23:13 2019 +0800\n\n net: hns3: implement .process_hw_error for hns3 client\n \n When hardware or IMP get specified error it may need the client\n to take some special operations.\n \n This patch implements the hns3 client\u0027s process_hw_errorx.\n \n Signed-off-by: Weihang Li \u003cliweihang@hisilicon.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit e8df45c281342b6f0193faa242148acfe95c6f09\nAuthor: Zhongzhu Liu \u003cliuzhongzhu@huawei.com\u003e\nDate: Wed Aug 28 22:23:12 2019 +0800\n\n net: hns3: optimize waiting time for TQP reset\n \n This patch optimizes the waiting time for TQP reset.\n \n Signed-off-by: Zhongzhu Liu \u003cliuzhongzhu@huawei.com\u003e\n Reviewed-by: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 82f7d0576fa6d511ecbfea73688176b700ae0e91\nAuthor: Guojia Liao \u003cliaoguojia@huawei.com\u003e\nDate: Wed Aug 28 22:23:11 2019 +0800\n\n net: hns3: fix incorrect type in assignment.\n \n This patch fixes some incorrect type in assignment reported by sparse.\n Those sparse warning as below:\n - warning : restricted __le16 degrades to integer\n - warning : cast from restricted __le32\n - warning : expected restricted __le32\n - warning : cast from restricted __be32\n - warning : cast from restricted __be16\n - warning : cast to restricted __le16\n \n Signed-off-by: Guojia Liao \u003cliaoguojia@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 199d2dd416df3a224c2e20da5499a9200a716829\nAuthor: Yonglong Liu \u003cliuyonglong@huawei.com\u003e\nDate: Wed Aug 28 22:23:10 2019 +0800\n\n net: hns3: make some reusable codes into a function\n \n In hclge_dcb.c, these pair of codes:\n hclge_notify_client(hdev, HNAE3_DOWN_CLIENT);\n hclge_notify_client(hdev, HNAE3_UNINIT_CLIENT);\n and\n hclge_notify_client(hdev, HNAE3_INIT_CLIENT);\n hclge_notify_client(hdev, HNAE3_UP_CLIENT);\n are called many times, so make them into a function.\n \n Signed-off-by: Yonglong Liu \u003cliuyonglong@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ed5b255ba679a2a6cf63a6891d9e802a9c77e5e1\nAuthor: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\nDate: Wed Aug 28 22:23:09 2019 +0800\n\n net: hns3: optimize some log printings\n \n To better identify abnormal conditions, this patch modifies or\n adds some logs to show driver status more accurately.\n \n Signed-off-by: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\n Signed-off-by: Zhongzhu Liu \u003cliuzhongzhu@huawei.com\u003e\n Signed-off-by: Guangbin Huang \u003chuangguangbin2@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 70a214903da9255b5e6aea35ca35c1a8e72dc5f6\nAuthor: Guojia Liao \u003cliaoguojia@huawei.com\u003e\nDate: Wed Aug 28 22:23:08 2019 +0800\n\n net: hns3: reduce the parameters of some functions\n \n This patch simplifies parameters of some functions by deleting\n unused parameter.\n \n Signed-off-by: Guojia Liao \u003cliaoguojia@huawei.com\u003e\n Signed-off-by: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\n Signed-off-by: Guangbin Huang \u003chuangguangbin2@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6125b52d2641497228e806c60b906769ab4b2c19\nAuthor: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\nDate: Wed Aug 28 22:23:07 2019 +0800\n\n net: hns3: modify base parameter of kstrtouint in hclge_dbg_dump_tm_map\n \n This patch replaces kstrtouint()\u0027s patameter base with 0 in the\n hclge_dbg_dump_tm_mac(), which makes it more flexible. Also\n uses a macro to replace string \"dump tm map\", since it has been\n used multiple times.\n \n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6f92bfd70a8ae545bc08941baa55316d8bfba514\nAuthor: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\nDate: Wed Aug 28 22:23:06 2019 +0800\n\n net: hns3: use macro instead of magic number\n \n This patch uses macro to replace some magic number.\n \n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a582b78dfc33d211ecaadda4cb5765b382d9d682\nAuthor: Zhongzhu Liu \u003cliuzhongzhu@huawei.com\u003e\nDate: Wed Aug 28 22:23:05 2019 +0800\n\n net: hns3: code optimization for debugfs related to \"dump reg\"\n \n For making the code more readable, this patch uses a array to\n keep the information about the dumping register, and then uses\n it to parse the parameter cmd_buf which passing into\n hclge_dbg_dump_reg_cmd().\n \n Also replaces parameter \"base\" of kstrtouint with 0 in the\n hclge_dbg_dump_reg_common(), which makes it more flexible.\n \n Signed-off-by: Zhongzhu Liu \u003cliuzhongzhu@huawei.com\u003e\n Signed-off-by: Guangbin Huang \u003chuangguangbin2@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit bf75a8db72e9454aadeca0276f6e6b0e9161861c\nAuthor: Ryan Grimm \u003cgrimm@linux.vnet.ibm.com\u003e\nDate: Mon Aug 19 23:13:26 2019 -0300\n\n powerpc/configs: Enable secure guest support in pseries and ppc64 defconfigs\n \n Enables running as a secure guest in platforms with an Ultravisor.\n \n Signed-off-by: Ryan Grimm \u003cgrimm@linux.vnet.ibm.com\u003e\n Signed-off-by: Ram Pai \u003clinuxram@us.ibm.com\u003e\n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820021326.6884-17-bauerman@linux.ibm.com\n\ncommit 2efbc58f157a39ad9e9199b92d9c47736023a2fe\nAuthor: Anshuman Khandual \u003ckhandual@linux.vnet.ibm.com\u003e\nDate: Mon Aug 19 23:13:24 2019 -0300\n\n powerpc/pseries/svm: Force SWIOTLB for secure guests\n \n SWIOTLB checks range of incoming CPU addresses to be bounced and sees if\n the device can access it through its DMA window without requiring bouncing.\n In such cases it just chooses to skip bouncing. But for cases like secure\n guests on powerpc platform all addresses need to be bounced into the shared\n pool of memory because the host cannot access it otherwise. Hence the need\n to do the bouncing is not related to device\u0027s DMA window and use of bounce\n buffers is forced by setting swiotlb_force.\n \n Also, connect the shared memory conversion functions into the\n ARCH_HAS_MEM_ENCRYPT hooks and call swiotlb_update_mem_attributes() to\n convert SWIOTLB\u0027s memory pool to shared memory.\n \n Signed-off-by: Anshuman Khandual \u003ckhandual@linux.vnet.ibm.com\u003e\n [ bauerman: Use ARCH_HAS_MEM_ENCRYPT hooks to share swiotlb memory pool. ]\n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820021326.6884-15-bauerman@linux.ibm.com\n\ncommit edea902c1c1efb855f77e041f9daf1abe7a9768a\nAuthor: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\nDate: Mon Aug 19 23:13:23 2019 -0300\n\n powerpc/pseries/iommu: Don\u0027t use dma_iommu_ops on secure guests\n \n Secure guest memory is inacessible to devices so regular DMA isn\u0027t\n possible.\n \n In that case set devices\u0027 dma_map_ops to NULL so that the generic\n DMA code path will use SWIOTLB to bounce buffers for DMA.\n \n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820021326.6884-14-bauerman@linux.ibm.com\n\ncommit 4edaac512c8a31b0aa715b58ee00ee06df642fa2\nAuthor: Sukadev Bhattiprolu \u003csukadev@linux.ibm.com\u003e\nDate: Mon Aug 19 23:13:22 2019 -0300\n\n powerpc/pseries/svm: Disable doorbells in SVM guests\n \n Normally, the HV emulates some instructions like MSGSNDP, MSGCLRP\n from a KVM guest. To emulate the instructions, it must first read\n the instruction from the guest\u0027s memory and decode its parameters.\n \n However for a secure guest (aka SVM), the page containing the\n instruction is in secure memory and the HV cannot access directly.\n It would need the Ultravisor (UV) to facilitate accessing the\n instruction and parameters but the UV currently does not have\n the support for such accesses.\n \n Until the UV has such support, disable doorbells in SVMs. This might\n incur a performance hit but that is yet to be quantified.\n \n With this patch applied (needed only in SVMs not needed for HV) we\n are able to launch SVM guests with multi-core support. Eg:\n \n qemu -smp sockets\u003d2,cores\u003d2,threads\u003d2.\n \n Fix suggested by Benjamin Herrenschmidt. Thanks to input from\n Paul Mackerras, Ram Pai and Michael Anderson.\n \n Signed-off-by: Sukadev Bhattiprolu \u003csukadev@linux.ibm.com\u003e\n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820021326.6884-13-bauerman@linux.ibm.com\n\ncommit 734560ac39aeb2516419c5878856df011f794a74\nAuthor: Ryan Grimm \u003cgrimm@linux.vnet.ibm.com\u003e\nDate: Mon Aug 19 23:13:21 2019 -0300\n\n powerpc/pseries/svm: Export guest SVM status to user space via sysfs\n \n User space might want to know it\u0027s running in a secure VM. It can\u0027t do\n a mfmsr because mfmsr is a privileged instruction.\n \n The solution here is to create a cpu attribute:\n \n /sys/devices/system/cpu/svm\n \n which will read 0 or 1 based on the S bit of the current CPU.\n \n Signed-off-by: Ryan Grimm \u003cgrimm@linux.vnet.ibm.com\u003e\n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820021326.6884-12-bauerman@linux.ibm.com\n\ncommit 256ba2c1689efd4f5383cf7ebe2f9970c198b79d\nAuthor: Ram Pai \u003clinuxram@us.ibm.com\u003e\nDate: Mon Aug 19 23:13:20 2019 -0300\n\n powerpc/pseries/svm: Unshare all pages before kexecing a new kernel\n \n A new kernel deserves a clean slate. Any pages shared with the hypervisor\n is unshared before invoking the new kernel. However there are exceptions.\n If the new kernel is invoked to dump the current kernel, or if there is a\n explicit request to preserve the state of the current kernel, unsharing\n of pages is skipped.\n \n NOTE: While testing crashkernel, make sure at least 256M is reserved for\n crashkernel. Otherwise SWIOTLB allocation will fail and crash kernel will\n fail to boot.\n \n Signed-off-by: Ram Pai \u003clinuxram@us.ibm.com\u003e\n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820021326.6884-11-bauerman@linux.ibm.com\n\ncommit d5394c059da9786043934bed474562cde124e7d3\nAuthor: Anshuman Khandual \u003ckhandual@linux.vnet.ibm.com\u003e\nDate: Mon Aug 19 23:13:19 2019 -0300\n\n powerpc/pseries/svm: Use shared memory for Debug Trace Log (DTL)\n \n Secure guests need to share the DTL buffers with the hypervisor. To that\n end, use a kmem_cache constructor which converts the underlying buddy\n allocated SLUB cache pages into shared memory.\n \n Signed-off-by: Anshuman Khandual \u003ckhandual@linux.vnet.ibm.com\u003e\n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820021326.6884-10-bauerman@linux.ibm.com\n\ncommit bd104e6db6f0ad124e507a9ecf1a468efe5697db\nAuthor: Anshuman Khandual \u003ckhandual@linux.vnet.ibm.com\u003e\nDate: Mon Aug 19 23:13:18 2019 -0300\n\n powerpc/pseries/svm: Use shared memory for LPPACA structures\n \n LPPACA structures need to be shared with the host. Hence they need to be in\n shared memory. Instead of allocating individual chunks of memory for a\n given structure from memblock, a contiguous chunk of memory is allocated\n and then converted into shared memory. Subsequent allocation requests will\n come from the contiguous chunk which will be always shared memory for all\n structures.\n \n While we are able to use a kmem_cache constructor for the Debug Trace Log,\n LPPACAs are allocated very early in the boot process (before SLUB is\n available) so we need to use a simpler scheme here.\n \n Introduce helper is_svm_platform() which uses the S bit of the MSR to tell\n whether we\u0027re running as a secure guest.\n \n Signed-off-by: Anshuman Khandual \u003ckhandual@linux.vnet.ibm.com\u003e\n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820021326.6884-9-bauerman@linux.ibm.com\n\ncommit e311a92da18cbdd4972dab0cda88b1b8484b8fef\nAuthor: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\nDate: Mon Aug 19 23:13:17 2019 -0300\n\n powerpc/pseries: Add and use LPPACA_SIZE constant\n \n Helps document what the hard-coded number means.\n \n Also take the opportunity to fix an #endif comment.\n \n Suggested-by: Alexey Kardashevskiy \u003caik@linux.ibm.com\u003e\n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820021326.6884-8-bauerman@linux.ibm.com\n\ncommit 7f70c3815a352f76d69fc5642ae9685033a428c0\nAuthor: Sukadev Bhattiprolu \u003csukadev@linux.vnet.ibm.com\u003e\nDate: Mon Aug 19 23:13:16 2019 -0300\n\n powerpc: Introduce the MSR_S bit\n \n Protected Execution Facility (PEF) is an architectural change for\n POWER 9 that enables Secure Virtual Machines (SVMs). When enabled,\n PEF adds a new higher privileged mode, called Ultravisor mode, to\n POWER architecture.\n \n The hardware changes include the following:\n \n * There is a new bit in the MSR that determines whether the current\n process is running in secure mode, MSR(S) bit 41. MSR(S)\u003d1, process\n is in secure mode, MSR(s)\u003d0 process is in normal mode.\n \n * The MSR(S) bit can only be set by the Ultravisor.\n \n * HRFID cannot be used to set the MSR(S) bit. If the hypervisor needs\n to return to a SVM it must use an ultracall. It can determine if\n the VM it is returning to is secure.\n \n * The privilege of a process is now determined by three MSR bits,\n MSR(S, HV, PR). In each of the tables below the modes are listed\n from least privilege to highest privilege. The higher privilege\n modes can access all the resources of the lower privilege modes.\n \n **Secure Mode MSR Settings**\n \n +---+---+---+---------------+\n | S | HV| PR|Privilege |\n +\u003d\u003d\u003d+\u003d\u003d\u003d+\u003d\u003d\u003d+\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d+\n | 1 | 0 | 1 | Problem |\n +---+---+---+---------------+\n | 1 | 0 | 0 | Privileged(OS)|\n +---+---+---+---------------+\n | 1 | 1 | 0 | Ultravisor |\n +---+---+---+---------------+\n | 1 | 1 | 1 | Reserved |\n +---+---+---+---------------+\n \n **Normal Mode MSR Settings**\n \n +---+---+---+---------------+\n | S | HV| PR|Privilege |\n +\u003d\u003d\u003d+\u003d\u003d\u003d+\u003d\u003d\u003d+\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d+\n | 0 | 0 | 1 | Problem |\n +---+---+---+---------------+\n | 0 | 0 | 0 | Privileged(OS)|\n +---+---+---+---------------+\n | 0 | 1 | 0 | Hypervisor |\n +---+---+---+---------------+\n | 0 | 1 | 1 | Problem (HV) |\n +---+---+---+---------------+\n \n Signed-off-by: Sukadev Bhattiprolu \u003csukadev@linux.vnet.ibm.com\u003e\n Signed-off-by: Ram Pai \u003clinuxram@us.ibm.com\u003e\n [ cclaudio: Update the commit message ]\n Signed-off-by: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820021326.6884-7-bauerman@linux.ibm.com\n\ncommit f7777e008cad17bf757ca256709070c07efd8283\nAuthor: Ram Pai \u003clinuxram@us.ibm.com\u003e\nDate: Mon Aug 19 23:13:15 2019 -0300\n\n powerpc/pseries/svm: Add helpers for UV_SHARE_PAGE and UV_UNSHARE_PAGE\n \n These functions are used when the guest wants to grant the hypervisor\n access to certain pages.\n \n Signed-off-by: Ram Pai \u003clinuxram@us.ibm.com\u003e\n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820021326.6884-6-bauerman@linux.ibm.com\n\ncommit 6a9c930bd7751bf0630d8b9b73b07af5c6842da6\nAuthor: Ram Pai \u003clinuxram@us.ibm.com\u003e\nDate: Mon Aug 19 23:13:14 2019 -0300\n\n powerpc/prom_init: Add the ESM call to prom_init\n \n Make the Enter-Secure-Mode (ESM) ultravisor call to switch the VM to secure\n mode. Pass kernel base address and FDT address so that the Ultravisor is\n able to verify the integrity of the VM using information from the ESM blob.\n \n Add \"svm\u003d\" command line option to turn on switching to secure mode.\n \n Signed-off-by: Ram Pai \u003clinuxram@us.ibm.com\u003e\n [ andmike: Generate an RTAS os-term hcall when the ESM ucall fails. ]\n Signed-off-by: Michael Anderson \u003candmike@linux.ibm.com\u003e\n [ bauerman: Cleaned up the code a bit. ]\n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820021326.6884-5-bauerman@linux.ibm.com\n\ncommit 528229d210781b2da66c6d257a326c21099982b5\nAuthor: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nDate: Mon Aug 19 23:13:13 2019 -0300\n\n powerpc: Add support for adding an ESM blob to the zImage wrapper\n \n For secure VMs, the signing tool will create a ticket called the \"ESM blob\"\n for the Enter Secure Mode ultravisor call with the signatures of the kernel\n and initrd among other things.\n \n This adds support to the wrapper script for adding that blob via the \"-e\"\n option to the zImage.pseries.\n \n It also adds code to the zImage wrapper itself to retrieve and if necessary\n relocate the blob, and pass its address to Linux via the device-tree, to be\n later consumed by prom_init.\n \n Signed-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n [ bauerman: Minor adjustments to some comments. ]\n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820021326.6884-4-bauerman@linux.ibm.com\n\ncommit 136bc0397ae21dbf63ca02e5775ad353a479cd2f\nAuthor: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\nDate: Mon Aug 19 23:13:12 2019 -0300\n\n powerpc/pseries: Introduce option to build secure virtual machines\n \n Introduce CONFIG_PPC_SVM to control support for secure guests and include\n Ultravisor-related helpers when it is selected\n \n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820021326.6884-3-bauerman@linux.ibm.com\n\ncommit 8eb3cef8d2642da6b72179da73344a442461cb58\nAuthor: Ioana Radulescu \u003cruxandra.radulescu@nxp.com\u003e\nDate: Wed Aug 28 17:08:15 2019 +0300\n\n dpaa2-eth: Add pause frame support\n \n Starting with firmware version MC10.18.0, we have support for\n L2 flow control. Asymmetrical configuration (Rx or Tx only) is\n supported, but not pause frame autonegotioation.\n \n Pause frame configuration is done via ethtool. By default, we start\n with flow control enabled on both Rx and Tx. Changes are propagated\n to hardware through firmware commands, using two flags (PAUSE,\n ASYM_PAUSE) to specify Rx and Tx pause configuration, as follows:\n \n PAUSE | ASYM_PAUSE | Rx pause | Tx pause\n ----------------------------------------\n 0 | 0 | disabled | disabled\n 0 | 1 | disabled | enabled\n 1 | 0 | enabled | enabled\n 1 | 1 | enabled | disabled\n \n The hardware can automatically send pause frames when the number\n of buffers in the pool goes below a predefined threshold. Due to\n this, flow control is incompatible with Rx frame queue taildrop\n (both mechanisms target the case when processing of ingress\n frames can\u0027t keep up with the Rx rate; for large frames, the number\n of buffers in the pool may never get low enough to trigger pause\n frames as long as taildrop is enabled). So we set pause frame\n generation and Rx FQ taildrop as mutually exclusive.\n \n Signed-off-by: Ioana Radulescu \u003cruxandra.radulescu@nxp.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit cce62943c08ef9fcf3880d7babf76f9e86c9cdbd\nAuthor: Ioana Radulescu \u003cruxandra.radulescu@nxp.com\u003e\nDate: Wed Aug 28 17:08:14 2019 +0300\n\n dpaa2-eth: Use stored link settings\n \n Whenever a link state change occurs, we get notified and save\n the new link settings in the device\u0027s private data. In ethtool\n get_link_ksettings, use the stored state instead of interrogating\n the firmware each time.\n \n Signed-off-by: Ioana Radulescu \u003cruxandra.radulescu@nxp.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f7fe7e3d19e8d529f33aec6b9d824156e1a6e0fc\nAuthor: Ioana Radulescu \u003cruxandra.radulescu@nxp.com\u003e\nDate: Wed Aug 28 17:08:13 2019 +0300\n\n dpaa2-eth: Remove support for changing link settings\n \n We only support fixed-link for now, so there is no point in\n offering users the option to change link settings via ethtool.\n \n Functionally there is no change, since firmware prevents us from\n changing link parameters anyway.\n \n Signed-off-by: Ioana Radulescu \u003cruxandra.radulescu@nxp.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 68e0aa8ec5cedec48dd5b11df84afc956c8f85be\nAuthor: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\nDate: Wed Aug 28 23:05:21 2019 +1000\n\n powerpc/powernv: Add ultravisor message log interface\n \n The ultravisor (UV) provides an in-memory console which follows the\n OPAL in-memory console structure.\n \n This patch extends the OPAL msglog code to initialize the UV memory\n console and provide the \"/sys/firmware/ultravisor/msglog\" interface\n for userspace to view the UV message log.\n \n Signed-off-by: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Tested-by: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190828130521.26764-2-mpe@ellerman.id.au\n\ncommit dea45ea7775240047f70e2631e468f6b65d09493\nAuthor: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\nDate: Wed Aug 28 23:05:20 2019 +1000\n\n powerpc/powernv/opal-msglog: Refactor memcons code\n \n This patch refactors the code in opal-msglog that operates on the OPAL\n memory console in order to make it cleaner and also allow the reuse of\n the new memcons_* functions.\n \n Signed-off-by: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Tested-by: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190828130521.26764-1-mpe@ellerman.id.au\n\ncommit 6c85b7bc637b64e681760f62c0eafba2f56745c6\nAuthor: Sukadev Bhattiprolu \u003csukadev@linux.vnet.ibm.com\u003e\nDate: Thu Aug 22 00:48:38 2019 -0300\n\n powerpc/kvm: Use UV_RETURN ucall to return to ultravisor\n \n When an SVM makes an hypercall or incurs some other exception, the\n Ultravisor usually forwards (a.k.a. reflects) the exceptions to the\n Hypervisor. After processing the exception, Hypervisor uses the\n UV_RETURN ultracall to return control back to the SVM.\n \n The expected register state on entry to this ultracall is:\n \n * Non-volatile registers are restored to their original values.\n * If returning from an hypercall, register R0 contains the return value\n (unlike other ultracalls) and, registers R4 through R12 contain any\n output values of the hypercall.\n * R3 contains the ultracall number, i.e UV_RETURN.\n * If returning with a synthesized interrupt, R2 contains the\n synthesized interrupt number.\n \n Thanks to input from Paul Mackerras, Ram Pai and Mike Anderson.\n \n Signed-off-by: Sukadev Bhattiprolu \u003csukadev@linux.vnet.ibm.com\u003e\n Signed-off-by: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\n Acked-by: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190822034838.27876-8-cclaudio@linux.ibm.com\n\ncommit 512a5a6452b6d742b6f713184414d28cb6413080\nAuthor: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\nDate: Thu Aug 22 00:48:37 2019 -0300\n\n powerpc/powernv: Access LDBAR only if ultravisor disabled\n \n LDBAR is a per-thread SPR populated and used by the thread-imc pmu\n driver to dump the data counter into memory. It contains memory along\n with few other configuration bits. LDBAR is populated and enabled only\n when any of the thread imc pmu events are monitored.\n \n In ultravisor enabled systems, LDBAR becomes ultravisor privileged and\n an attempt to write to it will cause a Hypervisor Emulation Assistance\n interrupt.\n \n In ultravisor enabled systems, the ultravisor is responsible to maintain\n the LDBAR (e.g. save and restore it).\n \n This restricts LDBAR access to only when ultravisor is disabled.\n \n Signed-off-by: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\n Reviewed-by: Ram Pai \u003clinuxram@us.ibm.com\u003e\n Reviewed-by: Ryan Grimm \u003cgrimm@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190822034838.27876-7-cclaudio@linux.ibm.com\n\ncommit 5223134029a87db925ecc9449f9501bad391a52e\nAuthor: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\nDate: Thu Aug 22 00:48:36 2019 -0300\n\n powerpc/mm: Write to PTCR only if ultravisor disabled\n \n In ultravisor enabled systems, PTCR becomes ultravisor privileged only\n for writing and an attempt to write to it will cause a Hypervisor\n Emulation Assitance interrupt.\n \n This patch uses the set_ptcr_when_no_uv() function to restrict PTCR\n writing to only when ultravisor is disabled.\n \n Signed-off-by: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190822034838.27876-6-cclaudio@linux.ibm.com\n\ncommit 139a1d2842ec181cf017502a46bb8d947682a960\nAuthor: Michael Anderson \u003candmike@linux.ibm.com\u003e\nDate: Thu Aug 22 00:48:35 2019 -0300\n\n powerpc/mm: Use UV_WRITE_PATE ucall to register a PATE\n \n When Ultravisor (UV) is enabled, the partition table is stored in secure\n memory and can only be accessed via the UV. The Hypervisor (HV) however\n maintains a copy of the partition table in normal memory to allow Nest MMU\n translations to occur (for normal VMs). The HV copy includes partition\n table entries (PATE)s for secure VMs which would currently be unused\n (Nest MMU translations cannot access secure memory) but they would be\n needed as we add functionality.\n \n This patch adds the UV_WRITE_PATE ucall which is used to update the PATE\n for a VM (both normal and secure) when Ultravisor is enabled.\n \n Signed-off-by: Michael Anderson \u003candmike@linux.ibm.com\u003e\n Signed-off-by: Madhavan Srinivasan \u003cmaddy@linux.vnet.ibm.com\u003e\n Signed-off-by: Ram Pai \u003clinuxram@us.ibm.com\u003e\n [ cclaudio: Write the PATE in HV\u0027s table before doing that in UV\u0027s ]\n Signed-off-by: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\n Reviewed-by: Ryan Grimm \u003cgrimm@linux.vnet.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190822034838.27876-5-cclaudio@linux.ibm.com\n\ncommit bb04ffe85eebebd64d5e673a9434d968e80f3aa1\nAuthor: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\nDate: Thu Aug 22 00:48:34 2019 -0300\n\n powerpc/powernv: Introduce FW_FEATURE_ULTRAVISOR\n \n In PEF enabled systems, some of the resources which were previously\n hypervisor privileged are now ultravisor privileged and controlled by\n the ultravisor firmware.\n \n This adds FW_FEATURE_ULTRAVISOR to indicate if PEF is enabled.\n \n The host kernel can use FW_FEATURE_ULTRAVISOR, for instance, to skip\n accessing resources (e.g. PTCR and LDBAR) in case PEF is enabled.\n \n Signed-off-by: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\n [ andmike: Device node name to \"ibm,ultravisor\" ]\n Signed-off-by: Michael Anderson \u003candmike@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190822034838.27876-4-cclaudio@linux.ibm.com\n\ncommit a49dddbdb0cca1d00fc9251e543a0aac09a6a65b\nAuthor: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\nDate: Thu Aug 22 00:48:33 2019 -0300\n\n powerpc/kernel: Add ucall_norets() ultravisor call handler\n \n The ultracalls (ucalls for short) allow the Secure Virtual Machines\n (SVM)s and hypervisor to request services from the ultravisor such as\n accessing a register or memory region that can only be accessed when\n running in ultravisor-privileged mode.\n \n This patch adds the ucall_norets() ultravisor call handler.\n \n The specific service needed from an ucall is specified in register\n R3 (the first parameter to the ucall). Other parameters to the\n ucall, if any, are specified in registers R4 through R12.\n \n Return value of all ucalls is in register R3. Other output values\n from the ucall, if any, are returned in registers R4 through R12.\n \n Each ucall returns specific error codes, applicable in the context\n of the ucall. However, like with the PowerPC Architecture Platform\n Reference (PAPR), if no specific error code is defined for a particular\n situation, then the ucall will fallback to an erroneous\n parameter-position based code. i.e U_PARAMETER, U_P2, U_P3 etc depending\n on the ucall parameter that may have caused the error.\n \n Every host kernel (powernv) needs to be able to do ucalls in case it\n ends up being run in a machine with ultravisor enabled. Otherwise, the\n kernel may crash early in boot trying to access ultravisor resources,\n for instance, trying to set the partition table entry 0. Secure guests\n also need to be able to do ucalls and its kernel may not have\n CONFIG_PPC_POWERNV\u003dy. For that reason, the ucall.S file is placed under\n arch/powerpc/kernel.\n \n If ultravisor is not enabled, the ucalls will be redirected to the\n hypervisor which must handle/fail the call.\n \n Thanks to inputs from Ram Pai and Michael Anderson.\n \n Signed-off-by: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190822034838.27876-3-cclaudio@linux.ibm.com\n\ncommit 250c6c31228d49f3b96855ec387cf37bbe7cb6a7\nAuthor: Sukadev Bhattiprolu \u003csukadev@linux.ibm.com\u003e\nDate: Thu Aug 22 00:48:32 2019 -0300\n\n Documentation/powerpc: Ultravisor API\n \n Protected Execution Facility (PEF) is an architectural change for\n POWER 9 that enables Secure Virtual Machines (SVMs). When enabled,\n PEF adds a new higher privileged mode, called Ultravisor mode, to POWER\n architecture. Along with the new mode there is new firmware called the\n Protected Execution Ultravisor (or Ultravisor for short).\n \n POWER 9 DD2.3 chips (PVR\u003d0x004e1203) or greater will be PEF-capable.\n \n Attached documentation provides an overview of PEF and defines the API\n for various interfaces that must be implemented in the Ultravisor\n firmware as well as in the KVM Hypervisor.\n \n Based on input from Mike Anderson, Thiago Bauermann, Claudio Carvalho,\n Ben Herrenschmidt, Guerney Hunt, Paul Mackerras.\n \n Signed-off-by: Sukadev Bhattiprolu \u003csukadev@linux.ibm.com\u003e\n Signed-off-by: Ram Pai \u003clinuxram@linux.ibm.com\u003e\n Signed-off-by: Guerney Hunt \u003cgdhh@linux.ibm.com\u003e\n Reviewed-by: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\n Reviewed-by: Michael Anderson \u003candmike@linux.ibm.com\u003e\n Reviewed-by: Thiago Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190822034838.27876-2-cclaudio@linux.ibm.com\n\ncommit 134cb3ab71c0736da0ec66d2cfd7f6ebba96c8cf\nAuthor: Maxiwell S. Garcia \u003cmaxiwell@linux.ibm.com\u003e\nDate: Thu Aug 29 12:50:21 2019 -0300\n\n Documentation/powerpc: Add ELF note documentation\n \n The ELF note documentation describes the types and descriptors to be\n used with the PowerPC namespace.\n \n Signed-off-by: Maxiwell S. Garcia \u003cmaxiwell@linux.ibm.com\u003e\n Signed-off-by: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190829155021.2915-3-maxiwell@linux.ibm.com\n\ncommit 70ed86f4de5bd74dd2d884dcd2f3275c4cfe665f\nAuthor: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\nDate: Thu Aug 29 12:50:20 2019 -0300\n\n powerpc: Add PowerPC Capabilities ELF note\n \n Add the PowerPC name and the PPC_ELFNOTE_CAPABILITIES type in the\n kernel binary ELF note. This type is a bitmap that can be used to\n advertise kernel capabilities to userland.\n \n This patch also defines PPCCAP_ULTRAVISOR_BIT as being the bit zero.\n \n Suggested-by: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\n Signed-off-by: Claudio Carvalho \u003ccclaudio@linux.ibm.com\u003e\n [ maxiwell: Define the \u0027PowerPC\u0027 type in the elfnote.h ]\n Signed-off-by: Maxiwell S. Garcia \u003cmaxiwell@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190829155021.2915-2-maxiwell@linux.ibm.com\n\ncommit a102f139aac54689eeb05883952742ae780159f3\nAuthor: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\nDate: Thu Aug 29 18:52:52 2019 +1000\n\n powerpc/powernv/ioda: Remove obsolete iommu_table_ops::exchange callbacks\n \n As now we have xchg_no_kill/tce_kill, these are not used anymore so\n remove them.\n \n Signed-off-by: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190829085252.72370-6-aik@ozlabs.ru\n\ncommit 021b7868113cdca56297f2356d0f6fe89985c285\nAuthor: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\nDate: Thu Aug 29 18:52:51 2019 +1000\n\n powerpc/pseries/iommu: Switch to xchg_no_kill\n \n This is the last implementation of iommu_table_ops::exchange() which\n we are about to remove.\n \n This implements xchg_no_kill() for pseries. Since it is paravirtual\n platform, the hypervisor does TCE invalidations and we do not have\n to deal with it here, hence no tce_kill() hook.\n \n Signed-off-by: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190829085252.72370-5-aik@ozlabs.ru\n\ncommit 650ab1e370cdb61ba5c7495006f4376e02374da0\nAuthor: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\nDate: Thu Aug 29 18:52:50 2019 +1000\n\n vfio/spapr_tce: Invalidate multiple TCEs at once\n \n Invalidating a TCE cache entry for each updated TCE is quite expensive.\n This makes use of the new iommu_table_ops::xchg_no_kill()/tce_kill()\n callbacks to bring down the time spent in mapping a huge guest DMA window.\n \n Signed-off-by: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190829085252.72370-4-aik@ozlabs.ru\n\ncommit 01b7d128b5a7f0d09626c090093ff44155f347c9\nAuthor: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\nDate: Thu Aug 29 18:52:49 2019 +1000\n\n KVM: PPC: Book3S: Invalidate multiple TCEs at once\n \n Invalidating a TCE cache entry for each updated TCE is quite expensive.\n This makes use of the new iommu_table_ops::xchg_no_kill()/tce_kill()\n callbacks to bring down the time spent in mapping a huge guest DMA window;\n roughly 20s to 10s for each guest\u0027s 100GB of DMA space.\n \n Signed-off-by: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\n Acked-by: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190829085252.72370-3-aik@ozlabs.ru\n\ncommit 35872480da47ec714fd9c4f2f3d2d83daf304851\nAuthor: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\nDate: Thu Aug 29 18:52:48 2019 +1000\n\n powerpc/powernv/ioda: Split out TCE invalidation from TCE updates\n \n At the moment updates in a TCE table are made by iommu_table_ops::exchange\n which update one TCE and invalidates an entry in the PHB/NPU TCE cache\n via set of registers called \"TCE Kill\" (hence the naming).\n Writing a TCE is a simple xchg() but invalidating the TCE cache is\n a relatively expensive OPAL call. Mapping a 100GB guest with PCI+NPU\n passed through devices takes about 20s.\n \n Thankfully we can do better. Since such big mappings happen at the boot\n time and when memory is plugged/onlined (i.e. not often), these requests\n come in 512 pages so we call call OPAL 512 times less which brings 20s\n from the above to less than 10s. Also, since TCE caches can be flushed\n entirely, calling OPAL for 512 TCEs helps skiboot [1] to decide whether\n to flush the entire cache or not.\n \n This implements 2 new iommu_table_ops callbacks:\n - xchg_no_kill() to update a single TCE with no TCE invalidation;\n - tce_kill() to invalidate multiple TCEs.\n This uses the same xchg_no_kill() callback for IODA1/2.\n \n This implements 2 new wrappers on top of the new callbacks similar to\n the existing iommu_tce_xchg().\n \n This does not use the new callbacks yet, the next patches will;\n so this should not cause any behavioral change.\n \n Signed-off-by: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190829085252.72370-2-aik@ozlabs.ru\n\ncommit 4f916593be9da38c5cf0d3a5c386b57beb70f422\nAuthor: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\nDate: Mon Aug 26 14:55:20 2019 +1000\n\n KVM: PPC: Book3S: Fix incorrect guest-to-user-translation error handling\n \n H_PUT_TCE_INDIRECT handlers receive a page with up to 512 TCEs from\n a guest. Although we verify correctness of TCEs before we do anything\n with the existing tables, there is a small window when a check in\n kvmppc_tce_validate might pass and right after that the guest alters\n the page with TCEs which can cause early exit from the handler and\n leave srcu_read_lock(\u0026vcpu-\u003ekvm-\u003esrcu) (virtual mode) or lock_rmap(rmap)\n (real mode) locked.\n \n This fixes the bug by jumping to the common exit code with an appropriate\n unlock.\n \n Fixes: 121f80ba68f1 (\"KVM: PPC: VFIO: Add in-kernel acceleration for VFIO\")\n Signed-off-by: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190826045520.92153-1-aik@ozlabs.ru\n\ncommit c4814af0b75cc6856f60e8a658d829000b156729\nAuthor: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nDate: Wed Aug 28 15:55:35 2019 -0700\n\n compiler_attributes.h: add note about __section\n \n GCC unescapes escaped string section names while Clang does not. Because\n __section uses the `#` stringification operator for the section name, it\n doesn\u0027t need to be escaped.\n \n Instead, we should:\n 1. Prefer __section(.section_name_no_quotes).\n 2. Only use __attribute__((__section__(\".section\"))) when creating the\n section name via C preprocessor (see the definition of __define_initcall\n in arch/um/include/shared/init.h).\n \n This antipattern was found with:\n $ grep -e __section\\(\\\" -e __section__\\(\\\" -r\n \n See the discussions in:\n Link: https://bugs.llvm.org/show_bug.cgi?id\u003d42950\n Link: https://marc.info/?l\u003dlinux-netdev\u0026m\u003d156412960619946\u0026w\u003d2\n Link: https://github.com/ClangBuiltLinux/linux/issues/619\n Tested-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Signed-off-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n [Improve a bit the formatting of the text to make it consistent]\n Signed-off-by: Miguel Ojeda \u003cmiguel.ojeda.sandonis@gmail.com\u003e\n\ncommit 5c75c72c1831efbfc4583182b723e72529a62d37\nAuthor: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nDate: Wed Aug 28 15:55:34 2019 -0700\n\n include/linux/compiler.h: remove unused KENTRY macro\n \n This macro is not used throughout the kernel. Delete it rather than\n update the __section to be a fully spelled out\n __attribute__((__section__())) to avoid\n https://bugs.llvm.org/show_bug.cgi?id\u003d42950.\n \n Tested-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Signed-off-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Miguel Ojeda \u003cmiguel.ojeda.sandonis@gmail.com\u003e\n\ncommit 08d54fc68c0aca5ed8fceffc3659ddea6e3b33ab\nAuthor: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nDate: Wed Aug 28 15:55:33 2019 -0700\n\n include/linux: prefer __section and __aligned from compiler_attributes.h\n \n GCC unescapes escaped string section names while Clang does not. Because\n __section uses the `#` stringification operator for the section name, it\n doesn\u0027t need to be escaped.\n \n Instead, we should:\n 1. Prefer __section(.section_name_no_quotes).\n 2. Only use __attribute__((__section__(\".section\"))) when creating the\n section name via C preprocessor (see the definition of __define_initcall\n in arch/um/include/shared/init.h).\n \n This antipattern was found with:\n $ grep -e __section\\(\\\" -e __section__\\(\\\" -r\n \n See the discussions in:\n Link: https://bugs.llvm.org/show_bug.cgi?id\u003d42950\n Link: https://marc.info/?l\u003dlinux-netdev\u0026m\u003d156412960619946\u0026w\u003d2\n Link: https://github.com/ClangBuiltLinux/linux/issues/619\n Acked-by: Will Deacon \u003cwill@kernel.org\u003e\n Reported-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Suggested-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Tested-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Signed-off-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n [Changed title to reflect the other changes]\n Signed-off-by: Miguel Ojeda \u003cmiguel.ojeda.sandonis@gmail.com\u003e\n\ncommit 1acab08505cba8577ffc2f74324cfb3e389f1b13\nAuthor: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nDate: Wed Aug 28 15:55:32 2019 -0700\n\n include/asm-generic: prefer __section from compiler_attributes.h\n \n GCC unescapes escaped string section names while Clang does not. Because\n __section uses the `#` stringification operator for the section name, it\n doesn\u0027t need to be escaped.\n \n Instead, we should:\n 1. Prefer __section(.section_name_no_quotes).\n 2. Only use __attribute__((__section__(\".section\"))) when creating the\n section name via C preprocessor (see the definition of __define_initcall\n in arch/um/include/shared/init.h).\n \n This antipattern was found with:\n $ grep -e __section\\(\\\" -e __section__\\(\\\" -r\n \n See the discussions in:\n Link: https://bugs.llvm.org/show_bug.cgi?id\u003d42950\n Link: https://marc.info/?l\u003dlinux-netdev\u0026m\u003d156412960619946\u0026w\u003d2\n Link: https://github.com/ClangBuiltLinux/linux/issues/619\n Acked-by: Naveen N. Rao \u003cnaveen.n.rao@linux.vnet.ibm.com\u003e\n Reported-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Suggested-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Tested-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Signed-off-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Miguel Ojeda \u003cmiguel.ojeda.sandonis@gmail.com\u003e\n\ncommit cfad365f03438ae35ca8cbab5a6b9a08dbf565e2\nAuthor: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nDate: Wed Aug 28 15:55:31 2019 -0700\n\n x86: prefer __section, __maybe_unused and __aligned from compiler_attributes.h\n \n GCC unescapes escaped string section names while Clang does not. Because\n __section uses the `#` stringification operator for the section name, it\n doesn\u0027t need to be escaped.\n \n Instead, we should:\n 1. Prefer __section(.section_name_no_quotes).\n 2. Only use __attribute__((__section__(\".section\"))) when creating the\n section name via C preprocessor (see the definition of __define_initcall\n in arch/um/include/shared/init.h).\n \n This antipattern was found with:\n $ grep -e __section\\(\\\" -e __section__\\(\\\" -r\n \n See the discussions in:\n Link: https://bugs.llvm.org/show_bug.cgi?id\u003d42950\n Link: https://marc.info/?l\u003dlinux-netdev\u0026m\u003d156412960619946\u0026w\u003d2\n Link: https://github.com/ClangBuiltLinux/linux/issues/619\n Reported-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Suggested-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Tested-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Signed-off-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n [Added back \"unused\" as __maybe_unused]\n [Changed title to reflect the other changes]\n Signed-off-by: Miguel Ojeda \u003cmiguel.ojeda.sandonis@gmail.com\u003e\n\ncommit f18b30a01e219fbeacc4b948fffec280bd4eba2b\nAuthor: John Pittman \u003cjpittman@redhat.com\u003e\nDate: Fri Aug 23 10:08:52 2019 -0400\n\n scsi: fnic: print port speed only at driver init or speed change\n \n Port speed printing was added by commit d948e6383ec3 (\"scsi: fnic: Add port\n speed stat to fnic debug stats\"). As currently configured, this will cause\n the port speed to be printed to syslog every 2 seconds. To prevent log\n spamming, only print the vnic port speed at driver initialization and if\n the speed changes. Also clean up a small typo in fnic_trace.c.\n \n Fixes: commit d948e6383ec3 (\"scsi: fnic: Add port speed stat to fnic debug stats\")\n Signed-off-by: John Pittman \u003cjpittman@redhat.com\u003e\n Reviewed-by: Satish Kharat \u003csatishkh@cisco.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 84769706dde5881b8da0dc3da666f7443be95a8a\nAuthor: zhengbin \u003czhengbin13@huawei.com\u003e\nDate: Fri Aug 23 21:22:53 2019 +0800\n\n scsi: bnx2fc: remove set but not used variables \u0027task\u0027,\u0027port\u0027,\u0027orig_task\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/scsi/bnx2fc/bnx2fc_hwi.c: In function bnx2fc_process_unsol_compl:\n drivers/scsi/bnx2fc/bnx2fc_hwi.c:636:30: warning: variable task set but not used [-Wunused-but-set-variable]\n drivers/scsi/bnx2fc/bnx2fc_hwi.c: In function bnx2fc_process_ofld_cmpl:\n drivers/scsi/bnx2fc/bnx2fc_hwi.c:1125:21: warning: variable port set but not used [-Wunused-but-set-variable]\n drivers/scsi/bnx2fc/bnx2fc_hwi.c: In function bnx2fc_init_seq_cleanup_task:\n drivers/scsi/bnx2fc/bnx2fc_hwi.c:1468:30: warning: variable orig_task set but not used [-Wunused-but-set-variable]\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: zhengbin \u003czhengbin13@huawei.com\u003e\n Acked-by: Saurav Kashyap \u003cskashyap@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 2f8eeaa258a9e8ac922f72957a5b0643c0d1b7f0\nAuthor: zhengbin \u003czhengbin13@huawei.com\u003e\nDate: Fri Aug 23 21:22:52 2019 +0800\n\n scsi: bnx2fc: remove set but not used variables \u0027lport\u0027,\u0027host\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/scsi/bnx2fc/bnx2fc_io.c: In function bnx2fc_initiate_seq_cleanup:\n drivers/scsi/bnx2fc/bnx2fc_io.c:932:19: warning: variable lport set but not used [-Wunused-but-set-variable]\n drivers/scsi/bnx2fc/bnx2fc_io.c: In function bnx2fc_initiate_cleanup:\n drivers/scsi/bnx2fc/bnx2fc_io.c:1001:19: warning: variable lport set but not used [-Wunused-but-set-variable]\n drivers/scsi/bnx2fc/bnx2fc_io.c: In function bnx2fc_process_scsi_cmd_compl:\n drivers/scsi/bnx2fc/bnx2fc_io.c:1882:20: warning: variable host set but not used [-Wunused-but-set-variable]\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: zhengbin \u003czhengbin13@huawei.com\u003e\n Acked-by: Saurav Kashyap \u003cskashyap@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 7444ad400bfd8c85a28678a7b68accc6c41041fd\nAuthor: zhengbin \u003czhengbin13@huawei.com\u003e\nDate: Fri Aug 23 21:22:51 2019 +0800\n\n scsi: bnx2fc: remove set but not used variable \u0027fh\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/scsi/bnx2fc/bnx2fc_fcoe.c: In function bnx2fc_rcv:\n drivers/scsi/bnx2fc/bnx2fc_fcoe.c:431:26: warning: variable fh set but not used [-Wunused-but-set-variable]\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: zhengbin \u003czhengbin13@huawei.com\u003e\n Acked-by: Saurav Kashyap \u003cskashyap@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 32dce48c054d345c8bce6ffe94aa39349fbf2147\nAuthor: Saurav Kashyap \u003cskashyap@marvell.com\u003e\nDate: Fri Aug 23 02:52:44 2019 -0700\n\n scsi: qedf: Update the version to 8.42.3.0\n \n Update the driver version to 8.42.3.0.\n \n Signed-off-by: Saurav Kashyap \u003cskashyap@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 3480e7a8c8e47a6ee101ab722c58d61761b67e4a\nAuthor: Saurav Kashyap \u003cskashyap@marvell.com\u003e\nDate: Fri Aug 23 02:52:43 2019 -0700\n\n scsi: qedf: Fix race betwen fipvlan request and response path\n \n There is a race b/w fipvlan request and response path:\n \n \u003d\u003d\u003d\u003d\u003d\n qedf_fcoe_process_vlan_resp:113]:2: VLAN response, vid\u003d0xffd.\n qedf_initiate_fipvlan_req:165]:2: vlan \u003d 0x6ffd already set.\n qedf_set_vlan_id:139]:2: Setting vlan_id\u003d0ffd prio\u003d3.\n \u003d\u003d\u003d\u003d\u003d\u003d\n \n The request thread sees that vlan is already set and fails to call\n ctrl_link_up.\n \n Fix:\n \n - While setting vlan_id use local variable and before setting vlan_id.\n \n - Call fcoe_ctlr_link_up in next iteration of fipvlan request.\n \n Signed-off-by: Saurav Kashyap \u003cskashyap@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 5797bcc897c363a775ac3d929b62d2579bdf8573\nAuthor: Hannes Reinecke \u003chare@suse.com\u003e\nDate: Fri Aug 23 02:52:42 2019 -0700\n\n scsi: qedf: Use discovery list to traverse rports\n \n The list of rports might become stale so we should rather traverse the\n discovery list when trying relogin.\n \n Signed-off-by: Hannes Reinecke \u003chare@suse.com\u003e\n Signed-off-by: Saurav Kashyap \u003cskashyap@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit f6d63678b07f935fefd7cfefe7b43f56bad274b6\nAuthor: Saurav Kashyap \u003cskashyap@marvell.com\u003e\nDate: Fri Aug 23 02:52:41 2019 -0700\n\n scsi: qedf: Decrease the LL2 MTU size to 2500\n \n Decrease the LL2 MTU size to 2500.\n \n Signed-off-by: Saurav Kashyap \u003cskashyap@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 0482262646e4e9e089c4e75f336244c5eaf360e1\nAuthor: Saurav Kashyap \u003cskashyap@marvell.com\u003e\nDate: Fri Aug 23 02:52:40 2019 -0700\n\n scsi: qedf: Check for module unloading bit before processing link update AEN\n \n Prevent race where we\u0027re removing the module and we get link update\n \n Signed-off-by: Saurav Kashyap \u003cskashyap@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit b29a907f105cc659a303bb2657959c7d0d188620\nAuthor: Saurav Kashyap \u003cskashyap@marvell.com\u003e\nDate: Fri Aug 23 02:52:39 2019 -0700\n\n scsi: qedf: Initiator fails to re-login to switch after link down\n \n Problem Statement:\n \n - Driver has fc_id of 0xcc0200\n \n - Driver gets link down (due to test) and calls fcoe_ctlr_link_down().\n \n - At this point, the fc_id of the initiator port is zeroed out.\n \n - Driver gets a link up 14 seconds later.\n \n - Driver performs FIP VLAN request, gets a response from the switch.\n \n - No change in VLAN is detected.\n \n - Driver then notifies libfcoe via fcoe_ctlr_link_up().\n \n - Libfcoe then issues a multicast discovery solicitation as expected.\n \n - Cisco FCF responds to that correctly.\n \n - Libfcoe at this point starts a 3 sec count-down to allow any other FCFs\n to be discovered. However, at this point, it has been 20 seconds since\n the last FKA from the driver (which would have been sent prior to\n backlink toggle), which causes the CVL to be issued from Cisco CVL from\n the switch is dropped by the driver as the vx_port identification\n descriptor is present and has value of 0xcc0200, which does not match\n the driver\u0027s value of 0. Libfcoe completes the 3 sec count down and\n proceeds to issue FLOGI as per protocol. Switch rejects FLogi request.\n All subsequent FLOGI requests from libfc are rejected by the switch\n (possibly because it is now expecting a new solicitation). This\n situation will continue until the next link toggle.\n \n Solution:\n \n The Vx_port descriptor in the CVL has three fields:\n \n MAC address\n Fabric ID\n Port Name\n \n Today, the code checks for both #1 and #2 above. In the case where we went\n through a link down, both these will be zero until FLOGI succeeds.\n \n We should change our code to check if any one of these 3 is valid and if\n so, handle the CVL (basically switching from AND to OR). The port name\n field is definitely expected to be valid always.\n \n Signed-off-by: Saurav Kashyap \u003cskashyap@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit aa5175a88cbb7eeab2aec874694e28a479d32bfe\nAuthor: Saurav Kashyap \u003cskashyap@marvell.com\u003e\nDate: Fri Aug 23 02:52:38 2019 -0700\n\n scsi: qedf: Add debug information for unsolicited processing\n \n Log s_id, d_id, type and command to the log message.\n \n [mkp: fixed warning]\n \n Signed-off-by: Saurav Kashyap \u003cskashyap@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 11988e393813f520974f69bf0e55c367a1cbac48\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 29 09:11:15 2019 +0100\n\n drm/i915/execlists: Try rearranging breadcrumb flush\n \n The addition of the DC_FLUSH failed to ensure sanctity of the post-sync\n write as CI immediately got a completion CS-event before the breadcrumb\n was coherent. So let\u0027s try the other idea of moving the post-sync write\n into the CS_STALL.\n \n References: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111514\n References: e8f6b4952ec5 (\"drm/i915/execlists: Flush the post-sync breadcrumb write harder\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Acked-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190829081150.10271-2-chris@chris-wilson.co.uk\n\ncommit 931285e0123c4f380a9e2ff06434c8307b92b726\nAuthor: Saurav Kashyap \u003cskashyap@marvell.com\u003e\nDate: Fri Aug 23 02:52:37 2019 -0700\n\n scsi: qedf: Add support for 20 Gbps speed\n \n The current code doeesn\u0027t support 20Gbps speed for current and supported\n speed. Add support for it.\n \n Signed-off-by: Saurav Kashyap \u003cskashyap@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit fa74f5e380b6d671735d72ab36646b781af3f876\nAuthor: Saurav Kashyap \u003cskashyap@marvell.com\u003e\nDate: Fri Aug 23 02:52:36 2019 -0700\n\n scsi: qedf: Interpret supported caps value correctly\n \n Driver was wrongly interpreting the supported cap value returned by qed.\n \n Solution: Use QED define macros instead of OS defined for interpreting\n supporting speeds.\n \n Signed-off-by: Saurav Kashyap \u003cskashyap@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 31696204c44ced38ca0a4b97069b4d8558c1483a\nAuthor: Saurav Kashyap \u003cskashyap@marvell.com\u003e\nDate: Fri Aug 23 02:52:35 2019 -0700\n\n scsi: qedf: Add shutdown callback handler\n \n Add shutdown callback handler.\n \n Signed-off-by: Saurav Kashyap \u003cskashyap@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit ec6350b800072db227585ee830672dc357f3983d\nAuthor: Nilesh Javali \u003cnjavali@marvell.com\u003e\nDate: Fri Aug 23 02:52:34 2019 -0700\n\n scsi: qedf: Update module description string\n \n Update module description.\n \n Signed-off-by: Nilesh Javali \u003cnjavali@marvell.com\u003e\n Signed-off-by: Saurav Kashyap \u003cskashyap@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 47aeee5549cf9326656a8f9190960dfd35c101e2\nAuthor: Arun Easi \u003caeasi@marvell.com\u003e\nDate: Fri Aug 23 02:52:33 2019 -0700\n\n scsi: qedf: Fix crash during sg_reset\n \n Driver was attempting to print cdb[0], which is not set for resets coming\n from SCSI ioctls. Check for cmd_len before accessing cmnd.\n \n Crash info:\n [84790.864747] BUG: unable to handle kernel NULL pointer dereference at (null)\n [84790.864783] IP: qedf_initiate_tmf+0x7a/0x6e0 [qedf]\n [84790.865204] Call Trace:\n [84790.865246] scsi_try_target_reset+0x2b/0x90 [scsi_mod]\n [84790.865266] scsi_ioctl_reset+0x20f/0x2a0 [scsi_mod]\n [84790.865284] scsi_ioctl+0x131/0x3a0 [scsi_mod]\n \n Signed-off-by: Arun Easi \u003caeasi@marvell.com\u003e\n Signed-off-by: Saurav Kashyap \u003cskashyap@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 4909e153e608360702ad34f76f72190bfaec3d9d\nAuthor: Saurav Kashyap \u003cskashyap@marvell.com\u003e\nDate: Fri Aug 23 02:52:32 2019 -0700\n\n scsi: qedf: Stop sending fipvlan request on unload\n \n - On some setups fipvlan can be retried for long duration and the\n connection to switch was not there so it was not getting any reply.\n \n - During unload this thread was hanging.\n \n Problem Resolution:\n \n Check if unload is in progress, then quit from fipvlan thread.\n \n Signed-off-by: Saurav Kashyap \u003cskashyap@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit e82e6ff7779a76ea9b2161ee5fe2cdd9387cf263\nAuthor: Saurav Kashyap \u003cskashyap@marvell.com\u003e\nDate: Fri Aug 23 02:52:31 2019 -0700\n\n scsi: qedf: Print message during bailout conditions\n \n Print messages during exiting condition to help debugging.\n \n Signed-off-by: Saurav Kashyap \u003cskashyap@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 391a2417942358de4c794846a4a7dde095010258\nAuthor: Don Brace \u003cdon.brace@microsemi.com\u003e\nDate: Thu Aug 22 15:40:05 2019 -0500\n\n scsi: smartpqi: bump version\n \n Reviewed-by: Scott Benesh \u003cscott.benesh@microsemi.com\u003e\n Reviewed-by: Gerry Morong \u003cgerry.morong@microsemi.com\u003e\n Signed-off-by: Don Brace \u003cdon.brace@microsemi.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 48edb87801004f2b93420701ad253a058dd48d49\nAuthor: Don Brace \u003cdon.brace@microsemi.com\u003e\nDate: Thu Aug 22 15:39:58 2019 -0500\n\n scsi: smartpqi: update copyright\n \n Reviewed-by: Scott Benesh \u003cscott.benesh@microsemi.com\u003e\n Reviewed-by: Kevin Barnett \u003ckevin.barnett@microsemi.com\u003e\n Signed-off-by: Don Brace \u003cdon.brace@microsemi.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 63a7956ae1e8cdb7bd026b584aef7d505cb393ee\nAuthor: Gilbert Wu \u003cgilbert.wu@microsemi.com\u003e\nDate: Thu Aug 22 15:39:51 2019 -0500\n\n scsi: smartpqi: add new pci ids\n \n Add:\n \n PM8222 VID_9005, DID_028F, SVID_1BD4 and SDID_004F\n 3101E-4i (1G, no GB) VID_9005, DID_028F, SVID_9005 and SDID_0808\n 3102E-8i (2G, no GB) VID_9005, DID_028F, SVID_9005 and SDID_0809\n \n Reviewed-by: Scott Benesh \u003cscott.benesh@microsemi.com\u003e\n Reviewed-by: Kevin Barnett \u003ckevin.barnett@microsemi.com\u003e\n Signed-off-by: Gilbert Wu \u003cgilbert.wu@microsemi.com\u003e\n Signed-off-by: Don Brace \u003cdon.brace@microsemi.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 9946a3987248353c32c94f3b5b13070aee0a1a6b\nAuthor: Murthy Bhat \u003cMurthy.Bhat@microsemi.com\u003e\nDate: Thu Aug 22 15:39:44 2019 -0500\n\n scsi: smartpqi: correct REGNEWD return status\n \n Return -EINPROGRESS when a rescan worker is queued.\n \n Reviewed-by: Scott Benesh \u003cscott.benesh@microsemi.com\u003e\n Reviewed-by: Kevin Barnett \u003ckevin.barnett@microsemi.com\u003e\n Signed-off-by: Murthy Bhat \u003cMurthy.Bhat@microsemi.com\u003e\n Signed-off-by: Don Brace \u003cdon.brace@microsemi.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 71ecc60d9ba47ee5af5a1d0f33cc92755995fb78\nAuthor: Gilbert Wu \u003cgilbert.wu@microsemi.com\u003e\nDate: Thu Aug 22 15:39:38 2019 -0500\n\n scsi: smartpqi: add gigabyte controller\n \n Reviewed-by: Scott Benesh \u003cscott.benesh@microsemi.com\u003e\n Reviewed-by: Kevin Barnett \u003ckevin.barnett@microsemi.com\u003e\n Signed-off-by: Gilbert Wu \u003cgilbert.wu@microsemi.com\u003e\n Signed-off-by: Don Brace \u003cdon.brace@microsemi.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 530dd8a7a2b1ff2ee6c862ba250ca54ccc8c8c3e\nAuthor: Mahesh Rajashekhara \u003cmahesh.rajashekhara@microsemi.com\u003e\nDate: Thu Aug 22 15:39:31 2019 -0500\n\n scsi: smartpqi: correct hang when deleting 32 lds\n \n When each ld is deleted, a rescan event is triggered in the driver. These\n can stack up waiting on mutex_lock.\n \n Change to mutex_try_lock and schedule a rescan for later.\n \n Reviewed-by: Scott Benesh \u003cscott.benesh@microsemi.com\u003e\n Reviewed-by: Kevin Barnett \u003ckevin.barnett@microsemi.com\u003e\n Signed-off-by: Mahesh Rajashekhara \u003cmahesh.rajashekhara@microsemi.com\u003e\n Signed-off-by: Don Brace \u003cdon.brace@microsemi.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 2d2ad4bc724e35459a19bbf77432facb9ac23f80\nAuthor: Gilbert Wu \u003cgilbert.wu@microsemi.com\u003e\nDate: Thu Aug 22 15:39:25 2019 -0500\n\n scsi: smartpqi: add bay identifier\n \n Return identify physical device \"Phys_Bay_in_Box\" as bay_identifier.\n \n Reviewed-by: Scott Benesh \u003cscott.benesh@microsemi.com\u003e\n Reviewed-by: Kevin Barnett \u003ckevin.barnett@microsemi.com\u003e\n Signed-off-by: Gilbert Wu \u003cgilbert.wu@microsemi.com\u003e\n Signed-off-by: Don Brace \u003cdon.brace@microsemi.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 6d90615f1346aed3674ebf825d1a6eda722e2e29\nAuthor: Murthy Bhat \u003cMurthy.Bhat@microsemi.com\u003e\nDate: Thu Aug 22 15:39:18 2019 -0500\n\n scsi: smartpqi: add sysfs entries\n \n - serial number\n - model\n - vendor\n \n Reviewed-by: Scott Benesh \u003cscott.benesh@microsemi.com\u003e\n Reviewed-by: Kevin Barnett \u003ckevin.barnett@microsemi.com\u003e\n Signed-off-by: Murthy Bhat \u003cMurthy.Bhat@microsemi.com\u003e\n Signed-off-by: Don Brace \u003cdon.brace@microsemi.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 522bc026f01381a91e65ed343781b1656f5e659a\nAuthor: Dave Carroll \u003cdavid.carroll@microsemi.com\u003e\nDate: Thu Aug 22 15:39:11 2019 -0500\n\n scsi: smartpqi: add module param to hide vsep\n \n Reviewed-by: Scott Benesh \u003cscott.benesh@microsemi.com\u003e\n Reviewed-by: Kevin Barnett \u003ckevin.barnett@microsemi.com\u003e\n Signed-off-by: Dave Carroll \u003cdavid.carroll@microsemi.com\u003e\n Signed-off-by: Don Brace \u003cdon.brace@microsemi.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 8bdb3b9c67c53eaeac2b97a6f0d2f86ccd3e5ba8\nAuthor: Gilbert Wu \u003cgilbert.wu@microsemi.com\u003e\nDate: Thu Aug 22 15:39:04 2019 -0500\n\n scsi: smartpqi: add pci ids for fiberhome controller\n \n Reviewed-by: Scott Benesh \u003cscott.benesh@microsemi.com\u003e\n Reviewed-by: Kevin Barnett \u003ckevin.barnett@microsemi.com\u003e\n Signed-off-by: Gilbert Wu \u003cgilbert.wu@microsemi.com\u003e\n Signed-off-by: Don Brace \u003cdon.brace@microsemi.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 5e6a9760f7da4dd86cca43ac6423695d6cb0dff4\nAuthor: Gilbert Wu \u003cgilbert.wu@microsemi.com\u003e\nDate: Thu Aug 22 15:38:58 2019 -0500\n\n scsi: smartpqi: add module param for exposure order\n \n Expose physical devices before logical devices.\n \n Reviewed-by: Scott Benesh \u003cscott.benesh@microsemi.com\u003e\n Reviewed-by: Kevin Barnett \u003ckevin.barnett@microsemi.com\u003e\n Signed-off-by: Gilbert Wu \u003cgilbert.wu@microsemi.com\u003e\n Signed-off-by: Don Brace \u003cdon.brace@microsemi.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 01f2ef6d18e70c92bbf3743be4b908f52a059263\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 28 16:19:11 2019 -0700\n\n scsi: lpfc: fix 12.4.0.0 GPF at boot\n \n The 12.4.0.0 patch that merged WQ/CQ pairs into single per-cpu pair\n contained a bug: a local variable was set to the queue pair by index. This\n should have allowed the local variable to be natively used. Instead, the\n code reused the index relative to the local variable, obtaining a random\n pointer value that when used eventually faulted the system\n \n Convert offending code to use local variable.\n \n Fixes: c00f62e6c546 (\"scsi: lpfc: Merge per-protocol WQ/CQ pairs into single per-cpu pair\")\n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Tested-by: Abdul Haleem \u003cabdhalee@linux.vnet.ibm.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 0622800d2ebccead42b3a85e255f7d473a36ec99\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Tue Aug 27 14:28:23 2019 -0700\n\n scsi: lpfc: Raise config max for lpfc_fcp_mq_threshold variable\n \n Raise the config max for lpfc_fcp_mq_threshold variable to 256.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n CC: Hannes Reinecke \u003chare@suse.de\u003e\n Reviewed-by: Hannes Reinecke \u003chare@suse.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 9db6c14c36fc8f7824ade95f53bcdad43df678b4\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Tue Aug 27 14:28:05 2019 -0700\n\n scsi: lpfc: Remove bg debugfs buffers\n \n Capturing and downloading dif command data and dif data was done a dozen\n years ago and no longer being used. Also creates a potential security hole.\n \n Remove the debugfs buffer for dif debugging.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n CC: KyleMahlkuch \u003ckmahlkuc@linux.vnet.ibm.com\u003e\n CC: Hannes Reinecke \u003chare@suse.de\u003e\n Reviewed-by: Hannes Reinecke \u003chare@suse.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 7f9989bace91a7a36051d64cbf46f7760c52bbf0\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Tue Aug 27 14:27:46 2019 -0700\n\n scsi: lpfc: Resolve checker warning for lpfc_new_io_buf()\n \n Per Dan Carpenter:\n \n The patch d79c9e9d4b3d: \"scsi: lpfc: Support dynamic unbounded SGL lists on\n G7 hardware.\" from Aug 14, 2019, leads to the following static checker\n warning:\n \n drivers/scsi/lpfc/lpfc_init.c:4107 lpfc_new_io_buf()\n error: not allocating enough data 784 vs 768\n \n There was no need to compare sizes nor to allocate size based on a define.\n \n Change allocation to use actual structure length\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n CC: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 844b17d7438bc6e8acbd85163b7c6b9bce0a6ea2\nAuthor: zhengbin \u003czhengbin13@huawei.com\u003e\nDate: Mon Aug 26 09:15:49 2019 +0800\n\n scsi: ufs: remove set but not used variable \u0027val\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/scsi/ufs/ufs-qcom.c: In function ufs_qcom_pwr_change_notify:\n drivers/scsi/ufs/ufs-qcom.c:808:6: warning: variable val set but not used [-Wunused-but-set-variable]\n \n Fixes: 1e1e465c6d23 (\"scsi/ufs: qcom: Remove ufs_qcom_phy_*() calls from host\")\n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: zhengbin \u003czhengbin13@huawei.com\u003e\n Acked-by: Avri Altman \u003cAvri.Altman@wdc.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 401fe8e99c71a51b61d24f237cb7c54842b9c662\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Thu Aug 22 22:56:51 2019 +0100\n\n scsi: qla2xxx: fix spelling mistake \"initializatin\" -\u003e \"initialization\"\n \n There is a spelling mistake in a ql_log message. Fix it.\n \n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 328bc6debf3dcaf8859dd1323882e8e24ec6e3f8\nAuthor: zhengbin \u003czhengbin13@huawei.com\u003e\nDate: Thu Aug 22 22:20:47 2019 +0800\n\n scsi: hisi_sas: remove set but not used variable \u0027irq_value\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/scsi/hisi_sas/hisi_sas_v1_hw.c: In function cq_interrupt_v1_hw:\n drivers/scsi/hisi_sas/hisi_sas_v1_hw.c:1542:6: warning: variable irq_value set but not used [-Wunused-but-set-variable]\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: zhengbin \u003czhengbin13@huawei.com\u003e\n Acked-by: John Garry \u003cjohn.garry@huawei.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 8dc8d29a00e29c679f027102fb9c6709347ca8ca\nAuthor: Sreekanth Reddy \u003csreekanth.reddy@broadcom.com\u003e\nDate: Thu Aug 22 02:19:01 2019 -0400\n\n scsi: mpt3sas: Introduce module parameter to override queue depth\n \n This patch provides a module parameter and sysfs interface to select\n whether the queue depth for each device should be based on the\n protocol-specific value set by the driver (the default) or the maximum\n supported by the controller (can_queue).\n \n Although we have a sysfs interface per sdev to change the queue depth\n of individual scsi devices, this implementation provides a single\n sysfs entry per shost to switch between the controller max and the\n driver default.\n \n [mkp: tweaked commit desc]\n \n Signed-off-by: Sreekanth Reddy \u003csreekanth.reddy@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit dd93b143706c636d3dfefe6cf355065d748b2473\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Mon Aug 19 19:18:36 2019 -0700\n\n scsi: qla2xxx: Fix a recently introduced kernel warning\n \n According to the firmware documentation a status type 0 IOCB can be\n followed by one or more status continuation type 0 IOCBs. Hence do not\n complain if the completion function is not called from inside the status\n type 0 IOCB handler.\n \n WARNING: CPU: 10 PID: 425 at drivers/scsi/qla2xxx/qla_isr.c:2784\n qla2x00_status_entry.isra.7+0x484/0x17b0 [qla2xxx]\n CPU: 10 PID: 425 Comm: kworker/10:1 Tainted: G E 5.3.0-rc4-next-20190813-autotest-autotest #1\n Workqueue: qla2xxx_wq qla25xx_free_rsp_que [qla2xxx]\n Call Trace:\n qla2x00_status_entry.isra.7+0x1484/0x17b0 [qla2xxx] (unreliable)\n qla24xx_process_response_queue+0x7d8/0xbd0 [qla2xxx]\n qla25xx_free_rsp_que+0x1a0/0x220 [qla2xxx]\n process_one_work+0x25c/0x520\n worker_thread+0x8c/0x5e0\n kthread+0x154/0x1a0\n ret_from_kernel_thread+0x5c/0x7c\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Cc: Abdul Haleem \u003cabdhalee@linux.vnet.ibm.com\u003e\n Reported-by: Abdul Haleem \u003cabdhalee@linux.vnet.ibm.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Abdul Haleem \u003cabdhalee@linux.vnet.ibm.com\u003e\n Reviewed-by: Lee Duncan \u003clduncan@suse.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit b557217c8475f40bc765ee20ff6b3b9124c8a4fe\nAuthor: Stanley Chu \u003cstanley.chu@mediatek.com\u003e\nDate: Mon Aug 19 21:43:28 2019 +0800\n\n scsi: ufs: fix broken hba-\u003eoutstanding_tasks\n \n Currently bits in hba-\u003eoutstanding_tasks are cleared only after their\n corresponding task management commands are successfully done by\n __ufshcd_issue_tm_cmd().\n \n If timeout happens in a task management command, its corresponding bit in\n hba-\u003eoutstanding_tasks will not be cleared until next task management\n command with the same tag used successfully finishes.\n \n This is wrong and can lead to some issues, like power issue. For example,\n ufshcd_release() and ufshcd_gate_work() will do nothing if\n hba-\u003eoutstanding_tasks is not zero even if both UFS host and devices are\n actually idle.\n \n Solution is referred from error handling of device commands: bits in\n hba-\u003eoutstanding_tasks shall be cleared regardless of their execution\n results.\n \n Signed-off-by: Stanley Chu \u003cstanley.chu@mediatek.com\u003e\n Signed-off-by: Chun-Hung Wu \u003cchun-hung.wu@mediatek.com\u003e\n Reviewed-by: Avri Altman \u003cavri.altman@wdc.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit e8fd0b9f6f487820a05a46ce3ac4b5df11ad6799\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Thu Aug 15 10:14:54 2019 +0100\n\n scsi: fcoe: remove redundant call to skb_transport_header\n \n Pointer fh is being assigned a return value from the call to\n skb_transport_header however this value is never read and fh is being\n re-assigned immediately afterwards with a new value. Since there are\n side-effects from calling skb_transport_header the call is redundant and\n can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit e55f24ffa936e09673905a9176a8a9d22cf68d40\nAuthor: Anil Varughese \u003caniljoy@cadence.com\u003e\nDate: Tue Aug 13 08:42:50 2019 +0100\n\n scsi: ufs: Disable local LCC in .link_startup_notify() in Cadence UFS\n \n Some UFS devices have issues if LCC is enabled. So we are setting\n PA_LOCAL_TX_LCC_Enable to 0 before link startup which will make sure that\n both host and device TX LCC are disabled once link startup is completed.\n \n Signed-off-by: Anil Varughese \u003caniljoy@cadence.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit c8a144b250dcd5fe6df95b8d101e9e2fe4c16201\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Thu Aug 29 14:31:05 2019 -0700\n\n Input: i8042 - enable wakeup on a stable struct device\n \n We don\u0027t know when the device will be added with device_add() in\n serio_add_port() because serio_add_port() is called from a workqueue\n that this driver schedules by calling serio_register_port(). The best we\n can know is that the device will definitely not have been added yet when\n the start callback is called on the serio device.\n \n While it hasn\u0027t been shown to be a problem, proactively move the wakeup\n enabling calls to the start hook so that we don\u0027t race with the\n workqueue calling device_add(). This will avoid racy situations where\n code tries to add wakeup sysfs attributes for this device from\n dpm_sysfs_add() but the path in device_set_wakeup_capable() has already\n done so.\n \n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 5f6b4e1e09331938a818cf8f371025ae03dee069\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Wed Jul 31 23:11:52 2019 +0100\n\n scsi: bfa: remove redundant assignment to variable error\n \n Variable error is being initialized with a value that is never read and\n error is being re-assigned a little later on. The assignment is redundant\n and hence can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 66d9cec8a6c9f3eb1d81507d5f045e28272359fb\nAuthor: Manasi Navare \u003cmanasi.d.navare@intel.com\u003e\nDate: Wed Aug 28 15:47:01 2019 -0700\n\n drm/i915/display: Move the commit_tail() disable sequence to separate function\n \n Create a new function intel_commit_modeset_disables() consistent\n with the naming in drm atomic helpers and similar to the enable function.\n This helps better organize the disable sequence in atomic_commit_tail()\n \n No functional change\n \n v4:\n * Do not create a function pointer, just a function (Maarten)\n v3:\n * Rebase (Manasi)\n v2:\n * Create a helper for old_crtc_state disables (Lucas)\n \n Suggested-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Signed-off-by: Manasi Navare \u003cmanasi.d.navare@intel.com\u003e\n Reviewed-by: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190828224701.422-1-manasi.d.navare@intel.com\n\ncommit 0c84127102ee240e3039d6e78a3dba0afff2f810\nAuthor: Manasi Navare \u003cmanasi.d.navare@intel.com\u003e\nDate: Tue Aug 27 15:17:34 2019 -0700\n\n drm/i915/display: Rename update_crtcs() to commit_modeset_enables()\n \n This patch has no functional changes. This just renames the update_crtcs()\n hooks to commit_modeset_enables() to match the drm_atomic helper naming\n conventions.\n \n v2:\n * Rebase on drm-tip\n \n Suggested-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Reviewed-by: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Signed-off-by: Manasi Navare \u003cmanasi.d.navare@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190827221735.29351-2-manasi.d.navare@intel.com\n\ncommit 31444afb460ee627cb7800acee46548449c52368\nAuthor: Matthew Auld \u003cmatthew.auld@intel.com\u003e\nDate: Thu Aug 29 21:19:19 2019 +0100\n\n drm/i915: s/for_each_sgt_dma/for_each_sgt_daddr/\n \n The sg_table for our backing store might contain addresses from\n stolen-memory or in the future local-memory, at which point this is no\n longer a dma-iterator. As a consequence we should now break on NULL\n iter.sgp, instead of dmap \u003d\u003d 0 which is considered an invalid dma\n address.\n \n As a bonus, gcc much prefers this construct,\n \n Function old new delta\n gen8_ggtt_insert_entries 211 192 -19\n gen6_ggtt_insert_entries 292 262 -30\n i915_error_object_create 996 954 -42\n \n Signed-off-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190829201919.21493-1-matthew.auld@intel.com\n\ncommit e1c14c43395c54c213a6bd415ba2db50fe503598\nAuthor: Roman Li \u003cRoman.Li@amd.com\u003e\nDate: Thu Aug 8 16:26:44 2019 -0400\n\n drm/amdgpu: Enable DC on Renoir\n \n Enable DC support for renoir.\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Roman Li \u003cRoman.Li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 542816ff168d8a3159dedfd10a5579e7d04f2130\nAuthor: Roman Li \u003cRoman.Li@amd.com\u003e\nDate: Thu Aug 8 15:14:25 2019 -0400\n\n drm/amd/display: Add DCN2.1 changes to DML\n \n Hook up the DML changes for renoir.\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Roman Li \u003cRoman.Li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f82effc4e59260f6826b497e5158742a8f5b59df\nAuthor: Roman Li \u003cRoman.Li@amd.com\u003e\nDate: Thu Aug 8 15:11:37 2019 -0400\n\n drm/amd/display: Correct order of RV family clk managers for Renoir\n \n Need to check for renoir first.\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Roman Li \u003cRoman.Li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 07842d54b95a1c236ba2bc237e32eeac476fc967\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Fri Jul 26 17:21:03 2019 -0400\n\n drm/amd/display: add Renoir to kconfig\n \n Add a kconfig option to enable renoir.\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 25f9955b15298c7b3c4fe0f28f469ddb5125e93b\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Fri Jul 26 17:20:21 2019 -0400\n\n drm/amd/display: build dcn21 blocks\n \n Enable the building of dcn21 support.\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit aa91916770d07cafafd24905b20e5350eb2dee30\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Fri Jul 26 17:16:47 2019 -0400\n\n drm/amd/display: add dcn21 core DC changes\n \n Add missing parameters, to make dcn21 compile\n without errors\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit dd44a63386e052304e965a36902bb3ce69ff665a\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Fri Jul 26 17:14:34 2019 -0400\n\n drm/amd/display: add dal_asic_id for renoir\n \n Add the rev id for renoir.\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1b2c7b2c6d615791a38be33d9e2d32a2736e70cb\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Fri Jul 26 17:09:35 2019 -0400\n\n drm/amd/display: call update_bw_bounding_box\n \n call update_bw_bounding_box in DC construct\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 30221ad8758b087c5eef6a62d492c7f98b1cd401\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Fri Jul 26 17:08:45 2019 -0400\n\n drm/amd/display: Handle Renoir in amdgpu_dm (v2)\n \n Hook up renoir support to KMS.\n \n v2: squash in \"Fixes for Renoir in amdgpu_dm\"\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e22ece54ee782ac81838fb6a4b1a45c82e33461c\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Fri Jul 26 17:06:02 2019 -0400\n\n drm/amd/display: Handle Renoir in DC\n \n add Renoir DCN version in DC and handle it\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit aad37f26064aa55876287e73a8bfcaf3ce341d99\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Fri Jul 26 16:59:04 2019 -0400\n\n drm/amd/display: Fix register names\n \n rename VM_CONTEXT0 to MMVM_CONTEXT0 as that is the name defined in\n the register files\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b04641a3f4c54b00dab7ccd49fd45909c42c3fc2\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Fri Jul 26 16:52:06 2019 -0400\n\n drm/amd/display: Add Renoir DML\n \n DML provides the display configuration validation as provided\n by the hw teams.\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 64ce485c4828799f223de9b3f930bed081ad4ec8\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Fri Jul 26 16:47:47 2019 -0400\n\n drm/amd/display: Add Renoir GPIO\n \n Misc display related configuration details.\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6f4e6361c3ff8457d45d2a898c418e3495e85e93\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Fri Jul 26 16:46:46 2019 -0400\n\n drm/amd/display: Add Renoir resource (v2)\n \n Manages the renoir display resources (crtcs, phys, plls, etc.).\n \n v2: rebase (Alex)\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4edb6fc91878603f325c79314ee7675558932f7f\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Fri Jul 26 16:43:53 2019 -0400\n\n drm/amd/display: Add Renoir clock manager\n \n Controls display clocks and interfaces with powerplay for\n clock and power requirements.\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6f451b60e044b1a60f93087000b460e779f2da51\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Thu Jul 25 16:31:50 2019 -0400\n\n drm/amd/display: Add Renoir Hubbub (v2)\n \n Controls the display hw\u0027s interface to memory.\n \n v2: rebase (Alex)\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 35b82ba8f2fa9b4ba01df0551e6ad69099b8e6a3\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Wed Aug 28 10:22:02 2019 -0500\n\n drm/amd/display: Add Renoir hubbub registers list\n \n These are the registers used to program the hubbub hw.\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit eced51f9babbccac23b398edf7215de82f944b04\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Thu Jul 25 16:19:17 2019 -0400\n\n drm/amd/display: Add hubp block for Renoir (v2)\n \n This provides the interface to memory for the display hw.\n \n v2: minor cleanup (Alex)\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1e768c5b0f44ab061f56216a44cd6cc02764f987\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Thu Jul 25 16:00:15 2019 -0400\n\n drm/amd/display: Add Renoir irq_services (v2)\n \n Provides the interface to configure display interrrupts on renoir.\n \n v2: rebase fix (Alex)\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 82f9146832c60e5d1f0e4ed7ee82c7a18dff5cc1\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Thu Jul 25 15:58:18 2019 -0400\n\n drm/amd/display: Add pp_smu functions for Renoir\n \n This defines the interface for communicating requirements\n between DC and powerplay.\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ab61831227869b049227fa3c0f2d1ede095e1288\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Wed Aug 21 16:56:52 2019 -0500\n\n drm/amd/display: Add Renoir hw_seq register list\n \n These are the registers used to for the hw sequences\n for modesetting.\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ff54ecb095b549a8e79ee87922cecdba69f6bb79\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Thu Jul 25 15:51:09 2019 -0400\n\n drm/amd/display: Add Renoir clock registers list\n \n These are the registers used to program the clock hw.\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b593bce59bfa25d9abbf220b6614396ccd965b1b\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Thu Jul 25 15:51:41 2019 -0400\n\n drm/amd/display: Add Renoir registers (v3)\n \n add registers for dcn, clk, and renoir ip offsets\n \n v2: header cleanup (Alex)\n v3: Add DPCS registers (Hersen)\n \n Acked-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit eee3258e8f8be822a6d7c15faa3f4a9e3081381e\nAuthor: Prike Liang \u003cPrike.Liang@amd.com\u003e\nDate: Thu Aug 15 16:53:08 2019 +0800\n\n drm/amd/powerplay: add the interface for getting ultimate frequency v3\n \n add the get_dpm_ultimate_freq for supporting different swSMU.\n -v2:\n Handle the unsupported clock type and read smc message failed case and return error code.\n Move the smu12 uclk frequency retrieved logic to renoir ppt.\n -v3:\n Use goto clause to handle invalidate clk index.\n Add the limited tag for smu_get_dpm_uclk to avoid other likewise interface introduced.\n \n Signed-off-by: Prike Liang \u003cPrike.Liang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 296ae1038d43c54e0b0e7e4f398c867ba4beaaac\nAuthor: Prike Liang \u003cPrike.Liang@amd.com\u003e\nDate: Fri Aug 16 11:22:13 2019 +0800\n\n drm/amd/powerplay: enable populate DPM clocks table for swSMU APU\n \n Should populate DPM clocks tables during hw init,otherwise will\n suffer from invalidate table.\n \n Signed-off-by: Prike Liang \u003cPrike.Liang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ffe61cd6423c9e5db973eb9d9f35accaeec0d9d8\nAuthor: Prike Liang \u003cPrike.Liang@amd.com\u003e\nDate: Thu Aug 15 09:39:06 2019 +0800\n\n drm/amd/powerplay: regards the APU always enable the dpm feature mask\n \n There is no driver message to enable/disable feature mask for APU.\n For the sake of APU reusing swSMU interface and assume APU supports all\n the feature.\n \n Signed-off-by: Prike Liang \u003cPrike.Liang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 334ffd0daa4357e998f43a4f914903013294f350\nAuthor: Prike Liang \u003cPrike.Liang@amd.com\u003e\nDate: Mon Aug 26 16:46:34 2019 +0800\n\n drm/amdgpu: Initialize and update SDMA power gating\n \n Init SDMA HW base configuration and enable idle INT for rn.\n \n Signed-off-by: Prike Liang \u003cPrike.Liang@amd.com\u003e\n Reviewed-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 12842d02c7fde1989616a490799bf8f86695b54a\nAuthor: Tianci.Yin \u003ctianci.yin@amd.com\u003e\nDate: Wed Aug 28 10:03:40 2019 +0800\n\n drm/amdgpu/psp: keep TMR in visible vram region for SRIOV\n \n Fix compute ring test failure in sriov scenario.\n \n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Tianci.Yin \u003ctianci.yin@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 994dcfaa7ee321a23ec398f6b0e709f11ab2f577\nAuthor: Tianci.Yin \u003ctianci.yin@amd.com\u003e\nDate: Wed Aug 28 18:51:19 2019 +0800\n\n drm/amdgpu: keep the stolen memory in visible vram region\n \n stolen memory should be fixed in visible region.\n \n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Tianci.Yin \u003ctianci.yin@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 92ead9fa6fc554e0ce43a4a83b26fac71f9b96ba\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Thu Aug 29 01:51:56 2019 +0100\n\n drm/amdgpu: fix spelling mistake \"jumpimng\" -\u003e \"jumping\"\n \n There is a spelling mistake in a DRM_DEBUG_DRIVER debug message.\n Fix it.\n \n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 53fd9b5ae87893540bbcd81a69ad809a13db5b89\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Wed Aug 28 08:47:34 2019 -0500\n\n drm/amdgpu/virtual_dce: drop error message in hw_init\n \n No need to add new asic cases. This is a sw display\n implementation, so just drop the error message so when\n we add new asics, all we have to do is add the virtual\n dce IP module.\n \n Reviewed-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 77efe48a729588527afb4d5811b9e0acb29f5e51\nAuthor: Jean Delvare \u003cjdelvare@suse.de\u003e\nDate: Wed Aug 28 17:05:57 2019 +0200\n\n drm/amdgpu/si: fix ASIC tests\n \n Comparing adev-\u003efamily with CHIP constants is not correct.\n adev-\u003efamily can only be compared with AMDGPU_FAMILY constants and\n adev-\u003easic_type is the struct member to compare with CHIP constants.\n They are separate identification spaces.\n \n Signed-off-by: Jean Delvare \u003cjdelvare@suse.de\u003e\n Fixes: 62a37553414a (\"drm/amdgpu: add si implementation v10\")\n Cc: Ken Wang \u003cQingqing.Wang@amd.com\u003e\n Cc: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Cc: \"Christian König\" \u003cchristian.koenig@amd.com\u003e\n Cc: \"David (ChunMing) Zhou\" \u003cDavid1.Zhou@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1cdd229bec36212261f69702464913b5c2581d4d\nAuthor: Jean Delvare \u003cjdelvare@suse.de\u003e\nDate: Wed Aug 28 10:27:29 2019 +0200\n\n drm/amd/amdgpu: hide voltage and power sensors on SI and KV parts\n \n The driver does not support these sensors yet and there is no point in\n creating sysfs attributes which will always return an error.\n \n Signed-off-by: Jean Delvare \u003cjdelvare@suse.de\u003e\n Cc: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Cc: \"Christian König\" \u003cchristian.koenig@amd.com\u003e\n Cc: \"David (ChunMing) Zhou\" \u003cDavid1.Zhou@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e352625796b343cc85b7648a74a57e739b12dc27\nAuthor: Monk Liu \u003cMonk.Liu@amd.com\u003e\nDate: Tue Aug 27 16:32:55 2019 +0800\n\n drm/amdgpu: introduce vram lost for reset (v2)\n \n for SOC15/vega10 the BACO reset \u0026 mode1 would introduce vram lost\n in high end address range, current kmd\u0027s vram lost checking cannot\n catch it since it only check very ahead visible frame buffer\n \n v2:\n cover NV as well\n \n Signed-off-by: Monk Liu \u003cMonk.Liu@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 514ad79103f5d278cb6b190eb0c1f83461fa70a9\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Aug 27 14:26:08 2019 +0800\n\n drm/amd/powerplay: enable jpeg powergating for navi1x\n \n jpeg pg depends on vcn pg\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5ef3b8acdcdf7fb61a304390e49a6d7df8720314\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Aug 27 11:06:13 2019 +0800\n\n drm/amdgpu: enable athub powergating for navi12\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c1653ea05bcec582c82e7b134f5535d8fad42f48\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Aug 27 11:05:23 2019 +0800\n\n drm/amdgpu: enable vcn powergating for navi12\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 317f9cc97bdcfef1e873ba4dfa2b7af87a17775f\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Tue Aug 27 17:13:47 2019 +0800\n\n drm/amdgpu: correct in_suspend setting for navi series\n \n in_suspend flag should be set in amdgpu_device_suspend/resume in pairs,\n instead of gfx10 ip suspend/resume function.\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7b60422cb796d40431337becf2129fd9944b2f05\nAuthor: Dan Williams \u003cdan.j.williams@intel.com\u003e\nDate: Mon Aug 26 17:55:05 2019 -0700\n\n libnvdimm/security: Consolidate \u0027security\u0027 operations\n \n The security operations are exported from libnvdimm/security.c to\n libnvdimm/dimm_devs.c, and libnvdimm/security.c is optionally compiled\n based on the CONFIG_NVDIMM_KEYS config symbol.\n \n Rather than export the operations across compile objects, just move the\n __security_store() entry point to live with the helpers.\n \n Acked-by: Jeff Moyer \u003cjmoyer@redhat.com\u003e\n Reviewed-by: Dave Jiang \u003cdave.jiang@intel.com\u003e\n Link: https://lore.kernel.org/r/156686730515.184120.10522747907309996674.stgit@dwillia2-desk3.amr.corp.intel.com\n Signed-off-by: Dan Williams \u003cdan.j.williams@intel.com\u003e\n\ncommit bc4f2199ca3107809df96cf72f618b9559b00a21\nAuthor: Dan Williams \u003cdan.j.williams@intel.com\u003e\nDate: Mon Aug 26 17:55:00 2019 -0700\n\n libnvdimm/security: Tighten scope of nvdimm-\u003ebusy vs security operations\n \n An attempt to freeze DIMMs currently runs afoul of default blocking of\n all security operations in the entry to the \u0027store\u0027 routine for the\n \u0027security\u0027 sysfs attribute.\n \n The blanket blocking of all security operations while the DIMM is in\n active use in a region is too restrictive. The only security operations\n that need to be aware of the -\u003ebusy state are those that mutate the\n state of data, i.e. erase and overwrite.\n \n Refactor the -\u003ebusy checks to be applied at the entry common entry point\n in __security_store() rather than each of the helper routines to enable\n freeze to be run regardless of busy state.\n \n Reviewed-by: Dave Jiang \u003cdave.jiang@intel.com\u003e\n Reviewed-by: Jeff Moyer \u003cjmoyer@redhat.com\u003e\n Link: https://lore.kernel.org/r/156686729996.184120.3458026302402493937.stgit@dwillia2-desk3.amr.corp.intel.com\n Signed-off-by: Dan Williams \u003cdan.j.williams@intel.com\u003e\n\ncommit d78c620a2e824d7b01a6e991208a8aa2c938cabe\nAuthor: Dan Williams \u003cdan.j.williams@intel.com\u003e\nDate: Mon Aug 26 17:54:54 2019 -0700\n\n libnvdimm/security: Introduce a \u0027frozen\u0027 attribute\n \n In the process of debugging a system with an NVDIMM that was failing to\n unlock it was found that the kernel is reporting \u0027locked\u0027 while the DIMM\n security interface is \u0027frozen\u0027. Unfortunately the security state is\n tracked internally as an enum which prevents it from communicating the\n difference between \u0027locked\u0027 and \u0027locked + frozen\u0027. It follows that the\n enum also prevents the kernel from communicating \u0027unlocked + frozen\u0027\n which would be useful for debugging why security operations like \u0027change\n passphrase\u0027 are disabled.\n \n Ditch the security state enum for a set of flags and introduce a new\n sysfs attribute explicitly for the \u0027frozen\u0027 state. The regression risk\n is low because the \u0027frozen\u0027 state was already blocked behind the\n \u0027locked\u0027 state, but will need to revisit if there were cases where\n applications need \u0027frozen\u0027 to show up in the primary \u0027security\u0027\n attribute. The expectation is that communicating \u0027frozen\u0027 is mostly a\n helper for debug and status monitoring.\n \n Reviewed-by: Dave Jiang \u003cdave.jiang@intel.com\u003e\n Reported-by: Jeff Moyer \u003cjmoyer@redhat.com\u003e\n Reviewed-by: Jeff Moyer \u003cjmoyer@redhat.com\u003e\n Link: https://lore.kernel.org/r/156686729474.184120.5835135644278860826.stgit@dwillia2-desk3.amr.corp.intel.com\n Signed-off-by: Dan Williams \u003cdan.j.williams@intel.com\u003e\n\ncommit 2b90cb223320a93b1be6c2616efe6f9ff14d8b28\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Jun 10 16:06:13 2019 -0500\n\n libnvdimm, region: Use struct_size() in kzalloc()\n \n One of the more common cases of allocation size calculations is finding\n the size of a structure that has a zero-sized array at the end, along\n with memory for some number of elements for that array. For example:\n \n struct nd_region {\n ...\n struct nd_mapping mapping[0];\n };\n \n instance \u003d kzalloc(sizeof(struct nd_region) + sizeof(struct nd_mapping) *\n count, GFP_KERNEL);\n \n Instead of leaving these open-coded and prone to type mistakes, we can\n now use the new struct_size() helper:\n \n instance \u003d kzalloc(struct_size(instance, mapping, count), GFP_KERNEL);\n \n This code was detected with the help of Coccinelle.\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Reviewed-by: Vishal Verma \u003cvishal.l.verma@intel.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190610210613.GA21989@embeddedor\n Signed-off-by: Dan Williams \u003cdan.j.williams@intel.com\u003e\n\ncommit b42090256fba05dce1a0482a4ccd9bb6464cc499\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 29 15:56:40 2019 -0300\n\n perf tools: Remove debug.h from header files not needing it\n \n And fix the fallout, adding it to places that must have it since they\n use its definitions.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-1s3jel4i26chq2g0lydoz7i3@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 0ac25fd0a04d8bd52ceac2476e71a4e497489987\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 29 15:42:40 2019 -0300\n\n perf tools: Remove perf.h from source files not needing it\n \n With the movement of lots of stuff out of perf.h to other headers we\n ended up not needing it in lots of places, remove it from those places.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-c718m0sxxwp73lp9d8vpihb4@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit c1a604dff486399ae0be95e6396e0158df95ad5d\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 29 15:20:59 2019 -0300\n\n perf tools: Remove needless perf.h include directive from headers\n \n Its not needed there, add it to the places that need it and were getting\n it via those headers.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-5yulx1u16vyd0zmrbg1tjhju@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit f37110205c3065546d6995b1463751c7bbb50e89\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 29 15:16:27 2019 -0300\n\n perf time-utils: Adopt rdclock() from perf.h\n \n Seems to be a better place for this function to live, further shrinking\n the hodge-podge that perf.h was.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-0zzt1u9rpyjukdy1ccr2u5r9@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 91854f9a077e18e43ed30ebe9c61f8089bec9166\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 29 14:59:50 2019 -0300\n\n perf tools: Move everything related to sys_perf_event_open() to perf-sys.h\n \n And remove unneeded include directives from perf-sys.h to prune the\n header dependency tree.\n \n Fixup the fallout in places where definitions were being used without\n the needed include directives that were being satisfied because they\n were in perf-sys.h.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-7b1zvugiwak4ibfa3j6ott7f@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit a77494026309711a5f1e4b078e353cd46c2dad9f\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 29 14:40:28 2019 -0300\n\n perf header: Move CPUINFO_PROC to the only file where it is used\n \n To reduce perf-sys.h and eventually nuke it.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-ars2j5m3if3gypsvkbbijucq@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 108a1bb9d1d88bff47d5eccd2cf18dc09a04fb9f\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 29 13:00:28 2019 -0300\n\n perf tools: Remove needless libtraceevent include directives\n \n Remove traceevent/event-parse.h and traceevent/trace-seq.h from places\n where it is not needed.\n \n Should avoid rebuilding those files when these traceevent headers get\n changed.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n Cc: Tzvetomir Stoyanov \u003ctstoyanov@vmware.com\u003e\n Link: https://lkml.kernel.org/n/tip-26hn75jn9rdealn4uqtzend6@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit df552793493ff83b2b7289389d29d417b3ef6d6d\nAuthor: Kyle Meyer \u003cmeyerk@hpe.com\u003e\nDate: Tue Aug 27 16:43:52 2019 -0500\n\n libperf: Warn when exceeding MAX_NR_CPUS in cpumap\n \n Display a warning when attempting to profile more than MAX_NR_CPU CPUs.\n This patch should not change any behavior.\n \n Signed-off-by: Kyle Meyer \u003ckyle.meyer@hpe.com\u003e\n Reviewed-by: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Russ Anderson \u003cruss.anderson@hpe.com\u003e\n Link: http://lore.kernel.org/lkml/20190827214352.94272-8-meyerk@stormcage.eag.rdlabs.hpecorp.net\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit dc84187f32a3e8eb86bd97f3b10494e1f1fe5e7f\nAuthor: Kyle Meyer \u003cmeyerk@hpe.com\u003e\nDate: Tue Aug 27 16:43:51 2019 -0500\n\n perf header: Replace MAX_NR_CPUS with cpu__max_cpu()\n \n The function cpu__max_cpu() returns the possible number of CPUs as\n defined in the sysfs and can be used as an alternative for MAX_NR_CPUS\n in write_cache.\n \n MAX_CACHES is replaced by cpu__max_cpu() * MAX_CACHE_LVL.\n \n Signed-off-by: Kyle Meyer \u003ckyle.meyer@hpe.com\u003e\n Reviewed-by: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Russ Anderson \u003cruss.anderson@hpe.com\u003e\n Link: http://lore.kernel.org/lkml/20190827214352.94272-7-meyerk@stormcage.eag.rdlabs.hpecorp.net\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 8c7274691f0de5fb56f3b9fe9208ce7e515a2d2c\nAuthor: Kyle Meyer \u003cmeyerk@hpe.com\u003e\nDate: Tue Aug 27 16:43:50 2019 -0500\n\n perf machine: Replace MAX_NR_CPUS with perf_env::nr_cpus_online\n \n nr_cpus, the number of CPUs online during a record session bound by\n MAX_NR_CPUS, can be used as a dynamic alternative for MAX_NR_CPUS in\n __machine__synthesize_threads and machine__set_current_tid.\n \n Signed-off-by: Kyle Meyer \u003ckyle.meyer@hpe.com\u003e\n Reviewed-by: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Russ Anderson \u003cruss.anderson@hpe.com\u003e\n Link: http://lore.kernel.org/lkml/20190827214352.94272-6-meyerk@stormcage.eag.rdlabs.hpecorp.net\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 7df4e36a4785618f0c63f3dc2bacb164780ab0f6\nAuthor: Kyle Meyer \u003cmeyerk@hpe.com\u003e\nDate: Tue Aug 27 16:43:49 2019 -0500\n\n perf session: Replace MAX_NR_CPUS with perf_env::nr_cpus_online\n \n nr_cpus, the number of CPUs online during a record session bound by\n MAX_NR_CPUS, can be used as a dynamic alternative for MAX_NR_CPUS in\n perf_session__cpu_bitmap.\n \n Signed-off-by: Kyle Meyer \u003ckyle.meyer@hpe.com\u003e\n Reviewed-by: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Russ Anderson \u003cruss.anderson@hpe.com\u003e\n Link: http://lore.kernel.org/lkml/20190827214352.94272-5-meyerk@stormcage.eag.rdlabs.hpecorp.net\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 92b5a1545ad51e8225e691e9a29ba33cc9fe37bc\nAuthor: Kyle Meyer \u003cmeyerk@hpe.com\u003e\nDate: Tue Aug 27 16:43:48 2019 -0500\n\n perf stat: Replace MAX_NR_CPUS with cpu__max_cpu()\n \n The function cpu__max_cpu() returns the possible number of CPUs as\n defined in the sysfs and can be used as an alternative for MAX_NR_CPUS\n in zero_per_pkg() and check_per_pkg().\n \n Signed-off-by: Kyle Meyer \u003ckyle.meyer@hpe.com\u003e\n Reviewed-by: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Russ Anderson \u003cruss.anderson@hpe.com\u003e\n Link: http://lore.kernel.org/lkml/20190827214352.94272-4-meyerk@stormcage.eag.rdlabs.hpecorp.net\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit f78f96676a256d7fa171a54b271a2ad2c6555c9c\nAuthor: Kyle Meyer \u003cmeyerk@hpe.com\u003e\nDate: Tue Aug 27 16:43:47 2019 -0500\n\n perf svghelper: Replace MAX_NR_CPUS with perf_env::nr_cpus_online\n \n \u0027nr_cpus\u0027, the number of CPUs online during a record session bound by\n MAX_NR_CPUS, can be used as a dynamic alternative for MAX_NR_CPUS in\n svg_build_topology_map().\n \n The value of nr_cpus can be passed into str_to_bitmap(),\n scan_core_topology(), and svg_build_topology_map() to replace\n MAX_NR_CPUS as well.\n \n Signed-off-by: Kyle Meyer \u003ckyle.meyer@hpe.com\u003e\n Reviewed-by: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Russ Anderson \u003cruss.anderson@hpe.com\u003e\n Link: http://lore.kernel.org/lkml/20190827214352.94272-3-meyerk@stormcage.eag.rdlabs.hpecorp.net\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 0ac1dd5b4a70cfc8591dd9426f800b484765badb\nAuthor: Kyle Meyer \u003cmeyerk@hpe.com\u003e\nDate: Tue Aug 27 16:43:46 2019 -0500\n\n perf timechart: Refactor svg_build_topology_map()\n \n Exchange the parameters of svg_build_topology_map() with \u0027struct\n perf_env *env\u0027 and adjust the function accordingly.\n \n This patch should not change any behavior, it is merely refactoring for\n the following patch.\n \n Committer notes:\n \n No need to include env.h from svghelper.h, all it needs is a forward\n declaration for \u0027struct perf_env\u0027, so move the include directive to\n svghelper.c, where it is really needed.\n \n Signed-off-by: Kyle Meyer \u003ckyle.meyer@hpe.com\u003e\n Reviewed-by: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Russ Anderson \u003cruss.anderson@hpe.com\u003e\n Link: http://lore.kernel.org/lkml/20190827214352.94272-2-meyerk@stormcage.eag.rdlabs.hpecorp.net\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 67260e8c0e681a9bb9ed861514b4c80c2d0eb2e5\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Tue Aug 20 16:02:19 2019 +0200\n\n perf c2c: Display proper cpu count in nodes column\n \n There\u0027s wrong bitmap considered when checking for cpu count of specific\n node.\n \n We do the needed computation for \u0027set\u0027 variable, but at the end we use\n the \u0027c2c_he-\u003ecpuset\u0027 weight, which shows misleading numbers.\n \n Fixes: 1e181b92a2da (\"perf c2c report: Add \u0027node\u0027 sort key\")\n Reported-by: Joe Mario \u003cjmario@redhat.com\u003e\n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lore.kernel.org/lkml/20190820140219.28338-1-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 0183eb8bb59d45f26ec4fc73aaa416067fe6c0be\nAuthor: Jean Delvare \u003cjdelvare@suse.de\u003e\nDate: Fri Aug 2 14:55:26 2019 +0200\n\n i2c: piix4: Add ACPI support\n \n Enable the i2c-piix4 SMBus controller driver to enumerate I2C slave\n devices using ACPI. It builds on the related I2C mux device work\n in commit 8eb5c87a92c0 (\"i2c: add ACPI support for I2C mux ports\")\n \n In the i2c-piix4 driver the adapters are enumerated as:\n Main SMBus adapter Port 0, Port 2, ..., aux port (i.e., ASF adapter)\n \n However, in the AMD BKDG documentation[1], the implied order of ports is:\n Main SMBus adapter Port 0, ASF adapter, Port 2, Port 3, ...\n \n This ordering difference is unfortunate. We assume that ACPI\n developers will use the AMD documentation ordering, so we have to\n pass an extra parameter to piix4_add_adapter().\n \n [1] 52740 BIOS and Kernel Developer\u0027s Guide (BKDG) for AMD Family 16h\n Models 30h-3Fh Processors\n \n Based on earlier work by Andrew Cooks.\n \n Reported-by: Andrew Cooks \u003candrew.cooks@opengear.com\u003e\n Signed-off-by: Jean Delvare \u003cjdelvare@suse.de\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit 528d53a1592b0e27c423f7cafc1df85f77fc1163\nAuthor: Jean Delvare \u003cjdelvare@suse.de\u003e\nDate: Fri Aug 2 14:54:38 2019 +0200\n\n i2c: piix4: Fix probing of reserved ports on AMD Family 16h Model 30h\n \n Prevent bus timeouts and resets on Family 16h Model 30h by not probing\n reserved Ports 3 and 4.\n \n According to the AMD BIOS and Kernel Developer\u0027s Guides (BKDG), Port 3\n and Port 4 are reserved on the following devices:\n - Family 15h Model 60h-6Fh\n - Family 15h Model 70h-7Fh\n - Family 16h Model 30h-3Fh\n \n Based on earlier work by Andrew Cooks.\n \n Reported-by: Andrew Cooks \u003candrew.cooks@opengear.com\u003e\n Signed-off-by: Jean Delvare \u003cjdelvare@suse.de\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit ba919403566dba52bf074851896ba0ca7f72c1e2\nAuthor: Federico Vaga \u003cfederico.vaga@cern.ch\u003e\nDate: Thu Aug 22 15:21:32 2019 +0200\n\n i2c: ocores: use request_any_context_irq() to register IRQ handler\n \n The i2c-ocores device is an HDL component that get instantiated in FPGA.\n The software stack used to drive an FPGA can be very different, and the\n i2c-ocore ip-core must work in different context. With respect to this\n patch the IRQ controller behind this device, and its driver, can have\n different implementations (nested threads). For this reason, it is safer\n to use `request_any_context_irq()` to avoid errors at probe time.\n \n Signed-off-by: Federico Vaga \u003cfederico.vaga@cern.ch\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Reviewed-by: Peter Korsgaard \u003cpeter@korsgaard.com\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit a6af48ec0712a0c98d8abe6b47c655b26026fceb\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Aug 19 13:31:30 2019 +0300\n\n i2c: designware: Fix optional reset error handling\n \n The commit bb475230b8e5 (\"reset: make optional functions really optional\")\n brought a missed part of the support for an optional reset handlers.\n \n Since that we don\u0027t need to have special error handling in the driver.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Jarkko Nikula \u003cjarkko.nikula@linux.intel.com\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit 71dc297ca9ab6311e7d39fc53811fd871d51854f\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Aug 19 13:24:23 2019 +0300\n\n i2c: designware: assert reset when error happen at -\u003eprobe()\n \n The commit c62ebb3d5f0d (\"i2c: designware: Add support for an interface clock\")\n introduced an optional clock while missed correct error handling.\n assert reset line back if error happen at -\u003eprobe().\n \n Fixes: c62ebb3d5f0d (\"i2c: designware: Add support for an interface clock\")\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Jarkko Nikula \u003cjarkko.nikula@linux.intel.com\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit f9bf7a899412b5f8c96b084343bde4168dc12efa\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Aug 19 13:16:01 2019 +0530\n\n i2c: taos-evm: Make structure tsl2550_info constant\n \n Static structure tsl2550_info, of type i2c_board_info, is referenced\n only twice: the first time in arguments to dev_info() (which does not\n modify it) and the second time as the last argument to function\n i2c_new_device() (where the corresponding parameter is declared as\n const). As tsl2550_info is therefore never modified, make it const to\n protect it from unintended modifications.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Reviewed-by: Jean Delvare \u003cjdelvare@suse.de\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit 70fb95e213147a88417eaae5b93ba6e59be087c0\nAuthor: Jarkko Nikula \u003cjarkko.nikula@linux.intel.com\u003e\nDate: Thu Aug 15 17:29:44 2019 +0300\n\n i2c: designware-pci: Add support for Elkhart Lake PSE I2C\n \n Add support for Intel(R) Programmable Services Engine (Intel(R) PSE) I2C\n controller in Intel Elkhart Lake when interface is assigned to the host\n processor.\n \n Signed-off-by: Jarkko Nikula \u003cjarkko.nikula@linux.intel.com\u003e\n Reviewed-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit 21aa3983d619055cc3ecd0f0ceea4981dc8ee751\nAuthor: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\nDate: Thu Aug 15 17:29:43 2019 +0300\n\n i2c: designware-pci: Switch over to MSI interrupts\n \n Some devices support MSI interrupts. Let\u0027s at least try to use them in\n platforms that provide MSI capability.\n \n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n Signed-off-by: Jarkko Nikula \u003cjarkko.nikula@linux.intel.com\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit bc31c1eea99de9a8e65b011483716236af52f7ed\nAuthor: Israel Rukshin \u003cisraelr@mellanox.com\u003e\nDate: Wed Aug 28 14:11:50 2019 +0300\n\n nvme-rdma: Use rq_dma_dir macro\n \n Remove code duplication.\n \n Signed-off-by: Israel Rukshin \u003cisraelr@mellanox.com\u003e\n Reviewed-by: Max Gurtovoy \u003cmaxg@mellanox.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit f15872c5dce43b69c3dee7739d7d3f54c54fc527\nAuthor: Israel Rukshin \u003cisraelr@mellanox.com\u003e\nDate: Wed Aug 28 14:11:49 2019 +0300\n\n nvme-fc: Use rq_dma_dir macro\n \n Remove code duplication.\n \n Signed-off-by: Israel Rukshin \u003cisraelr@mellanox.com\u003e\n Reviewed-by: Max Gurtovoy \u003cmaxg@mellanox.com\u003e\n Reviewed-by: James Smart \u003cjames.smart@broadcom.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit f2fa006f81cc5e7b5b9ef2c06dfeb50f52a47250\nAuthor: Israel Rukshin \u003cisraelr@mellanox.com\u003e\nDate: Wed Aug 28 14:11:48 2019 +0300\n\n nvme-pci: Tidy up nvme_unmap_data\n \n Remove pointless local variable and use rq_dma_dir macro.\n \n Signed-off-by: Israel Rukshin \u003cisraelr@mellanox.com\u003e\n Reviewed-by: Max Gurtovoy \u003cmaxg@mellanox.com\u003e\n Reviewed-by: Keith Busch \u003ckbusch@kernel.org\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit e7832cb48a654cd12b2bc9181b2f0ad49d526ac6\nAuthor: Sagi Grimberg \u003csagi@grimberg.me\u003e\nDate: Fri Aug 2 19:33:59 2019 -0700\n\n nvme: make fabrics command run on a separate request queue\n \n We have a fundamental issue that fabric commands use the admin_q.\n The reason is, that admin-connect, register reads and writes and\n admin commands cannot be guaranteed ordering while we are running\n controller resets.\n \n For example, when we reset a controller we perform:\n 1. disable the controller\n 2. teardown the admin queue\n 3. re-establish the admin queue\n 4. enable the controller\n \n In order to perform (3), we need to unquiesce the admin queue, however\n we may have some admin commands that are already pending on the\n quiesced admin_q and will immediate execute when we unquiesce it before\n we execute (4). The host must not send admin commands to the controller\n before enabling the controller.\n \n To fix this, we have the fabric commands (admin connect and property\n get/set, but not I/O queue connect) use a separate fabrics_q and make\n sure to quiesce the admin_q before we disable the controller, and\n unquiesce it only after we enable the controller.\n \n This fixes the error prints from nvmet in a controller reset storm test:\n kernel: nvmet: got cmd 6 while CC.EN \u003d\u003d 0 on qid \u003d 0\n Which indicate that the host is sending an admin command when the\n controller is not enabled.\n \n Reviewed-by: James Smart \u003cjames.smart@broadcom.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit d38e9f04ebf667d9cb8185b45bff747485f1d3e9\nAuthor: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nDate: Wed Aug 7 17:51:22 2019 +1000\n\n nvme-pci: Support shared tags across queues for Apple 2018 controllers\n \n Another issue with the Apple T2 based 2018 controllers seem to be\n that they blow up (and shut the machine down) if there\u0027s a tag\n collision between the IO queue and the Admin queue.\n \n My suspicion is that they use our tags for their internal tracking\n and don\u0027t mix them with the queue id. They also seem to not like\n when tags go beyond the IO queue depth, ie 128 tags.\n \n This adds a quirk that marks tags 0..31 of the IO queue reserved\n \n Signed-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n Reviewed-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Acked-by: Keith Busch \u003ckeith.busch@intel.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit 66341331ba0d2de4ff421cdc401a1e34de50502a\nAuthor: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nDate: Wed Aug 7 17:51:21 2019 +1000\n\n nvme-pci: Add support for Apple 2018+ models\n \n Based on reverse engineering and original patch by\n \n Paul Pawlowski \u003cpaul@mrarm.io\u003e\n \n This adds support for Apple weird implementation of NVME in their\n 2018 or later machines. It accounts for the twice-as-big SQ entries\n for the IO queues, and the fact that only interrupt vector 0 appears\n to function properly.\n \n Signed-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n Reviewed-by: Minwoo Im \u003cminwoo.im.dev@gmail.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit c1e0cc7e1d319936271dfdd0a9405275c8091381\nAuthor: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nDate: Wed Aug 7 17:51:20 2019 +1000\n\n nvme-pci: Add support for variable IO SQ element size\n \n The size of a submission queue element should always be 6 (64 bytes)\n by spec.\n \n However some controllers such as Apple\u0027s are not properly implementing\n the standard and require a different size.\n \n This provides the ground work for the subsequent quirks for these\n controllers.\n \n Signed-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n Reviewed-by: Minwoo Im \u003cminwoo.im.dev@gmail.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit 8a1d09a668e7e245b8e4131cc9017c63fee02ee5\nAuthor: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nDate: Wed Aug 7 17:51:19 2019 +1000\n\n nvme-pci: Pass the queue to SQ_SIZE/CQ_SIZE macros\n \n This will make it easier to handle variable queue entry sizes\n later. No functional change.\n \n Signed-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Minwoo Im \u003cminwoo.im.dev@gmail.com\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit 35fe0d12c8a3d5e45f297562732ddc9ba9dc58dd\nAuthor: Hannes Reinecke \u003chare@suse.de\u003e\nDate: Wed Jul 24 15:47:55 2019 +0200\n\n nvme: trace bio completion\n \n When native multipathing is enabled we cannot enable blktrace for\n the underlying paths, so any completion is never traced.\n \n Signed-off-by: Hannes Reinecke \u003chare@suse.com\u003e\n [fixed-up by Mikhail for non-multipath-build]\n Signed-off-by: Mikhail Skorzhinskii \u003cmskorzhinskiy@solarflare.com\u003e\n Reviewed-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit e01f91dff91c7b16a6e3faf2565017d497a73f83\nAuthor: Anton Eidelman \u003canton@lightbitslabs.com\u003e\nDate: Fri Aug 16 13:00:10 2019 -0700\n\n nvme-multipath: fix ana log nsid lookup when nsid is not found\n \n ANA log parsing invokes nvme_update_ana_state() per ANA group desc.\n This updates the state of namespaces with nsids in desc-\u003ensids[].\n \n Both ctrl-\u003enamespaces list and desc-\u003ensids[] array are sorted by nsid.\n Hence nvme_update_ana_state() performs a single walk over ctrl-\u003enamespaces:\n - if current namespace matches the current desc-\u003ensids[n],\n this namespace is updated, and n is incremented.\n - the process stops when it encounters the end of either\n ctrl-\u003enamespaces end or desc-\u003ensids[]\n \n In case desc-\u003ensids[n] does not match any of ctrl-\u003enamespaces,\n the remaining nsids following desc-\u003ensids[n] will not be updated.\n Such situation was considered abnormal and generated WARN_ON_ONCE.\n \n However ANA log MAY contain nsids not (yet) found in ctrl-\u003enamespaces.\n For example, lets consider the following scenario:\n - nvme0 exposes namespaces with nsids \u003d [2, 3] to the host\n - a new namespace nsid \u003d 1 is added dynamically\n - also, a ANA topology change is triggered\n - NS_CHANGED aen is generated and triggers scan_work\n - before scan_work discovers nsid\u003d1 and creates a namespace, a NOTICE_ANA\n aen was issues and ana_work receives ANA log with nsids\u003d[1, 2, 3]\n \n Result: ana_work fails to update ANA state on existing namespaces [2, 3]\n \n Solution:\n Change the way nvme_update_ana_state() namespace list walk\n checks the current namespace against desc-\u003ensids[n] as follows:\n a) ns-\u003ehead-\u003ens_id \u003c desc-\u003ensids[n]: keep walking ctrl-\u003enamespaces.\n b) ns-\u003ehead-\u003ens_id \u003d\u003d desc-\u003ensids[n]: match, update the namespace\n c) ns-\u003ehead-\u003ens_id \u003e\u003d desc-\u003ensids[n]: skip to desc-\u003ensids[n+1]\n \n This enables correct operation in the scenario described above.\n This also allows ANA log to contain nsids currently invisible\n to the host, i.e. inactive nsids.\n \n Signed-off-by: Anton Eidelman \u003canton@lightbitslabs.com\u003e\n Reviewed-by: James Smart \u003cjames.smart@broadcom.com\u003e\n Reviewed-by: Hannes Reinecke \u003chare@suse.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit 89275a9659fe57a3c7eef6778ec64f9e435c75eb\nAuthor: Israel Rukshin \u003cisraelr@mellanox.com\u003e\nDate: Sun Aug 18 12:08:55 2019 +0300\n\n nvmet-tcp: Add TOS for tcp transport\n \n Set the outgoing packets type of service (TOS) according to the\n receiving TOS.\n \n Signed-off-by: Israel Rukshin \u003cisraelr@mellanox.com\u003e\n Suggested-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n Reviewed-by: Max Gurtovoy \u003cmaxg@mellanox.com\u003e\n Reviewed-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit bb13985d5a55b2830095640a2e64145bcb34929b\nAuthor: Israel Rukshin \u003cisraelr@mellanox.com\u003e\nDate: Sun Aug 18 12:08:54 2019 +0300\n\n nvme-tcp: Add TOS for tcp transport\n \n TOS provide clients the ability to segregate traffic flows for\n different type of data.\n One of the TOS usage is bandwidth management which allows setting bandwidth\n limits for QoS classes, e.g. 80% bandwidth to controllers at QoS class A\n and 20% to controllers at QoS class B.\n \n usage examples:\n nvme connect --tos\u003d0 --transport\u003dtcp --traddr\u003d10.0.1.1 --nqn\u003dtest-nvme\n \n Signed-off-by: Israel Rukshin \u003cisraelr@mellanox.com\u003e\n Reviewed-by: Max Gurtovoy \u003cmaxg@mellanox.com\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit 9924b0304ab278406aaee9184a1a2032b4778c65\nAuthor: Israel Rukshin \u003cisraelr@mellanox.com\u003e\nDate: Sun Aug 18 12:08:53 2019 +0300\n\n nvme-tcp: Use struct nvme_ctrl directly\n \n This patch doesn\u0027t change any functionality.\n \n Signed-off-by: Israel Rukshin \u003cisraelr@mellanox.com\u003e\n Reviewed-by: Max Gurtovoy \u003cmaxg@mellanox.com\u003e\n Reviewed-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit e63440d6a3134f7ae74bfb00bfc01db3efb8d3aa\nAuthor: Israel Rukshin \u003cisraelr@mellanox.com\u003e\nDate: Sun Aug 18 12:08:52 2019 +0300\n\n nvme-rdma: Add TOS for rdma transport\n \n For RDMA transports, TOS is an extension of IB QoS to provide clients\n the ability to segregate traffic flows for different type of data.\n RDMA CM abstract it for ULPs using rdma_set_service_type().\n Internally, each traffic flow is represented by a connection with all of\n its independent resources like that of a normal connection, and is\n differentiated by service type. In other words, there can be multiple qp\n connections between an IP pair and each supports a unique service type.\n \n One of the TOS usage is bandwidth management which allows setting bandwidth\n limits for QoS classes, e.g. 80% bandwidth to controllers at QoS class A\n and 20% to controllers at QoS class B.\n \n Note: In addition to the TOS configuration, QOS must be configured on the\n relevant HCA on the target (send RDMA commands) and initiator to effect\n the traffic.\n \n usage examples:\n nvme connect --tos\u003d0 --transport\u003drdma --traddr\u003d10.0.1.1 --nqn\u003dtest-nvme\n \n Signed-off-by: Israel Rukshin \u003cisraelr@mellanox.com\u003e\n Reviewed-by: Max Gurtovoy \u003cmaxg@mellanox.com\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit 52b4451a9e5ae1e9ae739db16e8af61c77805389\nAuthor: Israel Rukshin \u003cisraelr@mellanox.com\u003e\nDate: Sun Aug 18 12:08:51 2019 +0300\n\n nvme-fabrics: Add type of service (TOS) configuration\n \n TOS is user-defined and needs to be configured via nvme-cli.\n It must be set before initiating any traffic and once set the TOS\n cannot be changed.\n \n Signed-off-by: Israel Rukshin \u003cisraelr@mellanox.com\u003e\n Reviewed-by: Max Gurtovoy \u003cmaxg@mellanox.com\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit 35d1a938dcdaeb8e1d860f061a0cd11f67f42774\nAuthor: Sagi Grimberg \u003csagi@grimberg.me\u003e\nDate: Fri Aug 2 20:29:11 2019 -0700\n\n nvmet-tcp: fix possible memory leak\n \n when we uninit a command in error flow we also need to\n free an iovec if it was allocated.\n \n Reviewed-by: Max Gurtovoy \u003cmaxg@mellanox.com\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit b627200762c7e8153fe1620fdd52a68f4ca2f8a5\nAuthor: Sagi Grimberg \u003csagi@grimberg.me\u003e\nDate: Fri Aug 2 20:23:38 2019 -0700\n\n nvmet-tcp: fix possible NULL deref\n \n We must only call sgl_free for sgl that we actually\n allocated.\n \n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit 42df26d4df7b4437db7d3847c36abc3e5aa237f1\nAuthor: Minwoo Im \u003cminwoo.im.dev@gmail.com\u003e\nDate: Sun Aug 4 16:50:50 2019 +0900\n\n nvmet: trace: parse Get LBA Status command in detail\n \n Four different fields are in CDWs of Get LBA Status command which means\n it would be great if we can see in detail when tracing in target side\n also.\n \n Signed-off-by: Minwoo Im \u003cminwoo.im.dev@gmail.com\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit 177b06ed0997bdaa9671c615a37649ab9aec889e\nAuthor: Minwoo Im \u003cminwoo.im.dev@gmail.com\u003e\nDate: Sun Aug 4 16:50:49 2019 +0900\n\n nvme: trace: parse Get LBA Status command in detail\n \n Four different fields are in CDWs of Get LBA Status command which means\n it would be great if we can see in detail when tracing.\n \n Signed-off-by: Minwoo Im \u003cminwoo.im.dev@gmail.com\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit a5ef757204bab6f80268a7437556cb57744ab7d4\nAuthor: Minwoo Im \u003cminwoo.im.dev@gmail.com\u003e\nDate: Sun Aug 4 16:50:48 2019 +0900\n\n nvme: trace: support for Get LBA Status opcode parsed\n \n This patch adds Get LBA Status command\u0027s opcode to the macro that is\n used by the trace feature. Now we can see \"get_lba_status\" instead of\n the opcode value itself.\n \n Signed-off-by: Minwoo Im \u003cminwoo.im.dev@gmail.com\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit c638984521f19ba218477d5ef9f10f9a6206bab6\nAuthor: Minwoo Im \u003cminwoo.im.dev@gmail.com\u003e\nDate: Sun Aug 4 16:50:47 2019 +0900\n\n nvme: add Get LBA Status command opcode\n \n NVMe 1.4 added Get LBA Status command with opcode 0x86.\n \n Signed-off-by: Minwoo Im \u003cminwoo.im.dev@gmail.com\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit 3bec2e3754becebd4c452999adb49bc62c575ea4\nAuthor: Tom Wu \u003ctomwu@mellanox.com\u003e\nDate: Thu Aug 8 02:22:36 2019 +0000\n\n nvmet: fix data units read and written counters in SMART log\n \n In nvme spec 1.3 there is a definition for data write/read counters\n from SMART log, (See section 5.14.1.2):\n This value is reported in thousands (i.e., a value of 1\n corresponds to 1000 units of 512 bytes read) and is rounded up.\n \n However, in nvme target where value is reported with actual units,\n but not thousands of units as the spec requires.\n \n Signed-off-by: Tom Wu \u003ctomwu@mellanox.com\u003e\n Reviewed-by: Israel Rukshin \u003cisraelr@mellanox.com\u003e\n Reviewed-by: Max Gurtovoy \u003cmaxg@mellanox.com\u003e\n Reviewed-by: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit 1a9460cef571100487cde9c44110a5afff2e9ea2\nAuthor: Sagi Grimberg \u003csagi@grimberg.me\u003e\nDate: Wed Jul 3 14:08:04 2019 -0700\n\n nvme-tcp: support simple polling\n \n Simple polling support via socket busy_poll interface.\n Although we do not shutdown interrupts but simply hammer\n the socket poll, we can sometimes find completions faster\n than the normal interrupt driven RX path.\n \n We add per queue nr_cqe counter that resets every time\n RX path is invoked such that .poll callback can return it\n to stay consistent with the semantics.\n \n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit 79fd751d61aa8b9979d51357236890d69989ba04\nAuthor: Minwoo Im \u003cminwoo.im.dev@gmail.com\u003e\nDate: Sun Jul 14 17:18:42 2019 +0900\n\n nvme: tcp: selects CRYPTO_CRC32C for nvme-tcp\n \n The tcp host module is now taking those APIs from crypto ahash:\n (1) crypto_ahash_final()\n (2) crypto_ahash_digest()\n (3) crypto_alloc_ahash()\n \n nvme-tcp should depends on CRYPTO_CRC32C.\n \n Cc: Christoph Hellwig \u003chch@lst.de\u003e\n Cc: Keith Busch \u003ckbusch@kernel.org\u003e\n Cc: Jens Axboe \u003caxboe@fb.com\u003e\n Reviewed-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Minwoo Im \u003cminwoo.im.dev@gmail.com\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit b5b0504878b884fb38e8983f1637a7be1d07fad3\nAuthor: Sagi Grimberg \u003csagi@grimberg.me\u003e\nDate: Mon Jul 22 17:06:54 2019 -0700\n\n nvme: don\u0027t pass cap to nvme_disable_ctrl\n \n All seem to call it with ctrl-\u003ecap so no need to pass it\n at all.\n \n Reviewed-by: Minwoo Im \u003cminwoo.im.dev@gmail.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit c0f2f45be2976abe973c8cd544f38e2d928771b0\nAuthor: Sagi Grimberg \u003csagi@grimberg.me\u003e\nDate: Mon Jul 22 17:06:53 2019 -0700\n\n nvme: move sqsize setting to the core\n \n nvme_enable_ctrl reads the cap register right after, so\n no need to do that locally in the transport driver. Have\n sqsize setting in nvme_init_identify.\n \n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit aa22c8e6650d29a00196087caa2bbb32dc6117bc\nAuthor: Sagi Grimberg \u003csagi@grimberg.me\u003e\nDate: Thu Aug 22 10:51:17 2019 -0700\n\n nvme-pci: set ctrl sqsize to the device q_depth\n \n Align with what the rest of the transports are doing.\n \n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit 4fba445828fc047c095a3a955b4ceac62cd56964\nAuthor: Sagi Grimberg \u003csagi@grimberg.me\u003e\nDate: Mon Jul 22 17:06:52 2019 -0700\n\n nvme: have nvme_init_identify set ctrl-\u003ecap\n \n No need to use a stack cap variable.\n \n Reviewed-by: Minwoo Im \u003cminwoo.im.dev@gmail.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit 10407ec9b42d30a6ebc49f7f84e2bb2131438699\nAuthor: Potnuri Bharat Teja \u003cbharat@chelsio.com\u003e\nDate: Mon Jul 8 15:22:00 2019 +0530\n\n nvme-tcp: Use protocol specific operations while reading socket\n \n Using socket specific read_sock() calls instead of directly calling\n tcp_read_sock() helps lld module registered handlers if any, to be called\n from nvme-tcp host.\n This patch therefore replaces the tcp_read_sock() with socket specific\n prot_ops.\n \n Signed-off-by: Potnuri Bharat Teja \u003cbharat@chelsio.com\u003e\n Acked-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit 6be182607db98f9672865fed3ee3bf8b98e62dd4\nAuthor: Sagi Grimberg \u003csagi@grimberg.me\u003e\nDate: Fri Jul 19 12:46:46 2019 -0700\n\n nvme-tcp: cleanup nvme_tcp_recv_pdu\n \n Can return directly in the switch statement\n \n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n\ncommit cc9bbb6cde6c321706f1eff79d077b4a91cd5d12\nAuthor: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\nDate: Wed Aug 28 11:51:02 2019 +0200\n\n ASoC: Intel: Baytrail: Fix implicit fallthrough warning\n \n Append fallthrough statement to fix warning reported during compilation.\n \n Fixes: b80d19c166c4 (\"ASoC: Intel: Restore Baytrail ADSP streams only when ADSP was in reset\")\n Signed-off-by: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190828095102.15737-1-cezary.rojewski@intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 696d05225cebffd172008d212657be90e823eac0\nAuthor: Shengjiu Wang \u003cshengjiu.wang@nxp.com\u003e\nDate: Wed Aug 28 13:20:17 2019 -0400\n\n ASoC: fsl_ssi: Fix clock control issue in master mode\n \n The test case is\n arecord -Dhw:0 -d 10 -f S16_LE -r 48000 -c 2 temp.wav \u0026\n aplay -Dhw:0 -d 30 -f S16_LE -r 48000 -c 2 test.wav\n \n There will be error after end of arecord:\n aplay: pcm_write:2051: write error: Input/output error\n \n Capture and Playback work in parallel in master mode, one\n substream stops, the other substream is impacted, the\n reason is that clock is disabled wrongly.\n \n The clock\u0027s reference count is not increased when second\n substream starts, the hw_param() function returns in the\n beginning because first substream is enabled, then in end\n of first substream, the hw_free() disables the clock.\n \n This patch is to move the clock enablement to the place\n before checking of the device enablement in hw_param().\n \n Signed-off-by: Shengjiu Wang \u003cshengjiu.wang@nxp.com\u003e\n Link: https://lore.kernel.org/r/1567012817-12625-1-git-send-email-shengjiu.wang@nxp.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit f0b576801d83cecab29888010017333babd61ede\nAuthor: Adamski, Krzysztof (Nokia - PL/Wroclaw) \u003ckrzysztof.adamski@nokia.com\u003e\nDate: Mon Aug 19 09:07:07 2019 +0000\n\n i2c: axxia: support slave mode\n \n This device contains both master and slave controllers which can be\n enabled simultaneously. Both controllers share the same SDA/SCL lines\n and interrupt source but has separate control and status registers.\n Controllers also works in loopback mode - slave device can communicate\n with its own master controller internally. The controller can handle up\n to two addresses, both of which may be 10 bit. Most of the logic\n (sending (N)ACK, handling repeated start or switching between\n write/read) is handled automatically which makes working with this\n controller quite easy.\n \n For simplicity, this patch adds basic support, limiting to only one\n slave address. Support for the 2nd device may be added in the future.\n \n Note that synchronize_irq() is used to ensure any running slave interrupt\n is finished to make sure slave i2c_client structure can be safely used\n by i2c_slave_event.\n \n Signed-off-by: Krzysztof Adamski \u003ckrzysztof.adamski@nokia.com\u003e\n Reviewed-by: Alexander Sverdlin \u003calexander.sverdlin@nokia.com\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit 724041ae15ed9639b72bb54dabfac3279c6b4e55\nAuthor: Max Staudt \u003cmax@enpas.org\u003e\nDate: Mon Aug 19 14:16:18 2019 +0200\n\n i2c: icy: Add LTC2990 present on 2019 board revision\n \n Since the 2019 a1k.org community re-print of these PCBs sports an\n LTC2990 hwmon chip as an example use case, let this driver autoprobe\n for that as well. If it is present, modprobing ltc2990 is sufficient.\n \n The property_entry enables the three additional inputs available on\n this particular board:\n \n in1 will be the voltage of the 5V rail, divided by 2.\n in2 will be the voltage of the 12V rail, divided by 4.\n temp3 will be measured using a PCB loop next the chip.\n \n Signed-off-by: Max Staudt \u003cmax@enpas.org\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit 4768e90ecaec6b503ff64229bda5d91186d2edd3\nAuthor: Max Staudt \u003cmax@enpas.org\u003e\nDate: Tue Aug 20 11:27:39 2019 +0200\n\n i2c: Add i2c-icy for I2C on m68k/Amiga\n \n This is the i2c-icy driver for the ICY board for Amiga computers.\n It connects a PCF8584 I2C controller to the Zorro bus, providing I2C\n connectivity. The original documentation can be found on Aminet:\n \n https://aminet.net/package/docs/hard/icy\n \n IRQ support is currently not implemented, as i2c-algo-pcf is built for\n the ISA bus and a straight implementation of the same stack locks up a\n Zorro machine.\n \n Signed-off-by: Max Staudt \u003cmax@enpas.org\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n [wsa: added a missing newline reported by checkpatch]\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit e2079e93f562c7f7a030eb7642017ee5eabaaa10\nAuthor: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\nDate: Mon Aug 26 17:41:55 2019 -0700\n\n kbuild: Do not enable -Wimplicit-fallthrough for clang for now\n \n This functionally reverts commit bfd77145f35c (\"Makefile: Convert\n -Wimplicit-fallthrough\u003d3 to just -Wimplicit-fallthrough for clang\").\n \n clang enabled support for -Wimplicit-fallthrough in C in r369414 [1],\n which causes a lot of warnings when building the kernel for two reasons:\n \n 1. Clang does not support the /* fall through */ comments. There seems\n to be a general consensus in the LLVM community that this is not\n something they want to support. Joe Perches wrote a script to convert\n all of the comments to a \"fallthrough\" keyword that will be added to\n compiler_attributes.h [2] [3], which catches the vast majority of the\n comments. There doesn\u0027t appear to be any consensus in the kernel\n community when to do this conversion.\n \n 2. Clang and GCC disagree about falling through to final case statements\n with no content or cases that simply break:\n \n https://godbolt.org/z/c8csDu\n \n This difference contributes at least 50 warnings in an allyesconfig\n build for x86, not considering other architectures. This difference\n will need to be discussed to see which compiler is right [4] [5].\n \n [1]: https://github.com/llvm/llvm-project/commit/1e0affb6e564b7361b0aadb38805f26deff4ecfc\n [2]: https://lore.kernel.org/lkml/61ddbb86d5e68a15e24ccb06d9b399bbf5ce2da7.camel@perches.com/\n [3]: https://lore.kernel.org/lkml/1d2830aadbe9d8151728a7df5b88528fc72a0095.1564549413.git.joe@perches.com/\n [4]: https://gcc.gnu.org/bugzilla/show_bug.cgi?id\u003d91432\n [5]: https://github.com/ClangBuiltLinux/linux/issues/636\n \n Given these two problems need discussion and coordination, do not enable\n -Wimplicit-fallthrough with clang right now. Add a comment to explain\n what is going on as well. This commit should be reverted once these two\n issues are fully flushed out and resolved.\n \n Suggested-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Acked-by: Miguel Ojeda \u003cmiguel.ojeda.sandonis@gmail.com\u003e\n Acked-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Acked-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 990784b57731192b7d90c8d4049e6318d81e887d\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 28 16:24:47 2019 +0200\n\n x86/mm/pti: Do not invoke PTI functions when PTI is disabled\n \n When PTI is disabled at boot time either because the CPU is not affected or\n PTI has been disabled on the command line, the boot code still calls into\n pti_finalize() which then unconditionally invokes:\n \n pti_clone_entry_text()\n pti_clone_kernel_text()\n \n pti_clone_kernel_text() was called unconditionally before the 32bit support\n was added and 32bit added the call to pti_clone_entry_text().\n \n The call has no side effects as cloning the page tables into the available\n second one, which was allocated for PTI does not create damage. But it does\n not make sense either and in case that this functionality would be extended\n later this might actually lead to hard to diagnose issues.\n \n Neither function should be called when PTI is runtime disabled. Make the\n invocation conditional.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Dave Hansen \u003cdave.hansen@linux.intel.com\u003e\n Acked-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190828143124.063353972@linutronix.de\n\ncommit 825d0b73cd7526b0bb186798583fae810091cbac\nAuthor: Song Liu \u003csongliubraving@fb.com\u003e\nDate: Wed Aug 28 23:54:55 2019 +0200\n\n x86/mm/pti: Handle unaligned address gracefully in pti_clone_pagetable()\n \n pti_clone_pmds() assumes that the supplied address is either:\n \n - properly PUD/PMD aligned\n or\n - the address is actually mapped which means that independently\n of the mapping level (PUD/PMD/PTE) the next higher mapping\n exists.\n \n If that\u0027s not the case the unaligned address can be incremented by PUD or\n PMD size incorrectly. All callers supply mapped and/or aligned addresses,\n but for the sake of robustness it\u0027s better to handle that case properly and\n to emit a warning.\n \n [ tglx: Rewrote changelog and added WARN_ON_ONCE() ]\n \n Signed-off-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/alpine.DEB.2.21.1908282352470.1938@nanos.tec.linutronix.de\n\ncommit 250212b59a8e33f1b47f15ad9afa3db19ba0fc8d\nAuthor: Stefan Wahren \u003cwahrenst@gmx.net\u003e\nDate: Tue Aug 27 19:04:16 2019 +0200\n\n i2c: bcm2835: Add full name of devicetree node to adapter name\n \n Inspired by Lori Hikichi\u0027s patch for iproc, this adds the full name of\n the devicetree node to the adapter name. With the introduction of\n BCM2711 it\u0027s very difficult to distinguish between the multiple instances.\n \n Signed-off-by: Stefan Wahren \u003cwahrenst@gmx.net\u003e\n Acked-by: Scott Branden \u003cscott.branden@broadcom.com\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit 67de10fbaa12dc7d46412c5ab80f784ea8b4d7f8\nAuthor: Stefan Wahren \u003cwahrenst@gmx.net\u003e\nDate: Tue Aug 27 19:04:15 2019 +0200\n\n i2c: bcm2835: Avoid clk stretch quirk for BCM2711\n \n The I2C block on the BCM2711 isn\u0027t affected by the clk stretching bug.\n So there is no need to apply the corresponding quirk.\n \n Signed-off-by: Stefan Wahren \u003cwahrenst@gmx.net\u003e\n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit ca85ee7457dc60771ec0173bfed58671177a74ab\nAuthor: Stefan Wahren \u003cwahrenst@gmx.net\u003e\nDate: Tue Aug 27 19:04:14 2019 +0200\n\n dt-bindings: i2c: bcm2835: Add brcm,bcm2711 compatible\n \n Add a new compatible for the BCM2711, which hasn\u0027t the clock stretch bug.\n \n Signed-off-by: Stefan Wahren \u003cwahrenst@gmx.net\u003e\n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit 9be02fde93e5b564663f10a48f19159fba718e3d\nAuthor: Fernando Pacheco \u003cfernando.pacheco@intel.com\u003e\nDate: Thu Aug 29 10:41:53 2019 -0700\n\n drm/i915/uc: Extract common code from GuC stop/disable comm\n \n During normal driver unload we attempt to disable GuC communication\n while it is currently stopped. This results in a nop\u0027d call to\n intel_guc_ct_disable within guc_disable_communication because\n stop/disable rely on the same flag to prevent further comms with CT.\n \n We can avoid the call to disable and still leave communication in a\n satisfactory state by extracting a set of shared steps from stop/disable.\n This set can include guc_disable_interrupts as we do not require the\n single caller of guc_stop_communication to be atomic:\n \"drm/i915/selftests: Fixup atomic reset checking\".\n \n This situation (stop -\u003e disable) only occurs during intel_uc_fini_hw,\n so during fini, call guc_disable_communication only if currently enabled.\n The symmetric calls to enable/disable remain unmodified for all other\n scenarios.\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d110943\n Signed-off-by: Fernando Pacheco \u003cfernando.pacheco@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190829174154.14675-1-fernando.pacheco@intel.com\n\ncommit 61c65f47f30a1f32e0a84fe8335cd0360a028b48\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Thu Aug 29 13:17:21 2019 -0500\n\n rpmsg: glink: Use struct_size() helper\n \n One of the more common cases of allocation size calculations is finding\n the size of a structure that has a zero-sized array at the end, along\n with memory for some number of elements for that array. For example:\n \n struct {\n ...\n struct intent_pair intents[];\n } __packed * msg;\n \n Make use of the struct_size() helper instead of an open-coded version\n in order to avoid any potential type mistakes.\n \n So, replace the following form:\n \n sizeof(*msg) + sizeof(struct intent_pair) * count\n \n with:\n \n struct_size(msg, intents, count)\n \n This code was detected with the help of Coccinelle.\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit e7b6affd0baef7b3c5e2acd94236093dcf437105\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 29 18:08:48 2019 +0100\n\n drm/i915/selftests: cond_resched() within the longer buddy tests\n \n Let the scheduler have a breather in between passes of the longer buddy\n tests. Important if we are running under kasan etc and this takes far\n longer than usual!\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190829170848.969-1-chris@chris-wilson.co.uk\n\ncommit cbc0425d3dd370a6f0bf23589dc7b6955a53a9ce\nAuthor: Mimi Zohar \u003czohar@linux.ibm.com\u003e\nDate: Sun Aug 25 09:58:16 2019 -0400\n\n sefltest/ima: support appended signatures (modsig)\n \n In addition to the PE/COFF and IMA xattr signatures, the kexec kernel\n image can be signed with an appended signature, using the same\n scripts/sign-file tool that is used to sign kernel modules.\n \n This patch adds support for detecting a kernel image signed with an\n appended signature and updates the existing test messages\n appropriately.\n \n Reviewed-by: Petr Vorel \u003cpvorel@suse.cz\u003e\n Acked-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n Reviewed-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Reviewed-by: Jordan Hand \u003cjorhand@linux.microsoft.com\u003e (x86_64 QEMU)\n Tested-by: Jordan Hand \u003cjorhand@linux.microsoft.com\u003e (x86_64 QEMU)\n Signed-off-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n\ncommit 87ce243206944a57383309dcbcdcc5750e6c905b\nAuthor: Benjamin Weis \u003cbenjamin.weis@gmx.com\u003e\nDate: Mon Jun 24 15:31:50 2019 +0200\n\n cpupower: update German translation\n \n Update the German translation of cpupower, and change the encoding\n to UTF-8.\n \n [skhan@linuxfoundation.org: fix merge conflicts]\n Signed-off-by: Benjamin Weis \u003cbenjamin.weis@gmx.com\u003e\n Acked-by: Dominik Brodowski \u003clinux@dominikbrodowski.net\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit 3532e7227243beb0b782266dc05c40b6184ad051\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Thu Aug 29 08:53:06 2019 -0700\n\n blkcg: fix missing free on error path of blk_iocost_init()\n \n blk_iocost_init() forgot to free its percpu stat on the error path.\n Fix it.\n \n Fixes: 7caa47151ab2 (\"blkcg: implement blk-iocost\")\n Reported-by: Hillf Danton \u003chdanton@sina.com\u003e\n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 6fd4fe9b496d9ba3382992ff4fde3871d1b6f63d\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Tue Jul 16 15:18:58 2019 +0800\n\n rtc: snvs: fix possible race condition\n \n The RTC IRQ is requested before the struct rtc_device is allocated,\n this may lead to a NULL pointer dereference in IRQ handler.\n \n To fix this issue, allocating the rtc_device struct before requesting\n the RTC IRQ using devm_rtc_allocate_device, and use rtc_register_device\n to register the RTC device.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190716071858.36750-1-Anson.Huang@nxp.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit a73f6e2fbe8077811ea9546e0d44a7533111f0ba\nAuthor: Sébastien Szymanski \u003csebastien.szymanski@armadeus.com\u003e\nDate: Fri Jun 7 17:02:55 2019 +0200\n\n tools/power/cpupower: fix 64bit detection when cross-compiling\n \n When cross-compiling cpupower, 64bit detection is done with the host\n compiler instead of the cross-compiler and libcpupower.so.0 ends up in\n /usr/lib64 instead of /usr/lib for 32bit target. Fix this by moving\n 64bit detection after CC is defined.\n \n Signed-off-by: Sébastien Szymanski \u003csebastien.szymanski@armadeus.com\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit 828f369d76d16bffa97133f6f63a64c13a401953\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Mon Jun 17 16:51:15 2019 +0200\n\n cpupower: Add missing newline at end of file\n \n \"git diff\" says:\n \n \\ No newline at end of file\n \n after modifying the files.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit 8d1c1560c383004e09c6a39498094671cc664e6b\nAuthor: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nDate: Thu Aug 29 09:43:34 2019 -0600\n\n blkcg: blk-iocost: predeclare used structs\n \n Fixes: 7caa47151ab2 (\"blkcg: implement blk-iocost\")\n Acked-by: Tejun Heo \u003ctj@kernel.org\u003e\n Signed-off-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit f155e79ab516c9ed2adb2f337fb4ff58b7474428\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Thu Aug 29 23:38:08 2019 +0900\n\n ALSA: firewire-lib: fix isoc cycle count to which rx packet is scheduled\n \n When introducing the list of packet descriptor, for rx packet, the\n calculation of scheduled isoc cycle is omitted. This commit fixes the\n bug.\n \n Fixes: f4f6ae7b7c1f (\"ALSA: firewire-lib: use packet descriptor for IT context\")\n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 6e26d193d15029a65e0904ad85d6d8634817c3a6\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Thu Aug 29 23:37:19 2019 +0900\n\n ALSA: dice: fix error path for initialization of stream structures\n \n When introducing AMDTP domain to ALSA dice driver, error path does not\n handle error correctly. This commit fixes the bug.\n \n Fixes: e9f21129b8d8 (\"ALSA: dice: support AMDTP domain\")\n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 4ca76945b037cf93648170ff8cb958dbf1040374\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sun Aug 25 10:31:39 2019 +0900\n\n kbuild: clean up subdir-ymn calculation in Makefile.clean\n \n Remove some variables.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 687ac1fa3164c75a3e1053ed8609f3a4e7a6186e\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sun Aug 25 10:31:28 2019 +0900\n\n kbuild: remove unneeded \u0027+\u0027 marker from cmd_clean\n \n This \u0027+\u0027 was added a long time ago:\n \n | commit c23e6bf05f7802e92fd3da69a1ed35e56f9c85bb (HEAD)\n | Author: Kai Germaschewski \u003ckai@tp1.ruhr-uni-bochum.de\u003e\n | Date: Mon Oct 28 01:16:34 2002 -0600\n |\n | kbuild: Fix a \"make -j\u003cN\u003e\" warning\n |\n | diff --git a/scripts/Makefile.clean b/scripts/Makefile.clean\n | index 2c843e0380bc..e7c392fd5788 100644\n | --- a/scripts/Makefile.clean\n | +++ b/scripts/Makefile.clean\n | @@ -42,7 +42,7 @@ quiet_cmd_clean \u003d CLEAN $(obj)\n |\n | __clean: $(subdir-ymn)\n | ifneq ($(strip $(__clean-files) $(clean-rule)),)\n | - $(call cmd,clean)\n | + +$(call cmd,clean)\n | else\n | @:\n | endif\n \n At that time, cmd_clean contained $(clean-rule), which was able to\n invoke sub-make. That was why cleaning with the -j option showed:\n warning: jobserver unavailable: using -j1. Add \u0027+\u0027 to parent make rule.\n \n It is not the case any more; cmd_clean now just runs the \u0027rm\u0027 command.\n The \u0027+\u0027 marker is pointless.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 1634f2bfdb846ed0a8b73131a9dff7c420fb3fe1\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sun Aug 25 10:31:27 2019 +0900\n\n kbuild: remove clean-dirs syntax\n \n The only the difference between clean-files and clean-dirs is the -r\n option passed to the \u0027rm\u0027 command.\n \n You can always pass -r, and then remove the clean-dirs syntax.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 36de077b20d05321466f8eaba0ae01b9b18ad93c\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Thu Aug 22 13:46:13 2019 +0900\n\n kbuild: check clean srctree even earlier\n \n Move the outputmakefile target to the leftmost in the prerequisite list\n so that this is checked first. GNU Make processes the prerequisites\n left to right.\n \n GNU Make will keep to stick to this behavior, and it seems even\n POSIX standard, according to this:\n https://lists.gnu.org/archive/html/bug-make/2019-08/msg00030.html\n \n The POSIX standard of make is available here:\n https://pubs.opengroup.org/onlinepubs/9699919799/utilities/make.html\n \n Of course, when the parallel option -j given, other targets will be run\n simultaneously but it is nice to show the error as early as possible.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit a5139fb368d2639afc6a3ea7e444f22a3e8ed3be\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Thu Aug 22 13:46:12 2019 +0900\n\n kbuild: remove prepare3 target\n \n Now prepare3 does nothing but depends on include/config/kernel.release\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit e8e83a236d36dac639b4846f17fa7c47013b4b95\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Thu Aug 22 13:46:11 2019 +0900\n\n kbuild: move the clean srctree check to the outputmakefile target\n \n With this commit, the error report is shown earlier, even before\n running kconfig.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 1a475d5486755fcf48d710cb7dc3a8a5f7bad573\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Thu Aug 22 13:46:10 2019 +0900\n\n kbuild: clarify where to run make mrproper when out-of-tree fails\n \n If you try out-of-tree build with an unclean source tree, Kbuild\n suggests to run make mrproper. The path to the source tree may be\n shown with a relative path, for example, \"make O\u003dfoo\" emits the\n following:\n \n .. is not clean, please run \u0027make mrproper\u0027\n in the \u0027..\u0027 directory.\n \n This is somewhat confusing if you ran \"make O\u003dfoo\" in the source tree.\n Using the absolute path will be clearer.\n \n This commit changes the error message like follows:\n \n ***\n *** The source tree is not clean, please run \u0027make mrproper\u0027\n *** in /absolute/path/to/linux\n ***\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 621da4ba0e244ba3cfb2e8a9a9055299ec0ab0d3\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Thu Aug 22 13:46:09 2019 +0900\n\n kbuild: Inform user to pass ARCH\u003d for make mrproper only when necessary\n \n Since commit 3a475b2166fd (\"kbuild: Inform user to pass ARCH\u003d for make\n mrproper\"), if you try out-of-tree build with an unclean source tree,\n it suggests to run \u0027make ARCH\u003d\u003cARCH\u003e mrproper\u0027.\n \n This looks odd when you are not cross-compiling the kernel. Show the\n \u0027ARCH\u003d\u003cARCH\u003e\u0027 part only when ARCH\u003d was given from the command line.\n If ARCH is the default (native build) or came from the environment,\n it should simply suggest \u0027make mrproper\u0027 as before.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 59747fb8693fee55d2184295d2a4b8b91e0203ee\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Thu Aug 22 13:46:08 2019 +0900\n\n kbuild: remove \u0027Using ... as source for kernel\u0027 message\n \n You already know the location of the source tree without this message.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 9c3ad4c14f7a3ac870d660bbc5f9961a8a79f356\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Thu Aug 22 12:59:11 2019 +0900\n\n kbuild: get rid of $(realpath ...) from scripts/mkmakefile\n \n Both relative path and absolute path have pros and cons. For example,\n we can move the source and objtree around together by using the\n relative path to the source tree.\n \n Do not force the absolute path to the source tree. If you prefer the\n absolute path, you can specify KBUILD_ABS_SRCTREE\u003d1.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit bc7b752a7a1c8498f5c48f5b1b63147e32f649dc\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Thu Aug 22 02:33:21 2019 +0900\n\n kbuild: remove unneeded dependency for $(DOC_TARGETS)\n \n Commit 415008af3219 (\"docs-rst: convert lsm from DocBook to ReST\")\n stopped using if_changed_rule. There is no more users of if_changed*\n for the doc targets. Hence, fixdep is unneeded. Remove the dependency\n on scripts_basic.\n \n All the doc targets are phony. The dependency on FORCE is not needed\n either.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit fc01adc41679b19ee35a79e2bd2e9176aeba20c8\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Thu Aug 22 02:32:47 2019 +0900\n\n kbuild: remove unneeded comments and code from scripts/basic/Makefile\n \n Kbuild descends into scripts/basic/ even before the Kconfig.\n I do not expect any other host programs added to this Makefile.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit eb3aabbfbfc203082d06a64517df97a3746ba9ea\nAuthor: Andrew Murray \u003candrew.murray@arm.com\u003e\nDate: Wed Aug 28 18:50:09 2019 +0100\n\n arm64: atomics: Remove atomic_ll_sc compilation unit\n \n We no longer fall back to out-of-line atomics on systems with\n CONFIG_ARM64_LSE_ATOMICS where ARM64_HAS_LSE_ATOMICS is not set.\n \n Remove the unused compilation unit which provided these symbols.\n \n Signed-off-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 3337cb5aea594e4090a660e3fc3250bb669b1305\nAuthor: Andrew Murray \u003candrew.murray@arm.com\u003e\nDate: Wed Aug 28 18:50:08 2019 +0100\n\n arm64: avoid using hard-coded registers for LSE atomics\n \n Now that we have removed the out-of-line ll/sc atomics we can give\n the compiler the freedom to choose its own register allocation.\n \n Remove the hard-coded use of x30.\n \n Signed-off-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit addfc38672c73efd5c4e559a2e455b086e3e20c5\nAuthor: Andrew Murray \u003candrew.murray@arm.com\u003e\nDate: Wed Aug 28 18:50:07 2019 +0100\n\n arm64: atomics: avoid out-of-line ll/sc atomics\n \n When building for LSE atomics (CONFIG_ARM64_LSE_ATOMICS), if the hardware\n or toolchain doesn\u0027t support it the existing code will fallback to ll/sc\n atomics. It achieves this by branching from inline assembly to a function\n that is built with special compile flags. Further this results in the\n clobbering of registers even when the fallback isn\u0027t used increasing\n register pressure.\n \n Improve this by providing inline implementations of both LSE and\n ll/sc and use a static key to select between them, which allows for the\n compiler to generate better atomics code. Put the LL/SC fallback atomics\n in their own subsection to improve icache performance.\n \n Signed-off-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit db91427b6502e8e46db4b616e4eaa9b9cf4e6363\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Aug 26 09:22:13 2019 +0200\n\n MIPS: document mixing \"slightly different CCAs\"\n \n Based on an email from Paul Burton, quoting section 4.8 \"Cacheability and\n Coherency Attributes and Access Types\" of \"MIPS Architecture Volume 1:\n Introduction to the MIPS32 Architecture\" (MD00080, revision 6.01).\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Acked-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit 3e4e1d3fb89193cc072858e1469d6f2926c603f7\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Sat Aug 3 12:38:31 2019 +0300\n\n arm64: document the choice of page attributes for pgprot_dmacoherent\n \n Based on an email from Will Deacon.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Acked-by: Will Deacon \u003cwill@kernel.org\u003e\n Acked-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n\ncommit 8e3a68fb55e00e0760bd8023883e064f1f93c62d\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Sat Aug 3 12:42:15 2019 +0300\n\n dma-mapping: make dma_atomic_pool_init self-contained\n \n The memory allocated for the atomic pool needs to have the same\n mapping attributes that we use for remapping, so use\n pgprot_dmacoherent instead of open coding it. Also deduct a\n suitable zone to allocate the memory from based on the presence\n of the DMA zones.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit 419e2f1838819e954071dfa1d1f820ab3386ada1\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Aug 26 09:03:44 2019 +0200\n\n dma-mapping: remove arch_dma_mmap_pgprot\n \n arch_dma_mmap_pgprot is used for two things:\n \n 1) to override the \"normal\" uncached page attributes for mapping\n memory coherent to devices that can\u0027t snoop the CPU caches\n 2) to provide the special DMA_ATTR_WRITE_COMBINE semantics on older\n arm systems and some mips platforms\n \n Replace one with the pgprot_dmacoherent macro that is already provided\n by arm and much simpler to use, and lift the DMA_ATTR_WRITE_COMBINE\n handling to common code with an explicit arch opt-in.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Acked-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e # m68k\n Acked-by: Paul Burton \u003cpaul.burton@mips.com\u003e # mips\n\ncommit deffa48fb014f06f7cf8c4b3ea3c96160be3c854\nAuthor: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\nDate: Thu Aug 29 15:16:35 2019 +0200\n\n s390/zcrypt: fix wrong handling of cca cipher keygenflags\n \n Tests showed that the keygenflags parameter is not handled\n correctly within the zcrypt ccamisc generate cca cipher key\n code. A similar code is used with cca cipher key value import\n and there the flags are handled correctly. For unknown reason\n these lines have not been updated for the generate function\n and so this patch now introduces these two lines of code.\n \n This affects only pkey with the use of CCA cipher keys and\n the use of additional key generate flags.\n \n Fixes: 4bc123b18ce6 (\"s390/zcrypt: Add low level functions for CCA AES cipher keys\")\n Signed-off-by: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 580fa1b874711d633f9b145b7777b0e83ebf3787\nAuthor: Andrew Murray \u003candrew.murray@arm.com\u003e\nDate: Wed Aug 28 18:50:06 2019 +0100\n\n arm64: Use correct ll/sc atomic constraints\n \n The A64 ISA accepts distinct (but overlapping) ranges of immediates for:\n \n * add arithmetic instructions (\u0027I\u0027 machine constraint)\n * sub arithmetic instructions (\u0027J\u0027 machine constraint)\n * 32-bit logical instructions (\u0027K\u0027 machine constraint)\n * 64-bit logical instructions (\u0027L\u0027 machine constraint)\n \n ... but we currently use the \u0027I\u0027 constraint for many atomic operations\n using sub or logical instructions, which is not always valid.\n \n When CONFIG_ARM64_LSE_ATOMICS is not set, this allows invalid immediates\n to be passed to instructions, potentially resulting in a build failure.\n When CONFIG_ARM64_LSE_ATOMICS is selected the out-of-line ll/sc atomics\n always use a register as they have no visibility of the value passed by\n the caller.\n \n This patch adds a constraint parameter to the ATOMIC_xx and\n __CMPXCHG_CASE macros so that we can pass appropriate constraints for\n each case, with uses updated accordingly.\n \n Unfortunately prior to GCC 8.1.0 the \u0027K\u0027 constraint erroneously accepted\n \u00274294967295\u0027, so we must instead force the use of a register.\n \n Signed-off-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 8f35eaa5f2de020073a48ad51112237c5932cfcc\nAuthor: Andrew Murray \u003candrew.murray@arm.com\u003e\nDate: Wed Aug 28 18:50:05 2019 +0100\n\n jump_label: Don\u0027t warn on __exit jump entries\n \n On architectures that discard .exit.* sections at runtime, a\n warning is printed for each jump label that is used within an\n in-kernel __exit annotated function:\n \n can\u0027t patch jump_label at ehci_hcd_cleanup+0x8/0x3c\n WARNING: CPU: 0 PID: 1 at kernel/jump_label.c:410 __jump_label_update+0x12c/0x138\n \n As these functions will never get executed (they are free\u0027d along\n with the rest of initmem) - we do not need to patch them and should\n not display any warnings.\n \n The warning is displayed because the test required to satisfy\n jump_entry_is_init is based on init_section_contains (__init_begin to\n __init_end) whereas the test in __jump_label_update is based on\n init_kernel_text (_sinittext to _einittext) via kernel_text_address).\n \n Fixes: 19483677684b (\"jump_label: Annotate entries that operate on __init code earlier\")\n Signed-off-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Acked-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit be6f55a60d28fb31864d3aab011cf4e1fc0dba04\nAuthor: Thomas Huth \u003cthuth@redhat.com\u003e\nDate: Thu Aug 29 15:07:32 2019 +0200\n\n KVM: selftests: Add a test for the KVM_S390_MEM_OP ioctl\n \n Check that we can write and read the guest memory with this s390x\n ioctl, and that some error cases are handled correctly.\n \n Signed-off-by: Thomas Huth \u003cthuth@redhat.com\u003e\n Link: https://lkml.kernel.org/r/20190829130732.580-1-thuth@redhat.com\n Signed-off-by: Christian Borntraeger \u003cborntraeger@de.ibm.com\u003e\n\ncommit b4d863c302e65515cc4706d184f1d54e6bbfc5a2\nAuthor: Cornelia Huck \u003ccohuck@redhat.com\u003e\nDate: Thu Aug 29 14:47:46 2019 +0200\n\n KVM: s390: improve documentation for S390_MEM_OP\n \n Explicitly specify the valid ranges for size and ar, and reword\n buf requirements a bit.\n \n Signed-off-by: Cornelia Huck \u003ccohuck@redhat.com\u003e\n Reviewed-by: Thomas Huth \u003cthuth@redhat.com\u003e\n Reviewed-by: David Hildenbrand \u003cdavid@redhat.com\u003e\n Link: https://lkml.kernel.org/r/20190829124746.28665-1-cohuck@redhat.com\n Signed-off-by: Christian Borntraeger \u003cborntraeger@de.ibm.com\u003e\n\ncommit a13b03bbb4575b350b46090af4dfd30e735aaed1\nAuthor: Thomas Huth \u003cthuth@redhat.com\u003e\nDate: Thu Aug 29 14:25:17 2019 +0200\n\n KVM: s390: Test for bad access register and size at the start of S390_MEM_OP\n \n If the KVM_S390_MEM_OP ioctl is called with an access register \u003e\u003d 16,\n then there is certainly a bug in the calling userspace application.\n We check for wrong access registers, but only if the vCPU was already\n in the access register mode before (i.e. the SIE block has recorded\n it). The check is also buried somewhere deep in the calling chain (in\n the function ar_translation()), so this is somewhat hard to find.\n \n It\u0027s better to always report an error to the userspace in case this\n field is set wrong, and it\u0027s safer in the KVM code if we block wrong\n values here early instead of relying on a check somewhere deep down\n the calling chain, so let\u0027s add another check to kvm_s390_guest_mem_op()\n directly.\n \n We also should check that the \"size\" is non-zero here (thanks to Janosch\n Frank for the hint!). If we do not check the size, we could call vmalloc()\n with this 0 value, and this will cause a kernel warning.\n \n Signed-off-by: Thomas Huth \u003cthuth@redhat.com\u003e\n Link: https://lkml.kernel.org/r/20190829122517.31042-1-thuth@redhat.com\n Reviewed-by: Cornelia Huck \u003ccohuck@redhat.com\u003e\n Reviewed-by: Janosch Frank \u003cfrankja@linux.ibm.com\u003e\n Reviewed-by: David Hildenbrand \u003cdavid@redhat.com\u003e\n Cc: stable@vger.kernel.org\n Signed-off-by: Christian Borntraeger \u003cborntraeger@de.ibm.com\u003e\n\ncommit bdf8613d75ea7b4f3eb0fd0bd55342fb25e7286b\nAuthor: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nDate: Wed Aug 28 15:55:29 2019 -0700\n\n sparc: prefer __section from compiler_attributes.h\n \n GCC unescapes escaped string section names while Clang does not. Because\n __section uses the `#` stringification operator for the section name, it\n doesn\u0027t need to be escaped.\n \n Instead, we should:\n 1. Prefer __section(.section_name_no_quotes).\n 2. Only use __attribute__((__section__(\".section\"))) when creating the\n section name via C preprocessor (see the definition of __define_initcall\n in arch/um/include/shared/init.h).\n \n This antipattern was found with:\n $ grep -e __section\\(\\\" -e __section__\\(\\\" -r\n \n See the discussions in:\n Link: https://bugs.llvm.org/show_bug.cgi?id\u003d42950\n Link: https://marc.info/?l\u003dlinux-netdev\u0026m\u003d156412960619946\u0026w\u003d2\n Link: https://github.com/ClangBuiltLinux/linux/issues/619\n Acked-by: David S. Miller \u003cdavem@davemloft.net\u003e\n Reported-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Suggested-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Signed-off-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Miguel Ojeda \u003cmiguel.ojeda.sandonis@gmail.com\u003e\n\ncommit ed20d0bcc6de0f7d1f8a773ef0e2a855a205da45\nAuthor: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nDate: Wed Aug 28 15:55:28 2019 -0700\n\n mips: prefer __section from compiler_attributes.h\n \n GCC unescapes escaped string section names while Clang does not. Because\n __section uses the `#` stringification operator for the section name, it\n doesn\u0027t need to be escaped.\n \n Instead, we should:\n 1. Prefer __section(.section_name_no_quotes).\n 2. Only use __attribute__((__section__(\".section\"))) when creating the\n section name via C preprocessor (see the definition of __define_initcall\n in arch/um/include/shared/init.h).\n \n This antipattern was found with:\n $ grep -e __section\\(\\\" -e __section__\\(\\\" -r\n \n See the discussions in:\n Link: https://bugs.llvm.org/show_bug.cgi?id\u003d42950\n Link: https://marc.info/?l\u003dlinux-netdev\u0026m\u003d156412960619946\u0026w\u003d2\n Link: https://github.com/ClangBuiltLinux/linux/issues/619\n Acked-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Reported-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Suggested-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Signed-off-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Miguel Ojeda \u003cmiguel.ojeda.sandonis@gmail.com\u003e\n\ncommit 2dfd7326c89cd20a8fefec6896c04286c696e1f2\nAuthor: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nDate: Wed Aug 28 15:55:27 2019 -0700\n\n arm: prefer __section from compiler_attributes.h\n \n GCC unescapes escaped string section names while Clang does not. Because\n __section uses the `#` stringification operator for the section name, it\n doesn\u0027t need to be escaped.\n \n Instead, we should:\n 1. Prefer __section(.section_name_no_quotes).\n 2. Only use __attribute__((__section__(\".section\"))) when creating the\n section name via C preprocessor (see the definition of __define_initcall\n in arch/um/include/shared/init.h).\n \n This antipattern was found with:\n $ grep -e __section\\(\\\" -e __section__\\(\\\" -r\n \n See the discussions in:\n Link: https://bugs.llvm.org/show_bug.cgi?id\u003d42950\n Link: https://marc.info/?l\u003dlinux-netdev\u0026m\u003d156412960619946\u0026w\u003d2\n Link: https://github.com/ClangBuiltLinux/linux/issues/619\n Reported-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Suggested-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Signed-off-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Miguel Ojeda \u003cmiguel.ojeda.sandonis@gmail.com\u003e\n\ncommit 90929d70f1153956dc801ca028a7e82a971ca5b9\nAuthor: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nDate: Wed Aug 28 15:55:26 2019 -0700\n\n ia64: prefer __section from compiler_attributes.h\n \n GCC unescapes escaped string section names while Clang does not. Because\n __section uses the `#` stringification operator for the section name, it\n doesn\u0027t need to be escaped.\n \n Instead, we should:\n 1. Prefer __section(.section_name_no_quotes).\n 2. Only use __attribute__((__section__(\".section\"))) when creating the\n section name via C preprocessor (see the definition of __define_initcall\n in arch/um/include/shared/init.h).\n \n This antipattern was found with:\n $ grep -e __section\\(\\\" -e __section__\\(\\\" -r\n \n See the discussions in:\n Link: https://bugs.llvm.org/show_bug.cgi?id\u003d42950\n Link: https://marc.info/?l\u003dlinux-netdev\u0026m\u003d156412960619946\u0026w\u003d2\n Link: https://github.com/ClangBuiltLinux/linux/issues/619\n Reported-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Suggested-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Signed-off-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Miguel Ojeda \u003cmiguel.ojeda.sandonis@gmail.com\u003e\n\ncommit 2e54e97cd5bbc8191291e5b6151558e7feb4b820\nAuthor: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nDate: Wed Aug 28 15:55:25 2019 -0700\n\n um: prefer __section from compiler_attributes.h\n \n GCC unescapes escaped string section names while Clang does not. Because\n __section uses the `#` stringification operator for the section name, it\n doesn\u0027t need to be escaped.\n \n Instead, we should:\n 1. Prefer __section(.section_name_no_quotes).\n 2. Only use __attribute__((__section__(\".section\"))) when creating the\n section name via C preprocessor (see the definition of __define_initcall\n in arch/um/include/shared/init.h).\n \n This antipattern was found with:\n $ grep -e __section\\(\\\" -e __section__\\(\\\" -r\n \n See the discussions in:\n Link: https://bugs.llvm.org/show_bug.cgi?id\u003d42950\n Link: https://marc.info/?l\u003dlinux-netdev\u0026m\u003d156412960619946\u0026w\u003d2\n Link: https://github.com/ClangBuiltLinux/linux/issues/619\n Reported-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Suggested-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Signed-off-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Miguel Ojeda \u003cmiguel.ojeda.sandonis@gmail.com\u003e\n\ncommit a43f3a6b2127c3f2c746f5814f06a59a1b076b49\nAuthor: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nDate: Wed Aug 28 15:55:24 2019 -0700\n\n parisc: prefer __section from compiler_attributes.h\n \n GCC unescapes escaped string section names while Clang does not. Because\n __section uses the `#` stringification operator for the section name, it\n doesn\u0027t need to be escaped.\n \n Instead, we should:\n 1. Prefer __section(.section_name_no_quotes).\n 2. Only use __attribute__((__section__(\".section\"))) when creating the\n section name via C preprocessor (see the definition of __define_initcall\n in arch/um/include/shared/init.h).\n \n This antipattern was found with:\n $ grep -e __section\\(\\\" -e __section__\\(\\\" -r\n \n See the discussions in:\n Link: https://bugs.llvm.org/show_bug.cgi?id\u003d42950\n Link: https://marc.info/?l\u003dlinux-netdev\u0026m\u003d156412960619946\u0026w\u003d2\n Link: https://github.com/ClangBuiltLinux/linux/issues/619\n Reported-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Suggested-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Signed-off-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Miguel Ojeda \u003cmiguel.ojeda.sandonis@gmail.com\u003e\n\ncommit c97e82b97f4bba00304905fe7965f923abd2d755\nAuthor: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nDate: Wed Aug 28 15:55:23 2019 -0700\n\n include/linux/compiler.h: prefer __section from compiler_attributes.h\n \n GCC unescapes escaped string section names while Clang does not. Because\n __section uses the `#` stringification operator for the section name, it\n doesn\u0027t need to be escaped.\n \n This fixes an Oops observed in distro\u0027s that use systemd and not\n net.core.bpf_jit_enable\u003d1, when their kernels are compiled with Clang.\n \n Instead, we should:\n 1. Prefer __section(.section_name_no_quotes).\n 2. Only use __attribute__((__section__(\".section\"))) when creating the\n section name via C preprocessor (see the definition of __define_initcall\n in arch/um/include/shared/init.h).\n \n This antipattern was found with:\n $ grep -e __section\\(\\\" -e __section__\\(\\\" -r\n \n See the discussions in:\n Link: https://bugs.llvm.org/show_bug.cgi?id\u003d42950\n Link: https://marc.info/?l\u003dlinux-netdev\u0026m\u003d156412960619946\u0026w\u003d2\n Link: https://github.com/ClangBuiltLinux/linux/issues/619\n Acked-by: Will Deacon \u003cwill@kernel.org\u003e\n Reported-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Suggested-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Tested-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Signed-off-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Miguel Ojeda \u003cmiguel.ojeda.sandonis@gmail.com\u003e\n\ncommit 1faa6d6339f3d5f82a3f6f6a409416cc6beb47c2\nAuthor: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nDate: Wed Aug 28 15:55:22 2019 -0700\n\n s390/boot: fix section name escaping\n \n GCC unescapes escaped string section names while Clang does not. Because\n __section uses the `#` stringification operator for the section name, it\n doesn\u0027t need to be escaped.\n \n Instead, we should:\n 1. Prefer __section(.section_name_no_quotes).\n 2. Only use __attribute__((__section__(\".section\"))) when creating the\n section name via C preprocessor (see the definition of __define_initcall\n in arch/um/include/shared/init.h).\n \n This antipattern was found with:\n $ grep -e __section\\(\\\" -e __section__\\(\\\" -r\n \n See the discussions in:\n Link: https://bugs.llvm.org/show_bug.cgi?id\u003d42950\n Link: https://marc.info/?l\u003dlinux-netdev\u0026m\u003d156412960619946\u0026w\u003d2\n Link: https://github.com/ClangBuiltLinux/linux/issues/619\n Reported-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Suggested-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Signed-off-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Miguel Ojeda \u003cmiguel.ojeda.sandonis@gmail.com\u003e\n\ncommit 042c1d29de8c0d4e4c9e77c64995b59be97aed33\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Tue Aug 20 19:45:47 2019 +0200\n\n s390/kasan: add kdump support\n \n If kasan enabled kernel is used as crash kernel it crashes itself with\n program check loop during kdump execution. The reason for that is that\n kasan shadow memory backed by pages beyond OLDMEM_SIZE. Make kasan memory\n allocator respect physical memory limit imposed by kdump.\n \n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit d0b319843baddc1a224ccdc73729aed55ec7787a\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Mon Aug 19 17:32:44 2019 +0200\n\n s390/setup: avoid using strncmp with hardcoded length\n \n Replace strncmp usage in console mode setup code with simple strcmp.\n Replace strncmp which is used for prefix comparison with str_has_prefix.\n \n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 54fb07d030e1b2fa9d903a748f84398bbe6f213d\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Mon Aug 19 17:13:04 2019 +0200\n\n s390/sclp: avoid using strncmp with hardcoded length\n \n \"earlyprintk\" option documentation does not clearly state which\n platform supports which additional values (e.g. \",keep\"). Preserve old\n option behaviour and reuse str_has_prefix instead of strncmp for prefix\n testing.\n \n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit b29cd7c4c482d895f4b13ef1dfe85550529f38c9\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Mon Aug 19 17:05:44 2019 +0200\n\n s390/module: avoid using strncmp with hardcoded length\n \n Reuse str_has_prefix instead of strncmp with hardcoded length to\n make the intent of a comparison more obvious.\n \n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 8dec6bd4ef327f48eb790a48cba0e833109d4e41\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Mon Aug 19 16:29:01 2019 +0200\n\n s390/pci: avoid using strncmp with hardcoded length\n \n Command line option values passed to __setup callbacks are always\n null-terminated and \"s390_iommu\u003d\" may only accept \"strict\" as value.\n So replace strncmp with strcmp.\n \n While at it also make s390_iommu_setup return 1, which means this\n command line option is handled by this callback.\n \n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 968bce2f59ce462d296af28610628fe7e03b120a\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 28 22:22:41 2019 -0300\n\n media: imx: remove unused including \u003clinux/version.h\u003e\n \n Remove including \u003clinux/version.h\u003e that don\u0027t need it.\n \n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 90499a230d16e48282fc78591fd33461d8741fb6\nAuthor: Markus Elfring \u003celfring@users.sourceforge.net\u003e\nDate: Tue Aug 27 11:00:13 2019 -0300\n\n media: stm32-dcmi: Delete an unnecessary of_node_put() call in dcmi_probe()\n \n A null pointer would be passed to a call of the function “of_node_put”\n immediately after a call of the function “of_graph_get_next_endpoint”\n failed at one place.\n Remove this superfluous function call.\n \n This issue was detected by using the Coccinelle software.\n \n Signed-off-by: Markus Elfring \u003celfring@users.sourceforge.net\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 0fb68ce02ae73626a61b8d03eab1625021057af5\nAuthor: Yizhuo \u003cyzhai003@ucr.edu\u003e\nDate: Mon Aug 26 20:55:28 2019 -0300\n\n media: pvrusb2: qctrl.flag will be uninitlaized if cx2341x_ctrl_query() returns error code\n \n Inside function ctrl_cx2341x_getv4lflags(), qctrl.flag\n will be uninitlaized if cx2341x_ctrl_query() returns -EINVAL.\n However, it will be used in the later if statement, which is\n potentially unsafe.\n \n Signed-off-by: Yizhuo \u003cyzhai003@ucr.edu\u003e\n Acked-by: Mike Isely \u003cisely@pobox.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit ecbce48f1ff2442371ebcd12ec0ecddb431fbd72\nAuthor: Markus Elfring \u003celfring@users.sourceforge.net\u003e\nDate: Mon Aug 26 08:14:02 2019 -0300\n\n media: em28xx: Fix exception handling in em28xx_alloc_urbs()\n \n A null pointer would be passed to a call of the function \"kfree\" directly\n after a call of the function \"kcalloc\" failed at one place.\n Pass the data structure member \"urb\" instead for which memory\n was allocated before (so that this resource will be properly cleaned up).\n \n This issue was detected by using the Coccinelle software.\n \n Fixes: d571b592c6206d33731f41aa710fa0f69ac8611b (\"media: em28xx: don\u0027t use coherent buffer for DMA transfers\")\n Signed-off-by: Markus Elfring \u003celfring@users.sourceforge.net\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 333f31436d3db19f4286f8862a00ea1d8d8420a1\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Thu Aug 29 09:52:02 2019 +0200\n\n ALSA: hda - Fix potential endless loop at applying quirks\n \n Since the chained quirks via chained_before flag is applied before the\n depth check, it may lead to the endless recursive calls, when the\n chain were set up incorrectly. Fix it by moving the depth check at\n the beginning of the loop.\n \n Fixes: 1f57825077dc (\"ALSA: hda - Add chained_before flag to the fixup entry\")\n Cc: \u003cstable@vger.kernel.org\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit f480b969448ee3896f55ef6d037ad4a2df3914bf\nAuthor: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\nDate: Tue Aug 27 12:36:53 2019 +0200\n\n mtd: rawnand: omap2: Fix number of bitflips reporting with ELM\n \n omap_elm_correct_data() returns the number of bitflips for the whole\n page. This is wrong, it should return the maximum number of bitflips\n found in each ECC step.\n \n In my case with a 4k page size NAND mtcdore reported -EUCLEAN with\n only 12 bitflips on a page where we could correct up to 128 bits per\n page (provided they are distributed equally on the 8 ECC steps)\n \n Signed-off-by: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n\ncommit 7f852cc1579297fd763789f8cd370639d0c654b6\nAuthor: Claire Lin \u003cclaire.lin@broadcom.com\u003e\nDate: Mon Aug 26 15:57:56 2019 -0400\n\n mtd: rawnand: brcmnand: Fix ecc chunk calculation for erased page bitfips\n \n In brcmstb_nand_verify_erased_page(), the ECC chunk pointer calculation\n while correcting erased page bitflips is wrong, fix it.\n \n Fixes: 02b88eea9f9c (\"mtd: brcmnand: Add check for erased page bitflips\")\n Signed-off-by: Claire Lin \u003cclaire.lin@broadcom.com\u003e\n Reviewed-by: Ray Jui \u003cray.jui@broadcom.com\u003e\n Signed-off-by: Kamal Dasu \u003ckdasu.kdev@gmail.com\u003e\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n\ncommit 301011ba622513cb41ced59973972204e0da2f71\nAuthor: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\nDate: Wed Aug 28 15:05:29 2019 -0400\n\n tools lib traceevent: Remove unneeded qsort and uses memmove instead\n \n While reading a trace data file that had 100,000s of tasks, the process\n took an extremely long time. I profiled it down to add_new_comm(), which\n was doing a qsort() call on an array that was pretty much already sorted\n (all but the last element. qsort() isn\u0027t very efficient when dealing\n with mostly sorted arrays, and this definitely showed its issues.\n \n When adding a new task to the task list, instead of using qsort(), do\n another bsearch() with a function that will find the element before\n where the new task will be inserted in. Then simply shift the rest of\n the array, and insert the task where it belongs.\n \n Fixes: f7d82350e597d (\"tools/events: Add files to create libtraceevent.a\")\n Signed-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n Cc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: linux-trace-devel@vger.kernel.org\n Link: http://lkml.kernel.org/r/20190828191820.127233764@goodmis.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit b0215e2d6a18d8331b2d4a8b38ccf3eff783edb1\nAuthor: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\nDate: Wed Aug 28 15:05:28 2019 -0400\n\n tools lib traceevent: Do not free tep-\u003ecmdlines in add_new_comm() on failure\n \n If the re-allocation of tep-\u003ecmdlines succeeds, then the previous\n allocation of tep-\u003ecmdlines will be freed. If we later fail in\n add_new_comm(), we must not free cmdlines, and also should assign\n tep-\u003ecmdlines to the new allocation. Otherwise when freeing tep, the\n tep-\u003ecmdlines will be pointing to garbage.\n \n Fixes: a6d2a61ac653a (\"tools lib traceevent: Remove some die() calls\")\n Signed-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n Cc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: linux-trace-devel@vger.kernel.org\n Cc: stable@vger.kernel.org\n Link: http://lkml.kernel.org/r/20190828191819.970121417@goodmis.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit b397f8468fa27f08b83b348ffa56a226f72453af\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 28 16:48:50 2019 -0300\n\n perf evlist: Use unshare(CLONE_FS) in sb threads to let setns(CLONE_NEWNS) work\n \n When we started using a thread to catch the PERF_RECORD_BPF_EVENT meta\n data events to then ask the kernel for further info (BTF, etc) for BPF\n programs shortly after they get loaded, we forgot to use\n unshare(CLONE_FS) as was done in:\n \n 868a832918f6 (\"perf top: Support lookup of symbols in other mount namespaces.\")\n \n Do it so that we can enter the namespaces to read the build-ids at the\n end of a \u0027perf record\u0027 session for the DSOs that had hits.\n \n Before:\n \n Starting a \u0027stress-ng --cpus 8\u0027 inside a container and then, outside the\n container running:\n \n # perf record -a --namespaces sleep 5\n # perf buildid-list | grep stress-ng\n #\n \n We would end up with a \u0027perf.data\u0027 file that had no entry in its\n build-id table for the /usr/bin/stress-ng binary inside the container\n that got tons of PERF_RECORD_SAMPLEs.\n \n After:\n \n # perf buildid-list | grep stress-ng\n f2ed02c68341183a124b9b0f6e2e6c493c465b29 /usr/bin/stress-ng\n #\n \n Then its just a matter of making sure that that binary debuginfo package\n gets available in a place that \u0027perf report\u0027 will look at build-id keyed\n ELF files, which, in my case, on a f30 notebook, was a matter of\n installing the debuginfo file for the distro used in the container,\n fedora 31:\n \n # rpm -ivh http://fedora.c3sl.ufpr.br/linux/development/31/Everything/x86_64/debug/tree/Packages/s/stress-ng-debuginfo-0.07.29-10.fc31.x86_64.rpm\n \n Then, because perf currently looks for those debuginfo files (richer ELF\n symtab) inside that namespace (look at the setns calls):\n \n openat(AT_FDCWD, \"/proc/self/ns/mnt\", O_RDONLY) \u003d 137\n openat(AT_FDCWD, \"/proc/13169/ns/mnt\", O_RDONLY) \u003d 139\n setns(139, CLONE_NEWNS) \u003d 0\n stat(\"/usr/bin/stress-ng\", {st_mode\u003dS_IFREG|0755, st_size\u003d3065416, ...}) \u003d 0\n openat(AT_FDCWD, \"/usr/bin/stress-ng\", O_RDONLY) \u003d 140\n fcntl(140, F_GETFD) \u003d 0\n fstat(140, {st_mode\u003dS_IFREG|0755, st_size\u003d3065416, ...}) \u003d 0\n mmap(NULL, 3065416, PROT_READ, MAP_PRIVATE, 140, 0) \u003d 0x7ff2fdc5b000\n munmap(0x7ff2fdc5b000, 3065416) \u003d 0\n close(140) \u003d 0\n stat(\"stress-ng-0.07.29-10.fc31.x86_64.debug\", 0x7fff45d71260) \u003d -1 ENOENT (No such file or directory)\n stat(\"/usr/bin/stress-ng-0.07.29-10.fc31.x86_64.debug\", 0x7fff45d71260) \u003d -1 ENOENT (No such file or directory)\n stat(\"/usr/bin/.debug/stress-ng-0.07.29-10.fc31.x86_64.debug\", 0x7fff45d71260) \u003d -1 ENOENT (No such file or directory)\n stat(\"/usr/lib/debug/usr/bin/stress-ng-0.07.29-10.fc31.x86_64.debug\", 0x7fff45d71260) \u003d -1 ENOENT (No such file or directory)\n stat(\"/root/.debug/.build-id/f2/ed02c68341183a124b9b0f6e2e6c493c465b29\", 0x7fff45d711e0) \u003d -1 ENOENT (No such file or directory)\n \n To only then go back to the \"host\" namespace to look just in the users\u0027s\n ~/.debug cache:\n \n setns(137, CLONE_NEWNS) \u003d 0\n chdir(\"/root\") \u003d 0\n close(137) \u003d 0\n close(139) \u003d 0\n stat(\"/root/.debug/.build-id/f2/ed02c68341183a124b9b0f6e2e6c493c465b29/elf\", 0x7fff45d732e0) \u003d -1 ENOENT (No such file or directory)\n \n It continues to fail to resolve symbols:\n \n # perf report | grep stress-ng | head -5\n 9.50% stress-ng-cpu stress-ng [.] 0x0000000000021ac1\n 8.58% stress-ng-cpu stress-ng [.] 0x0000000000021ab4\n 8.51% stress-ng-cpu stress-ng [.] 0x0000000000021489\n 7.17% stress-ng-cpu stress-ng [.] 0x00000000000219b6\n 3.93% stress-ng-cpu stress-ng [.] 0x0000000000021478\n #\n \n To overcome that we use:\n \n # perf buildid-cache -v --add /usr/lib/debug/usr/bin/stress-ng-0.07.29-10.fc31.x86_64.debug\n Adding f2ed02c68341183a124b9b0f6e2e6c493c465b29 /usr/lib/debug/usr/bin/stress-ng-0.07.29-10.fc31.x86_64.debug: Ok\n #\n # ls -la /root/.debug/.build-id/f2/ed02c68341183a124b9b0f6e2e6c493c465b29/elf\n -rw-r--r--. 3 root root 2401184 Jul 27 07:03 /root/.debug/.build-id/f2/ed02c68341183a124b9b0f6e2e6c493c465b29/elf\n # file /root/.debug/.build-id/f2/ed02c68341183a124b9b0f6e2e6c493c465b29/elf\n /root/.debug/.build-id/f2/ed02c68341183a124b9b0f6e2e6c493c465b29/elf: ELF 64-bit LSB shared object, x86-64, version 1 (SYSV), dynamically linked, interpreter \\004, BuildID[sha1]\u003df2ed02c68341183a124b9b0f6e2e6c493c465b29, for GNU/Linux 3.2.0, with debug_info, not stripped, too many notes (256)\n #\n \n Now it finally works:\n \n # perf report | grep stress-ng | head -5\n 23.59% stress-ng-cpu stress-ng [.] ackermann\n 23.33% stress-ng-cpu stress-ng [.] is_prime\n 17.36% stress-ng-cpu stress-ng [.] stress_cpu_sieve\n 6.08% stress-ng-cpu stress-ng [.] stress_cpu_correlate\n 3.55% stress-ng-cpu stress-ng [.] queens_try\n #\n \n I\u0027ll make sure that it looks for the build-id keyed files in both the\n \"host\" namespace (the namespace the user running \u0027perf record\u0027 was a the\n time of the recording) and in the container namespace, as it shouldn\u0027t\n matter where a content based key lookup finds the ELF file to use in\n resolving symbols, etc.\n \n Reported-by: Karl Rister \u003ckrister@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexei Starovoitov \u003cast@kernel.org\u003e\n Cc: Brendan Gregg \u003cbrendan.d.gregg@gmail.com\u003e\n Cc: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n Cc: Krister Johansen \u003ckjlx@templeofstupid.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Song Liu \u003csongliubraving@fb.com\u003e\n Cc: Stanislav Fomichev \u003csdf@google.com\u003e\n Cc: Thomas-Mich Richter \u003ctmricht@linux.vnet.ibm.com\u003e\n Fixes: 657ee5531903 (\"perf evlist: Introduce side band thread\")\n Link: https://lkml.kernel.org/n/tip-g79k0jz41adiaeuqud742t2l@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 653dd8e6e8e46591f864b7ad98e10819079d5a88\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:57:17 2019 +0200\n\n libperf: Move \u0027enum perf_user_event_type\u0027 to perf/event.h\n \n So it\u0027s available for libperf\u0027s users.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-24-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 72932371e78012cea96edb9e833d81f1c32dd892\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:57:16 2019 +0200\n\n libperf: Rename the PERF_RECORD_ structs to have a \"perf\" prefix\n \n Even more, to have a \"perf_record_\" prefix, so that they match the\n PERF_RECORD_ enum they map to.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-23-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 7510410a38c71eb5d45217a4934e60eef88c04e1\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:57:15 2019 +0200\n\n libperf: Add \u0027union perf_event\u0027 to perf/event.h\n \n So it\u0027s available for libperf\u0027s users.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-22-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit f5f684321791eb5ebb8c56d7d223e71fe08b5dd9\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:57:14 2019 +0200\n\n libperf: Add PERF_RECORD_COMPRESSED \u0027struct compressed_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_COMPRESSED event definition to libperf\u0027s event.h.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-21-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 1b8896fb296f4087c45c997c4b212544c639e272\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:57:13 2019 +0200\n\n libperf: Add PERF_RECORD_HEADER_FEATURE \u0027struct feature_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_HEADER_FEATURE event definition to libperf\u0027s\n event.h.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-20-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit bfd922d8f09a692f2a952a67451bc3eeaad3fb73\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:57:12 2019 +0200\n\n libperf: Add PERF_RECORD_TIME_CONV \u0027struct time_conv_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_TIME_CONV event definition to libperf\u0027s event.h.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-19-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 782adbe2964953803ea1a38b80f6255c336cdd7b\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:57:11 2019 +0200\n\n libperf: Add PERF_RECORD_STAT_ROUND \u0027struct stat_round_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_STAT_ROUND event definition to libperf\u0027s event.h.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-18-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 18a13a60f6f50f2fb1f7354f6d0b2ded01995443\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:57:10 2019 +0200\n\n libperf: Add PERF_RECORD_STAT \u0027struct stat_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_STAT event definition to libperf\u0027s event.h.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-17-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit c5f416e6c69e333207666a1ddab0b41c6f12e588\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:57:09 2019 +0200\n\n libperf: Add PERF_RECORD_STAT_CONFIG \u0027struct stat_config_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_STAT_CONFIG event definition to libperf\u0027s event.h.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-16-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 3e4c453f5cbac75afec37eedd0ffcd8dac82e23e\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:57:08 2019 +0200\n\n libperf: Add PERF_RECORD_THREAD_MAP \u0027struct thread_map_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_THREAD_MAP event definition to libperf\u0027s event.h.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-15-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 6b49aaebd05fdbb29a4ab6e28b4fa91539c64877\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:57:07 2019 +0200\n\n libperf: Add PERF_RECORD_SWITCH \u0027struct context_switch_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_SWITCH event definition to libperf\u0027s event.h.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-14-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit f279ad63a09da65766dfeaf03cfd659f95414936\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:57:06 2019 +0200\n\n libperf: Add PERF_RECORD_ITRACE_START \u0027struct itrace_start_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_ITRACE_START event definition to libperf\u0027s event.h.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-13-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit aedebdca09ca6efa7efbc0bf26d94cb235120ee4\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:57:05 2019 +0200\n\n libperf: Add PERF_RECORD_AUX \u0027struct aux_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_AUX event definition to libperf\u0027s event.h.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-12-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 3460efb2e842cccc4566756f194a1be0547f7098\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:57:04 2019 +0200\n\n libperf: Add PERF_RECORD_AUXTRACE_ERROR \u0027struct auxtrace_error_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_AUXTRACE_ERROR event definition to libperf\u0027s\n event.h.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-11-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 306c9d24c09d84d80ae54b36f7f907c8b8fa537a\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:57:03 2019 +0200\n\n libperf: Add PERF_RECORD_AUXTRACE \u0027struct auxtrace_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_AUXTRACE event definition to libperf\u0027s event.h.\n \n Ipn order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027\n types used events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-10-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 9a8dad0419552934573ddf94d11146faeda465b5\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:57:02 2019 +0200\n\n libperf: Add PERF_RECORD_AUXTRACE_INFO \u0027struct auxtrace_info_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_AUXTRACE_INFO event definition to libperf\u0027s\n event.h.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-9-jolsa@kernel.org\n [ Fix cs_etm__print_auxtrace_info() arg to be __u64 too to fix the CORESIGHT\u003d1 build ]\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit a2ed4fd685cd23e98922f933d5dbccfbe82a4f08\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Thu Aug 29 12:52:28 2019 +0200\n\n posix-cpu-timers: Make expiry_active check actually work correctly\n \n The state tracking changes broke the expiry active check by not writing to\n it and instead sitting timers_active, which is already set.\n \n That\u0027s not a big issue as the actual expiry is protected by sighand lock,\n so concurrent handling is not possible. That means that the second task\n which invokes that function executes the expiry code for nothing.\n \n Write to the proper flag.\n \n Also add a check whether the flag is set into check_process_timers(). That\n check had been missing in the code before the rework already. The check for\n another task handling the expiry of process wide timers was only done in\n the fastpath check. If the fastpath check returns true because a per task\n timer expired, then the checking of process wide timers was done in\n parallel which is as explained above just a waste of cycles.\n \n Fixes: 244d49e30653 (\"posix-cpu-timers: Move state tracking to struct posix_cputimers\")\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n\ncommit 3a960339e08e168c03f0587b788f05fadf4e396d\nAuthor: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\nDate: Wed Aug 28 10:35:17 2019 +0000\n\n mtd: spi-nor: remove superfluous pass of nor-\u003einfo-\u003esector_size\n \n We already pass a pointer to nor, we can obtain the sector_size\n by dereferencing it.\n \n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit bc605cd79edb68131d3be5b00b949aa312277d39\nAuthor: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\nDate: Thu Aug 29 18:44:17 2019 +1000\n\n powerpc/of/pci: Rewrite pci_parse_of_flags\n \n The existing code uses bunch of hardcoded values from the PCI Bus\n Binding to IEEE Std 1275 spec; and it does so in quite non-obvious\n way.\n \n This defines fields from the cell#0 of the \"reg\" property of a PCI\n device and uses them for parsing.\n \n This should cause no behavioral change.\n \n Signed-off-by: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\n [mpe: Unsplit some 80/81 char lines, space the code with some newlines]\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190829084417.71873-1-aik@ozlabs.ru\n\ncommit d9430f96c051ede4e4a3d5a20e68518181e9cd3c\nAuthor: Thierry Reding \u003ctreding@nvidia.com\u003e\nDate: Thu Aug 29 09:29:32 2019 +0200\n\n driver core: platform: Export platform_get_irq_optional()\n \n This function can be used by modules, so it needs to be exported.\n \n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 807f6c8472da716e29744e1fa35480b08af3e4bc\nAuthor: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nDate: Sat Aug 24 00:23:53 2019 +0300\n\n drm/ingenic: Hardcode panel type to DPI\n \n The ingenic driver supports DPI panels only at the moment, so hardcode\n their type to DPI instead of Unknown.\n \n Signed-off-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823212353.29369-1-laurent.pinchart@ideasonboard.com\n # *** extracted tags ***\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n\ncommit 6bbf87a131edd39040974f3843ff97e89310ebde\nAuthor: Pawel Laszczak \u003cpawell@cadence.com\u003e\nDate: Mon Aug 26 12:19:31 2019 +0100\n\n usb:cdns3 Fix for stuck packets in on-chip OUT buffer.\n \n Controller for OUT endpoints has shared on-chip buffers for all incoming\n packets, including ep0out. It\u0027s FIFO buffer, so packets must be handled\n by DMA in correct order. If the first packet in the buffer will not be\n handled, then the following packets directed for other endpoints and\n functions will be blocked.\n \n Additionally the packets directed to one endpoint can block entire on-chip\n buffers. In this case transfer to other endpoints also will blocked.\n \n To resolve this issue after raising the descriptor missing interrupt\n driver prepares internal usb_request object and use it to arm DMA\n transfer.\n \n The problematic situation was observed in case when endpoint has\n been enabled but no usb_request were queued. Driver try detects\n such endpoints and will use this workaround only for these endpoint.\n \n Driver use limited number of buffer. This number can be set by macro\n CDNS_WA2_NUM_BUFFERS.\n \n Such blocking situation was observed on ACM gadget. For this function\n host send OUT data packet but ACM function is not prepared for\n this packet. It\u0027s cause that buffer placed in on chip memory block\n transfer to other endpoints.\n \n Issue has been fixed for DEV_VER_V2 version of controller.\n \n Signed-off-by: Pawel Laszczak \u003cpawell@cadence.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 7733f6c32e36ff9d7adadf40001039bf219b1cbe\nAuthor: Pawel Laszczak \u003cpawell@cadence.com\u003e\nDate: Mon Aug 26 12:19:30 2019 +0100\n\n usb: cdns3: Add Cadence USB3 DRD Driver\n \n This patch introduce new Cadence USBSS DRD driver to Linux kernel.\n \n The Cadence USBSS DRD Controller is a highly configurable IP Core which\n can be instantiated as Dual-Role Device (DRD), Peripheral Only and\n Host Only (XHCI)configurations.\n \n The current driver has been validated with FPGA platform. We have\n support for PCIe bus, which is used on FPGA prototyping.\n \n The host side of USBSS-DRD controller is compliant with XHCI\n specification, so it works with standard XHCI Linux driver.\n \n Signed-off-by: Pawel Laszczak \u003cpawell@cadence.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit f72429fc88b5f48ee9ff2617224bbe098e40e86f\nAuthor: Pawel Laszczak \u003cpawell@cadence.com\u003e\nDate: Mon Aug 26 12:19:29 2019 +0100\n\n usb: common: Simplify usb_decode_get_set_descriptor function.\n \n Patch moves switch responsible for decoding descriptor type\n outside snprintf. It improves code readability a little.\n \n Signed-off-by: Pawel Laszczak \u003cpawell@cadence.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 010665b44d596d1b2aa0249b0e093fd3df774f6e\nAuthor: Pawel Laszczak \u003cpawell@cadence.com\u003e\nDate: Mon Aug 26 12:19:28 2019 +0100\n\n usb: common: Patch simplify usb_decode_set_clear_feature function.\n \n Patch adds usb_decode_test_mode and usb_decode_device_feature functions,\n which allow to make more readable and simplify the\n usb_decode_set_clear_feature function.\n \n Signed-off-by: Pawel Laszczak \u003cpawell@cadence.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 91f255a26bce80d27b5f3dcbb0333b852b5dbe2c\nAuthor: Pawel Laszczak \u003cpawell@cadence.com\u003e\nDate: Mon Aug 26 12:19:27 2019 +0100\n\n usb: common: Separated decoding functions from dwc3 driver.\n \n Patch moves some decoding functions from driver/usb/dwc3/debug.h driver\n to driver/usb/common/debug.c file. These moved functions include:\n dwc3_decode_get_status\n dwc3_decode_set_clear_feature\n dwc3_decode_set_address\n dwc3_decode_get_set_descriptor\n dwc3_decode_get_configuration\n dwc3_decode_set_configuration\n dwc3_decode_get_intf\n dwc3_decode_set_intf\n dwc3_decode_synch_frame\n dwc3_decode_set_sel\n dwc3_decode_set_isoch_delay\n dwc3_decode_ctrl\n \n These functions are used also in inroduced cdns3 driver.\n \n All functions prefixes were changed from dwc3 to usb.\n Also, function\u0027s parameters has been extended according to the name\n of fields in standard SETUP packet.\n Additionally, patch adds usb_decode_ctrl function to\n include/linux/usb/ch9.h file.\n \n Signed-off-by: Pawel Laszczak \u003cpawell@cadence.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit a7a3a6a941cf53d49c5f40529b669304d328caf4\nAuthor: Pawel Laszczak \u003cpawell@cadence.com\u003e\nDate: Mon Aug 26 12:19:26 2019 +0100\n\n dt-bindings: add binding for USBSS-DRD controller.\n \n This patch aim at documenting USB related dt-bindings for the\n Cadence USBSS-DRD controller.\n \n Signed-off-by: Pawel Laszczak \u003cpawell@cadence.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit dcc935b06f1f29aa9f93008df3d13ab84ab0bbbb\nAuthor: Zhuohao Lee \u003czhuohao@chromium.org\u003e\nDate: Wed Aug 28 09:08:02 2019 +0300\n\n mtd: spi-nor: enable the debugfs for the partname and partid\n \n This patch adds spi_nor_debugfs_init() for the debugfs initialization.\n With this patch, we can read the partname and partid through the\n debugfs.\n \n The output of new debugfs nodes on my device are:\n cat /sys/kernel/debug/mtd/mtd0/partid\n spi-nor:ef6017\n cat /sys/kernel/debug/mtd/mtd0/partname\n w25q64dw\n \n Signed-off-by: Zhuohao Lee \u003czhuohao@chromium.org\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n\ncommit 1018c94be6ea073115f6bcf993d6492138d2b8e3\nAuthor: Zhuohao Lee \u003czhuohao@chromium.org\u003e\nDate: Mon Jul 1 00:07:10 2019 +0800\n\n mtd: mtdcore: add debugfs nodes for querying the flash name and id\n \n Currently, we don\u0027t have vfs nodes for querying the underlying flash name\n and flash id. This information is important especially when we want to\n know the flash detail of the defective system. In order to support the\n query, we add mtd_debugfs_populate() to create two debugfs nodes\n (ie. partname and partid). The upper driver can assign the pointer to\n partname and partid before calling mtd_device_register().\n \n Signed-off-by: Zhuohao Lee \u003czhuohao@chromium.org\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n\ncommit 23d103ae3e061b466dc68dd10a035dcd3b2de99a\nAuthor: Chris Packham \u003cchris.packham@alliedtelesis.co.nz\u003e\nDate: Fri Jul 12 05:46:58 2019 +0100\n\n ARM: 8891/1: EDAC: armada_xp: Add support for more SoCs\n \n The Armada 38x and other integrated SoCs use a reduced pin count so the\n width of the SDRAM interface is smaller than the Armada XP SoCs. This\n means that the definition of \"full\" and \"half\" width is reduced from\n 64/32 to 32/16.\n \n Signed-off-by: Chris Packham \u003cchris.packham@alliedtelesis.co.nz\u003e\n Signed-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\n\ncommit 7f6998a41257a8930ee5b6866ba56a25230841ed\nAuthor: Jan Luebbe \u003cjlu@pengutronix.de\u003e\nDate: Fri Jul 12 05:46:57 2019 +0100\n\n ARM: 8888/1: EDAC: Add driver for the Marvell Armada XP SDRAM and L2 cache ECC\n \n Add support for the ECC functionality as found in the DDR RAM and L2\n cache controllers on the MV78230/MV78x60 SoCs. This driver has been\n tested on the MV78460 (on a custom board with a DDR3 ECC DIMM).\n \n [cp use SPDX license]\n \n Signed-off-by: Jan Luebbe \u003cjlu@pengutronix.de\u003e\n Signed-off-by: Chris Packham \u003cchris.packham@alliedtelesis.co.nz\u003e\n Reviewed-by: Borislav Petkov \u003cbp@suse.de\u003e\n Signed-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\n\ncommit 0ecace04a3b8444f3ffa6f18083e98198d9b9ec6\nAuthor: Jan Luebbe \u003cjlu@pengutronix.de\u003e\nDate: Fri Jul 12 05:46:56 2019 +0100\n\n ARM: 8892/1: EDAC: Add missing debugfs_create_x32 wrapper\n \n We already have wrappers for x8 and x16, so add the missing x32 one.\n \n Signed-off-by: Jan Luebbe \u003cjlu@pengutronix.de\u003e\n Reviewed-by: Borislav Petkov \u003cbp@suse.de\u003e\n Signed-off-by: Chris Packham \u003cchris.packham@alliedtelesis.co.nz\u003e\n Signed-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\n\ncommit c8abbd6f9dc47dffaea5adc7172ffa59a16401e3\nAuthor: Chris Packham \u003cchris.packham@alliedtelesis.co.nz\u003e\nDate: Fri Jul 12 05:46:55 2019 +0100\n\n ARM: 8890/1: l2x0: add marvell,ecc-enable property for aurora\n \n The aurora cache on the Marvell Armada-XP SoC supports ECC protection\n for the L2 data arrays. Add a \"marvell,ecc-enable\" device tree property\n which can be used to enable this.\n \n [jlu@pengutronix.de: use aurora specific define AURORA_ACR_ECC_EN]\n \n Signed-off-by: Chris Packham \u003cchris.packham@alliedtelesis.co.nz\u003e\n Signed-off-by: Jan Luebbe \u003cjlu@pengutronix.de\u003e\n Signed-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\n\ncommit 4bf4770db4f0dbf495ca6236644abb413501f6e0\nAuthor: Chris Packham \u003cchris.packham@alliedtelesis.co.nz\u003e\nDate: Fri Jul 12 05:46:54 2019 +0100\n\n ARM: 8889/1: dt-bindings: document marvell,ecc-enable binding\n \n Add documentation for the marvell,ecc-enable properties which can be\n used to enable ECC on the Marvell aurora cache.\n \n Signed-off-by: Chris Packham \u003cchris.packham@alliedtelesis.co.nz\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\n\ncommit fd3bbde717b00a2db75d0c93264f412c1176008f\nAuthor: Chris Packham \u003cchris.packham@alliedtelesis.co.nz\u003e\nDate: Fri Jul 12 05:46:53 2019 +0100\n\n ARM: 8886/1: l2x0: support parity-enable/disable on aurora\n \n The aurora cache on the Marvell Armada-XP SoC supports the same tag\n parity features as the other l2x0 cache implementations.\n \n [jlu@pengutronix.de: use aurora specific define AURORA_ACR_PARITY_EN]\n \n Signed-off-by: Chris Packham \u003cchris.packham@alliedtelesis.co.nz\u003e\n Signed-off-by: Jan Luebbe \u003cjlu@pengutronix.de\u003e\n Signed-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\n\ncommit 0770bc9214065e384080075ab9fcada8ad07038c\nAuthor: Jan Luebbe \u003cjlu@pengutronix.de\u003e\nDate: Fri Jul 12 05:46:52 2019 +0100\n\n ARM: 8885/1: aurora-l2: add defines for parity and ECC registers\n \n These defines will be used by subsequent patches to add support for the\n parity check and error correction functionality in the Aurora L2 cache\n controller.\n \n Signed-off-by: Jan Luebbe \u003cjlu@pengutronix.de\u003e\n Signed-off-by: Chris Packham \u003cchris.packham@alliedtelesis.co.nz\u003e\n Signed-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\n\ncommit 1a85cb4b0d2984a6d1afb03e2038855d654c9892\nAuthor: Jan Luebbe \u003cjlu@pengutronix.de\u003e\nDate: Fri Jul 12 05:46:51 2019 +0100\n\n ARM: 8887/1: aurora-l2: add prefix to MAX_RANGE_SIZE\n \n The macro name is too generic, so add a AURORA_ prefix.\n \n Signed-off-by: Jan Luebbe \u003cjlu@pengutronix.de\u003e\n Reviewed-by: Gregory CLEMENT \u003cgregory.clement@free-electrons.com\u003e\n Signed-off-by: Chris Packham \u003cchris.packham@alliedtelesis.co.nz\u003e\n Signed-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\n\ncommit 921a3fe5be2dd0aac70405faf32a970716f55e03\nAuthor: Jan Luebbe \u003cjlu@pengutronix.de\u003e\nDate: Tue Aug 27 21:49:22 2019 +0100\n\n ARM: 8902/1: l2c: move cache-aurora-l2.h to asm/hardware\n \n This include file will be used by the AURORA EDAC code.\n \n Signed-off-by: Jan Luebbe \u003cjlu@pengutronix.de\u003e\n Reviewed-by: Gregory CLEMENT \u003cgregory.clement@free-electrons.com\u003e\n Signed-off-by: Chris Packham \u003cchris.packham@alliedtelesis.co.nz\u003e\n Signed-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\n\ncommit 6dc5fd93b2f1ef75d5e50fced8cb193811f25f22\nAuthor: Nathan Huckleberry \u003cnhuck15@gmail.com\u003e\nDate: Thu Aug 22 21:26:53 2019 +0100\n\n ARM: 8900/1: UNWINDER_FRAME_POINTER implementation for Clang\n \n The stackframe setup when compiled with clang is different.\n Since the stack unwinder expects the gcc stackframe setup it\n fails to print backtraces. This patch adds support for the\n clang stackframe setup.\n \n Link: https://github.com/ClangBuiltLinux/linux/issues/35\n \n Cc: clang-built-linux@googlegroups.com\n Suggested-by: Tri Vo \u003ctrong@google.com\u003e\n Signed-off-by: Nathan Huckleberry \u003cnhuck@google.com\u003e\n Tested-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Reviewed-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\n\ncommit 8f2edb4a78f7f5fa35c025849152b1d2dfaee4eb\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Thu Aug 29 08:19:40 2019 +0200\n\n posix-timers: Unbreak CONFIG_POSIX_TIMERS\u003dn build\n \n The rework of the posix-cpu-timers patch series dropped the empty\n declaration of struct cpu_timer for the CONFIG_POSIX_TIMERS\u003dn case which\n causes the build to fail:\n \n ./include/linux/posix-timers.h:218:20: error: field \u0027cpu\u0027 has incomplete type\n \n Add it back.\n \n Fixes: 60bda037f1dd (\"posix-cpu-timers: Utilize timerqueue for storage\")\n Reported-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n\ncommit 8504dea783b044cab620acbaef87b86ee84646fe\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Wed Aug 28 15:06:00 2019 -0700\n\n blkcg: add tools/cgroup/iocost_coef_gen.py\n \n Add a script which can be used to generate device-specific iocost\n linear model coefficients.\n \n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 6954ff185ee0811cdd2e0f388ff4dd7df17f11af\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Wed Aug 28 15:05:59 2019 -0700\n\n blkcg: add tools/cgroup/iocost_monitor.py\n \n Instead of mucking with debugfs and -\u003epd_stat(), add drgn based\n monitoring script.\n \n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Cc: Omar Sandoval \u003cosandov@fb.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 7caa47151ab2e644dd221f741ec7578d9532c9a3\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Wed Aug 28 15:05:58 2019 -0700\n\n blkcg: implement blk-iocost\n \n This patchset implements IO cost model based work-conserving\n proportional controller.\n \n While io.latency provides the capability to comprehensively prioritize\n and protect IOs depending on the cgroups, its protection is binary -\n the lowest latency target cgroup which is suffering is protected at\n the cost of all others. In many use cases including stacking multiple\n workload containers in a single system, it\u0027s necessary to distribute\n IO capacity with better granularity.\n \n One challenge of controlling IO resources is the lack of trivially\n observable cost metric. The most common metrics - bandwidth and iops\n - can be off by orders of magnitude depending on the device type and\n IO pattern. However, the cost isn\u0027t a complete mystery. Given\n several key attributes, we can make fairly reliable predictions on how\n expensive a given stream of IOs would be, at least compared to other\n IO patterns.\n \n The function which determines the cost of a given IO is the IO cost\n model for the device. This controller distributes IO capacity based\n on the costs estimated by such model. The more accurate the cost\n model the better but the controller adapts based on IO completion\n latency and as long as the relative costs across differents IO\n patterns are consistent and sensible, it\u0027ll adapt to the actual\n performance of the device.\n \n Currently, the only implemented cost model is a simple linear one with\n a few sets of default parameters for different classes of device.\n This covers most common devices reasonably well. All the\n infrastructure to tune and add different cost models is already in\n place and a later patch will also allow using bpf progs for cost\n models.\n \n Please see the top comment in blk-iocost.c and documentation for\n more details.\n \n v2: Rebased on top of RQ_ALLOC_TIME changes and folded in Rik\u0027s fix\n for a divide-by-zero bug in current_hweight() triggered by zero\n inuse_sum.\n \n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Cc: Andy Newell \u003cnewella@fb.com\u003e\n Cc: Josef Bacik \u003cjbacik@fb.com\u003e\n Cc: Rik van Riel \u003criel@surriel.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 6f816b4b746c2241540e537682d30d8e9997d674\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Wed Aug 28 15:05:57 2019 -0700\n\n blk-mq: add optional request-\u003ealloc_time_ns\n \n There are currently two start time timestamps - start_time_ns and\n io_start_time_ns. The former marks the request allocation and and the\n second issue-to-device time. The planned io.weight controller needs\n to measure the total time bios take to execute after it leaves rq_qos\n including the time spent waiting for request to become available,\n which can easily dominate on saturated devices.\n \n This patch adds request-\u003ealloc_time_ns which records when the request\n allocation attempt started. As it isn\u0027t used for the usual stats,\n make it optional behind CONFIG_BLK_RQ_ALLOC_TIME and\n QUEUE_FLAG_RQ_ALLOC_TIME so that it can be compiled out when there are\n no users and it\u0027s active only on queues which need it even when\n compiled in.\n \n v2: s/pre_start_time/alloc_time/ and add CONFIG_BLK_RQ_ALLOC_TIME\n gating as suggested by Jens.\n \n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit beab17fc2a507e85dd18b3cef83820c5770c5f34\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Wed Aug 28 15:05:56 2019 -0700\n\n blkcg: s/RQ_QOS_CGROUP/RQ_QOS_LATENCY/\n \n io.weight is gonna be another rq_qos cgroup mechanism. Let\u0027s rename\n RQ_QOS_CGROUP which is being used by io.latency to RQ_QOS_LATENCY in\n preparation.\n \n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 9677a3e01f838622d2efc9a3ccb97090a2c3156a\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Wed Aug 28 15:05:55 2019 -0700\n\n block/rq_qos: implement rq_qos_ops-\u003equeue_depth_changed()\n \n wbt already gets queue depth changed notification through\n wbt_set_queue_depth(). Generalize it into\n rq_qos_ops-\u003equeue_depth_changed() so that other rq_qos policies can\n easily hook into the events too.\n \n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit d3e65ffff61c329fb2d0bf15736c440c2d0cfc97\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Wed Aug 28 15:05:54 2019 -0700\n\n block/rq_qos: add rq_qos_merge()\n \n Add a merge hook for rq_qos. This will be used by io.weight.\n \n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 015d254cb02b6d8eec4b3366274bf4672f9e0b64\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Wed Aug 28 15:05:53 2019 -0700\n\n blkcg: separate blkcg_conf_get_disk() out of blkg_conf_prep()\n \n Separate out blkcg_conf_get_disk() so that it can be used by blkcg\n policy interface file input parsers before the policy is actually\n enabled. This doesn\u0027t introduce any functional changes.\n \n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 86a5bba5c252e90d264c7460e29a0b9e633777e7\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Wed Aug 28 15:05:52 2019 -0700\n\n blkcg: make -\u003ecpd_init_fn() optional\n \n For policies which can do enough initialization from -\u003ecpd_alloc_fn(),\n make -\u003ecpd_init_fn() optional.\n \n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit cf09a8ee19ad1f78b4e18cdde9f2a61133efacf5\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Wed Aug 28 15:05:51 2019 -0700\n\n blkcg: pass @q and @blkcg into blkcg_pol_alloc_pd_fn()\n \n Instead of @node, pass in @q and @blkcg so that the alloc function has\n more context. This doesn\u0027t cause any behavior change and will be used\n by io.weight implementation.\n \n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit dc4f3eb08a123918e63843c6b10759598e7baf31\nAuthor: Petr Machata \u003cpetrm@mellanox.com\u003e\nDate: Wed Aug 28 18:54:37 2019 +0300\n\n mlxsw: spectrum_ptp: Add counters for GC events\n \n On Spectrum-1, timestamped PTP packets and the corresponding timestamps need to\n be kept in caches until both are available, at which point they are matched up\n and packets forwarded as appropriate. However, not all packets will ever see\n their timestamp, and not all timestamps will ever see their packet. It is\n necessary to dispose of such abandoned entries, so a garbage collector was\n introduced in commit 5d23e4159772 (\"mlxsw: spectrum: PTP: Garbage-collect\n unmatched entries\").\n \n If these GC events happen often, it is a sign of a problem. However because this\n whole mechanism is taking place behind the scenes, there is no direct way to\n determine whether garbage collection took place.\n \n Therefore to fix this, on Spectrum-1 only, expose four artificial ethtool\n counters for the GC events: GCd timestamps and packets, in TX and RX directions.\n \n Cc: Richard Cochran \u003crichardcochran@gmail.com\u003e\n Signed-off-by: Petr Machata \u003cpetrm@mellanox.com\u003e\n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 45bd634131fd17d5761876e09f81b921534bc2c4\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Wed Aug 28 18:54:36 2019 +0300\n\n mlxsw: Bump firmware version to 13.2000.1886\n \n The new version supports extended error reporting from firmware via a\n new TLV in the EMAD packet. Similar to netlink extended ack.\n \n It also fixes an issue in the PCI code that can result in false AER\n errors under high Tx rate.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 3f61967f4197547fb1888642d5178a64ca9fb191\nAuthor: Shalom Toledo \u003cshalomt@mellanox.com\u003e\nDate: Wed Aug 28 18:54:35 2019 +0300\n\n mlxsw: spectrum: Prevent auto negotiation on number of lanes\n \n After 50G-1-lane and 100G-2-lanes link modes were introduced, the driver\n is facing situations in which the hardware auto negotiates not only on\n speed and type, but also on number of lanes.\n \n Prevent auto negotiation on number of lanes by allowing only port speeds\n that can be supported on a given port according to its width.\n \n Signed-off-by: Shalom Toledo \u003cshalomt@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b97cd891268de9ccc5454c189232c74f262961ae\nAuthor: Amit Cohen \u003camitc@mellanox.com\u003e\nDate: Wed Aug 28 18:54:34 2019 +0300\n\n mlxsw: Remove 56G speed support\n \n Commit 275e928f1911 (\"mlxsw: spectrum: Prevent force of 56G\") prevented\n the driver from setting a speed of 56G when auto-negotiation is off.\n This is the only speed supported by mlxsw that cannot be set when\n auto-negotiation is off, which makes it difficult to write generic\n tests.\n \n Further, the speed is not supported by newer ASICs such as Spectrum-2\n and to the best of our knowledge it is not used by current users.\n \n Therefore, remove 56G support from mlxsw.\n \n Signed-off-by: Amit Cohen \u003camitc@mellanox.com\u003e\n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 99d7a74110ef4c97f921b90fa8d897c6d71610a9\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Fri Aug 23 01:20:39 2019 -0700\n\n drm/i915/tgl: PSR link standby is not supported anymore\n \n According to BSpc if link standby is set on TGL+, PSR will not be\n enabled. Vendors should not use panels that requires link standby and\n even if they do, panel should assert a PSR error that will cause PSR to\n be disabled.\n \n BSpec: 50434\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Anshuman Gupta \u003canshuman.gupta@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823082055.5992-8-lucas.demarchi@intel.com\n\ncommit 074c77e3ec636fa5e2664e3c79aba397ec569761\nAuthor: Dhinakaran Pandiyan \u003cdhinakaran.pandiyan@intel.com\u003e\nDate: Tue Aug 27 01:45:16 2019 -0700\n\n drm/i915/tgl: Gen-12 display loses Yf tiling and legacy CCS support\n \n Yf tiling was removed in gen-12, so do not expose Yf modifiers to user\n space. Gen-12 display also is incompatible with pre-gen12 Y-tiled\n CCS, so do not expose I915_FORMAT_MOD_Y_TILED_CCS.\n \n v2: Rebase to carry forward recently added gen11 formats.\n \n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Stanislav Lisovskiy \u003cstanislav.lisovskiy@intel.com\u003e\n Signed-off-by: Dhinakaran Pandiyan \u003cdhinakaran.pandiyan@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190827084516.6748-1-dhinakaran.pandiyan@intel.com\n\ncommit 95fb8bb3181bbe1ee87c95e91dff94f74f148c33\nAuthor: Jian Shen \u003cshenjian15@huawei.com\u003e\nDate: Wed Aug 28 09:34:47 2019 +0800\n\n net: phy: force phy suspend when calling phy_stop\n \n Some ethernet drivers may call phy_start() and phy_stop() from\n ndo_open() and ndo_close() respectively.\n \n When network cable is unconnected, and operate like below:\n step 1: ifconfig ethX up -\u003e ndo_open -\u003e phy_start -\u003estart\n autoneg, and phy is no link.\n step 2: ifconfig ethX down -\u003e ndo_close -\u003e phy_stop -\u003e just stop\n phy state machine.\n \n This patch forces phy suspend even phydev-\u003elink is off.\n \n Signed-off-by: Jian Shen \u003cshenjian15@huawei.com\u003e\n Reviewed-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 502a4440c0c1f94b32734744e4ab0947d53d31c6\nAuthor: Logan Gunthorpe \u003clogang@deltatee.com\u003e\nDate: Thu Aug 22 10:10:13 2019 -0600\n\n PCI: Force trailing new line to resource_alignment_param in sysfs\n \n When \u0027pci\u003dresource_alignment\u003d\u0027 is specified on the command line, there is\n no trailing new line. Then, when it\u0027s read through the corresponding sysfs\n attribute, there will be no newline and a cat command will not show\n correctly in a shell. If the parameter is set through sysfs a new line will\n be stored and it will \u0027cat\u0027 correctly.\n \n To solve this, append a new line character in the show function if one does\n not already exist.\n \n Link: https://lore.kernel.org/r/20190822161013.5481-4-logang@deltatee.com\n Signed-off-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit f137553186751a9484d9e9130e759d539dcfa520\nAuthor: Logan Gunthorpe \u003clogang@deltatee.com\u003e\nDate: Thu Aug 22 10:10:12 2019 -0600\n\n PCI: Move pci_[get|set]_resource_alignment_param() into their callers\n \n Both the functions pci_get_resource_alignment_param() and\n pci_set_resource_alignment_param() are now only called in one place:\n resource_alignment_show() and resource_alignment_store() respectively.\n \n There is no value in this extra set of functions so move both into their\n callers respectively.\n \n Link: https://lore.kernel.org/r/20190822161013.5481-3-logang@deltatee.com\n Signed-off-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit 41b5ef225daa9933540f2c554b0d35cc9d50e613\nAuthor: Logan Gunthorpe \u003clogang@deltatee.com\u003e\nDate: Thu Aug 22 10:10:11 2019 -0600\n\n PCI: Clean up resource_alignment parameter to not require static buffer\n \n Clean up the \u0027resource_alignment\u0027 parameter code to use kstrdup() in the\n initcall routine instead of a static buffer that wastes memory regardless\n of whether the feature is used. This allows us to drop \u0027COMMAND_LINE_SIZE\u0027\n bytes (typically 256-4096 depending on architecture) of static data.\n \n This is similar to what has been done for the \u0027disable_acs_redir\u0027\n parameter.\n \n Link: https://lore.kernel.org/r/20190822161013.5481-2-logang@deltatee.com\n Signed-off-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit 4a92352dd1ce6b6358c3d79730dab2e2f9f6cdc2\nAuthor: Krzysztof Wilczynski \u003ckw@linux.com\u003e\nDate: Mon Aug 26 17:14:36 2019 +0200\n\n PCI: Use static const struct, not const static struct\n \n Move the static keyword to the front of declarations of pci_regs_behavior[]\n and pcie_cap_regs_behavior[], which resolves compiler warnings when\n building with \"W\u003d1\":\n \n drivers/pci/pci-bridge-emul.c:41:1: warning: ‘static’ is not at beginning of\n declaration [-Wold-style-declaration]\n const static struct pci_bridge_reg_behavior pci_regs_behavior[] \u003d {\n ^\n drivers/pci/pci-bridge-emul.c:176:1: warning: ‘static’ is not at beginning of\n declaration [-Wold-style-declaration]\n const static struct pci_bridge_reg_behavior pcie_cap_regs_behavior[] \u003d {\n ^\n \n Link: https://lore.kernel.org/r/20190826151436.4672-1-kw@linux.com\n Link: https://lore.kernel.org/r/20190828131733.5817-1-kw@linux.com\n Signed-off-by: Krzysztof Wilczynski \u003ckw@linux.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Acked-by: Thomas Petazzoni \u003cthomas.petazzoni@bootlin.com\u003e\n\ncommit 3894793e4b1ae85e7af38cc5ab65de8749b9442a\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Tue Aug 27 21:50:32 2019 +0800\n\n phy: mdio-sun4i: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit bd51ce0583e20a6e962bfdbb62f396ff6a61d7b2\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Tue Aug 27 21:49:40 2019 +0800\n\n phy: mdio-mux-meson-g12a: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ea7076923bde9c3c692f5c8d8a46b04ec3d00680\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Tue Aug 27 21:48:04 2019 +0800\n\n phy: mdio-moxart: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ba869d3c40fc130649ba81326087802ec436c84f\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Tue Aug 27 21:47:22 2019 +0800\n\n phy: mdio-hisi-femac: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit bd301e05ba06548ef5afcd28d601dd3acf63fc92\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Tue Aug 27 21:46:16 2019 +0800\n\n phy: mdio-bcm-iproc: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Reviewed-by: Ray Jui \u003cray.jui@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a3a90244c471854bbe79884fd5a48b84953faa75\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Tue Aug 27 12:47:39 2019 +0100\n\n wimax/i2400m: remove redundant assignment to variable result\n \n Variable result is being assigned a value that is never read and result\n is being re-assigned a little later on. The assignment is redundant\n and hence can be removed.\n \n Addresses-Coverity: (\"Ununsed value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 14105c191e09b698782026827dbac966cbd30446\nAuthor: Eric Dumazet \u003cedumazet@google.com\u003e\nDate: Tue Aug 27 00:08:12 2019 -0700\n\n ipv6: shrink struct ipv6_mc_socklist\n \n Remove two holes on 64bit arches, to bring the size\n to one cache line exactly.\n \n Signed-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 146fa39943b1cf4ac60f5e1f90cfec9367d9469a\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Tue Aug 20 14:40:33 2019 -0500\n\n dt-bindings: Convert Arm Mali Utgard GPU to DT schema\n \n Convert the Arm Utgard GPU binding to DT schema format.\n \n \u0027allwinner,sun8i-a23-mali\u0027 compatible was not documented, so add it.\n \n The \u0027clocks\u0027 property is now required. This simplifies the schema as\n effectively all the users require \u0027clocks\u0027 already and the upstream\n driver requires clocks.\n \n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Acked-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 842b4aecfff6acec2f1abcde920a22df20830447\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Tue Aug 20 13:29:32 2019 -0500\n\n dt-bindings: Convert Arm Mali Bifrost GPU to DT schema\n \n Convert the Arm Bifrost GPU binding to DT schema format.\n \n The \u0027clocks\u0027 property is now required. This simplifies the schema as\n effectively all the users require \u0027clocks\u0027 already and the upstream\n driver requires at least one clock.\n \n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Acked-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 553cedf60056dc55366b060104cf9b694efe9bbf\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Fri Aug 16 16:51:46 2019 -0500\n\n dt-bindings: Convert Arm Mali Midgard GPU to DT schema\n \n Convert the Arm Midgard GPU binding to DT schema format.\n \n The \u0027clocks\u0027 property is now required. This simplifies the schema as\n effectively all the users require \u0027clocks\u0027 already and the upstream\n driver requires at least one clock.\n \n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Acked-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit eef3c2ba0a42a6aa709828e968b64bd11f4aeb19\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Fri Aug 23 11:04:15 2019 +0200\n\n soc: amlogic: Add support for Everything-Else power domains controller\n \n Add support for the General Purpose Amlogic Everything-Else Power controller,\n with the first support for G12A and SM1 SoCs dedicated to the VPU, PCIe,\n USB, NNA, GE2D and Ethernet Power Domains.\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Tested-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit b84efad58ea42a5cfe49831140b141d1008a3595\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Fri Aug 23 11:04:14 2019 +0200\n\n dt-bindings: power: add Amlogic Everything-Else power domains bindings\n \n Add the bindings for the Amlogic Everything-Else power domains,\n controlling the Everything-Else peripherals power domains.\n \n The bindings targets the Amlogic G12A and SM1 compatible SoCs,\n support for earlier SoCs will be added later.\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit fecb410030628d70401e06a98a585d735f61d7e2\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:57:01 2019 +0200\n\n libperf: Add PERF_RECORD_ID_INDEX \u0027struct id_index_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_ID_INDEX event definition to libperf\u0027s event.h.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Add the PRI_ld64 define, so we can use it in printf output.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-8-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit ffd337b45b1aedc86b1de3cf8de9a79c10fd3810\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:57:00 2019 +0200\n\n libperf: Add PERF_RECORD_HEADER_BUILD_ID \u0027struct build_id_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_HEADER_BUILD_ID event definition to libperf\u0027s event.h.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027\n types used events to their generic \u0027__u*\u0027 versions.\n \n Adding the fix value for build_id variable, because it will never\n change.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-7-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 4fd7a4d220421bc6b63a6f693e45d4397d4da055\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:56:59 2019 +0200\n\n libperf: Add PERF_RECORD_HEADER_TRACING_DATA \u0027struct tracing_data_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_HEADER_TRACING_DATA event definition to libperf\u0027s event.h.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027\n types used events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-6-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 0f5b1a28c03d416f7a66d47a9b2f18942e334fc6\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:56:58 2019 +0200\n\n libperf: Add PERF_RECORD_HEADER_EVENT_TYPE \u0027struct event_type_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_HEADER_EVENT_TYPE event definition to libperf\u0027s event.h.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027\n types used events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-5-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 5ded068e923837068e39c0fd4ab40c0dacaa08e8\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:56:57 2019 +0200\n\n libperf: Add PERF_RECORD_EVENT_UPDATE \u0027struct event_update_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_EVENT_UPDATE event definition to libperf\u0027s event.h.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027\n types used events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-4-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 78e5ea1620964c4e34e9bf43e98a3def77e6bcde\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:56:56 2019 +0200\n\n libperf: Add PERF_RECORD_CPU_MAP \u0027struct cpu_map_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_CPU_MAP event definition to libperf\u0027s event.h.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027\n types used events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-3-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit c78ad994ad99a4e03ecefe446d39dced48ba254f\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Wed Aug 28 15:56:55 2019 +0200\n\n libperf: Add PERF_RECORD_HEADER_ATTR \u0027struct attr_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_HEADER_ATTR event definition to libperf\u0027s event.h.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190828135717.7245-2-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit be5863b7d9281bbb932542d16b7d758357fde267\nAuthor: Namhyung Kim \u003cnamhyung@kernel.org\u003e\nDate: Wed Aug 28 08:15:55 2019 +0900\n\n perf top: Fix event group with more than two events\n \n The event group feature links relevant hist entries among events so that\n they can be displayed together. During the link process, each hist\n entry in non-leader events is connected to a hist entry in the leader\n event. This is done in order of events specified in the command line so\n it assumes that events are linked in the order.\n \n But \u0027perf top\u0027 can break the assumption since it does the link process\n multiple times. For example, a hist entry can be in the third event\n only at first so it\u0027s linked after the leader. Some time later, second\n event has a hist entry for it and it\u0027ll be linked after the entry of the\n third event.\n \n This makes the code compilicated to deal with such unordered entries.\n This patch simply unlink all the entries after it\u0027s printed so that they\n can assume the correct order after the repeated link process. Also it\u0027d\n be easy to deal with decaying old entries IMHO.\n \n Signed-off-by: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Reported-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Tested-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Link: http://lkml.kernel.org/r/20190827231555.121411-2-namhyung@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit ea4385f804eadce3f4fd8698d4ffd9e85fb6d5e0\nAuthor: Namhyung Kim \u003cnamhyung@kernel.org\u003e\nDate: Wed Aug 28 08:15:54 2019 +0900\n\n perf top: Decay all events in the evlist\n \n Currently perf top only decays entries in a selected evsel. I don\u0027t\n know whether it\u0027s intended (maybe due to performance reason?) but anyway\n it might show incorrect output when event group is used since users will\n see leader event is decayed but others are not.\n \n This patch moves the decay code into perf_top__resort_hists() so that\n stdio and TUI code shared the logic.\n \n Signed-off-by: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Link: http://lkml.kernel.org/r/20190827231555.121411-1-namhyung@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 630aec1a7fd60ac355d5f2d67b5454912c5971a6\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 28 09:59:10 2019 -0300\n\n perf clang: Delete needless util-cxx.h header\n \n It was put in place just to make sure the \u0027new\u0027 C++ operator wouldn\u0027t\n clash with some argument name in util.h, but there is not anymore any\n such argument and also the reason stated for util.h to be included there\n was to get the __maybe_unused definition, that is in linux/compiler.h,\n so use that instead and nuke util-cxx.h.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: He Kuang \u003chekuang@huawei.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Wang Nan \u003cwangnan0@huawei.com\u003e\n Link: https://lkml.kernel.org/n/tip-1r5tvfnwiydjxhukgqs6bi11@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 6e7e5c7fbc1c84e4ac657e0cb5bf9ca24492cc8f\nAuthor: Thierry Reding \u003ctreding@nvidia.com\u003e\nDate: Wed Aug 28 10:34:11 2019 +0200\n\n hwmon: pwm-fan: Use platform_get_irq_optional()\n \n The PWM fan interrupt is optional, so we don\u0027t want an error message in\n the kernel log if it wasn\u0027t specified.\n \n Signed-off-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Reviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Acked-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Acked-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://lore.kernel.org/r/20190828083411.2496-2-thierry.reding@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 8973ea47901c81a1912bd05f1577bed9b5b52506\nAuthor: Thierry Reding \u003ctreding@nvidia.com\u003e\nDate: Wed Aug 28 10:34:10 2019 +0200\n\n driver core: platform: Introduce platform_get_irq_optional()\n \n In some cases the interrupt line of a device is optional. Introduce a\n new platform_get_irq_optional() that works much like platform_get_irq()\n but does not output an error on failure to find the interrupt.\n \n Signed-off-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Reviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190828083411.2496-1-thierry.reding@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b2fcb285ade36c2339ee6638223a65829ee9907f\nAuthor: Chunfeng Yun \u003cchunfeng.yun@mediatek.com\u003e\nDate: Fri Aug 23 14:40:25 2019 +0800\n\n usb: xhci-mtk: add an optional xhci_ck clock\n \n Some SoCs may have an optional clock xhci_ck (125M or 200M), it\n usually uses the same PLL as sys_ck, so support it.\n \n Signed-off-by: Chunfeng Yun \u003cchunfeng.yun@mediatek.com\u003e\n Acked-by: Mathias Nyman \u003cmathias.nyman@linux.intel.com\u003e\n Reviewed-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Link: https://lore.kernel.org/r/1566542425-20082-2-git-send-email-chunfeng.yun@mediatek.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit e74e8bbff6e3654db55a56d3000ef581a6166944\nAuthor: Chunfeng Yun \u003cchunfeng.yun@mediatek.com\u003e\nDate: Fri Aug 23 14:40:24 2019 +0800\n\n dt-bindings: usb: mtk-xhci: add an optional xhci_ck clock\n \n Add a new optional clock xhci_ck\n \n Signed-off-by: Chunfeng Yun \u003cchunfeng.yun@mediatek.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://lore.kernel.org/r/1566542425-20082-1-git-send-email-chunfeng.yun@mediatek.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 630a0aa75ed6b69b9d882565677aa181b579781a\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Fri Aug 23 17:44:32 2019 +0200\n\n arm64: dts: meson: g12a: add reset to tdm formatters\n \n Add the reset to the TDM formatters of the g12a. This helps\n with channel mapping when a playback/capture uses more than 1 lane.\n \n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 58e6a511f5ad980b18a30e6ebb35d9c1e29ac8f1\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Fri Aug 23 17:44:31 2019 +0200\n\n arm64: dts: meson: g12a: audio clock controller provides resets\n \n The clock controller dedicated to audio clocks also provides reset lines\n on the g12 SoC family\n \n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit d4609acce187cd1404f0335f4b3593c67004e86c\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Mon Aug 26 09:25:39 2019 +0200\n\n arm64: dts: meson-sm1-sei610: enable DVFS\n \n This enables DVFS for the Amlogic SM1 based SEI610 board by:\n - Adding the SM1 SoC OPPs taken from the vendor tree\n - Selecting the SM1 Clock controller instead of the G12A one\n - Adding the CPU rail regulator, PWM and OPPs for each CPU nodes.\n \n Each power supply can achieve 0.69V to 1.05V using a single PWM\n output clocked at 666KHz with an inverse duty-cycle.\n \n DVFS has been tested by running the arm64 cpuburn at [1] and cycling\n between all the possible cpufreq translations of the cpu cluster and\n checking the final frequency using the clock-measurer, script at [2].\n \n [1] https://github.com/ssvb/cpuburn-arm/blob/master/cpuburn-a53.S\n [2] https://gist.github.com/superna9999/d4de964dbc0f84b7d527e1df2ddea25f\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Tested-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 5b3fd41e48fe7912f41faf6ae2613478a2c58f43\nAuthor: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\nDate: Sun Aug 25 08:01:28 2019 +0400\n\n arm64: dts: meson-gxm-khadas-vim2: use rc-khadas keymap\n \n Swap to the rc-khadas keymap that maps the mouse button to KEY_MUTE.\n \n Signed-off-by: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 34c4e5fd2b2b3f1bf9707312c2ff127ddddd256f\nAuthor: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\nDate: Sun Aug 25 08:01:27 2019 +0400\n\n arm64: dts: meson-gxl-s905w-tx3-mini: add rc-tx3mini keymap\n \n add the rc-tx3mini keymap to the ir node\n \n Signed-off-by: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 33cdbdaa86d968ce656569b168745f7752649cbe\nAuthor: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\nDate: Sun Aug 25 08:01:26 2019 +0400\n\n arm64: dts: meson-gxl-s905x-khadas-vim: use rc-khadas keymap\n \n Swap to the rc-khadas keymap that maps the mouse button to KEY_MUTE.\n \n Signed-off-by: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit aa23acef958990891b81baa79e60814ea7433269\nAuthor: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\nDate: Sun Aug 25 08:01:25 2019 +0400\n\n arm64: dts: meson-gxbb-wetek-play2: add rc-wetek-play2 keymap\n \n add the rc-wetek-play2 keymap to the ir node\n \n Signed-off-by: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 7fc697b8f2f009a01bf3a9a8293bb17cd67e9714\nAuthor: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\nDate: Sun Aug 25 08:01:24 2019 +0400\n\n arm64: dts: meson-gxbb-wetek-hub: add rc-wetek-hub keymap\n \n add the rc-wetek-hub keymap to the ir node\n \n Signed-off-by: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 24810b978b0d35fc56d2dc45f6f0082d9a4963a9\nAuthor: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\nDate: Sun Aug 25 08:01:23 2019 +0400\n\n arm64: dts: meson-g12a-x96-max: add rc-x96max keymap\n \n add the rc-x96max keymap to the ir node\n \n Signed-off-by: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit d3fb99dc8141cba30d20657f1256dc934cafca23\nAuthor: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\nDate: Sun Aug 25 08:01:22 2019 +0400\n\n arm64: dts: meson-g12b-odroid-n2: add rc-odroid keymap\n \n add the rc-odroid keymap to the ir node\n \n Signed-off-by: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 6197c74c05bdc9705f31dd3d5742b2d3033f0006\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Fri Aug 23 11:04:18 2019 +0200\n\n arm64: dts: meson-sm1-sei610: add USB support\n \n Add the USB properties for the Amlogic SM1 Based SEI610 Board in order to\n support the USB DRD Type-C port and the USB3 Type A port.\n \n The USB DRD Type-C controller uses the ID signal to toggle the USB role\n between the DWC3 Host controller and the DWC2 Device controller.\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 616b360846fe0c419b2638e9d0a9185ddb150ec9\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Fri Aug 23 11:04:17 2019 +0200\n\n arm64: dts: meson-sm1-sei610: add HDMI display support\n \n Add the HDMI support nodes for the Amlogic SM1 Based SEI610 Board.\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Acked-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 45ccf77a05751134cf47ea6cc3a410bec875925f\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Fri Aug 23 11:04:16 2019 +0200\n\n arm64: meson-g12: add Everything-Else power domain controller\n \n Replace the VPU-centric power domain controller by the generic system-wide\n Everything-Else power domain controller and setup the right power-domains\n properties on the VPU, Ethernet \u0026 USB nodes.\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit c6244339c09dc85da80dbb2a75476758824eb860\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Fri Aug 23 09:02:48 2019 +0200\n\n arm64: dts: meson: fix boards regulators states format\n \n This fixes the following DT schemas check errors:\n meson-gxbb-odroidc2.dt.yaml: gpio-regulator-tf_io: states:0: Additional items are not allowed (1800000, 1 were unexpected)\n meson-gxbb-odroidc2.dt.yaml: gpio-regulator-tf_io: states:0: [3300000, 0, 1800000, 1] is too long\n meson-gxbb-nexbox-a95x.dt.yaml: gpio-regulator: states:0: Additional items are not allowed (3300000, 1 were unexpected)\n meson-gxbb-nexbox-a95x.dt.yaml: gpio-regulator: states:0: [1800000, 0, 3300000, 1] is too long\n meson-gxbb-p200.dt.yaml: gpio-regulator: states:0: Additional items are not allowed (3300000, 1 were unexpected)\n meson-gxbb-p200.dt.yaml: gpio-regulator: states:0: [1800000, 0, 3300000, 1] is too long\n meson-gxl-s905x-hwacom-amazetv.dt.yaml: gpio-regulator: states:0: Additional items are not allowed (3300000, 1 were unexpected)\n meson-gxl-s905x-hwacom-amazetv.dt.yaml: gpio-regulator: states:0: [1800000, 0, 3300000, 1] is too long\n meson-gxbb-p201.dt.yaml: gpio-regulator: states:0: Additional items are not allowed (3300000, 1 were unexpected)\n meson-gxbb-p201.dt.yaml: gpio-regulator: states:0: [1800000, 0, 3300000, 1] is too long\n meson-g12b-odroid-n2.dt.yaml: gpio-regulator-tf_io: states:0: Additional items are not allowed (1800000, 1 were unexpected)\n meson-g12b-odroid-n2.dt.yaml: gpio-regulator-tf_io: states:0: [3300000, 0, 1800000, 1] is too long\n meson-gxl-s905x-nexbox-a95x.dt.yaml: gpio-regulator: states:0: Additional items are not allowed (3300000, 1 were unexpected)\n meson-gxl-s905x-nexbox-a95x.dt.yaml: gpio-regulator: states:0: [1800000, 0, 3300000, 1] is too long\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 2258af108994eeb764162d79709f98394edde05c\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Fri Aug 23 09:02:47 2019 +0200\n\n arm64: dts: meson-gxbb-p201: fix snps, reset-delays-us format\n \n This fixes the following DT schemas check errors:\n meson-gxbb-p201.dt.yaml: ethernet@c9410000: snps,reset-delays-us: [[0, 10000, 1000000]] is too short\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 35a8a061ad25a9c308c83577363b0aeeacbcb1b3\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Fri Aug 23 09:02:46 2019 +0200\n\n arm64: dts: meson-gxbb-nanopi-k2: add missing model\n \n This fixes the following DT schemas check errors:\n meson-gxbb-nanopi-k2.dt.yaml: /: \u0027model\u0027 is a required property\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit c61bf6cb5fd6de1bce7526d8e3d778dbcf30ce75\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Fri Aug 23 09:02:45 2019 +0200\n\n arm64: dts: meson-g12a-x96-max: fix compatible\n \n This fixes the following DT schemas check errors:\n meson-g12a-x96-max.dt.yaml: /: compatible: [\u0027amediatech,x96-max\u0027, \u0027amlogic,u200\u0027, \u0027amlogic,g12a\u0027] is not valid under any of the given schemas\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 382a77ca51d9f4d01d8891ffa7c7a5730cd2b322\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Fri Aug 23 09:02:44 2019 +0200\n\n arm64: dts: meson-g12a: fix reset controller compatible\n \n This fixes the following DT schemas check errors:\n meson-g12a-u200.dt.yaml: reset-controller@1004: compatible:0: \u0027amlogic,meson-g12a-reset\u0027 is not one of [\u0027amlogic,meson8b-reset\u0027, \u0027amlogic,meson-gxbb-reset\u0027, \u0027amlogic,meson-axg-reset\u0027]\n meson-g12a-sei510.dt.yaml: reset-controller@1004: compatible:0: \u0027amlogic,meson-g12a-reset\u0027 is not one of [\u0027amlogic,meson8b-reset\u0027, \u0027amlogic,meson-gxbb-reset\u0027, \u0027amlogic,meson-axg-reset\u0027]\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit c75b9f52939ddcdc9ab1e14c72f8fd9f1e5ac3f6\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Fri Aug 23 09:02:43 2019 +0200\n\n arm64: dts: meson-axg: fix MHU compatible\n \n This fixes the following DT schemas check errors:\n meson-axg-s400.dt.yaml: mailbox@ff63c404: compatible:0: \u0027amlogic,meson-gx-mhu\u0027 is not one of [\u0027amlogic,meson-gxbb-mhu\u0027]\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit f43860f4a5d068c1205098b582ff667903ebcf0e\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Fri Aug 23 09:02:42 2019 +0200\n\n arm64: dts: meson-gxl: fix internal phy compatible\n \n This fixes the following DT schemas check errors:\n meson-gxl-s805x-libretech-ac.dt.yaml: ethernet-phy@8: compatible: [\u0027ethernet-phy-id0181.4400\u0027, \u0027ethernet-phy-ieee802.3-c22\u0027] is not valid under any of the given schemas\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Acked-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 8671370cd71294e0d8436ab431855ff5e8da9fc6\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Fri Aug 23 09:02:41 2019 +0200\n\n arm64: dts: meson-gx: fix periphs bus node name\n \n This fixes the following DT schemas check errors:\n meson-gxbb-nanopi-k2.dt.yaml: periphs@c8834000: $nodename:0: \u0027periphs@c8834000\u0027 does not match \u0027^(bus|soc|axi|ahb|apb)(@[0-9a-f]+)?$\u0027\n meson-gxl-s805x-libretech-ac.dt.yaml: periphs@c8834000: $nodename:0: \u0027periphs@c8834000\u0027 does not match \u0027^(bus|soc|axi|ahb|apb)(@[0-9a-f]+)?$\u0027\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit c424d73194933d714a027e43ba33c58e8398a68e\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Fri Aug 23 09:02:40 2019 +0200\n\n arm64: dts: meson-gx: fix mhu compatible\n \n This fixes the following DT schemas check errors:\n meson-gxbb-nanopi-k2.dt.yaml: mailbox@404: compatible:0: \u0027amlogic,meson-gx-mhu\u0027 is not one of [\u0027amlogic,meson-gxbb-mhu\u0027]\n meson-gxl-s805x-libretech-ac.dt.yaml: mailbox@404: compatible:0: \u0027amlogic,meson-gx-mhu\u0027 is not one of [\u0027amlogic,meson-gxbb-mhu\u0027]\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 3d6f4dbee58d639809e2fffa70a6650868a34daa\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Fri Aug 23 09:02:39 2019 +0200\n\n arm64: dts: meson-gx: fix watchdog compatible\n \n This fixes the following DT schemas check errors:\n meson-gxbb-nanopi-k2.dt.yaml: watchdog@98d0: compatible:0: \u0027amlogic,meson-gx-wdt\u0027 is not one of [\u0027amlogic,meson-gxbb-wdt\u0027]\n meson-gxl-s805x-libretech-ac.dt.yaml: watchdog@98d0: compatible:0: \u0027amlogic,meson-gx-wdt\u0027 is not one of [\u0027amlogic,meson-gxbb-wdt\u0027]\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 9afc3c5d3bf338e072ba0ce3a7697f490b9440b2\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Fri Aug 23 09:02:38 2019 +0200\n\n arm64: dts: meson-gx: fix spifc compatible\n \n This fixes the following DT schemas check errors:\n meson-gxl-s805x-libretech-ac.dt.yaml: spi@8c80: compatible:0: \u0027amlogic,meson-gx-spifc\u0027 is not one of [\u0027amlogic,meson6-spifc\u0027, \u0027amlogic,meson-gxbb-spifc\u0027]\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 2e655d4a8027690fc2e29d5ccd4fde69d89efe25\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Fri Aug 23 09:02:37 2019 +0200\n\n arm64: dts: meson-gx: fix reset controller compatible\n \n This fixes the following DT schemas check errors:\n meson-gxbb-nanopi-k2.dt.yaml: reset-controller@4404: compatible:0: \u0027amlogic,meson-gx-reset\u0027 is not one of [\u0027amlogic,meson8b-reset\u0027, \u0027amlogic,meson-gxbb-reset\u0027, \u0027amlogic,meson-axg-reset\u0027]\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit e117ba5dbc769ea21879f01c3ad97abfe12c2d96\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Fri Aug 23 09:02:36 2019 +0200\n\n arm64: dts: meson-gx: drop the vpu dmc memory cell\n \n This fixes the following DT schemas check errors:\n meson-gxl-s805x-libretech-ac.dt.yaml: vpu@d0100000: reg-names: Additional items are not allowed (\u0027dmc\u0027 was unexpected)\n meson-gxl-s805x-libretech-ac.dt.yaml: vpu@d0100000: reg-names: [\u0027vpu\u0027, \u0027hhi\u0027, \u0027dmc\u0027] is too long\n \n The \u0027dmc\u0027 register area was replaced by the amlogic,canvas property\n which was introduced in commit f1726043426c73 (\"arm64: dts: meson-gx:\n add dmcbus and canvas nodes.\") and commit cf34287986d0b6 (\"arm64: dts:\n meson-gx: Add canvas provider node to the vpu\")\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 1e6714c713712cc0410e83f4a1e89d07570b9002\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Fri Aug 23 09:02:35 2019 +0200\n\n arm64: dts: meson: fix ethernet mac reg format\n \n This fixes the following DT schemas check errors:\n meson-axg-s400.dt.yaml: soc: ethernet@ff3f0000:reg:0: [0, 4282318848, 0, 65536, 0, 4284695872, 0, 8] is too long\n meson-axg-s400.dt.yaml: ethernet@ff3f0000: reg: [[0, 4282318848, 0, 65536, 0, 4284695872, 0, 8]] is too short\n meson-g12a-u200.dt.yaml: soc: ethernet@ff3f0000:reg:0: [0, 4282318848, 0, 65536, 0, 4284695872, 0, 8] is too long\n meson-g12a-u200.dt.yaml: ethernet@ff3f0000: reg: [[0, 4282318848, 0, 65536, 0, 4284695872, 0, 8]] is too short\n meson-gxbb-nanopi-k2.dt.yaml: soc: ethernet@c9410000:reg:0: [0, 3376480256, 0, 65536, 0, 3364046144, 0, 4] is too long\n meson-gxl-s805x-libretech-ac.dt.yaml: soc: ethernet@c9410000:reg:0: [0, 3376480256, 0, 65536, 0, 3364046144, 0, 4] is too lon\n \n while here, also drop the redundant reg property from meson-gxl.dtsi\n because it had the same value as meson-gx.dtsi from which it inherits.\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit c1442b4afc4461d24fedaec02db2898641f94a79\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Wed Mar 27 14:15:16 2019 +0000\n\n vfs: Create fs_context-aware mount_bdev() replacement\n \n Create a function, vfs_get_block_super(), that is fs_context-aware and a\n replacement for mount_bdev(). It caches the block device pointer and file\n open mode in the fs_context struct so that this information can be passed\n into sget_fc()\u0027s test and set functions.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n cc: Jens Axboe \u003caxboe@kernel.dk\u003e\n cc: linux-block@vger.kernel.org\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit b5efb8f672c97b638952998badb2717f1320a37a\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Wed Mar 20 23:31:05 2019 +0000\n\n vfs: Convert functionfs to use the new mount API\n \n Convert the functionfs filesystem to the new internal mount API as the old\n one will be obsoleted and removed. This allows greater flexibility in\n communication of mount parameters between userspace, the VFS and the\n filesystem.\n \n See Documentation/filesystems/mount_api.txt for more information.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Acked-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n Acked-by: Michal Nazarewicz \u003cmina86@mina86.com\u003e\n cc: linux-usb@vger.kernel.org\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 85ed84b1bce945ad9c8d674f7f02526bf0910a33\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Fri Mar 22 16:36:22 2019 +0000\n\n vfs: Convert ubifs to use the new mount API\n \n Convert the ubifs filesystem to the new internal mount API as the old\n one will be obsoleted and removed. This allows greater flexibility in\n communication of mount parameters between userspace, the VFS and the\n filesystem.\n \n See Documentation/filesystems/mount_api.txt for more information.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n cc: Richard Weinberger \u003crichard@nod.at\u003e\n cc: Artem Bityutskiy \u003cdedekind1@gmail.com\u003e\n cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n cc: linux-mtd@lists.infradead.org\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit cbc5ff0b36a00dcae9f03a9b3a735dbed7da5695\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Mar 25 15:41:33 2019 +0000\n\n vfs: Convert orangefs to use the new mount API\n \n Convert the orangefs filesystem to the new internal mount API as the old\n one will be obsoleted and removed. This allows greater flexibility in\n communication of mount parameters between userspace, the VFS and the\n filesystem.\n \n See Documentation/filesystems/mount_api.txt for more information.\n cc: Mike Marshall \u003chubcap@omnibond.com\u003e\n cc: Martin Brandenburg \u003cmartin@omnibond.com\u003e\n cc: devel@lists.orangefs.org\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit b51c9a932c460b13070917f3f74d58568ab4a543\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Fri Mar 22 14:58:36 2019 +0000\n\n vfs: Convert bpf to use the new mount API\n \n Convert the bpf filesystem to the new internal mount API as the old\n one will be obsoleted and removed. This allows greater flexibility in\n communication of mount parameters between userspace, the VFS and the\n filesystem.\n \n See Documentation/filesystems/mount_api.txt for more information.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n cc: Alexei Starovoitov \u003cast@kernel.org\u003e\n cc: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n cc: Martin KaFai Lau \u003ckafai@fb.com\u003e\n cc: Song Liu \u003csongliubraving@fb.com\u003e\n cc: Yonghong Song \u003cyhs@fb.com\u003e\n cc: netdev@vger.kernel.org\n cc: bpf@vger.kernel.org\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 74ed8bca963169f7e1e1eb7165e1c84f7dcafb4e\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Fri Mar 22 12:56:55 2019 +0000\n\n vfs: Convert devpts to use the new mount API\n \n Convert the devpts filesystem to the new internal mount API as the old\n one will be obsoleted and removed. This allows greater flexibility in\n communication of mount parameters between userspace, the VFS and the\n filesystem.\n \n See Documentation/filesystems/mount_api.txt for more information.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Acked-by: Christian Brauner \u003cchristian@brauner.io\u003e\n cc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 38de9fe6b1d1619ddd14adb3b104ce08946d441c\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Thu Mar 21 13:48:06 2019 +0000\n\n vfs: Convert autofs to use the new mount API\n \n Convert the autofs filesystem to the new internal mount API as the old\n one will be obsoleted and removed. This allows greater flexibility in\n communication of mount parameters between userspace, the VFS and the\n filesystem.\n \n See Documentation/filesystems/mount_api.txt for more information.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit d5eda3a44c62f89094336b5e89f4448d8b258af7\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Thu Mar 21 12:52:45 2019 +0000\n\n vfs: Convert coda to use the new mount API\n \n Convert the coda filesystem to the new internal mount API as the old\n one will be obsoleted and removed. This allows greater flexibility in\n communication of mount parameters between userspace, the VFS and the\n filesystem.\n \n See Documentation/filesystems/mount_api.txt for more information.\n \n Note this is slightly tricky as coda currently only has a binary mount data\n interface. This is handled through the parse_monolithic hook.\n \n Also add a more conventional interface with a parameter named \"fd\" that\n takes an fd that refers to a coda psdev, thereby specifying the index to\n use.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n cc: Jan Harkes \u003cjaharkes@cs.cmu.edu\u003e\n cc: coda@cs.cmu.edu\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit aecfd324a1f85ad94ac156d78cb317d26ec6c263\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Mar 25 16:38:31 2019 +0000\n\n vfs: Make fs_parse() handle fs_param_is_fd-type params better\n \n Make fs_parse() handle fs_param_is_fd-type parameters that are passed a\n string by converting it to an integer (in addition to handling direct fd\n specification).\n \n Also range check the integer.\n \n [fix from Yin Fengwei folded]\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 5088ec1c12d42c0da49d5bed00f5efa65473dae9\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Mar 25 16:38:26 2019 +0000\n\n vfs: Convert binderfs to use the new mount API\n \n Convert the binderfs filesystem to the new internal mount API as the old\n one will be obsoleted and removed. This allows greater flexibility in\n communication of mount parameters between userspace, the VFS and the\n filesystem.\n \n See Documentation/filesystems/mount_api.txt for more information.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Reviewed-by: Christian Brauner \u003cchristian@brauner.io\u003e\n cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n cc: \"Arve Hjønnevåg\" \u003carve@android.com\u003e\n cc: Todd Kjos \u003ctkjos@android.com\u003e\n cc: Martijn Coenen \u003cmaco@android.com\u003e\n cc: Joel Fernandes \u003cjoel@joelfernandes.org\u003e\n cc: devel@driverdev.osuosl.org\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 8bb3c61bafa8c1cd222ada602bb94ff23119e738\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Mar 25 16:38:31 2019 +0000\n\n vfs: Convert ramfs, shmem, tmpfs, devtmpfs, rootfs to use the new mount API\n \n Convert the ramfs, shmem, tmpfs, devtmpfs and rootfs filesystems to the new\n internal mount API as the old one will be obsoleted and removed. This\n allows greater flexibility in communication of mount parameters between\n userspace, the VFS and the filesystem.\n \n See Documentation/filesystems/mount_api.txt for more information.\n \n Note that tmpfs is slightly tricky as it can contain embedded commas, so it\n can\u0027t be trivially split up using strsep() to break on commas in\n generic_parse_monolithic(). Instead, tmpfs has to supply its own generic\n parser.\n \n However, if tmpfs changes, then devtmpfs and rootfs, which are wrappers\n around tmpfs or ramfs, must change too - and thus so must ramfs, so these\n had to be converted also.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n cc: Hugh Dickins \u003chughd@google.com\u003e\n cc: linux-mm@kvack.org\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 63228b974a6e1e398e1436edf34fc65d99b85dc9\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Sat Jun 1 18:56:53 2019 -0400\n\n make shmem_fill_super() static\n \n ... have callers use shmem_mount()\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit d154e3b55d9e6c58b6e7e01e47ac39344166d773\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Sat Jun 1 18:51:15 2019 -0400\n\n make ramfs_fill_super() static\n \n all users should just call ramfs_mount()\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit e941159e9d79e1412348dc62e6b8cf22bdb0fccb\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Sat Jun 1 18:43:09 2019 -0400\n\n devtmpfs: don\u0027t mix {ramfs,shmem}_fill_super() with mount_single()\n \n Create an internal-only type matching the current devtmpfs, never\n register it and have one kernel-internal mount done. That thing\n gets mounted only once, so it is free to use mount_nodev().\n \n The \"public\" devtmpfs (the one we do register, and only after\n the internal mount of the real thing is done) simply gets and\n returns an extra reference to the internal superblock.\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 4a2dbeddd3d54484d2e9c12965e4f9bfa1a0ee36\nAuthor: Krzysztof Wilczynski \u003ckw@linux.com\u003e\nDate: Tue Aug 27 11:49:51 2019 +0200\n\n PCI/ACPI: Remove unnecessary struct hotplug_program_ops\n \n Move the ACPI-specific structs hpx_type0, hpx_type1, hpx_type2 and\n hpx_type3 to drivers/pci/pci-acpi.c as they are not used anywhere else.\n Then remove the struct hotplug_program_ops that has been shared between\n drivers/pci/probe.c and drivers/pci/pci-acpi.c from drivers/pci/pci.h as it\n is no longer needed.\n \n The struct hotplug_program_ops was added by 87fcf12e846a (\"PCI/ACPI: Remove\n the need for \u0027struct hotplug_params\u0027\") and replaced previously used struct\n hotplug_params enabling the support for the _HPX Type 3 Setting Record that\n was added by f873c51a155a (\"PCI/ACPI: Implement _HPX Type 3 Setting\n Record\").\n \n The new struct allowed for the static functions such program_hpx_type0(),\n program_hpx_type1(), etc., from the drivers/pci/probe.c to be called from\n the function pci_acpi_program_hp_params() in the drivers/pci/pci-acpi.c.\n \n Previously a programming of _HPX Type 0 was as follows:\n \n drivers/pci/probe.c:\n \n program_hpx_type0()\n ...\n pci_configure_device()\n hp_ops \u003d {\n .program_type0 \u003d program_hpx_type0,\n ...\n }\n pci_acpi_program_hp_params(\u0026hp_ops)\n \n drivers/pci/pci-acpi.c:\n \n pci_acpi_program_hp_params(\u0026hp_ops)\n acpi_run_hpx(hp_ops)\n decode_type0_hpx_record()\n hp_ops-\u003eprogram_type0 # program_hpx_type0() called via hp_ops\n \n After the ACPI-specific functions, structs, enums, etc., have been moved to\n drivers/pci/pci-acpi.c there is no need for the hotplug_program_ops as all\n of the _HPX Type 0, 1, 2 and 3 are directly accessible.\n \n Link: https://lore.kernel.org/r/20190827094951.10613-4-kw@linux.com\n Signed-off-by: Krzysztof Wilczynski \u003ckw@linux.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit 8c3aac6e1b6146ce771b1cabd78e593136d3e5f2\nAuthor: Krzysztof Wilczynski \u003ckw@linux.com\u003e\nDate: Tue Aug 27 11:49:50 2019 +0200\n\n PCI/ACPI: Move _HPP \u0026 _HPX functions to pci-acpi.c\n \n Move program_hpx_type0(), program_hpx_type1(), etc., and enums\n hpx_type3_dev_type, hpx_type3_fn_type and hpx_type3_cfg_loc to\n drivers/pci/pci-acpi.c as these functions and enums are ACPI-specific.\n \n Move structs hpx_type0, hpx_type1, hpx_type2 and hpx_type3 to\n drivers/pci/pci.h as these are shared between drivers/pci/pci-acpi.c and\n drivers/pci/probe.c.\n \n Link: https://lore.kernel.org/r/20190827094951.10613-3-kw@linux.com\n Signed-off-by: Krzysztof Wilczynski \u003ckw@linux.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit 3982f1df00df57768f0782bc702fc4fbea347ca2\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Tue Aug 27 21:13:46 2019 +0800\n\n staging: most: sound: Fix error path of audio_init\n \n If most_register_configfs_subsys() fails, we should\n call most_deregister_component() do cleanup.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Fixes: 919c03ae11b9 (\"staging: most: enable configfs support\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190827131346.12704-1-yuehaibing@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a1617c8c3d39b1348e72d04548229e2ed9bf8121\nAuthor: Stephen Brennan \u003cstephen@brennan.io\u003e\nDate: Tue Aug 27 21:35:42 2019 -0700\n\n staging: rtl8192u: Fix indentation\n \n Checkpatch reports WARNING:SUSPECT_CODE_INDENT in several places. Fix\n this by aligning code properly with tabs.\n \n Signed-off-by: Stephen Brennan \u003cstephen@brennan.io\u003e\n Link: https://lore.kernel.org/r/20190828043542.3753-1-stephen@brennan.io\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 2da39f1cc36bff4cc53dc32a4afb3def488cc766\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Aug 27 11:51:18 2019 -0300\n\n perf evlist: Remove needless util.h from evlist.h\n \n There is no need for that util/util.h include there and, remove it,\n pruning the include tree, fix the fallout by adding necessary headers to\n places that were getting needed includes indirectly from evlist.h -\u003e\n util.h.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-s9f7uve8wvykr5itcm7m7d8q@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit efa73d37c11af5082a5605665186c368f1196381\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Aug 27 11:32:09 2019 -0300\n\n perf tools: Remove needless util.h include from builtin.h\n \n And fix up places where util.h is needed but was obtained indirectly via\n builtin.h.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-a01ig3c4t76ye5wkqmtgk9qn@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit d06e5fad8c4692c6e5f1bd626056f23716bfe4a6\nAuthor: Igor Lubashev \u003cilubashe@akamai.com\u003e\nDate: Mon Aug 26 21:39:16 2019 -0400\n\n perf tools: Warn that perf_event_paranoid can restrict kernel symbols\n \n Warn that /proc/sys/kernel/perf_event_paranoid can also restrict kernel\n symbols.\n \n Signed-off-by: Igor Lubashev \u003cilubashe@akamai.com\u003e\n Tested-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: James Morris \u003cjmorris@namei.org\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Suzuki Poulouse \u003csuzuki.poulose@arm.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Link: http://lkml.kernel.org/r/1566869956-7154-6-git-send-email-ilubashe@akamai.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 8859aedefefe7eeea5e67968b7fe39c828d589a0\nAuthor: Igor Lubashev \u003cilubashe@akamai.com\u003e\nDate: Mon Aug 26 21:39:15 2019 -0400\n\n perf symbols: Use CAP_SYSLOG with kptr_restrict checks\n \n The kernel is using CAP_SYSLOG capability instead of uid\u003d\u003d0 and euid\u003d\u003d0\n when checking kptr_restrict. Make perf do the same.\n \n Also, the kernel is a more restrictive than \"no restrictions\" in case of\n kptr_restrict\u003d\u003d0, so add the same logic to perf.\n \n Signed-off-by: Igor Lubashev \u003cilubashe@akamai.com\u003e\n Tested-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: James Morris \u003cjmorris@namei.org\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Suzuki Poulouse \u003csuzuki.poulose@arm.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Link: http://lkml.kernel.org/r/1566869956-7154-5-git-send-email-ilubashe@akamai.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit aa97293ff129f504e7c8589e56007ecfe3e3e835\nAuthor: Igor Lubashev \u003cilubashe@akamai.com\u003e\nDate: Mon Aug 26 21:39:14 2019 -0400\n\n perf evsel: Kernel profiling is disallowed only when perf_event_paranoid \u003e 1\n \n Perf was too restrictive about sysctl kernel.perf_event_paranoid. The\n kernel only disallows profiling when perf_event_paranoid \u003e 1. Make perf\n do the same.\n \n Committer testing:\n \n For a non-root user:\n \n $ id\n uid\u003d1000(acme) gid\u003d1000(acme) groups\u003d1000(acme),10(wheel) context\u003dunconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023\n $\n \n Before:\n \n We were restricting it to just userspace (:u suffix) even for a\n workload started by the user:\n \n $ perf record sleep 1\n [ perf record: Woken up 1 times to write data ]\n [ perf record: Captured and wrote 0.001 MB perf.data (8 samples) ]\n $ perf evlist\n cycles:u\n $ perf evlist -v\n cycles:u: size: 112, { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|PERIOD, read_format: ID, disabled: 1, inherit: 1, exclude_kernel: 1, mmap: 1, comm: 1, freq: 1, enable_on_exec: 1, task: 1, precise_ip: 3, sample_id_all: 1, exclude_guest: 1, mmap2: 1, comm_exec: 1, ksymbol: 1, bpf_event: 1\n $ perf report --stdio\n # To display the perf.data header info, please use --header/--header-only options.\n #\n # Total Lost Samples: 0\n #\n # Samples: 8 of event \u0027cycles:u\u0027\n # Event count (approx.): 1040396\n #\n # Overhead Command Shared Object Symbol\n # ........ ....... ................ ......................\n #\n 68.36% sleep libc-2.29.so [.] _dl_addr\n 27.33% sleep ld-2.29.so [.] dl_main\n 3.80% sleep ld-2.29.so [.] _dl_setup_hash\n #\n # (Tip: Order by the overhead of source file name and line number: perf report -s srcline)\n #\n $\n $\n \n After:\n \n When the kernel allows profiling the kernel in that scenario:\n \n $ perf record sleep 1\n [ perf record: Woken up 1 times to write data ]\n [ perf record: Captured and wrote 0.023 MB perf.data (11 samples) ]\n $ perf evlist\n cycles\n $ perf evlist -v\n cycles: size: 112, { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|PERIOD, read_format: ID, disabled: 1, inherit: 1, mmap: 1, comm: 1, freq: 1, enable_on_exec: 1, task: 1, precise_ip: 3, sample_id_all: 1, exclude_guest: 1, mmap2: 1, comm_exec: 1, ksymbol: 1, bpf_event: 1\n $\n $ perf report --stdio\n # To display the perf.data header info, please use --header/--header-only options.\n #\n # Total Lost Samples: 0\n #\n # Samples: 11 of event \u0027cycles\u0027\n # Event count (approx.): 1601964\n #\n # Overhead Command Shared Object Symbol\n # ........ ....... ................ ..........................\n #\n 28.14% sleep [kernel.vmlinux] [k] __rb_erase_color\n 27.21% sleep [kernel.vmlinux] [k] unmap_page_range\n 27.20% sleep ld-2.29.so [.] __tunable_get_val\n 15.24% sleep [kernel.vmlinux] [k] thp_get_unmapped_area\n 1.96% perf [kernel.vmlinux] [k] perf_event_exec\n 0.22% perf [kernel.vmlinux] [k] native_sched_clock\n 0.02% perf [kernel.vmlinux] [k] intel_bts_enable_local\n 0.00% perf [kernel.vmlinux] [k] native_write_msr\n #\n # (Tip: Boolean options have negative forms, e.g.: perf report --no-children)\n #\n $\n \n Reported-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Signed-off-by: Igor Lubashev \u003cilubashe@akamai.com\u003e\n Tested-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Tested-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: James Morris \u003cjmorris@namei.org\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Suzuki Poulouse \u003csuzuki.poulose@arm.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Link: http://lkml.kernel.org/r/1566869956-7154-4-git-send-email-ilubashe@akamai.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit dda1bf8ea78add78739d128a20b555c4a1a19c27\nAuthor: Igor Lubashev \u003cilubashe@akamai.com\u003e\nDate: Mon Aug 26 21:39:13 2019 -0400\n\n perf tools: Use CAP_SYS_ADMIN with perf_event_paranoid checks\n \n The kernel is using CAP_SYS_ADMIN instead of euid\u003d\u003d0 to override\n perf_event_paranoid check. Make perf do the same.\n \n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Acked-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Tested-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Reviewed-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e # coresight part\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: James Morris \u003cjmorris@namei.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Suzuki Poulouse \u003csuzuki.poulose@arm.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Link: http://lkml.kernel.org/r/1566869956-7154-3-git-send-email-ilubashe@akamai.com\n Signed-off-by: Igor Lubashev \u003cilubashe@akamai.com\u003e\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit e9a6882f267a8105461066e3ea6b4b6b9be1b807\nAuthor: Igor Lubashev \u003cilubashe@akamai.com\u003e\nDate: Mon Aug 26 21:39:12 2019 -0400\n\n perf event: Check ref_reloc_sym before using it\n \n Check for ref_reloc_sym before using it instead of checking\n symbol_conf.kptr_restrict and relying solely on that check.\n \n Reported-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Signed-off-by: Igor Lubashev \u003cilubashe@akamai.com\u003e\n Tested-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: James Morris \u003cjmorris@namei.org\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Suzuki Poulouse \u003csuzuki.poulose@arm.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Link: http://lkml.kernel.org/r/1566869956-7154-2-git-send-email-ilubashe@akamai.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit e2797ad31fb40f4ff59ebc4314d6f000d713bad9\nAuthor: Krzysztof Wilczynski \u003ckw@linux.com\u003e\nDate: Tue Aug 27 11:49:49 2019 +0200\n\n PCI/ACPI: Rename _HPX structs from hpp_* to hpx_*\n \n The names of the hpp_type0, hpp_type1 and hpp_type2 structs suggest that\n they\u0027re related to _HPP, when in fact they\u0027re related to _HPX.\n \n The struct hpp_type0 denotes an _HPX Type 0 setting record that supersedes\n the _HPP setting record, and it has been used interchangeably for _HPP as\n per the ACPI specification (see version 6.3, section 6.2.9.1) which states\n that it should be applied to PCI, PCI-X and PCI Express devices, with\n settings being ignored if they are not applicable.\n \n Rename them to hpx_type0, hpx_type1 and hpx_type2 to reflect their relation\n to _HPX rather than _HPP.\n \n Link: https://lore.kernel.org/r/20190827094951.10613-2-kw@linux.com\n Signed-off-by: Krzysztof Wilczynski \u003ckw@linux.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit 9770f22077981cac2bd2f4dd6caa6269892cd3f7\nAuthor: Madhumitha Tolakanahalli Pradeep \u003cmadhumitha.tolakanahalli.pradeep@intel.com\u003e\nDate: Thu Aug 22 17:46:55 2019 -0700\n\n drm/i915/tgl: Enabling DSC on Pipe A for TGL\n \n DSC was not supported on Pipe A for previous platforms. Tigerlake onwards,\n all the pipes support DSC. Hence, the DSC and FEC restriction on\n Pipe A needs to be removed.\n \n v2: Changes in the logic around removing the restriction around\n Pipe A (Manasi, Lucas)\n \n Cc: Manasi Navare \u003cmanasi.d.navare@intel.com\u003e\n Signed-off-by: Madhumitha Tolakanahalli Pradeep \u003cmadhumitha.tolakanahalli.pradeep@intel.com\u003e\n Reviewed-by: Manasi Navare \u003cmanasi.d.navare@intel.com\u003e\n Reviewed-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Signed-off-by: Manasi Navare \u003cmanasi.d.navare@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823004655.28905-1-madhumitha.tolakanahalli.pradeep@intel.com\n\ncommit 556d971bdae643de4cd7e2976e14f70ca2a3061d\nAuthor: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\nDate: Wed Aug 7 21:43:18 2019 -0300\n\n ima: Fix use after free in ima_read_modsig()\n \n If we can\u0027t parse the PKCS7 in the appended modsig, we will free the modsig\n structure and then access one of its members to determine the error value.\n \n Fixes: 39b07096364a (\"ima: Implement support for module-style appended signatures\")\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Reported-by: Julia Lawall \u003cjulia.lawall@lip6.fr\u003e\n Reported-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Reviewed-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n\ncommit 25948b87dda284664edeb3b3dab689df0a7dc889\nAuthor: Marina Varshaver \u003cmarinav@mellanox.com\u003e\nDate: Tue Aug 20 04:59:11 2019 +0300\n\n net/mlx5e: Support TSO and TX checksum offloads for IP-in-IP\n tunnels\n \n Add TX offloads support for IP-in-IP tunneled packets by reporting\n the needed netdev features.\n \n Signed-off-by: Marina Varshaver \u003cmarinav@mellanox.com\u003e\n Signed-off-by: Avihu Hagag \u003cavihuh@mellanox.com\u003e\n Reviewed-by: Aya Levin \u003cayal@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit e3a53bc536fc279de2ace13b8d6d54b071afb722\nAuthor: Marina Varshaver \u003cmarinav@mellanox.com\u003e\nDate: Tue Aug 20 03:36:29 2019 +0300\n\n net/mlx5e: Improve stateless offload capability check\n \n Use generic function for checking tunnel stateless offload capability\n instead of separate macros.\n \n Signed-off-by: Marina Varshaver \u003cmarinav@mellanox.com\u003e\n Reviewed-by: Aya Levin \u003cayal@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit a795d8db2a6d3c6f80e7002dd6357e6736dad1b6\nAuthor: Aya Levin \u003cayal@mellanox.com\u003e\nDate: Mon Apr 29 17:45:52 2019 +0300\n\n net/mlx5e: Support RSS for IP-in-IP and IPv6 tunneled packets\n \n Add support for inner header RSS on IP-in-IP and IPv6 tunneled packets.\n \n Add rules to the steering table regarding outer IP header, with\n IPv4/6-\u003eIP-in-IP. Tunneled packets with protocol numbers: 0x4 (IP-in-IP)\n and 0x29 (IPv6) are RSS-ed on the inner IP header.\n Separate FW dependencies between flow table inner IP capabilities and\n GRE offload support. Allowing this feature even if GRE offload is not\n supported. Tested with multi stream TCP traffic tunneled with IPnIP.\n Verified that:\n Without this patch, only a single RX ring was processing the traffic.\n With this patch, multiple RX rings were processing the traffic.\n Verified with and without GRE offload support.\n \n Signed-off-by: Aya Levin \u003cayal@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit a49e1f31ae155d64355d0cd0e0afa5b2bc8544cd\nAuthor: Aya Levin \u003cayal@mellanox.com\u003e\nDate: Thu Aug 8 16:16:28 2019 +0300\n\n net/mlx5e: Change function\u0027s position to a more fitting file\n \n Move function which indicates whether tunnel inner flow table is\n supported from en.h to en_fs.c. It fits better right after tunnel\n protocol rules definitions.\n \n Signed-off-by: Aya Levin \u003cayal@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 0cfafd4b4ddff99d6c21e028704a13ec1a578b8b\nAuthor: Moshe Shemesh \u003cmoshe@mellanox.com\u003e\nDate: Sun Aug 4 17:53:09 2019 +0300\n\n net/mlx5e: Add device out of buffer counter\n \n Added the following packets drop counter:\n Device out of buffer - counts packets which were dropped due to full\n device internal receive queue.\n This counter will be shown on ethtool as a new counter called\n dev_out_of_buffer.\n The counter is read from FW by command QUERY_VNIC_ENV.\n \n Signed-off-by: Moshe Shemesh \u003cmoshe@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 45f171b1182b9c4ab6d854d6f7fd7dd771fed591\nAuthor: Maxim Mikityanskiy \u003cmaximmi@mellanox.com\u003e\nDate: Wed Aug 7 17:46:15 2019 +0300\n\n net/mlx5e: Support LAG TX port affinity distribution\n \n When the VF LAG is in use, round-robin the TX affinity of channels among\n the different ports, if supported by the firmware. Create a set of TISes\n per port, while doing round-robin of the channels over the different\n sets. Let all SQs of a channel share the same set of TISes.\n \n If lag_tx_port_affinity HCA cap bit is supported, num_lag_ports \u003e 1 and\n we aren\u0027t the LACP owner (PF in the regular use), assign the affinities,\n otherwise use tx_affinity \u003d\u003d 0 in TIS context to let the FW assign the\n affinities itself. The TISes of the LACP owner are mapped only to the\n native physical port.\n \n For VFs, the starting port for round-robin is determined by its vhca_id,\n because a VF may have only one channel if attached to a single-core VM.\n \n Signed-off-by: Maxim Mikityanskiy \u003cmaximmi@mellanox.com\u003e\n Signed-off-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Signed-off-by: Mark Bloch \u003cmarkb@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 3c14562663c603bc523b6619a2b19a411e1cdc8e\nAuthor: Tariq Toukan \u003ctariqt@mellanox.com\u003e\nDate: Mon Jun 24 12:03:02 2019 +0300\n\n net/mlx5e: Expose new function for TIS destroy loop\n \n For better modularity and code sharing.\n Function internal change to be introduced in the next patches.\n \n Signed-off-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit d5e1c0ef550e4fa4577e6567f829a460a7f4ab6e\nAuthor: Erez Alfasi \u003cereza@mellanox.com\u003e\nDate: Wed Aug 7 17:56:02 2019 +0300\n\n net/mlx5e: ethtool, Fix a typo in WOL function names\n \n Fix a typo in \u0027mlx5e_refomrat_wol_mode_mlx5_to_linux\u0027 and\n \u0027mlx5e_refomrat_wol_mode_linux_to_mlx5\u0027 function names:\n \"refomrat\" -\u003e \"reformat\".\n \n Fixes: 928cfe8745a6 (\"net/mlx5e: Wake On LAN support\")\n Signed-off-by: Erez Alfasi \u003cereza@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 274b924088e93593c76fb122d24bc0ef18d0ddf4\nAuthor: Jeff Moyer \u003cjmoyer@redhat.com\u003e\nDate: Wed Aug 28 11:49:46 2019 -0400\n\n libnvdimm/pfn: Fix namespace creation on misaligned addresses\n \n Yi reported[1] that after commit a3619190d62e (\"libnvdimm/pfn: stop\n padding pmem namespaces to section alignment\"), it was no longer\n possible to create a device dax namespace with a 1G alignment. The\n reason was that the pmem region was not itself 1G-aligned. The code\n happily skips past the first 512M, but fails to account for a now\n misaligned end offset (since space was allocated starting at that\n misaligned address, and extending for size GBs). Reintroduce\n end_trunc, so that the code correctly handles the misaligned end\n address. This results in the same behavior as before the introduction\n of the offending commit.\n \n [1] https://lists.01.org/pipermail/linux-nvdimm/2019-July/022813.html\n \n Fixes: a3619190d62e (\"libnvdimm/pfn: stop padding pmem namespaces ...\")\n Reported-and-tested-by: Yi Zhang \u003cyi.zhang@redhat.com\u003e\n Signed-off-by: Jeff Moyer \u003cjmoyer@redhat.com\u003e\n Link: https://lore.kernel.org/r/x49ftll8f39.fsf@segfault.boston.devel.redhat.com\n Signed-off-by: Dan Williams \u003cdan.j.williams@intel.com\u003e\n\ncommit 0f7dc62068bb2b929e213e759dba1eb038448a5c\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 26 08:21:27 2019 +0100\n\n drm/i915: Protect our local workers against I915_FENCE_TIMEOUT\n \n Trust our own workers to not cause unnecessary delays and disable the\n automatic timeout on their asynchronous fence waits. (Along the same\n lines that we trust our own requests to complete eventually, if\n necessary by force.)\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190826072149.9447-6-chris@chris-wilson.co.uk\n\ncommit ce3b487f60189c56a4c3340ab7504971dac97338\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Wed Aug 21 02:09:39 2019 +0900\n\n init/Kconfig: rework help of CONFIG_CC_OPTIMIZE_FOR_SIZE\n \n CONFIG_CC_OPTIMIZE_FOR_SIZE was originally an independent boolean\n option, but commit 877417e6ffb9 (\"Kbuild: change CC_OPTIMIZE_FOR_SIZE\n definition\") turned it into a choice between _PERFORMANCE and _SIZE.\n \n The phrase \"If unsure, say N.\" sounds like an independent option.\n Reword the help text to make it appropriate for the choice menu.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit d20558d1de4e2d5bec236a8f0e99d5f1efdd39b8\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Wed Aug 14 19:54:00 2019 +0900\n\n docs: kbuild: remove cc-ldoption from document again\n \n Commit 055efab3120b (\"kbuild: drop support for cc-ldoption\") correctly\n removed the cc-ldoption from Documentation/kbuild/makefiles.txt, but\n commit cd238effefa2 (\"docs: kbuild: convert docs to ReST and rename\n to *.rst\") revived it. I guess it was a rebase mistake.\n \n Remove it again.\n \n Fixes: cd238effefa2 (\"docs: kbuild: convert docs to ReST and rename to *.rst\")\n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 4fef9dece5785f4ca522a13600de1014943d1fba\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Wed Aug 14 19:53:59 2019 +0900\n\n docs: kbuild: fix invalid ReST syntax\n \n I see the following warnings when I open this document with a ReST\n viewer, retext:\n \n /home/masahiro/ref/linux/Documentation/kbuild/makefiles.rst:1142: (WARNING/2) Inline emphasis start-string without end-string.\n /home/masahiro/ref/linux/Documentation/kbuild/makefiles.rst:1152: (WARNING/2) Inline emphasis start-string without end-string.\n /home/masahiro/ref/linux/Documentation/kbuild/makefiles.rst:1154: (WARNING/2) Inline emphasis start-string without end-string.\n \n These hunks were added by commit e846f0dc57f4 (\"kbuild: add support\n for ensuring headers are self-contained\") and commit 1e21cbfada87\n (\"kbuild: support header-test-pattern-y\"), respectively. They were\n written not for ReST but for the plain text, and merged via the\n kbuild tree.\n \n In the same development cycle, this document was converted to ReST\n by commit cd238effefa2 (\"docs: kbuild: convert docs to ReST and rename\n to *.rst\"), and merged via the doc sub-system.\n \n Merging them together into Linus\u0027 tree resulted in the current situation.\n \n To fix the syntax, surround the asterisks with back-quotes, and\n use :: for the code sample.\n \n Fixes: 39ceda5ce1b0 (\"Merge tag \u0027kbuild-v5.3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild\")\n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 8a84bacba19c154bdcb278657d0b7594ce2d939c\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Aug 23 13:07:11 2019 +0300\n\n drm/i915: Align power domain names with port names\n \n There is a difference in BSpec\u0027s and the driver\u0027s designation of DDI\n ports. BSpec uses the following names:\n - before GEN11:\n BSpec/driver:\n port A/B/C/D etc\n - GEN11:\n BSpec/driver:\n port A-F\n - GEN12:\n BSpec:\n port A/B/C for combo PHY ports\n port TC1-6 for Type C PHY ports\n driver:\n port A-I.\n The driver\u0027s port D name matches BSpec\u0027s TC1 port name.\n \n So far power domains were named according to the BSpec designation, to\n make it easier to match the code against the specification. That however\n can be confusing when a power domain needs to be matched to a port on\n GEN12+. To resolve that use the driver\u0027s port A-I designation for power\n domain names too and rename the corresponding power wells so that they\n reflect the mapping from the driver\u0027s to BSpec\u0027s port name.\n \n Cc: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: Stanislav Lisovskiy \u003cstanislav.lisovskiy@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823100711.27833-1-imre.deak@intel.com\n\ncommit 325b916a9e1b11d28c40b02e24734750946a1785\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 26 14:38:37 2019 +0100\n\n drm/i915/selftests: Ignore coherency failures on Broadwater\n \n We\u0027ve been ignoring similar coherency issues in IGT for Broadwater, and\n specifically Broadwater (original gen4) and not, for example, Crestline\n (same generation as Broadwater, but the mobile variant). Without any\n means to reproduce locally (I have a 965GM but alas no 965G), fixing will\n be slow, so tell CI to ignore any failure until we are ready with a fix.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Acked-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190826133837.6784-1-chris@chris-wilson.co.uk\n\ncommit 98c09b8c9767c61f97976c358c47ba17c026dc77\nAuthor: Lijun Ou \u003coulijun@huawei.com\u003e\nDate: Wed Aug 21 21:14:36 2019 +0800\n\n RDMA/hns: Fix wrong assignment of qp_access_flags\n \n We used wrong shifts when set qp_attr-\u003eqp_access_flag,\n this patch exchange V2_QP_RRE_S and V2_QP_RWE_S to fix it.\n \n Fixes: 2a3d923f8730 (\"RDMA/hns: Replace magic numbers with #defines\")\n Signed-off-by: Weihang Li \u003cliweihang@hisilicon.com\u003e\n Signed-off-by: Lijun Ou \u003coulijun@huawei.com\u003e\n Link: https://lore.kernel.org/r/1566393276-42555-10-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit afca2a2b837a8a93d8376f0843f225dd967aac12\nAuthor: Wenpeng Liang \u003cliangwenpeng@huawei.com\u003e\nDate: Wed Aug 21 21:14:35 2019 +0800\n\n RDMA/hns: Delete the not-used lines\n \n Delete the assignment of srq-\u003eibsrq.ext.xrc.srq_num, beacause this\n value is not used.\n \n Signed-off-by: Wenpeng Liang \u003cliangwenpeng@huawei.com\u003e\n Signed-off-by: Lijun Ou \u003coulijun@huawei.com\u003e\n Link: https://lore.kernel.org/r/1566393276-42555-9-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 18df508c7970c2c5e8bcaed48469bb28f9075e89\nAuthor: Wenpeng Liang \u003cliangwenpeng@huawei.com\u003e\nDate: Wed Aug 21 21:14:34 2019 +0800\n\n RDMA/hns: Remove if-else judgment statements for creating srq\n \n Because if the value of srqwqe_buf_pg_sz is zero, npages and\n ib_umem_page_count are equivalent, page_shif and PAGE_SHIFT\n are equivalent in hns_roce_create_srq. Here remove it.\n \n Signed-off-by: Wenpeng Liang \u003cliangwenpeng@huawei.com\u003e\n Signed-off-by: Lijun Ou \u003coulijun@huawei.com\u003e\n Link: https://lore.kernel.org/r/1566393276-42555-8-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit e075da5e7c47ee1c20b4e8f9a20c6105be99b10f\nAuthor: Lang Cheng \u003cchenglang@huawei.com\u003e\nDate: Wed Aug 21 21:14:33 2019 +0800\n\n RDMA/hns: Add reset process for function-clear\n \n If the hardware is resetting, the driver should not perform\n the mailbox operation.Function-clear needs to add relevant judgment.\n \n Signed-off-by: Lang Cheng \u003cchenglang@huawei.com\u003e\n Link: https://lore.kernel.org/r/1566393276-42555-7-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit bfe860351e31e71913d4e6c46aae5724b661a519\nAuthor: Lang Cheng \u003cchenglang@huawei.com\u003e\nDate: Wed Aug 21 21:14:32 2019 +0800\n\n RDMA/hns: Fix cast from or to restricted __le32 for driver\n \n Sparse is whining about the u32 and __le32 mixed usage in the driver.\n The roce_set_field() is used to __le32 data of hardware only.\n If a variable is not delivered to the hardware, the __le32 type and\n related operations are not required.\n \n Signed-off-by: Lang Cheng \u003cchenglang@huawei.com\u003e\n Link: https://lore.kernel.org/r/1566393276-42555-6-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 90c559b1864aa52518c10d5535bf5d5ea74189e5\nAuthor: Lijun Ou \u003coulijun@huawei.com\u003e\nDate: Wed Aug 21 21:14:31 2019 +0800\n\n RDMA/hns: Remove the some magic number\n \n Here uses the meaningful macro instead of the magic number\n for readability.\n \n Signed-off-by: Lijun Ou \u003coulijun@huawei.com\u003e\n Signed-off-by: Lang Chen \u003cchenglang@huawei.com\u003e\n Link: https://lore.kernel.org/r/1566393276-42555-5-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 82e620d9c3a096108e8a2f90ce7199bdad2040eb\nAuthor: Lang Cheng \u003cchenglang@huawei.com\u003e\nDate: Wed Aug 21 21:14:30 2019 +0800\n\n RDMA/hns: Modify the data structure of hns_roce_av\n \n we change type of some members to u32/u8 from __le32 as well as\n split sl_tclass_flowlabel into three variables in hns_roce_av.\n \n Signed-off-by: Lang Cheng \u003cchenglang@huawei.com\u003e\n Link: https://lore.kernel.org/r/1566393276-42555-4-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 7f313eda8fcc6a250803abb1aef3d6dc5c32f9ad\nAuthor: Eric Sandeen \u003csandeen@redhat.com\u003e\nDate: Tue Aug 27 17:35:12 2019 -0700\n\n xfs: log proper length of btree block in scrub/repair\n \n xfs_trans_log_buf() takes a final argument of the last byte to\n log in the buffer; b_length is in basic blocks, so this isn\u0027t\n the correct last byte. Fix it.\n \n Signed-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\n Reviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit ffb5696f7555e44c04b7c1212fcd83ddedc43498\nAuthor: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nDate: Mon Aug 26 17:06:05 2019 -0700\n\n xfs: reinitialize rm_flags when unpacking an offset into an rmap irec\n \n In xfs_rmap_irec_offset_unpack, we should always clear the contents of\n rm_flags before we begin unpacking the encoded (ondisk) offset into the\n incore rm_offset and incore rm_flags fields. Remove the open-coded\n field zeroing as this encourages api misuse.\n \n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Reviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n\ncommit 3e08f42ae7828bac2e7445a950f5de2b08203352\nAuthor: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nDate: Mon Aug 26 17:06:04 2019 -0700\n\n xfs: remove unnecessary int returns from deferred bmap functions\n \n Remove the return value from the functions that schedule deferred bmap\n operations since they never fail and do not return status.\n \n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Reviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n\ncommit 74b4c5d4a9c073162a37d1c20c95cb33152ca474\nAuthor: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nDate: Mon Aug 26 17:06:04 2019 -0700\n\n xfs: remove unnecessary int returns from deferred refcount functions\n \n Remove the return value from the functions that schedule deferred\n refcount operations since they never fail and do not return status.\n \n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Reviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n\ncommit bc46ac64713f11c86cbbe11a86abd2a71274b15f\nAuthor: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nDate: Mon Aug 26 17:06:03 2019 -0700\n\n xfs: remove unnecessary int returns from deferred rmap functions\n \n Remove the return value from the functions that schedule deferred rmap\n operations since they never fail and do not return status.\n \n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Reviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n\ncommit 2ca09177ab9dae7cc0a888137bd2421f5829cd87\nAuthor: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nDate: Mon Aug 26 17:06:02 2019 -0700\n\n xfs: remove unnecessary parameter from xfs_iext_inc_seq\n \n This function doesn\u0027t use the @state parameter, so get rid of it.\n \n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Reviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n\ncommit b521c89027f4b86af418083cb75db60a0e7a5997\nAuthor: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nDate: Mon Aug 26 17:06:02 2019 -0700\n\n xfs: fix sign handling problem in xfs_bmbt_diff_two_keys\n \n In xfs_bmbt_diff_two_keys, we perform a signed int64_t subtraction with\n two unsigned 64-bit quantities. If the second quantity is actually the\n \"maximum\" key (all ones) as used in _query_all, the subtraction\n effectively becomes addition of two positive numbers and the function\n returns incorrect results. Fix this with explicit comparisons of the\n unsigned values. Nobody needs this now, but the online repair patches\n will need this to work properly.\n \n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Reviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n\ncommit 7380e8fec16b3527baa1153255da9cc8d5d4d309\nAuthor: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nDate: Mon Aug 26 17:06:01 2019 -0700\n\n xfs: don\u0027t return _QUERY_ABORT from xfs_rmap_has_other_keys\n \n The xfs_rmap_has_other_keys helper aborts the iteration as soon as it\n has an answer. Don\u0027t let this abort leak out to callers.\n \n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Reviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n\ncommit c94613feefd7714ad0d03a1914632c15782114ff\nAuthor: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nDate: Mon Aug 26 17:06:01 2019 -0700\n\n xfs: fix maxicount division by zero error\n \n In xfs_ialloc_setup_geometry, it\u0027s possible for a malicious/corrupt fs\n image to set an unreasonably large value for sb_inopblog which will\n cause ialloc_blks to be zero. If sb_imax_pct is also set, this results\n in a division by zero error in the second do_div call. Therefore, force\n maxicount to zero if ialloc_blks is zero.\n \n Note that the kernel metadata verifiers will catch the garbage inopblog\n value and abort the fs mount long before it tries to set up the inode\n geometry; this is needed to avoid a crash in xfs_db while setting up the\n xfs_mount structure.\n \n Found by fuzzing sb_inopblog to 122 in xfs/350.\n \n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Reviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n Reviewed-by: Carlos Maiolino \u003ccmaiolino@redhat.com\u003e\n\ncommit e003f9af9b8dc110dd62b7eaa3a796fd81d4d256\nAuthor: Rui Miguel Silva \u003crui.silva@linaro.org\u003e\nDate: Wed Aug 28 13:48:25 2019 +0100\n\n staging: greybus: fix more header declarations\n \n More headers needed to be fixed when moving greybus out of staging and\n enabling the COMPILE_TEST option.\n \n Add forward declarations for the needed structures.\n \n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Rui Miguel Silva \u003crui.silva@linaro.org\u003e\n Reviewed-by: Johan Hovold \u003cjohan@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190828124825.20800-1-rui.silva@linaro.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 55b9a0549e4303ca43481f9ba2cdfa305277d97d\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Thu Aug 22 21:12:14 2019 -0500\n\n drm/panfrost: Use mutex_trylock in panfrost_gem_purge\n \n Lockdep reports a circular locking dependency with pages_lock taken in\n the shrinker callback. The deadlock can\u0027t actually happen with current\n users at least as a BO will never be purgeable when pages_lock is held.\n To be safe, let\u0027s use mutex_trylock() instead and bail if a BO is locked\n already.\n \n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823021216.5862-7-robh@kernel.org\n\ncommit edbd7f318cd66cec4588a28ba6985b014456dbca\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Thu Aug 22 21:12:13 2019 -0500\n\n drm/shmem: Use mutex_trylock in drm_gem_shmem_purge\n \n Lockdep reports a circular locking dependency with pages_lock taken in\n the shrinker callback. The deadlock can\u0027t actually happen with current\n users at least as a BO will never be purgeable when pages_lock is held.\n To be safe, let\u0027s use mutex_trylock() instead and bail if a BO is locked\n already.\n \n WARNING: possible circular locking dependency detected\n 5.3.0-rc1+ #100 Tainted: G L\n ------------------------------------------------------\n kswapd0/171 is trying to acquire lock:\n 000000009b9823fd (\u0026shmem-\u003epages_lock){+.+.}, at: drm_gem_shmem_purge+0x20/0x40\n \n but task is already holding lock:\n 00000000f82369b6 (fs_reclaim){+.+.}, at: __fs_reclaim_acquire+0x0/0x40\n \n which lock already depends on the new lock.\n \n the existing dependency chain (in reverse order) is:\n \n -\u003e #1 (fs_reclaim){+.+.}:\n fs_reclaim_acquire.part.18+0x34/0x40\n fs_reclaim_acquire+0x20/0x28\n __kmalloc_node+0x6c/0x4c0\n kvmalloc_node+0x38/0xa8\n drm_gem_get_pages+0x80/0x1d0\n drm_gem_shmem_get_pages+0x58/0xa0\n drm_gem_shmem_get_pages_sgt+0x48/0xd0\n panfrost_mmu_map+0x38/0xf8 [panfrost]\n panfrost_gem_open+0xc0/0xe8 [panfrost]\n drm_gem_handle_create_tail+0xe8/0x198\n drm_gem_handle_create+0x3c/0x50\n panfrost_gem_create_with_handle+0x70/0xa0 [panfrost]\n panfrost_ioctl_create_bo+0x48/0x80 [panfrost]\n drm_ioctl_kernel+0xb8/0x110\n drm_ioctl+0x244/0x3f0\n do_vfs_ioctl+0xbc/0x910\n ksys_ioctl+0x78/0xa8\n __arm64_sys_ioctl+0x1c/0x28\n el0_svc_common.constprop.0+0x90/0x168\n el0_svc_handler+0x28/0x78\n el0_svc+0x8/0xc\n \n -\u003e #0 (\u0026shmem-\u003epages_lock){+.+.}:\n __lock_acquire+0xa2c/0x1d70\n lock_acquire+0xdc/0x228\n __mutex_lock+0x8c/0x800\n mutex_lock_nested+0x1c/0x28\n drm_gem_shmem_purge+0x20/0x40\n panfrost_gem_shrinker_scan+0xc0/0x180 [panfrost]\n do_shrink_slab+0x208/0x500\n shrink_slab+0x10c/0x2c0\n shrink_node+0x28c/0x4d8\n balance_pgdat+0x2c8/0x570\n kswapd+0x22c/0x638\n kthread+0x128/0x130\n ret_from_fork+0x10/0x18\n \n other info that might help us debug this:\n \n Possible unsafe locking scenario:\n \n CPU0 CPU1\n ---- ----\n lock(fs_reclaim);\n lock(\u0026shmem-\u003epages_lock);\n lock(fs_reclaim);\n lock(\u0026shmem-\u003epages_lock);\n \n *** DEADLOCK ***\n \n 3 locks held by kswapd0/171:\n #0: 00000000f82369b6 (fs_reclaim){+.+.}, at: __fs_reclaim_acquire+0x0/0x40\n #1: 00000000ceb37808 (shrinker_rwsem){++++}, at: shrink_slab+0xbc/0x2c0\n #2: 00000000f31efa81 (\u0026pfdev-\u003eshrinker_lock){+.+.}, at: panfrost_gem_shrinker_scan+0x34/0x180 [panfrost]\n \n Fixes: 17acb9f35ed7 (\"drm/shmem: Add madvise state and purge helpers\")\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823021216.5862-6-robh@kernel.org\n\ncommit fe20d1cc95827e240cfcb21490b05ef38518ee4e\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Thu Aug 22 21:12:12 2019 -0500\n\n drm/shmem: Do dma_unmap_sg before purging pages\n \n Calling dma_unmap_sg() in drm_gem_shmem_free_object() is too late if the\n backing pages have already been released by the shrinker. The result is\n the following abort:\n \n Unable to handle kernel paging request at virtual address ffff8000098ed000\n Mem abort info:\n ESR \u003d 0x96000147\n Exception class \u003d DABT (current EL), IL \u003d 32 bits\n SET \u003d 0, FnV \u003d 0\n EA \u003d 0, S1PTW \u003d 0\n Data abort info:\n ISV \u003d 0, ISS \u003d 0x00000147\n CM \u003d 1, WnR \u003d 1\n swapper pgtable: 4k pages, 48-bit VAs, pgdp\u003d0000000002f51000\n [ffff8000098ed000] pgd\u003d00000000401f8003, pud\u003d00000000401f7003, pmd\u003d00000000401b1003, pte\u003d00e80000098ed712\n Internal error: Oops: 96000147 [#1] SMP\n Modules linked in: panfrost gpu_sched\n CPU: 5 PID: 902 Comm: gnome-shell Not tainted 5.3.0-rc1+ #95\n Hardware name: 96boards Rock960 (DT)\n pstate: 40000005 (nZcv daif -PAN -UAO)\n pc : __dma_inv_area+0x40/0x58\n lr : arch_sync_dma_for_cpu+0x28/0x30\n sp : ffff00001321ba30\n x29: ffff00001321ba30 x28: ffff00001321bd08\n x27: 0000000000000000 x26: 0000000000000009\n x25: 0000ffffc1f86170 x24: 0000000000000000\n x23: 0000000000000000 x22: 0000000000000000\n x21: 0000000000021000 x20: ffff80003bb2d810\n x19: 00000000098ed000 x18: 0000000000000000\n x17: 0000000000000000 x16: ffff800023fd9480\n x15: 0000000000000000 x14: 0000000000000000\n x13: 0000000000000000 x12: 0000000000000000\n x11: 00000000fffb9fff x10: 0000000000000000\n x9 : 0000000000000000 x8 : ffff800023fd9c18\n x7 : 0000000000000000 x6 : 00000000ffffffff\n x5 : 0000000000000000 x4 : 0000000000021000\n Purging 5693440 bytes\n x3 : 000000000000003f x2 : 0000000000000040\n x1 : ffff80000990e000 x0 : ffff8000098ed000\n Call trace:\n __dma_inv_area+0x40/0x58\n dma_direct_sync_single_for_cpu+0x7c/0x80\n dma_direct_unmap_page+0x80/0x88\n dma_direct_unmap_sg+0x54/0x80\n drm_gem_shmem_free_object+0xfc/0x108\n panfrost_gem_free_object+0x118/0x128 [panfrost]\n drm_gem_object_free+0x18/0x90\n drm_gem_object_put_unlocked+0x58/0x80\n drm_gem_object_handle_put_unlocked+0x64/0xb0\n drm_gem_object_release_handle+0x70/0x98\n drm_gem_handle_delete+0x64/0xb0\n drm_gem_close_ioctl+0x28/0x38\n drm_ioctl_kernel+0xb8/0x110\n drm_ioctl+0x244/0x3f0\n do_vfs_ioctl+0xbc/0x910\n ksys_ioctl+0x78/0xa8\n __arm64_sys_ioctl+0x1c/0x28\n el0_svc_common.constprop.0+0x88/0x150\n el0_svc_handler+0x28/0x78\n el0_svc+0x8/0xc\n Code: 8a230000 54000060 d50b7e20 14000002 (d5087620)\n \n Fixes: 17acb9f35ed7 (\"drm/shmem: Add madvise state and purge helpers\")\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823021216.5862-5-robh@kernel.org\n\ncommit aebe8c22a9121e89cdd5c725f2ebed1ee237ff16\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Thu Aug 22 21:12:09 2019 -0500\n\n drm/panfrost: Fix possible suspend in panfrost_remove\n \n Calls to panfrost_device_fini() access the h/w, but we already done a\n pm_runtime_put_sync_autosuspend() beforehand. This only works if the\n autosuspend delay is long enough. A 0ms delay will hang the system when\n removing the device. Fix this by moving the pm_runtime_put_sync_suspend()\n after the panfrost_device_fini() call.\n \n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823021216.5862-2-robh@kernel.org\n\ncommit c90f30812a79abeb9f7f0f78e69d314104b825be\nAuthor: Steven Price \u003csteven.price@arm.com\u003e\nDate: Thu Aug 22 10:32:18 2019 +0100\n\n drm/panfrost: Add missing check for pfdev-\u003eregulator\n \n When modifying panfrost_devfreq_target() to support a device without a\n regulator defined I missed the check on the error path. Let\u0027s add it.\n \n Reported-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Fixes: e21dd290881b (\"drm/panfrost: Enable devfreq to work without regulator\")\n Signed-off-by: Steven Price \u003csteven.price@arm.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190822093218.26014-1-steven.price@arm.com\n\ncommit 5b8f3df6239c3a9b625ab4bdc69c54d4768a4f06\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Wed Aug 28 16:19:55 2019 +0200\n\n pagewalk: use lockdep_assert_held for locking validation\n \n Use lockdep to check for held locks instead of using home grown asserts.\n \n Link: https://lore.kernel.org/r/20190828141955.22210-4-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 923bfc561e7535f7dc2be136da75690582268cf2\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Wed Aug 28 16:19:54 2019 +0200\n\n pagewalk: separate function pointers from iterator data\n \n The mm_walk structure currently mixed data and code. Split out the\n operations vectors into a new mm_walk_ops structure, and while we are\n changing the API also declare the mm_walk structure inside the\n walk_page_range and walk_page_vma functions.\n \n Based on patch from Linus Torvalds.\n \n Link: https://lore.kernel.org/r/20190828141955.22210-3-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 4e10e8c36663a011f77d39c937aaa473fad90de3\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Wed Aug 28 16:19:53 2019 +0200\n\n mm: split out a new pagewalk.h header from mm.h\n \n Add a new header for the two handful of users of the walk_page_range /\n walk_page_vma interface instead of polluting all users of mm.h with it.\n \n Link: https://lore.kernel.org/r/20190828141955.22210-2-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 318dd7698a435faad72d9971ddd589aa4c9743f8\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Mon Aug 26 22:14:25 2019 +0200\n\n mm/mmu_notifiers: annotate with might_sleep()\n \n Since mmu notifiers don\u0027t exist for many processes, but could block in\n interesting places, add some annotations. This should help make sure the\n core mm keeps up its end of the mmu notifier contract.\n \n The checks here are outside of all notifier checks because of that.\n They compile away without CONFIG_DEBUG_ATOMIC_SLEEP.\n \n Link: https://lore.kernel.org/r/20190826201425.17547-6-daniel.vetter@ffwll.ch\n Suggested-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 45256bbec203e6fc81a23eab24f50b098bbb5e2a\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Mon Aug 26 22:14:22 2019 +0200\n\n mm/mmu_notifiers: prime lockdep\n \n We want to teach lockdep that mmu notifiers can be called from direct\n reclaim paths, since on many CI systems load might never reach that\n level (e.g. when just running fuzzer or small functional tests).\n \n I\u0027ve put the annotation into mmu_notifier_register since only when we have\n mmu notifiers registered is there any point in teaching lockdep about\n them. Also, we already have a kmalloc(, GFP_KERNEL), so this is safe.\n \n Link: https://lore.kernel.org/r/20190826201425.17547-3-daniel.vetter@ffwll.ch\n Suggested-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit d2b219ed03d45a9799e4ba780c209edf9c510d3b\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Mon Aug 26 22:14:21 2019 +0200\n\n mm/mmu_notifiers: add a lockdep map for invalidate_range_start/end\n \n This is a similar idea to the fs_reclaim fake lockdep lock. It\u0027s fairly\n easy to provoke a specific notifier to be run on a specific range: Just\n prep it, and then munmap() it.\n \n A bit harder, but still doable, is to provoke the mmu notifiers for all\n the various callchains that might lead to them. But both at the same time\n is really hard to reliably hit, especially when you want to exercise paths\n like direct reclaim or compaction, where it\u0027s not easy to control what\n exactly will be unmapped.\n \n By introducing a lockdep map to tie them all together we allow lockdep to\n see a lot more dependencies, without having to actually hit them in a\n single challchain while testing.\n \n On Jason\u0027s suggestion this is is rolled out for both\n invalidate_range_start and invalidate_range_end. They both have the same\n calling context, hence we can share the same lockdep map. Note that the\n annotation for invalidate_ranage_start is outside of the\n mm_has_notifiers(), to make sure lockdep is informed about all paths\n leading to this context irrespective of whether mmu notifiers are present\n for a given context. We don\u0027t do that on the invalidate_range_end side to\n avoid paying the overhead twice, there the lockdep annotation is pushed\n down behind the mm_has_notifiers() check.\n \n Link: https://lore.kernel.org/r/20190826201425.17547-2-daniel.vetter@ffwll.ch\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit f0ade90a8aa1ea523eb366d1d1e8bd3463d9cf8a\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Wed Aug 28 16:21:09 2019 +0200\n\n mm/mmu_notifiers: remove the __mmu_notifier_invalidate_range_start/end exports\n \n No modular code uses these, which makes a lot of sense given the wrappers\n around them are only called by core mm code.\n \n Link: https://lore.kernel.org/r/20190828142109.29012-1-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit e8e4eb0fbeda570b16464208aebf5caccfb6eb95\nAuthor: Jonathan Neuschäfer \u003cj.neuschaefer@gmx.net\u003e\nDate: Thu Aug 8 20:19:48 2019 +0200\n\n asm-generic/div64: Fix documentation of do_div() parameter\n \n Contrary to the description, the first parameter (n) should not be passed\n as a pointer, but directly as an lvalue. This is possible because do_div() is\n a macro.\n \n Signed-off-by: Jonathan Neuschäfer \u003cj.neuschaefer@gmx.net\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Link: https://lkml.kernel.org/r/20190808181948.27659-1-j.neuschaefer@gmx.net\n\ncommit 67d94c21361959668da5b32ba1b1c1d4a3ddba21\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Wed Aug 28 16:34:36 2019 +0200\n\n ALSA: hda - Update descriptions about new position_fix values\n \n The new values for the recent Intel and AMD chips are missing in the\n documentation. Add the new descriptions.\n \n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 75e46fc02c975f401e70a53ecd55d475081d13a3\nAuthor: Michael Guralnik \u003cmichaelgur@mellanox.com\u003e\nDate: Mon Aug 19 15:08:15 2019 +0300\n\n IB/mlx5: Add page fault handler for DC initiator WQE\n \n Parsing DC initiator WQEs upon page fault requires skipping an address\n vector segment, as in UD WQEs.\n \n Link: https://lore.kernel.org/r/20190819120815.21225-4-leon@kernel.org\n Signed-off-by: Michael Guralnik \u003cmichaelgur@mellanox.com\u003e\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 29af94987ba68a0244b73e8a0b051addb8853f93\nAuthor: Michael Guralnik \u003cmichaelgur@mellanox.com\u003e\nDate: Mon Aug 19 15:08:14 2019 +0300\n\n IB/mlx5: Remove check of FW capabilities in ODP page fault handling\n \n As page fault handling is initiated by FW, there is no need to check that\n the ODP supports the operation and transport.\n \n Link: https://lore.kernel.org/r/20190819120815.21225-3-leon@kernel.org\n Signed-off-by: Michael Guralnik \u003cmichaelgur@mellanox.com\u003e\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 6350d0ce1b8ed53ac8d332cda15eb00bd78ee21d\nAuthor: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\nDate: Mon Jun 3 16:58:59 2019 +0200\n\n btrfs: add sha256 as another checksum algorithm\n \n Now that we everything in place, we can add SHA-256 as another checksum\n algorithm.\n \n SHA-256 was taken as it was the cryptographically strongest algorithm that\n can fit into the 32 Bytes we have left.\n \n Signed-off-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit eec947e0991c3b3fe4fb0a842f2d66b1ddb9f983\nAuthor: Qu Wenruo \u003cwqu@suse.com\u003e\nDate: Thu Jul 25 14:12:22 2019 +0800\n\n btrfs: ctree: Checking key orders before merged tree blocks\n \n [BUG]\n With crafted image, btrfs can panic at btrfs_del_csums().\n kernel BUG at fs/btrfs/ctree.c:3188!\n invalid opcode: 0000 [#1] SMP PTI\n CPU: 0 PID: 1156 Comm: btrfs-transacti Not tainted 5.0.0-rc8+ #9\n RIP: 0010:btrfs_set_item_key_safe+0x16c/0x180\n Code: b7 48 8d 7d bf 4c 89 fe 48 89 45 c8 0f b6 45 b6 88 45 c7 48 8b 45 ae 48 89 45 bf e8 ce f2 ff ff 85 c0 0f 8f 48 ff ff ff 0f 0b \u003c0f\u003e 0b e8 dd 8d be ff 0f 1f 00 66 2e 0f 1f 84 00 00 00 00 00 66 66\n RSP: 0018:ffff976141257ab8 EFLAGS: 00010202\n RAX: 0000000000000001 RBX: ffff898a6b890930 RCX: 0000000004b70000\n RDX: 0000000000000000 RSI: ffff976141257bae RDI: ffff976141257acf\n RBP: ffff976141257b10 R08: 0000000000001000 R09: ffff9761412579a8\n R10: 0000000000000000 R11: 0000000000000000 R12: ffff976141257abe\n R13: 0000000000000003 R14: ffff898a6a8be578 R15: ffff976141257bae\n FS: 0000000000000000(0000) GS:ffff898a77a00000(0000) knlGS:0000000000000000\n CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n CR2: 00007f779d9cd624 CR3: 000000022b2b4006 CR4: 00000000000206f0\n Call Trace:\n truncate_one_csum+0xac/0xf0\n btrfs_del_csums+0x24f/0x3a0\n __btrfs_free_extent.isra.72+0x5a7/0xbe0\n __btrfs_run_delayed_refs+0x539/0x1120\n btrfs_run_delayed_refs+0xdb/0x1b0\n btrfs_commit_transaction+0x52/0x950\n ? start_transaction+0x94/0x450\n transaction_kthread+0x163/0x190\n kthread+0x105/0x140\n ? btrfs_cleanup_transaction+0x560/0x560\n ? kthread_destroy_worker+0x50/0x50\n ret_from_fork+0x35/0x40\n Modules linked in:\n ---[ end trace 93bf9db00e6c374e ]---\n \n [CAUSE]\n This crafted image has a very tricky key order corruption:\n \n checksum tree key (CSUM_TREE ROOT_ITEM 0)\n node 29741056 level 1 items 14 free 107 generation 19 owner CSUM_TREE\n ...\n key (EXTENT_CSUM EXTENT_CSUM 73785344) block 29757440 gen 19\n key (EXTENT_CSUM EXTENT_CSUM 77594624) block 29753344 gen 19\n ...\n \n leaf 29757440 items 5 free space 150 generation 19 owner CSUM_TREE\n item 0 key (EXTENT_CSUM EXTENT_CSUM 73785344) itemoff 2323 itemsize 1672\n range start 73785344 end 75497472 length 1712128\n item 1 key (EXTENT_CSUM EXTENT_CSUM 75497472) itemoff 2319 itemsize 4\n range start 75497472 end 75501568 length 4096\n item 2 key (EXTENT_CSUM EXTENT_CSUM 75501568) itemoff 579 itemsize 1740\n range start 75501568 end 77283328 length 1781760\n item 3 key (EXTENT_CSUM EXTENT_CSUM 77283328) itemoff 575 itemsize 4\n range start 77283328 end 77287424 length 4096\n item 4 key (EXTENT_CSUM EXTENT_CSUM 4120596480) itemoff 275 itemsize 300 \u003c\u003c\u003c\n range start 4120596480 end 4120903680 length 307200\n leaf 29753344 items 3 free space 1936 generation 19 owner CSUM_TREE\n item 0 key (18446744073457893366 EXTENT_CSUM 77594624) itemoff 2323 itemsize 1672\n range start 77594624 end 79306752 length 1712128\n ...\n \n Note the item 4 key of leaf 29757440, which is obviously too large, and\n even larger than the first key of the next leaf.\n \n However it still follows the key order in that tree block, thus tree\n checker is unable to detect it at read time, since tree checker can only\n work inside a leaf, thus such complex corruption can\u0027t be rejected in\n advance.\n \n [FIX]\n The next timing to detect such problem is at tree block merge time,\n which is in push_node_left(), balance_node_right(), push_leaf_left() and\n push_leaf_right().\n \n Now we check if the key order of the right most key of the left node is\n larger than the left most key of the right node.\n \n By this we don\u0027t need to call the full tree-check, while still keeps the\n key order correct as key order in each node is already checked by tree\n checker thus we only need to check the above two slots.\n \n Link: https://bugzilla.kernel.org/show_bug.cgi?id\u003d202833\n Signed-off-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 6ee09c4f608974cc59fc86296d2c531756062564\nAuthor: Qu Wenruo \u003cwqu@suse.com\u003e\nDate: Thu Jul 25 14:12:21 2019 +0800\n\n btrfs: extent-tree: Kill the BUG_ON() in insert_inline_extent_backref()\n \n [BUG]\n With crafted image, btrfs can panic at insert_inline_extent_backref():\n kernel BUG at fs/btrfs/extent-tree.c:1857!\n invalid opcode: 0000 [#1] SMP PTI\n CPU: 0 PID: 1117 Comm: btrfs-transacti Not tainted 5.0.0-rc8+ #9\n RIP: 0010:insert_inline_extent_backref+0xcc/0xe0\n Code: 45 20 49 8b 7e 50 49 89 d8 4c 8b 4d 10 48 8b 55 c8 4c 89 e1 41 57 4c 89 ee 50 ff 75 18 e8 cc bf ff ff 31 c0 48 83 c4 18 eb b2 \u003c0f\u003e 0b e8 9d df bd ff 0f 1f 00 66 2e 0f 1f 84 00 00 00 00 00 66 66\n RSP: 0018:ffffac4dc1287be8 EFLAGS: 00010293\n RAX: 0000000000000000 RBX: 0000000000000007 RCX: 0000000000000001\n RDX: 0000000000001000 RSI: 0000000000000000 RDI: 0000000000000000\n RBP: ffffac4dc1287c28 R08: ffffac4dc1287ab8 R09: ffffac4dc1287ac0\n R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000\n R13: ffff8febef88a540 R14: ffff8febeaa7bc30 R15: 0000000000000000\n FS: 0000000000000000(0000) GS:ffff8febf7a00000(0000) knlGS:0000000000000000\n CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n CR2: 00007f663ace94c0 CR3: 0000000235698006 CR4: 00000000000206f0\n Call Trace:\n ? _cond_resched+0x1a/0x50\n __btrfs_inc_extent_ref.isra.64+0x7e/0x240\n ? btrfs_merge_delayed_refs+0xa5/0x330\n __btrfs_run_delayed_refs+0x653/0x1120\n btrfs_run_delayed_refs+0xdb/0x1b0\n btrfs_commit_transaction+0x52/0x950\n ? start_transaction+0x94/0x450\n transaction_kthread+0x163/0x190\n kthread+0x105/0x140\n ? btrfs_cleanup_transaction+0x560/0x560\n ? kthread_destroy_worker+0x50/0x50\n ret_from_fork+0x35/0x40\n Modules linked in:\n ---[ end trace 2ad8b3de903cf825 ]---\n \n [CAUSE]\n Due to extent tree corruption (still valid by itself, but bad cross ref),\n we can allocate an extent which is still in extent tree.\n The offending tree block of that case is from csum tree.\n The newly allocated tree block is also for csum tree.\n \n Then we will try to insert an tree block ref for the existing tree block\n ref.\n \n For btrfs tree extent item, a tree block can never be shared directly by\n the same tree twice.\n We have such BUG_ON() to prevent such problem, but BUG_ON() is\n definitely not good enough.\n \n [FIX]\n Replace that BUG_ON() with proper error message and leaf dump for debug\n build.\n \n Link: https://bugzilla.kernel.org/show_bug.cgi?id\u003d202829\n Signed-off-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 8d8ae38ea61b67ad6fdd2946f9994fb4beb2b320\nAuthor: Qu Wenruo \u003cwqu@suse.com\u003e\nDate: Thu Jul 25 14:12:20 2019 +0800\n\n btrfs: Detect unbalanced tree with empty leaf before crashing btree operations\n \n [BUG]\n With crafted image, btrfs will panic at btree operations:\n kernel BUG at fs/btrfs/ctree.c:3894!\n invalid opcode: 0000 [#1] SMP PTI\n CPU: 0 PID: 1138 Comm: btrfs-transacti Not tainted 5.0.0-rc8+ #9\n RIP: 0010:__push_leaf_left+0x6b6/0x6e0\n Code: 00 00 48 98 48 8d 04 80 48 8d 74 80 65 e8 42 5a 04 00 48 8b bd 78 ff ff ff 8b bf 90 d0 00 00 89 7d 98 83 ef 65 e9 06 ff ff ff \u003c0f\u003e 0b 0f 0b 48 8b 85 78 ff ff ff 8b 90 90 d0 00 00 e9 eb fe ff ff\n RSP: 0018:ffffc0bd4128b990 EFLAGS: 00010246\n RAX: 0000000000000000 RBX: ffffa0a4ab8f0e38 RCX: 0000000000000000\n RDX: ffffa0a280000000 RSI: 0000000000000000 RDI: ffffa0a4b3814000\n RBP: ffffc0bd4128ba38 R08: 0000000000001000 R09: ffffc0bd4128b948\n R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000240\n R13: ffffa0a4b556fb60 R14: ffffa0a4ab8f0af0 R15: ffffa0a4ab8f0af0\n FS: 0000000000000000(0000) GS:ffffa0a4b7a00000(0000) knlGS:0000000000000000\n CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n CR2: 00007f2461c80020 CR3: 000000022b32a006 CR4: 00000000000206f0\n Call Trace:\n ? _cond_resched+0x1a/0x50\n push_leaf_left+0x179/0x190\n btrfs_del_items+0x316/0x470\n btrfs_del_csums+0x215/0x3a0\n __btrfs_free_extent.isra.72+0x5a7/0xbe0\n __btrfs_run_delayed_refs+0x539/0x1120\n btrfs_run_delayed_refs+0xdb/0x1b0\n btrfs_commit_transaction+0x52/0x950\n ? start_transaction+0x94/0x450\n transaction_kthread+0x163/0x190\n kthread+0x105/0x140\n ? btrfs_cleanup_transaction+0x560/0x560\n ? kthread_destroy_worker+0x50/0x50\n ret_from_fork+0x35/0x40\n Modules linked in:\n ---[ end trace c2425e6e89b5558f ]---\n \n [CAUSE]\n The offending csum tree looks like this:\n checksum tree key (CSUM_TREE ROOT_ITEM 0)\n node 29741056 level 1 items 14 free 107 generation 19 owner CSUM_TREE\n ...\n key (EXTENT_CSUM EXTENT_CSUM 85975040) block 29630464 gen 17\n key (EXTENT_CSUM EXTENT_CSUM 89911296) block 29642752 gen 17 \u003c\u003c\u003c\n key (EXTENT_CSUM EXTENT_CSUM 92274688) block 29646848 gen 17\n ...\n \n leaf 29630464 items 6 free space 1 generation 17 owner CSUM_TREE\n item 0 key (EXTENT_CSUM EXTENT_CSUM 85975040) itemoff 3987 itemsize 8\n range start 85975040 end 85983232 length 8192\n ...\n leaf 29642752 items 0 free space 3995 generation 17 owner 0\n ^ empty leaf invalid owner ^\n \n leaf 29646848 items 1 free space 602 generation 17 owner CSUM_TREE\n item 0 key (EXTENT_CSUM EXTENT_CSUM 92274688) itemoff 627 itemsize 3368\n range start 92274688 end 95723520 length 3448832\n \n So we have a corrupted csum tree where one tree leaf is completely\n empty, causing unbalanced btree, thus leading to unexpected btree\n balance error.\n \n [FIX]\n For this particular case, we handle it in two directions to catch it:\n - Check if the tree block is empty through btrfs_verify_level_key()\n So that invalid tree blocks won\u0027t be read out through\n btrfs_search_slot() and its variants.\n \n - Check 0 tree owner in tree checker\n NO tree is using 0 as its tree owner, detect it and reject at tree\n block read time.\n \n Link: https://bugzilla.kernel.org/show_bug.cgi?id\u003d202821\n Signed-off-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 7e3c60ecf47f35784d3e19c1e64c6d607f882ffe\nAuthor: Qu Wenruo \u003cwqu@suse.com\u003e\nDate: Thu Jul 25 14:12:19 2019 +0800\n\n btrfs: extent-tree: Kill BUG_ON() in __btrfs_free_extent() and do better comment\n \n __btrfs_free_extent() is one of the best cases to show how optimization\n could make a function hard to read.\n \n In fact __btrfs_free_extent() is only doing two major works:\n 1. Reduce the refs number of an extent backref\n Either it\u0027s an inlined extent backref (inside EXTENT/METADATA item) or\n a keyed extent backref (SHARED_* item).\n We only need to locate that backref line, either reduce the number or\n remove the backref line completely.\n \n 2. Update the refs count in EXTENT/METADATA_ITEM\n \n But in real world, we do it in a complex but somewhat efficient way.\n During step 1), we will try to locate the EXTENT/METADATA_ITEM without\n triggering another btrfs_search_slot() as fast path.\n \n Only when we failed to locate that item, we will trigger another\n btrfs_search_slot() to get that EXTENT/METADATA_ITEM after we\n updated/deleted the backref line.\n \n And we have a lot of restrict check on things like refs_to_drop against\n extent refs and special case check for single ref extent.\n \n All of these results:\n - 7 BUG_ON()s in a single function\n Although all these BUG_ON() are doing correct check, they\u0027re super\n easy to get triggered for fuzzed images.\n It\u0027s never a good idea to piss the end user.\n \n - Near 300 lines without much useful comments but a lot of hidden\n conditions\n I believe even the author needs several minutes to recall what the\n code is doing\n Not to mention a lot of BUG_ON() conditions needs to go back tens of\n lines to find out why.\n \n This patch address all these problems by:\n - Introduce two examples to show what __btrfs_free_extent() is doing\n One inlined backref case and one keyed case.\n Should cover most cases.\n \n - Kill all BUG_ON()s with proper error message and optional leaf dump\n \n - Add comment to show the overall workflow\n \n Link: https://bugzilla.kernel.org/show_bug.cgi?id\u003d202819\n [ The report triggers one BUG_ON() in __btrfs_free_extent() ]\n Signed-off-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit ef271e612b5fea5708f3180ba57846c154a9d092\nAuthor: Qu Wenruo \u003cwqu@suse.com\u003e\nDate: Thu Jul 25 14:12:18 2019 +0800\n\n btrfs: extent_io: Do extra check for extent buffer read write functions\n \n Although we have start, len check for extent buffer reader/write (e.g.\n read_extent_buffer()), those checks has its limitations:\n - No overflow check\n Values like start \u003d 1024 len \u003d -1024 can still pass the basic\n (start + len) \u003e eb-\u003elen check.\n \n - Checks are not consistent\n For read_extent_buffer() we only check (start + len) against eb-\u003elen.\n While for memcmp_extent_buffer() we also check start against eb-\u003elen.\n \n - Different error reporting mechanism\n We use WARN() in read_extent_buffer() but BUG() in\n memcpy_extent_buffer().\n \n - Still modify memory if the request is obviously wrong\n In read_extent_buffer() even we find (start + len) \u003e eb-\u003elen, we still\n call memset(dst, 0, len), which can eaisly cause memory access error\n if start + len overflows.\n \n To address above problems, this patch creates a new common function to\n check such access, check_eb_range().\n - Add overflow check\n This function checks start, start + len against eb-\u003elen and overflow\n check.\n \n - Unified checks\n \n - Unified error reports\n Will call WARN() if CONFIG_BTRFS_DEBUG is configured.\n And also do btrfs_warn() message for non-debug build.\n \n - Exit ASAP if check fails\n No more possible memory corruption.\n \n - Add extra comment for @start @len used in those functions\n Even experienced developers sometimes get confused with the @start\n @len with logical address in those functions.\n I\u0027m not sure what\u0027s the cause, maybe it\u0027s the extent_buffer::start\n naming.\n For now, just add some comment.\n \n Link: https://bugzilla.kernel.org/show_bug.cgi?id\u003d202817\n [ Inspired by above report, the report itself is already addressed ]\n Signed-off-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 21113fbc17d6e9c20c5ccc5cdf3488c05a37e5dc\nAuthor: Chris Mason \u003cclm@fb.com\u003e\nDate: Wed Jul 10 12:28:18 2019 -0700\n\n Btrfs: extent_write_locked_range() should attach inode-\u003ei_wb\n \n extent_write_locked_range() is used when we\u0027re falling back to buffered\n IO from inside of compression. It allocates its own wbc and should\n associate it with the inode\u0027s i_wb to make sure the IO goes down from\n the correct cgroup.\n \n Signed-off-by: Chris Mason \u003cclm@fb.com\u003e\n Reviewed-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 29bc0f7f648a68d9e9b69a7f60e288b188962f53\nAuthor: Chris Mason \u003cclm@fb.com\u003e\nDate: Wed Jul 10 12:28:17 2019 -0700\n\n Btrfs: use REQ_CGROUP_PUNT for worker thread submitted bios\n \n Async CRCs and compression submit IO through helper threads, which\n means they have IO priority inversions when cgroup IO controllers are\n in use.\n \n This flags all of the writes submitted by btrfs helper threads as\n REQ_CGROUP_PUNT. submit_bio() will punt these to dedicated per-blkcg\n work items to avoid the priority inversion.\n \n For the compression code, we take a reference on the wbc\u0027s blkg css and\n pass it down to the async workers.\n \n For the async crcs, the bio already has the correct css, we just need to\n tell the block layer to use REQ_CGROUP_PUNT.\n \n Signed-off-by: Chris Mason \u003cclm@fb.com\u003e\n Modified-and-reviewed-by: Tejun Heo \u003ctj@kernel.org\u003e\n Reviewed-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit c08c353edfbc6a2e86a1a77e6a72fb120ab50cff\nAuthor: Chris Mason \u003cclm@fb.com\u003e\nDate: Wed Jul 10 12:28:16 2019 -0700\n\n Btrfs: only associate the locked page with one async_cow struct\n \n The btrfs writepages function collects a large range of pages flagged\n for delayed allocation, and then sends them down through the COW code\n for processing. When compression is on, we allocate one async_cow\n structure for every 512K, and then run those pages through the\n compression code for IO submission.\n \n writepages starts all of this off with a single page, locked by\n the original call to extent_write_cache_pages(), and it\u0027s important to\n keep track of this page because it has already been through\n clear_page_dirty_for_io().\n \n The btrfs async_cow struct has a pointer to the locked_page, and when\n we\u0027re redirtying the page because compression had to fallback to\n uncompressed IO, we use page-\u003eindex to decide if a given async_cow\n struct really owns that page.\n \n But, this is racey. If a given delalloc range is broken up into two\n async_cows (cow_A and cow_B), we can end up with something like this:\n \n compress_file_range(cowA)\n submit_compress_extents(cowA)\n submit compressed bios(cowA)\n put_page(locked_page)\n \n compress_file_range(cowB)\n ...\n \n The end result is that cowA is completed and cleaned up before cowB even\n starts processing. This means we can free locked_page() and reuse it\n elsewhere. If we get really lucky, it\u0027ll have the same page-\u003eindex in\n its new home as it did before.\n \n While we\u0027re processing cowB, we might decide we need to fall back to\n uncompressed IO, and so compress_file_range() will call\n __set_page_dirty_nobufers() on cowB-\u003elocked_page.\n \n Without cgroups in use, this creates as a phantom dirty page, which\n isn\u0027t great but isn\u0027t the end of the world. With cgroups in use, we\n might crash in the accounting code because page-\u003emapping-\u003ei_wb isn\u0027t\n set.\n \n [ 8308.523110] BUG: unable to handle kernel NULL pointer dereference at 00000000000000d0\n [ 8308.531084] IP: percpu_counter_add_batch+0x11/0x70\n [ 8308.538371] PGD 66534e067 P4D 66534e067 PUD 66534f067 PMD 0\n [ 8308.541750] Oops: 0000 [#1] SMP DEBUG_PAGEALLOC\n [ 8308.551948] CPU: 16 PID: 2172 Comm: rm Not tainted\n [ 8308.566883] RIP: 0010:percpu_counter_add_batch+0x11/0x70\n [ 8308.567891] RSP: 0018:ffffc9000a97bbe0 EFLAGS: 00010286\n [ 8308.568986] RAX: 0000000000000005 RBX: 0000000000000090 RCX: 0000000000026115\n [ 8308.570734] RDX: 0000000000000030 RSI: ffffffffffffffff RDI: 0000000000000090\n [ 8308.572543] RBP: 0000000000000000 R08: fffffffffffffff5 R09: 0000000000000000\n [ 8308.573856] R10: 00000000000260c0 R11: ffff881037fc26c0 R12: ffffffffffffffff\n [ 8308.580099] R13: ffff880fe4111548 R14: ffffc9000a97bc90 R15: 0000000000000001\n [ 8308.582520] FS: 00007f5503ced480(0000) GS:ffff880ff7200000(0000) knlGS:0000000000000000\n [ 8308.585440] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n [ 8308.587951] CR2: 00000000000000d0 CR3: 00000001e0459005 CR4: 0000000000360ee0\n [ 8308.590707] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\n [ 8308.592865] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\n [ 8308.594469] Call Trace:\n [ 8308.595149] account_page_cleaned+0x15b/0x1f0\n [ 8308.596340] __cancel_dirty_page+0x146/0x200\n [ 8308.599395] truncate_cleanup_page+0x92/0xb0\n [ 8308.600480] truncate_inode_pages_range+0x202/0x7d0\n [ 8308.617392] btrfs_evict_inode+0x92/0x5a0\n [ 8308.619108] evict+0xc1/0x190\n [ 8308.620023] do_unlinkat+0x176/0x280\n [ 8308.621202] do_syscall_64+0x63/0x1a0\n [ 8308.623451] entry_SYSCALL_64_after_hwframe+0x42/0xb7\n \n The fix here is to make asyc_cow-\u003elocked_page NULL everywhere but the\n one async_cow struct that\u0027s allowed to do things to the locked page.\n \n Signed-off-by: Chris Mason \u003cclm@fb.com\u003e\n Fixes: 771ed689d2cd (\"Btrfs: Optimize compressed writeback and reads\")\n Reviewed-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit e1353c578b59f5d90d786d34c86e8fa6681430ed\nAuthor: Chris Mason \u003cclm@fb.com\u003e\nDate: Wed Jul 10 12:28:15 2019 -0700\n\n Btrfs: delete the entire async bio submission framework\n \n Now that we\u0027re not using btrfs_schedule_bio() anymore, delete all the\n code that supported it.\n \n Reviewed-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Chris Mason \u003cclm@fb.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 8a33f02c3fb602decf3f28b3c366d9375114a6ff\nAuthor: Chris Mason \u003cclm@fb.com\u003e\nDate: Wed Jul 10 12:28:14 2019 -0700\n\n Btrfs: stop using btrfs_schedule_bio()\n \n btrfs_schedule_bio() hands IO off to a helper thread to do the actual\n submit_bio() call. This has been used to make sure async crc and\n compression helpers don\u0027t get stuck on IO submission. To maintain good\n performance, over time the IO submission threads duplicated some IO\n scheduler characteristics such as high and low priority IOs and they\n also made some ugly assumptions about request allocation batch sizes.\n \n All of this cost at least one extra context switch during IO submission,\n and doesn\u0027t fit well with the modern blkmq IO stack. So, this commit stops\n using btrfs_schedule_bio(). We may need to adjust the number of async\n helper threads for crcs and compression, but long term it\u0027s a better\n path.\n \n Reviewed-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Chris Mason \u003cclm@fb.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit de37970b3636bfef01bd96529b851babc24517a9\nAuthor: Qu Wenruo \u003cwqu@suse.com\u003e\nDate: Thu Aug 22 15:25:00 2019 +0800\n\n btrfs: transaction: Cleanup unused TRANS_STATE_BLOCKED\n \n The state is introduced in commit 4a9d8bdee368 (\"Btrfs: make the state\n of the transaction more readable\"), then in commit 302167c50b32\n (\"btrfs: don\u0027t end the transaction for delayed refs in throttle\") the\n state is completely removed.\n \n So we can just clean up the state since it\u0027s only compared but never\n set.\n \n Signed-off-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 680407965a133d060dbad6acd27963c8c590957d\nAuthor: Qu Wenruo \u003cwqu@suse.com\u003e\nDate: Thu Aug 22 15:24:59 2019 +0800\n\n btrfs: transaction: Add comment for btrfs transaction lifespan\n \n Just a overview of the basic btrfs transaction lifespan, including the\n following states:\n - No transaction states\n - Transaction N [[TRANS_STATE_RUNNING]]\n - Transaction N [[TRANS_STATE_COMMIT_START]]\n - Transaction N [[TRANS_STATE_COMMIT_DOING]]\n - Transaction N [[TRANS_STATE_UNBLOCKED]]\n - Transaction N [[TRANS_STATE_COMPLETED]]\n \n For each states, the comment will include:\n - Basic explaination about current state\n - How to go next stage\n - What will happen if we call variant start_transaction() functions\n - Relationship to transaction N+1\n \n This doesn\u0027t provide much tech details, but just a cheat sheet for\n reader to get into the code a little easier.\n \n Signed-off-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 03570b8126bc8384b92a0e2890654d17686f1cca\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 22 15:19:04 2019 -0400\n\n btrfs: add enospc debug messages for ticket failure\n \n When debugging weird enospc problems it\u0027s handy to be able to dump the\n space info when we wake up all tickets, and see what the ticket values\n are. This helped me figure out cases where we were enospc\u0027ing when we\n shouldn\u0027t have been.\n \n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit ea44a14e86087ee48376ad569258dc9807310998\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 22 15:19:03 2019 -0400\n\n btrfs: do not account global reserve in can_overcommit\n \n We ran into a problem in production where a box with plenty of space was\n getting wedged doing ENOSPC flushing. These boxes only had 20% of the\n disk allocated, but their metadata space + global reserve was right at\n the size of their metadata chunk.\n \n In this case can_overcommit should be allowing allocations without\n problem, but there\u0027s logic in can_overcommit that doesn\u0027t allow us to\n overcommit if there\u0027s not enough real space to satisfy the global\n reserve.\n \n This is for historical reasons. Before there were only certain places\n we could allocate chunks. We could go to commit the transaction and not\n have enough space for our pending delayed refs and such and be unable to\n allocate a new chunk. This would result in a abort because of ENOSPC.\n This code was added to solve this problem.\n \n However since then we\u0027ve gained the ability to always be able to\n allocate a chunk. So we can easily overcommit in these cases without\n risking a transaction abort because of ENOSPC.\n \n Also prior to now the global reserve really would be used because that\u0027s\n the space we relied on for delayed refs. With delayed refs being\n tracked separately we no longer have to worry about running out of\n delayed refs space while committing. We are much less likely to\n exhaust our global reserve space during transaction commit.\n \n Fix the can_overcommit code to simply see if our current usage + what we\n want is less than our current free space plus whatever slack space we\n have in the disk is. This solves the problem we were seeing in\n production and keeps us from flushing as aggressively as we approach our\n actual metadata size usage.\n \n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 14ca778ffa232c63b8d83a48a422ab8b9e661598\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 22 15:19:02 2019 -0400\n\n btrfs: use btrfs_try_granting_tickets in update_global_rsv\n \n We have some annoying xfstests tests that will create a very small fs,\n fill it up, delete it, and repeat to make sure everything works right.\n This trips btrfs up sometimes because we may commit a transaction to\n free space, but most of the free metadata space was being reserved by\n the global reserve. So we commit and update the global reserve, but the\n space is simply added to bytes_may_use directly, instead of trying to\n add it to existing tickets. This results in ENOSPC when we really did\n have space. Fix this by calling btrfs_try_granting_tickets once we add\n back our excess space to wake any pending tickets.\n \n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 002d0c8552ab2c05ef1136987a612c461e3135c5\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 22 15:19:01 2019 -0400\n\n btrfs: always reserve our entire size for the global reserve\n \n While messing with the overcommit logic I noticed that sometimes we\u0027d\n ENOSPC out when really we should have run out of space much earlier. It\n turns out it\u0027s because we\u0027ll only reserve up to the free amount left in\n the space info for the global reserve, but that doesn\u0027t make sense with\n overcommit because we could be well above our actual size. This results\n in the global reserve not carving out it\u0027s entire reservation, and thus\n not putting enough pressure on the rest of the infrastructure to do the\n right thing and ENOSPC out at a convenient time. Fix this by always\n taking our full reservation amount for the global reserve.\n \n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 3ae92b3782182d282a92573abe95c96d34ca6e73\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 22 15:19:00 2019 -0400\n\n btrfs: change the minimum global reserve size\n \n It made sense to have the global reserve set at 16M in the past, but\n since it is used less nowadays set the minimum size to the number of\n items we\u0027ll need to update the main trees we update during a transaction\n commit, plus some slop area so we can do unlinks if we need to.\n \n In practice this doesn\u0027t affect normal file systems, but for xfstests\n where we do things like fill up a fs and then rm * it can fall over in\n weird ways. This enables us for more sane behavior at extremely small\n file system sizes.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit d8c9054ce036903552956c7921ebdbcaedbc673f\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 22 15:11:02 2019 -0400\n\n btrfs: rename btrfs_space_info_add_old_bytes\n \n This name doesn\u0027t really fit with how the space reservation stuff works\n now, rename it to btrfs_space_info_free_bytes_may_use so it\u0027s clear what\n the function is doing.\n \n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit f10d2305e9acc586c74e6d66655534f352c0cd6c\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 22 15:11:01 2019 -0400\n\n btrfs: remove orig_bytes from reserve_ticket\n \n Now that we do not do partial filling of tickets simply remove\n orig_bytes, it is no longer needed.\n \n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 70b2ab4990ca5a1ee1b857bca2611a96cf7e630f\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 22 15:11:00 2019 -0400\n\n btrfs: fix may_commit_transaction to deal with no partial filling\n \n Now that we aren\u0027t partially filling tickets we may have some slack\n space left in the space_info. We need to account for this in\n may_commit_transaction, otherwise we may choose to not commit the\n transaction despite it actually having enough space to satisfy our\n ticket.\n \n Calculate the free space we have in the space_info, if any, and subtract\n this from the ticket we have and use that amount to determine if we will\n need to commit to reclaim enough space.\n \n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 5ab178e94efacc19b757d11f15ae21abe7bf00f7\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 22 15:10:59 2019 -0400\n\n btrfs: rework wake_all_tickets\n \n Now that we no longer partially fill tickets we need to rework\n wake_all_tickets to call btrfs_try_to_wakeup_tickets() in order to see\n if any subsequent tickets are able to be satisfied. If our tickets_id\n changes we know something happened and we can keep flushing.\n \n Also if we find a ticket that is smaller than the first ticket in our\n queue then we want to retry the flushing loop again in case\n may_commit_transaction() decides we could satisfy the ticket by\n committing the transaction.\n \n Rename this to maybe_fail_all_tickets() while we\u0027re at it, to better\n reflect what the function is actually doing.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 3c6fbd22afa83cb883553b1c4f0e07ac2d0c30c8\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 22 15:10:58 2019 -0400\n\n btrfs: refactor the ticket wakeup code\n \n Now that btrfs_space_info_add_old_bytes simply checks if we can make the\n reservation and updates bytes_may_use, there\u0027s no reason to have both\n helpers in place.\n \n Factor out the ticket wakeup logic into it\u0027s own helper, make\n btrfs_space_info_add_old_bytes() update bytes_may_use and then call the\n wakeup helper, and replace all calls to btrfs_space_info_add_new_bytes()\n with the wakeup helper.\n \n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 2b440fb2f677d8c581305c2fec3cf38dd546aee8\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 22 15:10:57 2019 -0400\n\n btrfs: rework btrfs_space_info_add_old_bytes\n \n If there are pending tickets and we are overcommitted we will simply\n return free\u0027d reservations to space_info-\u003ebytes_may_use if we cannot\n overcommit any more. This is problematic because we assume any free\n space would have been added to the tickets, and so if we go from an\n overcommitted state to not overcommitted we could have plenty of space\n in our space_info but be unable to make progress on our tickets because\n we only refill tickets from previous reservations.\n \n Consider the following example. We were allowed to overcommit to\n space_info-\u003etotal_bytes + 2mib. Now we\u0027ve allocated all of our chunks\n and are no longer allowed to overcommit those extra 2mib. Assume there\n is a 3mib reservation we are now freeing. Because we can no longer\n overcommit we do not partially refill the ticket with the 3mib, instead\n we subtract it from space_info-\u003ebytes_may_use. Now the total reserved\n space is 1mib less than total_bytes, meaning we have 1mib of space we\n could reserve. Now assume that our ticket is 2mib, and we only have\n 1mib of space to reclaim, so we have a partial refilling to 1mib. We\n keep trying to flush and eventually give up and ENOSPC the ticket, when\n there was the remaining 1mib left in the space_info for usage.\n \n Instead of doing this partial filling of tickets dance we need to simply\n add our space to the space_info, and then do the normal check to see if\n we can satisfy the whole reservation. If we can then we wake up the\n ticket and carry on. This solves the above problem and makes it much\n more straightforward to understand how the tickets are satisfied.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit e5a047c6fb50842486818a2cae7b46ca4a419d62\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 22 15:10:56 2019 -0400\n\n btrfs: add space reservation tracepoint for reserved bytes\n \n I noticed when folding the trace_btrfs_space_reservation() tracepoint\n into the btrfs_space_info_update_* helpers that we didn\u0027t emit a\n tracepoint when doing btrfs_add_reserved_bytes(). I know this is\n because we were swapping bytes_may_use for bytes_reserved, so in my mind\n there was no reason to have the tracepoint there. But now there is\n because we always emit the unreserve for the bytes_may_use side, and\n this would have broken if compression was on anyway. Add a tracepoint\n to cover the bytes_reserved counter so the math still comes out right.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 2dc70e4c48d71953dbe9f5dfc99023a4a71da746\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 22 15:10:55 2019 -0400\n\n btrfs: roll tracepoint into btrfs_space_info_update helper\n \n We duplicate this tracepoint everywhere we call these helpers, so update\n the helper to have the tracepoint as well.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 8d03977706c03ab320993d655ee2014d4fe125f6\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 22 15:10:54 2019 -0400\n\n btrfs: do not allow reservations if we have pending tickets\n \n If we already have tickets on the list we don\u0027t want to steal their\n reservations. This is a preparation patch for upcoming changes,\n technically this shouldn\u0027t happen today because of the way we add bytes\n to tickets before adding them to the space_info in most cases.\n \n This does not change the FIFO nature of reserve tickets, it simply\n allows us to enforce it in a different way. Previously it was enforced\n because any new space would be added to the first ticket on the list,\n which would result in new reservations getting a reserve ticket. This\n replaces that mechanism by simply checking to see if we have outstanding\n reserve tickets and skipping straight to adding a ticket for our\n reservation.\n \n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 592a7c5e082e237c7508f37b2c3494a731acad88\nAuthor: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\nDate: Fri Aug 9 17:56:53 2019 +0300\n\n drm/i915: Extend non readable mcr range\n \n Our current avoidance of non readable mcr range was not\n inclusive enough. Extend the start and end.\n \n References: HSDES#1405586840\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Signed-off-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Acked-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190809145653.2279-1-mika.kuoppala@linux.intel.com\n\ncommit 7a786b84790789eff5bad49e3f6c15f75b7bf691\nAuthor: Matthew Michilot \u003cmatthew.michilot@gmail.com\u003e\nDate: Thu Aug 15 17:40:54 2019 +0000\n\n USB: serial: ftdi_sio: add support for FT232H CBUS gpios\n \n Enable support for cbus gpios on FT232H. The cbus configuration is\n stored in two words in the EEPROM at byte-offset 0x1a with the mux\n config for ACBUS5, ACBUS6, ACBUS8 and ACBUS9 (only pins that can be\n configured as I/O mode).\n \n Tested using FT232H by configuring one ACBUS pin at a time.\n \n Reviewed-by: Tim Harvey \u003ctharvey@gateworks.com\u003e\n Signed-off-by: Matthew Michilot \u003cmatthew.michilot@gmail.com\u003e\n [ johan: fix copy-paste error in commit message ]\n Signed-off-by: Johan Hovold \u003cjohan@kernel.org\u003e\n\ncommit 3724e186fead350d1446d5202cd92fa6250bffda\nAuthor: Joakim Zhang \u003cqiangqing.zhang@nxp.com\u003e\nDate: Wed Aug 28 12:07:56 2019 +0000\n\n docs/perf: Add documentation for the i.MX8 DDR PMU\n \n Add some documentation describing the DDR PMU residing in the Freescale\n i.MDX SoC and its perf driver implementation in Linux.\n \n Signed-off-by: Joakim Zhang \u003cqiangqing.zhang@nxp.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 7ce2e76a0420801fb4b53b9e6850940e6b326433\nAuthor: Krzysztof Wilczynski \u003ckw@linux.com\u003e\nDate: Tue Aug 27 11:56:20 2019 +0200\n\n PCI: Move ASPM declarations to linux/pci.h\n \n Move ASPM definitions and function prototypes from include/linux/pci-aspm.h\n to include/linux/pci.h so users only need to include \u003clinux/pci.h\u003e:\n \n PCIE_LINK_STATE_L0S\n PCIE_LINK_STATE_L1\n PCIE_LINK_STATE_CLKPM\n pci_disable_link_state()\n pci_disable_link_state_locked()\n pcie_no_aspm()\n \n No functional changes intended.\n \n Link: https://lore.kernel.org/r/20190827095620.11213-1-kw@linux.com\n Signed-off-by: Krzysztof Wilczynski \u003ckw@linux.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit 0a56e0603fa13af08816d673f6f71b68cda2fb2e\nAuthor: Naveen N. Rao \u003cnaveen.n.rao@linux.vnet.ibm.com\u003e\nDate: Tue Aug 27 12:44:58 2019 +0530\n\n perf arch powerpc: Sync powerpc syscall.tbl\n \n Copy over powerpc syscall.tbl to grab changes from the below commits:\n \n commit cee3536d24a1 (\"powerpc: Wire up clone3 syscall\")\n commit 1a271a68e030 (\"arch: mark syscall number 435 reserved for clone3\")\n commit 7615d9e1780e (\"arch: wire-up pidfd_open()\")\n commit d8076bdb56af (\"uapi: Wire up the mount API syscalls on non-x86 arches [ver #2]\")\n commit 39036cd27273 (\"arch: add pidfd and io_uring syscalls everywhere\")\n commit 48166e6ea47d (\"y2038: add 64-bit time_t syscalls to all 32-bit architectures\")\n commit d33c577cccd0 (\"y2038: rename old time and utime syscalls\")\n commit 00bf25d693e7 (\"y2038: use time32 syscall names on 32-bit\")\n commit 8dabe7245bbc (\"y2038: syscalls: rename y2038 compat syscalls\")\n commit 0d6040d46817 (\"arch: add split IPC system calls where needed\")\n \n Reported-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Naveen N. Rao \u003cnaveen.n.rao@linux.vnet.ibm.com\u003e\n Cc: Ravi Bangoria \u003cravi.bangoria@linux.ibm.com\u003e\n Cc: linuxppc-dev@lists.ozlabs.org\n Link: http://lkml.kernel.org/r/20190827071458.19897-1-naveen.n.rao@linux.vnet.ibm.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit f2902a2fb40c589b886d21518ef8a1ee87f76b0c\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Wed Aug 14 15:22:30 2019 +0200\n\n powerpc: use the generic dma coherent remap allocator\n \n This switches to using common code for the DMA allocations, including\n potential use of the CMA allocator if configured.\n \n Switching to the generic code enables DMA allocations from atomic\n context, which is required by the DMA API documentation, and also\n adds various other minor features drivers start relying upon. It\n also makes sure we have on tested code base for all architectures\n that require uncached pte bits for coherent DMA allocations.\n \n Another advantage is that consistent memory allocations now share\n the general vmalloc pool instead of needing an explicit careout\n from it.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Tested-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e # tested on 8xx\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190814132230.31874-2-hch@lst.de\n\ncommit 555e28179d37e431e97d78d106fc917bec2c6f93\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Tue Aug 27 13:30:07 2019 +1000\n\n powerpc/64: remove support for kernel-mode syscalls\n \n There is support for the kernel to execute the \u0027sc 0\u0027 instruction and\n make a system call to itself. This is a relic that is unused in the\n tree, therefore untested. It\u0027s also highly questionable for modules to\n be doing this.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190827033010.28090-3-npiggin@gmail.com\n\ncommit facd04a904ff6cdc6ee85d6e85d500f478a1bec4\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Tue Aug 27 13:30:06 2019 +1000\n\n powerpc: convert to copy_thread_tls\n \n Commit 3033f14ab78c3 (\"clone: support passing tls argument via C rather\n than pt_regs magic\") introduced the HAVE_COPY_THREAD_TLS option. Use it\n to avoid a subtle assumption about the argument ordering of clone type\n syscalls.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190827033010.28090-2-npiggin@gmail.com\n\ncommit c7bf1252d5b3891b4ab7072b240a8422fb9da793\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Mon Aug 26 15:52:19 2019 +0000\n\n powerpc/32: don\u0027t use CPU_FTR_COHERENT_ICACHE\n \n Only 601 and E200 have CPU_FTR_COHERENT_ICACHE.\n \n Just use #ifdefs instead of feature fixup.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/5f3e92ccd64d06477b27626f6007a9da3b8da157.1566834712.git.christophe.leroy@c-s.fr\n\ncommit e0291f1decd6e8d447067f7d2cf01b1091b7cb3f\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Mon Aug 26 15:52:18 2019 +0000\n\n powerpc/32: drop CPU_FTR_UNIFIED_ID_CACHE\n \n Only 601 and e200 have unified I/D cache.\n \n Drop the feature and use CONFIG_PPC_BOOK3S_601 and CONFIG_E200.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/b5902144266d2f4eed1ffea53915bd0245841e02.1566834712.git.christophe.leroy@c-s.fr\n\ncommit 39097b9c6d762d3fcd6f753e05ee3e34ec250ff3\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Mon Aug 26 15:52:17 2019 +0000\n\n powerpc/32s: use CONFIG_PPC_BOOK3S_601 instead of reading PVR\n \n Use CONFIG_PPC_BOOK3S_601 instead of reading PVR to know if\n it is a 601 or not.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/909c26db9facd7fe454695b303f952e019dd9eda.1566834712.git.christophe.leroy@c-s.fr\n\ncommit 88fb309409ab454b497a6abb0f931ce3b6d9969c\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Mon Aug 26 15:52:16 2019 +0000\n\n powerpc/32s: drop CPU_FTR_USE_RTC feature\n \n CPU_FTR_USE_RTC feature only applies to powerpc601.\n \n Drop this feature and replace it with tests on CONFIG_PPC_BOOK3S_601.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/170411e2360861f4a95c21faad43519a08bc4040.1566834712.git.christophe.leroy@c-s.fr\n\ncommit 12c3f1fd87bf4e55f06d079a45d6f15e2f6f9750\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Mon Aug 26 15:52:14 2019 +0000\n\n powerpc/32s: get rid of CPU_FTR_601 feature\n \n Now that 601 is exclusive from other 6xx, CPU_FTR_601 and\n associated fixups are useless.\n \n Drop this feature and use #ifdefs instead.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/ecdb7194a17dbfa01865df6a82979533adc2c70b.1566834712.git.christophe.leroy@c-s.fr\n\ncommit f7a0bf7d904e902e13024986b7b357181ee30849\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Mon Aug 26 15:52:13 2019 +0000\n\n powerpc/32s: add an option to exclusively select powerpc 601\n \n Powerpc 601 is rather old powerpc which as some important\n limitations compared to other book3s/32 powerpcs:\n - No Timebase.\n - Common BATs for instruction and data.\n - No execution protection in segment registers.\n - No RI bit in MSR\n - ...\n \n It is starting to be difficult and cumbersome to maintain\n kernels that are compatible both with 601 and other 6xx cores.\n \n Create a compiletime option to exclusively select either powerpc 601\n or other 6xx.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/d644eaf7dff8cc149260066802af230bdf34fded.1566834712.git.christophe.leroy@c-s.fr\n\ncommit c12c0288e35a4693b1606e229dab54a62f1ad568\nAuthor: Joakim Zhang \u003cqiangqing.zhang@nxp.com\u003e\nDate: Wed Aug 28 12:07:52 2019 +0000\n\n perf/imx_ddr: Add support for AXI ID filtering\n \n AXI filtering is used by events 0x41 and 0x42 to count reads or writes\n with an ARID or AWID matching a specified filter. The filter is exposed\n to userspace as an (ID, MASK) pair, where each set bit in the mask\n causes the corresponding bit in the ID to be ignored when matching\n against the ID of memory transactions for the purposes of incrementing\n the counter.\n \n For example:\n \n # perf stat -a -e imx8_ddr0/axid-read,axi_mask\u003d0xff,axi_id\u003d0x800/ cmd\n \n will count all read transactions from AXI IDs 0x800 - 0x8ff. If the\n \u0027axi_mask\u0027 is omitted, then it is treated as 0x0 which means that the\n \u0027axi_id\u0027 will be matched exactly.\n \n Signed-off-by: Joakim Zhang \u003cqiangqing.zhang@nxp.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 3bd158c56a56e8767e569d7fbc66efbedc478077\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Sun Aug 4 02:38:52 2019 +0200\n\n spi: bcm2835: Convert to use CS GPIO descriptors\n \n This converts the BCM2835 SPI master driver to use GPIO\n descriptors for chip select handling.\n \n The BCM2835 driver was relying on the core to drive the\n CS high/low so very small changes were needed for this\n part. If it managed to request the CS from the device tree\n node, all is pretty straight forward.\n \n However for native GPIOs this driver has a quite unorthodox\n loopback to request some GPIOs from the SoC GPIO chip by\n looking it up from the device tree using gpiochip_find()\n and then offseting hard into its numberspace. This has\n been augmented a bit by using gpiochip_request_own_desc()\n but this code really needs to be verified. If \"native CS\"\n is actually an SoC GPIO, why is it even done this way?\n Should this GPIO not just be defined in the device tree\n like any other CS GPIO? I\u0027m confused.\n \n Cc: Lukas Wunner \u003clukas@wunner.de\u003e\n Cc: Stefan Wahren \u003cstefan.wahren@i2se.com\u003e\n Cc: Martin Sperl \u003ckernel@martin.sperl.org\u003e\n Cc: Chris Boot \u003cbootc@bootc.net\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190804003852.1312-1-linus.walleij@linaro.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 0f0581b24bd019dfe32878e4c1bde266c7364e02\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Sun Aug 4 02:35:39 2019 +0200\n\n spi: fsl: Convert to use CS GPIO descriptors\n \n This converts the Freescale SPI master driver to use GPIO\n descriptors for chip select handling.\n \n The Freescale (fsl) driver has a lot of quirks to look up\n \"gpios\" rather than \"cs-gpios\" from the device tree.\n After the prior patch that will make gpiolib return the\n GPIO descriptor for \"gpios\" in response to a request for\n \"cs-gpios\", this code can be cut down quite a bit.\n \n The driver has custom handling of chip select rather\n than using the core (which may be possible but not\n done in this patch) so it still needs to refer directly\n to spi-\u003ecs_gpiod to set the chip select.\n \n Cc: Shawn Guo \u003cshawnguo@kernel.org\u003e\n Cc: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\n Cc: Pengutronix Kernel Team \u003ckernel@pengutronix.de\u003e\n Cc: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Cc: NXP Linux Team \u003clinux-imx@nxp.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190804003539.985-1-linus.walleij@linaro.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit e8f6b4952ec54a9d7e43f908d39dc168b0310599\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Aug 27 13:06:15 2019 +0100\n\n drm/i915/execlists: Flush the post-sync breadcrumb write harder\n \n Quite rarely we see that the CS completion event fires before the\n breadcrumb is coherent, which presumably is a result of the CS_STALL not\n waiting for the post-sync operation. Try throwing in a DC_FLUSH into\n the following pipecontrol to see if that makes any difference.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Acked-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190827120615.31390-1-chris@chris-wilson.co.uk\n\ncommit 6b8249abb093551ef173d13a25ed0044d5dd33e0\nAuthor: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\nDate: Fri Aug 23 10:38:35 2019 +0100\n\n drivers: thermal: qcom: tsens: Fix memory leak from qfprom read\n \n memory returned as part of nvmem_read via qfprom_read should be\n freed by the consumer once done.\n Existing code is not doing it so fix it.\n \n Below memory leak detected by kmemleak\n [\u003cffffff80088b7658\u003e] kmemleak_alloc+0x50/0x84\n [\u003cffffff80081df120\u003e] __kmalloc+0xe8/0x168\n [\u003cffffff80086db350\u003e] nvmem_cell_read+0x30/0x80\n [\u003cffffff8008632790\u003e] qfprom_read+0x4c/0x7c\n [\u003cffffff80086335a4\u003e] calibrate_v1+0x34/0x204\n [\u003cffffff8008632518\u003e] tsens_probe+0x164/0x258\n [\u003cffffff80084e0a1c\u003e] platform_drv_probe+0x80/0xa0\n [\u003cffffff80084de4f4\u003e] really_probe+0x208/0x248\n [\u003cffffff80084de2c4\u003e] driver_probe_device+0x98/0xc0\n [\u003cffffff80084dec54\u003e] __device_attach_driver+0x9c/0xac\n [\u003cffffff80084dca74\u003e] bus_for_each_drv+0x60/0x8c\n [\u003cffffff80084de634\u003e] __device_attach+0x8c/0x100\n [\u003cffffff80084de6c8\u003e] device_initial_probe+0x20/0x28\n [\u003cffffff80084dcbb8\u003e] bus_probe_device+0x34/0x7c\n [\u003cffffff80084deb08\u003e] deferred_probe_work_func+0x6c/0x98\n [\u003cffffff80080c3da8\u003e] process_one_work+0x160/0x2f8\n \n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n Acked-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Signed-off-by: Zhang Rui \u003crui.zhang@intel.com\u003e\n\ncommit 743dac494d61d991967ebcfab92e4f80dc7583b3\nAuthor: Neil Horman \u003cnhorman@tuxdriver.com\u003e\nDate: Thu Aug 22 10:34:21 2019 -0400\n\n x86/apic/vector: Warn when vector space exhaustion breaks affinity\n \n On x86, CPUs are limited in the number of interrupts they can have affined\n to them as they only support 256 interrupt vectors per CPU. 32 vectors are\n reserved for the CPU and the kernel reserves another 22 for internal\n purposes. That leaves 202 vectors for assignement to devices.\n \n When an interrupt is set up or the affinity is changed by the kernel or the\n administrator, the vector assignment code attempts to honor the requested\n affinity mask. If the vector space on the CPUs in that affinity mask is\n exhausted the code falls back to a wider set of CPUs and assigns a vector\n on a CPU outside of the requested affinity mask silently.\n \n While the effective affinity is reflected in the corresponding\n /proc/irq/$N/effective_affinity* files the silent breakage of the requested\n affinity can lead to unexpected behaviour for administrators.\n \n Add a pr_warn() when this happens so that adminstrators get at least\n informed about it in the syslog.\n \n [ tglx: Massaged changelog and made the pr_warn() more informative ]\n \n Reported-by: djuran@redhat.com\n Signed-off-by: Neil Horman \u003cnhorman@tuxdriver.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Tested-by: djuran@redhat.com\n Link: https://lkml.kernel.org/r/20190822143421.9535-1-nhorman@tuxdriver.com\n\ncommit f32c7a8e45105bd0af76872bf6eef0438ff12fb2\nAuthor: Mark Rutland \u003cmark.rutland@arm.com\u003e\nDate: Tue Aug 27 18:12:57 2019 +0100\n\n arm64: kpti: ensure patched kernel text is fetched from PoU\n \n While the MMUs is disabled, I-cache speculation can result in\n instructions being fetched from the PoC. During boot we may patch\n instructions (e.g. for alternatives and jump labels), and these may be\n dirty at the PoU (and stale at the PoC).\n \n Thus, while the MMU is disabled in the KPTI pagetable fixup code we may\n load stale instructions into the I-cache, potentially leading to\n subsequent crashes when executing regions of code which have been\n modified at runtime.\n \n Similarly to commit:\n \n 8ec41987436d566f (\"arm64: mm: ensure patched kernel text is fetched from PoU\")\n \n ... we can invalidate the I-cache after enabling the MMU to prevent such\n issues.\n \n The KPTI pagetable fixup code itself should be clean to the PoC per the\n boot protocol, so no maintenance is required for this code.\n \n Signed-off-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Reviewed-by: James Morse \u003cjames.morse@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 1b90191826cbf2c8368c94c967e7fb775865802b\nAuthor: Omar Sandoval \u003cosandov@fb.com\u003e\nDate: Thu Aug 15 14:04:04 2019 -0700\n\n btrfs: stop clearing EXTENT_DIRTY in inode I/O tree\n \n Since commit fee187d9d9dd (\"Btrfs: do not set EXTENT_DIRTY along with\n EXTENT_DELALLOC\"), we never set EXTENT_DIRTY in inode-\u003eio_tree, so we\n can simplify and stop trying to clear it.\n \n Reviewed-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: Omar Sandoval \u003cosandov@fb.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 51da894c5e4ed5b22ce548b3af94bb8a5fefb32d\nAuthor: Omar Sandoval \u003cosandov@fb.com\u003e\nDate: Thu Aug 15 14:04:03 2019 -0700\n\n btrfs: treat RWF_{,D}SYNC writes as sync for CRCs\n \n The VFS indicates a synchronous write to -\u003ewrite_iter() via\n iocb-\u003eki_flags. The IOCB_{,D}SYNC flags may be set based on the file\n (see iocb_flags()) or the RWF_* flags passed to a syscall like\n pwritev2() (see kiocb_set_rw_flags()).\n \n However, in btrfs_file_write_iter(), we\u0027re checking if a write is\n synchronous based only on the file; we use this to decide when to bump\n the sync_writers counter and thus do CRCs synchronously. Make sure we do\n this for all synchronous writes as determined by the VFS.\n \n Reviewed-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: Omar Sandoval \u003cosandov@fb.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n [ add const ]\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 21f1bb3873bec4a8010c481f3637772ec6794674\nAuthor: Omar Sandoval \u003cosandov@fb.com\u003e\nDate: Thu Aug 15 14:04:02 2019 -0700\n\n btrfs: use correct count in btrfs_file_write_iter()\n \n generic_write_checks() may modify iov_iter_count(), so we must get the\n count after the call, not before. Using the wrong one has a couple of\n consequences:\n \n 1. We check a longer range in check_can_nocow() for nowait than we\u0027re\n actually writing.\n 2. We create extra hole extent maps in btrfs_cont_expand(). As far as I\n can tell, this is harmless, but I might be missing something.\n \n These issues are pretty minor, but let\u0027s fix it before something more\n important trips on it.\n \n Fixes: edf064e7c6fe (\"btrfs: nowait aio support\")\n Reviewed-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: Omar Sandoval \u003cosandov@fb.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit f7b15c74cffd760ec9959078982d8268a38456c4\nAuthor: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nDate: Wed Aug 28 10:03:53 2019 +0200\n\n input/vmmouse: Update the backdoor call with support for new instructions\n \n Use the definition provided by include/asm/vmware.h.\n \n Signed-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Reviewed-by: Doug Covelli \u003cdcovelli@vmware.com\u003e\n Acked-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n Cc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: linux-input@vger.kernel.org\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: VMware Graphics \u003clinux-graphics-maintainer@vmware.com\u003e\n Cc: \u003cpv-drivers@vmware.com\u003e\n Cc: x86-ml \u003cx86@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190828080353.12658-5-thomas_os@shipmail.org\n\ncommit 6abe3778cf5abd59b23b9037796f3eab8b7f1d98\nAuthor: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nDate: Wed Aug 28 10:03:52 2019 +0200\n\n drm/vmwgfx: Update the backdoor call with support for new instructions\n \n Use the definition provided by include/asm/vmware.h\n \n Signed-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Reviewed-by: Doug Covelli \u003cdcovelli@vmware.com\u003e\n Acked-by: Dave Airlie \u003cairlied@redhat.com\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: pv-drivers@vmware.com\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: VMware Graphics \u003clinux-graphics-maintainer@vmware.com\u003e\n Cc: x86-ml \u003cx86@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190828080353.12658-4-thomas_os@shipmail.org\n\ncommit b4dd4f6e3648dfd66576515f2222d885a9a765c0\nAuthor: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nDate: Wed Aug 28 10:03:51 2019 +0200\n\n x86/vmware: Add a header file for hypercall definitions\n \n The new header is intended to be used by drivers using the backdoor.\n Follow the KVM example using alternatives self-patching to choose\n between vmcall, vmmcall and io instructions.\n \n Also define two new CPU feature flags to indicate hypervisor support\n for vmcall- and vmmcall instructions. The new XF86_FEATURE_VMW_VMMCALL\n flag is needed because using XF86_FEATURE_VMMCALL might break QEMU/KVM\n setups using the vmmouse driver. They rely on XF86_FEATURE_VMMCALL\n on AMD to get the kvm_hypercall() right. But they do not yet implement\n vmmcall for the VMware hypercall used by the vmmouse driver.\n \n [ bp: reflow hypercall %edx usage explanation comment. ]\n \n Signed-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Reviewed-by: Doug Covelli \u003cdcovelli@vmware.com\u003e\n Cc: Aaron Lewis \u003caaronlewis@google.com\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: Fenghua Yu \u003cfenghua.yu@intel.com\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Cc: linux-graphics-maintainer@vmware.com\n Cc: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Cc: Nicolas Ferre \u003cnicolas.ferre@microchip.com\u003e\n Cc: Robert Hoo \u003crobert.hu@linux.intel.com\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: virtualization@lists.linux-foundation.org\n Cc: \u003cpv-drivers@vmware.com\u003e\n Cc: x86-ml \u003cx86@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190828080353.12658-3-thomas_os@shipmail.org\n\ncommit b333b0ba2346f2071390533b8f37f36f191d705d\nAuthor: Mark Rutland \u003cmark.rutland@arm.com\u003e\nDate: Tue Aug 27 16:57:08 2019 +0100\n\n arm64: fix fixmap copy for 16K pages and 48-bit VA\n \n With 16K pages and 48-bit VAs, the PGD level of table has two entries,\n and so the fixmap shares a PGD with the kernel image. Since commit:\n \n f9040773b7bbbd9e (\"arm64: move kernel image to base of vmalloc area\")\n \n ... we copy the existing fixmap to the new fine-grained page tables at\n the PUD level in this case. When walking to the new PUD, we forgot to\n offset the PGD entry and always used the PGD entry at index 0, but this\n worked as the kernel image and fixmap were in the low half of the TTBR1\n address space.\n \n As of commit:\n \n 14c127c957c1c607 (\"arm64: mm: Flip kernel VA space\")\n \n ... the kernel image and fixmap are in the high half of the TTBR1\n address space, and hence use the PGD at index 1, but we didn\u0027t update\n the fixmap copying code to account for this.\n \n Thus, we\u0027ll erroneously try to copy the fixmap slots into a PUD under\n the PGD entry at index 0. At the point we do so this PGD entry has not\n been initialised, and thus we\u0027ll try to write a value to a small offset\n from physical address 0, causing a number of potential problems.\n \n Fix this be correctly offsetting the PGD. This is split over a few steps\n for legibility.\n \n Fixes: 14c127c957c1c607 (\"arm64: mm: Flip kernel VA space\")\n Reported-by: Anshuman Khandual \u003canshuman.khandual@arm.com\u003e\n Cc: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Cc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Acked-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n Tested-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n Acked-by: Steve Capper \u003cSteve.Capper@arm.com\u003e\n Tested-by: Steve Capper \u003cSteve.Capper@arm.com\u003e\n Tested-by: Anshuman Khandual \u003canshuman.khandual@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 71fed982d63cb2bb88db6f36059e3b14a7913846\nAuthor: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\nDate: Fri Aug 23 13:38:45 2019 +0200\n\n tick: Mark sched_timer to expire in hard interrupt context\n \n sched_timer must be initialized with the _HARD mode suffix to ensure expiry\n in hard interrupt context on RT.\n \n The previous conversion to HARD expiry mode missed on one instance in\n tick_nohz_switch_to_nohz(). Fix it up.\n \n Fixes: 902a9f9c50905 (\"tick: Mark tick related hrtimers to expiry in hard interrupt context\")\n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190823113845.12125-3-bigeasy@linutronix.de\n\ncommit a67e408241783575716fcf3f79d0878f6cef0273\nAuthor: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\nDate: Fri Aug 23 13:38:44 2019 +0200\n\n hrtimer: Add kernel doc annotation for HRTIMER_MODE_HARD\n \n Add kernel doc annotation for HRTIMER_MODE_HARD.\n \n Fixes: ae6683d815895 (\"hrtimer: Introduce HARD expiry mode\")\n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190823113845.12125-2-bigeasy@linutronix.de\n\ncommit a8a652bfac7f4a60dfbe07a8c95c65506601b2e1\nAuthor: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\nDate: Thu Aug 22 10:56:51 2019 +0100\n\n ASoC: wcd9335: Fix primary interpolator max rate\n \n On this codec SLIMBus RX path supports 384000 rate on primary interpolator.\n Add this missing rate as supported rate.\n \n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190822095653.7200-3-srinivas.kandagatla@linaro.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 438b6c20e6161a1a7542490baa093c86732f77d6\nAuthor: David Jander \u003cdavid@protonic.nl\u003e\nDate: Tue Aug 27 06:46:29 2019 +0000\n\n gpio: pca953x: use pca953x_read_regs instead of regmap_bulk_read\n \n The register number needs to be translated for chips with more than 8\n ports. This patch fixes a bug causing all chips with more than 8 GPIO pins\n to not work correctly.\n \n Fixes: 0f25fda840a9 (\"gpio: pca953x: Zap ad-hoc reg_direction cache\")\n Cc: Cc: \u003cstable@vger.kernel.org\u003e\n Signed-off-by: David Jander \u003cdavid@protonic.nl\u003e\n Signed-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n\ncommit bc624a06f0c5190bc37fec7d22cd82b43a579698\nAuthor: David Jander \u003cdavid@protonic.nl\u003e\nDate: Tue Aug 27 06:46:28 2019 +0000\n\n gpio: pca953x: correct type of reg_direction\n \n The type of reg_direction needs to match the type of the regmap, which\n is u8.\n \n Fixes: 0f25fda840a9 (\"gpio: pca953x: Zap ad-hoc reg_direction cache\")\n Cc: Cc: \u003cstable@vger.kernel.org\u003e\n Signed-off-by: David Jander \u003cdavid@protonic.nl\u003e\n Signed-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n\ncommit 4cc4531c310e592cf624148ae59c64f930f12e39\nAuthor: Vidyakumar Athota \u003cvathota@codeaurora.org\u003e\nDate: Thu Aug 22 10:56:50 2019 +0100\n\n ALSA: pcm: add support for 352.8KHz and 384KHz sample rate\n \n Most of the modern codecs supports 352.8KHz and 384KHz sample rates.\n Currenlty HW params fails to set 352.8Kz and 384KHz sample rate\n as these are not in known rates list.\n Add these new rates to known list to allow them.\n \n This patch also adds defines in pcm.h so that drivers can use it.\n \n Signed-off-by: Vidyakumar Athota \u003cvathota@codeaurora.org\u003e\n Signed-off-by: Banajit Goswami \u003cbgoswami@codeaurora.org\u003e\n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190822095653.7200-2-srinivas.kandagatla@linaro.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit f67ff1bd58f0c102f0194c3888ddbc4a87dd1382\nAuthor: Hsin-Hsiung Wang \u003chsin-hsiung.wang@mediatek.com\u003e\nDate: Fri Aug 23 11:45:28 2019 +0800\n\n regulator: mt6358: Add support for MT6358 regulator\n \n The MT6358 is a regulator found on boards based on MediaTek MT8183 and\n probably other SoCs. It is a so called pmic and connects as a slave to\n SoC using SPI, wrapped inside the pmic-wrapper.\n \n Signed-off-by: Hsin-Hsiung Wang \u003chsin-hsiung.wang@mediatek.com\u003e\n Link: https://lore.kernel.org/r/1566531931-9772-8-git-send-email-hsin-hsiung.wang@mediatek.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit fa00eb4eb2361ffc594bd9c6d78c585e7112c5dc\nAuthor: Hsin-Hsiung Wang \u003chsin-hsiung.wang@mediatek.com\u003e\nDate: Fri Aug 23 11:45:26 2019 +0800\n\n regulator: Add document for MT6358 regulator\n \n add dt-binding document for MediaTek MT6358 PMIC\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Hsin-Hsiung Wang \u003chsin-hsiung.wang@mediatek.com\u003e\n Link: https://lore.kernel.org/r/1566531931-9772-6-git-send-email-hsin-hsiung.wang@mediatek.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 2c650fd5a53beace2e7507f478562f0dd1885b89\nAuthor: Jiaxin Yu \u003cjiaxin.yu@mediatek.com\u003e\nDate: Sat Aug 24 12:37:25 2019 +0800\n\n ASoC: mediatek: mt8183: fix tdm out data is valid on rising edge\n \n This patch correct tdm out bck inverse register to AUDIO_TOP_CON3[3].\n \n Signed-off-by: Jiaxin Yu \u003cjiaxin.yu@mediatek.com\u003e\n Link: https://lore.kernel.org/r/1566621445-26989-4-git-send-email-jiaxin.yu@mediatek.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 8e58c521bcb823bfd6e81f4a85b76afdce098b1d\nAuthor: Jiaxin Yu \u003cjiaxin.yu@mediatek.com\u003e\nDate: Sat Aug 24 12:37:24 2019 +0800\n\n ASoC: mediatek: mt8183: tdm hw support tdm out and 8ch i2s out\n \n This patch refined tdm driver code, and allow tdm hw to support two\n configurations in machine driver to output tdm signal or i2s signal.\n \n Signed-off-by: Jiaxin Yu \u003cjiaxin.yu@mediatek.com\u003e\n Link: https://lore.kernel.org/r/1566621445-26989-3-git-send-email-jiaxin.yu@mediatek.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 41cfe2a2a7f4fad5647031ad3a1da166452b5437\nAuthor: Tianyu Lan \u003cTianyu.Lan@microsoft.com\u003e\nDate: Wed Aug 28 16:07:47 2019 +0800\n\n x86/hyperv: Hide pv_ops access for CONFIG_PARAVIRT\u003dn\n \n hv_setup_sched_clock() references pv_ops which is only available when\n CONFIG_PARAVIRT\u003dY.\n \n Wrap it into a #ifdef\n \n Signed-off-by: Tianyu Lan \u003cTianyu.Lan@microsoft.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190828080747.204419-1-Tianyu.Lan@microsoft.com\n\ncommit 101f85b56d03b36418bbf867f67d81710839b0ec\nAuthor: Ming Lei \u003cming.lei@redhat.com\u003e\nDate: Wed Aug 28 16:58:15 2019 +0800\n\n genirq/affinity: Remove const qualifier from node_to_cpumask argument\n \n When CONFIG_CPUMASK_OFFSTACK isn\u0027t enabled, \u0027cpumask_var_t\u0027 is as\n \n \u0027typedef struct cpumask cpumask_var_t[1]\u0027,\n \n so the argument \u0027node_to_cpumask\u0027 alloc_nodes_vectors() can\u0027t be declared\n as \u0027const cpumask_var_t *\u0027\n \n Fixes the following warning:\n \n kernel/irq/affinity.c: In function \u0027__irq_build_affinity_masks\u0027:\n alloc_nodes_vectors(numvecs, node_to_cpumask, cpu_mask,\n ^\n kernel/irq/affinity.c:128:13: note: expected \u0027const struct cpumask (*)[1]\u0027 but argument is of type \u0027struct cpumask (*)[1]\u0027\n static void alloc_nodes_vectors(unsigned int numvecs,\n ^\n Fixes: b1a5a73e64e9 (\"genirq/affinity: Spread vectors on node according to nr_cpu ratio\")\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190828085815.19931-1-ming.lei@redhat.com\n\ncommit cca3854010c5aebf3a06be0de04f26c4f1ae5810\nAuthor: Thinh Nguyen \u003cThinh.Nguyen@synopsys.com\u003e\nDate: Mon Aug 19 18:36:12 2019 -0700\n\n usb: gadget: composite: Set recommended BESL values\n \n Set the recommended BESL deep and baseline values based on the gadget\u0027s\n configuration parameters to the extended BOS descriptor. This feature\n helps to optimize power savings by maximizing the opportunity for longer\n L1 residency time.\n \n Signed-off-by: Thinh Nguyen \u003cthinhn@synopsys.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 54fb5ba647f8f942e372dee45025bce4adc08c38\nAuthor: Thinh Nguyen \u003cThinh.Nguyen@synopsys.com\u003e\nDate: Mon Aug 19 18:36:06 2019 -0700\n\n usb: dwc3: gadget: Set BESL config parameter\n \n When operating with LPM signals, the controller asserts the deep\n low-power signal (utmi_l1_suspend_n) to the phy when the BESL value of\n the LPM token is equal to or greater than DCTL.HIRD_Thres[3:0] (and\n with DCTL.HIRD_Thres[4] set). Otherwise, the shallow low-power signal\n (utmi_sleep_n) is asserted. Set the recommended deep BESL equal to the\n controller\u0027s DCTL.HIRD_Thres[3:0] setting, and set the baseline BESL\n to 0 for the shallow low-power signal. This maximizes the opportunity\n for L1 residency and optimizes power savings.\n \n Signed-off-by: Thinh Nguyen \u003cthinhn@synopsys.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 16fe4f304015c39aa3a7aeef1beeac2ed585c7c6\nAuthor: Thinh Nguyen \u003cThinh.Nguyen@synopsys.com\u003e\nDate: Mon Aug 19 18:35:58 2019 -0700\n\n usb: dwc3: Separate field holding multiple properties\n \n dwc-\u003ehird_threshold field should store \"snps,hird_threshold\" property\n only and not a combination of multiple properties. Remove the value of\n \"snps,is-utmi-l1-suspend\" property from the field dwc-\u003ehird_threshold.\n \n Signed-off-by: Thinh Nguyen \u003cthinhn@synopsys.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 05f0b20b67f4185cc0022e49de08e1b9b2dbeb36\nAuthor: Thinh Nguyen \u003cThinh.Nguyen@synopsys.com\u003e\nDate: Mon Aug 19 18:35:47 2019 -0700\n\n usb: gadget: Export recommended BESL values\n \n Currently there\u0027s no option for the controller driver to report the\n recommended Best Effort Service Latency (BESL) when operating with LPM\n support. Add new fields in usb_dcd_config_params to export the\n recommended baseline and deep BESL values for the function drivers to\n set the proper BESL value in the BOS descriptor.\n \n Signed-off-by: Thinh Nguyen \u003cthinhn@synopsys.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit f1b19006f4c9e83c9e19c58102157ff1fdebca22\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Aug 19 13:12:34 2019 +0530\n\n usb: phy: phy-fsl-usb: Make structure fsl_otg_initdata constant\n \n Static structure fsl_otg_initdata, of type fsl_otg_config, is used only\n once, when the value of its field otg_port is assigned to another\n variable. As fsl_otg_initdata.otg_port is of type u8, any changes to\n its copy do not affect the original, and fsl_otg_initdata itself is\n never modified. Hence make fsl_otg_initdata constant to protect it from\n unintended modification.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit c33f141615c2850ab512f7dca654f8f0914ab9ab\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Tue Aug 20 21:16:27 2019 -0500\n\n usb: udc: lpc32xx: silence fall-through warning\n \n Silence the following fall-through warning by adding a break statement:\n \n drivers/usb/gadget/udc/lpc32xx_udc.c:2230:3: warning: this statement may\n fall through [-Wimplicit-fallthrough\u003d]\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 1cf084d161f40baf4aa6976bb2dcef114adbebd6\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Wed Aug 21 15:35:18 2019 +0200\n\n usb: dwc3: meson-g12a: fix suspend resume regulator unbalanced disables\n \n When going in suspend, in Device mode, then resuming back leads\n to the following:\n \n unbalanced disables for USB_PWR_EN\n WARNING: CPU: 0 PID: 163 at ../drivers/regulator/core.c:2590 _regulator_disable+0x104/0x180\n Hardware name: Amlogic Meson G12A U200 Development Board (DT)\n [...]\n pc : _regulator_disable+0x104/0x180\n lr : _regulator_disable+0x104/0x180\n [...]\n Call trace:\n _regulator_disable+0x104/0x180\n regulator_disable+0x40/0x78\n dwc3_meson_g12a_otg_mode_set+0x84/0xb0\n dwc3_meson_g12a_irq_thread+0x58/0xb8\n irq_thread_fn+0x28/0x80\n irq_thread+0x118/0x1b8\n kthread+0xf4/0x120\n ret_from_fork+0x10/0x18\n \n This disables the regulator if enabled on suspend, and the reverse on\n resume.\n \n Fixes: c99993376f72 (\"usb: dwc3: Add Amlogic G12A DWC3 glue\")\n Tested-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Reported-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 566aacaf80b8ae0cb0e6bc3d11d49a36ab7f32ca\nAuthor: Mao Wenan \u003cmaowenan@huawei.com\u003e\nDate: Thu Aug 22 15:05:57 2019 +0800\n\n usb: udc: lpc32xx: remove set but not used 3 variables\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n drivers/usb/gadget/udc/lpc32xx_udc.c: In function ‘udc_protocol_cmd_r’:\n drivers/usb/gadget/udc/lpc32xx_udc.c:744:6: warning: variable ‘tmp’ set but not used [-Wunused-but-set-variable]\n \n drivers/usb/gadget/udc/lpc32xx_udc.c: In function ‘udc_handle_dma_ep’:\n drivers/usb/gadget/udc/lpc32xx_udc.c:1994:14: warning: variable ‘epstatus’ set but not used [-Wunused-but-set-variable]\n \n drivers/usb/gadget/udc/lpc32xx_udc.c: In function ‘udc_handle_ep0_setup’:\n drivers/usb/gadget/udc/lpc32xx_udc.c:2200:22: warning: variable ‘wLength’ set but not used [-Wunused-but-set-variable]\n \n It is not used since commit 90fccb529d24 (\"usb: gadget: Gadget directory cleanup - group UDC drivers\")\n \n Signed-off-by: Mao Wenan \u003cmaowenan@huawei.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 163be6ff7739b12ff300d77897d340f661821da2\nAuthor: Roger Quadros \u003crogerq@ti.com\u003e\nDate: Thu Aug 22 16:40:28 2019 +0300\n\n usb: gadget: udc: core: Fix segfault if udc_bind_to_driver() for pending driver fails\n \n If a gadget driver is in the pending drivers list, a UDC\n becomes available and udc_bind_to_driver() fails, then it\n gets deleted from the pending list.\n i.e. list_del(\u0026driver-\u003epending) in check_pending_gadget_drivers().\n \n Then if that gadget driver is unregistered,\n usb_gadget_unregister_driver() does a list_del(\u0026driver-\u003epending)\n again thus causing a page fault as that list entry has been poisoned\n by the previous list_del().\n \n Fix this by using list_del_init() instead of list_del() in\n check_pending_gadget_drivers().\n \n Test case:\n \n - Make sure no UDC is available\n - modprobe g_mass_storage file\u003dwrongfile\n - Load UDC driver so it becomes available\n lun0: unable to open backing file: wrongfile\n - modprobe -r g_mass_storage\n \n [ 60.900431] Unable to handle kernel paging request at virtual address dead000000000108\n [ 60.908346] Mem abort info:\n [ 60.911145] ESR \u003d 0x96000044\n [ 60.914227] Exception class \u003d DABT (current EL), IL \u003d 32 bits\n [ 60.920162] SET \u003d 0, FnV \u003d 0\n [ 60.923217] EA \u003d 0, S1PTW \u003d 0\n [ 60.926354] Data abort info:\n [ 60.929228] ISV \u003d 0, ISS \u003d 0x00000044\n [ 60.933058] CM \u003d 0, WnR \u003d 1\n [ 60.936011] [dead000000000108] address between user and kernel address ranges\n [ 60.943136] Internal error: Oops: 96000044 [#1] PREEMPT SMP\n [ 60.948691] Modules linked in: g_mass_storage(-) usb_f_mass_storage libcomposite xhci_plat_hcd xhci_hcd usbcore ti_am335x_adc kfifo_buf omap_rng cdns3 rng_core udc_core crc32_ce xfrm_user crct10dif_ce snd_so6\n [ 60.993995] Process modprobe (pid: 834, stack limit \u003d 0x00000000c2aebc69)\n [ 61.000765] CPU: 0 PID: 834 Comm: modprobe Not tainted 4.19.59-01963-g065f42a60499 #92\n [ 61.008658] Hardware name: Texas Instruments SoC (DT)\n [ 61.014472] pstate: 60000005 (nZCv daif -PAN -UAO)\n [ 61.019253] pc : usb_gadget_unregister_driver+0x7c/0x108 [udc_core]\n [ 61.025503] lr : usb_gadget_unregister_driver+0x30/0x108 [udc_core]\n [ 61.031750] sp : ffff00001338fda0\n [ 61.035049] x29: ffff00001338fda0 x28: ffff800846d40000\n [ 61.040346] x27: 0000000000000000 x26: 0000000000000000\n [ 61.045642] x25: 0000000056000000 x24: 0000000000000800\n [ 61.050938] x23: ffff000008d7b0d0 x22: ffff0000088b07c8\n [ 61.056234] x21: ffff000001100000 x20: ffff000002020260\n [ 61.061530] x19: ffff0000010ffd28 x18: 0000000000000000\n [ 61.066825] x17: 0000000000000000 x16: 0000000000000000\n [ 61.072121] x15: 0000000000000000 x14: 0000000000000000\n [ 61.077417] x13: ffff000000000000 x12: ffffffffffffffff\n [ 61.082712] x11: 0000000000000030 x10: 7f7f7f7f7f7f7f7f\n [ 61.088008] x9 : fefefefefefefeff x8 : 0000000000000000\n [ 61.093304] x7 : ffffffffffffffff x6 : 000000000000ffff\n [ 61.098599] x5 : 8080000000000000 x4 : 0000000000000000\n [ 61.103895] x3 : ffff000001100020 x2 : ffff800846d40000\n [ 61.109190] x1 : dead000000000100 x0 : dead000000000200\n [ 61.114486] Call trace:\n [ 61.116922] usb_gadget_unregister_driver+0x7c/0x108 [udc_core]\n [ 61.122828] usb_composite_unregister+0x10/0x18 [libcomposite]\n [ 61.128643] msg_cleanup+0x18/0xfce0 [g_mass_storage]\n [ 61.133682] __arm64_sys_delete_module+0x17c/0x1f0\n [ 61.138458] el0_svc_common+0x90/0x158\n [ 61.142192] el0_svc_handler+0x2c/0x80\n [ 61.145926] el0_svc+0x8/0xc\n [ 61.148794] Code: eb03003f d10be033 54ffff21 a94d0281 (f9000420)\n [ 61.154869] ---[ end trace afb22e9b637bd9a7 ]---\n Segmentation fault\n \n Acked-by: Alan Stern \u003cstern@rowland.harvard.edu\u003e\n Signed-off-by: Roger Quadros \u003crogerq@ti.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit bceb418b65b917af6d36a92ea1eaca54cafc1c2e\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Aug 19 12:54:36 2019 +0530\n\n usb: dwc3: st: Add of_dev_put() in probe function\n \n In function st_dwc3_probe, variable child_pdev takes the value returned\n by of_find_device_by_node, which gets a device pointer but does not put\n it. If child_pdev is not put before the probe function returns, it may\n cause a reference leak. Hence put child_pdev after its last usage.\n Issue found with Coccinelle.\n \n Reviewed-by: Patrice Chotard \u003cpatrice.chotard@st.com\u003e\n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit e36721b90144bb46e1b6477be3ab63439c7fb79b\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Aug 19 12:54:35 2019 +0530\n\n usb: dwc3: st: Add of_node_put() before return in probe function\n \n The local variable child in the function st_dwc3_probe takes the return\n value of of_get_child_by_name, which gets a node and does not put it. If\n the function returns without releasing child, this could cause a memory\n error. Hence put child as soon as there is no more use for it. Also\n create a new label, err_node_put, just before label undo_softreset; so\n that err_node_put puts child. In between initialisation of child and its\n first put, modify all statements that go to undo_softreset to now go to\n err_node_put instead, from where they can fall through to\n undo_softreset.\n Issue found with Coccinelle.\n \n Reviewed-by: Patrice Chotard \u003cpatrice.chotard@st.com\u003e\n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit c09b73cfac2a9317f1104169045c519c6021aa1d\nAuthor: Roger Quadros \u003crogerq@ti.com\u003e\nDate: Mon Aug 26 16:10:58 2019 +0300\n\n usb: dwc3: don\u0027t set gadget-\u003eis_otg flag\n \n This reverts\n commit 6a4290cc28be1 (\"usb: dwc3: gadget: set the OTG flag in dwc3 gadget driver.\")\n \n We don\u0027t yet support any of the OTG mechanisms (HNP/SRP/ADP)\n and are not setting gadget-\u003eotg_caps, so don\u0027t set gadget-\u003eis_otg\n flag.\n \n If we do then we end up publishing a OTG1.0 descriptor in\n the gadget descriptor which causes device enumeration to fail\n if we are connected to a host with CONFIG_USB_OTG enabled.\n \n Host side log without this patch\n \n [ 96.720453] usb 1-1: new high-speed USB device number 2 using xhci-hcd\n [ 96.901391] usb 1-1: Dual-Role OTG device on non-HNP port\n [ 96.907552] usb 1-1: set a_alt_hnp_support failed: -32\n [ 97.060447] usb 1-1: new high-speed USB device number 3 using xhci-hcd\n [ 97.241378] usb 1-1: Dual-Role OTG device on non-HNP port\n [ 97.247536] usb 1-1: set a_alt_hnp_support failed: -32\n [ 97.253606] usb usb1-port1: attempt power cycle\n [ 97.960449] usb 1-1: new high-speed USB device number 4 using xhci-hcd\n [ 98.141383] usb 1-1: Dual-Role OTG device on non-HNP port\n [ 98.147540] usb 1-1: set a_alt_hnp_support failed: -32\n [ 98.300453] usb 1-1: new high-speed USB device number 5 using xhci-hcd\n [ 98.481391] usb 1-1: Dual-Role OTG device on non-HNP port\n [ 98.487545] usb 1-1: set a_alt_hnp_support failed: -32\n [ 98.493532] usb usb1-port1: unable to enumerate USB device\n \n Signed-off-by: Roger Quadros \u003crogerq@ti.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 9d6b4b871dcfd6796b46a05e002884e051687e47\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Sun Jul 21 18:33:58 2019 +0200\n\n thermal: tegra: Fix a typo\n \n s/sochterm/soctherm/\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Signed-off-by: Zhang Rui \u003crui.zhang@intel.com\u003e\n\ncommit b9cd1663fb49c9432c346be742f94ef25b406c89\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Mon Jul 8 20:34:09 2019 +0800\n\n thermal: rcar_gen3_thermal: Replace devm_add_action() followed by failure action with devm_add_action_or_reset()\n \n devm_add_action_or_reset() is introduced as a helper function which\n internally calls devm_add_action(). If devm_add_action() fails\n then it will execute the action mentioned and return the error code.\n This reduce source code size (avoid writing the action twice)\n and reduce the likelyhood of bugs.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Signed-off-by: Zhang Rui \u003crui.zhang@intel.com\u003e\n\ncommit 60bda037f1dd8151e0c9ee5b09f0c091a0f643cd\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Tue Aug 27 21:31:02 2019 +0200\n\n posix-cpu-timers: Utilize timerqueue for storage\n \n Using a linear O(N) search for timer insertion affects execution time and\n D-cache footprint badly with a larger number of timers.\n \n Switch the storage to a timerqueue which is already used for hrtimers and\n alarmtimers. It does not affect the size of struct k_itimer as it.alarm is\n still larger.\n \n The extra list head for the expiry list will go away later once the expiry\n is moved into task work context.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/alpine.DEB.2.21.1908272129220.1939@nanos.tec.linutronix.de\n\ncommit 244d49e30653658d4e7e9b2b8427777cbbc5affe\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:24 2019 +0200\n\n posix-cpu-timers: Move state tracking to struct posix_cputimers\n \n Put it where it belongs and clean up the ifdeffery in fork completely.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190821192922.743229404@linutronix.de\n\ncommit 8991afe2640d05a805eba01277856e8549cdc838\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:23 2019 +0200\n\n posix-cpu-timers: Deduplicate rlimit handling\n \n Both thread and process expiry functions have the same functionality for\n sending signals for soft and hard RLIMITs duplicated in 4 different\n ways.\n \n Split it out into a common function and cleanup the callsites.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192922.653276779@linutronix.de\n\ncommit dd6702241337bcd0bae01d2644b7bae1a496d937\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:22 2019 +0200\n\n posix-cpu-timers: Remove pointless comparisons\n \n The soft RLIMIT expiry code checks whether the soft limit is greater than\n the hard limit. That\u0027s pointless because if the soft RLIMIT is greater than\n the hard RLIMIT then that code cannot be reached as the hard RLIMIT check\n is before that and already killed the process.\n \n Remove it.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192922.548747613@linutronix.de\n\ncommit 8ea1de90a5eccdc18c8f05f8596bae8660a3ff9a\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:21 2019 +0200\n\n posix-cpu-timers: Get rid of 64bit divisions\n \n Instead of dividing A to match the units of B it\u0027s more efficient to\n multiply B to match the units of A.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192922.458286860@linutronix.de\n\ncommit 1cd07c0b94f2c320270d76edb7dd49bceb09c1df\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:20 2019 +0200\n\n posix-cpu-timers: Consolidate timer expiry further\n \n With the array based samples and expiry cache, the expiry function can use\n a loop to collect timers from the clock specific lists.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192922.365469982@linutronix.de\n\ncommit 2bbdbdae05167c688b6d3499a7dab74208b80a22\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:19 2019 +0200\n\n posix-cpu-timers: Get rid of zero checks\n \n Deactivation of the expiry cache is done by setting all clock caches to\n 0. That requires to have a check for zero in all places which update the\n expiry cache:\n \n if (cache \u003d\u003d 0 || new \u003c cache)\n cache \u003d new;\n \n Use U64_MAX as the deactivated value, which allows to remove the zero\n checks when updating the cache and reduces it to the obvious check:\n \n if (new \u003c cache)\n cache \u003d new;\n \n This also removes the weird workaround in do_prlimit() which was required\n to convert a RLIMIT_CPU value of 0 (immediate expiry) to 1 because handing\n in 0 to the posix CPU timer code would have effectively disarmed it.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192922.275086128@linutronix.de\n\ncommit 24db4dd90dd53ad6e3331b6f01cb985e466cface\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:18 2019 +0200\n\n rlimit: Rewrite non-sensical RLIMIT_CPU comment\n \n The comment above the function which arms RLIMIT_CPU in the posix CPU timer\n code makes no sense at all. It claims that the kernel does not return an\n error code when it rejected the attempt to set RLIMIT_CPU. That\u0027s clearly\n bogus as the code does an error check and the rlimit is only set and\n activated when the permission checks are ok. In case of a rejection an\n appropriate error code is returned.\n \n This is a historical and outdated comment which got dragged along even when\n the rlimit handling code was rewritten.\n \n Replace it with an explanation why the setup function is not called when\n the rlimit value is RLIM_INFINITY and how the \u0027disarming\u0027 is handled.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192922.185511287@linutronix.de\n\ncommit fe0517f893d36636de20d0a809fc0c788ca0cade\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:17 2019 +0200\n\n posix-cpu-timers: Respect INFINITY for hard RTTIME limit\n \n The RTIME limit expiry code does not check the hard RTTIME limit for\n INFINITY, i.e. being disabled. Add it.\n \n While this could be considered an ABI breakage if something would depend on\n this behaviour. Though it\u0027s highly unlikely to have an effect because\n RLIM_INFINITY is at minimum INT_MAX and the RTTIME limit is in seconds, so\n the timer would fire after ~68 years.\n \n Adding this obvious correct limit check also allows further consolidation\n of that code and is a prerequisite for cleaning up the 0 based checks and\n the rlimit setter code.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192922.078293002@linutronix.de\n\ncommit b7be4ef1365dcb56fdffc6689e41058b23f5996d\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:16 2019 +0200\n\n posix-cpu-timers: Switch thread group sampling to array\n \n That allows more simplifications in various places.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192921.988426956@linutronix.de\n\ncommit 87dc64480fb19a6a0fedbdff1e2557be50673287\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Aug 26 20:22:24 2019 +0200\n\n posix-cpu-timers: Restructure expiry array\n \n Now that the abused struct task_cputime is gone, it\u0027s more natural to\n bundle the expiry cache and the list head of each clock into a struct and\n have an array of those structs.\n \n Follow the hrtimer naming convention of \u0027bases\u0027 and rename the expiry cache\n to \u0027nextevt\u0027 and adapt all usage sites.\n \n Generates also better code .text size shrinks by 80 bytes.\n \n Suggested-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/alpine.DEB.2.21.1908262021140.1939@nanos.tec.linutronix.de\n\ncommit 46b883995c88520f2c4de6a64cccc04c69d59f83\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:14 2019 +0200\n\n posix-cpu-timers: Remove cputime_expires\n \n The last users of the magic struct cputime based expiry cache are\n gone. Remove the leftovers.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192921.790209622@linutronix.de\n\ncommit 001f7971433a53bb76443cd8f5827ca27b0bc8ca\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:13 2019 +0200\n\n posix-cpu-timers: Make expiry checks array based\n \n The expiry cache is an array indexed by clock ids. The new sample functions\n allow to retrieve a corresponding array of samples.\n \n Convert the fastpath expiry checks to make use of the new sample functions\n and do the comparisons on the sample and the expiry array.\n \n Make the check for the expiry array being zero array based as well.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192921.695481430@linutronix.de\n\ncommit b0d524f77956eec887b30732af1f5f98cbf62b9f\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:12 2019 +0200\n\n posix-cpu-timers: Provide array based sample functions\n \n Instead of using task_cputime and doing the addition of utime and stime at\n all call sites, it\u0027s way simpler to have a sample array which allows\n indexed based checks against the expiry cache array.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192921.590362974@linutronix.de\n\ncommit bbc9bae1e49bee9862c7f24101a728e73cd9f589\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:11 2019 +0200\n\n posix-cpu-timers: Remove the odd field rename defines\n \n The last users of the odd define based renaming of struct task_cputime\n members are gone. Good riddance.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192921.499058279@linutronix.de\n\ncommit c02b078e63a6f42029cb655d0aa3c991271637ac\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:10 2019 +0200\n\n posix-cpu-timers: Switch check_*_timers() to array cache\n \n Use the array based expiry cache in check_thread_timers() and convert the\n store in check_process_timers() for consistency.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192921.408222378@linutronix.de\n\ncommit 1b0dd96d0f07c482bf1d04987cc1b35e376a7518\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:09 2019 +0200\n\n posix-cpu-timers: Simplify set_process_cpu_timer()\n \n The expiry cache can now be accessed as an array. Replace the per clock\n checks with a simple comparison of the clock indexed array member.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192921.303316423@linutronix.de\n\ncommit 3b495b22d04df3220ccae829bf7c5cadb3059ccf\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:08 2019 +0200\n\n posix-cpu-timers: Simplify timer queueing\n \n Now that the expiry cache can be accessed as an array, the per clock\n checking can be reduced to just comparing the corresponding array elements.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192921.212129449@linutronix.de\n\ncommit 11b8462f7e1d25f639c88949a2746a9c2667a766\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:07 2019 +0200\n\n posix-cpu-timers: Provide array based access to expiry cache\n \n Using struct task_cputime for the expiry cache is a pretty odd choice and\n comes with magic defines to rename the fields for usage in the expiry\n cache.\n \n struct task_cputime is basically a u64 array with 3 members, but it has\n distinct members.\n \n The expiry cache content is different than the content of task_cputime\n because\n \n expiry[PROF] \u003d task_cputime.stime + task_cputime.utime\n expiry[VIRT] \u003d task_cputime.utime\n expiry[SCHED] \u003d task_cputime.sum_exec_runtime\n \n So there is no direct mapping between task_cputime and the expiry cache and\n the #define based remapping is just a horrible hack.\n \n Having the expiry cache array based allows further simplification of the\n expiry code.\n \n To avoid an all in one cleanup which is hard to review add a temporary\n anonymous union into struct task_cputime which allows array based access to\n it. That requires to reorder the members. Add a build time sanity check to\n validate that the members are at the same place.\n \n The union and the build time checks will be removed after conversion.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192921.105793824@linutronix.de\n\ncommit 3a245c0f110e2bfcf7f2cd2248a29005c78999e3\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:06 2019 +0200\n\n posix-cpu-timers: Move expiry cache into struct posix_cputimers\n \n The expiry cache belongs into the posix_cputimers container where the other\n cpu timers information is.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192921.014444012@linutronix.de\n\ncommit 9eacb5c7e6607aba00a7322b21cad83fc8b101c8\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:05 2019 +0200\n\n sched: Move struct task_cputime to types.h\n \n For upcoming posix-timer changes to avoid include recursion hell.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190821192920.909530418@linutronix.de\n\ncommit 2b69942f9021bf75bd1b001f53bd2578361fadf3\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:04 2019 +0200\n\n posix-cpu-timers: Create a container struct\n \n Per task/process data of posix CPU timers is all over the place which\n makes the code hard to follow and requires ifdeffery.\n \n Create a container to hold all this information in one place, so data is\n consolidated and the ifdeffery can be confined to the posix timer header\n file and removed from places like fork.\n \n As a first step, move the cpu_timers list head array into the new struct\n and clean up the initializers and simplify fork. The remaining #ifdef in\n fork will be removed later.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192920.819418976@linutronix.de\n\ncommit ab693c5a5e3107f035d5162e6ada9aaf7dd76a1d\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:03 2019 +0200\n\n posix-cpu-timers: Move prof/virt_ticks into caller\n \n The functions have only one caller left. No point in having them.\n \n Move the almost duplicated code into the caller and simplify it.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192920.729298382@linutronix.de\n\ncommit 0476ff2c151ee35bda2f938e8828b6f6113ebf55\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:02 2019 +0200\n\n posix-cpu-timers: Sample task times once in expiry check\n \n Sampling the task times twice does not make sense. Do it once.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192920.639878168@linutronix.de\n\ncommit 8c2d74f03705c2c993a57673bae8fd535eabede6\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:01 2019 +0200\n\n posix-cpu-timers: Get rid of pointer indirection\n \n Now that the sample functions have no return value anymore, the result can\n simply be returned instead of using pointer indirection.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192920.535079278@linutronix.de\n\ncommit 2092c1d4fed9f279d10600b4c1b5167dd8486a2a\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:09:00 2019 +0200\n\n posix-cpu-timers: Simplify sample functions\n \n All callers hand in a valdiated clock id. Remove the return value which was\n unchecked in most places anyway.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192920.430475832@linutronix.de\n\ncommit 5405d0051f7ca820d1781d87baf4d730ff58f208\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:08:59 2019 +0200\n\n posix-cpu-timers: Remove pointless return value check\n \n set_process_cpu_timer() checks already whether the clock id is valid. No\n point in checking the return value of the sample function. That allows to\n simplify the sample function later.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192920.339725769@linutronix.de\n\ncommit da020ce406b2a9b714b82de9123a4c5a4848647b\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:08:58 2019 +0200\n\n posix-cpu-timers: Use clock ID in posix_cpu_timer_rearm()\n \n Extract the clock ID (PROF/VIRT/SCHED) from the clock selector and use it\n as argument to the sample functions. That allows to simplify them once all\n callers are fixed.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192920.245357769@linutronix.de\n\ncommit 99093c5b81f58815fbfc1fe301c87206c6f5f730\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:08:57 2019 +0200\n\n posix-cpu-timers: Use clock ID in posix_cpu_timer_get()\n \n Extract the clock ID (PROF/VIRT/SCHED) from the clock selector and use it\n as argument to the sample functions. That allows to simplify them once all\n callers are fixed.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192920.155487201@linutronix.de\n\ncommit c7a37c6f4c651a531101c5721814333bae2804ec\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:08:56 2019 +0200\n\n posix-cpu-timers: Use clock ID in posix_cpu_timer_set()\n \n Extract the clock ID (PROF/VIRT/SCHED) from the clock selector and use it\n as argument to the sample functions. That allows to simplify them once all\n callers are fixed.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192920.050770464@linutronix.de\n\ncommit 24ab7f5a7b2c917e89fc6a87252f18faff91d6ce\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:08:55 2019 +0200\n\n posix-cpu-timers: Consolidate thread group sample code\n \n cpu_clock_sample_group() and cpu_timer_sample_group() are almost the\n same. Before the rename one called thread_group_cputimer() and the other\n thread_group_cputime(). Really intuitive function names.\n \n Consolidate the functions and also avoid the thread traversal when\n the thread group\u0027s accounting is already active.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192919.960966884@linutronix.de\n\ncommit c506bef424ca282f2ad357e86fee940c69018974\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:08:54 2019 +0200\n\n posix-cpu-timers: Rename thread_group_cputimer() and make it static\n \n thread_group_cputimer() is a complete misnomer. The function does two things:\n \n - For arming process wide timers it makes sure that the atomic time\n storage is up to date. If no cpu timer is armed yet, then the atomic\n time storage is not updated by the scheduler for performance reasons.\n \n In that case a full summing up of all threads needs to be done and the\n update needs to be enabled.\n \n - Samples the current time into the caller supplied storage.\n \n Rename it to thread_group_start_cputime(), make it static and fixup the\n callsite.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192919.869350319@linutronix.de\n\ncommit a324956fae05d863386c682830e917f6685f1d4f\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:08:53 2019 +0200\n\n posix-cpu-timers: Sample directly in timer check\n \n The thread group accounting is active, otherwise the expiry function would\n not be running. Sample the thread group time directly.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192919.780348088@linutronix.de\n\ncommit a34360d42434bbf28c0f375444c52c154ae3e6cf\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:08:52 2019 +0200\n\n itimers: Use quick sample function\n \n get_itimer() locks sighand lock and checks whether the timer is already\n expired. If it is not expired then the thread group cputime accounting is\n already enabled. Use the sampling function not the one which is meant for\n starting a timer.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192919.689713638@linutronix.de\n\ncommit 19298fbf453c90a6cf72288155f80c6f55e9139d\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:08:51 2019 +0200\n\n posix-cpu-timers: Provide quick sample function for itimer\n \n get_itimer() needs a sample of the current thread group cputime. It invokes\n thread_group_cputimer() - which is a misnomer. That function also starts\n eventually the group cputime accouting which is bogus because the\n accounting is already active when a timer is armed.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192919.599658199@linutronix.de\n\ncommit e5a8b65b4cb2fe024b83bdec0424269949cc0a27\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:08:50 2019 +0200\n\n posix-cpu-timers: Use common permission check in posix_cpu_timer_create()\n \n Yet another copy of the same thing gone...\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192919.505833418@linutronix.de\n\ncommit bfcf3e92c6c07cd1084624bad5622f3dad96328c\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:08:49 2019 +0200\n\n posix-cpu-timers: Use common permission check in posix_cpu_clock_get()\n \n Replace the next slightly different copy of permission checks. That also\n removes the necessarity to check the return value of the sample functions\n because the clock id is already validated.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190821192919.414813172@linutronix.de\n\ncommit 6ae40e3fdcd33a6ff3c490b9302d6a1861093f65\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Aug 21 21:08:48 2019 +0200\n\n posix-cpu-timers: Provide task validation functions\n \n The code contains three slightly different copies of validating whether a\n given clock resolves to a valid task and whether the current caller has\n permissions to access it.\n \n Create central functions. Replace check_clock() as a first step and rename\n it to something sensible.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190821192919.326097175@linutronix.de\n\ncommit 7ae2227b1c1928cfc8220ca6383c757b185fcf32\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 15 11:33:34 2019 +0530\n\n mtd: spi-nor: hisi-sfc: Add of_node_put() before break\n \n Each iteration of for_each_available_child_of_node puts the previous\n node, but in the case of a break from the middle of the loop, there\n is no put, thus causing a memory leak. Hence add an of_node_put before\n the break.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n\ncommit 6597f0b0516ab078b9ed2df4dd9aea17275995cd\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 8 13:21:04 2019 +0530\n\n mtd: spi-nor: aspeed-smc: Add of_node_put()\n \n Each iteration of for_each_available_child_of_node puts the previous\n node, but in the case of a break from the middle of the loop, there is\n no put, thus causing a memory leak. Upon termination of the loop\n (whether by break or a natural exit), either ret will have a non-zero\n value or child will be NULL. Hence add an of_node_put() that will\n execute only when ret has a non-zero value, as calling of_node_put() on\n a possible NULL value does not cause any further issues.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Reviewed-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n\ncommit 620df2497415a932abd0d9f59688af79f60de8bf\nAuthor: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\nDate: Sat Aug 24 12:19:24 2019 +0000\n\n mtd: spi-nor: Introduce spi_nor_get_flash_info()\n \n Dedicate a function for getting the pointer to the flash_info\n const struct. Trim a bit the spi_nor_scan() huge function.\n \n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit 696ce50f4e9377d68740ed46c0547be0e7f9d3f7\nAuthor: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\nDate: Sat Aug 24 12:19:22 2019 +0000\n\n mtd: spi-nor: Introduce spi_nor_set_addr_width()\n \n Parsing of flash parameters were interleaved with setting of the\n nor addr width. Dedicate a function for setting nor addr width.\n \n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit 07920dfcf0039695e0202c35b458e0ffce4de385\nAuthor: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\nDate: Sat Aug 24 12:19:20 2019 +0000\n\n mtd: spi-nor: Bring flash params init together\n \n Bring all flash parameters default initialization in\n spi_nor_legacy_params_init().\n \n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit ad3bba06b63697fd47525a13d29e8ac0312b0302\nAuthor: Boris Brezillon \u003cboris.brezillon@bootlin.com\u003e\nDate: Tue Jul 30 11:56:53 2019 +0300\n\n mtd: spi-nor: Add the SPI_NOR_XSR_RDY flag\n \n S3AN flashes use a specific opcode to read the status register.\n We currently use the SPI_S3AN flag to decide whether this specific\n SR read opcode should be used, but SPI_S3AN is about to disappear, so\n let\u0027s add a new flag.\n \n Note that we use the same bit as SPI_S3AN implies SPI_NOR_XSR_RDY and\n vice versa.\n \n Signed-off-by: Boris Brezillon \u003cboris.brezillon@bootlin.com\u003e\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit 641edddb4f43e8adfd851aea3e48ba35c3b81f4a\nAuthor: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\nDate: Sat Aug 24 12:07:16 2019 +0000\n\n mtd: spi-nor: Add s3an_post_sfdp_fixups()\n \n s3an_nor_scan() was overriding the opcode selection done in\n spi_nor_default_setup(). Set nor-\u003esetup() method in order to\n avoid the unnecessary call to spi_nor_default_setup().\n \n Now that the call to spi_nor_default_setup() is skipped, set\n mtd.erasesize to nor-\u003einfo-\u003esector_size, as it was when\n spi_nor_select_erase() was called.\n \n No dummy byte is required for the S3AN\u0027s Random Read command (0x03),\n so no need to set nor-\u003eread_dummy.\n \n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit 2d7ff858e5f683393f32b07e64e565877a2e4bcb\nAuthor: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\nDate: Sun Aug 25 22:48:36 2019 +0300\n\n mtd: spi-nor: Add a -\u003esetup() method\n \n nor-\u003eparams.setup() configures the SPI NOR memory. Useful for SPI NOR\n flashes that have peculiarities to the SPI NOR standard, e.g.\n different opcodes, specific address calculation, page size, etc.\n Right now the only user will be the S3AN chips, but other\n manufacturers can implement it if needed.\n \n Move spi_nor_setup() related code in order to avoid a forward\n declaration to spi_nor_default_setup().\n \n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit 36499596280359d34a0663afe0abbf34d80862e8\nAuthor: Boris Brezillon \u003cboris.brezillon@bootlin.com\u003e\nDate: Sat Aug 24 10:17:20 2019 +0300\n\n mtd: spi-nor: Add a -\u003econvert_addr() method\n \n In order to separate manufacturer quirks from the core we need to get\n rid of all the manufacturer specific flags, like the\n SNOR_F_S3AN_ADDR_DEFAULT one.\n \n This can easily be replaced by a -\u003econvert_addr() hook, which when\n implemented will provide the core with an easy way to convert an\n absolute address into something the flash understands.\n \n Right now the only user are the S3AN chips, but other manufacturers\n can implement it if needed.\n \n Signed-off-by: Boris Brezillon \u003cboris.brezillon@bootlin.com\u003e\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit 92094ebc385ef52d45e27d23e9a7c32abf8df44e\nAuthor: Boris Brezillon \u003cboris.brezillon@bootlin.com\u003e\nDate: Sat Aug 24 12:07:11 2019 +0000\n\n mtd: spi-nor: Add spansion_post_sfdp_fixups()\n \n Add a spansion_post_sfdp_fixups() function to fix the erase opcode,\n erase sector size and set the SNOR_F_4B_OPCODES flag.\n This way, all spansion related quirks are placed in the\n spansion_post_sfdp_fixups() function.\n \n Signed-off-by: Boris Brezillon \u003cboris.brezillon@bootlin.com\u003e\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit 2b12ae1f2fe5f1cd87e20b91231a22390afd0464\nAuthor: Boris Brezillon \u003cboris.brezillon@bootlin.com\u003e\nDate: Sat Aug 24 12:07:09 2019 +0000\n\n mtd: spi-nor: Add post_sfdp() hook to tweak flash config\n \n SFDP tables are sometimes wrong and we need a way to override the\n config chosen by the SFDP parsing logic without discarding all of it.\n \n Add a new hook called after the SFDP parsing has taken place to deal\n with such problems.\n \n Signed-off-by: Boris Brezillon \u003cboris.brezillon@bootlin.com\u003e\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit dff972458acb05ab919b6950da99f8172ab14836\nAuthor: Boris Brezillon \u003cboris.brezillon@bootlin.com\u003e\nDate: Sat Aug 24 09:22:03 2019 +0300\n\n mtd: spi-nor: Rework the SPI NOR lock/unlock logic\n \n Add the SNOR_F_HAS_LOCK flag and set it when SPI_NOR_HAS_LOCK is set\n in the flash_info entry or when it\u0027s a Micron or ST flash.\n \n Move the locking hooks in a separate struct so that we have just\n one field to update when we change the locking implementation.\n \n Signed-off-by: Boris Brezillon \u003cboris.brezillon@bootlin.com\u003e\n [tudor.ambarus@microchip.com: use -\u003edefault_init() hook, introduce\n spi_nor_late_init_params(), set ops in nor-\u003eparams]\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit 64c160f32235f725a5378c66fb3583c720357ab9\nAuthor: Boris Brezillon \u003cboris.brezillon@bootlin.com\u003e\nDate: Sat Aug 24 08:45:28 2019 +0300\n\n mtd: spi-nor: Create a -\u003eset_4byte() method\n \n The procedure used to enable 4 byte addressing mode depends on the NOR\n device, so let\u0027s provide a hook so that manufacturer specific handling\n can be implemented in a sane way.\n \n Signed-off-by: Boris Brezillon \u003cboris.brezillon@bootlin.com\u003e\n [tudor.ambarus@microchip.com: use nor-\u003eparams.set_4byte() instead of\n nor-\u003eset_4byte()]\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit 1c1d8d98e1c7067758e624a62eaae30b18683e32\nAuthor: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\nDate: Sat Aug 24 08:27:02 2019 +0300\n\n mtd: spi-nor: Split spi_nor_init_params()\n \n Add functions to delimit what the chunks of code do:\n \n static void spi_nor_init_params()\n {\n spi_nor_info_init_params()\n spi_nor_manufacturer_init_params()\n spi_nor_sfdp_init_params()\n }\n \n Add descriptions to all methods.\n \n spi_nor_init_params() becomes of type void, as all its children\n return void.\n \n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit 22f2eaac3f3128c4aa3acad83d43701265c1c150\nAuthor: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\nDate: Sat Aug 24 12:00:41 2019 +0000\n\n mtd: spi-nor: Move manufacturer quad_enable() in -\u003edefault_init()\n \n The goal is to move the quad_enable manufacturer specific init in the\n nor-\u003emanufacturer-\u003efixups-\u003edefault_init()\n \n The legacy quad_enable() implementation is spansion_quad_enable(),\n select this method by default.\n \n Set specific manufacturer fixups-\u003edefault_init() hooks to overwrite\n the default quad_enable() implementation when needed.\n \n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit 48e4d973aefeea6080e6f6dc69326a66004c7923\nAuthor: Boris Brezillon \u003cboris.brezillon@bootlin.com\u003e\nDate: Sat Aug 24 12:00:39 2019 +0000\n\n mtd: spi-nor: Add a default_init() fixup hook for gd25q256\n \n gd25q256 needs to tweak the -\u003equad_enable() implementation and the\n -\u003edefault_init() fixup hook is the perfect place to do that. This way,\n if we ever need to tweak more things for this flash, we won\u0027t have to\n add new fields in flash_info.\n \n We can get rid of the flash_info-\u003equad_enable field as gd25q256 was\n the only user.\n \n Signed-off-by: Boris Brezillon \u003cboris.brezillon@bootlin.com\u003e\n [tudor.ambarus@microchip.com: use -\u003edefault_init() hook instead of\n -\u003epost_sfdp()]\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit ce0b6f3f3c43271f4486c7255656277cc5996f6e\nAuthor: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\nDate: Sat Aug 24 12:00:37 2019 +0000\n\n mtd: spi-nor: Add default_init() hook to tweak flash parameters\n \n As of now, the flash parameters initialization logic is as following:\n \n a/ default flash parameters init in spi_nor_init_params()\n b/ manufacturer specific flash parameters updates, split across entire\n spi-nor core code\n c/ flash parameters updates based on SFDP tables\n d/ post BFPT flash parameter updates\n \n In the quest of removing the manufacturer specific code from the spi-nor\n core, we want to impose a timeline/priority on how the flash parameters\n are updated. The following sequence of calls is pursued:\n \n 1/ spi-nor core parameters init based on \u0027flash_info\u0027 struct:\n spi_nor_info_init_params()\n \n which can be overwritten by:\n 2/ MFR-based manufacturer flash parameters init:\n nor-\u003emanufacturer-\u003efixups-\u003edefault_init()\n \n which can be overwritten by:\n 3/ specific flash_info tweeks done when decisions can not be done just on\n MFR:\n nor-\u003einfo-\u003efixups-\u003edefault_init()\n \n which can be overwritten by:\n 4/ SFDP tables flash parameters init - SFDP knows better:\n spi_nor_sfdp_init_params()\n \n which can be overwritten by:\n 5/ post SFDP tables flash parameters updates - in case manufacturers get\n the serial flash tables wrong or incomplete.\n nor-\u003einfo-\u003efixups-\u003epost_sfdp()\n The later can be extended to nor-\u003emanufacturer-\u003efixups-\u003epost_sfdp() if\n needed.\n \n This patch opens doors for steps 2/ and 3/.\n \n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit c46872170a54c902425c8580fd0a75a56ac1d147\nAuthor: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\nDate: Fri Aug 23 17:36:03 2019 +0300\n\n mtd: spi-nor: Move erase_map to \u0027struct spi_nor_flash_parameter\u0027\n \n All flash parameters and settings should reside inside\n \u0027struct spi_nor_flash_parameter\u0027. Move the SMPT parsed erase map\n from \u0027struct spi_nor\u0027 to \u0027struct spi_nor_flash_parameter\u0027.\n \n Please note that there is a roll-back mechanism for the flash\n parameter and settings, for cases when SFDP parser fails. The SFDP\n parser receives a Stack allocated copy of nor-\u003eparams, called\n sfdp_params, and uses it to retrieve the serial flash discoverable\n parameters. JESD216 SFDP is a standard and has a higher priority\n than the default initialized flash parameters, so will overwrite the\n sfdp_params data when needed. All SFDP code uses the local copy of\n nor-\u003eparams, that will overwrite it in the end, if the parser succeds.\n \n Saving and restoring the nor-\u003eparams.erase_map is no longer needed,\n since the SFDP code does not touch it.\n \n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit 42f5994724bcbcdbeb743dcc3e2756ec582cb86b\nAuthor: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\nDate: Fri Aug 23 15:53:39 2019 +0000\n\n mtd: spi-nor: Drop quad_enable() from \u0027struct spi-nor\u0027\n \n All flash parameters and settings should reside inside\n \u0027struct spi_nor_flash_parameter\u0027. Drop the local copy of\n quad_enable() and use the one from \u0027struct spi_nor_flash_parameter\u0027.\n \n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit 1e35a56781b4b5d83f428c2c40a1b6fa5bca8cd1\nAuthor: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\nDate: Fri Aug 23 15:53:37 2019 +0000\n\n mtd: spi-nor: Use nor-\u003eparams\n \n The Flash parameters and settings are now stored in \u0027struct spi_nor\u0027.\n Use this instead of the stack allocated params.\n \n Few functions stop passing pointer to params, as they can get it from\n \u0027struct spi_nor\u0027. spi_nor_parse_sfdp() and children will keep passing\n pointer to params because of the roll-back mechanism: in case the\n parsing of SFDP fails, the legacy flash parameter and settings will be\n restored.\n \n Zeroing params is no longer needed because all SPI NOR users kzalloc\n \u0027struct spi_nor\u0027.\n \n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit 47599127a2e8c42843e27fc370e63cc5344a2a38\nAuthor: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\nDate: Fri Aug 23 15:53:35 2019 +0000\n\n mtd: spi-nor: Regroup flash parameter and settings\n \n The scope is to move all [FLASH-SPECIFIC] parameters and settings\n from \u0027struct spi_nor\u0027 to \u0027struct spi_nor_flash_parameter\u0027.\n \n \u0027struct spi_nor_flash_parameter\u0027 describes the hardware capabilities\n and associated settings of the SPI NOR flash memory. It includes\n legacy flash parameters and settings that can be overwritten by the\n spi_nor_fixups hooks, or dynamically when parsing the JESD216\n Serial Flash Discoverable Parameters (SFDP) tables. All SFDP params\n and settings will fit inside \u0027struct spi_nor_flash_parameter\u0027.\n \n Move spi_nor_hwcaps related code to avoid forward declarations.\n Add a forward declaration that we can\u0027t avoid: \u0027struct spi_nor\u0027 will\n be used in \u0027struct spi_nor_flash_parameter\u0027.\n \n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit 92b6d38f1a3d59fb43d36c2d9681b9b8f1e46c05\nAuthor: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\nDate: Thu Aug 22 16:28:15 2019 +0300\n\n mtd: spi-nor: Remove unused macro\n \n Remove leftover from nor-\u003ecmd_buf.\n \n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n\ncommit 42880f726c66f13ae1d9ac9ce4c43abe64ecac84\nAuthor: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\nDate: Tue Aug 6 11:46:01 2019 +0300\n\n perf/x86/intel: Support PEBS output to PT\n \n If PEBS declares ability to output its data to Intel PT stream, use the\n aux_output attribute bit to enable PEBS data output to PT. This requires\n a PT event to be present and scheduled in the same context. Unlike the\n DS area, the kernel does not extract PEBS records from the PT stream to\n generate corresponding records in the perf stream, because that would\n require real time in-kernel PT decoding, which is not feasible. The PMI,\n however, can still be used.\n \n The output setting is per-CPU, so all PEBS events must be either writing\n to PT or to the DS area, therefore, in case of conflict, the conflicting\n event will fail to schedule, allowing the rotation logic to alternate\n between the PEBS-\u003ePT and PEBS-\u003eDS events.\n \n Signed-off-by: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: kan.liang@linux.intel.com\n Link: https://lkml.kernel.org/r/20190806084606.4021-3-alexander.shishkin@linux.intel.com\n\ncommit ab43762ef010967e4ccd53627f70a2eecbeafefb\nAuthor: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\nDate: Tue Aug 6 11:46:00 2019 +0300\n\n perf: Allow normal events to output AUX data\n \n In some cases, ordinary (non-AUX) events can generate data for AUX events.\n For example, PEBS events can come out as records in the Intel PT stream\n instead of their usual DS records, if configured to do so.\n \n One requirement for such events is to consistently schedule together, to\n ensure that the data from the \"AUX output\" events isn\u0027t lost while their\n corresponding AUX event is not scheduled. We use grouping to provide this\n guarantee: an \"AUX output\" event can be added to a group where an AUX event\n is a group leader, and provided that the former supports writing to the\n latter.\n \n Signed-off-by: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: kan.liang@linux.intel.com\n Link: https://lkml.kernel.org/r/20190806084606.4021-2-alexander.shishkin@linux.intel.com\n\ncommit a3d8c0d13bdedf84fe74259f6949b2cdffd80e55\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Tue Aug 27 21:48:25 2019 +0200\n\n x86/intel: Add common OPTDIFFs\n \n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Reviewed-by: Tony Luck \u003ctony.luck@intel.com\u003e\n Cc: x86@kernel.org\n Cc: Dave Hansen \u003cdave.hansen@intel.com\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Borislav Petkov \u003cbp@alien8.de\u003e\n Link: https://lkml.kernel.org/r/20190827195122.731530141@infradead.org\n\ncommit 5ebb34edbefa8ea6a7e109179d5fc7b3529dbeba\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Tue Aug 27 21:48:24 2019 +0200\n\n x86/intel: Aggregate microserver naming\n \n Currently big microservers have _XEON_D while small microservers have\n _X, Make it uniformly: _D.\n \n for i in `git grep -l \"\\(INTEL_FAM6_\\|VULNWL_INTEL\\|INTEL_CPU_FAM6\\).*_\\(X\\|XEON_D\\)\"`\n do\n sed -i -e \u0027s/\\(\\(INTEL_FAM6_\\|VULNWL_INTEL\\|INTEL_CPU_FAM6\\).*ATOM.*\\)_X/\\1_D/g\u0027 \\\n -e \u0027s/\\(\\(INTEL_FAM6_\\|VULNWL_INTEL\\|INTEL_CPU_FAM6\\).*\\)_XEON_D/\\1_D/g\u0027 ${i}\n done\n \n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Reviewed-by: Tony Luck \u003ctony.luck@intel.com\u003e\n Cc: x86@kernel.org\n Cc: Dave Hansen \u003cdave.hansen@intel.com\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Borislav Petkov \u003cbp@alien8.de\u003e\n Link: https://lkml.kernel.org/r/20190827195122.677152989@infradead.org\n\ncommit 5e741407eab7c602ee5a2b06afb0070a02f4412f\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Tue Aug 27 21:48:23 2019 +0200\n\n x86/intel: Aggregate big core graphics naming\n \n Currently big core clients with extra graphics on have:\n \n - _G\n - _GT3E\n \n Make it uniformly: _G\n \n for i in `git grep -l \"\\(INTEL_FAM6_\\|VULNWL_INTEL\\|INTEL_CPU_FAM6\\).*_GT3E\"`\n do\n sed -i -e \u0027s/\\(\\(INTEL_FAM6_\\|VULNWL_INTEL\\|INTEL_CPU_FAM6\\).*\\)_GT3E/\\1_G/g\u0027 ${i}\n done\n \n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Reviewed-by: Tony Luck \u003ctony.luck@intel.com\u003e\n Cc: x86@kernel.org\n Cc: Dave Hansen \u003cdave.hansen@intel.com\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Borislav Petkov \u003cbp@alien8.de\u003e\n Link: https://lkml.kernel.org/r/20190827195122.622802314@infradead.org\n\ncommit af239c44e3f976762e9bc052f0d5796b90ea530b\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Tue Aug 27 21:48:22 2019 +0200\n\n x86/intel: Aggregate big core mobile naming\n \n Currently big core mobile chips have either:\n \n - _L\n - _ULT\n - _MOBILE\n \n Make it uniformly: _L.\n \n for i in `git grep -l \"\\(INTEL_FAM6_\\|VULNWL_INTEL\\|INTEL_CPU_FAM6\\).*_\\(MOBILE\\|ULT\\)\"`\n do\n sed -i -e \u0027s/\\(\\(INTEL_FAM6_\\|VULNWL_INTEL\\|INTEL_CPU_FAM6\\).*\\)_\\(MOBILE\\|ULT\\)/\\1_L/g\u0027 ${i}\n done\n \n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Reviewed-by: Tony Luck \u003ctony.luck@intel.com\u003e\n Cc: x86@kernel.org\n Cc: Dave Hansen \u003cdave.hansen@intel.com\u003e\n Cc: Borislav Petkov \u003cbp@alien8.de\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190827195122.568978530@infradead.org\n\ncommit c66f78a6de4de6cb520b15cf6a1b586617b9add5\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Tue Aug 27 21:48:21 2019 +0200\n\n x86/intel: Aggregate big core client naming\n \n Currently the big core client models either have:\n \n - no OPTDIFF\n - _CORE\n - _DESKTOP\n \n Make it uniformly: \u0027no OPTDIFF\u0027.\n \n for i in `git grep -l \"\\(INTEL_FAM6_\\|VULNWL_INTEL\\|INTEL_CPU_FAM6\\).*_\\(CORE\\|DESKTOP\\)\"`\n do\n sed -i -e \u0027s/\\(\\(INTEL_FAM6_\\|VULNWL_INTEL\\|INTEL_CPU_FAM6\\).*\\)_\\(CORE\\|DESKTOP\\)/\\1/g\u0027 ${i}\n done\n \n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Reviewed-by: Tony Luck \u003ctony.luck@intel.com\u003e\n Cc: x86@kernel.org\n Cc: Dave Hansen \u003cdave.hansen@intel.com\u003e\n Cc: Borislav Petkov \u003cbp@alien8.de\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190827195122.513945586@infradead.org\n\ncommit 77c84dd1881d0f0176cb678d770bfbda26c54390\nAuthor: Douglas RAILLARD \u003cdouglas.raillard@arm.com\u003e\nDate: Wed Aug 7 16:33:40 2019 +0100\n\n sched/cpufreq: Align trace event behavior of fast switching\n \n Fast switching path only emits an event for the CPU of interest, whereas the\n regular path emits an event for all the CPUs that had their frequency changed,\n i.e. all the CPUs sharing the same policy.\n \n With the current behavior, looking at cpu_frequency event for a given CPU that\n is using the fast switching path will not give the correct frequency signal.\n \n Signed-off-by: Douglas RAILLARD \u003cdouglas.raillard@arm.com\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit d15ce412737accaba5e4c7d653b184772da47365\nAuthor: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\nDate: Wed Aug 28 14:20:13 2019 +0530\n\n ACPI: cpufreq: Switch to QoS requests instead of cpufreq notifier\n \n The cpufreq core now takes the min/max frequency constraints via QoS\n requests and the CPUFREQ_ADJUST notifier shall get removed later on.\n \n Switch over to using the QoS request for maximum frequency constraint\n for acpi driver.\n \n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 4c4cdc4c63853fee48c02e25c8605fb65a6c9924\nAuthor: Al Stone \u003cahs3@redhat.com\u003e\nDate: Tue Aug 27 18:21:20 2019 -0600\n\n ACPI / CPPC: do not require the _PSD method\n \n According to the ACPI 6.3 specification, the _PSD method is optional\n when using CPPC. The underlying assumption is that each CPU can change\n frequency independently from all other CPUs; _PSD is provided to tell\n the OS that some processors can NOT do that.\n \n However, the acpi_get_psd() function returns ENODEV if there is no _PSD\n method present, or an ACPI error status if an error occurs when evaluating\n _PSD, if present. This makes _PSD mandatory when using CPPC, in violation\n of the specification, and only on Linux.\n \n This has forced some firmware writers to provide a dummy _PSD, even though\n it is irrelevant, but only because Linux requires it; other OSPMs follow\n the spec. We really do not want to have OS specific ACPI tables, though.\n \n So, correct acpi_get_psd() so that it does not return an error if there\n is no _PSD method present, but does return a failure when the method can\n not be executed properly. This allows _PSD to be optional as it should\n be.\n \n Signed-off-by: Al Stone \u003cahs3@redhat.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit bac7b4e843232a3a49a042410cf743341eb0887e\nAuthor: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nDate: Wed Aug 28 10:03:50 2019 +0200\n\n x86/vmware: Update platform detection code for VMCALL/VMMCALL hypercalls\n \n Vmware has historically used an INL instruction for this, but recent\n hardware versions support using VMCALL/VMMCALL instead, so use this\n method if supported at platform detection time. Explicitly code separate\n macro versions since the alternatives self-patching has not been\n performed at platform detection time.\n \n Also put tighter constraints on the assembly input parameters.\n \n Co-developed-by: Doug Covelli \u003cdcovelli@vmware.com\u003e\n Signed-off-by: Doug Covelli \u003cdcovelli@vmware.com\u003e\n Signed-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Reviewed-by: Doug Covelli \u003cdcovelli@vmware.com\u003e\n Cc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: linux-graphics-maintainer@vmware.com\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: virtualization@lists.linux-foundation.org\n Cc: \u003cpv-drivers@vmware.com\u003e\n Cc: x86-ml \u003cx86@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190828080353.12658-2-thomas_os@shipmail.org\n\ncommit 9aee3713135a6733f6e58ec84c1ce24514579039\nAuthor: Nathan Huckleberry \u003cnhuck@google.com\u003e\nDate: Thu Jun 13 11:49:23 2019 -0700\n\n thermal: armada: Fix -Wshift-negative-value\n \n Clang produces the following warning\n \n drivers/thermal/armada_thermal.c:270:33: warning: shifting a negative\n signed value is undefined [-Wshift-negative-value]\n 1 warning reg \u0026\u003d ~CONTROL1_TSEN_AVG_MASK \u003c\u003c\n CONTROL1_TSEN_AVG_SHIFT; generated\n .\n ~~~~~~~~~~~~~~~~~~~~~~~ ^\n \n CONTROL1_TSEN_AVG_SHIFT is defined to be zero.\n Since shifting by zero does nothing this variable can be removed.\n \n Cc: clang-built-linux@googlegroups.com\n Link: https://github.com/ClangBuiltLinux/linux/issues/532\n Signed-off-by: Nathan Huckleberry \u003cnhuck@google.com\u003e\n Reviewed-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Acked-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Zhang Rui \u003crui.zhang@intel.com\u003e\n\ncommit 11f787b0840e26648b5e01fad99bd2aa35098d5d\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Tue Jul 30 10:21:26 2019 +0800\n\n dt-bindings: thermal: qoriq: Add optional clocks property\n \n Some platforms like i.MX8M series SoCs have clock control for TMU,\n add optional clocks property to the binding doc.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Signed-off-by: Zhang Rui \u003crui.zhang@intel.com\u003e\n\ncommit aea591970f659bd6d792ca4c46dd0d251331a397\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Tue Jul 30 10:21:25 2019 +0800\n\n thermal: qoriq: Use __maybe_unused instead of #if CONFIG_PM_SLEEP\n \n Use __maybe_unused for power management related functions\n instead of #if CONFIG_PM_SLEEP to simply the code.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Signed-off-by: Zhang Rui \u003crui.zhang@intel.com\u003e\n\ncommit 4d82000af007acda6b46729ea368f6b2823c532c\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Tue Jul 30 10:21:24 2019 +0800\n\n thermal: qoriq: Use devm_platform_ioremap_resource() instead of of_iomap()\n \n Use devm_platform_ioremap_resource() instead of of_iomap() to\n save the iounmap() call in error handle path;\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Signed-off-by: Zhang Rui \u003crui.zhang@intel.com\u003e\n\ncommit 11f0cdc8bd621fcded06c951b05076e618c5c717\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Tue Jul 30 10:21:23 2019 +0800\n\n thermal: qoriq: Fix error path of calling qoriq_tmu_register_tmu_zone fail\n \n When registering tmu zone failed, the error path should be err_tmu\n instead of err_iomap, as iounmap() needs to be called.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Signed-off-by: Zhang Rui \u003crui.zhang@intel.com\u003e\n\ncommit 51904045d4aa07fbccf6ff18d56d2064a7676d35\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Tue Jul 30 10:21:22 2019 +0800\n\n thermal: qoriq: Add clock operations\n \n Some platforms like i.MX8MQ has clock control for this module,\n need to add clock operations to make sure the driver is working\n properly.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n Reviewed-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Signed-off-by: Zhang Rui \u003crui.zhang@intel.com\u003e\n\ncommit c4e6488120e9ef1ceced8fb0caf134c5242ddaf4\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Aug 27 17:17:26 2019 +0100\n\n drm/i915/selftests: Try to recycle context allocations\n \n igt_ctx_exec allocates a new context for each iteration, keeping them\n all allocated until the end. Instead, release the local ctx reference at\n the end of each iteration, allowing ourselves to reap those if under\n mempressure.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190827161726.3640-2-chris@chris-wilson.co.uk\n\ncommit f2085c8e950d536c7982182b0c9c015804011dd2\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Aug 27 17:17:25 2019 +0100\n\n drm/i915/selftests: Remove accidental serialization between gpu_fill\n \n Upon object creation for live_gem_contexts, we fill the object with\n known scratch and flush it out of the CPU cache. Before performing the\n GPU fill, we don\u0027t need to flush it again and so avoid serialising with\n previous fills.\n \n However, we do need some throttling on the internal interfaces if we do\n not want to run out of memory!\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190827161726.3640-1-chris@chris-wilson.co.uk\n\ncommit 00679b631eddaa0aa0ceba719fcb1f60c65da5a3\nAuthor: Michael Guralnik \u003cmichaelgur@mellanox.com\u003e\nDate: Mon Aug 19 15:08:13 2019 +0300\n\n net/mlx5: Set ODP capabilities for DC transport to max\n \n In mlx5_core initialization, query max ODP capabilities for DC transport\n from FW and set as current capabilities.\n \n Signed-off-by: Michael Guralnik \u003cmichaelgur@mellanox.com\u003e\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n\ncommit cbb1133b563a63901cf778220eb17e3ff1425aed\nAuthor: Cao Jin \u003ccaoj.fnst@cn.fujitsu.com\u003e\nDate: Wed Aug 28 14:11:00 2019 +0800\n\n x86/cpufeature: Explain the macro duplication\n \n Explain the intent behind the duplication of the\n \n BUILD_BUG_ON_ZERO(NCAPINTS !\u003d n)\n \n check in *_MASK_CHECK and its immediate use in the *MASK_BIT_SET macros\n too.\n \n [ bp: Massage. ]\n \n Suggested-by: Borislav Petkov \u003cbp@alien8.de\u003e\n Signed-off-by: Cao Jin \u003ccaoj.fnst@cn.fujitsu.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Cc: Dave Hansen \u003cdave.hansen@intel.com\u003e\n Cc: Fenghua Yu \u003cfenghua.yu@intel.com\u003e\n Cc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: Jann Horn \u003cjannh@google.com\u003e\n Cc: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Cc: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Cc: Nadav Amit \u003cnamit@vmware.com\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: x86-ml \u003cx86@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190828061100.27032-1-caoj.fnst@cn.fujitsu.com\n\ncommit 0067e154b11e236d62a7a8205f321b097c21a35b\nAuthor: Jussi Laako \u003cjussi@sonarnerd.net\u003e\nDate: Wed Aug 28 00:08:47 2019 +0300\n\n ALSA: usb-audio: Update DSD support quirks for Oppo and Rotel\n \n Oppo has issued firmware updates that change alt setting used for DSD\n support. However, these devices seem to support auto-detection, so\n support is moved from explicit whitelisting to auto-detection.\n \n Also Rotel devices have USB interfaces that support DSD with\n auto-detection.\n \n Signed-off-by: Jussi Laako \u003cjussi@sonarnerd.net\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit eb7505d52a2f8b0cfc3fd7146d8cb2dab5a73f0d\nAuthor: Jussi Laako \u003cjussi@sonarnerd.net\u003e\nDate: Wed Aug 28 00:08:46 2019 +0300\n\n ALSA: usb-audio: DSD auto-detection for Playback Designs\n \n Add DSD support auto-detection for newer Playback Designs devices. Older\n device generations have a different USB interface implementation.\n \n Keep the auto-detection VID whitelist sorted.\n \n Signed-off-by: Jussi Laako \u003cjussi@sonarnerd.net\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 190f73ab4c43ecfc8e93843fe249efeff7d69a90\nAuthor: Voon Weifeng \u003cweifeng.voon@intel.com\u003e\nDate: Tue Aug 27 09:38:11 2019 +0800\n\n net: stmmac: setup higher frequency clk support for EHL \u0026 TGL\n \n EHL DW EQOS is running on a 200MHz clock. Setting up stmmac-clk,\n ptp clock and ptp_max_adj to 200MHz.\n \n Signed-off-by: Voon Weifeng \u003cweifeng.voon@intel.com\u003e\n Signed-off-by: Ong Boon Leong \u003cboon.leong.ong@intel.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f6256585fecc9b9d2f0a335a92e864ccae98ea24\nAuthor: Voon Weifeng \u003cweifeng.voon@intel.com\u003e\nDate: Tue Aug 27 09:38:10 2019 +0800\n\n net: stmmac: add EHL RGMII 1Gbps PCI info and PCI ID\n \n Added EHL RGMII 1Gbps PCI ID. Different MII and speed will have\n different PCI ID.\n \n Signed-off-by: Voon Weifeng \u003cweifeng.voon@intel.com\u003e\n Signed-off-by: Ong Boon Leong \u003cboon.leong.ong@intel.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit e125dcef755612efa4f057dd7955cd8edd3fadb1\nAuthor: Voon Weifeng \u003cweifeng.voon@intel.com\u003e\nDate: Tue Aug 27 09:38:09 2019 +0800\n\n net: stmmac: add TGL SGMII 1Gbps PCI info and PCI ID\n \n Added TGL SGMII 1Gbps PCI ID. Different MII and speed will have\n different PCI ID.\n \n Signed-off-by: Voon Weifeng \u003cweifeng.voon@intel.com\u003e\n Signed-off-by: Ong Boon Leong \u003cboon.leong.ong@intel.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 99122836d26a366847f92f9a211602a44a038c61\nAuthor: Voon Weifeng \u003cweifeng.voon@intel.com\u003e\nDate: Tue Aug 27 09:38:08 2019 +0800\n\n net: stmmac: add EHL SGMII 1Gbps PCI info and PCI ID\n \n Added EHL SGMII 1Gbps PCI ID. Different MII and speed will have\n different PCI ID.\n \n Signed-off-by: Voon Weifeng \u003cweifeng.voon@intel.com\u003e\n Signed-off-by: Ong Boon Leong \u003cboon.leong.ong@intel.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7a3007d22e8dc7d0c14f711c5d370dc41226ac55\nAuthor: Marek Behún \u003cmarek.behun@nic.cz\u003e\nDate: Mon Aug 26 23:31:55 2019 +0200\n\n net: dsa: mv88e6xxx: fully support SERDES on Topaz family\n \n Currently we support SERDES on the Topaz family in a limited way: no\n IRQs and the cmode is not writable, thus the mode is determined by\n strapping pins.\n \n Marvell\u0027s examples though show how to make cmode writable on port 5 and\n support SGMII autonegotiation. It is done by writing hidden registers,\n for which we already have code.\n \n This patch adds support for making the cmode for the SERDES port\n writable on the Topaz family, via a new chip operation,\n .port_set_cmode_writable, which is called from mv88e6xxx_port_setup_mac\n just before .port_set_cmode.\n \n SERDES IRQs are also enabled for Topaz.\n \n Tested on Turris Mox.\n \n Signed-off-by: Marek Behún \u003cmarek.behun@nic.cz\u003e\n Reviewed-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 3bbb8867f87d915312e7904b4a670dbaf062fb80\nAuthor: Marek Behún \u003cmarek.behun@nic.cz\u003e\nDate: Mon Aug 26 23:31:54 2019 +0200\n\n net: dsa: mv88e6xxx: rename port cmode macro\n \n This is a cosmetic update. We are removing the last underscore from\n macros MV88E6XXX_PORT_STS_CMODE_100BASE_X and\n MV88E6XXX_PORT_STS_CMODE_1000BASE_X. The 2500base-x version does not\n have that underscore. Also PHY_INTERFACE_MODE_ macros do not have it\n there.\n \n Signed-off-by: Marek Behún \u003cmarek.behun@nic.cz\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d3cf7d8f20b49331e198172e8c6b6ebadc72fa9c\nAuthor: Marek Behún \u003cmarek.behun@nic.cz\u003e\nDate: Mon Aug 26 23:31:53 2019 +0200\n\n net: dsa: mv88e6xxx: simplify SERDES code for Topaz and Peridot\n \n By adding an additional serdes_get_lane implementation (for Topaz), we\n can merge the implementations of other SERDES functions (powering and\n IRQs). We can skip checking port numbers, since the serdes_get_lane()\n methods inform if there is no lane on a port or if the lane cannot be\n used for given cmode.\n \n Signed-off-by: Marek Behún \u003cmarek.behun@nic.cz\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 17deaf5cb37a365fd65aaa0e14d1a2e1a50418c7\nAuthor: Marek Behún \u003cmarek.behun@nic.cz\u003e\nDate: Mon Aug 26 23:31:52 2019 +0200\n\n net: dsa: mv88e6xxx: create serdes_get_lane chip operation\n \n Create a serdes_get_lane() method in the mv88e6xxx operations structure.\n Use it instead of calling the different implementations.\n Also change the methods so that their return value is used only for\n error. The lane number is put into a place referred to by a pointer\n given as argument. If the port does not have a lane, return -ENODEV.\n Lanes are phy addresses, so use u8 as their type.\n \n Signed-off-by: Marek Behún \u003cmarek.behun@nic.cz\u003e\n Reviewed-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 609070133aff1f2b533561758aa97c947fa534c8\nAuthor: Marek Behún \u003cmarek.behun@nic.cz\u003e\nDate: Mon Aug 26 23:31:51 2019 +0200\n\n net: dsa: mv88e6xxx: update code operating on hidden registers\n \n This patch moves the functions operating on the hidden debug registers\n into it\u0027s own file, port_hidden.c. The functions prefix is renamed from\n mv88e6390_hidden_ to mv88e6xxx_port_hidden_, to be consistent with the\n rest of this driver. The macros are prefixed with MV88E6XXX_ prefix, and\n are changed not to use the BIT() macro nor bit shifts, since the rest of\n the port.h file does not use it.\n \n We also add the support for setting the Block Address field when\n operating hidden registers. Marvell\u0027s mdio examples for SERDES settings\n on Topaz use Block Address 0x7 when reading/writing hidden registers,\n and although the specification says that block must be set to 0xf, those\n settings are reachable only with Block Address 0x7.\n \n Signed-off-by: Marek Behún \u003cmarek.behun@nic.cz\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Tested-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4e6da7969ffecf52a974b165975eb6a3149eeb73\nAuthor: Marek Behún \u003cmarek.behun@nic.cz\u003e\nDate: Mon Aug 26 23:31:50 2019 +0200\n\n net: dsa: mv88e6xxx: support 2500base-x in SGMII IRQ handler\n \n The mv88e6390_serdes_irq_link_sgmii IRQ handler reads the SERDES PHY\n status register to determine speed, among other things. If cmode of the\n port is set to 2500base-x, though, the PHY still reports 1000 Mbps (the\n PHY register itself does not differentiate between 1000 Mbps and 2500\n Mbps - it thinks it is running at 1000 Mbps, although clock is 2.5x\n faster).\n Look at the cmode and set SPEED_2500 if cmode is set to 2500base-x.\n Also tell mv88e6xxx_port_setup_mac the PHY interface mode corresponding\n to current cmode in terms of phy_interface_t.\n \n Signed-off-by: Marek Behún \u003cmarek.behun@nic.cz\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Tested-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d5886b919a720ff859aebf569cb0f353b1d977a6\nAuthor: Xin Long \u003clucien.xin@gmail.com\u003e\nDate: Mon Aug 26 16:30:04 2019 +0800\n\n sctp: allow users to set ep ecn flag by sockopt\n \n SCTP_ECN_SUPPORTED sockopt will be added to allow users to change\n ep ecn flag, and it\u0027s similar with other feature flags.\n \n Signed-off-by: Xin Long \u003clucien.xin@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 2f5268a9249b87d49ab3c7050f99064e6d1a8bb0\nAuthor: Xin Long \u003clucien.xin@gmail.com\u003e\nDate: Mon Aug 26 16:30:03 2019 +0800\n\n sctp: allow users to set netns ecn flag with sysctl\n \n sysctl net.sctp.ecn_enable is added in this patch. It will allow\n users to change the default sctp ecn flag, net.sctp.ecn_enable.\n \n This feature was also required on this thread:\n \n http://lkml.iu.edu/hypermail/linux/kernel/0812.1/01858.html\n \n Signed-off-by: Xin Long \u003clucien.xin@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1b0b8114b9549dee6490e728cd787f808b586158\nAuthor: Xin Long \u003clucien.xin@gmail.com\u003e\nDate: Mon Aug 26 16:30:02 2019 +0800\n\n sctp: make ecn flag per netns and endpoint\n \n This patch is to add ecn flag for both netns_sctp and sctp_endpoint,\n net-\u003esctp.ecn_enable is set 1 by default, and ep-\u003eecn_enable will\n be initialized with net-\u003esctp.ecn_enable.\n \n asoc-\u003epeer.ecn_capable will be set during negotiation only when\n ep-\u003eecn_enable is set on both sides.\n \n Signed-off-by: Xin Long \u003clucien.xin@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7add83d93a94aaac79645414fd57d3c4b707d3be\nAuthor: Mao Wenan \u003cmaowenan@huawei.com\u003e\nDate: Mon Aug 26 09:31:18 2019 +0800\n\n net: mediatek: remove set but not used variable \u0027status\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n drivers/net/ethernet/mediatek/mtk_eth_soc.c: In function mtk_handle_irq:\n drivers/net/ethernet/mediatek/mtk_eth_soc.c:1951:6: warning: variable status set but not used [-Wunused-but-set-variable]\n \n Fixes: 296c9120752b (\"net: ethernet: mediatek: Add MT7628/88 SoC support\")\n Signed-off-by: Mao Wenan \u003cmaowenan@huawei.com\u003e\n Reviewed-by: Stefan Roese \u003csr@denx.de\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit e9bf96943b408e6c99dd13fb01cb907335787c61\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Sun Aug 25 22:46:30 2019 +0300\n\n net: dsa: sja1105: Clear VLAN filtering offload netdev feature\n \n The switch barely supports traffic I/O, and it does that by repurposing\n VLANs when there is no bridge that is taking control of them.\n \n Letting DSA declare this netdev feature as supported (see\n dsa_slave_create) would mean that VLAN sub-interfaces created on sja1105\n switch ports will be hardware offloaded. That means that\n net/8021q/vlan_core.c would install the VLAN into the filter tables of\n the switch, potentially interfering with the tag_8021q VLANs.\n \n We need to prevent that from happening and not let the 8021q core\n offload VLANs to the switch hardware tables. In vlan_filtering\u003d0 modes\n of operation, the switch ports can pass through VLAN-tagged frames with\n no problem.\n \n Suggested-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9b236d2a69da41da4aef3531cc3a034bf9dfc26d\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Sun Aug 25 22:46:29 2019 +0300\n\n net: dsa: Advertise the VLAN offload netdev ability only if switch supports it\n \n When adding a VLAN sub-interface on a DSA slave port, the 8021q core\n checks NETIF_F_HW_VLAN_CTAG_FILTER and, if the netdev is capable of\n filtering, calls .ndo_vlan_rx_add_vid or .ndo_vlan_rx_kill_vid to\n configure the VLAN offloading.\n \n DSA sets this up counter-intuitively: it always advertises this netdev\n feature, but the underlying driver may not actually support VLAN table\n manipulation. In that case, the DSA core is forced to ignore the error,\n because not being able to offload the VLAN is still fine - and should\n result in the creation of a non-accelerated VLAN sub-interface.\n \n Change this so that the netdev feature is only advertised for switch\n drivers that support VLAN manipulation, instead of checking for\n -EOPNOTSUPP at runtime.\n \n Suggested-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit bd69baaace06e659299dc3cacc6eaa26102ca3fc\nAuthor: René van Dorst \u003copensource@vdorst.com\u003e\nDate: Sun Aug 25 19:43:41 2019 +0200\n\n dt-bindings: net: ethernet: Update mt7622 docs and dts to reflect the new phylink API\n \n This patch the removes the recently added mediatek,physpeed property.\n Use the fixed-link property speed \u003d \u003c2500\u003e to set the phy in 2.5Gbit.\n See mt7622-bananapi-bpi-r64.dts for a working example.\n \n Signed-off-by: René van Dorst \u003copensource@vdorst.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7e538372694b3e449783eed3981d59d2597c2882\nAuthor: René van Dorst \u003copensource@vdorst.com\u003e\nDate: Sun Aug 25 19:43:40 2019 +0200\n\n net: ethernet: mediatek: Re-add support SGMII\n \n * Re-add SGMII support but now with PHYLINK API support\n So the SGMII changes are more clear\n * Move SGMII block setup from mtk_gmac_sgmii_path_setup() to\n mtk_mac_config()\n * Merge mtk_setup_hw_path() into mtk_mac_config()\n * Remove mediatek,physpeed property, fixed-link supports now any speed so\n speed \u003d \u003c2500\u003e; is now valid with PHYLINK\n * Demagic SGMII register values\n * Use phylink state to setup fixed-link mode\n \n Signed-off-by: René van Dorst \u003copensource@vdorst.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b8fc9f30821ec0ca018b3716fbed55ab6910a8f1\nAuthor: René van Dorst \u003copensource@vdorst.com\u003e\nDate: Sun Aug 25 19:43:39 2019 +0200\n\n net: ethernet: mediatek: Add basic PHYLINK support\n \n This convert the basics to PHYLINK API.\n SGMII support is not in this patch.\n \n Signed-off-by: René van Dorst \u003copensource@vdorst.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b9499904f363342f56fd2039605aac9f7388e795\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Sun Aug 25 13:25:20 2019 -0400\n\n net: dsa: clear VLAN PVID flag for CPU port\n \n When the bridge offloads a VLAN on a slave port, we also need to\n program its dedicated CPU port as a member of the VLAN.\n \n Drivers may handle the CPU port\u0027s membership as they want. For example,\n Marvell as a special \"Unmodified\" mode to pass frames as is through\n such ports.\n \n Even though DSA expects the drivers to handle the CPU port membership,\n it does not make sense to program user VLANs as PVID on the CPU port.\n This patch clears this flag before programming the CPU port.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Suggested-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7e1741b47f2441e7c570284d60ba936e3d41529a\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Sun Aug 25 13:25:19 2019 -0400\n\n net: dsa: program VLAN on CPU port from slave\n \n DSA currently programs a VLAN on the CPU port implicitly after the\n related notifier is received by a switch.\n \n While we still need to do this transparent programmation of the DSA\n links in the fabric, programming the CPU port this way may cause\n problems in some corners such as the tag_8021q driver.\n \n Because the dedicated CPU port is specific to a slave, make their\n programmation explicit a few layers up, in the slave code.\n \n Note that technically, DSA links have a dedicated CPU port as well,\n but since they are only used as conduit between interconnected switches\n of a fabric, programming them transparently this way is what we want.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c5335d737ff30f1cb23d245ef9e20ec23cc2d7ba\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Sun Aug 25 13:25:18 2019 -0400\n\n net: dsa: check bridge VLAN in slave operations\n \n The bridge VLANs are not offloaded by dsa_port_vlan_* if the port is\n not bridged or if its bridge is not VLAN aware.\n \n This is a good thing but other corners of DSA, such as the tag_8021q\n driver, may need to program VLANs regardless the bridge state.\n \n And also because bridge_dev is specific to user ports anyway, move\n these checks were it belongs, one layer up in the slave code.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Suggested-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit bdcff080f7ae1932893436f7891528aa8fc59cb9\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Sun Aug 25 13:25:17 2019 -0400\n\n net: dsa: add slave VLAN helpers\n \n Add dsa_slave_vlan_add and dsa_slave_vlan_del helpers to handle\n SWITCHDEV_OBJ_ID_PORT_VLAN switchdev objects. Also copy the\n switchdev_obj_port_vlan structure on add since we will modify it in\n future patches.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit cf360866b11901eca5317e8a5c26c3e15b7bf1f2\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Sun Aug 25 13:25:16 2019 -0400\n\n net: dsa: do not skip -EOPNOTSUPP in dsa_port_vid_add\n \n Currently dsa_port_vid_add returns 0 if the switch returns -EOPNOTSUPP.\n \n This function is used in the tag_8021q.c code to offload the PVID of\n ports, which would simply not work if .port_vlan_add is not supported\n by the underlying switch.\n \n Do not skip -EOPNOTSUPP in dsa_port_vid_add but only when necessary,\n that is to say in dsa_slave_vlan_rx_add_vid.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit e65d45cc351ac5f1c11e6bac5669e536df753664\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Sun Aug 25 13:25:15 2019 -0400\n\n net: dsa: remove bitmap operations\n \n The bitmap operations were introduced to simplify the switch drivers\n in the future, since most of them could implement the common VLAN and\n MDB operations (add, del, dump) with simple functions taking all target\n ports at once, and thus limiting the number of hardware accesses.\n \n Programming an MDB or VLAN this way in a single operation would clearly\n simplify the drivers a lot but would require a new get-set interface\n in DSA. The usage of such bitmap from the stack also raised concerned\n in the past, leading to the dynamic allocation of a new ds-\u003e_bitmap\n member in the dsa_switch structure. So let\u0027s get rid of them for now.\n \n This commit nicely wraps the ds-\u003eops-\u003eport_{mdb,vlan}_{prepare,add}\n switch operations into new dsa_switch_{mdb,vlan}_{prepare,add}\n variants not using any bitmap argument anymore.\n \n New dsa_switch_{mdb,vlan}_match helpers have been introduced to make\n clear which local port of a switch must be programmed with the target\n object. While the targeted user port is an obvious candidate, the\n DSA links must also be programmed, as well as the CPU port for VLANs.\n \n While at it, also remove local variables that are only used once.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 05559f10ed797b79f7fa47313682c48919a2b111\nAuthor: Li Jun \u003cjun.li@nxp.com\u003e\nDate: Mon Aug 26 18:25:12 2019 +0800\n\n usb: chipidea: add role switch class support\n \n USB role is fully controlled by usb role switch consumer(e.g. typec),\n usb port can be at host mode(USB_ROLE_HOST), device mode connected to\n host(USB_ROLE_DEVICE), or not connecting any partner(USB_ROLE_NONE).\n \n Signed-off-by: Li Jun \u003cjun.li@nxp.com\u003e\n Signed-off-by: Peter Chen \u003cpeter.chen@nxp.com\u003e\n\ncommit 71fcb8bdf5ab68e91c05b69e80c2d0ab4c51d78b\nAuthor: Peter Chen \u003cpeter.chen@nxp.com\u003e\nDate: Mon Jan 28 15:37:30 2019 +0800\n\n dt-binding: usb: usbmisc-imx: add imx7ulp compatible\n \n Add imx7ulp compatible\n \n Signed-off-by: Peter Chen \u003cpeter.chen@nxp.com\u003e\n\ncommit 58fb8beda201f00b139e8993cbda937cd9a4e603\nAuthor: Peter Chen \u003cpeter.chen@nxp.com\u003e\nDate: Mon Jan 28 15:36:09 2019 +0800\n\n dt-binding: usb: ci-hdrc-usb2: add imx7ulp compatible\n \n Add imx7ulp and imx8qm compatible\n \n Signed-off-by: Peter Chen \u003cpeter.chen@nxp.com\u003e\n\ncommit 3bbd2343734e44de92238eea1a5cd3ad32a6baf0\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Wed Aug 21 10:20:51 2019 +0000\n\n powerpc/8xx: set STACK_END_MAGIC earlier on the init_stack\n \n Today, the STACK_END_MAGIC is set on init_stack in start_kernel().\n \n To avoid a false \u0027Thread overran stack, or stack corrupted\u0027 message\n on early Oopses, setup STACK_END_MAGIC as soon as possible.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/54f67bb7ac486c1350f2fa8905cd279f94b9dfb1.1566382841.git.christophe.leroy@c-s.fr\n\ncommit a04565741284f695db4cfe5a3e61940d2259cb8f\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Wed Aug 21 20:00:34 2019 +0000\n\n powerpc/8xx: drop unused self-modifying code alternative to FixupDAR.\n \n The code which fixups the DAR on TLB errors for dbcX instructions\n has a self-modifying code alternative that has never been used.\n \n Drop it.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Reviewed-by: Joakim Tjernlund \u003cjoakim.tjernlund@infinera.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/b095e12c82fcba1ac4c09fc3b85d969f36614746.1566417610.git.christophe.leroy@c-s.fr\n\ncommit 63ce271b5e377deaddace4bac6dafb6e79d2bee4\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Mon Aug 26 11:10:23 2019 +0000\n\n powerpc/prom: convert PROM_BUG() to standard trap\n \n Prior to commit 1bd98d7fbaf5 (\"ppc64: Update BUG handling based on\n ppc32\"), BUG() family was using BUG_ILLEGAL_INSTRUCTION which\n was an invalid instruction opcode to trap into program check\n exception.\n \n That commit converted them to using standard trap instructions,\n but prom/prom_init and their PROM_BUG() macro were left over.\n head_64.S and exception-64s.S were left aside as well.\n \n Convert them to using the standard BUG infrastructure.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/cdaf4bbbb64c288a077845846f04b12683f8875a.1566817807.git.christophe.leroy@c-s.fr\n\ncommit 6737cd20eec4b1ed0ecfc24931f71ed9f7501b02\nAuthor: Krzysztof Wilczynski \u003ckw@linux.com\u003e\nDate: Mon Aug 26 00:46:16 2019 +0200\n\n PCI: Add pci_info_ratelimited() to ratelimit PCI separately\n \n Do not use printk_ratelimit() in drivers/pci/pci.c as it shares the rate\n limiting state with all other callers to the printk_ratelimit().\n \n Add pci_info_ratelimited() (similar to pci_notice_ratelimited() added in\n the commit a88a7b3eb076 (\"vfio: Use dev_printk() when possible\")) and use\n it instead of printk_ratelimit() + pci_info().\n \n Link: https://lore.kernel.org/r/20190825224616.8021-1-kw@linux.com\n Signed-off-by: Krzysztof Wilczynski \u003ckw@linux.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit 47ee6e86e0a3e3a15fbdd6d94aab39e46013c961\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Mon Aug 26 15:27:12 2019 -0700\n\n selftests/bpf: remove wrong nhoff in flow dissector test\n \n .nhoff \u003d 0 is (correctly) reset to ETH_HLEN on the next line so let\u0027s\n drop it.\n \n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 86ccc384cfcac22f45b1158b57d9ef5b4079e027\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Wed Aug 21 16:44:27 2019 -0700\n\n selftests/bpf: test_progs: remove unused ret\n \n send_signal test returns static codes from the subtests which\n nobody looks at, let\u0027s rely on the CHECK macros instead.\n \n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 62d69f24fe5eca23410b6a21334a7267b0c8838b\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Wed Aug 21 16:44:26 2019 -0700\n\n selftests/bpf: test_progs: remove asserts from subtests\n \n Otherwise they can bring the whole process down.\n \n Cc: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit d38835b75f67df16cef65c14aa64796a1832e6b4\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Wed Aug 21 16:44:25 2019 -0700\n\n selftests/bpf: test_progs: remove global fail/success counts\n \n Now that we have a global per-test/per-environment state, there\n is no longer need to have global fail/success counters (and there\n is no need to save/get the diff before/after the test).\n \n Introduce CHECK_FAIL macro (suggested by Andrii) and covert existing tests\n to it. CHECK_FAIL uses new test__fail() to record the failure.\n \n Cc: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit cd9c21d76879a06216597052d3b45c51be977aaa\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Wed Aug 21 16:44:24 2019 -0700\n\n selftests/bpf: test_progs: test__skip\n \n Export test__skip() to indicate skipped tests and use it in\n test_send_signal_nmi().\n \n Cc: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 411cdb4569013cc1089c00b2cd279033d6278f61\nAuthor: Alexei Starovoitov \u003cast@kernel.org\u003e\nDate: Thu Aug 22 22:52:15 2019 -0700\n\n selftests/bpf: add precision tracking test\n \n Copy-paste of existing test\n \"calls: cross frame pruning - liveness propagation\"\n but ran with different parentage chain heuristic\n which stresses different path in precision tracking logic.\n \n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit e8c13c4d9b36065903a025f163db87a7afff7307\nAuthor: Alexei Starovoitov \u003cast@kernel.org\u003e\nDate: Thu Aug 22 22:52:14 2019 -0700\n\n selftests/bpf: verifier precise tests\n \n Use BPF_F_TEST_STATE_FREQ flag to check that precision\n tracking works as expected by comparing every step it takes.\n \n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 0fc2e0b84ba725c5e6ee66059936638389e67c80\nAuthor: Alexei Starovoitov \u003cast@kernel.org\u003e\nDate: Thu Aug 22 22:52:13 2019 -0700\n\n tools/bpf: sync bpf.h\n \n sync bpf.h from kernel/ to tools/\n \n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 10d274e880eb208ec6a76261a9f8f8155020f771\nAuthor: Alexei Starovoitov \u003cast@kernel.org\u003e\nDate: Thu Aug 22 22:52:12 2019 -0700\n\n bpf: introduce verifier internal test flag\n \n Introduce BPF_F_TEST_STATE_FREQ flag to stress test parentage chain\n and state pruning.\n \n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit c18ce674d548c00faa6b7e760bacbaf1f39315f3\nAuthor: Ralph Campbell \u003crcampbell@nvidia.com\u003e\nDate: Fri Aug 23 15:17:53 2019 -0700\n\n mm/hmm: hmm_range_fault() infinite loop\n \n Normally, callers to handle_mm_fault() are supposed to check the\n vma-\u003evm_flags first. hmm_range_fault() checks for VM_READ but doesn\u0027t\n check for VM_WRITE if the caller requests a page to be faulted in with\n write permission (via the hmm_range.pfns[] value). If the vma is write\n protected, this can result in an infinite loop:\n \n hmm_range_fault()\n walk_page_range()\n ...\n hmm_vma_walk_hole()\n hmm_vma_walk_hole_()\n hmm_vma_do_fault()\n handle_mm_fault(FAULT_FLAG_WRITE)\n /* returns VM_FAULT_WRITE */\n /* returns -EBUSY */\n /* returns -EBUSY */\n /* returns -EBUSY */\n /* loops on -EBUSY and range-\u003evalid */\n \n Prevent this by checking for vma-\u003evm_flags \u0026 VM_WRITE before calling\n handle_mm_fault().\n \n Link: https://lore.kernel.org/r/20190823221753.2514-3-rcampbell@nvidia.com\n Signed-off-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 6c64f2bbe79cf3b770ac60ae79442322bd76d55e\nAuthor: Ralph Campbell \u003crcampbell@nvidia.com\u003e\nDate: Fri Aug 23 15:17:52 2019 -0700\n\n mm/hmm: hmm_range_fault() NULL pointer bug\n \n Although hmm_range_fault() calls find_vma() to make sure that a vma exists\n before calling walk_page_range(), hmm_vma_walk_hole() can still be called\n with walk-\u003evma \u003d\u003d NULL if the start and end address are not contained\n within the vma range.\n \n hmm_range_fault() /* calls find_vma() but no range check */\n walk_page_range() /* calls find_vma(), sets walk-\u003evma \u003d NULL */\n __walk_page_range()\n walk_pgd_range()\n walk_p4d_range()\n walk_pud_range()\n hmm_vma_walk_hole()\n hmm_vma_walk_hole_()\n hmm_vma_do_fault()\n handle_mm_fault(vma\u003d0)\n \n Link: https://lore.kernel.org/r/20190823221753.2514-2-rcampbell@nvidia.com\n Signed-off-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit de4064af76537f13d74a814a962f4524e81436ac\nAuthor: Suman Anna \u003cs-anna@ti.com\u003e\nDate: Tue Oct 23 20:19:09 2018 -0500\n\n rpmsg: virtio_rpmsg_bus: replace \"%p\" with \"%pK\"\n \n The virtio_rpmsg_bus driver uses the \"%p\" format-specifier for\n printing the vring buffer address. This prints only a hashed\n pointer even for previliged users. Use \"%pK\" instead so that\n the address can be printed during debug using kptr_restrict\n sysctl.\n \n Signed-off-by: Suman Anna \u003cs-anna@ti.com\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit bb487d29ce005416ed355dea213bbc39c03dabe7\nAuthor: Qian Cai \u003ccai@lca.pw\u003e\nDate: Fri Aug 23 15:56:23 2019 -0400\n\n net/mlx5: fix a -Wstringop-truncation warning\n \n In file included from ./arch/powerpc/include/asm/paca.h:15,\n from ./arch/powerpc/include/asm/current.h:13,\n from ./include/linux/thread_info.h:21,\n from ./include/asm-generic/preempt.h:5,\n from ./arch/powerpc/include/generated/asm/preempt.h:1,\n from ./include/linux/preempt.h:78,\n from ./include/linux/spinlock.h:51,\n from ./include/linux/wait.h:9,\n from ./include/linux/completion.h:12,\n from ./include/linux/mlx5/driver.h:37,\n from\n drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h:6,\n from\n drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c:33:\n In function \u0027strncpy\u0027,\n inlined from \u0027mlx5_fw_tracer_save_trace\u0027 at\n drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c:549:2,\n inlined from \u0027mlx5_tracer_print_trace\u0027 at\n drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c:574:2:\n ./include/linux/string.h:305:9: warning: \u0027__builtin_strncpy\u0027 output may\n be truncated copying 256 bytes from a string of length 511\n [-Wstringop-truncation]\n return __builtin_strncpy(p, q, size);\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n \n Fix it by using the new strscpy_pad() since the commit 458a3bf82df4\n (\"lib/string: Add strscpy_pad() function\") which will always\n NUL-terminate the string, and avoid possibly leak data through the ring\n buffer where non-admin account might enable these events through perf.\n \n Fixes: fd1483fe1f9f (\"net/mlx5: Add support for FW reporter dump\")\n Signed-off-by: Qian Cai \u003ccai@lca.pw\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 8a9a982767b7c89b2f23290e4f0f21f194b79dfe\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Tue Aug 27 11:58:05 2019 -0700\n\n drm/i915: use a separate context for gpu relocs\n \n The CS pre-parser can pre-fetch commands across memory sync points and\n starting from gen12 it is able to pre-fetch across BB_START and BB_END\n boundaries as well, so when we emit gpu relocs the pre-parser might\n fetch the target location of the reloc before the memory write lands.\n \n The parser can\u0027t pre-fetch across the ctx switch, so we use a separate\n context to guarantee that the memory is synchronized before the parser\n can get to it.\n \n Note that there is no risk of the CS doing a lite restore from the reloc\n context to the user context, even if the two have the same hw_id,\n because since gen11 the CS also checks the LRCA when deciding if it can\n lite-restore.\n \n v2: limit new context to gen12+, release in eb_destroy, add a comment\n in emit_fini_breadcrumb (Chris).\n \n Suggested-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190827185805.21799-1-daniele.ceraolospurio@intel.com\n\ncommit b60c2c4aa512f3e95f4525c02df57dcc71a0cdc7\nAuthor: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\nDate: Tue Aug 27 08:49:45 2019 +0000\n\n regulator: sy8824x: add SY20278 support\n \n The differences between SY8824C and SY20278 are different regs\n for mode/enable.\n \n Signed-off-by: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\n Link: https://lore.kernel.org/r/20190827163830.2c94f29b@xhacker.debian\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit f27b425d13a34c8b9767c666b7fab30692960276\nAuthor: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\nDate: Tue Aug 27 08:49:09 2019 +0000\n\n dt-bindings: sy8824x: Document SY20278 support\n \n SY20276 is an I2C-controlled adjustable voltage regulator made by\n Silergy Corp. The differences between SY8824C and SY20278 are\n different regs for mode/enable.\n \n Signed-off-by: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\n Link: https://lore.kernel.org/r/20190827163754.170cf130@xhacker.debian\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 92fd0db2cd414c5c78f8f959ed7325e2b24f9445\nAuthor: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\nDate: Tue Aug 27 08:48:36 2019 +0000\n\n regulator: sy8824x: add SY20276 support\n \n The differences between SY8824C and SY20276 are different vsel_min,\n vsel_step, vsel_count and regs for mode/enable.\n \n Signed-off-by: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\n Link: https://lore.kernel.org/r/20190827163721.1947f7a0@xhacker.debian\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 253f6cb8b893f5260198d8be244bfc11c3b54d76\nAuthor: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\nDate: Tue Aug 27 08:48:04 2019 +0000\n\n dt-bindings: sy8824x: Document SY20276 support\n \n SY20276 is an I2C-controlled adjustable voltage regulator made by\n Silergy Corp. The differences between SY8824C and SY20276 are\n different vsel_min, vsel_step, vsel_count and regs for mode/enable.\n \n Signed-off-by: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\n Link: https://lore.kernel.org/r/20190827163650.47ed1213@xhacker.debian\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 67a7b05b297eba72ee4e0655966ae7cc769d3a26\nAuthor: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\nDate: Tue Aug 27 08:47:30 2019 +0000\n\n regulator: sy8824x: add SY8824E support\n \n The only difference between SY8824E and SY8824C/D is the vsel_min.\n \n Signed-off-by: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\n Link: https://lore.kernel.org/r/20190827163537.52023c4e@xhacker.debian\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 90de3ae410aaaa63ee7518fac3dabd38f15972c4\nAuthor: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\nDate: Tue Aug 27 08:46:21 2019 +0000\n\n dt-bindings: sy8824x: Document SY8824E support\n \n SY8824E is an I2C-controlled adjustable voltage regulator made by\n Silergy Corp. The only difference between SY8824C and SY8824E is the\n vsel_min.\n \n Signed-off-by: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\n Link: https://lore.kernel.org/r/20190827163505.361890af@xhacker.debian\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit d5a2b8220335a605ce22357dd71864f942dda2b9\nAuthor: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\nDate: Tue Aug 27 08:45:33 2019 +0000\n\n regulator: add support for SY8824C regulator\n \n SY8824C is an I2C attached single output regulator made by Silergy Corp,\n which is used on several Synaptics berlin platforms to control the\n power supply of the ARM cores.\n \n Add a driver for it.\n \n Signed-off-by: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\n Link: https://lore.kernel.org/r/20190827163418.1a32fc48@xhacker.debian\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit d3733bc80a5c992679ba9c117c2649ba305338ca\nAuthor: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\nDate: Tue Aug 27 08:45:00 2019 +0000\n\n regulator: add binding for the SY8824C voltage regulator\n \n SY8824C is an I2C-controlled adjustable voltage regulator made by\n Silergy Corp.\n \n Add its device tree binding.\n \n Signed-off-by: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\n Link: https://lore.kernel.org/r/20190827163341.61df63a7@xhacker.debian\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit bfb3943bed670a0655aa2f9dcf82222ff09d6dd1\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 27 21:41:16 2019 +0200\n\n Revert \"driver core: Add support for linking devices during device addition\"\n \n This reverts commit 5302dd7dd0b6d04c63cdce51d1e9fda9ef0be886.\n \n Based on a lot of email and in-person discussions, this patch series is\n being reworked to address a number of issues that were pointed out that\n needed to be taken care of before it should be merged. It will be\n resubmitted with those changes hopefully soon.\n \n Cc: Frank Rowand \u003cfrowand.list@gmail.com\u003e\n Cc: Saravana Kannan \u003csaravanak@google.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 33cbfe54499338af08ab906a99afac247ea533f6\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 27 21:41:06 2019 +0200\n\n Revert \"driver core: Add edit_links() callback for drivers\"\n \n This reverts commit 134b23eec9e3a3c795a6ceb0efe2fa63e87983b2.\n \n Based on a lot of email and in-person discussions, this patch series is\n being reworked to address a number of issues that were pointed out that\n needed to be taken care of before it should be merged. It will be\n resubmitted with those changes hopefully soon.\n \n Cc: Frank Rowand \u003cfrowand.list@gmail.com\u003e\n Cc: Saravana Kannan \u003csaravanak@google.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d77b3f07a1d7127f1ecaf82a95a36b8a3e8fc689\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 27 21:40:58 2019 +0200\n\n Revert \"of/platform: Add functional dependency link from DT bindings\"\n \n This reverts commit 690ff7881b2681afca1c3c063f4a5cb7c71d9d8b.\n \n Based on a lot of email and in-person discussions, this patch series is\n being reworked to address a number of issues that were pointed out that\n needed to be taken care of before it should be merged. It will be\n resubmitted with those changes hopefully soon.\n \n Cc: Frank Rowand \u003cfrowand.list@gmail.com\u003e\n Cc: Saravana Kannan \u003csaravanak@google.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 376142b7bbc1edef2900ddc300a006f8c056b0d5\nAuthor: Michał Mirosław \u003cmirq-linux@rere.qmqm.pl\u003e\nDate: Sat Aug 24 22:26:54 2019 +0200\n\n ASoC: atmel_ssc_dai: implement left-justified data mode\n \n Enable support for left-justified data mode for SSC-codec link.\n \n Signed-off-by: Michał Mirosław \u003cmirq-linux@rere.qmqm.pl\u003e\n Reviewed-by: Codrin Ciubotariu \u003ccodrin.ciubotariu@microchip.com\u003e\n Link: https://lore.kernel.org/r/44fa6b700421e80778f20ff9ead2b148cf6d2e92.1566677788.git.mirq-linux@rere.qmqm.pl\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit bcca686c11cdad3b4566ed1818cd2688e7901def\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 27 21:40:50 2019 +0200\n\n Revert \"driver core: Add sync_state driver/bus callback\"\n \n This reverts commit 8f8184d6bf676a8680d6f441e40317d166b46f73.\n \n Based on a lot of email and in-person discussions, this patch series is\n being reworked to address a number of issues that were pointed out that\n needed to be taken care of before it should be merged. It will be\n resubmitted with those changes hopefully soon.\n \n Cc: Frank Rowand \u003cfrowand.list@gmail.com\u003e\n Cc: Saravana Kannan \u003csaravanak@google.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit db8e3e2091c3ccc4e1dde0f78b3111301cf129e9\nAuthor: Michał Mirosław \u003cmirq-linux@rere.qmqm.pl\u003e\nDate: Sat Aug 24 22:26:53 2019 +0200\n\n ASoC: atmel_ssc_dai: rework DAI format configuration\n \n Rework DAI format calculation in preparation for adding more formats\n later. As a side-effect this enables all CBM/CBS x CFM/CFS combinations\n for supported formats. (Note: the additional modes are not tested.)\n \n Note: this changes FSEDGE to POSITIVE for I2S CBM_CFS mode as the TXSYN\n interrupt is not used anyway.\n \n Signed-off-by: Michał Mirosław \u003cmirq-linux@rere.qmqm.pl\u003e\n Link: https://lore.kernel.org/r/f5949b0326fdcdca072f3ed03f77de9e207631cd.1566677788.git.mirq-linux@rere.qmqm.pl\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 77df8e790026d6f1a60c83e5fe822bd35d54a002\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 27 21:40:42 2019 +0200\n\n Revert \"of/platform: Pause/resume sync state during init and of_platform_populate()\"\n \n This reverts commit 21871a99b34c65c56a24193c277a4981529c306f.\n \n Based on a lot of email and in-person discussions, this patch series is\n being reworked to address a number of issues that were pointed out that\n needed to be taken care of before it should be merged. It will be\n resubmitted with those changes hopefully soon.\n \n Cc: Frank Rowand \u003cfrowand.list@gmail.com\u003e\n Cc: Saravana Kannan \u003csaravanak@google.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b1c46e118b21dfaa4112a897ead12d7b141cb83b\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 27 21:40:35 2019 +0200\n\n Revert \"of/platform: Create device links for all child-supplier depencencies\"\n \n This reverts commit 709fb8297358f08a125b770d1518a95d03b541db.\n \n Based on a lot of email and in-person discussions, this patch series is\n being reworked to address a number of issues that were pointed out that\n needed to be taken care of before it should be merged. It will be\n resubmitted with those changes hopefully soon.\n \n Cc: Frank Rowand \u003cfrowand.list@gmail.com\u003e\n Cc: Saravana Kannan \u003csaravanak@google.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 40e05e6acdc07f1a513f206c1a8c5fce16ae5a81\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 27 21:40:17 2019 +0200\n\n Revert \"of/platform: Don\u0027t create device links for default busses\"\n \n This reverts commit b3173c2292fbaf24ff7062d366830b012ed04269.\n \n Based on a lot of email and in-person discussions, this patch series is\n being reworked to address a number of issues that were pointed out that\n needed to be taken care of before it should be merged. It will be\n resubmitted with those changes hopefully soon.\n \n Cc: Frank Rowand \u003cfrowand.list@gmail.com\u003e\n Cc: Saravana Kannan \u003csaravanak@google.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f6680a66a3ca3ed41a5870ead7591a27827d5632\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 27 21:40:08 2019 +0200\n\n Revert \"of/platform: Fix fn definitons for of_link_is_valid() and of_link_property()\"\n \n This reverts commit 0ad0bdd8ba6489d05044aa5ca6c6f7491be7743c.\n \n Based on a lot of email and in-person discussions, this patch series is\n being reworked to address a number of issues that were pointed out that\n needed to be taken care of before it should be merged. It will be\n resubmitted with those changes hopefully soon.\n \n Cc: Frank Rowand \u003cfrowand.list@gmail.com\u003e\n Cc: Saravana Kannan \u003csaravanak@google.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 18291410557fb860706d1889505cc15d6e31d790\nAuthor: Michał Mirosław \u003cmirq-linux@rere.qmqm.pl\u003e\nDate: Sat Aug 24 22:26:52 2019 +0200\n\n ASoC: atmel: enable SOC_SSC_PDC and SOC_SSC_DMA in Kconfig\n \n Allow SSC to be used on platforms described using audio-graph-card\n in Device Tree.\n \n Signed-off-by: Michał Mirosław \u003cmirq-linux@rere.qmqm.pl\u003e\n Reviewed-by: Codrin Ciubotariu \u003ccodrin.ciubotariu@microchip.com\u003e\n Acked-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Link: https://lore.kernel.org/r/233d5461f4448df151755de7b69a0cd3ad310d5c.1566677788.git.mirq-linux@rere.qmqm.pl\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 8ef4df5ae1091af6c8aa65938a80f9ad9fa4460f\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 27 21:40:00 2019 +0200\n\n Revert \"of/platform: Fix device_links_supplier_sync_state_resume() warning\"\n \n This reverts commit 5adf5781019dd21233d6d13ab0d78bf03a13d4f1.\n \n Based on a lot of email and in-person discussions, this patch series is\n being reworked to address a number of issues that were pointed out that\n needed to be taken care of before it should be merged. It will be\n resubmitted with those changes hopefully soon.\n \n Cc: Frank Rowand \u003cfrowand.list@gmail.com\u003e\n Cc: Saravana Kannan \u003csaravanak@google.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b429ca49406501e2c48c2a137eab12910c21ad0c\nAuthor: Shengjiu Wang \u003cshengjiu.wang@nxp.com\u003e\nDate: Tue Aug 27 18:13:46 2019 -0400\n\n ASoC: cs42xx8: Force suspend/resume during system suspend/resume\n \n Use force_suspend/resume to make sure clocks are disabled/enabled\n accordingly during system suspend/resume.\n \n Signed-off-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Signed-off-by: Shengjiu Wang \u003cshengjiu.wang@nxp.com\u003e\n Reviewed-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Link: https://lore.kernel.org/r/1566944026-18113-1-git-send-email-shengjiu.wang@nxp.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit caec05b6688284bd9745804d54aab5ef62b7ac78\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 27 21:39:43 2019 +0200\n\n Revert \"of/platform: Disable generic device linking code for PowerPC\"\n \n This reverts commit 9c1c5e0bc5ec032e78bf15c302f3b20152f865ad.\n \n Based on a lot of email and in-person discussions, this patch series is\n being reworked to address a number of issues that were pointed out that\n needed to be taken care of before it should be merged. It will be\n resubmitted with those changes hopefully soon.\n \n Cc: Frank Rowand \u003cfrowand.list@gmail.com\u003e\n Cc: Saravana Kannan \u003csaravanak@google.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d2ac1fe0342ce2b44621f60d053800e2d94f1d21\nAuthor: Jiaxin Yu \u003cjiaxin.yu@mediatek.com\u003e\nDate: Sat Aug 24 12:58:46 2019 +0800\n\n ASoC: mediatek: mt8183: fix tdm out data align issue\n \n Mt8183 tdm out support S16_LE/S24_LE/S32_LE formats. When output S32_LE,\n we need set hd_align so that memif can output MSB 24bits. When output\n S24_LE, we need reset hd_align so that memif can output LSB 24bits.\n \n Signed-off-by: Jiaxin Yu \u003cjiaxin.yu@mediatek.com\u003e\n Link: https://lore.kernel.org/r/1566622726-27113-1-git-send-email-jiaxin.yu@mediatek.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit e9e897d5fdba9bda3a8c80ea39761413725f198c\nAuthor: Shuming Fan \u003cshumingf@realtek.com\u003e\nDate: Mon Aug 26 17:00:52 2019 +0800\n\n ASoC: rt1011: add mutex protection to set_fmt/set_tdm_slot\n \n The calibration process at booting will reset registers and bypass cache\n to make sure the calibration is done.\n We add mutex protection to avoid unexpected settings while\n the registration process and calibration are interleaved.\n \n Signed-off-by: Shuming Fan \u003cshumingf@realtek.com\u003e\n Link: https://lore.kernel.org/r/20190826090052.1875-1-shumingf@realtek.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit f6e77921969003eaf5dbae9c0b5feeb20c6caf50\nAuthor: Katsuhiro Suzuki \u003ckatsuhiro@katsuster.net\u003e\nDate: Tue Aug 27 00:39:00 2019 +0900\n\n ASoC: es8316: fix inverted L/R of headphone mixer volume\n \n This patch fixes inverted Left-Right channel of headphone mixer\n volume by wrong shift_left, shift_right values.\n \n Signed-off-by: Katsuhiro Suzuki \u003ckatsuhiro@katsuster.net\u003e\n Reviewed-by: Daniel Drake \u003cdrake@endlessm.com\u003e\n Link: https://lore.kernel.org/r/20190826153900.25969-2-katsuhiro@katsuster.net\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit f972d02fee2496024cfd6f59021c9d89d54922a6\nAuthor: Katsuhiro Suzuki \u003ckatsuhiro@katsuster.net\u003e\nDate: Tue Aug 27 00:38:59 2019 +0900\n\n ASoC: es8316: fix headphone mixer volume table\n \n This patch fix setting table of Headphone mixer volume.\n Current code uses 4 ... 7 values but these values are prohibited.\n \n Correct settings are the following:\n 0000 -12dB\n 0001 -10.5dB\n 0010 -9dB\n 0011 -7.5dB\n 0100 -6dB\n 1000 -4.5dB\n 1001 -3dB\n 1010 -1.5dB\n 1011 0dB\n \n Signed-off-by: Katsuhiro Suzuki \u003ckatsuhiro@katsuster.net\u003e\n Reviewed-by: Daniel Drake \u003cdrake@endlessm.com\u003e\n Link: https://lore.kernel.org/r/20190826153900.25969-1-katsuhiro@katsuster.net\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 0009fad033370802de75e4cedab54f4d86450e22\nAuthor: Nigel Croxon \u003cncroxon@redhat.com\u003e\nDate: Wed Aug 21 09:27:08 2019 -0400\n\n raid5 improve too many read errors msg by adding limits\n \n Often limits can be changed by admin. When discussing such things\n it helps if you can provide \"self-sustained\" facts. Also\n sometimes the admin thinks he changed a limit, but it did not\n take effect for some reason or he changed the wrong thing.\n \n V3: Only pr_warn when Faulty is 0.\n V2: Add read_errors value to pr_warn.\n \n Signed-off-by: Nigel Croxon \u003cncroxon@redhat.com\u003e\n Signed-off-by: Song Liu \u003csongliubraving@fb.com\u003e\n\ncommit 9d4b45d6af442237560d0bb5502a012baa5234b7\nAuthor: NeilBrown \u003cneilb@suse.com\u003e\nDate: Tue Aug 20 10:21:09 2019 +1000\n\n md: don\u0027t report active array_state until after revalidate_disk() completes.\n \n Until revalidate_disk() has completed, the size of a new md array will\n appear to be zero.\n So we shouldn\u0027t report, through array_state, that the array is active\n until that time.\n udev rules check array_state to see if the array is ready. As soon as\n it appear to be zero, fsck can be run. If it find the size to be\n zero, it will fail.\n \n So add a new flag to provide an interlock between do_md_run() and\n array_state_show(). This flag is set while do_md_run() is active and\n it prevents array_state_show() from reporting that the array is\n active.\n \n Before do_md_run() is called, -\u003epers will be NULL so array is\n definitely not active.\n After do_md_run() is called, revalidate_disk() will have run and the\n array will be completely ready.\n \n We also move various sysfs_notify*() calls out of md_run() into\n do_md_run() after MD_NOT_READY is cleared. This ensure the\n information is ready before the notification is sent.\n \n Prior to v4.12, array_state_show() was called with the\n mddev-\u003ereconfig_mutex held, which provided exclusion with do_md_run().\n \n Note that MD_NOT_READY cleared twice. This is deliberate to cover\n both success and error paths with minimal noise.\n \n Fixes: b7b17c9b67e5 (\"md: remove mddev_lock() from md_attr_show()\")\n Cc: stable@vger.kernel.org (v4.12++)\n Signed-off-by: NeilBrown \u003cneilb@suse.com\u003e\n Signed-off-by: Song Liu \u003csongliubraving@fb.com\u003e\n\ncommit 480523feae581ab714ba6610388a3b4619a2f695\nAuthor: NeilBrown \u003cneilb@suse.com\u003e\nDate: Tue Aug 20 10:21:09 2019 +1000\n\n md: only call set_in_sync() when it is expected to succeed.\n \n Since commit 4ad23a976413 (\"MD: use per-cpu counter for\n writes_pending\"), set_in_sync() is substantially more expensive: it\n can wait for a full RCU grace period which can be 10s of milliseconds.\n \n So we should only call it when the cost is justified.\n \n md_check_recovery() currently calls set_in_sync() every time it finds\n anything to do (on non-external active arrays). For an array\n performing resync or recovery, this will be quite often.\n Each call will introduce a delay to the md thread, which can noticeable\n affect IO submission latency.\n \n In md_check_recovery() we only need to call set_in_sync() if\n \u0027safemode\u0027 was non-zero at entry, meaning that there has been not\n recent IO. So we save this \"safemode was nonzero\" state, and only\n call set_in_sync() if it was non-zero.\n \n This measurably reduces mean and maximum IO submission latency during\n resync/recovery.\n \n Reported-and-tested-by: Jack Wang \u003cjinpu.wang@cloud.ionos.com\u003e\n Fixes: 4ad23a976413 (\"MD: use per-cpu counter for writes_pending\")\n Cc: stable@vger.kernel.org (v4.12+)\n Signed-off-by: NeilBrown \u003cneilb@suse.com\u003e\n Signed-off-by: Song Liu \u003csongliubraving@fb.com\u003e\n\ncommit 1c4691199fbef086ee7953a4f4db6505b106e79b\nAuthor: Andreas Gruenbacher \u003cagruenba@redhat.com\u003e\nDate: Thu Aug 22 18:07:09 2019 +0200\n\n gfs2: Fix recovery slot bumping\n \n Get rid of the assumption that the number of slots can at most increase by\n RECOVER_SIZE_INC (16) in set_recover_size.\n \n Signed-off-by: Andreas Gruenbacher \u003cagruenba@redhat.com\u003e\n\ncommit 3c9347351a6ea1234aa647b36f89052de050d2a2\nAuthor: Robin Murphy \u003crobin.murphy@arm.com\u003e\nDate: Thu Aug 1 16:22:45 2019 +0100\n\n perf/smmuv3: Validate groups for global filtering\n \n With global filtering, it becomes possible for users to construct\n self-contradictory groups with conflicting filters. Make sure we\n cover that when initially validating events.\n \n Signed-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 33e84ea4330da8a16bda8a871d0cd3c872bcd89f\nAuthor: Robin Murphy \u003crobin.murphy@arm.com\u003e\nDate: Thu Aug 1 16:22:44 2019 +0100\n\n perf/smmuv3: Validate group size\n \n Ensure that a group will actually fit into the available counters.\n \n Signed-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 6f43791713c5bb1212ec8497034275dfdbba5cbd\nAuthor: Amadeusz Sławiński \u003camadeuszx.slawinski@linux.intel.com\u003e\nDate: Tue Aug 27 16:17:11 2019 +0200\n\n ASoC: Intel: Skylake: Release topology when we are done with it\n \n Currently topology is kept in memory while driver is running. It\u0027s\n unnecessary, as it\u0027s only needed during parsing.\n \n Signed-off-by: Amadeusz Sławiński \u003camadeuszx.slawinski@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190827141712.21015-6-amadeuszx.slawinski@linux.intel.com\n Reviewed-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 0cc33cecb4a1de5b55d382b8e24a128a3551a031\nAuthor: Amadeusz Sławiński \u003camadeuszx.slawinski@intel.com\u003e\nDate: Tue Aug 27 16:17:10 2019 +0200\n\n ASoC: Intel: Skylake: Print module type instead of id\n \n When we are printing module params, we were actually printing module id\n instead of type, but debug message was saying that number we get is type.\n So print module type, as it is useful when debugging paths, but also\n keep printing module id, as it is used in all other logs.\n \n Signed-off-by: Amadeusz Sławiński \u003camadeuszx.slawinski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190827141712.21015-5-amadeuszx.slawinski@linux.intel.com\n Reviewed-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 855a06da37a773fd073d51023ac9d07988c87da8\nAuthor: Amadeusz Sławiński \u003camadeuszx.slawinski@intel.com\u003e\nDate: Tue Aug 27 16:17:12 2019 +0200\n\n ASoC: Intel: NHLT: Fix debug print format\n \n oem_table_id is 8 chars long, so we need to limit it, otherwise it\n may print some unprintable characters into dmesg.\n \n Signed-off-by: Amadeusz Sławiński \u003camadeuszx.slawinski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190827141712.21015-7-amadeuszx.slawinski@linux.intel.com\n Reviewed-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n Cc: stable@vger.kernel.org\n\ncommit db33f00d15a63d269e283bad3f6f61eb00d2bc9d\nAuthor: Amadeusz Sławiński \u003camadeuszx.slawinski@intel.com\u003e\nDate: Tue Aug 27 16:17:09 2019 +0200\n\n ASoC: dapm: Expose snd_soc_dapm_new_control_unlocked properly\n \n We use snd_soc_dapm_new_control_unlocked for topology and have local\n declaration, instead declare it properly in header like already declared\n snd_soc_dapm_new_control.\n \n Signed-off-by: Amadeusz Sławiński \u003camadeuszx.slawinski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190827141712.21015-4-amadeuszx.slawinski@linux.intel.com\n Reviewed-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 810f3b860850148788fc1ed8a6f5f807199fed65\nAuthor: Amadeusz Sławiński \u003camadeuszx.slawinski@intel.com\u003e\nDate: Tue Aug 27 16:17:08 2019 +0200\n\n ASoC: Intel: Fix use of potentially uninitialized variable\n \n If ipc-\u003eops.reply_msg_match is NULL, we may end up using uninitialized\n mask value.\n \n reported by smatch:\n sound/soc/intel/common/sst-ipc.c:266 sst_ipc_reply_find_msg() error: uninitialized symbol \u0027mask\u0027.\n \n Signed-off-by: Amadeusz Sławiński \u003camadeuszx.slawinski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190827141712.21015-3-amadeuszx.slawinski@linux.intel.com\n Reviewed-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n Cc: stable@vger.kernel.org\n\ncommit 17d29ff98fd4b70e9ccdac5e95e18a087e2737ef\nAuthor: Amadeusz Sławiński \u003camadeuszx.slawinski@intel.com\u003e\nDate: Tue Aug 27 16:17:07 2019 +0200\n\n ASoC: Intel: Skylake: Use correct function to access iomem space\n \n For copying from __iomem, we should use __ioread32_copy.\n \n reported by sparse:\n sound/soc/intel/skylake/skl-debug.c:437:34: warning: incorrect type in argument 1 (different address spaces)\n sound/soc/intel/skylake/skl-debug.c:437:34: expected void [noderef] \u003casn:2\u003e *to\n sound/soc/intel/skylake/skl-debug.c:437:34: got unsigned char *\n sound/soc/intel/skylake/skl-debug.c:437:51: warning: incorrect type in argument 2 (different address spaces)\n sound/soc/intel/skylake/skl-debug.c:437:51: expected void const *from\n sound/soc/intel/skylake/skl-debug.c:437:51: got void [noderef] \u003casn:2\u003e *[assigned] fw_reg_addr\n \n Signed-off-by: Amadeusz Sławiński \u003camadeuszx.slawinski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190827141712.21015-2-amadeuszx.slawinski@linux.intel.com\n Reviewed-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n Cc: stable@vger.kernel.org\n\ncommit a325c7bc3a92f1d5ba8cff10d7ab93a989cdeb80\nAuthor: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nDate: Mon Aug 26 16:18:55 2019 +0300\n\n ASoC: SOF: imx8: Fix an is IS_ERR() vs NULL check\n \n The device_link_add() function only returns NULL on error, it doesn\u0027t\n return error pointers.\n \n Fixes: 202acc565a1f (\"ASoC: SOF: imx: Add i.MX8 HW support\")\n Signed-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Reviewed-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190826131855.GA6840@mwanda\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9ec05d4723bf83dd272cef5ccf508e5fe4d30fa3\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Tue Aug 27 14:31:31 2019 +0200\n\n ASoC: sun4i: Revert A83t description\n \n The last set of reworks included some fixes to change the A83t behaviour\n and \"fix\" it.\n \n It turns out that the controller described in the datasheet and the one\n supported here are not the same, yet the A83t has the two of them, and the\n one supported in the driver wasn\u0027t the one described in the datasheet.\n \n Fix this by reintroducing the proper quirks.\n \n Fixes: 69e450e50ca6 (\"ASoC: sun4i-i2s: Fix the LRCK period on A83t\")\n Fixes: bf943d527987 (\"ASoC: sun4i-i2s: Fix MCLK Enable bit offset on A83t\")\n Fixes: 2e04fc4dbf50 (\"ASoC: sun4i-i2s: Fix WSS and SR fields for the A83t\")\n Fixes: 515fcfbc7736 (\"ASoC: sun4i-i2s: Fix LRCK and BCLK polarity offsets on newer SoCs\")\n Fixes: c1d3a921d72b (\"ASoC: sun4i-i2s: Fix the MCLK and BCLK dividers on newer SoCs\")\n Fixes: fb19739d7f68 (\"ASoC: sun4i-i2s: Use module clock as BCLK parent on newer SoCs\")\n Fixes: 71137bcd0a9a (\"ASoC: sun4i-i2s: Move the format configuration to a callback\")\n Fixes: d70be625f25a (\"ASoC: sun4i-i2s: Move the channel configuration to a callback\")\n Reported-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\n Tested-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190827123131.29129-2-mripard@kernel.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 393151c211006cc5ac6af26ecd9982dd916a8104\nAuthor: Wei Yongjun \u003cweiyongjun1@huawei.com\u003e\nDate: Mon Aug 26 12:00:03 2019 +0000\n\n ASoC: SOF: imx8: Fix return value check in imx8_probe()\n \n In case of error, the function devm_ioremap_wc() returns NULL pointer\n not ERR_PTR(). The IS_ERR() test in the return value check should be\n replaced with NULL test.\n \n Fixes: 202acc565a1f (\"ASoC: SOF: imx: Add i.MX8 HW support\")\n Signed-off-by: Wei Yongjun \u003cweiyongjun1@huawei.com\u003e\n Reviewed-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190826120003.183279-1-weiyongjun1@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 3cdc3c9fd9fa92bcef777512d9f4eef854f9a4ed\nAuthor: Tzung-Bi Shih \u003ctzungbi@google.com\u003e\nDate: Mon Aug 26 11:26:42 2019 +0800\n\n ASoC: mediatek: mt8183: remove unused DAPM-related variables\n \n Remove unused variable mt8183_mt6358_ts3a227_max98357_dapm_widgets and\n mt8183_mt6358_ts3a227_max98357_dapm_routes. They are accidentially\n included when rebasing commits.\n \n Fixes: 6191cbde5ff0 (\"ASoC: mediatek: mt8183: switch tdm pins gpio function when playback on or off\")\n Signed-off-by: Tzung-Bi Shih \u003ctzungbi@google.com\u003e\n Link: https://lore.kernel.org/r/20190826032642.27324-1-tzungbi@google.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit eb748213be482730af41ba2381da79f1edd1f07d\nAuthor: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\nDate: Mon Aug 26 11:41:09 2019 -0600\n\n coresight: tmc-etr: Add barrier packets when moving offset forward\n \n This patch adds barrier packets in the trace stream when the offset in the\n data buffer needs to be moved forward. Otherwise the decoder isn\u0027t aware\n of the break in the stream and can\u0027t synchronise itself with the trace\n data.\n \n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Tested-by: Yabin Cui \u003cyabinc@google.com\u003e\n Reviewed-by: Leo Yan \u003cleo.yan@linaro.org\u003e\n\ncommit aebbd3b1f3adc8a4b113c69cdc11cd64e76cab2d\nAuthor: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\nDate: Mon Aug 26 11:19:05 2019 -0600\n\n coresight: tmc-etr: Decouple buffer sync and barrier packet insertion\n \n If less space is available in the perf ring buffer than the ETR buffer,\n barrier packets inserted in the trace stream by tmc_sync_etr_buf() are\n skipped over when the head of the buffer is moved forward, resulting in\n traces that can\u0027t be decoded.\n \n This patch decouples the process of syncing ETR buffers and the addition\n of barrier packets in order to perform the latter once the offset in the\n trace buffer has been properly computed.\n \n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Reviewed-by: Leo Yan \u003cleo.yan@linaro.org\u003e\n\ncommit 1aab1cab46c037a6d6e89b5a7ad4ee852eff1475\nAuthor: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\nDate: Thu Aug 22 14:08:39 2019 -0600\n\n coresight: tmc: Make memory width mask computation into a function\n \n Make the computation of a memory mask representing the width of the memory\n bus into a function so that it can be re-used by the ETR driver.\n \n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Reviewed-by: Leo Yan \u003cleo.yan@linaro.org\u003e\n\ncommit 92604e825304ce170955df44b00973843054dcd5\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Wed Aug 21 22:54:41 2019 -0700\n\n smack: use GFP_NOFS while holding inode_smack::smk_lock\n \n inode_smack::smk_lock is taken during smack_d_instantiate(), which is\n called during a filesystem transaction when creating a file on ext4.\n Therefore to avoid a deadlock, all code that takes this lock must use\n GFP_NOFS, to prevent memory reclaim from waiting for the filesystem\n transaction to complete.\n \n Reported-by: syzbot+0eefc1e06a77d327a056@syzkaller.appspotmail.com\n Cc: stable@vger.kernel.org\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n Signed-off-by: Casey Schaufler \u003ccasey@schaufler-ca.com\u003e\n\ncommit 1d4a24debb088e30c5f7e6743abbf9b275396d25\nAuthor: Jia-Ju Bai \u003cbaijiaju1990@gmail.com\u003e\nDate: Tue Jul 23 18:00:15 2019 +0800\n\n security: smack: Fix possible null-pointer dereferences in smack_socket_sock_rcv_skb()\n \n In smack_socket_sock_rcv_skb(), there is an if statement\n on line 3920 to check whether skb is NULL:\n if (skb \u0026\u0026 skb-\u003esecmark !\u003d 0)\n \n This check indicates skb can be NULL in some cases.\n \n But on lines 3931 and 3932, skb is used:\n ad.a.u.net-\u003enetif \u003d skb-\u003eskb_iif;\n ipv6_skb_to_auditdata(skb, \u0026ad.a, NULL);\n \n Thus, possible null-pointer dereferences may occur when skb is NULL.\n \n To fix these possible bugs, an if statement is added to check skb.\n \n These bugs are found by a static analysis tool STCheck written by us.\n \n Signed-off-by: Jia-Ju Bai \u003cbaijiaju1990@gmail.com\u003e\n Signed-off-by: Casey Schaufler \u003ccasey@schaufler-ca.com\u003e\n\ncommit aa2e9da31431c2152689e8160bbb6809137b1e8d\nAuthor: luanshi \u003czhangliguang@linux.alibaba.com\u003e\nDate: Fri Jul 5 10:35:20 2019 +0800\n\n smack: fix some kernel-doc notations\n \n Fix/add kernel-doc notation and fix typos in security/smack/.\n \n Signed-off-by: Liguang Zhang \u003czhangliguang@linux.alibaba.com\u003e\n Signed-off-by: Casey Schaufler \u003ccasey@schaufler-ca.com\u003e\n\ncommit c6e22efdb6bb269c4a70227987d6fbc98c2b6d36\nAuthor: Jann Horn \u003cjannh@google.com\u003e\nDate: Thu Jul 4 20:44:44 2019 +0200\n\n Smack: Don\u0027t ignore other bprm-\u003eunsafe flags if LSM_UNSAFE_PTRACE is set\n \n There is a logic bug in the current smack_bprm_set_creds():\n If LSM_UNSAFE_PTRACE is set, but the ptrace state is deemed to be\n acceptable (e.g. because the ptracer detached in the meantime), the other\n -\u003eunsafe flags aren\u0027t checked. As far as I can tell, this means that\n something like the following could work (but I haven\u0027t tested it):\n \n - task A: create task B with fork()\n - task B: set NO_NEW_PRIVS\n - task B: install a seccomp filter that makes open() return 0 under some\n conditions\n - task B: replace fd 0 with a malicious library\n - task A: attach to task B with PTRACE_ATTACH\n - task B: execve() a file with an SMACK64EXEC extended attribute\n - task A: while task B is still in the middle of execve(), exit (which\n destroys the ptrace relationship)\n \n Make sure that if any flags other than LSM_UNSAFE_PTRACE are set in\n bprm-\u003eunsafe, we reject the execve().\n \n Cc: stable@vger.kernel.org\n Fixes: 5663884caab1 (\"Smack: unify all ptrace accesses in the smack\")\n Signed-off-by: Jann Horn \u003cjannh@google.com\u003e\n Signed-off-by: Casey Schaufler \u003ccasey@schaufler-ca.com\u003e\n\ncommit 116f21bb967fcef1fa360fe591a2947481788020\nAuthor: Ondrej Mosnacek \u003comosnace@redhat.com\u003e\nDate: Wed Aug 14 15:33:20 2019 +0200\n\n selinux: avoid atomic_t usage in sidtab\n \n As noted in Documentation/atomic_t.txt, if we don\u0027t need the RMW atomic\n operations, we should only use READ_ONCE()/WRITE_ONCE() +\n smp_rmb()/smp_wmb() where necessary (or the combined variants\n smp_load_acquire()/smp_store_release()).\n \n This patch converts the sidtab code to use regular u32 for the counter\n and reverse lookup cache and use the appropriate operations instead of\n atomic_get()/atomic_set(). Note that when reading/updating the reverse\n lookup cache we don\u0027t need memory barriers as it doesn\u0027t need to be\n consistent or accurate. We can now also replace some atomic ops with\n regular loads (when under spinlock) and stores (for conversion target\n fields that are always accessed under the master table\u0027s spinlock).\n \n We can now also bump SIDTAB_MAX to U32_MAX as we can use the full u32\n range again.\n \n Suggested-by: Jann Horn \u003cjannh@google.com\u003e\n Signed-off-by: Ondrej Mosnacek \u003comosnace@redhat.com\u003e\n Reviewed-by: Jann Horn \u003cjannh@google.com\u003e\n Signed-off-by: Paul Moore \u003cpaul@paul-moore.com\u003e\n\ncommit a45dddad3e2b360c5f21fdf12ac2bd32db77ea5c\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Wed Aug 21 10:21:38 2019 +0200\n\n dt-bindings: irq: Convert Allwinner NMI Controller to a schema\n \n The Allwinner SoCs have an interrupt controller called NMI supported in\n Linux, with a matching Device Tree binding.\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for that controller over to a YAML schemas.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 2eccb5655660a2bcdf2eecbff7068bc22d478915\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Wed Aug 21 10:21:37 2019 +0200\n\n dt-bindings: irq: Convert Allwinner IRQ Controller to a schema\n \n The Allwinner SoCs have an interrupt controller supported in Linux, with a\n matching Device Tree binding.\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for that controller over to a YAML schemas.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 9519893caf16cab28608c08985ae1521b1b222b5\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Fri Aug 9 17:48:21 2019 +0200\n\n btrfs: tie extent buffer and it\u0027s token together\n \n Further simplifaction of the get/set helpers is possible when the token\n is uniquely tied to an extent buffer. A condition and an assignment can\n be avoided.\n \n The initializations are moved closer to the first use when the extent\n buffer is valid. There\u0027s one exception in __push_leaf_left where the\n token is reused.\n \n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 6ec5cc7971295e972548ddd8512ce6e57677262b\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Fri Aug 9 17:30:23 2019 +0200\n\n btrfs: assume valid token for btrfs_set/get_token helpers\n \n Now that we can safely assume that the token is always a valid pointer,\n remove the branches that check that.\n \n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 1c4079cc2828faf7b8b8af599e5b88edfc588a33\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Fri Aug 9 17:12:38 2019 +0200\n\n btrfs: define separate btrfs_set/get_XX helpers\n \n There are helpers for all type widths defined via macro and optionally\n can use a token which is a cached pointer to avoid repeated mapping of\n the extent buffer.\n \n The token value is known at compile time, when it\u0027s valid it\u0027s always\n address of a local variable, otherwise it\u0027s NULL passed by the\n token-less helpers.\n \n This can be utilized to remove some branching as the helpers are used\n frequenlty.\n \n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 92af2b696119e491a95d77acdd8832b582d300d4\nAuthor: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nDate: Fri Aug 23 17:37:17 2019 +0100\n\n arm64: Relax Documentation/arm64/tagged-pointers.rst\n \n On AArch64 the TCR_EL1.TBI0 bit is set by default, allowing userspace\n (EL0) to perform memory accesses through 64-bit pointers with a non-zero\n top byte. However, such pointers were not allowed at the user-kernel\n syscall ABI boundary.\n \n With the Tagged Address ABI patchset, it is now possible to pass tagged\n pointers to the syscalls. Relax the requirements described in\n tagged-pointers.rst to be compliant with the behaviours guaranteed by\n the AArch64 Tagged Address ABI.\n \n Cc: Will Deacon \u003cwill.deacon@arm.com\u003e\n Cc: Szabolcs Nagy \u003cszabolcs.nagy@arm.com\u003e\n Cc: Kevin Brodsky \u003ckevin.brodsky@arm.com\u003e\n Acked-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\n Signed-off-by: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\n Co-developed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit e8688ba3874915b7d1661a57adf15a9c9e877df7\nAuthor: James Morse \u003cjames.morse@arm.com\u003e\nDate: Tue Aug 27 18:06:46 2019 +0100\n\n arm64: KVM: Device mappings should be execute-never\n \n Since commit 2f6ea23f63cca (\"arm64: KVM: Avoid marking pages as XN in\n Stage-2 if CTR_EL0.DIC is set\"), KVM has stopped marking normal memory\n as execute-never at stage2 when the system supports D-\u003eI Coherency at\n the PoU. This avoids KVM taking a trap when the page is first executed,\n in order to clean it to PoU.\n \n The patch that added this change also wrapped PAGE_S2_DEVICE mappings\n up in this too. The upshot is, if your CPU caches support DIC ...\n you can execute devices.\n \n Revert the PAGE_S2_DEVICE change so PTE_S2_XN is always used\n directly.\n \n Fixes: 2f6ea23f63cca (\"arm64: KVM: Avoid marking pages as XN in Stage-2 if CTR_EL0.DIC is set\")\n Signed-off-by: James Morse \u003cjames.morse@arm.com\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit c10bf3921e743dadb11a6cf59ffaf38cdbeb281b\nAuthor: Rui Miguel Silva \u003crui.silva@linaro.org\u003e\nDate: Tue Aug 27 16:53:02 2019 +0100\n\n staging: greybus: add missing includes\n \n Before moving greybus core out of staging and moving header files to\n include/linux some greybus header files were missing the necessary\n includes. This would trigger compilation faillures with some example\n errors logged bellow for with CONFIG_KERNEL_HEADER_TEST\u003dy.\n \n So, add the necessary headers to compile clean before relocating the\n header files.\n \n ./include/linux/greybus/hd.h:23:50: error: unknown type name \u0027u16\u0027\n int (*cport_disable)(struct gb_host_device *hd, u16 cport_id); ^~~\n ./include/linux/greybus/greybus_protocols.h:1314:2: error: unknown type name \u0027__u8\u0027\n __u8 data[0];\n ^~~~\n ./include/linux/greybus/hd.h:24:52: error: unknown type name \u0027u16\u0027\n int (*cport_connected)(struct gb_host_device *hd, u16 cport_id); ^~~\n ./include/linux/greybus/hd.h:25:48: error: unknown type name \u0027u16\u0027\n int (*cport_flush)(struct gb_host_device *hd, u16 cport_id); ^~~\n ./include/linux/greybus/hd.h:26:51: error: unknown type name \u0027u16\u0027\n int (*cport_shutdown)(struct gb_host_device *hd, u16 cport_id, ^~~\n ./include/linux/greybus/hd.h:27:5: error: unknown type name \u0027u8\u0027\n u8 phase, unsigned int timeout);\n ^~\n ./include/linux/greybus/hd.h:28:50: error: unknown type name \u0027u16\u0027\n int (*cport_quiesce)(struct gb_host_device *hd, u16 cport_id, ^~~\n ./include/linux/greybus/hd.h:29:5: error: unknown type name \u0027size_t\u0027\n size_t peer_space, unsigned int timeout);\n ^~~~~~\n ./include/linux/greybus/hd.h:29:5: note: \u0027size_t\u0027 is defined in header \u0027\u003cstddef.h\u003e\u0027; did you forget to \u0027#include \u003cstddef.h\u003e\u0027?\n ./include/linux/greybus/hd.h:1:1:\n +#include \u003cstddef.h\u003e\n /* SPDX-License-Identifier: GPL-2.0 */\n ./include/linux/greybus/hd.h:29:5:\n size_t peer_space, unsigned int timeout);\n ^~~~~~\n ./include/linux/greybus/hd.h:30:48: error: unknown type name \u0027u16\u0027\n int (*cport_clear)(struct gb_host_device *hd, u16 cport_id); ^~~\n ./include/linux/greybus/hd.h:32:49: error: unknown type name \u0027u16\u0027\n int (*message_send)(struct gb_host_device *hd, u16 dest_cport_id, ^~~\n ./include/linux/greybus/hd.h:33:32: error: unknown type name \u0027gfp_t\u0027\n struct gb_message *message, gfp_t gfp_mask); ^~~~~\n ./include/linux/greybus/hd.h:35:55: error: unknown type name \u0027u16\u0027\n int (*latency_tag_enable)(struct gb_host_device *hd, u16 cport_id);\n \n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Reported-by: Gao Xiang \u003chsiangkao@aol.com\u003e\n Signed-off-by: Rui Miguel Silva \u003crmfrfs@gmail.com\u003e\n Signed-off-by: Rui Miguel Silva \u003crui.silva@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190827155302.25704-1-rui.silva@linaro.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b81beec9cb2d586412c7166c893894930f19965e\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sun Aug 25 07:54:29 2019 +0200\n\n staging: greybus: move es2 to drivers/greybus/\n \n The es2 Greybus host controller has long been stable, so move it out of\n drivers/staging/ to drivers/greybus/\n \n Cc: Johan Hovold \u003cjohan@kernel.org\u003e\n Cc: greybus-dev@lists.linaro.org\n Cc: linux-kernel@vger.kernel.org\n Acked-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Acked-by: Alex Elder \u003celder@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190825055429.18547-10-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 8465def499c70d041a234087eff380108da7e830\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sun Aug 25 07:54:28 2019 +0200\n\n staging: greybus: move the greybus core to drivers/greybus\n \n The Greybus core code has been stable for a long time, and has been\n shipping for many years in millions of phones. With the advent of a\n recent Google Summer of Code project, and a number of new devices in the\n works from various companies, it is time to get the core greybus code\n out of staging as it really is going to be with us for a while.\n \n Cc: Johan Hovold \u003cjohan@kernel.org\u003e\n Cc: linux-kernel@vger.kernel.org\n Cc: greybus-dev@lists.linaro.org\n Acked-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Acked-by: Alex Elder \u003celder@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190825055429.18547-9-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit ec0ad868173da8a75121f9dc116a5d5478ff614d\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sun Aug 25 07:54:27 2019 +0200\n\n staging: greybus: move core include files to include/linux/greybus/\n \n With the goal of moving the core of the greybus code out of staging, the\n include files need to be moved to include/linux/greybus.h and\n include/linux/greybus/\n \n Cc: Vaibhav Hiremath \u003chvaibhav.linux@gmail.com\u003e\n Cc: Johan Hovold \u003cjohan@kernel.org\u003e\n Cc: Vaibhav Agarwal \u003cvaibhav.sr@gmail.com\u003e\n Cc: Rui Miguel Silva \u003crmfrfs@gmail.com\u003e\n Cc: David Lin \u003cdtwlin@gmail.com\u003e\n Cc: \"Bryan O\u0027Donoghue\" \u003cpure.logic@nexus-software.ie\u003e\n Cc: greybus-dev@lists.linaro.org\n Cc: devel@driverdev.osuosl.org\n Acked-by: Mark Greer \u003cmgreer@animalcreek.com\u003e\n Acked-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Acked-by: Alex Elder \u003celder@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190825055429.18547-8-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 9c31973907fb157ac91f61f479e8bf43bb43aded\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sun Aug 25 07:54:26 2019 +0200\n\n staging: greybus: loopback: Fix up some alignment checkpatch issues\n \n Some function prototypes do not match the expected alignment formatting\n so fix that up so that checkpatch is happy.\n \n Cc: \"Bryan O\u0027Donoghue\" \u003cpure.logic@nexus-software.ie\u003e\n Cc: Johan Hovold \u003cjohan@kernel.org\u003e\n Cc: greybus-dev@lists.linaro.org\n Cc: devel@driverdev.osuosl.org\n Acked-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Acked-by: Alex Elder \u003celder@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190825055429.18547-7-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 9238238589a3e5d289cfbf90f7ce65c6554e1614\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sun Aug 25 07:54:25 2019 +0200\n\n staging: greybus: log: Fix up some alignment checkpatch issues\n \n Some function prototypes do not match the expected alignment formatting\n so fix that up so that checkpatch is happy.\n \n Cc: David Lin \u003cdtwlin@gmail.com\u003e\n Cc: Johan Hovold \u003cjohan@kernel.org\u003e\n Cc: greybus-dev@lists.linaro.org\n Cc: devel@driverdev.osuosl.org\n Acked-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Acked-by: Alex Elder \u003celder@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190825055429.18547-6-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit ee989b020404023a27bfe34003748dc199835f23\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sun Aug 25 07:54:24 2019 +0200\n\n staging: greybus: manifest: Fix up some alignment checkpatch issues\n \n Some function prototypes do not match the expected alignment formatting\n so fix that up so that checkpatch is happy.\n \n Cc: Johan Hovold \u003cjohan@kernel.org\u003e\n Cc: greybus-dev@lists.linaro.org\n Cc: devel@driverdev.osuosl.org\n Acked-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Acked-by: Alex Elder \u003celder@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190825055429.18547-5-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a11ac9ef4b1604ce5d068932ab85b0906d1109c0\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sun Aug 25 07:54:23 2019 +0200\n\n staging: greybus: hd: Fix up some alignment checkpatch issues\n \n Some function prototypes do not match the expected alignment formatting\n so fix that up so that checkpatch is happy.\n \n Cc: Johan Hovold \u003cjohan@kernel.org\u003e\n Cc: greybus-dev@lists.linaro.org\n Cc: devel@driverdev.osuosl.org\n Acked-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Acked-by: Alex Elder \u003celder@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190825055429.18547-4-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 137bfbd006b2aad72fa90b6fbe096f3a1db5b32a\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sun Aug 25 07:54:22 2019 +0200\n\n staging: greybus: remove license \"boilerplate\"\n \n When the greybus drivers were converted to SPDX identifiers for the\n license text, some license boilerplate was not removed. Clean this up\n by removing this unneeded text now.\n \n Cc: Johan Hovold \u003cjohan@kernel.org\u003e\n Cc: Vaibhav Agarwal \u003cvaibhav.sr@gmail.com\u003e\n Cc: \"Bryan O\u0027Donoghue\" \u003cpure.logic@nexus-software.ie\u003e\n Cc: greybus-dev@lists.linaro.org\n Cc: devel@driverdev.osuosl.org\n Acked-by: Mark Greer \u003cmgreer@animalcreekk.com\u003e\n Acked-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Acked-by: Alex Elder \u003celder@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190825055429.18547-3-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit baeeb02ccd704e3a6cdff95a78a41916e6a85b21\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sun Aug 25 07:54:21 2019 +0200\n\n staging: greybus: fix up SPDX comment in .h files\n \n When these files originally got an SPDX tag, I used // instead of /* */\n for the .h files. Fix this up to use // properly.\n \n Cc: Johan Hovold \u003cjohan@kernel.org\u003e\n Cc: greybus-dev@lists.linaro.org\n Cc: devel@driverdev.osuosl.org\n Acked-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Acked-by: Alex Elder \u003celder@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190825055429.18547-2-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a1db98f20b818edf15f4ad70072f771b6f185949\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Mon Aug 26 21:26:53 2019 +0800\n\n erofs: fix compile warnings when moving out include/trace/events/erofs.h\n \n As Stephon reported [1], many compile warnings are raised when\n moving out include/trace/events/erofs.h:\n \n In file included from include/trace/events/erofs.h:8,\n from \u003ccommand-line\u003e:\n include/trace/events/erofs.h:28:37: warning: \u0027struct dentry\u0027 declared inside parameter list will not be visible outside of this definition or declaration\n TP_PROTO(struct inode *dir, struct dentry *dentry, unsigned int flags),\n ^~~~~~\n include/linux/tracepoint.h:233:34: note: in definition of macro \u0027__DECLARE_TRACE\u0027\n static inline void trace_##name(proto) \\\n ^~~~~\n include/linux/tracepoint.h:396:24: note: in expansion of macro \u0027PARAMS\u0027\n __DECLARE_TRACE(name, PARAMS(proto), PARAMS(args), \\\n ^~~~~~\n include/linux/tracepoint.h:532:2: note: in expansion of macro \u0027DECLARE_TRACE\u0027\n DECLARE_TRACE(name, PARAMS(proto), PARAMS(args))\n ^~~~~~~~~~~~~\n include/linux/tracepoint.h:532:22: note: in expansion of macro \u0027PARAMS\u0027\n DECLARE_TRACE(name, PARAMS(proto), PARAMS(args))\n ^~~~~~\n include/trace/events/erofs.h:26:1: note: in expansion of macro \u0027TRACE_EVENT\u0027\n TRACE_EVENT(erofs_lookup,\n ^~~~~~~~~~~\n include/trace/events/erofs.h:28:2: note: in expansion of macro \u0027TP_PROTO\u0027\n TP_PROTO(struct inode *dir, struct dentry *dentry, unsigned int flags),\n ^~~~~~~~\n \n That makes me very confused since most original EROFS tracepoint code\n was taken from f2fs, and finally I found\n \n commit 43c78d88036e (\"kbuild: compile-test kernel headers to ensure they are self-contained\")\n \n It seems these warnings are generated from KERNEL_HEADER_TEST feature and\n ext4/f2fs tracepoint files were in blacklist.\n \n Anyway, let\u0027s fix these issues for KERNEL_HEADER_TEST feature instead\n of adding to blacklist...\n \n [1] https://lore.kernel.org/lkml/20190826162432.11100665@canb.auug.org.au/\n \n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190826132653.100731-1-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 48d911f29e9f4e51eda87e0dc3c944be79771a0d\nAuthor: Sumera Priyadarsini \u003csylphrenadin@gmail.com\u003e\nDate: Mon Aug 26 23:52:27 2019 +0530\n\n staging: rtl8192u: Add or remove blank lines as required\n \n This patch fixes the file r8190_rtl8256.c to avoid the following\n checkpatch.pl warnings:\n CHECK: Blank lines aren\u0027t necessary before a close brace \u0027}\u0027\n CHECK: Please use a blank line after function/struct/union/enum declarations\n CHECK: Please don\u0027t use multiple blank line\n \n Signed-off-by: Sumera Priyadarsini \u003csylphrenadin@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190826182227.30738-1-sylphrenadin@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 8c10170fe141c989b204a840921a612bfc7ca72d\nAuthor: Sumera Priyadarsini \u003csylphrenadin@gmail.com\u003e\nDate: Mon Aug 26 23:39:09 2019 +0530\n\n staging: rtl8192u: Add or remove spaces to fix style issues\n \n This patch fixes the file r8190_rtl8256.c to avoid the following\n checkpatch.pl warnings:\n CHECK: spaces preferred around that \u0027\u003c\u003c\u0027 (ctx:VxV)\n CHECK: spaces preferred around that \u0027-\u0027 (ctx:VxV)\n CHECK: No space is necessary after a cast\n \n Signed-off-by: Sumera Priyadarsini \u003csylphrenadin@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190826180909.27775-1-sylphrenadin@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 8e480989dff378bc0b2418e3ccf47294fa77f5bf\nAuthor: Taihsiang Ho (tai271828) \u003ctai271828@gmail.com\u003e\nDate: Mon Aug 26 13:20:18 2019 +0800\n\n staging: rtl8712: wifi: checkpatch style fix\n \n Remove multiple blank lines.\n \n Signed-off-by: Taihsiang Ho (tai271828) \u003ctai271828@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190826052018.18649-1-tai271828@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f419f889e2f9bd991afb5c9c9025f89d726cdafe\nAuthor: Peikan Tsai \u003cpeikantsai@gmail.com\u003e\nDate: Mon Aug 26 01:58:49 2019 +0800\n\n staging: most-core: Fix checkpatch warnings\n \n Hi,\n \n This patch solves the following checkpatch.pl\u0027s messages in drivers/staging/most/core.c.\n \n WARNING: line over 80 characters\n + return snprintf(buf, PAGE_SIZE, \"%s\", ch_data_type[i].name);\n \n CHECK: Please use a blank line after function/struct/union/enum declarations\n +}\n +/**\n \n Signed-off-by: Peikan Tsai \u003cpeikantsai@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190825175849.GA74586@MarkdeMacBook-Pro.local\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 75b28affdd6aed1c68073ef53907c7bd822aff84\nAuthor: Hristo Venev \u003christo@venev.name\u003e\nDate: Mon Aug 26 17:23:46 2019 +0000\n\n io_uring: allocate the two rings together\n \n Both the sq and the cq rings have sizes just over a power of two, and\n the sq ring is significantly smaller. By bundling them in a single\n alllocation, we get the sq ring for free.\n \n This also means that IORING_OFF_SQ_RING and IORING_OFF_CQ_RING now mean\n the same thing. If we indicate this to userspace, we can save a mmap\n call.\n \n Signed-off-by: Hristo Venev \u003christo@venev.name\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 27c4d3a3252fa661b049b8f57a548bba331133e0\nAuthor: John Hubbard \u003cjhubbard@nvidia.com\u003e\nDate: Sun Aug 4 19:32:06 2019 -0700\n\n fs/io_uring.c: convert put_page() to put_user_page*()\n \n For pages that were retained via get_user_pages*(), release those pages\n via the new put_user_page*() routines, instead of via put_page() or\n release_pages().\n \n This is part a tree-wide conversion, as described in commit fc1d8e7cca2d\n (\"mm: introduce put_user_page*(), placeholder versions\").\n \n Cc: Alexander Viro \u003cviro@zeniv.linux.org.uk\u003e\n Cc: Jens Axboe \u003caxboe@kernel.dk\u003e\n Cc: linux-fsdevel@vger.kernel.org\n Cc: linux-block@vger.kernel.org\n Signed-off-by: John Hubbard \u003cjhubbard@nvidia.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit cecf5d87ff2035127bb5a9ee054d0023a4a7cad3\nAuthor: Ming Lei \u003cming.lei@redhat.com\u003e\nDate: Tue Aug 27 19:01:48 2019 +0800\n\n block: split .sysfs_lock into two locks\n \n The kernfs built-in lock of \u0027kn-\u003ecount\u0027 is held in sysfs .show/.store\n path. Meantime, inside block\u0027s .show/.store callback, q-\u003esysfs_lock is\n required.\n \n However, when mq \u0026 iosched kobjects are removed via\n blk_mq_unregister_dev() \u0026 elv_unregister_queue(), q-\u003esysfs_lock is held\n too. This way causes AB-BA lock because the kernfs built-in lock of\n \u0027kn-count\u0027 is required inside kobject_del() too, see the lockdep warning[1].\n \n On the other hand, it isn\u0027t necessary to acquire q-\u003esysfs_lock for\n both blk_mq_unregister_dev() \u0026 elv_unregister_queue() because\n clearing REGISTERED flag prevents storing to \u0027queue/scheduler\u0027\n from being happened. Also sysfs write(store) is exclusive, so no\n necessary to hold the lock for elv_unregister_queue() when it is\n called in switching elevator path.\n \n So split .sysfs_lock into two: one is still named as .sysfs_lock for\n covering sync .store, the other one is named as .sysfs_dir_lock\n for covering kobjects and related status change.\n \n sysfs itself can handle the race between add/remove kobjects and\n showing/storing attributes under kobjects. For switching scheduler\n via storing to \u0027queue/scheduler\u0027, we use the queue flag of\n QUEUE_FLAG_REGISTERED with .sysfs_lock for avoiding the race, then\n we can avoid to hold .sysfs_lock during removing/adding kobjects.\n \n [1] lockdep warning\n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n WARNING: possible circular locking dependency detected\n 5.3.0-rc3-00044-g73277fc75ea0 #1380 Not tainted\n ------------------------------------------------------\n rmmod/777 is trying to acquire lock:\n 00000000ac50e981 (kn-\u003ecount#202){++++}, at: kernfs_remove_by_name_ns+0x59/0x72\n \n but task is already holding lock:\n 00000000fb16ae21 (\u0026q-\u003esysfs_lock){+.+.}, at: blk_unregister_queue+0x78/0x10b\n \n which lock already depends on the new lock.\n \n the existing dependency chain (in reverse order) is:\n \n -\u003e #1 (\u0026q-\u003esysfs_lock){+.+.}:\n __lock_acquire+0x95f/0xa2f\n lock_acquire+0x1b4/0x1e8\n __mutex_lock+0x14a/0xa9b\n blk_mq_hw_sysfs_show+0x63/0xb6\n sysfs_kf_seq_show+0x11f/0x196\n seq_read+0x2cd/0x5f2\n vfs_read+0xc7/0x18c\n ksys_read+0xc4/0x13e\n do_syscall_64+0xa7/0x295\n entry_SYSCALL_64_after_hwframe+0x49/0xbe\n \n -\u003e #0 (kn-\u003ecount#202){++++}:\n check_prev_add+0x5d2/0xc45\n validate_chain+0xed3/0xf94\n __lock_acquire+0x95f/0xa2f\n lock_acquire+0x1b4/0x1e8\n __kernfs_remove+0x237/0x40b\n kernfs_remove_by_name_ns+0x59/0x72\n remove_files+0x61/0x96\n sysfs_remove_group+0x81/0xa4\n sysfs_remove_groups+0x3b/0x44\n kobject_del+0x44/0x94\n blk_mq_unregister_dev+0x83/0xdd\n blk_unregister_queue+0xa0/0x10b\n del_gendisk+0x259/0x3fa\n null_del_dev+0x8b/0x1c3 [null_blk]\n null_exit+0x5c/0x95 [null_blk]\n __se_sys_delete_module+0x204/0x337\n do_syscall_64+0xa7/0x295\n entry_SYSCALL_64_after_hwframe+0x49/0xbe\n \n other info that might help us debug this:\n \n Possible unsafe locking scenario:\n \n CPU0 CPU1\n ---- ----\n lock(\u0026q-\u003esysfs_lock);\n lock(kn-\u003ecount#202);\n lock(\u0026q-\u003esysfs_lock);\n lock(kn-\u003ecount#202);\n \n *** DEADLOCK ***\n \n 2 locks held by rmmod/777:\n #0: 00000000e69bd9de (\u0026lock){+.+.}, at: null_exit+0x2e/0x95 [null_blk]\n #1: 00000000fb16ae21 (\u0026q-\u003esysfs_lock){+.+.}, at: blk_unregister_queue+0x78/0x10b\n \n stack backtrace:\n CPU: 0 PID: 777 Comm: rmmod Not tainted 5.3.0-rc3-00044-g73277fc75ea0 #1380\n Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS ?-20180724_192412-buildhw-07.phx4\n Call Trace:\n dump_stack+0x9a/0xe6\n check_noncircular+0x207/0x251\n ? print_circular_bug+0x32a/0x32a\n ? find_usage_backwards+0x84/0xb0\n check_prev_add+0x5d2/0xc45\n validate_chain+0xed3/0xf94\n ? check_prev_add+0xc45/0xc45\n ? mark_lock+0x11b/0x804\n ? check_usage_forwards+0x1ca/0x1ca\n __lock_acquire+0x95f/0xa2f\n lock_acquire+0x1b4/0x1e8\n ? kernfs_remove_by_name_ns+0x59/0x72\n __kernfs_remove+0x237/0x40b\n ? kernfs_remove_by_name_ns+0x59/0x72\n ? kernfs_next_descendant_post+0x7d/0x7d\n ? strlen+0x10/0x23\n ? strcmp+0x22/0x44\n kernfs_remove_by_name_ns+0x59/0x72\n remove_files+0x61/0x96\n sysfs_remove_group+0x81/0xa4\n sysfs_remove_groups+0x3b/0x44\n kobject_del+0x44/0x94\n blk_mq_unregister_dev+0x83/0xdd\n blk_unregister_queue+0xa0/0x10b\n del_gendisk+0x259/0x3fa\n ? disk_events_poll_msecs_store+0x12b/0x12b\n ? check_flags+0x1ea/0x204\n ? mark_held_locks+0x1f/0x7a\n null_del_dev+0x8b/0x1c3 [null_blk]\n null_exit+0x5c/0x95 [null_blk]\n __se_sys_delete_module+0x204/0x337\n ? free_module+0x39f/0x39f\n ? blkcg_maybe_throttle_current+0x8a/0x718\n ? rwlock_bug+0x62/0x62\n ? __blkcg_punt_bio_submit+0xd0/0xd0\n ? trace_hardirqs_on_thunk+0x1a/0x20\n ? mark_held_locks+0x1f/0x7a\n ? do_syscall_64+0x4c/0x295\n do_syscall_64+0xa7/0x295\n entry_SYSCALL_64_after_hwframe+0x49/0xbe\n RIP: 0033:0x7fb696cdbe6b\n Code: 73 01 c3 48 8b 0d 1d 20 0c 00 f7 d8 64 89 01 48 83 c8 ff c3 66 2e 0f 1f 84 00 00 008\n RSP: 002b:00007ffec9588788 EFLAGS: 00000206 ORIG_RAX: 00000000000000b0\n RAX: ffffffffffffffda RBX: 0000559e589137c0 RCX: 00007fb696cdbe6b\n RDX: 000000000000000a RSI: 0000000000000800 RDI: 0000559e58913828\n RBP: 0000000000000000 R08: 00007ffec9587701 R09: 0000000000000000\n R10: 00007fb696d4eae0 R11: 0000000000000206 R12: 00007ffec95889b0\n R13: 00007ffec95896b3 R14: 0000559e58913260 R15: 0000559e589137c0\n \n Cc: Christoph Hellwig \u003chch@infradead.org\u003e\n Cc: Hannes Reinecke \u003chare@suse.com\u003e\n Cc: Greg KH \u003cgregkh@linuxfoundation.org\u003e\n Cc: Mike Snitzer \u003csnitzer@redhat.com\u003e\n Reviewed-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 58c898ba370e68d39470cd0d932b524682c1f9be\nAuthor: Ming Lei \u003cming.lei@redhat.com\u003e\nDate: Tue Aug 27 19:01:47 2019 +0800\n\n block: add helper for checking if queue is registered\n \n There are 4 users which check if queue is registered, so add one helper\n to check it.\n \n Cc: Christoph Hellwig \u003chch@infradead.org\u003e\n Cc: Hannes Reinecke \u003chare@suse.com\u003e\n Cc: Greg KH \u003cgregkh@linuxfoundation.org\u003e\n Cc: Mike Snitzer \u003csnitzer@redhat.com\u003e\n Cc: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Reviewed-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit c6ba933358f0d7a6a042b894dba20cc70396a6d3\nAuthor: Ming Lei \u003cming.lei@redhat.com\u003e\nDate: Tue Aug 27 19:01:46 2019 +0800\n\n blk-mq: don\u0027t hold q-\u003esysfs_lock in blk_mq_map_swqueue\n \n blk_mq_map_swqueue() is called from blk_mq_init_allocated_queue()\n and blk_mq_update_nr_hw_queues(). For the former caller, the kobject\n isn\u0027t exposed to userspace yet. For the latter caller, hctx sysfs entries\n and debugfs are un-registered before updating nr_hw_queues.\n \n On the other hand, commit 2f8f1336a48b (\"blk-mq: always free hctx after\n request queue is freed\") moves freeing hctx into queue\u0027s release\n handler, so there won\u0027t be race with queue release path too.\n \n So don\u0027t hold q-\u003esysfs_lock in blk_mq_map_swqueue().\n \n Cc: Christoph Hellwig \u003chch@infradead.org\u003e\n Cc: Hannes Reinecke \u003chare@suse.com\u003e\n Cc: Greg KH \u003cgregkh@linuxfoundation.org\u003e\n Cc: Mike Snitzer \u003csnitzer@redhat.com\u003e\n Cc: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Reviewed-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit c48dac137a62a5d6fa1ef3fa445cbd9c43655a76\nAuthor: Ming Lei \u003cming.lei@redhat.com\u003e\nDate: Tue Aug 27 19:01:45 2019 +0800\n\n block: don\u0027t hold q-\u003esysfs_lock in elevator_init_mq\n \n The original comment says:\n \n q-\u003esysfs_lock must be held to provide mutual exclusion between\n elevator_switch() and here.\n \n Which is simply wrong. elevator_init_mq() is only called from\n blk_mq_init_allocated_queue, which is always called before the request\n queue is registered via blk_register_queue(), for dm-rq or normal rq\n based driver. However, queue\u0027s kobject is only exposed and added to sysfs\n in blk_register_queue(). So there isn\u0027t such race between elevator_switch()\n and elevator_init_mq().\n \n So avoid to hold q-\u003esysfs_lock in elevator_init_mq().\n \n Cc: Christoph Hellwig \u003chch@infradead.org\u003e\n Cc: Hannes Reinecke \u003chare@suse.com\u003e\n Cc: Greg KH \u003cgregkh@linuxfoundation.org\u003e\n Cc: Mike Snitzer \u003csnitzer@redhat.com\u003e\n Cc: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Cc: Damien Le Moal \u003cdamien.lemoal@wdc.com\u003e\n Reviewed-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 9685b2270211628e27ea7880a02b52efd4524099\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Tue Aug 27 19:01:44 2019 +0800\n\n block: Remove blk_mq_register_dev()\n \n This function has no callers. Hence remove it.\n \n Cc: Christoph Hellwig \u003chch@infradead.org\u003e\n Cc: Ming Lei \u003cming.lei@redhat.com\u003e\n Cc: Hannes Reinecke \u003chare@suse.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 5c062ef4155b60018c547552ca48823297d00998\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Thu Aug 22 17:21:21 2019 +0100\n\n arm64: kvm: Replace hardcoded \u00271\u0027 with SYS_PAR_EL1_F\n \n Now that we have a definition for the \u0027F\u0027 field of PAR_EL1, use that\n instead of coding the immediate directly.\n \n Acked-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n Reviewed-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 42f91093b043332ad75cea7aeafecda6fe81814c\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Thu Aug 22 17:22:14 2019 +0100\n\n arm64: mm: Ignore spurious translation faults taken from the kernel\n \n Thanks to address translation being performed out of order with respect to\n loads and stores, it is possible for a CPU to take a translation fault when\n accessing a page that was mapped by a different CPU.\n \n For example, in the case that one CPU maps a page and then sets a flag to\n tell another CPU:\n \n CPU 0\n -----\n \n MOV X0, \u003cvalid pte\u003e\n STR X0, [Xptep] // Store new PTE to page table\n DSB ISHST\n ISB\n MOV X1, #1\n STR X1, [Xflag] // Set the flag\n \n CPU 1\n -----\n \n loop: LDAR X0, [Xflag] // Poll flag with Acquire semantics\n CBZ X0, loop\n LDR X1, [X2] // Translates using the new PTE\n \n then the final load on CPU 1 can raise a translation fault because the\n translation can be performed speculatively before the read of the flag and\n marked as \"faulting\" by the CPU. This isn\u0027t quite as bad as it sounds\n since, in reality, code such as:\n \n CPU 0 CPU 1\n ----- -----\n spin_lock(\u0026lock); spin_lock(\u0026lock);\n *ptr \u003d vmalloc(size); if (*ptr)\n spin_unlock(\u0026lock); foo \u003d **ptr;\n spin_unlock(\u0026lock);\n \n will not trigger the fault because there is an address dependency on CPU 1\n which prevents the speculative translation. However, more exotic code where\n the virtual address is known ahead of time, such as:\n \n CPU 0 CPU 1\n ----- -----\n spin_lock(\u0026lock); spin_lock(\u0026lock);\n set_fixmap(0, paddr, prot); if (mapped)\n mapped \u003d true; foo \u003d *fix_to_virt(0);\n spin_unlock(\u0026lock); spin_unlock(\u0026lock);\n \n could fault. This can be avoided by any of:\n \n * Introducing broadcast TLB maintenance on the map path\n * Adding a DSB;ISB sequence after checking a flag which indicates\n that a virtual address is now mapped\n * Handling the spurious fault\n \n Given that we have never observed a problem due to this under Linux and\n future revisions of the architecture are being tightened so that\n translation table walks are effectively ordered in the same way as explicit\n memory accesses, we no longer treat spurious kernel faults as fatal if an\n AT instruction indicates that the access does not trigger a translation\n fault.\n \n Reviewed-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit e8620cff99946ea1f7891d7bec071a23a1fdaef3\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Thu Aug 22 17:19:17 2019 +0100\n\n arm64: sysreg: Add some field definitions for PAR_EL1\n \n PAR_EL1 is a mysterious creature, but sometimes it\u0027s necessary to read\n it when translating addresses in situations where we cannot walk the\n page table directly.\n \n Add a couple of system register definitions for the fault indication\n field (\u0027F\u0027) and the fault status code (\u0027FST\u0027).\n \n Reviewed-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit eb6a4dcce33925ac95023bbe5199474f8db40ba7\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Fri Aug 23 13:03:55 2019 +0100\n\n arm64: mm: Add ISB instruction to set_pgd()\n \n Commit 6a4cbd63c25a (\"Revert \"arm64: Remove unnecessary ISBs from\n set_{pte,pmd,pud}\"\") reintroduced ISB instructions to some of our\n page table setter functions in light of a recent clarification to the\n Armv8 architecture. Although \u0027set_pgd()\u0027 isn\u0027t currently used to update\n a live page table, add the ISB instruction there too for consistency\n with the other macros and to provide some future-proofing if we use it\n on live tables in the future.\n \n Reported-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Reviewed-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 51696d346c49c6cf4f29e9b20d6e15832a2e3408\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Thu Aug 22 15:03:45 2019 +0100\n\n arm64: tlb: Ensure we execute an ISB following walk cache invalidation\n \n 05f2d2f83b5a (\"arm64: tlbflush: Introduce __flush_tlb_kernel_pgtable\")\n added a new TLB invalidation helper which is used when freeing\n intermediate levels of page table used for kernel mappings, but is\n missing the required ISB instruction after completion of the TLBI\n instruction.\n \n Add the missing barrier.\n \n Cc: \u003cstable@vger.kernel.org\u003e\n Fixes: 05f2d2f83b5a (\"arm64: tlbflush: Introduce __flush_tlb_kernel_pgtable\")\n Reviewed-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit d0b7a302d58abe24ed0f32a0672dd4c356bb73db\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Thu Aug 22 14:58:37 2019 +0100\n\n Revert \"arm64: Remove unnecessary ISBs from set_{pte,pmd,pud}\"\n \n This reverts commit 24fe1b0efad4fcdd32ce46cffeab297f22581707.\n \n Commit 24fe1b0efad4fcdd (\"arm64: Remove unnecessary ISBs from\n set_{pte,pmd,pud}\") removed ISB instructions immediately following updates\n to the page table, on the grounds that they are not required by the\n architecture and a DSB alone is sufficient to ensure that subsequent data\n accesses use the new translation:\n \n DDI0487E_a, B2-128:\n \n | ... no instruction that appears in program order after the DSB\n | instruction can alter any state of the system or perform any part of\n | its functionality until the DSB completes other than:\n |\n | * Being fetched from memory and decoded\n | * Reading the general-purpose, SIMD and floating-point,\n | Special-purpose, or System registers that are directly or indirectly\n | read without causing side-effects.\n \n However, the same document also states the following:\n \n DDI0487E_a, B2-125:\n \n | DMB and DSB instructions affect reads and writes to the memory system\n | generated by Load/Store instructions and data or unified cache\n | maintenance instructions being executed by the PE. Instruction fetches\n | or accesses caused by a hardware translation table access are not\n | explicit accesses.\n \n which appears to claim that the DSB alone is insufficient. Unfortunately,\n some CPU designers have followed the second clause above, whereas in Linux\n we\u0027ve been relying on the first. This means that our mapping sequence:\n \n MOV X0, \u003cvalid pte\u003e\n STR X0, [Xptep] // Store new PTE to page table\n DSB ISHST\n LDR X1, [X2] // Translates using the new PTE\n \n can actually raise a translation fault on the load instruction because the\n translation can be performed speculatively before the page table update and\n then marked as \"faulting\" by the CPU. For user PTEs, this is ok because we\n can handle the spurious fault, but for kernel PTEs and intermediate table\n entries this results in a panic().\n \n Revert the offending commit to reintroduce the missing barriers.\n \n Cc: \u003cstable@vger.kernel.org\u003e\n Fixes: 24fe1b0efad4fcdd (\"arm64: Remove unnecessary ISBs from set_{pte,pmd,pud}\")\n Reviewed-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit ebef746543fd1aa162216b0e484eb9062b65741d\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Aug 27 15:54:56 2019 +0100\n\n arm64: smp: Treat unknown boot failures as being \u0027stuck in kernel\u0027\n \n When we fail to bring a secondary CPU online and it fails in an unknown\n state, we should assume the worst and increment \u0027cpus_stuck_in_kernel\u0027\n so that things like kexec() are disabled.\n \n Reviewed-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 5b1cfe3a0ba74c1f2b83b607712a217b9f9463a2\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Aug 27 14:36:38 2019 +0100\n\n arm64: smp: Don\u0027t enter kernel with NULL stack pointer or task struct\n \n Although SMP bringup is inherently racy, we can significantly reduce\n the window during which secondary CPUs can unexpectedly enter the\n kernel by sanity checking the \u0027stack\u0027 and \u0027task\u0027 fields of the\n \u0027secondary_data\u0027 structure. If the booting CPU gave up waiting for us,\n then they will have been cleared to NULL and we should spin in a WFE; WFI\n loop instead.\n \n Reviewed-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 0e1645557d19fc6d88d3c40431f63a3c3a4c417b\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Aug 27 14:25:44 2019 +0100\n\n arm64: smp: Increase secondary CPU boot timeout value\n \n When many debug options are enabled simultaneously (e.g. PROVE_LOCKING,\n KMEMLEAK, DEBUG_PAGE_ALLOC, KASAN etc), it is possible for us to timeout\n when attempting to boot a secondary CPU and give up. Unfortunately, the\n CPU will /eventually/ appear, and sit in the background happily stuck\n in a recursive exception due to a NULL stack pointer.\n \n Increase the timeout to 5s, which will of course be enough for anybody.\n \n Reviewed-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 28abbba36a5a9f35bfaa75bf8ee0ebf2fe40b9e3\nAuthor: Bruno Thomsen \u003cbruno.thomsen@gmail.com\u003e\nDate: Tue Aug 27 16:36:56 2019 +0200\n\n rtc: pcf2127: bugfix: watchdog build dependency\n \n Disable watchdog registation when kernel is build without\n watchdog functionality, and enable watchdog core otherwise.\n This removes compile errors like the one below:\n \n drivers/rtc/rtc-pcf2127.o: in function `pcf2127_probe.constprop.3\u0027:\n rtc-pcf2127.c:(.text.unlikely+0x2c8): undefined reference to\n `devm_watchdog_register_device\u0027\n \n Watchdog feature in chip will always be configured as\n this is safe to do in both cases and minimize code churn.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Reported-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Fixes: bbc597561ce1 (\"rtc: pcf2127: add watchdog feature support\")\n Signed-off-by: Bruno Thomsen \u003cbruno.thomsen@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190827143656.4734-1-bruno.thomsen@gmail.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 03623b4b041cde396641399c343f52ce840d349d\nAuthor: Bruno Thomsen \u003cbruno.thomsen@gmail.com\u003e\nDate: Thu Aug 22 15:19:36 2019 +0200\n\n rtc: pcf2127: add tamper detection support\n \n Add support for integrated tamper detection function in both PCF2127 and\n PCF2129 chips. This patch implements the feature by adding an additional\n timestamp0 file to sysfs device path. This file contains seconds since\n epoch, if an event occurred, or is empty, if none occurred.\n Interface should match ISL1208 and RV3028 RTC drivers.\n \n Signed-off-by: Bruno Thomsen \u003cbruno.thomsen@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190822131936.18772-5-bruno.thomsen@gmail.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 0e735eaae1650a2e26835cc776f496e06a87be9f\nAuthor: Bruno Thomsen \u003cbruno.thomsen@gmail.com\u003e\nDate: Thu Aug 22 15:19:35 2019 +0200\n\n rtc: pcf2127: add watchdog feature support\n \n Add partial support for the watchdog functionality of\n both PCF2127 and PCF2129 chips.\n \n The programmable watchdog timer is currently using a fixed\n clock source of 1Hz. This result in a selectable range of\n 1-255 seconds, which covers most embedded Linux use-cases.\n \n Clock sources of 4096Hz, 64Hz and 1/60Hz is mostly useful\n in MCU use-cases.\n \n Countdown timer not available when using watchdog feature.\n \n Signed-off-by: Bruno Thomsen \u003cbruno.thomsen@gmail.com\u003e\n Acked-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Link: https://lore.kernel.org/r/20190822131936.18772-4-bruno.thomsen@gmail.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit a7bcae591f595a727feea9a5a389756015579072\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 15 11:52:18 2019 +0530\n\n of: unittest: Add of_node_put() before return\n \n The local variable np in function of_unittest_platform_populate takes\n the return value of of_find_node_by_path, which gets a node but does not\n put it. If np is not put before return it may cause a memory leak. Hence\n put np before a return statement.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit fd1a52f38c23c5d99fc2e6178ac4b74c5cbcf793\nAuthor: Markus Elfring \u003celfring@users.sourceforge.net\u003e\nDate: Wed Aug 21 19:30:09 2019 +0200\n\n RDMA/iwpm: Delete unnecessary checks before the macro call \"dev_kfree_skb\"\n \n The dev_kfree_skb() function performs also input parameter validation.\n Thus the test around the shown calls is not needed.\n \n This issue was detected by using the Coccinelle software.\n \n Link: https://lore.kernel.org/r/16df4c50-1f61-d7c4-3fc8-3073666d281d@web.de\n Signed-off-by: Markus Elfring \u003celfring@users.sourceforge.net\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 1bc5ba836e3ba02b8c7981a1fb453fe33513526d\nAuthor: Gal Pressman \u003cgalpress@amazon.com\u003e\nDate: Mon Aug 26 14:53:50 2019 +0300\n\n RDMA/efa: Use existing FIELD_SIZEOF macro\n \n Use FIELD_SIZEOF macro instead of hard coding it in field_avail macro.\n \n Link: https://lore.kernel.org/r/20190826115350.21718-3-galpress@amazon.com\n Reviewed-by: Daniel Kranzdorf \u003cdkkranzd@amazon.com\u003e\n Reviewed-by: Firas JahJah \u003cfirasj@amazon.com\u003e\n Signed-off-by: Gal Pressman \u003cgalpress@amazon.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 958b6813f0c077c45a36d0a10b5bdcd27216eabe\nAuthor: Gal Pressman \u003cgalpress@amazon.com\u003e\nDate: Mon Aug 26 14:53:49 2019 +0300\n\n RDMA/efa: Remove umem check on dereg MR flow\n \n EFA driver is not a kverbs provider, the check for MR umem is redundant.\n \n Link: https://lore.kernel.org/r/20190826115350.21718-2-galpress@amazon.com\n Reviewed-by: Firas JahJah \u003cfirasj@amazon.com\u003e\n Reviewed-by: Yossi Leybovich \u003csleybo@amazon.com\u003e\n Signed-off-by: Gal Pressman \u003cgalpress@amazon.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 476646e09668cc85138adec726a94f76ca83740d\nAuthor: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\nDate: Tue Aug 13 16:45:18 2019 +0300\n\n dt-bindings: smiapp: Align documentation with current practices\n \n - The clock-lanes property is not needed for the sensors do not support\n lane reordering. (The information possibly present in existing clock-lane\n properties is simply not used.)\n \n - There\u0027s no need to refer to the sensor device in the DT example, thus\n remove the label.\n \n - Rename the \"camera\" device node as \"camera-sensor\".\n \n - Rename the endpoint label as \"smiapp_ep\" (was: \"smiapp_1_1\"). There is\n in practice only one anyway.\n \n - Remove the remote-endpoint documentation (it is covered by\n graph.txt to which video-interfaces.txt refers to).\n \n - Add a note on the port and endpoint nodes.\n \n These changes make the smiapp bindings a better example.\n \n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 34be1d84763280ac13e3e34afc79eb35a262c9ea\nAuthor: Nikolay Borisov \u003cnborisov@suse.com\u003e\nDate: Tue Aug 27 14:46:29 2019 +0300\n\n btrfs: Make btrfs_find_name_in_ext_backref return struct btrfs_inode_extref\n \n btrfs_find_name_in_ext_backref returns either 0/1 depending on whether it\n found a backref for the given name. If it returns true then the actual\n inode_ref struct is returned in one of its parameters. That\u0027s pointless,\n instead refactor the function such that it returns either a pointer\n to the btrfs_inode_extref or NULL it it didn\u0027t find anything. This\n streamlines the function calling convention.\n \n Signed-off-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 45e9c829ebeaa13b3b999f76963b9920d2147128\nAuthor: Michel Thierry \u003cmichel.thierry@intel.com\u003e\nDate: Fri Aug 23 01:20:50 2019 -0700\n\n drm/i915/tgl/perf: use the same oa ctx_id format as icl\n \n Compared to Icelake, Tigerlake\u0027s MAX_CONTEXT_HW_ID is smaller by one, but\n since we just use the upper 32 bits of the lrc_desc, it\u0027s guaranteed OA\n will use the correct one.\n \n Cc: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Signed-off-by: Michel Thierry \u003cmichel.thierry@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Umesh Nerlige Ramappa \u003cumesh.nerlige.ramappa@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823082055.5992-19-lucas.demarchi@intel.com\n\ncommit a8ff5d405e90e7044ab45f315f2dbcbe6aebf5f8\nAuthor: Michel Thierry \u003cmichel.thierry@intel.com\u003e\nDate: Fri Aug 23 01:20:48 2019 -0700\n\n drm/i915/tgl: Do not apply WaIncreaseDefaultTLBEntries from GEN12 onwards\n \n Workaround no longer needed (plus L3_LRA_1_GPGPU doesn\u0027t exist).\n \n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Signed-off-by: Michel Thierry \u003cmichel.thierry@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823082055.5992-17-lucas.demarchi@intel.com\n\ncommit 99389390fef50c880d2ca0b6667c12423b3d5260\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Fri Aug 23 01:20:47 2019 -0700\n\n drm/i915/tgl: Implement TGL DisplayPort training sequence\n \n On TGL some registers moved from DDI to transcoder and the\n DisplayPort training sequence has a separate BSpec page.\n \n I started adding \u0027ifs\u0027 to the original intel_ddi_pre_enable_dp() but\n it was becoming really hard to follow, so a new and cleaner function\n for TGL was added with comments of all steps. It\u0027s similar to ICL,\n but different enough to deserve a new function.\n \n The rest of DisplayPort enable and the whole disable sequences\n remained the same.\n \n v2: FEC and DSC should be enabled on sink side before start link\n training(Maarten reported and Manasi confirmed the DSC part)\n \n v3: Add call to enable FEC on step 7.l(Manasi)\n \n BSpec: 49190\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Manasi Navare \u003cmanasi.d.navare@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823082055.5992-16-lucas.demarchi@intel.com\n\ncommit 9c722e17c1b91d7dc5ec77cb4fcaaa6b30653fe3\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Fri Aug 23 01:20:44 2019 -0700\n\n drm/i915: Disable pipes in reverse order\n \n Disable CRTC/pipes in reverse order because some features (MST in\n TGL+) requires master and slave relationship between pipes, so it\n should always pick the lowest pipe as master as it will be enabled\n first and disable in the reverse order so the master will be the last\n one to be disabled.\n \n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Mika Kahola \u003cmika.kahola@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823082055.5992-13-lucas.demarchi@intel.com\n\ncommit 8ec107de22f057c9e300df720d492589780e8da3\nAuthor: Nikolay Borisov \u003cnborisov@suse.com\u003e\nDate: Tue Aug 27 14:46:28 2019 +0300\n\n btrfs: Make btrfs_find_name_in_backref return btrfs_inode_ref struct\n \n btrfs_find_name_in_backref returns either 0/1 depending on whether it\n found a backref for the given name. If it returns true then the actual\n inode_ref struct is returned in one of its parameters. That\u0027s pointless,\n instead refactor the function such that it returns either a pointer\n to the btrfs_inode_ref or NULL it it didn\u0027t find anything. This\n streamlines the function calling convention.\n \n Signed-off-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit c072b0c24e6b8a1951b26b3a575e9e0491bd7bfe\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Tue Aug 27 22:59:45 2019 +0800\n\n drm/amdgpu: fix GFXOFF on Picasso and Raven2\n \n For picasso(adev-\u003epdev-\u003edevice \u003d\u003d 0x15d8)\u0026raven2(adev-\u003erev_id \u003e\u003d 0x8),\n firmware is sufficient to support gfxoff.\n In commit 98f58ada2d37e, for picasso\u0026raven2,\n return directly and cause gfxoff disabled.\n \n Fixes: 98f58ada2d37 (\"drm/amdgpu/gfx9: update pg_flags after determining if gfx off is possible\")\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 3e627fe6d56a4bbed9177ee20229c6aae61e11c5\nAuthor: John Wang \u003cwangzqbj@inspur.com\u003e\nDate: Sat Aug 10 17:58:36 2019 +0800\n\n dt-bindings: Add vendor prefix for Inspur Corporation\n \n Signed-off-by: John Wang \u003cwangzqbj@inspur.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 0456417ef680aab9a2aa93b950443e25104101b9\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Fri Aug 23 01:20:43 2019 -0700\n\n drm: Add for_each_oldnew_intel_crtc_in_state_reverse()\n \n Same as for_each_oldnew_intel_crtc_in_state() but iterates in reverse\n order.\n \n v2: Fix additional blank line\n v3: Rebase\n \n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Mika Kahola \u003cmika.kahola@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823082055.5992-12-lucas.demarchi@intel.com\n\ncommit f7b3c22619a44ce7599fc9fa1df5df6281a26217\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Fri Aug 23 01:20:41 2019 -0700\n\n drm/i915/tgl: Add maximum resolution supported by PSR2 HW\n \n TGL PSR2 HW supports a bigger resolution, so lets add it\n \n BSpec: 50422, 49199\n Cc: Dhinakaran Pandiyan \u003cdhinakaran.pandiyan@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Anshuman Gupta \u003canshuman.gupta@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823082055.5992-10-lucas.demarchi@intel.com\n\ncommit 0f81e645eb1ed224af43bb7b347806cc74d189f6\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Sat Aug 17 02:38:33 2019 -0700\n\n drm/i915: Do not read PSR2 register in transcoders without PSR2\n \n This fix unclaimed access warnings:\n \n [ 245.525788] ------------[ cut here ]------------\n [ 245.525884] Unclaimed read from register 0x62900\n [ 245.526154] WARNING: CPU: 0 PID: 1234 at drivers/gpu/drm/i915/intel_uncore.c:1100 __unclaimed_reg_debug+0x40/0x50 [i915]\n [ 245.526160] Modules linked in: i915 x86_pkg_temp_thermal ax88179_178a coretemp usbnet crct10dif_pclmul mii crc32_pclmul ghash_clmulni_intel e1000e [last unloaded: i915]\n [ 245.526191] CPU: 0 PID: 1234 Comm: kms_fullmodeset Not tainted 5.1.0-rc6+ #915\n [ 245.526197] Hardware name: Intel Corporation Tiger Lake Client Platform/TigerLake U DDR4 SODIMM RVP, BIOS TGLSFWR1.D00.2081.A10.1904182155 04/18/2019\n [ 245.526273] RIP: 0010:__unclaimed_reg_debug+0x40/0x50 [i915]\n [ 245.526281] Code: 74 05 5b 5d 41 5c c3 45 84 e4 48 c7 c0 76 97 21 a0 48 c7 c6 6c 97 21 a0 89 ea 48 0f 44 f0 48 c7 c7 7f 97 21 a0 e8 4f 1e fe e0 \u003c0f\u003e 0b 83 2d 6f d9 1c 00 01 5b 5d 41 5c c3 66 90 41 57 41 56 41 55\n [ 245.526288] RSP: 0018:ffffc900006bf7d8 EFLAGS: 00010086\n [ 245.526297] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000\n [ 245.526304] RDX: 0000000000000007 RSI: 0000000000000000 RDI: 00000000ffffffff\n [ 245.526310] RBP: 0000000000061900 R08: 0000000000000000 R09: 0000000000000001\n [ 245.526317] R10: 0000000000000006 R11: 0000000000000000 R12: 0000000000000001\n [ 245.526324] R13: 0000000000000000 R14: ffff8882914f0d58 R15: 0000000000000206\n [ 245.526332] FS: 00007fed2a3c39c0(0000) GS:ffff8882a8600000(0000) knlGS:0000000000000000\n [ 245.526340] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n [ 245.526347] CR2: 00007fed28dff000 CR3: 00000002a086c006 CR4: 0000000000760ef0\n [ 245.526354] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\n [ 245.526361] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\n [ 245.526367] PKRU: 55555554\n [ 245.526373] Call Trace:\n [ 245.526454] gen11_fwtable_read32+0x219/0x250 [i915]\n [ 245.526576] intel_psr_activate+0x57/0x400 [i915]\n [ 245.526697] intel_psr_enable_locked+0x367/0x4b0 [i915]\n [ 245.526828] intel_psr_enable+0xa4/0xd0 [i915]\n [ 245.526946] intel_enable_ddi+0x127/0x2f0 [i915]\n [ 245.527075] intel_encoders_enable.isra.79+0x62/0x90 [i915]\n [ 245.527202] haswell_crtc_enable+0x2a2/0x850 [i915]\n [ 245.527337] intel_update_crtc+0x51/0x360 [i915]\n [ 245.527466] skl_update_crtcs+0x26c/0x300 [i915]\n [ 245.527603] intel_atomic_commit_tail+0x3e5/0x13c0 [i915]\n [ 245.527757] intel_atomic_commit+0x24d/0x2d0 [i915]\n [ 245.527782] drm_atomic_helper_set_config+0x7b/0x90\n [ 245.527799] drm_mode_setcrtc+0x1b4/0x6f0\n [ 245.527856] ? drm_mode_getcrtc+0x180/0x180\n [ 245.527867] drm_ioctl_kernel+0xad/0xf0\n [ 245.527886] drm_ioctl+0x2f4/0x3b0\n [ 245.527902] ? drm_mode_getcrtc+0x180/0x180\n [ 245.527935] ? rcu_read_lock_sched_held+0x6f/0x80\n [ 245.527956] do_vfs_ioctl+0xa0/0x6d0\n [ 245.527970] ? __task_pid_nr_ns+0xb6/0x200\n [ 245.527991] ksys_ioctl+0x35/0x70\n [ 245.528009] __x64_sys_ioctl+0x11/0x20\n [ 245.528020] do_syscall_64+0x55/0x180\n [ 245.528034] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n [ 245.528042] RIP: 0033:0x7fed2cc7c3c7\n [ 245.528050] Code: 00 00 90 48 8b 05 c9 3a 0d 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 b8 10 00 00 00 0f 05 \u003c48\u003e 3d 01 f0 ff ff 73 01 c3 48 8b 0d 99 3a 0d 00 f7 d8 64 89 01 48\n [ 245.528057] RSP: 002b:00007ffe36944378 EFLAGS: 00000246 ORIG_RAX: 0000000000000010\n [ 245.528067] RAX: ffffffffffffffda RBX: 00007ffe369443b0 RCX: 00007fed2cc7c3c7\n [ 245.528074] RDX: 00007ffe369443b0 RSI: 00000000c06864a2 RDI: 0000000000000003\n [ 245.528081] RBP: 00007ffe369443b0 R08: 0000000000000000 R09: 0000564c0173ae98\n [ 245.528088] R10: 0000564c0173aeb8 R11: 0000000000000246 R12: 00000000c06864a2\n [ 245.528095] R13: 0000000000000003 R14: 0000000000000000 R15: 0000000000000000\n [ 245.528128] irq event stamp: 140866\n [ 245.528138] hardirqs last enabled at (140865): [\u003cffffffff819a63dc\u003e] _raw_spin_unlock_irqrestore+0x4c/0x60\n [ 245.528148] hardirqs last disabled at (140866): [\u003cffffffff819a624d\u003e] _raw_spin_lock_irqsave+0xd/0x50\n [ 245.528158] softirqs last enabled at (140860): [\u003cffffffff81c0038c\u003e] __do_softirq+0x38c/0x499\n [ 245.528170] softirqs last disabled at (140853): [\u003cffffffff810b4a09\u003e] irq_exit+0xa9/0xc0\n [ 245.528247] WARNING: CPU: 0 PID: 1234 at drivers/gpu/drm/i915/intel_uncore.c:1100 __unclaimed_reg_debug+0x40/0x50 [i915]\n [ 245.528254] ---[ end trace 366069676e98a410 ]---\n \n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823082055.5992-7-lucas.demarchi@intel.com\n\ncommit 6056517ab8c30fb239b08c06ef577677cbcd409c\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Fri Aug 23 01:20:37 2019 -0700\n\n drm/i915/tgl: Guard and warn if more than one eDP panel is present\n \n On TGL+ it\u0027s possible to have PSR1 enabled in other ports besides DDIA.\n PSR2 is still limited to DDIA. However currently we handle only one\n instance of PSR struct. Lets guard intel_psr_init_dpcd() against\n multiple eDP panels and warn about it.\n \n v2: Reword commit message to be TGL+ only and with the info where\n PSR1/PSR2 are supported (Lucas)\n \n Cc: Dhinakaran Pandiyan \u003cdhinakaran.pandiyan@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Anshuman Gupta \u003canshuman.gupta@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823082055.5992-6-lucas.demarchi@intel.com\n\ncommit cccdce1dd061c0b8ae156f026a3ee2c9d58613d3\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Aug 27 14:59:35 2019 +0100\n\n drm/i915: Make engine\u0027s batch pool safe for use with virtual engines\n \n A virtual engine itself does not have a batch pool, but we can gleefully\n use any of its siblings instead.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190827135935.3831-1-chris@chris-wilson.co.uk\n\ncommit f52c6d0df6909a0412c347c8e442acc22ce94747\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Aug 27 14:26:31 2019 +0100\n\n drm/i915: Only activate i915_active debugobject once\n \n The point of debug_object_activate is to mark the first, and only the\n first, acquisition. The object then remains active until the last\n release. However, we marked up all successful first acquires even though\n we allowed concurrent parties to try and acquire the i915_active\n simultaneously (serialised by the i915_active.mutex).\n \n Testcase: igt/gem_mmap_gtt/fault-concurrent\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190827132631.18627-1-chris@chris-wilson.co.uk\n\ncommit d0a8d877da976c244092ce859683b2fa116217db\nAuthor: Ander Juaristi \u003ca@juaristi.eus\u003e\nDate: Sat Aug 17 13:26:52 2019 +0200\n\n netfilter: nft_dynset: support for element deletion\n \n This patch implements the delete operation from the ruleset.\n \n It implements a new delete() function in nft_set_rhash. It is simpler\n to use than the already existing remove(), because it only takes the set\n and the key as arguments, whereas remove() expects a full\n nft_set_elem structure.\n \n Signed-off-by: Ander Juaristi \u003ca@juaristi.eus\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit 97b27821b4854ca744946dae32a3f2fd55bcd5bc\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Mon Aug 26 09:06:56 2019 -0700\n\n writeback, memcg: Implement foreign dirty flushing\n \n There\u0027s an inherent mismatch between memcg and writeback. The former\n trackes ownership per-page while the latter per-inode. This was a\n deliberate design decision because honoring per-page ownership in the\n writeback path is complicated, may lead to higher CPU and IO overheads\n and deemed unnecessary given that write-sharing an inode across\n different cgroups isn\u0027t a common use-case.\n \n Combined with inode majority-writer ownership switching, this works\n well enough in most cases but there are some pathological cases. For\n example, let\u0027s say there are two cgroups A and B which keep writing to\n different but confined parts of the same inode. B owns the inode and\n A\u0027s memory is limited far below B\u0027s. A\u0027s dirty ratio can rise enough\n to trigger balance_dirty_pages() sleeps but B\u0027s can be low enough to\n avoid triggering background writeback. A will be slowed down without\n a way to make writeback of the dirty pages happen.\n \n This patch implements foreign dirty recording and foreign mechanism so\n that when a memcg encounters a condition as above it can trigger\n flushes on bdi_writebacks which can clean its pages. Please see the\n comment on top of mem_cgroup_track_foreign_dirty_slowpath() for\n details.\n \n A reproducer follows.\n \n write-range.c::\n \n #include \u003cstdio.h\u003e\n #include \u003cstdlib.h\u003e\n #include \u003cunistd.h\u003e\n #include \u003cfcntl.h\u003e\n #include \u003csys/types.h\u003e\n \n static const char *usage \u003d \"write-range FILE START SIZE\\n\";\n \n int main(int argc, char **argv)\n {\n int fd;\n unsigned long start, size, end, pos;\n char *endp;\n char buf[4096];\n \n if (argc \u003c 4) {\n fprintf(stderr, usage);\n return 1;\n }\n \n fd \u003d open(argv[1], O_WRONLY);\n if (fd \u003c 0) {\n perror(\"open\");\n return 1;\n }\n \n start \u003d strtoul(argv[2], \u0026endp, 0);\n if (*endp !\u003d \u0027\\0\u0027) {\n fprintf(stderr, usage);\n return 1;\n }\n \n size \u003d strtoul(argv[3], \u0026endp, 0);\n if (*endp !\u003d \u0027\\0\u0027) {\n fprintf(stderr, usage);\n return 1;\n }\n \n end \u003d start + size;\n \n while (1) {\n for (pos \u003d start; pos \u003c end; ) {\n long bread, bwritten \u003d 0;\n \n if (lseek(fd, pos, SEEK_SET) \u003c 0) {\n perror(\"lseek\");\n return 1;\n }\n \n bread \u003d read(0, buf, sizeof(buf) \u003c end - pos ?\n sizeof(buf) : end - pos);\n if (bread \u003c 0) {\n perror(\"read\");\n return 1;\n }\n if (bread \u003d\u003d 0)\n return 0;\n \n while (bwritten \u003c bread) {\n long this;\n \n this \u003d write(fd, buf + bwritten,\n bread - bwritten);\n if (this \u003c 0) {\n perror(\"write\");\n return 1;\n }\n \n bwritten +\u003d this;\n pos +\u003d bwritten;\n }\n }\n }\n }\n \n repro.sh::\n \n #!/bin/bash\n \n set -e\n set -x\n \n sysctl -w vm.dirty_expire_centisecs\u003d300000\n sysctl -w vm.dirty_writeback_centisecs\u003d300000\n sysctl -w vm.dirtytime_expire_seconds\u003d300000\n echo 3 \u003e /proc/sys/vm/drop_caches\n \n TEST\u003d/sys/fs/cgroup/test\n A\u003d$TEST/A\n B\u003d$TEST/B\n \n mkdir -p $A $B\n echo \"+memory +io\" \u003e $TEST/cgroup.subtree_control\n echo $((1\u003c\u003c30)) \u003e $A/memory.high\n echo $((32\u003c\u003c30)) \u003e $B/memory.high\n \n rm -f testfile\n touch testfile\n fallocate -l 4G testfile\n \n echo \"Starting B\"\n \n (echo $BASHPID \u003e $B/cgroup.procs\n pv -q --rate-limit 70M \u003c /dev/urandom | ./write-range testfile $((2\u003c\u003c30)) $((2\u003c\u003c30))) \u0026\n \n echo \"Waiting 10s to ensure B claims the testfile inode\"\n sleep 5\n sync\n sleep 5\n sync\n echo \"Starting A\"\n \n (echo $BASHPID \u003e $A/cgroup.procs\n pv \u003c /dev/urandom | ./write-range testfile 0 $((2\u003c\u003c30)))\n \n v2: Added comments explaining why the specific intervals are being used.\n \n v3: Use 0 @nr when calling cgroup_writeback_by_id() to use best-effort\n flushing while avoding possible livelocks.\n \n v4: Use get_jiffies_64() and time_before/after64() instead of raw\n jiffies_64 and arthimetic comparisons as suggested by Jan.\n \n Reviewed-by: Jan Kara \u003cjack@suse.cz\u003e\n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit d62241c7a406f0680d702bd974f6f17e28ab8e5d\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Mon Aug 26 09:06:55 2019 -0700\n\n writeback, memcg: Implement cgroup_writeback_by_id()\n \n Implement cgroup_writeback_by_id() which initiates cgroup writeback\n from bdi and memcg IDs. This will be used by memcg foreign inode\n flushing.\n \n v2: Use wb_get_lookup() instead of wb_get_create() to avoid creating\n spurious wbs.\n \n v3: Interpret 0 @nr as 1.25 * nr_dirty to implement best-effort\n flushing while avoding possible livelocks.\n \n Reviewed-by: Jan Kara \u003cjack@suse.cz\u003e\n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit ed288dc0d4aa29f65bd25b31b5cb866aa5664ff9\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Mon Aug 26 09:06:54 2019 -0700\n\n writeback: Separate out wb_get_lookup() from wb_get_create()\n \n Separate out wb_get_lookup() which doesn\u0027t try to create one if there\n isn\u0027t already one from wb_get_create(). This will be used by later\n patches.\n \n Reviewed-by: Jan Kara \u003cjack@suse.cz\u003e\n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 34f8fe501f0624de115d087680c84000b5d9abc9\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Mon Aug 26 09:06:53 2019 -0700\n\n bdi: Add bdi-\u003eid\n \n There currently is no way to universally identify and lookup a bdi\n without holding a reference and pointer to it. This patch adds an\n non-recycling bdi-\u003eid and implements bdi_get_by_id() which looks up\n bdis by their ids. This will be used by memcg foreign inode flushing.\n \n I left bdi_list alone for simplicity and because while rb_tree does\n support rcu assignment it doesn\u0027t seem to guarantee lossless walk when\n walk is racing aginst tree rebalance operations.\n \n Reviewed-by: Jan Kara \u003cjack@suse.cz\u003e\n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 5b9cce4c7eb0696558dfd4946074ae1fb9d8f05d\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Mon Aug 26 09:06:52 2019 -0700\n\n writeback: Generalize and expose wb_completion\n \n wb_completion is used to track writeback completions. We want to use\n it from memcg side for foreign inode flushes. This patch updates it\n to remember the target waitq instead of assuming bdi-\u003ewb_waitq and\n expose it outside of fs-writeback.c.\n \n Reviewed-by: Jan Kara \u003cjack@suse.cz\u003e\n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 2ddb134d97633e96b8020fd054e00ecddfdbb5f1\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Tue Aug 27 15:44:25 2019 +0900\n\n drm/amd: remove meaningless descending into amd/amdkfd/\n \n Since commit 04d5e2765802 (\"drm/amdgpu: Merge amdkfd into amdgpu\"),\n drivers/gpu/drm/amd/amdkfd/Makefile does not contain any syntax that\n is understood by the build system.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c7b33cfb3ca57f80ff4eaa55c6022b8a4c34d6ea\nAuthor: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\nDate: Tue Aug 27 17:33:32 2019 +0800\n\n drm/amdgpu: Add APTX quirk for Dell Latitude 5495\n \n Needs ATPX rather than _PR3 to really turn off the dGPU. This can save\n ~5W when dGPU is runtime-suspended.\n \n Signed-off-by: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b202c0ff184a6203779e32aeb46b50e32e92d7ee\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Tue Aug 27 15:09:25 2019 +0800\n\n drm/amd/display: remove unused function setFieldWithMask\n \n After commit a9f54ce3c603 (\"drm/amd/display: Refactoring VTEM\"),\n there is no caller in tree.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Reviewed-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 57be797c484ad8c7593c75f878322263db19a989\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Wed Aug 21 16:35:12 2019 +0800\n\n drm/amd/powerplay: correct the pp_feature output on Arcturus\n \n Fix for the commit below:\n drm/amd/powerplay: implment sysfs feature status function in smu\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 08d544ab54dfcc4baa17082dadf10ab1bb6f8c7e\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Mon Aug 19 13:17:53 2019 +0800\n\n drm/amd/powerplay: correct Vega20 dpm level related settings\n \n Correct the settings for auto mode and skip the unnecessary\n settings for dcefclk and fclk.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 644763224169e9ca2e3010c24bf4b81a6be64959\nAuthor: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\nDate: Thu Aug 1 00:32:33 2019 +0200\n\n ARM: dts: kirkwood: ts219: disable the SoC\u0027s RTC\n \n The internal RTC doesn\u0027t work, loading the driver only yields\n \n rtc-mv f1010300.rtc: internal RTC not ticking\n \n . So disable it.\n \n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\n Acked-by: Martin Michlmayr \u003ctbm@cyrius.com\u003e\n Acked-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Signed-off-by: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n\ncommit c00bc38354cf81ce83b678ff13ecf01e75d0e8da\nAuthor: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\nDate: Tue Sep 11 22:57:26 2018 +0200\n\n arm64: dts: marvell: Add cpu clock node on Armada 7K/8K\n \n Add cpu clock node on AP\n \n Signed-off-by: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n\ncommit dd23e1d566d0f74aa3b68ab3237927bb15f0e644\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Tue Aug 27 16:37:50 2019 +0200\n\n ALSA: hda - Allow runtime PM for controller if component notifier is used\n \n Currently we disallow the runtime PM of the HD-audio controller if\n it\u0027s bound with HDMI/DP on Nvidia / AMD unless it\u0027s for dGPU. This is\n for keeping the link up to get the proper notification for ELD\n hotplug.\n \n As explained in the commit 37a3a98ef601 (\"ALSA: hda - Enable runtime\n PM only for discrete GPU\"), this keep-power-up behavior is rather a\n stop-gap solution until the ELD notification via audio component.\n And now we finally got the audio component for these graphics drivers\n via commit ade49db337a9 (\"ALSA: hda/hdmi - Allow audio component for\n AMD/ATI and Nvidia HDMI\"), so it\u0027s time to change.\n \n This patch makes HD-audio controller again runtime-suspendable when\n the device gets bound with audio component in HDMI codec driver. For\n making it easier to access from the codec driver, move the flag into\n the common hda_bus object instead of hda_intel flag. Also rename it\n to keep_power, to indicate the actual meaning.\n \n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit b1a5a73e64e99faa5f4deef2ae96d7371a0fb5d0\nAuthor: Ming Lei \u003cming.lei@redhat.com\u003e\nDate: Fri Aug 16 10:28:49 2019 +0800\n\n genirq/affinity: Spread vectors on node according to nr_cpu ratio\n \n Now __irq_build_affinity_masks() spreads vectors evenly per node, but there\n is a case that not all vectors have been spread when each numa node has a\n different number of CPUs which triggers the warning in the spreading code.\n \n Improve the spreading algorithm by\n \n - assigning vectors according to the ratio of the number of CPUs on a node\n to the number of remaining CPUs.\n \n - running the assignment from smaller nodes to bigger nodes to guarantee\n that every active node gets allocated at least one vector.\n \n This ensures that all vectors are spread out. Asided of that the spread\n becomes more fair if the nodes have different number of CPUs.\n \n For example, on the following machine:\n CPU(s): 16\n On-line CPU(s) list: 0-15\n Thread(s) per core: 1\n Core(s) per socket: 8\n Socket(s): 2\n NUMA node(s): 2\n ...\n NUMA node0 CPU(s): 0,1,3,5-9,11,13-15\n NUMA node1 CPU(s): 2,4,10,12\n \n When a driver requests to allocate 8 vectors, the following spread results:\n \n irq 31, cpu list 2,4\n irq 32, cpu list 10,12\n irq 33, cpu list 0-1\n irq 34, cpu list 3,5\n irq 35, cpu list 6-7\n irq 36, cpu list 8-9\n irq 37, cpu list 11,13\n irq 38, cpu list 14-15\n \n So Node 0 has now 6 and Node 1 has 2 vectors assigned. The original\n algorithm assigned 4 vectors on each node which was unfair versus Node 0.\n \n [ tglx: Massaged changelog ]\n \n Reported-by: Jon Derrick \u003cjonathan.derrick@intel.com\u003e\n Signed-off-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Keith Busch \u003ckbusch@kernel.org\u003e\n Reviewed-by: Jon Derrick \u003cjonathan.derrick@intel.com\u003e\n Link: https://lkml.kernel.org/r/20190816022849.14075-3-ming.lei@redhat.com\n\ncommit 53c1788b7d7720565214a466afffdc818d8c6e5f\nAuthor: Ming Lei \u003cming.lei@redhat.com\u003e\nDate: Fri Aug 16 10:28:48 2019 +0800\n\n genirq/affinity: Improve __irq_build_affinity_masks()\n \n One invariant of __irq_build_affinity_masks() is that all CPUs in the\n specified masks (cpu_mask AND node_to_cpumask for each node) should be\n covered during the spread. Even though all requested vectors have been\n reached, it\u0027s still required to spread vectors among remained CPUs. A\n similar policy has been taken in case of \u0027numvecs \u003c\u003d nodes\u0027 already.\n \n So remove the following check inside the loop:\n \n if (done \u003e\u003d numvecs)\n break;\n \n Meantime assign at least 1 vector for remaining nodes if \u0027numvecs\u0027 vectors\n have been handled already.\n \n Also, if the specified cpumask for one numa node is empty, simply do not\n spread vectors on this node.\n \n Signed-off-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190816022849.14075-2-ming.lei@redhat.com\n\ncommit 96018a6fafb158036372cfd9e6046b352485f143\nAuthor: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nDate: Wed Jul 31 14:21:26 2019 +0200\n\n arm64: dts: marvell: Convert 7k/8k usb-phy properties to phy-supply\n \n Update Aramda 7k/8k DTs to use the phy-supply property of the (recent)\n generic PHY framework instead of the (legacy) usb-phy preperty. Both\n enable the supply when the PHY is enabled.\n \n The COMPHY nodes only provide SERDES lanes configuration. The power\n supply that is represented by the phy-supply property is just a\n regulator wired to the USB connector, hence the creation of connector\n nodes as child of the COMPHY nodes and the supply attached to it.\n \n Cc: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n\ncommit ce55522c035e98803832eff8938f621f25b0f4f1\nAuthor: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nDate: Wed Jul 31 14:21:25 2019 +0200\n\n arm64: dts: marvell: Add 7k/8k PHYs in PCIe nodes\n \n Fill-in the missing PCIe phys/phy-names DT properties of Armada 7k/8k\n based boards.\n \n The MacchiatoBin is a bit particular as the Armada8k-PCI IP supports\n x4 link widths and in this case the PHY for each lane must be\n referenced.\n \n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n\ncommit 01d0deba28f6dc3b31c505dccb9559ee19dfade3\nAuthor: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nDate: Wed Jul 31 14:21:24 2019 +0200\n\n arm64: dts: marvell: Add 7k/8k PHYs in USB3 nodes\n \n Fill-in the missing USB3 phys/phy-names DT properties of Armada 7k/8k\n based boards. Only update nodes actually enabling USB3 in the default\n (mainline) configuration. A few USB nodes are enabled but there is\n only USB2 working on them.\n \n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n\ncommit 691bac9d093b13abf39f95bd82db0430a152246c\nAuthor: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\nDate: Wed May 1 10:57:14 2019 -0400\n\n drm/amdgpu: Vega20 SMU I2C HW engine controller.\n \n Implement HW I2C enigne controller to be used by the RAS EEPROM\n table manager. This is based on code from ATITOOLs.\n \n v2:\n Rename the file and all function prefixes to smu_v11_0_i2c\n \n By Luben\u0027s observation always fill the TX fifo to full so\n we don\u0027t have garbadge interpreted by the slave as valid data.\n \n v3:\n Remove preemption disable as the HW I2C controller will not\n stop the clock on empty TX fifo and so it\u0027s not critical to\n keep not empty queue.\n Switch to fast mode 400 khz SCL clock for faster read and write.\n \n v5:\n Restore clock gating before releasing I2C bus and fix some\n style comments.\n \n v6:\n squash in warning fix, fix includes (Alex)\n \n Signed-off-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Reviewed-by: Luben Tuikov \u003cLuben.Tuikov@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d04abe999e20a57c96b305d329e480ffe057c9ec\nAuthor: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nDate: Wed Jul 31 14:21:23 2019 +0200\n\n arm64: dts: marvell: Add 7k/8k per-port PHYs in SATA nodes\n \n Fill-in the missing SATA phys/phy-names DT properties of Armada 7k/8k\n based boards.\n \n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n\ncommit 93ab25e6f5da60c7678bf54746836f83e7729284\nAuthor: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nDate: Wed Jul 31 14:21:22 2019 +0200\n\n arm64: dts: marvell: Add CP110 COMPHY clocks\n \n Declare the three clocks feeding the COMPHY block.\n \n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n\ncommit e8cac0556158506f5f522b88cc0c85b9a672bbbf\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Wed Aug 21 20:05:32 2019 +0200\n\n btrfs: move dev_stats helpers to volumes.c\n \n The other dev stats functions are already there and the helpers are not\n used by anything else.\n \n Reviewed-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit dd3e0904bbdd3a3c3e0d1426c40642d6dff96491\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Wed Aug 21 19:57:04 2019 +0200\n\n btrfs: move struct io_ctl to free-space-cache.h\n \n The io_ctl structure is used for free space management, and used only by\n the v1 space cache code, but unfortunatlly the full definition is\n required by block-group.h so it can\u0027t be moved to free-space-cache.c\n without additional changes.\n \n Reviewed-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 9ae853f5bfbb2b14fa1e905560f81e19941da4ab\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Wed Aug 21 19:12:59 2019 +0200\n\n btrfs: move functions for tree compare to send.c\n \n Send is the only user of tree_compare, we can move it there along with\n the other helpers and definitions.\n \n Reviewed-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit cf8805e3a2f041c5d8274ad5c3a6b349d5c523f6\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Wed Aug 21 19:16:27 2019 +0200\n\n btrfs: rename and export read_node_slot\n \n Preparatory work for code that will be moved out of ctree and uses this\n function.\n \n Reviewed-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 445cea2a13763856f451933425a941c87844185d\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Wed Aug 21 19:06:17 2019 +0200\n\n btrfs: move private raid56 definitions from ctree.h\n \n Reviewed-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 8d517ca01b0895ebef7054d805b2f63cfc6dcbcf\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Wed Aug 21 18:54:28 2019 +0200\n\n btrfs: move math functions to misc.h\n \n Reviewed-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 9f69c146911776d0892a0b97e2db61d60531d27d\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Wed Aug 21 18:48:25 2019 +0200\n\n btrfs: move cond_wake_up functions out of ctree\n \n The file ctree.h serves as a header for everything and has become quite\n bloated. Split some helpers that are generic and create a new file that\n should be the catch-all for code that\u0027s not btrfs-specific.\n \n Reviewed-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 535462c244f954eb96b95dc186db7360f7e6a471\nAuthor: Marek Behún \u003cmarek.behun@nic.cz\u003e\nDate: Mon Aug 12 18:30:03 2019 +0200\n\n arm64: dts: marvell: armada-37xx: add mailbox node\n \n This adds the rWTM BIU mailbox node for communication with the secure\n processor. The driver already exists in\n drivers/mailbox/armada-37xx-rwtm-mailbox.c.\n \n Signed-off-by: Marek Behún \u003cmarek.behun@nic.cz\u003e\n Cc: Gregory Clement \u003cgregory.clement@bootlin.com\u003e\n Cc: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n\ncommit 04f718389ab15d49138a59f921cab1e0c25af98b\nAuthor: Ramuthevar Vadivel Muruganx \u003cvadivel.muruganx.ramuthevar@linux.intel.com\u003e\nDate: Mon Aug 26 15:28:00 2019 +0800\n\n mmc: sdhci-of-arasan: Add Support for Intel LGM eMMC\n \n The current arasan sdhci PHY configuration isn\u0027t compatible\n with the PHY on Intel\u0027s LGM(Lightning Mountain) SoC devices.\n \n Therefore, add a new compatible, to adapt the Intel\u0027s LGM\n eMMC PHY with arasan-sdhc controller to configure the PHY.\n \n Signed-off-by: Ramuthevar Vadivel Muruganx \u003cvadivel.muruganx.ramuthevar@linux.intel.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit 71e3fef67e24e86e4ea9ab5a0fab8ebbea0b9676\nAuthor: Ramuthevar Vadivel Murugan \u003cvadivel.muruganx.ramuthevar@linux.intel.com\u003e\nDate: Mon Aug 26 15:27:59 2019 +0800\n\n dt-bindings: mmc: sdhci-of-arasan: Add new compatible for Intel LGM eMMC\n \n Add a new compatible to use the sdhc-arasan host controller driver\n with the eMMC PHY on Intel\u0027s Lightning Mountain SoC.\n \n Signed-off-by: Ramuthevar Vadivel Murugan \u003cvadivel.muruganx.ramuthevar@linux.intel.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit 7b7de08a14b59d45cc4720bfb21732063ccfc8f6\nAuthor: Wei Yongjun \u003cweiyongjun1@huawei.com\u003e\nDate: Mon Aug 26 12:00:13 2019 +0000\n\n mmc: sdhci-of-aspeed: Fix return value check in aspeed_sdc_probe()\n \n In case of error, the function of_platform_device_create() returns\n NULL pointer not ERR_PTR(). The IS_ERR() test in the return value\n check should be replaced with NULL test.\n \n Signed-off-by: Wei Yongjun \u003cweiyongjun1@huawei.com\u003e\n Reviewed-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit 96fc9a50b212d71023573868c13d7ae5bf1e601a\nAuthor: Andrew Jeffery \u003candrew@aj.id.au\u003e\nDate: Mon Aug 26 11:46:19 2019 +0930\n\n dt-bindings: mmc: sdhci-of-aspeed: Update example ranges property\n \n The example node in the binding uses the AST2500 compatible string for\n the SD controller with a 64kiB ranges property, but the SD controller is\n allocated 128kiB of MMIO space according to the AST2500 datasheet. Fix\n the example to correctly reflect the hardware in the AST2500, however it\n should be noted that the MMIO region is reduced to 64kiB in the AST2600\n where a second SD controller block has been introduced into the address\n space.\n \n Also add the IBM copyright header that I left out of the initial patch.\n \n Suggested-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n Signed-off-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Acked-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit 9c1135a5e2a511adfc0f890330f15882a3a27878\nAuthor: Stefan Wahren \u003cwahrenst@gmx.net\u003e\nDate: Sat Aug 24 08:31:04 2019 +0200\n\n mmc: bcm2835: Take SWIOTLB memory size limitation into account\n \n Make sure the sdhost driver doesn\u0027t use requests bigger than SWIOTLB\n can handle.\n \n Signed-off-by: Stefan Wahren \u003cwahrenst@gmx.net\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit 4eb7b5288e6cb38063726a83f594a4ec7629caef\nAuthor: Andrew Jeffery \u003candrew@aj.id.au\u003e\nDate: Wed Aug 7 10:06:29 2019 +0930\n\n mmc: sdhci-of-aspeed: Add support for the ASPEED SD controller\n \n Add a minimal driver for ASPEED\u0027s SD controller, which exposes two\n SDHCIs.\n \n The ASPEED design implements a common register set for the SDHCIs, and\n moves some of the standard configuration elements out to this common\n area (e.g. 8-bit mode, and card detect configuration which is not\n currently supported).\n \n The SD controller has a dedicated hardware interrupt that is shared\n between the slots. The common register set exposes information on which\n slot triggered the interrupt; early revisions of the patch introduced an\n irqchip for the register, but reality is it doesn\u0027t behave as an\n irqchip, and the result fits awkwardly into the irqchip APIs. Instead\n I\u0027ve taken the simple approach of using the IRQ as a shared IRQ with\n some minor performance impact for the second slot.\n \n Ryan was the original author of the patch - I\u0027ve taken his work and\n massaged it to drop the irqchip support and rework the devicetree\n integration. The driver has been smoke tested under qemu against a\n minimal SD controller model and lightly tested on an ast2500-evb.\n \n Signed-off-by: Ryan Chen \u003cryanchen.aspeed@gmail.com\u003e\n Signed-off-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Acked-by: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Reviewed-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit 77502eeb08850ea101b143d32fea05af2b062b06\nAuthor: Andrew Jeffery \u003candrew@aj.id.au\u003e\nDate: Wed Aug 7 10:06:28 2019 +0930\n\n dt-bindings: mmc: sdhci-of-aspeed: Document Aspeed SD controller\n \n The ASPEED SD/SDIO/MMC controller exposes two slots implementing the\n SDIO Host Specification v2.00, with 1 or 4 bit data buses, or an 8 bit\n data bus if only a single slot is enabled.\n \n Signed-off-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit c0226660e48b9326189675d61533b9be373635de\nAuthor: Anand Jain \u003canand.jain@oracle.com\u003e\nDate: Tue Aug 27 15:40:45 2019 +0800\n\n btrfs: use proper error values on allocation failure in clone_fs_devices\n \n Fix the fake ENOMEM return error code to the actual error in\n clone_fs_devices().\n \n Reviewed-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Anand Jain \u003canand.jain@oracle.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 6acaa6af1501d17d40bb9aa5d76d5bb0b4936ed9\nAuthor: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\nDate: Wed May 1 18:19:48 2019 -0400\n\n drm/amd/powerplay: Add interface to lock SMU HW I2C.\n \n v2:\n PPSMC_MSG_RequestI2CBus seems not to work and so to avoid conflict\n over I2C bus and engine disable thermal control access to\n force SMU stop using the I2C bus until the issue is reslolved.\n \n Expose and call vega20_is_smc_ram_running to skip locking when SMU\n FW is not yet loaded.\n \n v3:\n Remove the prevoius hack as the SMU found the bug.\n \n v5: Typo fix\n \n Signed-off-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6a3068065fa4d6f931a12573d4bda5d85261cee2\nAuthor: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\nDate: Tue Apr 30 13:58:56 2019 -0400\n\n drm/amd: Import smuio_11_0 headers for EEPROM access on Vega20\n \n v3: Merge CKSVII2C_IC regs into exsisting headers.\n \n Signed-off-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 64f55e629237e4752db18df4d6969a69e3f4835a\nAuthor: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\nDate: Thu May 30 11:01:26 2019 -0400\n\n drm/amdgpu: Add RAS EEPROM table.\n \n Add RAS EEPROM table manager to eanble RAS errors to be stored\n upon appearance and retrived on driver load.\n \n v2: Fix some prints.\n \n v3:\n Fix checksum calculation.\n Make table record and header structs packed to do correct byte value sum.\n Fix record crossing EEPROM page boundry.\n \n v4:\n Fix byte sum val calculation for record - look at sizeof(record).\n Fix some style comments.\n \n v5: Add description to EEPROM_TABLE_RECORD_SIZE and syntax fixes.\n \n Signed-off-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Reviewed-by: Luben Tuikov \u003cLuben.Tuikov@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 250af743c044c342916a36517187e6e42a281efe\nAuthor: Gang Ba \u003cgaba@amd.com\u003e\nDate: Fri Aug 23 16:01:11 2019 -0400\n\n Revert \"drm/amdgpu: free up the first paging queue v2\"\n \n This reverts commit 4f8bc72fbf10f2dc8bca74d5da08b3a981b2e5cd.\n \n It turned out that a single reserved queue wouldn\u0027t be\n sufficient for page fault handling.\n \n Signed-off-by: Gang Ba \u003cgaba@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit da26ded3b2fff646d28559004195abe353bce49b\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Mon Aug 26 16:57:07 2019 +0800\n\n drm/amdgpu/display: fix build error without CONFIG_DRM_AMD_DC_DSC_SUPPORT\n \n If CONFIG_DRM_AMD_DC_DSC_SUPPORT is not set, build fails:\n \n drivers/gpu/drm/amd/amdgpu/../display/dc/dcn20/dcn20_hwseq.c: In function dcn20_hw_sequencer_construct:\n drivers/gpu/drm/amd/amdgpu/../display/dc/dcn20/dcn20_hwseq.c:2099:28:\n error: dcn20_dsc_pg_control undeclared (first use in this function); did you mean dcn20_dpp_pg_control?\n dc-\u003ehwss.dsc_pg_control \u003d dcn20_dsc_pg_control;\n ^~~~~~~~~~~~~~~~~~~~\n dcn20_dpp_pg_control\n \n Use CONFIG_DRM_AMD_DC_DSC_SUPPORT to guard this.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Fixes: 8a31820b1218 (\"drm/amd/display: Make init_hw and init_pipes generic for seamless boot\")\n Reviewed-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5556b9fed171bbfb53b08b56aef65201fcc4b4b1\nAuthor: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nDate: Mon Aug 26 16:20:12 2019 +0300\n\n drm/amd/powerplay: Fix an off by one in navi10_get_smu_msg_index()\n \n The navi10_message_map[] array has SMU_MSG_MAX_COUNT elements so the \"\u003e\"\n has to be changed to \"\u003e\u003d\" to prevent reading one element beyond the end\n of the array.\n \n Signed-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ed8a5fb2a606b6dfc6e417da6e64efcbb899e4cc\nAuthor: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\nDate: Mon Aug 26 10:11:50 2019 -0400\n\n drm/amd/display: Fix error message\n \n Since reservation_object_wait_timeout_rcu is called with\n interruptable set to false it\u0027s wrong to say\n \u0027or interrupted\u0027 in the error message.\n \n Signed-off-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Reviewed-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 534991731cb5fa94b5519957646cf849ca10d17d\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Fri Aug 16 16:13:28 2019 +0800\n\n drm/amdgpu: add dummy read for some GCVM status registers\n \n The GRBM register interface is now capable of bursting 1 cycle per\n register wr-\u003ewr, wr-\u003erd much faster than previous muticycle per\n transaction done interface. This has caused a problem where status\n registers requiring HW to update have a 1 cycle delay, due to the\n register update having to go through GRBM.\n \n SW may operate on an incorrect value if they write a register and\n immediately check the corresponding status register.\n \n Registers requiring HW to clear or set fields may be delayed by 1 cycle.\n For example,\n \n 1. write VM_INVALIDATE_ENG0_REQ mask \u003d 5a\n 2. read VM_INVALIDATE_ENG0_ACK till the ack is same as the request mask \u003d 5a\n a. HW will reset VM_INVALIDATE_ENG0_ACK \u003d 0 until invalidation is complete\n 3. write VM_INVALIDATE_ENG0_REQ mask \u003d 5a\n 4. read VM_INVALIDATE_ENG0_ACK till the ack is same as the request mask \u003d 5a\n a. First read of VM_INVALIDATE_ENG0_ACK \u003d 5a instead of 0\n b. Second read of VM_INVALIDATE_ENG0_ACK \u003d 0 because\n the remote GRBM h/w register takes one extra cycle to be cleared\n c. In this case, SW will see a false ACK if they exit on first read\n \n Affected registers (only GC variant) | Recommended Dummy Read\n --------------------------------------+----------------------------\n VM_INVALIDATE_ENG*_ACK | VM_INVALIDATE_ENG*_REQ\n VM_L2_STATUS | VM_L2_STATUS\n VM_L2_PROTECTION_FAULT_STATUS | VM_L2_PROTECTION_FAULT_STATUS\n VM_L2_PROTECTION_FAULT_ADDR_HI/LO32 | VM_L2_PROTECTION_FAULT_ADDR_HI/LO32\n VM_L2_IH_LOG_BUSY | VM_L2_IH_LOG_BUSY\n MC_VM_L2_PERFCOUNTER_HI/LO | MC_VM_L2_PERFCOUNTER_HI/LO\n ATC_L2_PERFCOUNTER_HI/LO | ATC_L2_PERFCOUNTER_HI/LO\n ATC_L2_PERFCOUNTER2_HI/LO | ATC_L2_PERFCOUNTER2_HI/LO\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a454a2e56a0b0ebfdb788996ea48c7edb12efcb6\nAuthor: Anand Jain \u003canand.jain@oracle.com\u003e\nDate: Tue Aug 27 15:40:44 2019 +0800\n\n btrfs: proper error handling when invalid device is found in find_next_devid\n \n In a corrupted tree, if search for next devid finds the device with\n devid \u003d -1, then report the error -EUCLEAN back to the parent function\n to fail gracefully.\n \n The tree checker will not catch this in case the devids are created\n using the following script:\n \n umount /btrfs\n dev1\u003d/dev/sdb\n dev2\u003d/dev/sdc\n mkfs.btrfs -fq -dsingle -msingle $dev1\n mount $dev1 /btrfs\n \n _fail()\n {\n echo $1\n exit 1\n }\n \n while true; do\n btrfs dev add -f $dev2 /btrfs || _fail \"add failed\"\n btrfs dev del $dev1 /btrfs || _fail \"del failed\"\n dev_tmp\u003d$dev1\n dev1\u003d$dev2\n dev2\u003d$dev_tmp\n done\n \n With output:\n \n BTRFS critical (device sdb): corrupt leaf: root\u003d3 block\u003d313739198464 slot\u003d1 devid\u003d1 invalid devid: has\u003d507 expect\u003d[0, 506]\n BTRFS error (device sdb): block\u003d313739198464 write time tree block corruption detected\n BTRFS: error (device sdb) in btrfs_commit_transaction:2268: errno\u003d-5 IO failure (Error while writing out transaction)\n BTRFS warning (device sdb): Skipping commit of aborted transaction.\n BTRFS: error (device sdb) in cleanup_transaction:1827: errno\u003d-5 IO failure\n \n Reviewed-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Reviewed-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Anand Jain \u003canand.jain@oracle.com\u003e\n [ add script and messages ]\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 21b0c32bdaba7c2e365d9faca0536fc40dd056d4\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Aug 27 10:49:33 2019 +0100\n\n drm/i915/selftests: Markup impossible error pointers\n \n If we create a new live_context() we should have a mapping for each\n engine. Document that assumption with an assertion.\n \n Reported-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190827094933.13778-1-chris@chris-wilson.co.uk\n\ncommit 405efc5980f2590a9520dc66cfe295456b8c9818\nAuthor: Christopher M. Riedl \u003ccmr@informatik.wtf\u003e\nDate: Mon Aug 12 22:13:14 2019 -0500\n\n powerpc/spinlocks: Fix oops in __spin_yield() on bare metal\n \n Booting w/ppc64le_defconfig + CONFIG_PREEMPT on bare metal results in\n the oops below due to calling into __spin_yield() when not running in\n an SPLPAR, which means lppaca pointers are NULL.\n \n We fixed a similar case previously in commit a6201da34ff9 (\"powerpc:\n Fix oops due to bad access of lppaca on bare metal\"), by adding SPLPAR\n checks in lppaca_shared_proc(). However when PREEMPT is enabled we can\n call __spin_yield() directly from arch_spin_yield().\n \n To fix it add spin_yield() and rw_yield() which check that\n shared-processor LPAR is enabled before calling the SPLPAR-only\n implementation of each.\n \n BUG: Kernel NULL pointer dereference at 0x00000100\n Faulting instruction address: 0xc000000000097f88\n Oops: Kernel access of bad area, sig: 7 [#1]\n LE PAGE_SIZE\u003d64K MMU\u003dRadix MMU\u003dHash PREEMPT SMP NR_CPUS\u003d2048 NUMA PowerNV\n Modules linked in:\n CPU: 0 PID: 2 Comm: kthreadd Not tainted 5.2.0-rc6-00491-g249155c20f9b #28\n NIP: c000000000097f88 LR: c000000000c07a88 CTR: c00000000015ca10\n REGS: c0000000727079f0 TRAP: 0300 Not tainted (5.2.0-rc6-00491-g249155c20f9b)\n MSR: 9000000002009033 \u003cSF,HV,VEC,EE,ME,IR,DR,RI,LE\u003e CR: 84000424 XER: 20040000\n CFAR: c000000000c07a84 DAR: 0000000000000100 DSISR: 00080000 IRQMASK: 1\n GPR00: c000000000c07a88 c000000072707c80 c000000001546300 c00000007be38a80\n GPR04: c0000000726f0c00 0000000000000002 c00000007279c980 0000000000000100\n GPR08: c000000001581b78 0000000080000001 0000000000000008 c00000007279c9b0\n GPR12: 0000000000000000 c000000001730000 c000000000142558 0000000000000000\n GPR16: 0000000000000000 0000000000000000 0000000000000000 0000000000000000\n GPR20: 0000000000000000 0000000000000000 0000000000000000 0000000000000000\n GPR24: c00000007be38a80 c000000000c002f4 0000000000000000 0000000000000000\n GPR28: c000000072221a00 c0000000726c2600 c00000007be38a80 c00000007be38a80\n NIP [c000000000097f88] __spin_yield+0x48/0xa0\n LR [c000000000c07a88] __raw_spin_lock+0xb8/0xc0\n Call Trace:\n [c000000072707c80] [c000000072221a00] 0xc000000072221a00 (unreliable)\n [c000000072707cb0] [c000000000bffb0c] __schedule+0xbc/0x850\n [c000000072707d70] [c000000000c002f4] schedule+0x54/0x130\n [c000000072707da0] [c0000000001427dc] kthreadd+0x28c/0x2b0\n [c000000072707e20] [c00000000000c1cc] ret_from_kernel_thread+0x5c/0x70\n Instruction dump:\n 4d9e0020 552a043e 210a07ff 79080fe0 0b080000 3d020004 3908b878 794a1f24\n e8e80000 7ce7502a e8e70000 38e70100 \u003c7ca03c2c\u003e 70a70001 78a50020 4d820020\n ---[ end trace 474d6b2b8fc5cb7e ]---\n \n Fixes: 499dcd41378e (\"powerpc/64s: Allocate LPPACAs individually\")\n Signed-off-by: Christopher M. Riedl \u003ccmr@informatik.wtf\u003e\n [mpe: Reword change log a bit]\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190813031314.1828-4-cmr@informatik.wtf\n\ncommit dd753d961c4844a39f947be115b3d81e10376ee5\nAuthor: Hsin-Yi Wang \u003chsinyi@chromium.org\u003e\nDate: Tue Aug 27 18:33:53 2019 +0800\n\n fdt: Update CRC check for rng-seed\n \n Commit 428826f5358c (\"fdt: add support for rng-seed\") moves of_fdt_crc32\n from early_init_dt_verify() to early_init_dt_scan() since\n early_init_dt_scan_chosen() may modify fdt to erase rng-seed.\n \n However, arm and some other arch won\u0027t call early_init_dt_scan(), they\n call early_init_dt_verify() then early_init_dt_scan_nodes().\n \n Restore of_fdt_crc32 to early_init_dt_verify() then update it in\n early_init_dt_scan_chosen() if fdt if updated.\n \n Fixes: 428826f5358c (\"fdt: add support for rng-seed\")\n Reported-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Hsin-Yi Wang \u003chsinyi@chromium.org\u003e\n Tested-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit f6a7053ddcf1c05c443715d627507f0ab9a0b491\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Fri Aug 9 15:26:49 2019 +0200\n\n gpio: zynq: Pass irqchip when adding gpiochip\n \n We need to convert all old gpio irqchips to pass the irqchip\n setup along when adding the gpio_chip. For more info see\n drivers/gpio/TODO.\n \n For chained irqchips this is a pretty straight-forward\n conversion.\n \n Cc: Michal Simek \u003cmichal.simek@xilinx.com\u003e\n Cc: Shubhrajyoti Datta \u003cshubhrajyoti.datta@xilinx.com\u003e\n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190809132649.25176-1-linus.walleij@linaro.org\n\ncommit 455b1d42e82c6027b9763f0055b54e45ff6cd7fd\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Tue Aug 27 11:32:05 2019 +0200\n\n ASoC: sunxi: Revert initial A83t support\n \n This reverts commit 3e9acd7ac6933c (ASoC: sun4i-i2s: Remove\n duplicated quirks structure\").\n \n It turns out that while one I2S controller is described in the A83t\n datasheet, the driver supports another, undocumented, one that has been\n inherited from the older SoCs, while the documented one uses the new\n design.\n \n Fixes: 3e9acd7ac693 (\"ASoC: sun4i-i2s: Remove duplicated quirks structure\")\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190827093206.17919-1-mripard@kernel.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9573820eb1951e0cb0f329886abcb4153f2ea798\nAuthor: Shengjiu Wang \u003cshengjiu.wang@nxp.com\u003e\nDate: Tue Aug 27 11:55:15 2019 -0400\n\n ASoC: imx-audmix: register the card on a proper dev\n \n This platform device is registered from \"fsl_audmix\", which is\n its parent device. If use pdev-\u003edev.parent for the priv-\u003ecard.dev,\n the value set by dev_set_drvdata in parent device will be covered\n by the value in child device.\n \n Fixes: b86ef5367761 (\"ASoC: fsl: Add Audio Mixer machine driver\")\n Signed-off-by: Viorel Suman \u003cviorel.suman@nxp.com\u003e\n Signed-off-by: Shengjiu Wang \u003cshengjiu.wang@nxp.com\u003e\n Reviewed-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Link: https://lore.kernel.org/r/1566921315-23402-1-git-send-email-shengjiu.wang@nxp.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit b9a4805a4ac1200326ee0c78df0ae3faceabad49\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 23 20:59:39 2019 +0800\n\n ASoC: SOF: imx8: Make some functions static\n \n Fix sparse warnings:\n \n sound/soc/sof/imx/imx8.c:104:6: warning: symbol \u0027imx8_dsp_handle_reply\u0027 was not declared. Should it be static?\n sound/soc/sof/imx/imx8.c:115:6: warning: symbol \u0027imx8_dsp_handle_request\u0027 was not declared. Should it be static?\n sound/soc/sof/imx/imx8.c:336:5: warning: symbol \u0027imx8_get_bar_index\u0027 was not declared. Should it be static?\n sound/soc/sof/imx/imx8.c:341:6: warning: symbol \u0027imx8_ipc_msg_data\u0027 was not declared. Should it be static?\n sound/soc/sof/imx/imx8.c:348:5: warning: symbol \u0027imx8_ipc_pcm_params\u0027 was not declared. Should it be static?\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Reviewed-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190823125939.30012-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 00f3e689518b8f8d36e333d3b447db18ef22abb3\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Tue Aug 27 07:23:34 2019 +0000\n\n MIPS: Octeon: remove duplicated include from dma-octeon.c\n \n Remove duplicated include.\n \n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: Jiaxun Yang \u003cjiaxun.yang@flygoat.com\u003e\n Cc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n Cc: Mike Rapoport \u003crppt@linux.ibm.com\u003e\n Cc: Michal Hocko \u003cmhocko@suse.com\u003e\n Cc: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Cc: \u003clinux-mips@vger.kernel.org\u003e\n Cc: \u003ckernel-janitors@vger.kernel.org\u003e\n\ncommit ebfdf5cd806b3bbf1ff79e69bce6a28df8bbe39d\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 26 14:07:50 2019 +0100\n\n drm/i915: Use NOEVICT for first pass on attemping to pin a GGTT mmap\n \n The intention is that we first try to pin the current vma into the\n mappable aperture only if it is already in use or it fits in the free\n space and will not cause contention. The first attempt was meant to be\n using PIN_NOEVICT to reuse the current vma if possible, following up\n with different eviction strategies.\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111485\n Fixes: 6846895fde05 (\"drm/i915: Replace PIN_NONFAULT with calls to PIN_NOEVICT\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190826130750.17272-1-chris@chris-wilson.co.uk\n\ncommit 1d5b7773314994d23f2fc4acfa4278929b03ffd5\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 26 08:21:22 2019 +0100\n\n drm/i915/selftests: Add the usual batch vma managements to st_workarounds\n \n To properly handle asynchronous migration of batch objects, we need to\n couple the fences on the incoming batch into the request and should not\n assume that they always start idle.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190826072149.9447-1-chris@chris-wilson.co.uk\n\ncommit 6030f42d20cedc04df019891851320f3e257038b\nAuthor: Axel Lin \u003caxel.lin@ingics.com\u003e\nDate: Fri Aug 16 15:08:25 2019 +0800\n\n i3c: master: Use dev_to_i3cmaster()\n \n Since we have dev_to_i3cmaster() available, use it.\n \n Signed-off-by: Axel Lin \u003caxel.lin@ingics.com\u003e\n Signed-off-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n\ncommit 058bf149b92a735b2b17a3797938fa455f9ed189\nAuthor: Yilun Lin \u003cyllin@chromium.org\u003e\nDate: Wed Aug 14 16:17:57 2019 +0800\n\n platform/chrome: cros_ec_rpmsg: Add host command AP sleep state support\n \n Add EC host command to inform EC of AP suspend/resume status.\n \n Signed-off-by: Yilun Lin \u003cyllin@chromium.org\u003e\n Tested-by: Pi-Hsun Shih \u003cpihsun@chromium.org\u003e\n Signed-off-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n\ncommit ae2bdbb45d38ee2ec67a958bf221138b58817c82\nAuthor: Henry Tieman \u003chenry.w.tieman@intel.com\u003e\nDate: Fri Aug 2 01:25:24 2019 -0700\n\n ice: fix adminq calls during remove\n \n The order of operations was incorrect in ice_remove(). The code would\n try to use adminq operations after the adminq was disabled. This caused\n all adminq calls to fail and possibly timeout waiting.\n \n Signed-off-by: Henry Tieman \u003chenry.w.tieman@intel.com\u003e\n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 152b978a1f904f252ae3df548bc2e183e7368c6d\nAuthor: Anirudh Venkataramanan \u003canirudh.venkataramanan@intel.com\u003e\nDate: Fri Aug 2 01:25:23 2019 -0700\n\n ice: Rework ice_ena_msix_range\n \n The current implementation of ice_ena_msix_range is difficult to read\n and has subtle issues. This patch reworks the said function for\n clarity and correctness.\n \n More specifically,\n \n 1. Add more checks to bail out of \u0027needed\u0027 is greater than \u0027v_left\u0027.\n \n 2. Simplify fallback logic\n \n 3. Do not set pf-\u003enum_avail_sw_msix in ice_ena_msix_range as it\n gets overwritten by ice_init_interrupt_scheme.\n \n Signed-off-by: Anirudh Venkataramanan \u003canirudh.venkataramanan@intel.com\u003e\n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit cb6a8dc07827086a7ce4151c62367d959c6ad5f1\nAuthor: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\nDate: Fri Aug 2 01:25:22 2019 -0700\n\n ice: Fix VF configuration issues due to reset\n \n This patch fixes a critical reset issue that resulting to the server\n reboot when an Admin changes VF configuration on the host, for example\n changing VF to Trusted/non_Trusted mode, the PF driver send reset\n notification to AVF driver while also continue with reset flow. However,\n AVF driver schedule another reset due to notification, which causes two\n concurrent reset going on, and trigger lock up in the FW, with AQ call to\n delete VSI.\n \n Signed-off-by: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 78b5713ac12417d796bbdcabf67dd94584f6102b\nAuthor: Anirudh Venkataramanan \u003canirudh.venkataramanan@intel.com\u003e\nDate: Fri Aug 2 01:25:21 2019 -0700\n\n ice: Alloc queue management bitmaps and arrays dynamically\n \n The total number of queues available on the device is divided between\n multiple physical functions (PF) in the firmware and provided to the\n driver when it gets function capabilities from the firmware. Thus\n each PF knows how many Tx/Rx queues it has. These queues are then\n doled out to different VSIs (for LAN traffic, SR-IOV VF traffic, etc.)\n \n To track usage of these queues at the PF level, the driver uses two\n bitmaps avail_txqs and avail_rxqs. At the VSI level (i.e. struct ice_vsi\n instances) the driver uses two arrays txq_map and rxq_map, to track\n ownership of VSIs\u0027 queues in avail_txqs and avail_rxqs respectively.\n \n The aforementioned bitmaps and arrays should be allocated dynamically,\n because the number of queues supported by a PF is only available once\n function capabilities have been queried. The current static allocation\n consumes way more memory than required.\n \n This patch removes the DECLARE_BITMAP for avail_txqs and avail_rxqs\n and instead uses bitmap_zalloc to allocate the bitmaps during init.\n Similarly txq_map and rxq_map are now allocated in ice_vsi_alloc_arrays.\n As a result ICE_MAX_TXQS and ICE_MAX_RXQS defines are no longer needed.\n Also as txq_map and rxq_map are now allocated and freed, some code\n reordering was required in ice_vsi_rebuild for correct functioning.\n \n Signed-off-by: Anirudh Venkataramanan \u003canirudh.venkataramanan@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 77ca27c417050376b703cec52810aaa10e17d9ef\nAuthor: Paul Greenwalt \u003cpaul.greenwalt@intel.com\u003e\nDate: Fri Aug 2 01:25:20 2019 -0700\n\n ice: add support for virtchnl_queue_select.[tx|rx]_queues bitmap\n \n The VF driver can call VIRTCHNL_OP_[ENABLE|DISABLE]_QUEUES separately\n for each queue. Add support for virtchnl_queue_select.[tx|rx]_queues\n bitmap which is used to indicate which queues to enable and disable.\n \n Add tracing of VF Tx/Rx per queue enable state to avoid enabling enabled\n queues and disabling disabled queues. Add total queues enabled count and\n clear ICE_VF_STATE_QS_ENA when count is zero.\n \n Signed-off-by: Paul Greenwalt \u003cpaul.greenwalt@intel.com\u003e\n Signed-off-by: Peng Huang \u003cpeng.huang@intel.com\u003e\n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit d02f734cb713aee558f61883ea694c056b01beb0\nAuthor: Maciej Fijalkowski \u003cmaciej.fijalkowski@intel.com\u003e\nDate: Fri Aug 2 01:25:19 2019 -0700\n\n ice: add support for enabling/disabling single queues\n \n Refactor the queue handling functions that are going through queue\n arrays in a way that the logic done for a single queue is pulled out and\n it will be called for each ring when traversing ring array. This implies\n that when disabling Tx rings we won\u0027t fill up q_ids, q_teids and\n q_handles arrays. Drop also \u0027offset\u0027 parameter; the value from vsi\u0027s\n txq_map is stored in ring-\u003ereg_idx and that drops the need for mentioned\n parameter. Introduce the ice_vsi_cfg_txq, ice_vsi_stop_tx_ring and\n ice_vsi_ctrl_rx_ring that are the functions with pulled out logic.\n \n There\u0027s several Tx queue meta data (q_id, q_handle, q_teid and other)\n that need to be set up during Tx queue disablement, so let\u0027s as well add\n a helper structure that wraps it up and a function that will be filling\n it up.\n \n Signed-off-by: Maciej Fijalkowski \u003cmaciej.fijalkowski@intel.com\u003e\n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit a1199d679af4103acdc0fdc17ae9c1274733d673\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Aug 2 16:52:17 2019 +0100\n\n ice: fix potential infinite loop\n \n The loop counter of a for-loop is a u8 however this is being compared\n to an int upper bound and this can lead to an infinite loop if the\n upper bound is greater than 255 since the loop counter will wrap back\n to zero. Fix this potential issue by making the loop counter an int.\n \n Addresses-Coverity: (\"Infinite loop\")\n Fixes: c7aeb4d1b9bf (\"ice: Disable VFs until reset is completed\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 35b4f4372f91d79632e8881b8c4a8f7802ff12f4\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Thu Jul 25 02:54:01 2019 -0700\n\n ice: fix ice_is_tc_ena\n \n ice_is_tc_ena is used to check whether a given traffic class is\n enabled. Because there are only 8 traffic classes, the function took\n a u8 bitmap. This causes problems because it is cast to an unsigned\n long causing a static analysis warning regarding Out-of-bounds read.\n \n Fix this by simply updating ice_is_tc_ena to take an unsigned long.\n Passing a u8 to this function should implicitly convert the value.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 9c7dd7566d189426dd013edc3ddd00c1d981a8a0\nAuthor: Michal Swiatkowski \u003cmichal.swiatkowski@intel.com\u003e\nDate: Thu Jul 25 02:53:59 2019 -0700\n\n ice: add validation in OP_CONFIG_VSI_QUEUES VF message\n \n Check num_queue_pairs to avoid access to unallocated field of\n vsi-\u003etx_rings/vsi-\u003erx_rings. Without this validation we can set\n vsi-\u003ealloc_txq/vsi-\u003ealloc_rxq to value smaller than ICE_MAX_BASE_QS_PER_VF\n and send this command with num_queue_pairs greater than\n vsi-\u003ealloc_txq/vsi-\u003ealloc_rxq. This lead to access to unallocated memory.\n \n In VF vsi alloc_txq and alloc_rxq should be the same. Get minimum\n because looks more readable.\n \n Also add validation for ring_len param. It should be greater than 32 and\n be multiple of 32. Incorrect value leads to hang traffic on PF.\n \n Signed-off-by: Michal Swiatkowski \u003cmichal.swiatkowski@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit e63a1dbdc7df4b6b48af920a86630c20a44340cf\nAuthor: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\nDate: Thu Jul 25 02:53:58 2019 -0700\n\n ice: Don\u0027t clog kernel debug log with VF MDD events errors\n \n In case of MDD events on VF, don\u0027t clog kernel log with unlimited VF MDD\n events message \"VF 0 has had 1018 MDD events since last boot\" - limit\n events log message to 30, based on the observation in some experimentation\n with sending malicious packet once, and number of events reported before\n device stopped observing MDD events.\n \n Also removed defunct macro \"ICE_DFLT_NUM_MDD_EVENTS_ALLOWED\" for tracking\n number of MDD events allowed before disabling the interface...\n \n Signed-off-by: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 0c39e403808cedd86adc0f505189fc6ea513bfe2\nAuthor: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\nDate: Mon Aug 26 23:16:44 2019 -0700\n\n MAINTAINERS: hwspinlock: update git tree location\n \n Update the maintainers file to reflect the move of the hwspinlock tree\n to kernel.org\n \n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 4425e0531c433e3a9414d09c48b82a0f372978b2\nAuthor: Krzysztof Kazimierczak \u003ckrzysztof.kazimierczak@intel.com\u003e\nDate: Thu Jul 25 02:53:57 2019 -0700\n\n ice: Introduce a local variable for a VSI in the rebuild path\n \n When a VSI is accessed inside the ice_for_each_vsi macro in the rebuild\n path (ice_vsi_rebuild_all() and ice_vsi_replay_all()), it is referred to\n as pf-\u003evsi[i]. Introduce local variables to improve readability.\n \n Signed-off-by: Krzysztof Kazimierczak \u003ckrzysztof.kazimierczak@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 150997fac770c37b12459ec52fdf67a5dc1366f5\nAuthor: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\nDate: Mon Aug 26 23:16:44 2019 -0700\n\n MAINTAINERS: remoteproc: update git tree location\n \n Update the maintainers file to reflect the move of the remoteproc tree\n to kernel.org\n \n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 13ef71f074837b53a25822e0a6912c2aefe55019\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Fri Feb 22 22:20:16 2019 -0600\n\n MAINTAINERS: rpmsg: fix git tree location\n \n The current information isn\u0027t up-to-date, use Bjorn Andersson\u0027s\n kernel.org repo instead.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n [bjorn: Replaced github URL]\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit dc67039b3d1159f8feba34e6fdb798603505d5d6\nAuthor: Jesse Brandeburg \u003cjesse.brandeburg@intel.com\u003e\nDate: Thu Jul 25 02:53:56 2019 -0700\n\n ice: shorten local and add debug prints\n \n Add some verbose debugging for dyndbg to help us when\n we are having issues with link and/or PHY.\n \n While there, shorten some strings used by locals that\n were causing long line wrapping.\n \n Signed-off-by: Jesse Brandeburg \u003cjesse.brandeburg@intel.com\u003e\n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 9ff166def8c1f5759555c2c94ddd0fef11a18c2b\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Fri Feb 22 22:20:17 2019 -0600\n\n rpmsg: core: fix comments\n \n Minor typos, grammar and copy/paste issues. Fix for consistency. No\n functional or semantic change.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 5af67635c36ed92ef172c7bbf4d711364bc3bdf7\nAuthor: Matt Pelland \u003cmpelland@starry.com\u003e\nDate: Thu Aug 1 15:50:59 2019 -0400\n\n phy: marvell: phy-mvebu-cp110-comphy: rename instances of DLT\n \n The documentation for Marvell\u0027s cp110 phy refers to these\n registers/register regions as DTL control, DTL frequency loop enable,\n etc. This patch aligns the relevant code for these accordingly.\n \n Signed-off-by: Matt Pelland \u003cmpelland@starry.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit f2a857aa2ad7335a54bd7b306ce02488eb269d58\nAuthor: Matt Pelland \u003cmpelland@starry.com\u003e\nDate: Thu Aug 1 15:50:58 2019 -0400\n\n phy: marvell: phy-mvebu-cp110-comphy: implement RXAUI support\n \n Marvell\u0027s cp110 phy supports RXAUI on lanes 2, 3, 4, and 5 when\n connected to port zero. When used in this mode, lanes operate in pairs\n of two (2 and 3, 4 and 5).\n \n Signed-off-by: Matt Pelland \u003cmpelland@starry.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit 0c79cf1f486135929f4370b6d919be0facdc1c8e\nAuthor: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nDate: Wed Jul 31 14:21:21 2019 +0200\n\n dt-bindings: pci: add PHY properties to Armada 7K/8K controller bindings\n \n Armada CP110 PCIe controller can have from one to four PHYs for\n configuring SERDES lanes (PCIe x1, PCIe x2 or PCIe x4). Describe the\n phys and phy-names properties in the bindings.\n \n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit 06a09dc318e4ee91be630a99dc7cd6b92aa44d3d\nAuthor: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nDate: Wed Jul 31 14:21:20 2019 +0200\n\n dt-bindings: phy: Add Marvell COMPHY clocks\n \n Marvell CP110 COMPHY block is fed by 3 clocks. Describe each of them in the\n bindings.\n \n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit 4e19a76ec08e52dd3ad43dcd49bbb704a2ff420b\nAuthor: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nDate: Wed Jul 31 14:21:19 2019 +0200\n\n phy: mvebu-cp110-comphy: Update comment about powering off all lanes at boot\n \n Now that all COMPHY modes are supported by the driver, update the\n comment stating that mvebu_comphy_power_off() should be called for\n each lane. This is still wrong because for compatibility reasons, it\n might break users running an old firmware (the driver only uses SMC\n calls for SATA, USB and PCIe configuration, there is no code in Linux\n to fallback on in these cases.\n \n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit 652488760ea613af0c1207169aeafc9c91203c38\nAuthor: Grzegorz Jaszczyk \u003cjaz@semihalf.com\u003e\nDate: Wed Jul 31 14:21:18 2019 +0200\n\n phy: mvebu-cp110-comphy: Add PCIe support\n \n Add PCIe support by filling the COMPHY modes table.\n \n Also add a new macro to generate the right value for the firmware\n depending on the width (PCI x1, x2, x4, etc). The width will be passed\n by the core as the \"submode\" argument of the -\u003eset_mode() callback. If\n this argument is zero, default to x1 mode.\n \n Signed-off-by: Grzegorz Jaszczyk \u003cjaz@semihalf.com\u003e\n [miquel.raynal@bootlin.com: adapt the content to the mainline driver]\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit 1eb9157ab3ef64e845e10fe40d49638fb408119e\nAuthor: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nDate: Wed Jul 31 14:21:17 2019 +0200\n\n phy: mvebu-cp110-comphy: Cosmetic change in a helper\n \n Before adding more logic, simplify a bit the writing of the\n mvebu_comphy_get_mode() helper by using a pointer instead of\n referencing a configuration with the entire table name.\n \n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit ef0ac9f24b65bac389a21b2d37ab90200f285062\nAuthor: Grzegorz Jaszczyk \u003cjaz@semihalf.com\u003e\nDate: Wed Jul 31 14:21:16 2019 +0200\n\n phy: mvebu-cp110-comphy: Add SATA support\n \n Add the corresponding entries in the COMPHY modes table.\n \n SATA support does not need any additional care.\n \n Signed-off-by: Grzegorz Jaszczyk \u003cjaz@semihalf.com\u003e\n [miquel.raynal@bootlin.com: adapt the content to the mainline driver]\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit c527a636d6200b0583caa7dee2427b0de218fb2c\nAuthor: Grzegorz Jaszczyk \u003cjaz@semihalf.com\u003e\nDate: Wed Jul 31 14:21:15 2019 +0200\n\n phy: mvebu-cp110-comphy: Add USB3 host/device support\n \n Add USB3 host/device support by adding the right entries in the COMPHY\n modes table. A new macro is created to instantiate a \"generic\" mode\n ie. not an Ethernet one. This macro will be re-used when adding SATA\n support.\n \n Signed-off-by: Grzegorz Jaszczyk \u003cjaz@semihalf.com\u003e\n [miquel.raynal@bootlin.com: adapt the content to the mainline driver]\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit 96888aed3d09862c1d80b3ccb405b4bcf6d827c5\nAuthor: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nDate: Wed Jul 31 14:21:14 2019 +0200\n\n phy: mvebu-cp110-comphy: Allow non-Ethernet modes to be configured\n \n The COMPHY can configure the SERDES lanes in several non-Ethernet\n modes: SATA, USB3, PCIe. Drop the condition limiting the driver to\n Ethernet modes only before adding support for more.\n \n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit c2afb2fef595805607633b6bffd5600d836e4ead\nAuthor: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nDate: Wed Jul 31 14:21:13 2019 +0200\n\n phy: mvebu-cp110-comphy: Rename the macro handling only Ethernet modes\n \n Before adding support for other PHY modes (not Ethernet ones), let\u0027s\n rename the MVEBU_COMPHY_CONF macro to a more specific (and shorter)\n appellation.\n \n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit 461324f0382cf63637d7158da53a5419ba51be54\nAuthor: Grzegorz Jaszczyk \u003cjaz@semihalf.com\u003e\nDate: Wed Jul 31 14:21:12 2019 +0200\n\n phy: mvebu-cp110-comphy: Add RXAUI support\n \n Add support for RXAUI mode by adding an entry in the COMPHY modes list.\n \n There is no user for this mode yet so we can enforce an up-to-date\n firmware and return an error otherwise without breaking anywone.\n \n Signed-off-by: Grzegorz Jaszczyk \u003cjaz@semihalf.com\u003e\n [miquel.raynal@bootlin.com: adapt the content to the mainline driver]\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit ea678b4b637ebe9a7edd758f2480b194c8fbae31\nAuthor: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nDate: Wed Jul 31 14:21:11 2019 +0200\n\n phy: mvebu-cp110-comphy: List already supported Ethernet modes\n \n Currently, the driver supports setting lanes to 1000BASEX, 2500BASEX,\n 10GKR. Complete the COMPHY modes list by adding two (already\n supported) cases for lane 4.\n \n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit eb6a1fcb53e2036608b80269429719025f5f1370\nAuthor: Grzegorz Jaszczyk \u003cjaz@semihalf.com\u003e\nDate: Wed Jul 31 14:21:10 2019 +0200\n\n phy: mvebu-cp110-comphy: Add SMC call support\n \n Keep the exact same list of supported configurations but first try to\n use the firmware\u0027s implementation. If it fails, try the legacy method:\n Linux implementation.\n \n Signed-off-by: Grzegorz Jaszczyk \u003cjaz@semihalf.com\u003e\n [miquel.raynal@bootlin.com: adapt the content to the mainline driver]\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Tested-by: Maxime Chevallier \u003cmaxime.chevallier@bootlin.com\u003e\n Tested-by: Grzegorz Jaszczyk \u003cjaz@semihalf.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit d4eda9d847ac3c32012bd28ac15ed68802c14b63\nAuthor: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nDate: Wed Jul 31 14:21:09 2019 +0200\n\n phy: mvebu-cp110-comphy: Explicitly initialize the lane submode\n \n Explicitly set the lane submode (enum) to a known invalid value.\n \n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit 0629d57bbdbf73aed45f057741b19bdfdefe8f5b\nAuthor: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nDate: Wed Jul 31 14:21:08 2019 +0200\n\n phy: mvebu-cp110-comphy: Add clocks support\n \n There is no public clock tree that implies such dependencies between\n the MG/MG-core/AXI clocks and the COMPHY IP but accessing the COMPHY\n registers while one of the three clocks are disabled stalls the CPU.\n \n This happens if, for instance, the COMPHY driver probe is deferred\n (eg. the USB Vbus regulator driver is not yet visible). The MVPP2\n driver which also needs these clocks (among others) will\n prepare/enable the clocks, then be deferred, and disable/unprepare\n them. Next COMPHY lane to be configured would produce an infinite\n stall.\n \n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Tested-by: Grzegorz Jaszczyk \u003cjaz@semihalf.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit 64ea59577f68667141609c6573a392d54fc7edbd\nAuthor: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\nDate: Wed Aug 7 12:23:05 2019 -0700\n\n phy-rockchip-inno-hdmi: Fix RK3328_TERM_RESISTOR_CALIB_SPEED_7_0\u0027s third value\n \n After commit \"linux/bits.h: Add compile time sanity check of GENMASK\n inputs\" [1], arm64 defconfig builds started failing:\n \n In file included from ../include/linux/bits.h:22,\n from ../include/linux/bitops.h:5,\n from ../include/linux/kernel.h:12,\n from ../include/linux/clk.h:13,\n from ../drivers/phy/rockchip/phy-rockchip-inno-hdmi.c:9:\n ../drivers/phy/rockchip/phy-rockchip-inno-hdmi.c: In function \u0027inno_hdmi_phy_rk3328_power_on\u0027:\n ../include/linux/build_bug.h:16:45: error: negative width in bit-field \u0027\u003canonymous\u003e\u0027\n 16 | #define BUILD_BUG_ON_ZERO(e) (sizeof(struct { int:(-!!(e)); }))\n | ^\n ../include/linux/bits.h:24:18: note: in expansion of macro \u0027BUILD_BUG_ON_ZERO\u0027\n 24 | ((unsigned long)BUILD_BUG_ON_ZERO(__builtin_choose_expr( \\\n | ^~~~~~~~~~~~~~~~~\n ../include/linux/bits.h:39:3: note: in expansion of macro \u0027GENMASK_INPUT_CHECK\u0027\n 39 | (GENMASK_INPUT_CHECK(high, low) + __GENMASK(high, low))\n | ^~~~~~~~~~~~~~~~~~~\n ../drivers/phy/rockchip/phy-rockchip-inno-hdmi.c:24:42: note: in expansion of macro \u0027GENMASK\u0027\n 24 | #define UPDATE(x, h, l) (((x) \u003c\u003c (l)) \u0026 GENMASK((h), (l)))\n | ^~~~~~~\n ../drivers/phy/rockchip/phy-rockchip-inno-hdmi.c:201:50: note: in expansion of macro \u0027UPDATE\u0027\n 201 | #define RK3328_TERM_RESISTOR_CALIB_SPEED_7_0(x) UPDATE(x, 7, 9)\n | ^~~~~~\n ../drivers/phy/rockchip/phy-rockchip-inno-hdmi.c:1046:26: note: in expansion of macro \u0027RK3328_TERM_RESISTOR_CALIB_SPEED_7_0\u0027\n 1046 | inno_write(inno, 0xc6, RK3328_TERM_RESISTOR_CALIB_SPEED_7_0(v));\n | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n \n As pointed out by Robin and Guenter, inno_write\u0027s val argument is an\n 8-bit value so having a mask larger than that doesn\u0027t make sense. This\n also matches the rest of the *_7_0 macros in this driver.\n \n [1]: https://lore.kernel.org/lkml/20190801230358.4193-2-rikard.falkeborn@gmail.com/\n \n Reported-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Reported-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Reported-by: kernelci.org bot \u003cbot@kernelci.org\u003e\n Reported-by: Naresh Kamboju \u003cnaresh.kamboju@linaro.org\u003e\n Suggested-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Suggested-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n Reviewed-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Tested-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit f27db2e65e1134c4173ced2db4aebade47485dc5\nAuthor: Anirudh Venkataramanan \u003canirudh.venkataramanan@intel.com\u003e\nDate: Thu Jul 25 02:53:55 2019 -0700\n\n ice: Sanitize ice_ena_vsi and ice_dis_vsi\n \n 1. ndo_open and ndo_stop are implemented by ice_open and ice_stop\n respectively. When enabling/disabling VSIs, just call\n ice_open/ice_stop instead of ndo_open/ndo_stop.\n \n 2. Rework logic around rtnl_lock/rtnl_unlock\n \n 3. In ice_ena_vsi, remove an unnecessary stack variable and return\n 0 instead of err when __ICE_NEEDS_RESTART is not set.\n \n Signed-off-by: Anirudh Venkataramanan \u003canirudh.venkataramanan@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 2935824873890de1503c0798c20833bb72a6c7c6\nAuthor: Victor Raj \u003cvictor.raj@intel.com\u003e\nDate: Thu Jul 25 02:53:54 2019 -0700\n\n ice: added sibling head to parse nodes\n \n There was a bug in the previous code which never traverses all the\n children to get the first node of the requested layer. Add a sibling\n head pointer to point the first node of each layer per TC. This helps\n traverse easier and quicker and also removes the recursion.\n \n Signed-off-by: Victor Raj \u003cvictor.raj@intel.com\u003e\n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 9e7a5d1746222238712df19ce9833e574d4cac8e\nAuthor: Usha Ketineni \u003cusha.k.ketineni@intel.com\u003e\nDate: Thu Jul 25 02:53:53 2019 -0700\n\n ice: Fix ethtool port and PFC stats for 4x25G cards\n \n This patch fixes the issue where port and PFC statistics counters are\n incrementing at the wrong port with 4x25G cards.\n Read the GLPRT port registers using lport parameter instead of pf_id to\n update the statistics otherwise the pf_ids are flipped for ports 2 and 3\n when read from the HW register PF_FUNC_RID and this is expected as per\n hardware specification.\n \n Signed-off-by: Usha Ketineni \u003cusha.k.ketineni@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 9a703eb72059530941ad32e2f99eccb70071f3f4\nAuthor: Suman Anna \u003cs-anna@ti.com\u003e\nDate: Fri Aug 9 11:27:10 2019 -0500\n\n samples/rpmsg: Introduce a module parameter for message count\n \n The current rpmsg_client_sample uses a fixed number of messages to\n be sent to each instance. This is currently set at 100. Introduce\n an optional module parameter \u0027count\u0027 so that the number of messages\n to be exchanged can be made flexible.\n \n Signed-off-by: Suman Anna \u003cs-anna@ti.com\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 2519fbb39711e5e6696685f29fe049af93c5987c\nAuthor: Suman Anna \u003cs-anna@ti.com\u003e\nDate: Fri Aug 9 11:27:09 2019 -0500\n\n samples/rpmsg: Replace print_hex_dump() with print_hex_dump_debug()\n \n Replace the raw print_hex_dump() call in the rpmsg_sample_cb() function\n with the equivalent print_hex_dump_debug() better suited for dynamic\n debug. This switch allows flexibility of controlling this trace through\n dynamic debug when enabled.\n \n Signed-off-by: Suman Anna \u003cs-anna@ti.com\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit d446609df212938c0d4660f8a328f694aa11e5ef\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:38 2019 -0700\n\n remoteproc: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Ohad Ben-Cohen \u003cohad@wizery.com\u003e\n Cc: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Cc: linux-remoteproc@vger.kernel.org\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit ff42df49e75f053a8a6b4c2533100cdcc23afe69\nAuthor: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\nDate: Tue Aug 27 11:35:40 2019 +1000\n\n KVM: PPC: Book3S HV: Don\u0027t lose pending doorbell request on migration on P9\n \n On POWER9, when userspace reads the value of the DPDES register on a\n vCPU, it is possible for 0 to be returned although there is a doorbell\n interrupt pending for the vCPU. This can lead to a doorbell interrupt\n being lost across migration. If the guest kernel uses doorbell\n interrupts for IPIs, then it could malfunction because of the lost\n interrupt.\n \n This happens because a newly-generated doorbell interrupt is signalled\n by setting vcpu-\u003earch.doorbell_request to 1; the DPDES value in\n vcpu-\u003earch.vcore-\u003edpdes is not updated, because it can only be updated\n when holding the vcpu mutex, in order to avoid races.\n \n To fix this, we OR in vcpu-\u003earch.doorbell_request when reading the\n DPDES value.\n \n Cc: stable@vger.kernel.org # v4.13+\n Fixes: 579006944e0d (\"KVM: PPC: Book3S HV: Virtualize doorbell facility on POWER9\")\n Signed-off-by: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\n Tested-by: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\n\ncommit d28eafc5a64045c78136162af9d4ba42f8230080\nAuthor: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\nDate: Tue Aug 27 11:31:37 2019 +1000\n\n KVM: PPC: Book3S HV: Check for MMU ready on piggybacked virtual cores\n \n When we are running multiple vcores on the same physical core, they\n could be from different VMs and so it is possible that one of the\n VMs could have its arch.mmu_ready flag cleared (for example by a\n concurrent HPT resize) when we go to run it on a physical core.\n We currently check the arch.mmu_ready flag for the primary vcore\n but not the flags for the other vcores that will be run alongside\n it. This adds that check, and also a check when we select the\n secondary vcores from the preempted vcores list.\n \n Cc: stable@vger.kernel.org # v4.14+\n Fixes: 38c53af85306 (\"KVM: PPC: Book3S HV: Fix exclusion between HPT resizing and other HPT updates\")\n Signed-off-by: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\n\ncommit 31391ff7ea1ef557a804475436501f33ff0ead95\nAuthor: Christopher M. Riedl \u003ccmr@informatik.wtf\u003e\nDate: Mon Aug 12 22:13:13 2019 -0500\n\n powerpc/spinlocks: Rename SPLPAR-only spinlocks\n \n The __rw_yield and __spin_yield locks only pertain to SPLPAR mode.\n Rename them to make this relationship obvious.\n \n Signed-off-by: Christopher M. Riedl \u003ccmr@informatik.wtf\u003e\n Reviewed-by: Andrew Donnellan \u003cajd@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190813031314.1828-3-cmr@informatik.wtf\n\ncommit d57b78353a99f5d813248954dd7d0527a01751ac\nAuthor: Christopher M. Riedl \u003ccmr@informatik.wtf\u003e\nDate: Mon Aug 12 22:13:12 2019 -0500\n\n powerpc/spinlocks: Refactor SHARED_PROCESSOR\n \n Determining if a processor is in shared processor mode is not a constant\n so don\u0027t hide it behind a #define.\n \n Signed-off-by: Christopher M. Riedl \u003ccmr@informatik.wtf\u003e\n Reviewed-by: Andrew Donnellan \u003cajd@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190813031314.1828-2-cmr@informatik.wtf\n\ncommit d7fb5b18a540efaf05da2b980fc11d50ba775677\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Tue Aug 20 14:34:14 2019 +0000\n\n powerpc/64: optimise LOAD_REG_IMMEDIATE_SYM()\n \n Optimise LOAD_REG_IMMEDIATE_SYM() using a temporary register to\n parallelise operations.\n \n It reduces the path from 5 to 3 instructions.\n \n Suggested-by: Segher Boessenkool \u003csegher@kernel.crashing.org\u003e\n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/bad41ed02531bb0382420cbab50a0d7153b71767.1566311636.git.christophe.leroy@c-s.fr\n\ncommit ba18025fb03306ccdf3557a1e7b8a5b39b474872\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Tue Aug 20 14:34:13 2019 +0000\n\n powerpc/32: replace LOAD_MSR_KERNEL() by LOAD_REG_IMMEDIATE()\n \n LOAD_MSR_KERNEL() and LOAD_REG_IMMEDIATE() are doing the same thing\n in the same way. Drop LOAD_MSR_KERNEL()\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/8f04a6df0bc8949517fd8236d50c15008ccf9231.1566311636.git.christophe.leroy@c-s.fr\n\ncommit c691b4b83b6a348f7b9d13c36916e73c2e1d85e4\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Tue Aug 20 14:34:12 2019 +0000\n\n powerpc: rewrite LOAD_REG_IMMEDIATE() as an intelligent macro\n \n Today LOAD_REG_IMMEDIATE() is a basic #define which loads all\n parts on a value into a register, including the parts that are NUL.\n \n This means always 2 instructions on PPC32 and always 5 instructions\n on PPC64. And those instructions cannot run in parallele as they are\n updating the same register.\n \n Ex: LOAD_REG_IMMEDIATE(r1,THREAD_SIZE) in head_64.S results in:\n \n 3c 20 00 00 lis r1,0\n 60 21 00 00 ori r1,r1,0\n 78 21 07 c6 rldicr r1,r1,32,31\n 64 21 00 00 oris r1,r1,0\n 60 21 40 00 ori r1,r1,16384\n \n Rewrite LOAD_REG_IMMEDIATE() with GAS macro in order to skip\n the parts that are NUL.\n \n Rename existing LOAD_REG_IMMEDIATE() as LOAD_REG_IMMEDIATE_SYM()\n and use that one for loading value of symbols which are not known\n at compile time.\n \n Now LOAD_REG_IMMEDIATE(r1,THREAD_SIZE) in head_64.S results in:\n \n 38 20 40 00 li r1,16384\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/d60ce8dd3a383c7adbfc322bf1d53d81724a6000.1566311636.git.christophe.leroy@c-s.fr\n\ncommit 163918fc5741d755cf9d477ebfcb761f09b82982\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Tue Aug 20 14:07:20 2019 +0000\n\n powerpc/mm: split out early ioremap path.\n \n ioremap does things differently depending on whether\n SLAB is available or not at different levels.\n \n Try to separate the early path from the beginning.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/3acd2dbe04b04f111475e7a59f2b6f2ab9b95ab6.1566309263.git.christophe.leroy@c-s.fr\n\ncommit 4a45b7460cf458012a6930f675e141256b81dcf4\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Tue Aug 20 14:07:19 2019 +0000\n\n powerpc/mm: refactor ioremap vm area setup.\n \n PPC32 and PPC64 are doing the same once SLAB is available.\n Create a do_ioremap() function that calls get_vm_area and\n do the mapping.\n \n For PPC64, we add the 4K PFN hack sanity check to __ioremap_caller()\n in order to avoid using __ioremap_at(). Other checks in __ioremap_at()\n are irrelevant for __ioremap_caller().\n \n On PPC64, VM area is allocated in the range [ioremap_bot ; IOREMAP_END]\n On PPC32, VM area is allocated in the range [VMALLOC_START ; VMALLOC_END]\n \n Lets define IOREMAP_START is ioremap_bot for PPC64, and alias\n IOREMAP_START/END to VMALLOC_START/END on PPC32\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/42e7e36ad32e0fdf76692426cc642799c9f689b8.1566309263.git.christophe.leroy@c-s.fr\n\ncommit 191e42063a7241e5c3a1d1f36896a20b147517e9\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Tue Aug 20 14:07:18 2019 +0000\n\n powerpc/mm: refactor ioremap_range() and use ioremap_page_range()\n \n book3s64\u0027s ioremap_range() is almost same as fallback ioremap_range(),\n except that it calls radix__ioremap_range() when radix is enabled.\n \n radix__ioremap_range() is also very similar to the other ones, expect\n that it calls ioremap_page_range when slab is available.\n \n PPC32 __ioremap_caller() have a loop doing the same thing as\n ioremap_range() so use it on PPC32 as well.\n \n Lets keep only one version of ioremap_range() which calls\n ioremap_page_range() on all platforms when slab is available.\n \n At the same time, drop the nid parameter which is not used.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/4b1dca7096b01823b101be7338983578641547f1.1566309263.git.christophe.leroy@c-s.fr\n\ncommit f381d5711f091facd8847a54a2377cc0d1519df2\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Tue Aug 20 14:07:17 2019 +0000\n\n powerpc/mm: Move ioremap functions out of pgtable_32/64.c\n \n Create ioremap_32.c and ioremap_64.c and move respective ioremap\n functions out of pgtable_32.c and pgtable_64.c\n \n In the meantime, fix a few comments and changes a printk() to\n pr_warn(). Also fix a few oversplitted lines.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/b5c8b02ccefd4ede64c61b53cf64fb5dacb35740.1566309263.git.christophe.leroy@c-s.fr\n\ncommit 7cd9b317b630683b0c8eb2dfcfb046003ad6b97b\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Tue Aug 20 14:07:16 2019 +0000\n\n powerpc/mm: make ioremap_bot common to all\n \n Drop multiple definitions of ioremap_bot and make one common to\n all subarches.\n \n Only CONFIG_PPC_BOOK3E_64 had a global static init value for\n ioremap_bot. Now ioremap_bot is set in early_init_mmu_global().\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/920eebfd9f36f14c79d1755847f5bf7c83703bdd.1566309262.git.christophe.leroy@c-s.fr\n\ncommit edfe1a5679263827ff94eb478c6ee0d65d467adf\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Tue Aug 20 14:07:15 2019 +0000\n\n powerpc/mm: move ioremap_prot() into ioremap.c\n \n Both ioremap_prot() are idenfical, move them into ioremap.c\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/0b3eb0e0f1490a99fd6c983e166fb8946233f151.1566309262.git.christophe.leroy@c-s.fr\n\ncommit 4634c375db7a082b2522621519a5fb6eba977584\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Tue Aug 20 14:07:14 2019 +0000\n\n powerpc/mm: move common 32/64 bits ioremap functions into ioremap.c\n \n ioremap(), ioremap_wc() and ioremap_coherent() are now identical on\n PPC32 and PPC64 as iowa_is_active() will always return false on\n PPC32. Move them into a new common location called ioremap.c\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/6223803ce024d6ab4dfaa919f44098aed5b4bc33.1566309262.git.christophe.leroy@c-s.fr\n\ncommit 14b4d97669b79d1ac83e64d6795098394e15ab1b\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Tue Aug 20 14:07:13 2019 +0000\n\n powerpc/mm: rework io-workaround invocation.\n \n ppc_md.ioremap() is only used for I/O workaround on CELL platform,\n so indirect function call can be avoided.\n \n This patch reworks the io-workaround and ioremap() functions to\n use the global \u0027io_workaround_inited\u0027 flag for the activation\n of io-workaround.\n \n When CONFIG_PPC_IO_WORKAROUNDS or CONFIG_PPC_INDIRECT_MMIO are not\n selected, the I/O workaround ioremap() voids and the global flag is\n not used.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/5fa3ef069fbd0f152512afaae19e7a60161454cf.1566309262.git.christophe.leroy@c-s.fr\n\ncommit 492643e81e58e7585cccb22c37814060e3f59268\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Tue Aug 20 14:07:12 2019 +0000\n\n powerpc/mm: drop function __ioremap()\n \n __ioremap() is not used anymore, drop it.\n \n Suggested-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/ccc439f481a0884e00a6be1bab44bab2a4477fea.1566309262.git.christophe.leroy@c-s.fr\n\ncommit 8aee077292a34eaca717445290f12d23f8bd7732\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Tue Aug 20 14:07:11 2019 +0000\n\n powerpc/mm: drop ppc_md.iounmap() and __iounmap()\n \n ppc_md.iounmap() is never set, drop it.\n \n Once ppc_md.iounmap() is gone, iounmap() remains the only user of\n __iounmap() and iounmap() does nothing else than calling __iounmap().\n So drop iounmap() and make __iounmap() the new iounmap().\n \n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/d73ba92bb7a387cc58cc34666d7f5158a45851b0.1566309262.git.christophe.leroy@c-s.fr\n\ncommit 6f57e6631d7c6ae34a5e811f7f49cb63647a9cab\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Tue Aug 20 14:07:10 2019 +0000\n\n powerpc/ps3: replace __ioremap() by ioremap_prot()\n \n __ioremap() is similar to ioremap_prot() except that ioremap_prot()\n does a few sanity changes in addition.\n \n The flags used by PS3 are not impacted by those changes so for\n PS3 both functions are equivalent.\n \n At the same time, drop parts of the comment that have been invalid\n since commit e58e87adc8bf (\"powerpc/mm: Update _PAGE_KERNEL_RO\")\n \n Suggested-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/36bff5d875ff562889c5e12dab63e5d7c5d1fbd8.1566309262.git.christophe.leroy@c-s.fr\n\ncommit f0f8d7ae3924ed93453e30123e4aaf6f888ca555\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 20 14:07:09 2019 +0000\n\n powerpc: remove the ppc44x ocm.c file\n \n The on chip memory allocator is entirely unused in the kernel tree.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Acked-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/7b1668941ad1041d08b19167030868de5840b153.1566309262.git.christophe.leroy@c-s.fr\n\ncommit b4645ffc49cfe34f67feda20c34bd7a859c78312\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Thu Aug 22 16:44:05 2019 +0000\n\n powerpc/64: don\u0027t select ARCH_HAS_SCALED_CPUTIME on book3E\n \n Book3E doesn\u0027t have SPRN_SPURR/SPRN_PURR.\n \n Activating ARCH_HAS_SCALED_CPUTIME is just wasting CPU time.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://github.com/linuxppc/issues/issues/171\n Link: https://lore.kernel.org/r/a8b567c569aa521a7cf1beb061d43d79070e850c.1566492229.git.christophe.leroy@c-s.fr\n\ncommit d8f0e0b073e1ec52a05f0c2a56318b47387d2f10\nAuthor: Christopher M. Riedl \u003ccmr@informatik.wtf\u003e\nDate: Thu May 23 21:46:48 2019 -0500\n\n powerpc/64s: support nospectre_v2 cmdline option\n \n Add support for disabling the kernel implemented spectre v2 mitigation\n (count cache flush on context switch) via the nospectre_v2 and\n mitigations\u003doff cmdline options.\n \n Suggested-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Signed-off-by: Christopher M. Riedl \u003ccmr@informatik.wtf\u003e\n Reviewed-by: Andrew Donnellan \u003cajd@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190524024647.381-1-cmr@informatik.wtf\n\ncommit 2ad7a27deaf6d78545d97ab80874584f6990360e\nAuthor: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\nDate: Mon Aug 26 16:21:21 2019 +1000\n\n KVM: PPC: Book3S: Enable XIVE native capability only if OPAL has required functions\n \n There are some POWER9 machines where the OPAL firmware does not support\n the OPAL_XIVE_GET_QUEUE_STATE and OPAL_XIVE_SET_QUEUE_STATE calls.\n The impact of this is that a guest using XIVE natively will not be able\n to be migrated successfully. On the source side, the get_attr operation\n on the KVM native device for the KVM_DEV_XIVE_GRP_EQ_CONFIG attribute\n will fail; on the destination side, the set_attr operation for the same\n attribute will fail.\n \n This adds tests for the existence of the OPAL get/set queue state\n functions, and if they are not supported, the XIVE-native KVM device\n is not created and the KVM_CAP_PPC_IRQ_XIVE capability returns false.\n Userspace can then either provide a software emulation of XIVE, or\n else tell the guest that it does not have a XIVE controller available\n to it.\n \n Cc: stable@vger.kernel.org # v5.2+\n Fixes: 3fab2d10588e (\"KVM: PPC: Book3S HV: XIVE: Activate XIVE exploitation mode\")\n Reviewed-by: David Gibson \u003cdavid@gibson.dropbear.id.au\u003e\n Reviewed-by: Cédric Le Goater \u003cclg@kaod.org\u003e\n Signed-off-by: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\n\ncommit 1dd70853f8131f1674c4496b1d24de7251cad42c\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Mon Aug 26 23:12:33 2019 +0530\n\n arm64: dts: sdm845: Add parent clock for rpmhcc\n \n RPM clock controller has parent as xo, so specify that in DT node for\n rpmhcc\n \n Reviewed-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 519e5869d50d1e41b39cef4d757973469393fac7\nAuthor: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nDate: Mon Aug 26 17:06:00 2019 -0700\n\n xfs: bmap scrub should only scrub records once\n \n The inode block mapping scrub function does more work for btree format\n extent maps than is absolutely necessary -- first it will walk the bmbt\n and check all the entries, and then it will load the incore tree and\n check every entry in that tree, possibly for a second time.\n \n Simplify the code and decrease check runtime by separating the two\n responsibilities. The bmbt walk will make sure the incore extent\n mappings are loaded, check the shape of the bmap btree (via xchk_btree)\n and check that every bmbt record has a corresponding incore extent map;\n and the incore extent map walk takes all the responsibility for checking\n the mapping records and cross referencing them with other AG metadata.\n \n This enables us to clean up some messy parameter handling and reduce\n redundant code. Rename a few functions to make the split of\n responsibilities clearer.\n \n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Reviewed-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n Reviewed-by: Brian Foster \u003cbfoster@redhat.com\u003e\n\ncommit 71912e08e06b7041ba3b40945328b84798dfe6fc\nAuthor: zhengbin \u003czhengbin13@huawei.com\u003e\nDate: Mon Aug 26 12:08:40 2019 -0700\n\n xfs: remove excess function parameter description in \u0027xfs_btree_sblock_v5hdr_verify\u0027\n \n Fixes gcc warning:\n \n fs/xfs/libxfs/xfs_btree.c:4475: warning: Excess function parameter \u0027max_recs\u0027 description in \u0027xfs_btree_sblock_v5hdr_verify\u0027\n fs/xfs/libxfs/xfs_btree.c:4475: warning: Excess function parameter \u0027pag_max_level\u0027 description in \u0027xfs_btree_sblock_v5hdr_verify\u0027\n \n Fixes: c5ab131ba0df (\"libxfs: refactor short btree block verification\")\n Signed-off-by: zhengbin \u003czhengbin13@huawei.com\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit f8f9ee479439c1be9e33c4404912a2a112c46200\nAuthor: Dave Chinner \u003cdchinner@redhat.com\u003e\nDate: Mon Aug 26 12:08:39 2019 -0700\n\n xfs: add kmem_alloc_io()\n \n Memory we use to submit for IO needs strict alignment to the\n underlying driver contraints. Worst case, this is 512 bytes. Given\n that all allocations for IO are always a power of 2 multiple of 512\n bytes, the kernel heap provides natural alignment for objects of\n these sizes and that suffices.\n \n Until, of course, memory debugging of some kind is turned on (e.g.\n red zones, poisoning, KASAN) and then the alignment of the heap\n objects is thrown out the window. Then we get weird IO errors and\n data corruption problems because drivers don\u0027t validate alignment\n and do the wrong thing when passed unaligned memory buffers in bios.\n \n TO fix this, introduce kmem_alloc_io(), which will guaranteeat least\n 512 byte alignment of buffers for IO, even if memory debugging\n options are turned on. It is assumed that the minimum allocation\n size will be 512 bytes, and that sizes will be power of 2 mulitples\n of 512 bytes.\n \n Use this everywhere we allocate buffers for IO.\n \n This no longer fails with log recovery errors when KASAN is enabled\n due to the brd driver not handling unaligned memory buffers:\n \n # mkfs.xfs -f /dev/ram0 ; mount /dev/ram0 /mnt/test\n \n Signed-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit d916275aa4ddfea69b5480e292f398d58bdb40fc\nAuthor: Dave Chinner \u003cdchinner@redhat.com\u003e\nDate: Mon Aug 26 12:08:38 2019 -0700\n\n xfs: get allocation alignment from the buftarg\n \n Needed to feed into the allocation routine to guarantee the memory\n buffers we add to bios are correctly aligned to the underlying\n device.\n \n Signed-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit 0ad95687c3adb91e762d0f6ea50a6b1137300e19\nAuthor: Dave Chinner \u003cdchinner@redhat.com\u003e\nDate: Mon Aug 26 12:08:10 2019 -0700\n\n xfs: add kmem allocation trace points\n \n When trying to correlate XFS kernel allocations to memory reclaim\n behaviour, it is useful to know what allocations XFS is actually\n attempting. This information is not directly available from\n tracepoints in the generic memory allocation and reclaim\n tracepoints, so these new trace points provide a high level\n indication of what the XFS memory demand actually is.\n \n There is no per-filesystem context in this code, so we just trace\n the type of allocation, the size and the allocation constraints.\n The kmem code also doesn\u0027t include much of the common XFS headers,\n so there are a few definitions that need to be added to the trace\n headers and a couple of types that need to be made common to avoid\n needing to include the whole world in the kmem code.\n \n Signed-off-by: Dave Chinner \u003cdchinner@redhat.com\u003e\n Reviewed-by: Brian Foster \u003cbfoster@redhat.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit d00ee466a07eb9182ad3caf6140c7ebb527b4c64\nAuthor: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\nDate: Mon Aug 26 15:30:41 2019 -0700\n\n nfp: add AMDA0058 boards to firmware list\n \n Add MODULE_FIRMWARE entries for AMDA0058 boards.\n \n Signed-off-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Reviewed-by: Dirk van der Merwe \u003cdirk.vandermerwe@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 049b405029c00f3fd9e4ffa269bdd29b429c4672\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Mon Aug 26 16:02:56 2019 -0700\n\n MAINTAINERS: Update from paulmck@linux.ibm.com to paulmck@kernel.org\n \n Note that the paulmck@linux.ibm.com still works most of the time.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 24691069a348f82a95e0fa9697bb5656c6d8c48c\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Thu Aug 22 10:53:43 2019 +0900\n\n rcu: Don\u0027t include \u003clinux/ktime.h\u003e in rcutiny.h\n \n The kbuild reported a built failure due to a header loop when RCUTINY is\n enabled with my pending riscv-nommu port. Switch rcutiny.h to only\n include the minimal required header to get HZ instead.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 74a1e863eb73dcc9f069b671dfb40650f3832116\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Aug 26 19:31:06 2019 -0300\n\n perf evsel: Rename perf_missing_features::bpf_event to ::bpf\n \n No need for that _event suffix, do just like all the other meta events\n and do away with that.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Song Liu \u003csongliubraving@fb.com\u003e\n Link: https://lkml.kernel.org/n/tip-bvc83f380dva83wlg52yd10t@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 3f604b5f61dbff80725392c99827d6617f7bb180\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Aug 26 19:28:13 2019 -0300\n\n perf tool: Rename perf_tool::bpf_event to bpf\n \n No need for that _event suffix, do just like all the other meta event\n handlers and suppress that suffix.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Song Liu \u003csongliubraving@fb.com\u003e\n Link: https://lkml.kernel.org/n/tip-03spzxtqafbabbbmnm7y4xfx@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 6a1b359821eb8d929c4dd9f53178da84888d79ec\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Aug 26 19:20:35 2019 -0300\n\n perf tools: Rename perf_event::bpf_event to perf_event::bpf\n \n Just like all the other meta events, that extra _event suffix is just\n redundant, ditch it.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Song Liu \u003csongliubraving@fb.com\u003e\n Link: https://lkml.kernel.org/n/tip-505qwpaizq1k0t6pk13v1ibd@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit ebdba16e95f728e94dba07fe0f1221b0e8efdb9d\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Aug 26 19:15:18 2019 -0300\n\n perf tools: Rename perf_event::ksymbol_event to perf_event::ksymbol\n \n Just like all the other meta events, that extra _event suffix is just\n redundant, ditch it.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Song Liu \u003csongliubraving@fb.com\u003e\n Link: https://lkml.kernel.org/n/tip-0q8b2xnfs17q0g523oej75s0@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 69d81f09e1607b577346c0579bf938c1194bff3a\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Aug 26 19:02:31 2019 -0300\n\n libperf: Rename the PERF_RECORD_ structs to have a \"perf\" suffix\n \n Even more, to have a \"perf_record_\" prefix, so that they match the\n PERF_RECORD_ enum they map to.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-qbabmcz2a0pkzt72liyuz3p8@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit b1fcd190bb3fc1234dca60390d171a4cc75b21b2\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Aug 25 20:17:52 2019 +0200\n\n libperf: Add PERF_RECORD_SAMPLE \u0027struct sample_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_SAMPLE event definition to libperf\u0027s event.h header\n include.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190825181752.722-13-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit b1b510142283c02991f48b27d399852364f7d89b\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Aug 25 20:17:51 2019 +0200\n\n libperf: Add PERF_RECORD_BPF_EVENT \u0027struct bpf_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_BPF_EVENT event definition to libperf\u0027s event.h.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027\n types used events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190825181752.722-12-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit f15e3c25a1b40794a0ef2647360afe873fe34f54\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Aug 25 20:17:50 2019 +0200\n\n libperf: Add PERF_RECORD_KSYMBOL \u0027struct ksymbol_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_KSYMBOL event definition into libperf\u0027s event.h\n header include.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Perf added \u0027u*\u0027 types mainly to ease up printing __u64 values\n as stated in the linux/types.h comment:\n \n /*\n * We define u64 as uint64_t for every architecture\n * so that we can print it with \"%\"PRIx64 without getting warnings.\n *\n * typedef __u64 u64;\n * typedef __s64 s64;\n */\n \n Add and use new PRI_lu64 and PRI_lx64 macros for that. Use extra \u0027_\u0027 to\n ease up the reading and differentiate them from standard PRI*64 macros.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190825181752.722-11-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 003c66fec28fea52825b60cad98af8cf11074d76\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Aug 25 20:17:49 2019 +0200\n\n libperf: Add PERF_RECORD_THROTTLE \u0027struct throttle_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_THROTTLE event definition into libperf\u0027s event.h\n header include.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Perf added \u0027u*\u0027 types mainly to ease up printing __u64 values as stated\n in the linux/types.h comment:\n \n /*\n * We define u64 as uint64_t for every architecture\n * so that we can print it with \"%\"PRIx64 without getting warnings.\n *\n * typedef __u64 u64;\n * typedef __s64 s64;\n */\n \n Add and use new PRI_lu64 and PRI_lx64 macros for that. Use extra \u0027_\u0027 to\n ease up the reading and differentiate them from standard PRI*64 macros.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190825181752.722-10-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 213a6c1d20687d44acaa1cb4f77ce5bae4f1dd8f\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Aug 25 20:17:48 2019 +0200\n\n libperf: Add PERF_RECORD_READ \u0027struct read_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_READ event definition to libperf\u0027s event.h header\n include.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Perf added \u0027u*\u0027 types mainly to ease up printing __u64 values\n as stated in the linux/types.h comment:\n \n /*\n * We define u64 as uint64_t for every architecture\n * so that we can print it with \"%\"PRIx64 without getting warnings.\n *\n * typedef __u64 u64;\n * typedef __s64 s64;\n */\n \n Add and use new PRI_lu64 and PRI_lx64 macros for that. Use extra \u0027_\u0027 to\n ease up the reading and differentiate them from standard PRI*64 macros.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190825181752.722-9-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit a2e254d84172f7eb638261a83024d849f78c89e9\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Aug 25 20:17:47 2019 +0200\n\n libperf: Add PERF_RECORD_LOST_SAMPLES \u0027struct lost_samples_event\u0027 to perf/event.h\n \n Move the PERF_RECORD_LOST_SAMPLES event definition into libperf\u0027s\n event.h header include.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Perf added \u0027u*\u0027 types mainly to ease up printing __u64 values\n as stated in the linux/types.h comment:\n \n /*\n * We define u64 as uint64_t for every architecture\n * so that we can print it with \"%\"PRIx64 without getting warnings.\n *\n * typedef __u64 u64;\n * typedef __s64 s64;\n */\n \n Add and use new PRI_lu64 and PRI_lx64 macros for that. Use extra \u0027_\u0027 to\n ease up the reading and differentiate them from standard PRI*64 macros.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190825181752.722-8-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 5290ed6955ebc481d5cd62f7175e8514931058bc\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Aug 25 20:17:46 2019 +0200\n\n libperf: Add PERF_RECORD_LOST \u0027struct lost_event\u0027 to perf/event.h\n \n Move the lost_event event definition to libperf\u0027s event.h header\n include.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Perf added \u0027u*\u0027 types mainly to ease up printing __u64 values as stated\n in the linux/types.h comment:\n \n /*\n * We define u64 as uint64_t for every architecture\n * so that we can print it with \"%\"PRIx64 without getting warnings.\n *\n * typedef __u64 u64;\n * typedef __s64 s64;\n */\n \n Add and use new PRI_lu64 and PRI_lx64 macros for that. Use extra \u0027_\u0027 to\n ease up the reading and differentiate them from standard PRI*64 macros.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190825181752.722-7-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit bceb59b1f28d1ca812d816fd8f33c72b1a8378fb\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Aug 25 20:17:45 2019 +0200\n\n libperf: Add PERF_RECORD_FORK \u0027struct fork_event\u0027 to perf/event.h\n \n Move the fork_event event definition into libperf\u0027s event.h header\n include.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Perf added \u0027u*\u0027 types mainly to ease up printing __u64 values\n as stated in the linux/types.h comment:\n \n /*\n * We define u64 as uint64_t for every architecture\n * so that we can print it with \"%\"PRIx64 without getting warnings.\n *\n * typedef __u64 u64;\n * typedef __s64 s64;\n */\n \n Add and use new PRI_lu64 and PRI_lx64 macros for that. Using extra \u0027_\u0027\n to ease up the reading and differentiate them from standard PRI*64\n macros.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190825181752.722-6-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 19d1765a3ed9a8f78d93909120f6d39398809f75\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Aug 25 20:17:44 2019 +0200\n\n libperf: Add PERF_RECORD_NAMESPACES \u0027struct namespaces_event\u0027 to perf/event.h\n \n Move the namespaces_event event definition into libperf\u0027s event.h header\n include.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190825181752.722-5-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 002dda32a831b30d22a3620eb619a0d103e78e81\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Aug 25 20:17:43 2019 +0200\n\n libperf: Add PERF_RECORD_COMM \u0027struct comm_event\u0027 to perf/event.h\n \n Moving comm_event event definition into libperf\u0027s event.h\n header include.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027\n types used events to their generic \u0027__u*\u0027 versions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190825181752.722-4-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit b66ced19c9f64dbe707cf318fae9fca82b999564\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Aug 25 20:17:42 2019 +0200\n\n libperf: Add PERF_RECORD_MMAP2 \u0027struct mmap2_event\u0027 to perf/event.h\n \n Moving mmap2_event event definition into libperf\u0027s event.h header\n include.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Perf added \u0027u*\u0027 types mainly to ease up printing __u64 values\n as stated in the linux/types.h comment:\n \n /*\n * We define u64 as uint64_t for every architecture\n * so that we can print it with \"%\"PRIx64 without getting warnings.\n *\n * typedef __u64 u64;\n * typedef __s64 s64;\n */\n \n Adding and using new PRI_lu64 and PRI_lx64 macros to be used for\n that. Using extra \u0027_\u0027 to ease up the reading and differentiate\n them from standard PRI*64 macros.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/n/tip-ufs9ityr5w2xqwtd5w3p6dm4@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 1345e2ee87a83c758f336f03f7fb305bc5e24490\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Aug 25 20:17:41 2019 +0200\n\n libperf: Add PERF_RECORD_MMAP \u0027struct mmap_event\u0027 to perf/event.h\n \n Move the mmap_event event definition to libperf\u0027s event.h header\n include.\n \n In order to keep libperf simple, we switch \u0027u64/u32/u16/u8\u0027 types used\n events to their generic \u0027__u*\u0027 versions.\n \n Perf added \u0027u*\u0027 types mainly to ease up printing __u64 values as stated\n in the linux/types.h comment:\n \n /*\n * We define u64 as uint64_t for every architecture\n * so that we can print it with \"%\"PRIx64 without getting warnings.\n *\n * typedef __u64 u64;\n * typedef __s64 s64;\n */\n \n Add and use new PRI_lu64 and PRI_lx64 macros for that. Use extra \u0027_\u0027\n to ease up reading and differentiate them from standard PRI*64 macros.\n \n Committer notes:\n \n Fixup the PRI_l[ux]64 macros on 32-bit arches, conditionally defining it\n with that extra \u0027l\u0027 modifier only on arches where __u64 is long long,\n leaving it aside on 32-bit arches.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190825181752.722-2-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 19d608458f4f3bb3a1f89bd7e4814c3fd30dbec7\nAuthor: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\nDate: Tue Aug 20 21:36:07 2019 +0900\n\n clocksource/drivers/sh_cmt: Document \"cmt-48\" as deprecated\n \n Update the CMT driver to mark \"renesas,cmt-48\" as deprecated.\n \n Instead of documenting a theoretical hardware device based on current software\n support level, define DT bindings top-down based on available data sheet\n information and make use of part numbers in the DT compat string.\n \n In case of the only in-tree users r8a7740 and sh73a0 the compat strings\n \"renesas,r8a7740-cmt1\" and \"renesas,sh73a0-cmt1\" may be used instead.\n \n Signed-off-by: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 8c1afba285a86b9dbb0637f8c70a34fe2d88569e\nAuthor: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\nDate: Tue Aug 20 21:35:56 2019 +0900\n\n clocksource/drivers/sh_cmt: r8a7740 and sh73a0 SoC-specific match\n \n Add SoC-specific matching for CMT1 on r8a7740 and sh73a0.\n \n This allows us to move away from the old DT bindings such as\n - \"renesas,cmt-48-sh73a0\"\n - \"renesas,cmt-48-r8a7740\"\n - \"renesas,cmt-48\"\n in favour for the now commonly used format \"renesas,\u003csoc\u003e-\u003cdevice\u003e\"\n \n Signed-off-by: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 1be8c9fd2ac9ad730cf537b8909f66c357866c5d\nAuthor: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\nDate: Tue Aug 20 21:35:46 2019 +0900\n\n dt-bindings: timer: renesas, cmt: Update R-Car Gen3 CMT1 usage\n \n The R-Car Gen3 SoCs so far come with a total for 4 on-chip CMT devices:\n - CMT0\n - CMT1\n - CMT2\n - CMT3\n \n CMT0 includes two rather basic 32-bit timer channels. The rest of the on-chip\n CMT devices support 48-bit counters and have 8 channels each.\n \n Based on the data sheet information \"CMT2/3 are exactly same as CMT1\"\n it seems that CMT2 and CMT3 now use the CMT1 compat string in the DTSI.\n \n Clarify this in the DT binding documentation by describing R-Car Gen3 and\n RZ/G2 CMT1 as \"48-bit CMT devices\".\n \n Signed-off-by: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 53933bc3a69e0f07a1af2fea16fda9c816ffcf87\nAuthor: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\nDate: Tue Aug 20 21:35:36 2019 +0900\n\n dt-bindings: timer: renesas, cmt: Add CMT0 and CMT1 to r8a77995\n \n This patch adds DT binding documentation for the CMT devices on\n the R-Car Gen3 D3 (r8a77995) SoC.\n \n Signed-off-by: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 649dd060334f13792f624ec3fa8a0024ed1e02bc\nAuthor: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\nDate: Tue Aug 20 21:35:25 2019 +0900\n\n dt-bindings: timer: renesas, cmt: Add CMT0 and CMT1 to r8a7792\n \n This patch adds DT binding documentation for the CMT devices on\n the R-Car Gen2 V2H (r8a7792) SoC.\n \n Signed-off-by: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 81b604c39997de91f4b2912f803074c85045fe36\nAuthor: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\nDate: Tue Aug 20 21:35:14 2019 +0900\n\n dt-bindings: timer: renesas, cmt: Update CMT1 on sh73a0 and r8a7740\n \n This patch reworks the DT binding documentation for the 6-channel\n 48-bit CMTs known as CMT1 on r8a7740 and sh73a0.\n \n After the update the same style of DT binding as the rest of the upstream\n SoCs will now also be used by r8a7740 and sh73a0. The DT binding \"cmt-48\"\n is removed from the DT binding documentation, however software support for\n this deprecated binding will still remain in the CMT driver for some time.\n \n Signed-off-by: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit c90d37c9c41a572ea7183299951341b4640d5b4b\nAuthor: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\nDate: Tue Aug 20 21:35:03 2019 +0900\n\n dt-bindings: timer: renesas, cmt: Add CMT0234 to sh73a0 and r8a7740\n \n Document the on-chip CMT devices included in r8a7740 and sh73a0.\n \n Included in this patch is DT binding documentation for 32-bit CMTs\n CMT0, CMT2, CMT3 and CMT4. They all contain a single channel and are\n quite similar however some minor differences still exist:\n - \"Counter input clock\" (clock input and on-device divider)\n One example is that RCLK 1/1 is supported by CMT2, CMT3 and CMT4.\n - \"Wakeup request\" (supported by CMT0 and CMT2)\n \n Because of this one unique compat string per CMT device is selected.\n \n Signed-off-by: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 14e019df1e64c8b19ce8e0b3da25b6f40c8716be\nAuthor: Jon Hunter \u003cjonathanh@nvidia.com\u003e\nDate: Wed Aug 21 16:02:41 2019 +0100\n\n clocksource/drivers: Do not warn on probe defer\n \n Deferred probe is an expected return value on many platforms and so\n there\u0027s no need to output a warning that may potentially confuse users.\n \n Signed-off-by: Jon Hunter \u003cjonathanh@nvidia.com\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 763719771e84b8c8c2f53af668cdc905faa608de\nAuthor: Jon Hunter \u003cjonathanh@nvidia.com\u003e\nDate: Wed Aug 21 16:02:40 2019 +0100\n\n clocksource/drivers/timer-of: Do not warn on deferred probe\n \n Deferred probe is an expected return value for clk_get() on many\n platforms. The driver deals with it properly, so there\u0027s no need\n to output a warning that may potentially confuse users.\n \n Signed-off-by: Jon Hunter \u003cjonathanh@nvidia.com\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit a2b58537b4a1cc08fd254fb8d1c24191ce286ae1\nAuthor: Avi Fishman \u003cavifishman70@gmail.com\u003e\nDate: Mon Jul 29 20:03:54 2019 +0300\n\n clocksource/drivers/npcm: Fix GENMASK and timer operation\n \n NPCM7XX_Tx_OPER GENMASK bits are wrong, fix them.\n \n Hopefully the NPCM7XX_REG_TICR0 register reset value of those bits was 0,\n so it did not cause an issue.\n \n The function npcm7xx_timer_oneshot() reads the register\n NPCM7XX_REG_TCSR0, modifies it and then reads it again overwriting the\n previous changes. Remove the extra read which is pointless.\n \n The function npcm7xx_timer_periodic() is correct but the code writes\n to the NPCM7XX_REG_TICR0 register while it is dealing with the\n NPCM7XX_REG_TCSR0 register, that is confusing. Separate the write to\n the registers in the code for the sake of clarity.\n \n Fixes: 1c00289ecd12 (\"clocksource/drivers/npcm: Add NPCM7xx timer driver\")\n Signed-off-by: Avi Fishman \u003cavifishman70@gmail.com\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 8d18f6fc75c86b2efa5865655a0630a083c8a1fd\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Aug 7 10:46:33 2019 +0200\n\n clocksource/drivers/renesas-ostm: Use DIV_ROUND_CLOSEST() helper\n \n Use the DIV_ROUND_CLOSEST() helper instead of open-coding the same\n operation.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 24e8a5db8ae46bf021d3b4063c005f443282ab4f\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Thu Aug 15 20:38:44 2019 -0400\n\n arm64: dts: imx8mq: Add system counter node\n \n Add i.MX8MQ system counter node to enable timer-imx-sysctr\n broadcast timer driver.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 5b0221bf7b08163030e6801255ffbaf52775df01\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Thu Aug 15 20:38:43 2019 -0400\n\n arm64: dts: imx8mm: Add system counter node\n \n Add i.MX8MM system counter node to enable timer-imx-sysctr\n broadcast timer driver.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 4419e19d8117dda6bccfbc62090e766f469ff20a\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Thu Aug 15 20:38:42 2019 -0400\n\n clocksource/drivers/imx-sysctr: Add internal clock divider handle\n \n The system counter block guide states that the base clock is\n internally divided by 3 before use, that means the clock input of\n system counter defined in DT should be base clock which is normally\n from OSC, and then internally divided by 3 before use.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 1ce861cec03c79a68bae81a7e039edae46b2c493\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Tue Aug 13 15:30:50 2019 +0200\n\n clocksource/drivers/tcb_clksrc: Register delay timer\n \n Implement and register delay timer to allow get_cycles() to work properly.\n \n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit d9b51093cca430f75a054d78eb35a6e13c1540cb\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jul 22 10:12:22 2019 +0200\n\n dt-bindings: timer: Convert Allwinner A13 HSTimer to a schema\n \n The newer Allwinner SoCs have a High Speed Timer supported in Linux, with a\n matching Device Tree binding.\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for that controller over to a YAML schemas.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit bca4e08426e48d1f59b9bb7a8835b08140525828\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jul 22 10:12:21 2019 +0200\n\n clocksource: sun4i: Add missing compatibles\n \n Newer Allwinner SoCs have different number of interrupts, let\u0027s add\n different compatibles for all of them to deal with this properly.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Acked-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 7fccfcd678e80cc8cf131922296eccf72e19a69c\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jul 22 10:12:20 2019 +0200\n\n dt-bindings: timer: Add missing compatibles\n \n Newer Allwinner SoCs have different number of interrupts, let\u0027s add\n different compatibles for all of them to deal with this properly.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit a08bda2d27f271aacc92b75f4f260d3136cf6f58\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jul 22 10:12:19 2019 +0200\n\n dt-bindings: timer: Convert Allwinner A10 Timer to a schema\n \n The older Allwinner SoCs have a Timer supported in Linux, with a matching\n Device Tree binding.\n \n While the original binding only mentions one interrupt, the timer actually\n has 6 of them.\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for that controller over to a YAML schemas.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 9f475d084c032116cbecd4dc840003dc36465db5\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:04 2019 -0700\n\n clocksource: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit d333de3707785b10c4f42e14f3ff254ac15cd7dc\nAuthor: Fabien Dessenne \u003cfabien.dessenne@st.com\u003e\nDate: Mon Aug 26 17:38:29 2019 +0200\n\n remoteproc: stm32: manage the get_irq probe defer case\n \n Manage the -EPROBE_DEFER error case for \"wdg\" IRQ.\n \n Signed-off-by: Fabien Dessenne \u003cfabien.dessenne@st.com\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit abbe429d56168b90c03bc548403d72505e4fd6b2\nAuthor: Fabien Dessenne \u003cfabien.dessenne@st.com\u003e\nDate: Mon Aug 26 17:38:06 2019 +0200\n\n remoteproc: stm32: clear MCU PDDS at firmware start\n \n When remoteproc stops a firmware, it sets the MCU PDDS flag, allowing\n the platform to reach the Standby power mode if needed.\n Symmetrically, clear this flag before running a firmware.\n This fixes the issue where the system goes to Standby mode when the\n MCU crashes while MPU is in CSTOP mode.\n \n Signed-off-by: Fabien Dessenne \u003cfabien.dessenne@st.com\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit f2583fde2719362039a6efb3b4094f46284df541\nAuthor: Sibi Sankar \u003csibis@codeaurora.org\u003e\nDate: Wed Aug 21 23:35:48 2019 +0530\n\n remoteproc: qcom: q6v5-mss: fixup q6v5_pds_enable error handling\n \n dev_pm_domain_attach_by_name will return NULL if the requested\n power-domain is not a part device node. This could result in\n NULL pointer de-reference in q6v5_pds_enable. Fix this by\n checking for IS_ERR_OR_NULL and forward the appropriate error\n code.\n \n Fixes: 4760a896be88e (\"remoteproc: q6v5-mss: Vote for rpmh power domains\")\n Signed-off-by: Sibi Sankar \u003csibis@codeaurora.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 6ed756aa0148a5ad0dbdced6f14f22e2f5748d35\nAuthor: Suman Anna \u003cs-anna@ti.com\u003e\nDate: Fri Aug 9 17:20:57 2019 -0500\n\n remoteproc: Add a sysfs interface for name\n \n This patch adds a sysfs interface that provides the name of the\n remote processor to userspace. This allows the userspace to identify\n a remote processor as the remoteproc devices themselves are created\n based on probe order and can change from one boot to another or\n at runtime.\n \n The name is made available in debugfs originally, and is being\n retained for now. This can be cleaned up after couple of releases\n once users get familiar with the new interface.\n \n Signed-off-by: Suman Anna \u003cs-anna@ti.com\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 3c95e5013b7f204a03757691367857b7b668163a\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Mon Aug 26 22:52:36 2019 +0200\n\n r8169: improve DMA handling in rtl_rx\n \n Move the call to dma_sync_single_for_cpu after calling napi_alloc_skb.\n This avoids calling dma_sync_single_for_cpu w/o handing control back\n to device if the memory allocation should fail.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 918190f50eb630032ca0fb21f14f2e9c5c325626\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Mon Aug 26 16:45:06 2019 +0300\n\n net: sched: flower: don\u0027t take rtnl lock for cls hw offloads API\n \n Don\u0027t manually take rtnl lock in flower classifier before calling cls\n hardware offloads API. Instead, pass rtnl lock status via \u0027rtnl_held\u0027\n parameter.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1444c175a37443d3f6d3db825df050741452c3c3\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Mon Aug 26 16:45:05 2019 +0300\n\n net: sched: copy tunnel info when setting flow_action entry-\u003etunnel\n \n In order to remove dependency on rtnl lock, modify tc_setup_flow_action()\n to copy tunnel info, instead of just saving pointer to tunnel_key action\n tunnel info. This is necessary to prevent concurrent action overwrite from\n releasing tunnel info while it is being used by rtnl-unlocked driver.\n \n Implement helper tcf_tunnel_info_copy() that is used to copy tunnel info\n with all its options to dynamically allocated memory block. Modify\n tc_cleanup_flow_action() to free dynamically allocated tunnel info.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 5a6ff4b13d598573fc954f672cd2a267b76a01ec\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Mon Aug 26 16:45:04 2019 +0300\n\n net: sched: take reference to action dev before calling offloads\n \n In order to remove dependency on rtnl lock when calling hardware offload\n API, take reference to action mirred dev when initializing flow_action\n structure in tc_setup_flow_action(). Implement function\n tc_cleanup_flow_action(), use it to release the device after hardware\n offload API is done using it.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9838b20a7fb28c69fa66ac8e68d967ffe1d0ecad\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Mon Aug 26 16:45:03 2019 +0300\n\n net: sched: take rtnl lock in tc_setup_flow_action()\n \n In order to allow using new flow_action infrastructure from unlocked\n classifiers, modify tc_setup_flow_action() to accept new \u0027rtnl_held\u0027\n argument. Take rtnl lock before accessing tc_action data. This is necessary\n to protect from concurrent action replace.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 11bd634da25735a3f2f12112d02661d462a76792\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Mon Aug 26 16:45:02 2019 +0300\n\n net: sched: conditionally obtain rtnl lock in cls hw offloads API\n \n In order to remove dependency on rtnl lock from offloads code of\n classifiers, take rtnl lock conditionally before executing driver\n callbacks. Only obtain rtnl lock if block is bound to devices that require\n it.\n \n Block bind/unbind code is rtnl-locked and obtains block-\u003ecb_lock while\n holding rtnl lock. Obtain locks in same order in tc_setup_cb_*() functions\n to prevent deadlock.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c9f14470d04830de217f9d28fcd0deffd7e8c0b1\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Mon Aug 26 16:45:01 2019 +0300\n\n net: sched: add API for registering unlocked offload block callbacks\n \n Extend struct flow_block_offload with \"unlocked_driver_cb\" flag to allow\n registering and unregistering block hardware offload callbacks that do not\n require caller to hold rtnl lock. Extend tcf_block with additional\n lockeddevcnt counter that is incremented for each non-unlocked driver\n callback attached to device. This counter is necessary to conditionally\n obtain rtnl lock before calling hardware callbacks in following patches.\n \n Register mlx5 tc block offload callbacks as \"unlocked\".\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a449a3e77a85fc8b31fef7238451dc87af8ff1af\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Mon Aug 26 16:45:00 2019 +0300\n\n net: sched: notify classifier on successful offload add/delete\n \n To remove dependency on rtnl lock, extend classifier ops with new\n ops-\u003ehw_add() and ops-\u003ehw_del() callbacks. Call them from cls API while\n holding cb_lock every time filter if successfully added to or deleted from\n hardware.\n \n Implement the new API in flower classifier. Use it to manage hw_filters\n list under cb_lock protection, instead of relying on rtnl lock to\n synchronize with concurrent fl_reoffload() call.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 401192113730947572d280ec465555ab9ff5a597\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Mon Aug 26 16:44:59 2019 +0300\n\n net: sched: refactor block offloads counter usage\n \n Without rtnl lock protection filters can no longer safely manage block\n offloads counter themselves. Refactor cls API to protect block offloadcnt\n with tcf_block-\u003ecb_lock that is already used to protect driver callback\n list and nooffloaddevcnt counter. The counter can be modified by concurrent\n tasks by new functions that execute block callbacks (which is safe with\n previous patch that changed its type to atomic_t), however, block\n bind/unbind code that checks the counter value takes cb_lock in write mode\n to exclude any concurrent modifications. This approach prevents race\n conditions between bind/unbind and callback execution code but allows for\n concurrency for tc rule update path.\n \n Move block offload counter, filter in hardware counter and filter flags\n management from classifiers into cls hardware offloads API. Make functions\n tcf_block_offload_{inc|dec}() and tc_cls_offload_cnt_update() to be cls API\n private. Implement following new cls API to be used instead:\n \n tc_setup_cb_add() - non-destructive filter add. If filter that wasn\u0027t\n already in hardware is successfully offloaded, increment block offloads\n counter, set filter in hardware counter and flag. On failure, previously\n offloaded filter is considered to be intact and offloads counter is not\n decremented.\n \n tc_setup_cb_replace() - destructive filter replace. Release existing\n filter block offload counter and reset its in hardware counter and flag.\n Set new filter in hardware counter and flag. On failure, previously\n offloaded filter is considered to be destroyed and offload counter is\n decremented.\n \n tc_setup_cb_destroy() - filter destroy. Unconditionally decrement block\n offloads counter.\n \n tc_setup_cb_reoffload() - reoffload filter to single cb. Execute cb() and\n call tc_cls_offload_cnt_update() if cb() didn\u0027t return an error.\n \n Refactor all offload-capable classifiers to atomically offload filters to\n hardware, change block offload counter, and set filter in hardware counter\n and flag by means of the new cls API functions.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 97394bef5622cb32fd1e5d152251090da6c238b9\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Mon Aug 26 16:44:58 2019 +0300\n\n net: sched: change tcf block offload counter type to atomic_t\n \n As a preparation for running proto ops functions without rtnl lock, change\n offload counter type to atomic. This is necessary to allow updating the\n counter by multiple concurrent users when offloading filters to hardware\n from unlocked classifiers.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4f8116c85057239ff37519debdd5d45b38ad8130\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Mon Aug 26 16:44:57 2019 +0300\n\n net: sched: protect block offload-related fields with rw_semaphore\n \n In order to remove dependency on rtnl lock, extend tcf_block with \u0027cb_lock\u0027\n rwsem and use it to protect flow_block-\u003ecb_list and related counters from\n concurrent modification. The lock is taken in read mode for read-only\n traversal of cb_list in tc_setup_cb_call() and write mode in all other\n cases. This approach ensures that:\n \n - cb_list is not changed concurrently while filters is being offloaded on\n block.\n \n - block-\u003enooffloaddevcnt is checked while holding the lock in read mode,\n but is only changed by bind/unbind code when holding the cb_lock in write\n mode to prevent concurrent modification.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b5e55556182d2e43da035df1bffbd492c72a7994\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Aug 26 12:57:25 2019 +0200\n\n libata: switch remaining drivers to use dma_set_mask_and_coherent\n \n Use dma_set_mask_and_coherent instead of separate dma_set_mask and\n dma_set_coherent_mask calls.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit dcc02c19cc06bd7bc1b6db0aa0087a2b6eb05b94\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Aug 26 12:57:24 2019 +0200\n\n sata_sil24: use dma_set_mask_and_coherent\n \n Use the dma_set_mask_and_coherent helper to set the DMA mask. Rely\n on the relatively recent change that setting a larger than required\n mask will never fail to avoid the need for the boilerplate 32-bit\n fallback code.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 440bd77f09b8a11a5c729ef61f346421e4d10b87\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Aug 26 12:57:23 2019 +0200\n\n sata_qstor: use dma_set_mask_and_coherent\n \n Use the dma_set_mask_and_coherent helper to set the DMA mask. Rely\n on the relatively recent change that setting a larger than required\n mask will never fail to avoid the need for the boilerplate 32-bit\n fallback code.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 51872b6606eb52a5c8b82e79c644d352fdb41d04\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Aug 26 12:57:22 2019 +0200\n\n sata_nv: use dma_set_mask_and_coherent\n \n Use the dma_set_mask_and_coherent helper to set the DMA mask. Rely\n on the relatively recent change that setting a larger than required\n mask will never fail to avoid the need for the boilerplate 32-bit\n fallback code.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 496d4575e7ac57f4ee3457c10be214e6b3ce0bd0\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Aug 26 12:57:21 2019 +0200\n\n sata_mv: use dma_set_mask_and_coherent\n \n Use the dma_set_mask_and_coherent helper to set the DMA mask. Rely\n on the relatively recent change that setting a larger than required\n mask will never fail to avoid the need for the boilerplate 32-bit\n fallback code.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 94c58148a6bcb1f2c342fb806409d35019b4bb12\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Aug 26 12:57:20 2019 +0200\n\n pdc_adma: use dma_set_mask_and_coherent\n \n Use the dma_set_mask_and_coherent helper to set the DMA mask. Rely\n on the relatively recent change that setting a larger than required\n mask will never fail to avoid the need for the boilerplate 32-bit\n fallback code.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit b1716871c00dd48a8efa8ee93007490c939fc8c3\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Aug 26 12:57:19 2019 +0200\n\n ahci: use dma_set_mask_and_coherent\n \n Use the dma_set_mask_and_coherent helper to set the DMA mask. Rely\n on the relatively recent change that setting a larger than required\n mask will never fail to avoid the need for the boilerplate 32-bit\n fallback code.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 759ad0979808a586816cf4f01bf5784080343d31\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Aug 26 12:57:18 2019 +0200\n\n acard_ahci: use dma_set_mask_and_coherent\n \n Use the dma_set_mask_and_coherent helper to set the DMA mask. Rely\n on the relatively recent change that setting a larger than required\n mask will never fail to avoid the need for the boilerplate 32-bit\n fallback code.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit c1499a044d7c2bbd9ea3d9c17e17f16595ae6d69\nAuthor: ZhangXiaoxu \u003czhangxiaoxu5@huawei.com\u003e\nDate: Sun Jun 9 15:52:00 2019 +0800\n\n dm space map common: remove check for impossible sm_find_free() return value\n \n The function sm_find_free() just returns -ENOSPC and 0.\n So remove lone caller\u0027s check for some other error.\n \n Signed-off-by: ZhangXiaoxu \u003czhangxiaoxu5@huawei.com\u003e\n Signed-off-by: Mike Snitzer \u003csnitzer@redhat.com\u003e\n\ncommit 707e0ddaf67e8942448ebdd16b523e409ebe40ce\nAuthor: Tetsuo Handa \u003cpenguin-kernel@I-love.SAKURA.ne.jp\u003e\nDate: Mon Aug 26 12:06:22 2019 -0700\n\n fs: xfs: Remove KM_NOSLEEP and KM_SLEEP.\n \n Since no caller is using KM_NOSLEEP and no callee branches on KM_SLEEP,\n we can remove KM_NOSLEEP and replace KM_SLEEP with 0.\n \n Signed-off-by: Tetsuo Handa \u003cpenguin-kernel@I-love.SAKURA.ne.jp\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit de4c5bafb6d5cd85e1dc0afd9cd170b8c37ddc76\nAuthor: Ilya Dryomov \u003cidryomov@gmail.com\u003e\nDate: Wed Aug 21 13:57:18 2019 +0200\n\n libceph: drop unused con parameter of calc_target()\n \n This bit was omitted from a561372405cf (\"libceph: fix PG split vs OSD\n (re)connect race\") to avoid backport conflicts.\n \n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit 4dcaccf64c6c25cd2e52acdbcffb23fcd8adf34f\nAuthor: John Hubbard \u003cjhubbard@nvidia.com\u003e\nDate: Thu Aug 8 20:56:47 2019 -0700\n\n ceph: use release_pages() directly\n \n release_pages() has been available to modules since Oct, 2010,\n when commit 0be8557bcd34 (\"fuse: use release_pages()\") added\n EXPORT_SYMBOL(release_pages). However, this ceph code was still\n using a workaround.\n \n Remove the workaround, and call release_pages() directly.\n \n Signed-off-by: John Hubbard \u003cjhubbard@nvidia.com\u003e\n Reviewed-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit 3abcabf3165d343ba1d94f3067447af698f88c0b\nAuthor: Dongsheng Yang \u003cdongsheng.yang@easystack.cn\u003e\nDate: Fri Aug 9 07:05:27 2019 +0000\n\n rbd: fix response length parameter for encoded strings\n \n rbd_dev_image_id() allocates space for length but passes a smaller\n value to rbd_obj_method_sync(). rbd_dev_v2_object_prefix() doesn\u0027t\n allocate space for length. Fix both to be consistent.\n \n Signed-off-by: Dongsheng Yang \u003cdongsheng.yang@easystack.cn\u003e\n Reviewed-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit 85a14ed5fb5025213a6294c74f2e05fd9e04d1e2\nAuthor: Jeff Layton \u003cjlayton@kernel.org\u003e\nDate: Tue Aug 6 10:55:41 2019 -0400\n\n ceph: allow arbitrary security.* xattrs\n \n Most filesystems don\u0027t limit what security.* xattrs can be set or\n fetched. I see no reason that we need to limit that on cephfs either.\n \n Drop the special xattr handler for \"security.\" xattrs, and allow the\n \"other\" xattr handler to handle security xattrs as well.\n \n In addition to fixing xfstest generic/093, this allows us to support\n per-file capabilities (a\u0027la setcap(8)).\n \n Link: https://tracker.ceph.com/issues/41135\n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit ce4b27e84c6017775adc8601928857583e28a837\nAuthor: Jeff Layton \u003cjlayton@kernel.org\u003e\nDate: Tue Aug 6 10:41:40 2019 -0400\n\n ceph: only set CEPH_I_SEC_INITED if we got a MAC label\n \n __ceph_getxattr will set the CEPH_I_SEC_INITED flag whenever it gets\n any xattr that starts with \"security.\". We only want to set that flag\n when fetching the MAC label for the currently-active LSM, however.\n \n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit 9f72c3964ce71dde66c5698d8f14666513e9e659\nAuthor: Jeff Layton \u003cjlayton@kernel.org\u003e\nDate: Tue Aug 6 09:07:51 2019 -0400\n\n ceph: turn ceph_security_invalidate_secctx into static inline\n \n No need to do an extra jump here. Also add some comments on the endifs.\n \n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit 668835e1e0bfb50a88650a733a042f252e151ea4\nAuthor: Jeff Layton \u003cjlayton@kernel.org\u003e\nDate: Fri Aug 2 13:15:39 2019 -0400\n\n ceph: add buffered/direct exclusionary locking for reads and writes\n \n xfstest generic/451 intermittently fails. The test does O_DIRECT writes\n to a file, and then reads back the result using buffered I/O, while\n running a separate set of tasks that are also doing buffered reads.\n \n The client will invalidate the cache prior to a direct write, but it\u0027s\n easy for one of the other readers\u0027 replies to race in and reinstantiate\n the invalidated range with stale data.\n \n To fix this, we must to serialize direct I/O writes and buffered reads.\n We could just sprinkle in some shared locks on the i_rwsem for reads,\n and increase the exclusive footprint on the write side, but that would\n cause O_DIRECT writes to end up serialized vs. other direct requests.\n \n Instead, borrow the scheme used by nfs.ko. Buffered writes take the\n i_rwsem exclusively, but buffered reads take a shared lock, allowing\n them to run in parallel.\n \n O_DIRECT requests also take a shared lock, but we need for them to not\n run in parallel with buffered reads. A flag on the ceph_inode_info is\n used to indicate whether it\u0027s in direct or buffered I/O mode. When a\n conflicting request is submitted, it will block until the inode can be\n flipped to the necessary mode.\n \n Link: https://tracker.ceph.com/issues/40985\n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Reviewed-by: \"Yan, Zheng\" \u003czyan@redhat.com\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit c4fa8b84b811139c77632af6fa6f0aacc3e3fa40\nAuthor: David Disseldorp \u003cddiss@suse.de\u003e\nDate: Wed Jul 3 17:59:20 2019 +0200\n\n libceph: handle OSD op ceph_pagelist_append() errors\n \n osd_req_op_cls_init() and osd_req_op_xattr_init() currently propagate\n ceph_pagelist_alloc() ENOMEM errors but ignore ceph_pagelist_append()\n memory allocation failures. Add these checks and cleanup on error.\n \n Signed-off-by: David Disseldorp \u003cddiss@suse.de\u003e\n Reviewed-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit f8ff616b64225fca7f3d42a12faa2ea0c6a77a4a\nAuthor: John Hubbard \u003cjhubbard@nvidia.com\u003e\nDate: Thu Aug 1 18:06:58 2019 -0700\n\n ceph: don\u0027t return a value from void function\n \n This fixes a build warning to that effect.\n \n Fixes: 1a829ff2a6c3 (\"ceph: no need to check return value of debugfs_create functions\")\n Signed-off-by: John Hubbard \u003cjhubbard@nvidia.com\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit 831bcd9be6ed946c5032df4f4e9468dc97b297a6\nAuthor: Jeff Layton \u003cjlayton@kernel.org\u003e\nDate: Thu Aug 1 10:06:40 2019 -0400\n\n ceph: don\u0027t freeze during write page faults\n \n Prevent freezing operations during write page faults. This is good\n practice for most filesystems, but especially for ceph since we\u0027re\n monkeying with the signal table here.\n \n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Reviewed-by: \"Yan, Zheng\" \u003czyan@redhat.com\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit cf491ee7cb50c7e02aac8d92fba844edf0ddaf5b\nAuthor: Jia-Ju Bai \u003cbaijiaju1990@gmail.com\u003e\nDate: Wed Jul 24 17:43:06 2019 +0800\n\n libceph: don\u0027t call crypto_free_sync_skcipher() on a NULL tfm\n \n In set_secret(), key-\u003etfm is assigned to NULL on line 55, and then\n ceph_crypto_key_destroy(key) is executed.\n \n ceph_crypto_key_destroy(key)\n crypto_free_sync_skcipher(key-\u003etfm)\n crypto_free_skcipher(\u0026tfm-\u003ebase);\n \n This happens to work because crypto_sync_skcipher is a trivial wrapper\n around crypto_skcipher: \u0026tfm-\u003ebase is still 0 and crypto_free_skcipher()\n handles that. Let\u0027s not rely on the layout of crypto_sync_skcipher.\n \n This bug is found by a static analysis tool STCheck written by us.\n \n Fixes: 69d6302b65a8 (\"libceph: Remove VLA usage of skcipher\").\n Signed-off-by: Jia-Ju Bai \u003cbaijiaju1990@gmail.com\u003e\n Reviewed-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit 7cc259d32779f567d3b066ad004d7210cbc36ccc\nAuthor: Jeff Layton \u003cjlayton@kernel.org\u003e\nDate: Thu Jul 25 13:03:32 2019 -0400\n\n ceph: update the mtime when truncating up\n \n If we have Fx caps, and the we\u0027re truncating the size to be larger, then\n we\u0027ll cache the size attribute change, but the mtime won\u0027t be updated.\n \n Move the size handling before the mtime, and add ATTR_MTIME to ia_valid\n in that case to make sure the mtime also gets updated.\n \n This fixes xfstest generic/313.\n \n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit 1d84fc7085d4e6310144de8480d8c46443f88ff0\nAuthor: Ilya Dryomov \u003cidryomov@gmail.com\u003e\nDate: Mon Jul 29 11:33:08 2019 +0200\n\n ceph: fix indentation in __get_snap_name()\n \n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Reported-by: Julia Lawall \u003cjulia.lawall@lip6.fr\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit f4b8a1665810eb88fd5e53d64908ce7c0ee78e53\nAuthor: Jeff Layton \u003cjlayton@kernel.org\u003e\nDate: Fri Jul 5 13:11:51 2019 -0400\n\n ceph: remove incorrect comment above __send_cap\n \n It doesn\u0027t do anything to invalidate the cache when dropping RD caps.\n \n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Reviewed-by: \"Yan, Zheng\" \u003czyan@redhat.com\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit 1ca05ac5da7b7c45cbb8d485774457853ac57553\nAuthor: Jeff Layton \u003cjlayton@kernel.org\u003e\nDate: Fri Jul 5 10:55:38 2019 -0400\n\n ceph: remove CEPH_I_NOFLUSH\n \n Nothing sets this flag.\n \n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Reviewed-by: \"Yan, Zheng\" \u003czyan@redhat.com\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit c2e547dd46e53533bbfe269870077b5827e533a4\nAuthor: Jeff Layton \u003cjlayton@kernel.org\u003e\nDate: Fri Jul 5 13:26:29 2019 -0400\n\n ceph: remove unneeded test in try_flush_caps\n \n cap-\u003esession is always non-NULL, so we can just do a single test for\n equality w/o testing explicitly for a NULL pointer.\n \n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Reviewed-by: \"Yan, Zheng\" \u003czyan@redhat.com\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit dc8b8b4ecded3ac3cc7c8b77eaca17b48d6eae42\nAuthor: Jeff Layton \u003cjlayton@kernel.org\u003e\nDate: Mon Jul 8 12:27:57 2019 -0400\n\n ceph: have __mark_caps_flushing return flush_tid\n \n Currently, this function returns ci-\u003ei_dirty_caps, but the callers have\n to check that that isn\u0027t 0 before calling this function. Have the\n callers grab that value directly out of the inode, and have\n __mark_caps_flushing return the flush_tid instead.\n \n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Reviewed-by: \"Yan, Zheng\" \u003czyan@redhat.com\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit e1b5cb971b17b15b293bafb8fc1672c909eb710a\nAuthor: Jeff Layton \u003cjlayton@kernel.org\u003e\nDate: Fri Jul 19 09:41:02 2019 -0400\n\n ceph: fix comments over ceph_add_cap\n \n We actually need the ci-\u003ei_ceph_lock here. The necessity of the s_mutex\n is less clear. Also add a lockdep assertion for the i_ceph_lock.\n \n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Reviewed-by: \"Yan, Zheng\" \u003czyan@redhat.com\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit 71fa86036b15cd4c108da5ee41bb40e7bc275548\nAuthor: Jeff Layton \u003cjlayton@kernel.org\u003e\nDate: Fri Jul 19 15:22:28 2019 -0400\n\n ceph: eliminate session-\u003es_trim_caps\n \n It\u0027s only used to keep count of caps being trimmed, but that requires\n that we hold the session-\u003es_mutex to prevent multiple trimming\n operations from running concurrently.\n \n We can achieve the same effect using an integer on the stack, which\n allows us to (eventually) not need the s_mutex.\n \n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Reviewed-by: \"Yan, Zheng\" \u003czyan@redhat.com\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit aa9c8987317d79796b3eeb7b26044991ed122bc0\nAuthor: Jeff Layton \u003cjlayton@kernel.org\u003e\nDate: Mon Jul 22 13:12:01 2019 -0400\n\n ceph: fetch cap_gen under spinlock in ceph_add_cap\n \n It\u0027s protected by the s_gen_ttl_lock, so we should fetch under it\n and ensure that we\u0027re using the same generation in both places.\n \n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Reviewed-by: \"Yan, Zheng\" \u003czyan@redhat.com\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit 42c4946539a27b44ed1144240948bcd17f4fd881\nAuthor: Jeff Layton \u003cjlayton@kernel.org\u003e\nDate: Tue Jul 23 15:09:50 2019 -0400\n\n ceph: remove ceph_get_cap_mds and __ceph_get_cap_mds\n \n Nothing calls these routines.\n \n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit f9d685e1299bcf5a27ed95263a40dbae46fb63b8\nAuthor: Jeff Layton \u003cjlayton@kernel.org\u003e\nDate: Tue Jul 2 12:35:52 2019 -0400\n\n ceph: don\u0027t SetPageError on writepage errors\n \n We already mark the mapping in that case, and doing this can cause\n false positives to occur at fsync time, as well as spurious read\n errors.\n \n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit 8e4133936f30a39cde4c6f836af79bbf466ba4a4\nAuthor: Yan, Zheng \u003czyan@redhat.com\u003e\nDate: Thu Jul 25 20:16:47 2019 +0800\n\n ceph: auto reconnect after blacklisted\n \n Make client use osd reply and session message to infer if itself is\n blacklisted. Client reconnect to cluster using new entity addr if it\n is blacklisted. Auto reconnect is limited to once every 30 minutes.\n \n Auto reconnect is disabled by default. It can be enabled/disabled by\n recover_session\u003d\u003cno|clean\u003e mount option. In \u0027clean\u0027 mode, client drops\n any dirty data/metadata, invalidates page caches and invalidates all\n writable file handles. After reconnect, file locks become stale because\n MDS loses track of them. If an inode contains any stale file locks,\n read/write on the indoe are not allowed until applications release all\n stale file locks.\n \n Signed-off-by: \"Yan, Zheng\" \u003czyan@redhat.com\u003e\n Reviewed-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit 5f01ed63482fb7e83cb12ccdb22e7b4b361ce15e\nAuthor: Yan, Zheng \u003czyan@redhat.com\u003e\nDate: Thu Jul 25 20:16:46 2019 +0800\n\n ceph: invalidate all write mode filp after reconnect\n \n Signed-off-by: \"Yan, Zheng\" \u003czyan@redhat.com\u003e\n Reviewed-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit d306b90e6b6d01bcd9dba8124b71ce5c6f0cb49d\nAuthor: Yan, Zheng \u003czyan@redhat.com\u003e\nDate: Thu Jul 25 20:16:45 2019 +0800\n\n ceph: return -EIO if read/write against filp that lost file locks\n \n After mds evicts session, file locks get lost sliently. It\u0027s not safe to\n let programs continue to do read/write.\n \n Signed-off-by: \"Yan, Zheng\" \u003czyan@redhat.com\u003e\n Reviewed-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit 4beefaa67a4fd99fcd86daeb26f813c50db46109\nAuthor: Yan, Zheng \u003czyan@redhat.com\u003e\nDate: Thu Jul 25 20:16:44 2019 +0800\n\n ceph: add helper function that forcibly reconnects to ceph cluster.\n \n It closes mds sessions, drop all caps and invalidates page caches,\n then use new entity address to reconnect to the cluster.\n \n After reconnect, all dirty data/metadata are dropped, file locks\n get lost sliently. Open files continue to work because client will\n try renewing caps on later read/write.\n \n Signed-off-by: \"Yan, Zheng\" \u003czyan@redhat.com\u003e\n Reviewed-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit d75595772be3c0c0cedacbf440b19e3b5366aae3\nAuthor: Yan, Zheng \u003czyan@redhat.com\u003e\nDate: Thu Jul 25 20:16:43 2019 +0800\n\n ceph: pass filp to ceph_get_caps()\n \n Also change several other functions\u0027 arguments, no logical changes.\n This is preparetion for later patch that checks filp error.\n \n Signed-off-by: \"Yan, Zheng\" \u003czyan@redhat.com\u003e\n Reviewed-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit f112962eb3d62c6980e9209de52193abd4207e00\nAuthor: Yan, Zheng \u003czyan@redhat.com\u003e\nDate: Thu Jul 25 20:16:42 2019 +0800\n\n ceph: track and report error of async metadata operation\n \n Use errseq_t to track and report errors of async metadata operations,\n similar to how kernel handles errors during writeback.\n \n If any dirty caps or any unsafe request gets dropped during session\n eviction, record -EIO in corresponding inode\u0027s i_meta_err. The error\n will be reported by subsequent fsync,\n \n Signed-off-by: \"Yan, Zheng\" \u003czyan@redhat.com\u003e\n Reviewed-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit 54602778cc074065e1b3b7c53d932e7db5d36997\nAuthor: Yan, Zheng \u003czyan@redhat.com\u003e\nDate: Thu Jul 25 20:16:41 2019 +0800\n\n ceph: allow closing session in restarting/reconnect state\n \n CEPH_MDS_SESSION_{RESTARTING,RECONNECTING} are for for mds failover,\n they are sub-states of CEPH_MDS_SESSION_OPEN. So __close_session()\n should send close request for session in these two state.\n \n Signed-off-by: \"Yan, Zheng\" \u003czyan@redhat.com\u003e\n Reviewed-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit 3a6694eca4a940aafdb09b28c8752bbf5696d856\nAuthor: Yan, Zheng \u003czyan@redhat.com\u003e\nDate: Thu Jul 25 20:16:40 2019 +0800\n\n libceph: add function that clears osd client\u0027s abort_err\n \n Signed-off-by: \"Yan, Zheng\" \u003czyan@redhat.com\u003e\n Reviewed-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit 60751bf8dd6615947c4fc69fd4de24610a443972\nAuthor: Yan, Zheng \u003czyan@redhat.com\u003e\nDate: Thu Jul 25 20:16:39 2019 +0800\n\n libceph: add function that reset client\u0027s entity addr\n \n This function also re-open connections to OSD/MON, and re-send in-flight\n OSD requests after re-opening connections to OSD.\n \n Signed-off-by: \"Yan, Zheng\" \u003czyan@redhat.com\u003e\n Reviewed-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit fdc5a123442bf1b7881f248236c2c5ae772d0cff\nAuthor: Jeff Layton \u003cjlayton@kernel.org\u003e\nDate: Wed Jul 24 12:46:20 2019 -0400\n\n ceph: don\u0027t list vxattrs in listxattr()\n \n Most filesystems that provide virtual xattrs (e.g. CIFS) don\u0027t display\n them via listxattr(). Ceph does, and that causes some of the tests in\n xfstests to fail.\n \n Have cephfs stop listing vxattrs in listxattr. Userspace can always\n query them directly when the name is known.\n \n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Acked-by: David Disseldorp \u003cddiss@suse.de\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit 6bd826f4b05b1efb0d6742f771161e6f2e4eab75\nAuthor: Jeff Layton \u003cjlayton@kernel.org\u003e\nDate: Wed Jul 24 07:59:51 2019 -0400\n\n ceph: allow copy_file_range when src and dst inode are same\n \n There is no reason to prevent this. The OSD should be able to handle\n this as long as the objects are different, and the existing code falls\n back when the offset into the object is different.\n \n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Acked-by: Luis Henriques \u003clhenriques@suse.com\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit c029bd6cfbd06397d9f76272a7d50e02f24bcac5\nAuthor: Luis Henriques \u003clhenriques@suse.com\u003e\nDate: Tue Jul 23 16:50:20 2019 +0100\n\n ceph: fix directories inode i_blkbits initialization\n \n When filling an inode with info from the MDS, i_blkbits is being\n initialized using fl_stripe_unit, which contains the stripe unit in\n bytes. Unfortunately, this doesn\u0027t make sense for directories as they\n have fl_stripe_unit set to \u00270\u0027. This means that i_blkbits will be set\n to 0xff, causing an UBSAN undefined behaviour in i_blocksize():\n \n UBSAN: Undefined behaviour in ./include/linux/fs.h:731:12\n shift exponent 255 is too large for 32-bit type \u0027int\u0027\n \n Fix this by initializing i_blkbits to CEPH_BLOCK_SHIFT if fl_stripe_unit\n is zero.\n \n Signed-off-by: Luis Henriques \u003clhenriques@suse.com\u003e\n Reviewed-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n Signed-off-by: Ilya Dryomov \u003cidryomov@gmail.com\u003e\n\ncommit 1dbb9fb4082ce2a2f1cf9596881ddece062d15d0\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Fri Aug 16 13:52:29 2019 +0300\n\n leds: Allow to call led_classdev_unregister() unconditionally\n \n If in the certain driver the LED is optional, and it\u0027s a majority of them,\n the call of led_classdev_unregister() still requires some additional\n checks.\n \n The usual pattern on unregistering is to check for NULL, but we also check\n for IS_ERR() in case device_create_with_groups() fails.\n \n The change will reduce a burden in a lot of drivers to repeatedly check\n for above conditions.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 517ea49a7b23c6730ed1a23f152fcd32173eb300\nAuthor: Dan Murphy \u003cdmurphy@ti.com\u003e\nDate: Tue Aug 20 14:53:07 2019 -0500\n\n leds: lm3532: Add full scale current configuration\n \n Allow the full scale current to be configured at init.\n Valid rangles are 5mA-\u003e29.8mA.\n \n Signed-off-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit aa01d255cf101d027f3fabae778cf3eca4d0d899\nAuthor: Dan Murphy \u003cdmurphy@ti.com\u003e\nDate: Tue Aug 20 14:53:06 2019 -0500\n\n dt: lm3532: Add property for full scale current.\n \n Add a property for each control bank to configure the\n full scale current setting for the device.\n \n Signed-off-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 6559ac32998248182572e1ccae79dc2eb40ac7c6\nAuthor: Dan Murphy \u003cdmurphy@ti.com\u003e\nDate: Tue Aug 20 14:53:05 2019 -0500\n\n leds: lm3532: Fixes for the driver for stability\n \n Fixed misspelled words, added error check during probe\n on the init of the registers, and fixed ALS/I2C control\n mode.\n \n Fixes: bc1b8492c764 (\"leds: lm3532: Introduce the lm3532 LED driver\")\n Reported-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 4c9054506cf970759a9ded0e45c400dd03f37f51\nAuthor: Dan Murphy \u003cdmurphy@ti.com\u003e\nDate: Tue Aug 20 14:53:04 2019 -0500\n\n leds: lm3532: Change the define for the fs current register\n \n Change the define name of the full scale current registers.\n \n Signed-off-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 1312394023dd288946757ba5d7ee89d70709fc33\nAuthor: Dan Murphy \u003cdmurphy@ti.com\u003e\nDate: Tue Aug 20 14:53:03 2019 -0500\n\n leds: lm3532: Fix brightness control for i2c mode\n \n Fix the brightness control for I2C mode. Instead of\n changing the full scale current register update the ALS target\n register for the appropriate banks.\n \n In addition clean up some code errors and random misspellings found\n during coding.\n \n Tested on Droid4 as well as LM3532 EVM connected to a BeagleBoneBlack\n \n Fixes: bc1b8492c764 (\"leds: lm3532: Introduce the lm3532 LED driver\")\n Reported-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 1170f235834bfb858f691670b127f8673af5923b\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Mon Aug 26 10:48:29 2019 -0700\n\n ARM: dts: Drop legacy custom hwmods property for dra7 gpio\n \n With recent ti-sysc driver changes, we can probe most devices with device\n tree data only and drop the custom \"ti,hwmods\" property.\n \n We have already added the related device tree data earlier, and have\n already dropped the platform data. But we have been still dynamically\n allocating the platform data based on \"ti,hwmods\" property. With recent\n ti-sysc driver changes this is no longer needed.\n \n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 54b7026fc2015797e1013341c34e54ce63b76238\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Mon Aug 26 10:59:19 2019 -0700\n\n ARM: dts: Drop legacy custom hwmods property for dra7 mmc\n \n With recent ti-sysc driver changes, we can probe most devices with device\n tree data only and drop the custom \"ti,hwmods\" property.\n \n We have already added the related device tree data earlier, and have\n already dropped the platform data. But we have been still dynamically\n allocating the platform data based on \"ti,hwmods\" property. With recent\n ti-sysc driver changes this is no longer needed.\n \n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 44a2cdf0f14b6188f426524ecb98464ea4bbe1f1\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Mon Aug 26 10:59:19 2019 -0700\n\n ARM: dts: Drop legacy custom hwmods property for dra7 i2c\n \n With recent ti-sysc driver changes, we can probe most devices with device\n tree data only and drop the custom \"ti,hwmods\" property.\n \n We have already added the related device tree data earlier, and have\n already dropped the platform data. But we have been still dynamically\n allocating the platform data based on \"ti,hwmods\" property. With recent\n ti-sysc driver changes this is no longer needed.\n \n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 71feab2866f427f5d6991dc233a7c4f68a26b0db\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Mon Aug 26 10:59:19 2019 -0700\n\n ARM: dts: Drop legacy custom hwmods property for dra7 uart\n \n With recent ti-sysc driver changes, we can probe most devices with device\n tree data only and drop the custom \"ti,hwmods\" property.\n \n We have already added the related device tree data earlier, and have\n already dropped the platform data. But we have been still dynamically\n allocating the platform data based on \"ti,hwmods\" property. With recent\n ti-sysc driver changes this is no longer needed.\n \n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit b8637a68254fe8910b00241d6c21021dc6534cd5\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Mon Aug 26 10:59:19 2019 -0700\n\n ARM: OMAP2+: Drop legacy platform data for dra7 mcasp\n \n With recent ti-sysc driver changes, we can probe most devices with device\n tree data only and drop the custom \"ti,hwmods\" property.\n \n Let\u0027s drop the legacy platform data and custom \"ti,hwmods\" property. We\n want to do this in a single patch as the \"ti,hwmods\" property is used to\n allocate platform data dynamically that we no longer want to do.\n \n Cc: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit d1daf63b0bd016f0bc90c31a424e560c2965a210\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Mon Aug 26 10:59:19 2019 -0700\n\n ARM: OMAP2+: Drop legacy platform data for dra7 mcspi\n \n With recent ti-sysc driver changes, we can probe most devices with device\n tree data only and drop the custom \"ti,hwmods\" property.\n \n Let\u0027s drop the legacy platform data and custom \"ti,hwmods\" property. We\n want to do this in a single patch as the \"ti,hwmods\" property is used to\n allocate platform data dynamically that we no longer want to do.\n \n Cc: Vignesh R \u003cvigneshr@ti.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 95c520690f5fafb2cda2ec17f8c76ab3422b0174\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Thu Aug 22 11:16:42 2019 -0300\n\n media: don\u0027t do a 31 bit shift on a signed int\n \n On 32-bits archs, a signed integer has 31 bits plus on extra\n bit for signal. Due to that, touching the 32th bit with something\n like:\n \n int bar \u003d 1 \u003c\u003c 31;\n \n has an undefined behavior in C on 32 bit architectures, as it\n touches the signal bit. This is warned by cppcheck.\n \n Instead, force the numbers to be unsigned, in order to solve this\n issue.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit cce8ccca80d8388982133192d0a6d9dc2e8ed712\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Thu Aug 22 11:06:32 2019 -0300\n\n media: use the BIT() macro\n \n As warned by cppcheck:\n \n [drivers/media/dvb-frontends/cx24123.c:434]: (error) Shifting signed 32-bit value by 31 bits is undefined behaviour\n [drivers/media/pci/bt8xx/bttv-input.c:87]: (error) Shifting signed 32-bit value by 31 bits is undefined behaviour\n [drivers/media/pci/bt8xx/bttv-input.c:98]: (error) Shifting signed 32-bit value by 31 bits is undefined behaviour\n ...\n [drivers/media/v4l2-core/v4l2-ioctl.c:1391]: (error) Shifting signed 32-bit value by 31 bits is undefined behaviour\n \n There are lots of places where we\u0027re doing 1 \u003c\u003c 31. That\u0027s bad,\n as, depending on the architecture, this has an undefined behavior.\n \n The BIT() macro is already prepared to handle this, so, let\u0027s\n just switch all \"1 \u003c\u003c number\" macros by BIT(number) at the header files\n with has 1 \u003c\u003c 31.\n \n Reviewed-by: Sylwester Nawrocki \u003cs.nawrocki@samsung.com\u003e # exynos4-is and s3c-camif\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e # omap3isp, vsp1, xilinx, wl128x and ipu3\n Reviewed-by: Benoit Parrot \u003cbparrot@ti.com\u003e # am437x and ti-vpe\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 093347abc7a4e0490e3c962ecbde2dc272a8f708\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Thu Aug 22 11:55:07 2019 -0300\n\n media: ov9650: add a sanity check\n \n As pointed by cppcheck:\n \n [drivers/media/i2c/ov9650.c:706]: (error) Shifting by a negative value is undefined behaviour\n [drivers/media/i2c/ov9650.c:707]: (error) Shifting by a negative value is undefined behaviour\n [drivers/media/i2c/ov9650.c:721]: (error) Shifting by a negative value is undefined behaviour\n \n Prevent mangling with gains with invalid values.\n \n As pointed by Sylvester, this should never happen in practice,\n as min value of V4L2_CID_GAIN control is 16 (gain is always \u003e\u003d 16\n and m is always \u003e\u003d 0), but it is too hard for a static analyzer\n to get this, as the logic with validates control min/max is\n elsewhere inside V4L2 core.\n \n Reviewed-by: Sylwester Nawrocki \u003cs.nawrocki@samsung.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 31b8b0bd6e55c3ea5a08bb8141fa5d3c90600e3b\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Thu Aug 22 11:46:40 2019 -0300\n\n media: aspeed-video: address a protential usage of an unitialized var\n \n While this might not occur in practice, if the device is doing\n the right thing, it would be teoretically be possible to have\n both hsync_counter and vsync_counter negatives.\n \n If this ever happen, ctrl will be undefined, but the driver\n will still call:\n \n aspeed_video_update(video, VE_CTRL, 0, ctrl);\n \n Change the code to prevent this to happen.\n \n This was warned by cppcheck:\n \n [drivers/media/platform/aspeed-video.c:653]: (error) Uninitialized variable: ctrl\n \n Reviewed-by: Eddie James \u003ceajames@linux.ibm.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit cb8045072a34d113e51fcd19cec74359b4e8daa9\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Thu Aug 22 11:41:45 2019 -0300\n\n media: vicodec: make life easier for static analyzers\n \n cppcheck incorrectly produces an error here:\n [drivers/media/platform/vicodec/vicodec-core.c:1677]: (error) Pointer addition with NULL pointer.\n \n While this is actually a false positive, it doesn\u0027t hurt to\n reorder the checks to make the code simpler, handling first\n the error patch, where no color or alpha components are there.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 59364e971c1c4e2bcf45ee118507d60609337de8\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Thu Aug 22 10:24:12 2019 -0300\n\n media: remove include stdarg.h from some drivers\n \n It doesn\u0027t make any sense to have gcc\u0027s stdarg.h included\n inside the Linux Kernel.\n \n Get rid of those includes.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 50f57ef83d836d727902a6a7e70cca61b0d02655\nAuthor: Markus Elfring \u003celfring@users.sourceforge.net\u003e\nDate: Mon Aug 26 15:05:31 2019 +0200\n\n ARM: OMAP2+: Delete an unnecessary kfree() call in omap_hsmmc_pdata_init()\n \n A null pointer would be passed to a call of the function \"kfree\" directly\n after a call of the function \"kzalloc\" failed at one place.\n Remove this superfluous function call.\n \n This issue was detected by using the Coccinelle software.\n \n Signed-off-by: Markus Elfring \u003celfring@users.sourceforge.net\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit ccf4975dca233b1d6a74752d6ab35c239edc0d58\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Fri Aug 23 11:58:08 2019 +0900\n\n ARM: OMAP2+: move platform-specific asm-offset.h to arch/arm/mach-omap2\n \n \u003cgenerated/ti-pm-asm-offsets.h\u003e is only generated and included by\n arch/arm/mach-omap2/, so it does not need to reside in the globally\n visible include/generated/.\n \n I renamed it to arch/arm/mach-omap2/pm-asm-offsets.h since the prefix\n \u0027ti-\u0027 is just redundant in mach-omap2/.\n \n My main motivation of this change is to avoid the race condition for\n the parallel build (-j) when CONFIG_IKHEADERS is enabled.\n \n When it is enabled, all the headers under include/ are archived into\n kernel/kheaders_data.tar.xz and exposed in the sysfs.\n \n In the parallel build, we have no idea in which order files are built.\n \n - If ti-pm-asm-offsets.h is built before kheaders_data.tar.xz,\n the header will be included in the archive. Probably nobody will\n use it, but it is harmless except that it will increase the archive\n size needlessly.\n \n - If kheaders_data.tar.xz is built before ti-pm-asm-offsets.h,\n the header will not be included in the archive. However, in the next\n build, the archive will be re-generated to include the newly-found\n ti-pm-asm-offsets.h. This is not nice from the build system point\n of view.\n \n - If ti-pm-asm-offsets.h and kheaders_data.tar.xz are built at the\n same time, the corrupted header might be included in the archive,\n which does not look nice either.\n \n This commit fixes the race.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Tested-by: Keerthy \u003cj-keerthy@ti.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 6cb0ac0fb9786d01f140dcfcd3d09ce850dd0a64\nAuthor: Adam Ford \u003caford173@gmail.com\u003e\nDate: Mon Aug 26 08:47:09 2019 -0700\n\n ARM: dts: ARM: dts: Configure interconnect target module for am3517sgx\n \n Based on Tony Lindgren\u0027s work for omap34xx, this patch applies the same\n functionality to the AM3517.\n \n The following can be tested via sysfs with the following to ensure the SGX\n module gets enabled and disabled properly:\n \n 0x00010201\n \n Bus error\n \n Cc: Filip Matijević \u003cfilip.matijevic.pz@gmail.com\u003e\n Cc: \"H. Nikolaus Schaller\" \u003chns@goldelico.com\u003e\n Cc: Ivaylo Dimitrov \u003civo.g.dimitrov.75@gmail.com\u003e\n Cc: moaz korena \u003cmoaz@korena.xyz\u003e\n Cc: Merlijn Wajer \u003cmerlijn@wizzup.org\u003e\n Cc: Paweł Chmiel \u003cpawel.mikolaj.chmiel@gmail.com\u003e\n Cc: Philipp Rossak \u003cembed3d@gmail.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Signed-off-by: Adam Ford \u003caford173@gmail.com\u003e\n [tony@atomide.com: updated subject, dropped rstctrl info]\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 3b72fc895a2e57f70276b46f386f35d752adf555\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Mon Aug 26 08:47:08 2019 -0700\n\n ARM: dts: Configure interconnect target module for omap3 sgx\n \n Looks like omap34xx OCP registers are not readable unlike on omap36xx.\n We use SGX revision register instead of the OCP revision register for\n 34xx and do not configure any SYSCONFIG register unlike for 36xx.\n \n I\u0027ve tested that the interconnect target module enables and idles\n just fine with PM runtime control via sys:\n \n # echo on \u003e $(find /sys -name control | grep \\/5000); rwmem 0x5000fe10\n # rwmem 0x50000014 # SGX revision register on 36xx\n 0x50000014 \u003d 0x00010205\n # echo auto \u003e $(find /sys -name control | grep \\/5000)\n # rwmem 0x5000fe00\n And when idled, it will produce \"Bus error\" as expected.\n \n Cc: Adam Ford \u003caford173@gmail.com\u003e\n Cc: Filip Matijević \u003cfilip.matijevic.pz@gmail.com\u003e\n Cc: \"H. Nikolaus Schaller\" \u003chns@goldelico.com\u003e\n Cc: Ivaylo Dimitrov \u003civo.g.dimitrov.75@gmail.com\u003e\n Cc: moaz korena \u003cmoaz@korena.xyz\u003e\n Cc: Merlijn Wajer \u003cmerlijn@wizzup.org\u003e\n Cc: Paweł Chmiel \u003cpawel.mikolaj.chmiel@gmail.com\u003e\n Cc: Philipp Rossak \u003cembed3d@gmail.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Tested-by: Adam Ford \u003caford173@gmail.com\u003e #logicpd-torpedo-37xx-devkit\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 394534cb07d8f89b6e621e8a1e6da23571725aef\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Mon Aug 26 08:47:08 2019 -0700\n\n ARM: dts: Configure sgx for omap5\n \n I\u0027ve tested that the interconnect target module enables and idles\n just fine when probed with ti-sysc with PM runtime control via sys:\n \n # echo on \u003e $(find /sys -name control | grep \\/5600)\n # rwmem 0x5600fe00 # OCP Revision\n 0x5600fe00 \u003d 0x40000000\n # echo auto \u003e $(find /sys -name control | grep \\/5600)\n # rwmem 0x5600fe10\n # rwmem 0x56000024\n \n Cc: Adam Ford \u003caford173@gmail.com\u003e\n Cc: Filip Matijević \u003cfilip.matijevic.pz@gmail.com\u003e\n Cc: \"H. Nikolaus Schaller\" \u003chns@goldelico.com\u003e\n Cc: Ivaylo Dimitrov \u003civo.g.dimitrov.75@gmail.com\u003e\n Cc: moaz korena \u003cmoaz@korena.xyz\u003e\n Cc: Merlijn Wajer \u003cmerlijn@wizzup.org\u003e\n Cc: Paweł Chmiel \u003cpawel.mikolaj.chmiel@gmail.com\u003e\n Cc: Philipp Rossak \u003cembed3d@gmail.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit fd5683749472bb3370487009c91c49182eb99694\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Mon Aug 26 08:47:07 2019 -0700\n\n clk: ti: add clkctrl data omap5 sgx\n \n Looks like we have sgx clock missing currently so let\u0027s add it.\n \n Cc: Adam Ford \u003caford173@gmail.com\u003e\n Cc: Filip Matijević \u003cfilip.matijevic.pz@gmail.com\u003e\n Cc: \"H. Nikolaus Schaller\" \u003chns@goldelico.com\u003e\n Cc: Ivaylo Dimitrov \u003civo.g.dimitrov.75@gmail.com\u003e\n Cc: moaz korena \u003cmoaz@korena.xyz\u003e\n Cc: Merlijn Wajer \u003cmerlijn@wizzup.org\u003e\n Cc: Michael Turquette \u003cmturquette@baylibre.com\u003e\n Cc: Paweł Chmiel \u003cpawel.mikolaj.chmiel@gmail.com\u003e\n Cc: Philipp Rossak \u003cembed3d@gmail.com\u003e\n Cc: Stephen Boyd \u003csboyd@kernel.org\u003e\n Cc: Tero Kristo \u003ct-kristo@ti.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: linux-clk@vger.kernel.org\n Acked-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 1b95c5a48f5ea3a23618b5894f44124bfa98171a\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Mon Aug 26 08:47:07 2019 -0700\n\n ARM: OMAP2+: Drop legacy platform data for omap4 gpu\n \n I\u0027ve tested that the interconnect target module enables and idles\n just fine when probed with ti-sysc with PM runtime control via sys:\n \n # echo on \u003e $(find /sys -name control | grep \\/5601)\n # rwmem 0x56000024\n 0x56000024 \u003d 0x00010200 # SGX540 CORE_REVISION\n # echo auto \u003e $(find /sys -name control | grep \\/5601)\n # rwmem 0x56000024\n And when idled, it will produce \"Bus error\" as expected.\n \n Cc: Adam Ford \u003caford173@gmail.com\u003e\n Cc: Filip Matijević \u003cfilip.matijevic.pz@gmail.com\u003e\n Cc: \"H. Nikolaus Schaller\" \u003chns@goldelico.com\u003e\n Cc: Ivaylo Dimitrov \u003civo.g.dimitrov.75@gmail.com\u003e\n Cc: moaz korena \u003cmoaz@korena.xyz\u003e\n Cc: Merlijn Wajer \u003cmerlijn@wizzup.org\u003e\n Cc: Paweł Chmiel \u003cpawel.mikolaj.chmiel@gmail.com\u003e\n Cc: Philipp Rossak \u003cembed3d@gmail.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit af055ade3490958ddd24608c93f3b796b04aebde\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Mon Aug 26 08:41:17 2019 -0700\n\n ARM: OMAP2+: Drop legacy platform data for omap4 d2d\n \n We can now probe devices with ti-sysc interconnect driver and dts\n data. Let\u0027s drop the related platform data and custom ti,hwmods\n dts property.\n \n As we\u0027re just dropping data, and the early platform data init\n is based on the custom ti,hwmods property, we want to drop both\n the platform data and ti,hwmods property in a single patch.\n \n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit d73d7667bb4ff282c3d00a21c365c40b6a7efa91\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Mon Aug 26 08:41:16 2019 -0700\n\n ARM: dts: Configure d2d dts data for omap4\n \n We can now probe devices with ti-sysc interconnect driver and dts\n data. Let\u0027s configure the related dts data based on what we have\n defined in the legacy platform data.\n \n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 84087b3d633d04716cf1a83eb537f732a540a5b5\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Mon Aug 26 08:41:16 2019 -0700\n\n ARM: OMAP2+: Drop legacy watchdog platform data for omap4\n \n We can now probe devices with ti-sysc interconnect driver and dts\n data. Let\u0027s drop the related platform data and custom ti,hwmods\n dts property.\n \n As we\u0027re just dropping data, and the early platform data init\n is based on the custom ti,hwmods property, we want to drop both\n the platform data and ti,hwmods property in a single patch.\n \n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit e47f534dae5fc2e8167d8a2273f6ad1d39448526\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Mon Aug 26 08:41:16 2019 -0700\n\n ARM: dts: Drop custom hwmod property for omap4 i2c\n \n We can now probe devices with ti-sysc interconnect driver and dts\n data. Let\u0027s drop the custom ti,hwmods dts property. We have already\n dropped the platform data earlier and have been allocating it\n dynamically.\n \n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 3f95f51b1602f89cf9c0269bdfb8b1021de3e3df\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Mon Aug 26 08:41:15 2019 -0700\n\n ARM: OMAP2+: Drop legacy platform data for cpsw on dra7\n \n We can now probe cpsw with ti-sysc interconnect driver and dts\n data. Let\u0027s drop the related platform data and custom ti,hwmods\n dts property for am3 and am4.\n \n As we\u0027re just dropping data, and the early platform data init\n is based on the custom ti,hwmods property, we want to drop both\n the platform data and ti,hwmods property in a single patch.\n \n Cc: Grygorii Strashko \u003cgrygorii.strashko@ti.com\u003e\n Cc: Ivan Khoronzhuk \u003civan.khoronzhuk@linaro.org\u003e\n Cc: Keerthy \u003cj-keerthy@ti.com\u003e\n Reviewed-by: Grygorii Strashko \u003cgrygorii.strashko@ti.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 7abe746191a095e8e8f51615af6a4690bd9b192a\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Mon Aug 26 08:41:15 2019 -0700\n\n ARM: OMAP2+: Drop legacy platform data for cpsw on am3 and am4\n \n We can now probe cpsw with ti-sysc interconnect driver and dts\n data. Let\u0027s drop the related platform data and custom ti,hwmods\n dts property for am3 and am4.\n \n As we\u0027re just dropping data, and the early platform data init\n is based on the custom ti,hwmods property, we want to drop both\n the platform data and ti,hwmods property in a single patch.\n \n Cc: Grygorii Strashko \u003cgrygorii.strashko@ti.com\u003e\n Cc: Ivan Khoronzhuk \u003civan.khoronzhuk@linaro.org\u003e\n Cc: Keerthy \u003cj-keerthy@ti.com\u003e\n Reviewed-by: Grygorii Strashko \u003cgrygorii.strashko@ti.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 1faa415c9c6e41a5350d3067307e7985b546ac3a\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Mon Aug 26 08:41:14 2019 -0700\n\n ARM: dts: Add fck for cpsw mdio for omap variants\n \n In preparation for dropping legacy platform data and custom ti,hwmods\n property, we need to make functional clock available for mdio for the\n SoCs so the the mdio driver can find it.\n \n The mdio hardware currently relies on a mdio_hwmod to manage the clock\n for omap variants. This is wrong though as there are no separate\n sysconfig registers for mdio. All the cpsw related components are just\n children of the gmac module.\n \n Note that since mdio is a child of cpsw, just doing pm_runtime_get()\n in the mdio driver enables the clock. However, since mdio is also used\n by davinci that does not implement runtime PM, let\u0027s just add the fck\n for now.\n \n Also note that am437x mdio already has a clock, let\u0027s update it to\n not use the legacy clock naming to unify things further.\n \n Cc: Grygorii Strashko \u003cgrygorii.strashko@ti.com\u003e\n Cc: Ivan Khoronzhuk \u003civan.khoronzhuk@linaro.org\u003e\n Cc: Keerthy \u003cj-keerthy@ti.com\u003e\n Reviewed-by: Grygorii Strashko \u003cgrygorii.strashko@ti.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 369bd269f7d534d4da3e1e8759f875d0eef6a06d\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Wed Aug 21 15:05:55 2019 +0000\n\n btrfs: fix allocation of free space cache v1 bitmap pages\n \n Various notifications of type \"BUG kmalloc-4096 () : Redzone\n overwritten\" have been observed recently in various parts of the kernel.\n After some time, it has been made a relation with the use of BTRFS\n filesystem and with SLUB_DEBUG turned on.\n \n [ 22.809700] BUG kmalloc-4096 (Tainted: G W ): Redzone overwritten\n \n [ 22.810286] INFO: 0xbe1a5921-0xfbfc06cd. First byte 0x0 instead of 0xcc\n [ 22.810866] INFO: Allocated in __load_free_space_cache+0x588/0x780 [btrfs] age\u003d22 cpu\u003d0 pid\u003d224\n [ 22.811193] __slab_alloc.constprop.26+0x44/0x70\n [ 22.811345] kmem_cache_alloc_trace+0xf0/0x2ec\n [ 22.811588] __load_free_space_cache+0x588/0x780 [btrfs]\n [ 22.811848] load_free_space_cache+0xf4/0x1b0 [btrfs]\n [ 22.812090] cache_block_group+0x1d0/0x3d0 [btrfs]\n [ 22.812321] find_free_extent+0x680/0x12a4 [btrfs]\n [ 22.812549] btrfs_reserve_extent+0xec/0x220 [btrfs]\n [ 22.812785] btrfs_alloc_tree_block+0x178/0x5f4 [btrfs]\n [ 22.813032] __btrfs_cow_block+0x150/0x5d4 [btrfs]\n [ 22.813262] btrfs_cow_block+0x194/0x298 [btrfs]\n [ 22.813484] commit_cowonly_roots+0x44/0x294 [btrfs]\n [ 22.813718] btrfs_commit_transaction+0x63c/0xc0c [btrfs]\n [ 22.813973] close_ctree+0xf8/0x2a4 [btrfs]\n [ 22.814107] generic_shutdown_super+0x80/0x110\n [ 22.814250] kill_anon_super+0x18/0x30\n [ 22.814437] btrfs_kill_super+0x18/0x90 [btrfs]\n [ 22.814590] INFO: Freed in proc_cgroup_show+0xc0/0x248 age\u003d41 cpu\u003d0 pid\u003d83\n [ 22.814841] proc_cgroup_show+0xc0/0x248\n [ 22.814967] proc_single_show+0x54/0x98\n [ 22.815086] seq_read+0x278/0x45c\n [ 22.815190] __vfs_read+0x28/0x17c\n [ 22.815289] vfs_read+0xa8/0x14c\n [ 22.815381] ksys_read+0x50/0x94\n [ 22.815475] ret_from_syscall+0x0/0x38\n \n Commit 69d2480456d1 (\"btrfs: use copy_page for copying pages instead of\n memcpy\") changed the way bitmap blocks are copied. But allthough bitmaps\n have the size of a page, they were allocated with kzalloc().\n \n Most of the time, kzalloc() allocates aligned blocks of memory, so\n copy_page() can be used. But when some debug options like SLAB_DEBUG are\n activated, kzalloc() may return unaligned pointer.\n \n On powerpc, memcpy(), copy_page() and other copying functions use\n \u0027dcbz\u0027 instruction which provides an entire zeroed cacheline to avoid\n memory read when the intention is to overwrite a full line. Functions\n like memcpy() are writen to care about partial cachelines at the start\n and end of the destination, but copy_page() assumes it gets pages. As\n pages are naturally cache aligned, copy_page() doesn\u0027t care about\n partial lines. This means that when copy_page() is called with a\n misaligned pointer, a few leading bytes are zeroed.\n \n To fix it, allocate bitmaps through kmem_cache instead of using kzalloc()\n The cache pool is created with PAGE_SIZE alignment constraint.\n \n Reported-by: Erhard F. \u003cerhard_f@mailbox.org\u003e\n Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id\u003d204371\n Fixes: 69d2480456d1 (\"btrfs: use copy_page for copying pages instead of memcpy\")\n Cc: stable@vger.kernel.org # 4.19+\n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n [ rename to btrfs_free_space_bitmap ]\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 7edd00f71f4b91ca31dbfa08926495fe5e77aab4\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Wed Aug 14 05:18:16 2019 -0700\n\n bus: ti-sysc: Detect d2d when debug is enabled\n \n We want to see what modules probe when debug is enabled.\n \n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit d7f563db7794a6a271b6e9dd6e65a437d6a1d933\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Wed Aug 14 05:18:16 2019 -0700\n\n bus: ti-sysc: Add module enable quirk for SGX on omap36xx\n \n Add module enable quirk for SGX needed on omap36xx.\n \n Cc: Adam Ford \u003caford173@gmail.com\u003e\n Cc: Filip Matijević \u003cfilip.matijevic.pz@gmail.com\u003e\n Cc: \"H. Nikolaus Schaller\" \u003chns@goldelico.com\u003e\n Cc: Ivaylo Dimitrov \u003civo.g.dimitrov.75@gmail.com\u003e\n Cc: moaz korena \u003cmoaz@korena.xyz\u003e\n Cc: Merlijn Wajer \u003cmerlijn@wizzup.org\u003e\n Cc: Paweł Chmiel \u003cpawel.mikolaj.chmiel@gmail.com\u003e\n Cc: Philipp Rossak \u003cembed3d@gmail.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit c6e78d701966b94370fc114209c87c9345993af0\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 15 11:16:47 2019 +0530\n\n bus: ti-sysc: Change return types of functions\n \n Change return type of functions sysc_check_one_child() and\n sysc_check_children() from int to void as neither ever returns an error.\n Modify call sites of both functions accordingly.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Acked-by: Roger Quadros \u003crogerq@ti.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit acd674af95d3f627062007429b9c195c6b32361d\nAuthor: Lyude Paul \u003clyude@redhat.com\u003e\nDate: Fri Aug 23 16:52:51 2019 -0400\n\n drm/i915: Call dma_set_max_seg_size() in i915_driver_hw_probe()\n \n Currently, we don\u0027t call dma_set_max_seg_size() for i915 because we\n intentionally do not limit the segment length that the device supports.\n However, this results in a warning being emitted if we try to map\n anything larger than SZ_64K on a kernel with CONFIG_DMA_API_DEBUG_SG\n enabled:\n \n [ 7.751926] DMA-API: i915 0000:00:02.0: mapping sg segment longer\n than device claims to support [len\u003d98304] [max\u003d65536]\n [ 7.751934] WARNING: CPU: 5 PID: 474 at kernel/dma/debug.c:1220\n debug_dma_map_sg+0x20f/0x340\n \n This was originally brought up on\n https://bugs.freedesktop.org/show_bug.cgi?id\u003d108517 , and the consensus\n there was it wasn\u0027t really useful to set a limit (and that dma-debug\n isn\u0027t really all that useful for i915 in the first place). Unfortunately\n though, CONFIG_DMA_API_DEBUG_SG is enabled in the debug configs for\n various distro kernels. Since a WARN_ON() will disable automatic problem\n reporting (and cause any CI with said option enabled to start\n complaining), we really should just fix the problem.\n \n Note that as me and Chris Wilson discussed, the other solution for this\n would be to make DMA-API not make such assumptions when a driver hasn\u0027t\n explicitly set a maximum segment size. But, taking a look at the commit\n which originally introduced this behavior, commit 78c47830a5cb\n (\"dma-debug: check scatterlist segments\"), there is an explicit mention\n of this assumption and how it applies to devices with no segment size:\n \n Conversely, devices which are less limited than the rather\n conservative defaults, or indeed have no limitations at all\n (e.g. GPUs with their own internal MMU), should be encouraged to\n set appropriate dma_parms, as they may get more efficient DMA\n mapping performance out of it.\n \n So unless there\u0027s any concerns (I\u0027m open to discussion!), let\u0027s just\n follow suite and call dma_set_max_seg_size() with UINT_MAX as our limit\n to silence any warnings.\n \n Changes since v3:\n * Drop patch for enabling CONFIG_DMA_API_DEBUG_SG in CI. It looks like\n just turning it on causes the kernel to spit out bogus WARN_ONs()\n during some igt tests which would otherwise require teaching igt to\n disable the various DMA-API debugging options causing this. This is\n too much work to be worth it, since DMA-API debugging is useless for\n us. So, we\u0027ll just settle with this single patch to squelch WARN_ONs()\n during driver load for users that have CONFIG_DMA_API_DEBUG_SG turned\n on for some reason.\n * Move dma_set_max_seg_size() call into i915_driver_hw_probe() - Chris\n Wilson\n \n Signed-off-by: Lyude Paul \u003clyude@redhat.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: \u003cstable@vger.kernel.org\u003e # v4.18+\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823205251.14298-1-lyude@redhat.com\n\ncommit 1d194d2039a9ddb3fb8dda2448787d4c5a53f865\nAuthor: Qu Wenruo \u003cwqu@suse.com\u003e\nDate: Thu Aug 22 10:14:15 2019 +0800\n\n btrfs: Detect unbalanced tree with empty leaf before crashing btree operations\n \n [BUG]\n With crafted image, btrfs will panic at btree operations:\n \n kernel BUG at fs/btrfs/ctree.c:3894!\n invalid opcode: 0000 [#1] SMP PTI\n CPU: 0 PID: 1138 Comm: btrfs-transacti Not tainted 5.0.0-rc8+ #9\n RIP: 0010:__push_leaf_left+0x6b6/0x6e0\n RSP: 0018:ffffc0bd4128b990 EFLAGS: 00010246\n RAX: 0000000000000000 RBX: ffffa0a4ab8f0e38 RCX: 0000000000000000\n RDX: ffffa0a280000000 RSI: 0000000000000000 RDI: ffffa0a4b3814000\n RBP: ffffc0bd4128ba38 R08: 0000000000001000 R09: ffffc0bd4128b948\n R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000240\n R13: ffffa0a4b556fb60 R14: ffffa0a4ab8f0af0 R15: ffffa0a4ab8f0af0\n FS: 0000000000000000(0000) GS:ffffa0a4b7a00000(0000) knlGS:0000000000000000\n CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n CR2: 00007f2461c80020 CR3: 000000022b32a006 CR4: 00000000000206f0\n Call Trace:\n ? _cond_resched+0x1a/0x50\n push_leaf_left+0x179/0x190\n btrfs_del_items+0x316/0x470\n btrfs_del_csums+0x215/0x3a0\n __btrfs_free_extent.isra.72+0x5a7/0xbe0\n __btrfs_run_delayed_refs+0x539/0x1120\n btrfs_run_delayed_refs+0xdb/0x1b0\n btrfs_commit_transaction+0x52/0x950\n ? start_transaction+0x94/0x450\n transaction_kthread+0x163/0x190\n kthread+0x105/0x140\n ? btrfs_cleanup_transaction+0x560/0x560\n ? kthread_destroy_worker+0x50/0x50\n ret_from_fork+0x35/0x40\n Modules linked in:\n ---[ end trace c2425e6e89b5558f ]---\n \n [CAUSE]\n The offending csum tree looks like this:\n \n checksum tree key (CSUM_TREE ROOT_ITEM 0)\n node 29741056 level 1 items 14 free 107 generation 19 owner CSUM_TREE\n ...\n key (EXTENT_CSUM EXTENT_CSUM 85975040) block 29630464 gen 17\n key (EXTENT_CSUM EXTENT_CSUM 89911296) block 29642752 gen 17 \u003c\u003c\u003c\n key (EXTENT_CSUM EXTENT_CSUM 92274688) block 29646848 gen 17\n ...\n \n leaf 29630464 items 6 free space 1 generation 17 owner CSUM_TREE\n item 0 key (EXTENT_CSUM EXTENT_CSUM 85975040) itemoff 3987 itemsize 8\n range start 85975040 end 85983232 length 8192\n ...\n leaf 29642752 items 0 free space 3995 generation 17 owner 0\n ^ empty leaf invalid owner ^\n \n leaf 29646848 items 1 free space 602 generation 17 owner CSUM_TREE\n item 0 key (EXTENT_CSUM EXTENT_CSUM 92274688) itemoff 627 itemsize 3368\n range start 92274688 end 95723520 length 3448832\n \n So we have a corrupted csum tree where one tree leaf is completely\n empty, causing unbalanced btree, thus leading to unexpected btree\n balance error.\n \n [FIX]\n For this particular case, we handle it in two directions to catch it:\n - Check if the tree block is empty through btrfs_verify_level_key()\n So that invalid tree blocks won\u0027t be read out through\n btrfs_search_slot() and its variants.\n \n - Check 0 tree owner in tree checker\n NO tree is using 0 as its tree owner, detect it and reject at tree\n block read time.\n \n Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id\u003d202821\n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit bcd676542cdd44b3a91e0b695cc1d7a1d35679cd\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Aug 5 19:18:25 2019 -0500\n\n dm raid1: use struct_size() with kzalloc()\n \n One of the more common cases of allocation size calculations is finding\n the size of a structure that has a zero-sized array at the end, along\n with memory for some number of elements for that array. For example:\n \n struct mirror_set {\n ...\n struct mirror mirror[0];\n };\n \n size \u003d sizeof(struct mirror_set) + count * sizeof(struct mirror);\n instance \u003d kzalloc(size, GFP_KERNEL)\n \n Instead of leaving these open-coded and prone to type mistakes, we can\n now use the new struct_size() helper:\n \n instance \u003d kzalloc(struct_size(instance, mirror, count), GFP_KERNEL)\n \n Notice that, in this case, variable len is not necessary, hence it\n is removed.\n \n This code was detected with the help of Coccinelle.\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Mike Snitzer \u003csnitzer@redhat.com\u003e\n\ncommit b20ff35ef4a248ad32999370570fe45f5d8e3b87\nAuthor: Nikolay Borisov \u003cnborisov@suse.com\u003e\nDate: Mon Aug 26 17:34:24 2019 +0300\n\n btrfs: Deprecate BTRFS_SUBVOL_CREATE_ASYNC flag\n \n Support for asynchronous snapshot creation was originally added in\n 72fd032e9424 (\"Btrfs: add SNAP_CREATE_ASYNC ioctl\") to cater for\n ceph\u0027s backend needs. However, since Ceph has deprecated support for\n btrfs there is no longer need for that support in btrfs. Additionally,\n this was never supported by btrfs-progs, the official userspace tools.\n \n Reviewed-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Reviewed-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Signed-off-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 5229b4896e8f32bda4bfe29ff91e594ae7aa8a75\nAuthor: Huaisheng Ye \u003cyehs1@lenovo.com\u003e\nDate: Sun Aug 25 15:24:33 2019 +0800\n\n dm writecache: optimize performance by sorting the blocks for writeback_all\n \n During the process of writeback, the blocks, which have been placed in wbl.list\n for writeback soon, are partially ordered for the contiguous ones.\n \n When writeback_all has been set, for most cases, also by default, there will be\n a lot of blocks in pmem need to writeback at the same time.\n For this case, we could optimize the performance by sorting all blocks in\n wbl.list. writecache_writeback doesn\u0027t need to get blocks from the tail of\n wc-\u003elru, whereas from the first rb_node from the rb_tree.\n \n The benefit is that, writecache_writeback doesn\u0027t need to have any cost to sort\n the blocks, because of all blocks are incremental originally in rb_tree.\n There will be a writecache_flush when writeback_all begins to work, that will\n eliminate duplicate blocks in cache by committed/uncommitted.\n \n Testing platform: Thinksystem SR630 with persistent memory.\n The cache comes from pmem, which has 1006MB size. The origin device is HDD, 2GB\n of which for using.\n \n Testing steps:\n 1) dmsetup create mycache --table \u00270 4194304 writecache p /dev/sdb1 /dev/pmem4 4096 0\u0027\n 2) fio -filename\u003d/dev/mapper/mycache -direct\u003d1 -iodepth\u003d20 -rw\u003drandwrite\n -ioengine\u003dlibaio -bs\u003d4k -loops\u003d1 -size\u003d2g -group_reporting -name\u003dmytest1\n 3) time dmsetup message /dev/mapper/mycache 0 flush\n \n Here is the results below,\n With the patch:\n # fio -filename\u003d/dev/mapper/mycache -direct\u003d1 -iodepth\u003d20 -rw\u003drandwrite\n -ioengine\u003dlibaio -bs\u003d4k -loops\u003d1 -size\u003d2g -group_reporting -name\u003dmytest1\n iops : min\u003d 1582, max\u003d199470, avg\u003d5305.94, stdev\u003d21273.44, samples\u003d197\n # time dmsetup message /dev/mapper/mycache 0 flush\n real 0m44.020s\n user 0m0.002s\n sys 0m0.003s\n \n Without the patch:\n # fio -filename\u003d/dev/mapper/mycache -direct\u003d1 -iodepth\u003d20 -rw\u003drandwrite\n -ioengine\u003dlibaio -bs\u003d4k -loops\u003d1 -size\u003d2g -group_reporting -name\u003dmytest1\n iops : min\u003d 1202, max\u003d197650, avg\u003d4968.67, stdev\u003d20480.17, samples\u003d211\n # time dmsetup message /dev/mapper/mycache 0 flush\n real 1m39.221s\n user 0m0.001s\n sys 0m0.003s\n \n I also have checked the data accuracy with this patch by making EXT4 filesystem\n on mycache, then mount it for checking md5 of files on that.\n The test result is positive, with this patch it could save more than half of time\n when writeback_all.\n \n Signed-off-by: Huaisheng Ye \u003cyehs1@lenovo.com\u003e\n Signed-off-by: Mike Snitzer \u003csnitzer@redhat.com\u003e\n\ncommit 3b4acbb92dbda4829e021e5c6d5410658849fa1c\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Apr 8 11:27:48 2019 -0500\n\n perf script: Fix memory leaks in list_scripts()\n \n In case memory resources for *buf* and *paths* were allocated, jump to\n *out* and release them before return.\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Addresses-Coverity-ID: 1444328 (\"Resource leak\")\n Fixes: 6f3da20e151f (\"perf report: Support builtin perf script in scripts menu\")\n Link: http://lkml.kernel.org/r/20190408162748.GA21008@embeddedor\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 3dab6ac080dcd7f71cb9ceb84ad7dafecd6f7c07\nAuthor: Andi Kleen \u003cak@linux.intel.com\u003e\nDate: Fri Aug 23 14:03:38 2019 -0700\n\n perf report: Fix --ns time sort key output\n \n If the user specified --ns, the column to print the sort time stamp\n wasn\u0027t wide enough to actually print the full nanoseconds.\n \n Widen the time key column width when --ns is specified.\n \n Before:\n \n % perf record -a sleep 1\n % perf report --sort time,overhead,symbol --stdio --ns\n ...\n 2.39% 187851.10000 [k] smp_call_function_single - -\n 1.53% 187851.10000 [k] intel_idle - -\n 0.59% 187851.10000 [.] __wcscmp_ifunc - -\n 0.33% 187851.10000 [.] 0000000000000000 - -\n 0.28% 187851.10000 [k] cpuidle_enter_state - -\n \n After:\n \n % perf report --sort time,overhead,symbol --stdio --ns\n ...\n 2.39% 187851.100000000 [k] smp_call_function_single - -\n 1.53% 187851.100000000 [k] intel_idle - -\n 0.59% 187851.100000000 [.] __wcscmp_ifunc - -\n 0.33% 187851.100000000 [.] 0000000000000000 - -\n 0.28% 187851.100000000 [k] cpuidle_enter_state - -\n \n Signed-off-by: Andi Kleen \u003cak@linux.intel.com\u003e\n Tested-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Link: http://lkml.kernel.org/r/20190823210338.12360-2-andi@firstfloor.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 092804ae092fc6097348f5c09b62cde040717aa1\nAuthor: Andi Kleen \u003cak@linux.intel.com\u003e\nDate: Fri Aug 23 14:03:37 2019 -0700\n\n perf report: Use timestamp__scnprintf_nsec() for time sort key\n \n Use timestamp__scnprintf_nsec() to print nanoseconds for the time sort\n key, instead of open coding.\n \n Signed-off-by: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Link: http://lkml.kernel.org/r/20190823210338.12360-1-andi@firstfloor.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit b4de344b25b9fbd4db7b84da43808a6b7daac887\nAuthor: Souptick Joarder \u003cjrdr.linux@gmail.com\u003e\nDate: Sat Aug 24 21:45:19 2019 +0530\n\n perf tools: Remove duplicate headers\n \n Removed headers which are included twice.\n \n Signed-off-by: Souptick Joarder \u003cjrdr.linux@gmail.com\u003e\n Reviewed-by: Mukesh Ojha \u003cmojha@codeaurora.org\u003e\n Cc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/1566663319-4283-1-git-send-email-jrdr.linux@gmail.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit e051c2f69850e571cb408b555aaffb00c07dbb05\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 23 15:54:02 2019 -0300\n\n perf augmented_raw_syscalls: Reduce perf_event_output() boilerplate\n \n Add a augmented__output() helper to reduce the boilerplate of sending\n the augmented tracepoint to the PERF_EVENT_ARRAY BPF map associated with\n the bpf-output event used to communicate with the userspace perf trace\n tool.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-ln99gt0j4fv0kw0778h6vphm@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit c265784de7adff9e07934b712204556fb0b3ba43\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 23 15:45:54 2019 -0300\n\n perf augmented_raw_syscalls: Introduce helper to get the scratch space\n \n We need more than the BPF stack can give us to format the\n raw_syscalls:sys_enter augmented tracepoint, so we use a PERCPU_ARRAY\n map for that, use a helper to shorten the sequence to access that area.\n \n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 01128065ca5112123e6992dc0522484349c6ced7\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 23 15:38:08 2019 -0300\n\n perf augmented_raw_syscalls: Postpone tmp map lookup to after pid_filter\n \n No sense in doing that lookup before figuring out if it will be used,\n i.e. if the pid is being filtered that tmp space lookup will be useless.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-o74yggieorucfg4j74tb6rta@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 2ad926db78777148b07fced1e4bc88e20ad00268\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 23 10:37:47 2019 -0300\n\n perf augmented_raw_syscalls: Rename augmented_filename to augmented_arg\n \n Because it is not used only for strings, we already use it for sockaddr\n structs and will use it for all other types.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-w9nkt3tvmyn5i4qnwng3ap1k@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit b92675f4a9c02dd78052645597dac9e270679ddf\nAuthor: Benjamin Peterson \u003cbenjamin@python.org\u003e\nDate: Thu Aug 22 20:36:25 2019 -0700\n\n perf trace beauty ioctl: Fix off-by-one error in cmd-\u003estring table\n \n While tracing a program that calls isatty(3), I noticed that strace\n reported TCGETS for the request argument of the underlying ioctl(2)\n syscall while perf trace reported TCSETS. strace is corrrect. The bug in\n perf was due to the tty ioctl beauty table starting at 0x5400 rather\n than 0x5401.\n \n Committer testing:\n \n Using augmented_raw_syscalls.o and settings to make \u0027perf trace\u0027\n use strace formatting, i.e. with this in ~/.perfconfig\n \n # cat ~/.perfconfig\n [trace]\n add_events \u003d /home/acme/git/linux/tools/perf/examples/bpf/augmented_raw_syscalls.c\n show_zeros \u003d yes\n show_duration \u003d no\n no_inherit \u003d yes\n show_timestamp \u003d no\n show_arg_names \u003d no\n args_alignment \u003d 40\n show_prefix \u003d yes\n \n # strace -e ioctl stty \u003e /dev/null\n ioctl(0, TCGETS, {B38400 opost isig icanon echo ...}) \u003d 0\n ioctl(1, TIOCGWINSZ, 0x7fff8a9b0860) \u003d -1 ENOTTY (Inappropriate ioctl for device)\n ioctl(1, TCGETS, 0x7fff8a9b0540) \u003d -1 ENOTTY (Inappropriate ioctl for device)\n +++ exited with 0 +++\n #\n \n Before:\n \n # perf trace -e ioctl stty \u003e /dev/null\n ioctl(0, TCSETS, 0x7fff2cf79f20) \u003d 0\n ioctl(1, TIOCSWINSZ, 0x7fff2cf79f40) \u003d -1 ENOTTY (Inappropriate ioctl for device)\n ioctl(1, TCSETS, 0x7fff2cf79c20) \u003d -1 ENOTTY (Inappropriate ioctl for device)\n #\n \n After:\n \n # perf trace -e ioctl stty \u003e /dev/null\n ioctl(0, TCGETS, 0x7ffed0763920) \u003d 0\n ioctl(1, TIOCGWINSZ, 0x7ffed0763940) \u003d -1 ENOTTY (Inappropriate ioctl for device)\n ioctl(1, TCGETS, 0x7ffed0763620) \u003d -1 ENOTTY (Inappropriate ioctl for device)\n #\n \n Signed-off-by: Benjamin Peterson \u003cbenjamin@python.org\u003e\n Tested-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Fixes: 1cc47f2d46206d67285aea0ca7e8450af571da13 (\"perf trace beauty ioctl: Improve \u0027cmd\u0027 beautifier\")\n Link: http://lkml.kernel.org/r/20190823033625.18814-1-benjamin@python.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit d93fc7ac88c7347d1c31ca158d1466d1ce2097e1\nAuthor: James Clark \u003cJames.Clark@arm.com\u003e\nDate: Thu Aug 22 13:55:15 2019 +0000\n\n perf tests: Fixes hang in zstd compression test by changing the source of random data\n \n Running \u0027perf test\u0027 with zstd compression linked will hang at the test\n \u0027Zstd perf.data compression/decompression\u0027 because /dev/random blocks\n reads until there is enough entropy. This means that the test will\n appear to never complete unless the mouse is continually moved while\n running it.\n \n Signed-off-by: James Clark \u003cjames.clark@arm.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Jeremy Linton \u003cjeremy.linton@arm.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/3d8cc701-df4e-f949-1715-5118b530e990@arm.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 185bcb92c80eae2d85ec834ea76a144fd163e2af\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 22 17:11:39 2019 -0300\n\n perf sort: Remove needless headers from sort.h, provide fwd struct decls\n \n Reducing the includes hell a bit more, speeding up the build and\n avoiding needless rebuilds when just one of those files gets updated.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-u63el2vqsovsmnhebx1rcixo@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 97b9d866a66cf9884cea623cde3300073815873d\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 22 17:10:08 2019 -0300\n\n perf srcline: Add missing srcline.h header to files needing its defs\n \n When srcline was introduced it wrongly added the include to util/sort.h,\n even with that header not needing the definitions it provides, fix it by\n adding it to the places that need it as a pre patch to remove srcline.h\n from sort.h.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-shuebppedtye8hrgxk15qe3x@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 125009026bfc9ec929975d35344bf69d2c636e95\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 22 16:58:29 2019 -0300\n\n perf cacheline: Move cacheline related routines to separate files\n \n To disentangle util/sort.h a bit more.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-6kbf2cauas06rbqp15pyter5@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit aeb00b1aeab6dadd72c24f93bea51a46e109c2ba\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 22 15:40:29 2019 -0300\n\n perf record: Move record_opts and other record decls out of perf.h\n \n And into a separate util/record.h, to better isolate things and make\n sure that those who use record_opts and the other moved declarations\n are explicitly including the necessary header.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-31q8mei1qkh74qvkl9nwidfq@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 62421b3880c71da7ecbf9c4072dc64ef3e65ad0a\nAuthor: Huaisheng Ye \u003cyehs1@lenovo.com\u003e\nDate: Sun Aug 25 15:24:32 2019 +0800\n\n dm writecache: add unlikely for getting two block with same LBA\n \n In function writecache_writeback, entries g and f has same original\n sector only happens at entry f has been committed, but entry g has\n NOT yet.\n \n The probability of this happening is very low in the following\n 256 blocks at most of entry e.\n \n Signed-off-by: Huaisheng Ye \u003cyehs1@lenovo.com\u003e\n Acked-by: Mikulas Patocka \u003cmpatocka@redhat.com\u003e\n Signed-off-by: Mike Snitzer \u003csnitzer@redhat.com\u003e\n\ncommit 58912dbce6becd31c09b02a1a26b76a5ee23763b\nAuthor: Huaisheng Ye \u003cyehs1@lenovo.com\u003e\nDate: Sun Aug 25 15:24:31 2019 +0800\n\n dm writecache: remove unused member pointer in writeback_struct\n \n The stucture member pointer page in writeback_struct never has been\n used actually. Remove it.\n \n Signed-off-by: Huaisheng Ye \u003cyehs1@lenovo.com\u003e\n Acked-by: Mikulas Patocka \u003cmpatocka@redhat.com\u003e\n Signed-off-by: Mike Snitzer \u003csnitzer@redhat.com\u003e\n\ncommit 248d327ed7b6d3ebda2ac7b38482a306108bcd3e\nAuthor: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\nDate: Mon Aug 26 09:13:12 2019 +0000\n\n x86/ftrace: Remove mcount() declaration\n \n Commit 562e14f72292 (\"ftrace/x86: Remove mcount support\") removed the\n support for mcount, but forgot to remove the mcount() declaration.\n \n Clean it up.\n \n Signed-off-by: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r20190826170150.10f101ba@xhacker.debian\n\ncommit 0c8e9c2d668278652af028c3cc068c65f66342f4\nAuthor: Mikulas Patocka \u003cmpatocka@redhat.com\u003e\nDate: Mon Aug 26 02:41:17 2019 -0400\n\n dm zoned: fix invalid memory access\n \n Commit 75d66ffb48efb30f2dd42f041ba8b39c5b2bd115 (\"dm zoned: properly\n handle backing device failure\") triggers a coverity warning:\n \n *** CID 1452808: Memory - illegal accesses (USE_AFTER_FREE)\n /drivers/md/dm-zoned-target.c: 137 in dmz_submit_bio()\n 131 clone-\u003ebi_private \u003d bioctx;\n 132\n 133 bio_advance(bio, clone-\u003ebi_iter.bi_size);\n 134\n 135 refcount_inc(\u0026bioctx-\u003eref);\n 136 generic_make_request(clone);\n \u003e\u003e\u003e CID 1452808: Memory - illegal accesses (USE_AFTER_FREE)\n \u003e\u003e\u003e Dereferencing freed pointer \"clone\".\n 137 if (clone-\u003ebi_status \u003d\u003d BLK_STS_IOERR)\n 138 return -EIO;\n 139\n 140 if (bio_op(bio) \u003d\u003d REQ_OP_WRITE \u0026\u0026 dmz_is_seq(zone))\n 141 zone-\u003ewp_block +\u003d nr_blocks;\n 142\n \n The \"clone\" bio may be processed and freed before the check\n \"clone-\u003ebi_status \u003d\u003d BLK_STS_IOERR\" - so this check can access invalid\n memory.\n \n Fixes: 75d66ffb48efb3 (\"dm zoned: properly handle backing device failure\")\n Cc: stable@vger.kernel.org\n Signed-off-by: Mikulas Patocka \u003cmpatocka@redhat.com\u003e\n Reviewed-by: Damien Le Moal \u003cdamien.lemoal@wdc.com\u003e\n Signed-off-by: Mike Snitzer \u003csnitzer@redhat.com\u003e\n\ncommit d8abe88450beb96a66e434323eb6ab737654b840\nAuthor: Mark Zhang \u003cmarkz@mellanox.com\u003e\nDate: Mon Aug 19 14:36:26 2019 +0300\n\n RDMA/mlx5: RDMA_RX flow type support for user applications\n \n Currently user applications can only steer TCP/IP(NIC RX/RX) traffic.\n This patch adds RDMA_RX as a new flow type to allow the user to insert\n steering rules to control RDMA traffic.\n Two destinations are supported(but not set at the same time): devx\n flow table object and QP.\n \n Signed-off-by: Mark Zhang \u003cmarkz@mellanox.com\u003e\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Link: https://lore.kernel.org/r/20190819113626.20284-4-leon@kernel.org\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit cc14c00c478ce503c28856e7b2425aae5bc7b31e\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Mon Aug 26 11:01:25 2019 -0300\n\n v4l2-core: fix coding style for the two new c files\n \n As I2C and SPI parts of the V4L2 core got split, let\u0027s take\n the chance and solve the CodingStyle issues there, as reported\n by checkpatch --strict.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 4c098dab121a4164e9daa4d1ddbc59e93054e79e\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Aug 26 22:54:15 2019 +0900\n\n ALSA: oxfw: fix NULL pointer dereference to unused stream structure\n \n ALSA oxfw driver supports Griffin FireWave. This device supports one\n isochronous stream for PCM playback. Current driver executes code to\n wait event for stream structure of opposite direction. This causes\n NULL pointer dereference.\n \n This commit fixes the bug.\n \n Fixes: ac5d77864cfc (\"ALSA: oxfw: support AMDTP domain\")\n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 1a065ee3e97cc58ab900ccb7eb2b421fdab1d0dc\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Thu Aug 15 13:48:06 2019 -0300\n\n media: v4l2-core: Remove BUG() from i2c and spi helpers\n \n Currently, the i2c and spi subdev creation helpers\n are calling BUG() when passed a NULL v4l2_device parameter.\n \n This makes little sense; simply returning NULL seems more\n sensible.\n \n These two helpers may already return NULL on error, so callers\n should already be checking for this, or at least be prepared\n for a NULL result.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 51ff392c280733aa9e6bd47b3f5e83e32bfdf16a\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Thu Aug 15 13:48:05 2019 -0300\n\n media: v4l2-core: introduce a helper to unregister a i2c subdev\n \n Introduce a new video4linux2 i2c helper, to unregister a subdev.\n This allows to get rid of yet another ifdef.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n [hverkuil-cisco@xs4all.nl: fix checkpatch warning]\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit b39e077fcb283dd96dd251a3abeba585402c61fe\nAuthor: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nDate: Mon Aug 26 16:45:50 2019 +0300\n\n ALSA: usb-audio: remove some dead code\n \n We recently cleaned up the error handling in commit 52c3e317a857 (\"ALSA:\n usb-audio: Unify the release of usb_mixer_elem_info objects\") but\n accidentally left this stray return.\n \n Fixes: 52c3e317a857 (\"ALSA: usb-audio: Unify the release of usb_mixer_elem_info objects\")\n Signed-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit a9cff393c1d78ecbbc33e6196e79bb05ccb4a709\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Thu Aug 15 13:48:04 2019 -0300\n\n media: v4l2-core: introduce a helper to unregister a spi subdev\n \n Introduce a new video4linux2 spi helper, to unregister a subdev.\n This allows to get rid of some more ifdefs.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 02283b98b1ac47659b17d575ea24e2b92ead7ede\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Thu Aug 15 13:48:03 2019 -0300\n\n media: v4l2-core: move i2c helpers out of v4l2-common.c\n \n Separate the i2c helpers to v4l2-i2c.c, in order to get rid\n of the ifdefery. No functional changes intended, this is\n just a cosmetic change to organize the code better.\n \n Given I2C is a tristate symbol, a hidden boolean symbol\n is introduced, to make the conditional build easier.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 7c795df5f344482c2ab8c52ebc1d94302d2b9082\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Thu Aug 15 13:48:02 2019 -0300\n\n media: v4l2-core: move spi helpers out of v4l2-common.c\n \n Separate the spi helpers to v4l2-spi.c, in order to get rid\n of the ifdefery. No functional changes intended, this is\n just a cosmetic change to organize the code better.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit ff35213fda9e2c6ba38979229b42d49dafc6e7d9\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Thu Aug 15 13:48:01 2019 -0300\n\n media: v4l2-core: Module re-organization\n \n videodev.ko and v4l2-common.ko driver are built under\n the same conditions. Therefore, it doesn\u0027t make much sense\n to split them in two different modules.\n \n Splitting v4l2-common to its own driver was done many years ago:\n \n commit a9254475bbfbed5f0596d952c6a3c9806e19dd0b\n Author: Mauro Carvalho Chehab \u003cmchehab@infradead.org\u003e\n Date: Tue Jan 29 18:32:35 2008 -0300\n \n V4L/DVB (7115): Fix bug #9833: regression when compiling V4L without I2C\n \n Back then, the subsystem organization was different and the module split\n was needed. However, with the current organization, there is no issue\n compiling V4L2 with I2C as y/m/n.\n \n This commit makes v4l2-common part of our V4L2 core driver (videodev.ko).\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 9ca6f5120c89cdfcb6250449092692dd96db9c64\nAuthor: Souptick Joarder \u003cjrdr.linux@gmail.com\u003e\nDate: Sun Aug 25 17:25:41 2019 -0300\n\n media: usbvision: Remove dead code\n \n These codes are commented since v4.2-rc1. If there is no plan\n to enable the #else part in future, these can be removed\n forever.\n \n Signed-off-by: Souptick Joarder \u003cjrdr.linux@gmail.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 578c5ab074981b1e74331e58df4b8ef4439df563\nAuthor: Jae Hyun Yoo \u003cjae.hyun.yoo@linux.intel.com\u003e\nDate: Fri Aug 23 18:29:57 2019 -0300\n\n media: aspeed: fix an incorrect return code on buffer allocation failure\n \n It returns \u00270\u0027 even when a failure happens on jpeg buffer allocation\n so this commit fixes the issue.\n \n Signed-off-by: Jae Hyun Yoo \u003cjae.hyun.yoo@linux.intel.com\u003e\n Reviewed-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 834a1cfee8340775798296eb693964c883e239a2\nAuthor: Jonas Karlman \u003cjonas@kwiboo.se\u003e\nDate: Wed Aug 21 14:27:47 2019 -0300\n\n media: hantro: Enable VP8 decoding on RK3328\n \n RK3328 SoC has the same decoder IP block as RK3399,\n lets enable VP8 decoding on RK3328.\n \n Signed-off-by: Jonas Karlman \u003cjonas@kwiboo.se\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 9d802222a3405599d6e1984d9324cddf592ea1f4\nAuthor: Maciej S. Szmigiero \u003cmail@maciej.szmigiero.name\u003e\nDate: Tue Aug 20 19:05:55 2019 -0300\n\n media: saa7134: fix terminology around saa7134_i2c_eeprom_md7134_gate()\n \n saa7134_i2c_eeprom_md7134_gate() function and the associated comment uses\n an inverted i2c gate open / closed terminology.\n Let\u0027s fix this.\n \n Signed-off-by: Maciej S. Szmigiero \u003cmail@maciej.szmigiero.name\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n [hverkuil-cisco@xs4all.nl: fix alignment checkpatch warning]\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit c1849372a68145b277c917fb5db8fb3f0a50e11e\nAuthor: Maciej S. Szmigiero \u003cmail@maciej.szmigiero.name\u003e\nDate: Tue Aug 20 19:05:54 2019 -0300\n\n media: saa7134: keep demod i2c gate closed on Medion 7134\n \n Medion 7134 has two i2c eeproms on the same i2c bus sharing the same bus\n addresses: the first one for SAA7134 chip config and the second one behind\n TDA10046 DVB-T demod chip i2c gate storing its firmware.\n The TV tuner on this board is not behind this i2c gate.\n \n Due to the bus conflict described above, the card PCI SVID / SSID sometimes\n gets garbled after a reboot, which makes it necessary to specify the card\n model manually as an insmod option in order for it to be detected reliably.\n To avoid this, let\u0027s just leave the gate permanently closed so the eeprom\n chips won\u0027t clash.\n \n The demod firmware load is done with its i2c gate closed anyway so it is\n not affected by this change.\n \n Signed-off-by: Maciej S. Szmigiero \u003cmail@maciej.szmigiero.name\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit b5654c9425157706b86bdddb10b3532d09ff72e2\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Tue Aug 20 12:34:41 2019 -0300\n\n media: i2c: adv7842: drop check because i2c_unregister_device() is NULL safe\n \n No need to check the argument of i2c_unregister_device() because the\n function itself does it.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 086c51e2405e75bb5a7575e3bdc784ae14c8a4e6\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Tue Aug 20 12:34:40 2019 -0300\n\n media: i2c: adv7511-v4l2: drop check because i2c_unregister_device() is NULL safe\n \n No need to check the argument of i2c_unregister_device() because the\n function itself does it.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit eb0a6fb25c05679d32698c2503ee39bdbbc13663\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Sun Aug 18 15:36:08 2019 -0300\n\n media: hdpvr: remove redundant assignment to retval\n \n Variable retval is initialized to a value that is never read\n so this assignment is redundant. Clean up the code by removing\n retval completely and just return the error return from the\n call to i2c_add_adapter.\n \n Addresses-Coverity: (\"Unused value\")\n \n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 7e0bb5828311f811309bed5749528ca04992af2f\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Sun Aug 18 12:03:23 2019 -0300\n\n media: sn9c20x: Add MSI MS-1039 laptop to flip_dmi_table\n \n Like a bunch of other MSI laptops the MS-1039 uses a 0c45:627b\n SN9C201 + OV7660 webcam which is mounted upside down.\n \n Add it to the sn9c20x flip_dmi_table to deal with this.\n \n Cc: stable@vger.kernel.org\n Reported-by: Rui Salvaterra \u003crsalvaterra@gmail.com\u003e\n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 6ca5989e44d3a33710aef8c3f3e8e72156913412\nAuthor: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nDate: Sun Aug 18 03:40:04 2019 -0300\n\n media: ti-vpe: Add cleanup in vpdma_list_cleanup()\n \n If an error occurs in this function, no cleanup is executed, leading to\n memory/resource leaks. To fix this issue, introduce two labels to perform\n the cleanup work.\n \n Signed-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 51d47e578a74c687bd71e462ef7db039ce38a964\nAuthor: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nDate: Sun Aug 18 02:58:53 2019 -0300\n\n media: fdp1: Fix a memory leak bug\n \n In fdp1_open(), \u0027ctx\u0027 is allocated through kzalloc(). However, it is not\n deallocated if v4l2_ctrl_new_std() fails, leading to a memory leak bug. To\n fix this issue, free \u0027ctx\u0027 before going to the \u0027done\u0027 label.\n \n Signed-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\n Reviewed-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 42e64117d3b4a759013f77bbcf25ab6700e55de7\nAuthor: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nDate: Sun Aug 18 02:40:14 2019 -0300\n\n media: saa7146: add cleanup in hexium_attach()\n \n If saa7146_register_device() fails, no cleanup is executed, leading to\n memory/resource leaks. To fix this issue, perform necessary cleanup work\n before returning the error.\n \n Signed-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 725a28290a53bcc02a3de1c1bf64a416d567b283\nAuthor: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nDate: Sat Aug 17 03:14:54 2019 -0300\n\n media: usb: cx231xx-417: fix a memory leak bug\n \n In cx231xx_load_firmware(), \u0027p_buffer\u0027 is allocated through vmalloc() to\n hold the firmware. However, after the usage, it is not deallocated, leading\n to a memory leak bug.\n \n Signed-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 1c770f0f52dca1a2323c594f01f5ec6f1dddc97f\nAuthor: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nDate: Sat Aug 17 02:27:46 2019 -0300\n\n media: cpia2_usb: fix memory leaks\n \n In submit_urbs(), \u0027cam-\u003esbuf[i].data\u0027 is allocated through kmalloc_array().\n However, it is not deallocated if the following allocation for urbs fails.\n To fix this issue, free \u0027cam-\u003esbuf[i].data\u0027 if usb_alloc_urb() fails.\n \n Signed-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 38d3cfbc95624fa2e8e84cd36c10bfef31d90f0c\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Tue Aug 20 17:34:49 2019 +0200\n\n platform/chrome: chromeos_laptop: drop checks of NULL-safe functions\n \n No need to check the argument of i2c_unregister_device() and\n property_entries_free() because the functions do check it.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n\ncommit 560ec60464462c8896e6dd81a27f23789222acdb\nAuthor: Nikolay Borisov \u003cnborisov@suse.com\u003e\nDate: Thu Aug 22 17:24:20 2019 +0300\n\n btrfs: improve error handling in run_delalloc_nocow\n \n Correctly handle failure cases when adding an ordered extents in case\n of REGULAR or PREALLOC extents. Remove the BUG_ON.\n \n Signed-off-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit ebe48fd0caafb580a6e5b52082fb2304a8aea90e\nAuthor: Nikolay Borisov \u003cnborisov@suse.com\u003e\nDate: Thu Aug 22 17:25:23 2019 +0300\n\n btrfs: comment and minor simplifications in run_delalloc_nocow\n \n Add a comment explaining why we keep the BUG also use the already read\n and cached value of extent ram bytes stored in \u0027ram_bytes\u0027.\n \n Signed-off-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 081fad2b15ab10854ec3d20d066750e11960c09a\nAuthor: Nikolay Borisov \u003cnborisov@suse.com\u003e\nDate: Mon Aug 5 17:47:06 2019 +0300\n\n btrfs: streamline code in run_delalloc_nocow in case of inline extents\n \n The extent range check right after the \"out_check\" label is redundant,\n because the only way it can trigger is if we have an inline extent. In\n this case it makes more sense to actually move it in the branch\n explictly dealing with inlines extents.\n \n What\u0027s more, the nested \u0027if (nocow)\u0027 can never be true because for\n inline extents we always do COW and there is no chance \u0027nocow\u0027 can be\n true, just remove that check.\n \n Signed-off-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 1985e22630972ca74972868620b397085e1f7378\nAuthor: Nikolay Borisov \u003cnborisov@suse.com\u003e\nDate: Mon Aug 5 17:47:05 2019 +0300\n\n btrfs: simplify extent type checks in run_delalloc_nocow\n \n There is no point in checking the type of the extent again just to set\n the \u0027type\u0027 variable, when this check has already been performed before.\n Instead, extend the original if branch with an \u0027else\u0027 clause. This\n allows to remove one local variable and make it obvious how the code\n flow differs for prealloc/regular extents.\n \n Reviewed-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Signed-off-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 0430f316feb0564207a81dcb1b5c8aa761053549\nAuthor: Nikolay Borisov \u003cnborisov@suse.com\u003e\nDate: Wed Aug 21 10:42:57 2019 +0300\n\n btrfs: improve comments around nocow path\n \n run_delalloc_nocow contains numerous, somewhat subtle, checks when\n figuring out whether a particular extent should be CoW\u0027ed or not. This\n patch explicitly states the assumptions those checks verify. As a\n result also document 2 of the more subtle checks in check_committed_ref\n as well.\n \n Signed-off-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit bfcf241bece10f1789b97962167eca4f4f950335\nAuthor: Nikolay Borisov \u003cnborisov@suse.com\u003e\nDate: Wed Aug 21 10:42:03 2019 +0300\n\n btrfs: refactor variable scope in run_delalloc_nocow\n \n Of the 22 (!!!) local variables declared in this function only 9 have\n function-wide context. Of the remaining 13, 12 are needed in the main\n while loop of the function and 1 is needed in a tiny if branch, only in\n case we have prealloc extent. This commit reduces the lifespan of every\n variable to its bare minimum. It also renames the \u0027nolock\u0027 boolean to\n freespace_inode to clearly indicate its purpose.\n \n Signed-off-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit e031372472c3896c96c84c50dbf78c20bdc27e40\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 22 15:14:34 2019 -0400\n\n btrfs: only reserve metadata_size for inodes\n \n Historically we reserved worst case for every btree operation, and\n generally speaking we want to do that in cases where it could be the\n worst case. However for updating inodes we know the inode items are\n already in the tree, so it will only be an update operation and never an\n insert operation. This allows us to always reserve only the\n metadata_size amount for inode updates rather than the\n insert_metadata_size amount.\n \n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 98586765c7ec5c043afd09d9e9e0da3da6dd9d84\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 22 15:14:33 2019 -0400\n\n btrfs: rename the btrfs_calc_*_metadata_size helpers\n \n btrfs_calc_trunc_metadata_size differs from trans_metadata_size in that\n it doesn\u0027t take into account any splitting at the levels, because\n truncate will never split nodes. However truncate _and_ changing will\n never split nodes, so rename btrfs_calc_trunc_metadata_size to\n btrfs_calc_metadata_size. Also btrfs_calc_trans_metadata_size is purely\n for inserting items, so rename this to btrfs_calc_insert_metadata_size.\n Making these clearer will help when I start using them differently in\n upcoming patches.\n \n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit ce7eb70f9fd7fa579ade73d1c1a3358e72df8ba5\nAuthor: Qu Wenruo \u003cwqu@suse.com\u003e\nDate: Fri Aug 9 09:24:24 2019 +0800\n\n btrfs: tree-checker: Add EXTENT_DATA_REF check\n \n EXTENT_DATA_REF is a little like DIR_ITEM which contains hash in its\n key-\u003eoffset.\n \n This patch will check the following contents:\n - Key-\u003eobjectid\n Basic alignment check.\n \n - Hash\n Hash of each extent_data_ref item must match key-\u003eoffset.\n \n - Offset\n Basic alignment check.\n \n Signed-off-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit a206cdef8c8543bb24e86e76aadb12d95689c2a5\nAuthor: Qu Wenruo \u003cwqu@suse.com\u003e\nDate: Fri Aug 9 09:24:23 2019 +0800\n\n btrfs: tree-checker: Add simple keyed refs check\n \n For TREE_BLOCK_REF, SHARED_DATA_REF and SHARED_BLOCK_REF we need to\n check:\n | TREE_BLOCK_REF | SHARED_BLOCK_REF | SHARED_BLOCK_REF\n --------------+----------------+-----------------+------------------\n key-\u003eobjectid | Alignment | Alignment | Alignment\n key-\u003eoffset | Any value | Alignment | Alignment\n item_size | 0 | 0 | sizeof(le32) (*)\n \n *: sizeof(struct btrfs_shared_data_ref)\n \n So introduce a check to check all these 3 key types together.\n \n Signed-off-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 537d313887157b15dce5c7aeaacd2fcdf56f807b\nAuthor: Qu Wenruo \u003cwqu@suse.com\u003e\nDate: Fri Aug 9 09:24:22 2019 +0800\n\n btrfs: tree-checker: Add EXTENT_ITEM and METADATA_ITEM check\n \n This patch introduces the ability to check extent items.\n \n This check involves:\n - key-\u003eobjectid check\n Basic alignment check.\n \n - key-\u003etype check\n Against btrfs_extent_item::type and SKINNY_METADATA feature.\n \n - key-\u003eoffset alignment check for EXTENT_ITEM\n \n - key-\u003eoffset check for METADATA_ITEM\n \n - item size check\n Both against minimal size and stepping check.\n \n - btrfs_extent_item check\n Checks its flags and generation.\n \n - btrfs_extent_inline_ref checks\n Against 4 types inline ref.\n Checks bytenr alignment and tree level.\n \n - btrfs_extent_item::refs check\n Check against total refs found in inline refs.\n \n This check would be the most complex single item check due to its nature\n of inlined items.\n \n Signed-off-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 4e157e87c51f00de0b4b20c9727a5dc307bf73f7\nAuthor: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nDate: Fri Aug 9 17:07:39 2019 +0300\n\n btrfs: fix errror pointer check in __btrfs_map_block()\n \n The btrfs_get_chunk_map() never returns NULL, it returns error pointers.\n \n Fixes: 89b798ad1b42 (\"btrfs: Use btrfs_get_io_geometry appropriately\")\n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 83e7547fe93dbffa68bad65883da120fd20bef5a\nAuthor: Anand Jain \u003canand.jain@oracle.com\u003e\nDate: Wed Aug 21 17:26:32 2019 +0800\n\n btrfs: dev stat drop useless goto\n \n In the function btrfs_init_dev_stats() goto out is not needed, because the\n alloc has failed. So just return -ENOMEM.\n \n Signed-off-by: Anand Jain \u003canand.jain@oracle.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit d5231f91d7906bc15318a875ac6239d1f8a413bf\nAuthor: Anand Jain \u003canand.jain@oracle.com\u003e\nDate: Wed Aug 21 17:26:33 2019 +0800\n\n btrfs: dev stats item key conversion per cpu type is not needed\n \n %found_key is not used, drop it since it hasn\u0027t been used since the\n beginning in 733f4fbbc108 (\"Btrfs: read device stats on mount, write\n modified ones during commit\").\n \n Signed-off-by: Anand Jain \u003canand.jain@oracle.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit dbbec2d13992b68cebe163e3ea5bfdf68dc213b7\nAuthor: Nikolay Borisov \u003cnborisov@suse.com\u003e\nDate: Wed Aug 21 16:38:15 2019 +0300\n\n btrfs: Make reada_tree_block_flagged private\n \n This function is used only for the readahead machinery. It makes no\n sense to keep it external to reada.c file. Place it above its sole\n caller and make it static. No functional changes.\n \n Reviewed-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Signed-off-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit c03012cb9554f6f28f9cda4db2e2fbe121d0dcfb\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Fri Aug 9 16:49:06 2019 +0200\n\n btrfs: compression: replace set_level callbacks by a common helper\n \n The set_level callbacks do not do anything special and can be replaced\n by a helper that uses the levels defined in the tables.\n \n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Reviewed-by: Anand Jain \u003canand.jain@oracle.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit ba00c45c8af267cac53a722e11e06555595b426f\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Fri Aug 9 16:25:34 2019 +0200\n\n btrfs: define compression levels statically\n \n The maximum and default levels do not change and can be defined\n directly. The set_level callback was a temporary solution and will be\n removed.\n \n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Reviewed-by: Anand Jain \u003canand.jain@oracle.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit bcc4baa7b2d2001c7cb2032ede0c7f8e2d886a06\nAuthor: Filipe Manana \u003cfdmanana@suse.com\u003e\nDate: Mon Aug 12 19:14:29 2019 +0100\n\n Btrfs: fix use-after-free when using the tree modification log\n \n At ctree.c:get_old_root(), we are accessing a root\u0027s header owner field\n after we have freed the respective extent buffer. This results in an\n use-after-free that can lead to crashes, and when CONFIG_DEBUG_PAGEALLOC\n is set, results in a stack trace like the following:\n \n [ 3876.799331] stack segment: 0000 [#1] SMP DEBUG_PAGEALLOC PTI\n [ 3876.799363] CPU: 0 PID: 15436 Comm: pool Not tainted 5.3.0-rc3-btrfs-next-54 #1\n [ 3876.799385] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.0-0-ga698c8995f-prebuilt.qemu.org 04/01/2014\n [ 3876.799433] RIP: 0010:btrfs_search_old_slot+0x652/0xd80 [btrfs]\n (...)\n [ 3876.799502] RSP: 0018:ffff9f08c1a2f9f0 EFLAGS: 00010286\n [ 3876.799518] RAX: ffff8dd300000000 RBX: ffff8dd85a7a9348 RCX: 000000038da26000\n [ 3876.799538] RDX: 0000000000000000 RSI: ffffe522ce368980 RDI: 0000000000000246\n [ 3876.799559] RBP: dae1922adadad000 R08: 0000000008020000 R09: ffffe522c0000000\n [ 3876.799579] R10: ffff8dd57fd788c8 R11: 000000007511b030 R12: ffff8dd781ddc000\n [ 3876.799599] R13: ffff8dd9e6240578 R14: ffff8dd6896f7a88 R15: ffff8dd688cf90b8\n [ 3876.799620] FS: 00007f23ddd97700(0000) GS:ffff8dda20200000(0000) knlGS:0000000000000000\n [ 3876.799643] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n [ 3876.799660] CR2: 00007f23d4024000 CR3: 0000000710bb0005 CR4: 00000000003606f0\n [ 3876.799682] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\n [ 3876.799703] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\n [ 3876.799723] Call Trace:\n [ 3876.799735] ? do_raw_spin_unlock+0x49/0xc0\n [ 3876.799749] ? _raw_spin_unlock+0x24/0x30\n [ 3876.799779] resolve_indirect_refs+0x1eb/0xc80 [btrfs]\n [ 3876.799810] find_parent_nodes+0x38d/0x1180 [btrfs]\n [ 3876.799841] btrfs_check_shared+0x11a/0x1d0 [btrfs]\n [ 3876.799870] ? extent_fiemap+0x598/0x6e0 [btrfs]\n [ 3876.799895] extent_fiemap+0x598/0x6e0 [btrfs]\n [ 3876.799913] do_vfs_ioctl+0x45a/0x700\n [ 3876.799926] ksys_ioctl+0x70/0x80\n [ 3876.799938] ? trace_hardirqs_off_thunk+0x1a/0x20\n [ 3876.799953] __x64_sys_ioctl+0x16/0x20\n [ 3876.799965] do_syscall_64+0x62/0x220\n [ 3876.799977] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n [ 3876.799993] RIP: 0033:0x7f23e0013dd7\n (...)\n [ 3876.800056] RSP: 002b:00007f23ddd96ca8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010\n [ 3876.800078] RAX: ffffffffffffffda RBX: 00007f23d80210f8 RCX: 00007f23e0013dd7\n [ 3876.800099] RDX: 00007f23d80210f8 RSI: 00000000c020660b RDI: 0000000000000003\n [ 3876.800626] RBP: 000055fa2a2a2440 R08: 0000000000000000 R09: 00007f23ddd96d7c\n [ 3876.801143] R10: 00007f23d8022000 R11: 0000000000000246 R12: 00007f23ddd96d80\n [ 3876.801662] R13: 00007f23ddd96d78 R14: 00007f23d80210f0 R15: 00007f23ddd96d80\n (...)\n [ 3876.805107] ---[ end trace e53161e179ef04f9 ]---\n \n Fix that by saving the root\u0027s header owner field into a local variable\n before freeing the root\u0027s extent buffer, and then use that local variable\n when needed.\n \n Fixes: 30b0463a9394d9 (\"Btrfs: fix accessing the root pointer in tree mod log functions\")\n CC: stable@vger.kernel.org # 3.10+\n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Reviewed-by: Anand Jain \u003canand.jain@oracle.com\u003e\n Signed-off-by: Filipe Manana \u003cfdmanana@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 07993e6a3fda24724ad5f5c8cc4e65d665e2c050\nAuthor: Anand Jain \u003canand.jain@oracle.com\u003e\nDate: Thu Aug 8 12:32:44 2019 +0800\n\n btrfs: replace: BTRFS_DEV_REPLACE_ITEM_STATE_x defines should go\n \n The BTRFS_DEV_REPLACE_ITEM_STATE_x defines, as shown in [1], are\n unused in both kernel and btrfs-progs (except for one instance of\n BTRFS_DEV_REPLACE_ITEM_STATE_NEVER_STARTED in kernel).\n \n [1]\n btrfs.h:#define BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED 2\n btrfs.h:#define BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED 3\n btrfs.h:#define BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED 4\n \n Further these define-values are different form its counterpart\n BTRFS_IOCTL_DEV_REPLACE_STATE_x series as shown in [2].\n \n [2]\n btrfs_tree.h:#define BTRFS_DEV_REPLACE_ITEM_STATE_SUSPENDED 2\n btrfs_tree.h:#define BTRFS_DEV_REPLACE_ITEM_STATE_FINISHED 3\n btrfs_tree.h:#define BTRFS_DEV_REPLACE_ITEM_STATE_CANCELED 4\n \n So this patch deletes the BTRFS_DEV_REPLACE_ITEM_STATE_x altogether, and\n one instance of BTRFS_DEV_REPLACE_ITEM_STATE_NEVER_STARTED is replaced\n with BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED in the kernel.\n \n Signed-off-by: Anand Jain \u003canand.jain@oracle.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 3c5e938e2ba78724def159e96234648e62e1b959\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 1 18:19:37 2019 -0400\n\n btrfs: introduce an evict flushing state\n \n We have this weird space flushing loop inside inode.c for evict where\n we\u0027ll do the normal LIMIT flush, and then commit the transaction and\n hope we get our space. This is super janky, and in fact there\u0027s really\n nothing stopping us from using FLUSH_ALL except that we run delayed\n iputs, which means we could deadlock. So introduce a new flush state\n for eviction that does the normal priority flushing with all of the\n states that are safe for eviction.\n \n The nice side-effect of this is that we\u0027ll try harder for evictions.\n Previously if (for example generic/269) you had a bunch of other\n operations happening on the fs you could race with those reservations\n when committing the transaction, and eventually miss getting a\n reservation for the evict. With this code we\u0027ll have our ticket in\n place through the transaction commit, so any pinned bytes will go to our\n pending evictions first.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 0049fef6d51fa889c0f9f76ce89d4d8de36f0794\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 1 18:19:36 2019 -0400\n\n btrfs: refactor priority_reclaim_metadata_space\n \n With the eviction flushing stuff we\u0027ll want to allow for different\n states, but still work basically the same way that\n priority_reclaim_metadata_space works currently. Refactor this to take\n the flushing states and size as an argument so we can use the same logic\n for limit flushing and eviction flushing.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 324ae0d055339b12d3f3c724933957c56e681638\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 1 18:19:35 2019 -0400\n\n btrfs: factor out the ticket flush handling\n \n We\u0027re going to make this logic a little more complicated for evict, so\n factor the ticket flushing/waiting code out of __reserve_metadata_bytes.\n This has no functional change.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit e934c7ea82c5b9cda32191ab9003431735f63c58\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 1 18:19:34 2019 -0400\n\n btrfs: unify error handling for ticket flushing\n \n Currently we handle the cleanup of errored out tickets in both the\n priority flush path and the normal flushing path. This is the same code\n in both places, so just refactor so we don\u0027t duplicate the cleanup work.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit e97583e121d6930c0767de2276de76810cdf3dfd\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Aug 1 18:19:33 2019 -0400\n\n btrfs: add a flush step for delayed iputs\n \n Delayed iputs could very well free up enough space without needing to\n commit the transaction, so make this step it\u0027s own step. This will\n allow us to skip the step for evictions in a later patch.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit c9ba1590bc564f5a3464645b7fdf95ea2b83f643\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:38:07 2019 -0400\n\n btrfs: unexport the temporary exported functions\n \n These were renamed and exported to facilitate logical migration of\n different code chunks into block-group.c. Now that all the users are in\n one file go ahead and rename them back, move the code around, and make\n them static.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 20fbb669b69d01b31e8ad04e9bf28a3721bfd98a\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:38:06 2019 -0400\n\n btrfs: migrate the block group cleanup code\n \n This can now be easily migrated as well.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n [ refresh on top of sysfs cleanups ]\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 5f47ae74756b23172af7cd3f4c6dbdedecff7e89\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:38:05 2019 -0400\n\n btrfs: migrate the alloc_profile helpers\n \n These feel more at home in block-group.c.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n [ refresh, adjust btrfs_get_alloc_profile exports ]\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit c54b98f3255d92d2025c5dde1ecc0a080a2877bb\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:38:04 2019 -0400\n\n btrfs: migrate the chunk allocation code\n \n This feels more at home in block-group.c than in extent-tree.c.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003ei\n [ refresh ]\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 097baabcaab02aad9027dc2063df750b9867bb6e\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:38:02 2019 -0400\n\n btrfs: migrate the block group space accounting helpers\n \n We can now easily migrate this code as well.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 4fddd4bfeb84fc6e47a5e98f6515f2d126bd4d1d\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:38:01 2019 -0400\n\n btrfs: export block group accounting helpers\n \n Want to move these functions into block-group.c, so export them.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 64392def4b398b2a461f9c2593873fb1f4695767\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:38:00 2019 -0400\n\n btrfs: migrate the dirty bg writeout code\n \n This can be easily migrated over now.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n [ update comments ]\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit ef6bad0ef908b23390fbdd20ff6006e98796fb11\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:37:59 2019 -0400\n\n btrfs: migrate inc/dec_block_group_ro code\n \n This can easily be moved now.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n [ refresh ]\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit fcb4018fe61ad75c7016d36dffbb60c2340faded\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:37:58 2019 -0400\n\n btrfs: temporarily export btrfs_get_restripe_target\n \n This gets used by a few different logical chunks of the block group\n code, export it while we move things around.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 73193b9fcc0ca1368fb4c6ac77ae9221d6b8e7a4\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:37:57 2019 -0400\n\n btrfs: migrate the block group read/creation code\n \n All of the prep work has been done so we can now cleanly move this chunk\n over.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n [ refresh, add btrfs_get_alloc_profile export, comment updates ]\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 738b2932d153204f5a5964c4a2819c78eababdc3\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:37:55 2019 -0400\n\n btrfs: migrate the block group removal code\n \n This is the removal code and the unused bgs code.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n [ refresh, move clear_incompat_bg_bits ]\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 8f0eb63c08e05962673515c2f17c4b7f1c2a5205\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:37:54 2019 -0400\n\n btrfs: temporarily export inc_block_group_ro\n \n This is used in a few logical parts of the block group code, temporarily\n export it so we can move things in pieces.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit a9c1df4a32da4fcef594148a89d2ee02f55f1baa\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Tue Aug 6 16:43:19 2019 +0200\n\n btrfs: migrate the block group caching code\n \n We can now just copy it over to block-group.c.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit fc520073f1c95e1baf4cb3f162efd96d36f2a48f\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Fri Aug 2 12:52:48 2019 +0200\n\n btrfs: sysfs: move helper macros to sysfs.c\n \n None of the macros is used outside of sysfs.c.\n \n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 2119b4d24ab274eb96cc13db7fd0c8c4b366ddcb\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Fri Aug 2 13:07:38 2019 +0200\n\n btrfs: sysfs: move type conversion helpers to sysfs.c\n \n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 284cba7b6c7f355f74e8fe1ba0b3f8ba0a28f8e9\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Thu Aug 1 19:46:20 2019 +0200\n\n btrfs: cleanup kobject.h includes\n \n The kobject should be pulled in via sysfs.h and that needs to include it\n because it needs various definitions like kobj_attribute or kobject.\n \n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 889d59ac4c3d2a071152e6e05c0e9df793bcae48\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Thu Aug 1 18:50:16 2019 +0200\n\n btrfs: factor out sysfs code for updating sprout fsid\n \n Wrap the fsid renaming code and move it to sysfs.c.\n \n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit af9a04c07fc4b91eb9ffc223f8434a329c936b9b\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Thu Aug 1 18:50:16 2019 +0200\n\n btrfs: factor out sysfs code for deleting block group and space infos\n \n The helpers to create block group and space info directories already\n live in sysfs.c, move the deletion part there too.\n \n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 3636b215832f4a87185f9f207f4fecff0cd22678\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Thu Aug 1 18:50:16 2019 +0200\n\n btrfs: factor out sysfs code for sending device uevent\n \n The device uevent belongs to the sysfs API.\n \n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 246ccbf4153ff6ad933a4713004ca6cda1d7cdd1\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Thu Aug 1 19:07:55 2019 +0200\n\n btrfs: sysfs: replace direct access to feature set names with a helper\n \n In order to unexport the feature type array, add a helper for the\n enum-to-string conversion.\n \n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit cff80313ccba787f30120507ffcf8c39da74e9f6\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Thu Aug 1 17:55:55 2019 +0200\n\n btrfs: sysfs: unexport space_info_ktype\n \n The last non-sysfs usage of space_info_ktype has been moved to a private\n helper in previous patch so the variable can be made static.\n \n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 6993522520d0638a9629a8f73bcae2d6efcb428f\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Thu Aug 1 18:50:16 2019 +0200\n\n btrfs: factor out sysfs code for creating space infos\n \n Move creation of data/metadata/system space info directories to sysfs.c.\n \n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 230c953d27091a070b1b80e41a2e0700a61fc241\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Thu Aug 1 17:55:55 2019 +0200\n\n btrfs: sysfs: unexport btrfs_raid_ktype\n \n The last non-sysfs usage of btrfs_raid_ktype has been moved to a private\n helper in previous patch so the variable can be made static.\n \n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 532171e1e421edfd016a45a21763cc39a2c61938\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Thu Aug 1 17:49:55 2019 +0200\n\n btrfs: factor sysfs code out of link_block_group\n \n The part of link_block_group that just creates the sysfs object is\n independent and can be factored out to a helper.\n \n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 2ff3a8e3849141085bbf11fdb183fb3609618ab8\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Thu Aug 1 17:34:41 2019 +0200\n\n btrfs: move sysfs declarations out of ctree.h\n \n As the header for sysfs code already exists, use it to clean up ctree.h.\n \n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 08e8d734a8980aa7e7a9ca2a3176326426be8d38\nAuthor: Anand Jain \u003canand.jain@oracle.com\u003e\nDate: Wed Aug 7 16:21:20 2019 +0800\n\n btrfs: opencode to reset all devices stat\n \n __btrfs_reset_dev_stats() is a small helper function to reset devices stat\n values, and is used only once, instead just open code it.\n \n Signed-off-by: Anand Jain \u003canand.jain@oracle.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 95b8ff9cb77229d79caba6d4f6a019bf8c9de330\nAuthor: Anand Jain \u003canand.jain@oracle.com\u003e\nDate: Wed Aug 7 16:21:19 2019 +0800\n\n btrfs: reset device stat using btrfs_dev_stat_set\n \n btrfs_dev_stat_reset() is an overdo in terms of wrapping. So this patch\n open codes btrfs_dev_stat_reset().\n \n Signed-off-by: Anand Jain \u003canand.jain@oracle.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 664d0aed1c5879f361f1704129fddf454f01c4db\nAuthor: Qu Wenruo \u003cwqu@suse.com\u003e\nDate: Tue Aug 6 22:05:07 2019 +0800\n\n btrfs: qgroup: Try our best to delete qgroup relations\n \n When we try to delete qgroups, we\u0027re pretty cautious, we make sure both\n qgroups exist and there is a relationship between them, then try to\n delete the relation.\n \n This behavior is OK, but the problem is we need to two relation items,\n and if we failed the first item deletion, we error out, leaving the\n other relation item in qgroup tree.\n \n Sometimes the error from del_qgroup_relation_item() could just be\n -ENOENT, thus we can ignore that error and continue without any problem.\n \n Further more, such cautious behavior makes qgroup relation deletion\n impossible for orphan relation items.\n \n This patch will enhance __del_qgroup_relation():\n - If both qgroups and their relation items exist\n Go the regular deletion routine and update their accounting if needed.\n \n - If any qgroup or relation item doesn\u0027t exist\n Then we still try to delete the orphan items anyway, but don\u0027t trigger\n the accounting update.\n \n By this, we try our best to remove relation items, and can handle orphan\n relation items properly, while still keep the existing behavior for good\n qgroup tree.\n \n Reported-by: Andrei Borzenkov \u003carvidjaar@gmail.com\u003e\n Signed-off-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 05d73916855e6fe5977fbba0d841b1c9b2f2cfc5\nAuthor: Hans van Kranenburg \u003chans.van.kranenburg@mendix.com\u003e\nDate: Sat Aug 3 23:36:34 2019 +0200\n\n btrfs: clarify btrfs_ioctl_get_dev_stats padding\n \n In commit c11d2c236cc26 (\"Btrfs: add ioctl to get and reset the device\n stats\") the get_dev_stats ioctl was added.\n \n Shortly thereafter, in commit b27f7c0c150f7 (\"btrfs: join DEV_STATS\n ioctls to one\") , the flags field was added. However, the calculation\n for unused padding space was not updated, which also invalidated the\n comment.\n \n Clarify what happened to reduce confusion and wasted time for anyone\n implementing this.\n \n Signed-off-by: Hans van Kranenburg \u003chans.van.kranenburg@mendix.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 190030d09f652350c417fda22436188eb17fd752\nAuthor: Filipe Manana \u003cfdmanana@suse.com\u003e\nDate: Mon Aug 5 10:57:41 2019 +0100\n\n Btrfs: make test_find_first_clear_extent_bit fail on incorrect results\n \n If any call to find_first_clear_extent_bit() returns an unexpected result,\n the test should fail and not just print an error message, otherwise it\n makes detection of regressions much harder to notice.\n \n Fixes: 1eaebb341d2b41 (\"btrfs: Don\u0027t trim returned range based on input value in find_first_clear_extent_bit\")\n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Filipe Manana \u003cfdmanana@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit cd31bbc2e36d628b15eaaef07ffc1dc99993ad10\nAuthor: Filipe Manana \u003cfdmanana@suse.com\u003e\nDate: Sat Aug 3 09:53:16 2019 +0100\n\n Btrfs: fix memory leaks in the test test_find_first_clear_extent_bit\n \n The test creates an extent io tree and sets several ranges with the\n CHUNK_ALLOCATED and CHUNK_TRIMMED bits, resulting in the allocation of\n several extent state structures. However the test never clears those\n ranges, resulting in memory leaks of the extent state structures.\n \n This is detected when CONFIG_BTRFS_DEBUG is set once we remove the\n btrfs module (rmmod btrfs):\n \n [57399.787918] BTRFS: state leak: start 67108864 end 75497471 state 1 in tree 1 refs 1\n [57399.790155] BTRFS: state leak: start 33554432 end 67108863 state 33 in tree 1 refs 1\n [57399.791941] BTRFS: state leak: start 1048576 end 4194303 state 33 in tree 1 refs 1\n [57399.793753] BTRFS: state leak: start 67108864 end 75497471 state 1 in tree 1 refs 1\n [57399.795188] BTRFS: state leak: start 33554432 end 67108863 state 33 in tree 1 refs 1\n [57399.796453] BTRFS: state leak: start 1048576 end 4194303 state 33 in tree 1 refs 1\n [57399.797765] BTRFS: state leak: start 67108864 end 75497471 state 1 in tree 1 refs 1\n [57399.799049] BTRFS: state leak: start 33554432 end 67108863 state 33 in tree 1 refs 1\n [57399.800142] BTRFS: state leak: start 1048576 end 4194303 state 33 in tree 1 refs 1\n [57399.801126] BTRFS: state leak: start 67108864 end 75497471 state 1 in tree 1 refs 1\n [57399.802106] BTRFS: state leak: start 33554432 end 67108863 state 33 in tree 1 refs 1\n [57399.803119] BTRFS: state leak: start 1048576 end 4194303 state 33 in tree 1 refs 1\n [57399.804153] BTRFS: state leak: start 67108864 end 75497471 state 1 in tree 1 refs 1\n [57399.805196] BTRFS: state leak: start 33554432 end 67108863 state 33 in tree 1 refs 1\n [57399.806191] BTRFS: state leak: start 1048576 end 4194303 state 33 in tree 1 refs 1\n \n The start and end offsets reported correspond exactly to the ranges\n used by the test.\n \n So fix that by clearing all the ranges when the test finishes.\n \n Fixes: 1eaebb341d2b41 (\"btrfs: Don\u0027t trim returned range based on input value in find_first_clear_extent_bit\")\n Signed-off-by: Filipe Manana \u003cfdmanana@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 361cbe432822bcd2739232d0d44b1e2ff622d3a6\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Thu Jun 13 17:27:36 2019 +0200\n\n btrfs: delete debugfs code\n \n Replaced by the sysfs exports that provide a more fine grained interface\n for filesystem debugging.\n \n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 91e4331a20613846821c6c4335f7dbb46f5b1b18\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Thu Jun 13 17:23:02 2019 +0200\n\n btrfs: sysfs: add debugging exports\n \n Add \u0027debug\u0027 directories to global sysfs and per-filesystem. This will\n replace the debugfs directory. The sysfs location is simpler and builds\n on top of the existing file hierarchy so there will hopefully be no more\n questions about the sample debugfs file.\n \n The directory is called \u0027debug\u0027 and only present under\n CONFIG_BTRFS_DEBUG so this will not affect productions builds.\n \n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 775ec996e6520955b6966b088683465f102d4ceb\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:37:52 2019 -0400\n\n btrfs: make caching_thread use btrfs_find_next_key\n \n extent-tree.c has a find_next_key that just walks up the path to find\n the next key, but it is used for both the caching stuff and the snapshot\n delete stuff. The snapshot deletion stuff is special so it can\u0027t really\n use btrfs_find_next_key, but the caching thread stuff can. We just need\n to fix btrfs_find_next_key to deal with -\u003eskip_locking and then it works\n exactly the same as the private find_next_key helper.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit be0928f3b34fddea0d83186503674954d3ad67ac\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:37:51 2019 -0400\n\n btrfs: temporarily export fragment_free_space\n \n This is used in caching and reading block groups, so export it while we\n move these chunks independently.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 1dca663a54c5a23b8b756637c0727cc437e7b6f4\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:37:50 2019 -0400\n\n btrfs: export the caching control helpers\n \n Man a lot of people use this stuff.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 3727e01782a1765a9492f250eedb2722e0de1223\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:37:49 2019 -0400\n\n btrfs: export the excluded extents helpers\n \n We\u0027ll need this to move the caching stuff around.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit f178ac0d6fe611022ef6291ca03a7bb876e47e8c\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:37:48 2019 -0400\n\n btrfs: export the block group caching helpers\n \n This will make it so we can move them easily.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n [ coding style updates ]\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 7233f2581553f2cab2d4e1b39a7cf70eca57ba28\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:37:47 2019 -0400\n\n btrfs: migrate nocow and reservation helpers\n \n These are relatively straightforward as well.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit bbd9497cb836cd5542e6fd1cc4aa3ab01e630b7d\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:37:46 2019 -0400\n\n btrfs: migrate the block group ref counting stuff\n \n Another easy set to move over to block-group.c.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit b3f9b3d6301e19cb4b6f2d10d6a3fa27592924db\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:37:45 2019 -0400\n\n btrfs: migrate the block group lookup code\n \n Move these bits first as they are the easiest to move. Export two of\n the helpers so they can be moved all at once.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n [ minor style updates ]\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 0a9ede2c53f641bbc084fae46b27b7a656ba7956\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:37:44 2019 -0400\n\n btrfs: move basic block_group definitions to their own header\n \n This is prep work for moving all of the block group cache code into its\n own file.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n [ minor comment updates ]\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 27e9780a41b20249b9865eafaaed96b2b5e6c00f\nAuthor: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nDate: Thu Jun 20 15:37:43 2019 -0400\n\n btrfs: move btrfs_add_free_space out of a header file\n \n This is prep work for moving block_group_cache around. Having this in\n the header file makes the header file include need to be in a certain\n order, which is awkward, so just move it into free-space-cache.c and\n then we can re-arrange later.\n \n Signed-off-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 40b98c622a27fe9780658473a8155d26178a9777\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Thu Aug 1 14:50:35 2019 +0200\n\n btrfs: tree-log: use symbolic name for first replay stage\n \n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Reviewed-by: Anand Jain \u003canand.jain@oracle.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 246a902589f809d30f3dc58da80053d36721f902\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Thu Aug 1 14:50:33 2019 +0200\n\n btrfs: async-thread: convert defines to enums\n \n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Reviewed-by: Anand Jain \u003canand.jain@oracle.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit a49cc13611be10ceab72043d54a90637d56dc614\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Thu Aug 1 14:50:30 2019 +0200\n\n btrfs: tree-log: convert defines to enums\n \n Used only for in-memory state tracking.\n \n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Reviewed-by: Anand Jain \u003canand.jain@oracle.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 35fd76d67b38148fb304d2f4a6337a9d5da472b3\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Thu Aug 1 19:53:04 2019 +0200\n\n btrfs: remove unused key type set/get helpers\n \n The switch to open coded set/get has happend long time ago in\n 962a298f3511 (\"btrfs: kill the key type accessor helpers\"), remove the\n stray helpers.\n \n Reviewed-by: Anand Jain \u003canand.jain@oracle.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 839101047eb3d83f34c23ff41487d1d53295d8e1\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Thu Aug 1 19:53:02 2019 +0200\n\n btrfs: remove unused btrfs_device::flush_bio_sent\n \n The status of flush bio is tracked as a status bit, changed in commit\n 1c3063b6dbfa (\"btrfs: cleanup device states define\n BTRFS_DEV_STATE_FLUSH_SENT\"), the flush_bio_sent was forgotten.\n \n Reviewed-by: Anand Jain \u003canand.jain@oracle.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 0ea00cfed3cab58b1c7ddc77524e875b2fb6f142\nAuthor: Filipe Manana \u003cfdmanana@suse.com\u003e\nDate: Tue Jul 2 15:23:07 2019 +0100\n\n Btrfs: remove unnecessary condition in btrfs_clone() to avoid too much nesting\n \n The bulk of the work done when cloning extents, at ioctl.c:btrfs_clone(),\n is done inside an if statement that checks if the found key has the type\n BTRFS_EXTENT_DATA_KEY. That if statement is redundant however, because\n btrfs_search_slot() always leaves us in a leaf slot that points to a key\n that is always greater then or equals to the search key, and our search\n key here has that type, and because we bail out before that if statement\n if the key at the given leaf slot is greater then BTRFS_EXTENT_DATA_KEY.\n \n Therefore just remove that if statement, not only because it is useless\n but mostly because it increases the nesting/indentation level in this\n function which is quite big and makes things a bit awkward whenever I need\n to fix something that requires changing btrfs_clone() (and it has been\n like that for many years already).\n \n Signed-off-by: Filipe Manana \u003cfdmanana@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit dfa698e083911293ea16d9fc171aa86c740e8ff3\nAuthor: Nikolay Borisov \u003cnborisov@suse.com\u003e\nDate: Wed Jul 3 15:32:59 2019 +0300\n\n btrfs: Refactor btrfs_calc_avail_data_space\n \n Simplify the code by removing variables that don\u0027t bring any real value\n as well as simplifying the checks when buidling the candidate list of\n devices. No functional changes.\n \n Signed-off-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit bf9274fee5efe1b298f59bbe4885e8a1bf673863\nAuthor: Nikolay Borisov \u003cnborisov@suse.com\u003e\nDate: Thu May 23 14:51:26 2019 +0300\n\n btrfs: Remove unnecessary check from join_running_log_trans\n \n join_running_log_trans checks btrfs_root::log_root outside of\n btrfs_root::log_mutex to avoid contention on the mutex. Turns out this\n check is not necessary because the two callers of join_running_log_trans\n (both of which deal with removing entries from the tree-log during\n unlink) explicitly check whether the respective inode has been logged in\n the current transaction.\n \n If it hasn\u0027t then it won\u0027t have any items in the tree-log and call path\n will return before calling join_running_log_trans. If the check passes,\n however, then it\u0027s guaranteed that btrfs_root::log_root is set because\n the inode is logged.\n \n Those guarantees allows us to remove the speculative as well as the\n implicity and tricky memory barrier.\n \n Reviewed-by: Filipe Manana \u003cfdmanana@suse.com\u003e\n Signed-off-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 01fc6c825d4ad8ae75cf778d523b9cab94814d82\nAuthor: Filipe Manana \u003cfdmanana@suse.com\u003e\nDate: Thu Jul 4 16:25:00 2019 +0100\n\n Btrfs: wake up inode cache waiters sooner to reduce waiting time\n \n If we need to start an inode caching thread, because none currently exists\n on disk, we can wake up all waiters as soon as we mark the range starting\n at root\u0027s highest objectid + 1 and ending at BTRFS_LAST_FREE_OBJECTID as\n free, so that they don\u0027t need to wait for the caching thread to start and\n do some progress. We follow the same approach within the caching thread,\n since as soon as it finds a free range and marks it as free space in the\n cache, it wakes up all waiters. So improve this by adding such a wakeup\n call after marking that initial range as free space.\n \n Fixes: a47d6b70e28040 (\"Btrfs: setup free ino caching in a more asynchronous way\")\n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Filipe Manana \u003cfdmanana@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 9d78a062e7a5c32083789bb821e3034eb2dcabc6\nAuthor: Filipe Manana \u003cfdmanana@suse.com\u003e\nDate: Thu Jul 4 16:24:44 2019 +0100\n\n Btrfs: fix inode cache waiters hanging on path allocation failure\n \n If the caching thread fails to allocate a path, it returns without waking\n up any cache waiters, leaving them hang forever. Fix this by following the\n same approach as when we fail to start the caching thread: print an error\n message, disable inode caching and make the wakers fallback to non-caching\n mode behaviour (calling btrfs_find_free_objectid()).\n \n Fixes: 581bb050941b4f (\"Btrfs: Cache free inode numbers in memory\")\n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Filipe Manana \u003cfdmanana@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit e8d5cd5eef2c3a1ef1eab4929980b699c77fe97d\nAuthor: Filipe Manana \u003cfdmanana@suse.com\u003e\nDate: Thu Jul 4 16:24:32 2019 +0100\n\n Btrfs: fix inode cache waiters hanging on failure to start caching thread\n \n If we fail to start the inode caching thread, we print an error message\n and disable the inode cache, however we never wake up any waiters, so they\n hang forever waiting for the caching to finish. Fix this by waking them\n up and have them fallback to a call to btrfs_find_free_objectid().\n \n Fixes: e60efa84252c05 (\"Btrfs: avoid triggering bug_on() when we fail to start inode caching task\")\n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Filipe Manana \u003cfdmanana@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 1e327ffc9f14feae150dab5a842439f69692ed93\nAuthor: Filipe Manana \u003cfdmanana@suse.com\u003e\nDate: Thu Jul 4 16:24:19 2019 +0100\n\n Btrfs: fix inode cache block reserve leak on failure to allocate data space\n \n If we failed to allocate the data extent(s) for the inode space cache, we\n were bailing out without releasing the previously reserved metadata. This\n was triggering the following warnings when unmounting a filesystem:\n \n $ cat -n fs/btrfs/inode.c\n (...)\n 9268 void btrfs_destroy_inode(struct inode *inode)\n 9269 {\n (...)\n 9276 WARN_ON(BTRFS_I(inode)-\u003eblock_rsv.reserved);\n 9277 WARN_ON(BTRFS_I(inode)-\u003eblock_rsv.size);\n (...)\n 9281 WARN_ON(BTRFS_I(inode)-\u003ecsum_bytes);\n 9282 WARN_ON(BTRFS_I(inode)-\u003edefrag_bytes);\n (...)\n \n Several fstests test cases triggered this often, such as generic/083,\n generic/102, generic/172, generic/269 and generic/300 at least, producing\n stack traces like the following in dmesg/syslog:\n \n [82039.079546] WARNING: CPU: 2 PID: 13167 at fs/btrfs/inode.c:9276 btrfs_destroy_inode+0x203/0x270 [btrfs]\n (...)\n [82039.081543] CPU: 2 PID: 13167 Comm: umount Tainted: G W 5.2.0-rc4-btrfs-next-50 #1\n [82039.081912] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.11.2-0-gf9626ccb91-prebuilt.qemu-project.org 04/01/2014\n [82039.082673] RIP: 0010:btrfs_destroy_inode+0x203/0x270 [btrfs]\n (...)\n [82039.083913] RSP: 0018:ffffac0b426a7d30 EFLAGS: 00010206\n [82039.084320] RAX: ffff8ddf77691158 RBX: ffff8dde29b34660 RCX: 0000000000000002\n [82039.084736] RDX: 0000000000000000 RSI: 0000000000000001 RDI: ffff8dde29b34660\n [82039.085156] RBP: ffff8ddf5fbec000 R08: 0000000000000000 R09: 0000000000000000\n [82039.085578] R10: ffffac0b426a7c90 R11: ffffffffb9aad768 R12: ffffac0b426a7db0\n [82039.086000] R13: ffff8ddf5fbec0a0 R14: dead000000000100 R15: 0000000000000000\n [82039.086416] FS: 00007f8db96d12c0(0000) GS:ffff8de036b00000(0000) knlGS:0000000000000000\n [82039.086837] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n [82039.087253] CR2: 0000000001416108 CR3: 00000002315cc001 CR4: 00000000003606e0\n [82039.087672] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\n [82039.088089] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\n [82039.088504] Call Trace:\n [82039.088918] destroy_inode+0x3b/0x70\n [82039.089340] btrfs_free_fs_root+0x16/0xa0 [btrfs]\n [82039.089768] btrfs_free_fs_roots+0xd8/0x160 [btrfs]\n [82039.090183] ? wait_for_completion+0x65/0x1a0\n [82039.090607] close_ctree+0x172/0x370 [btrfs]\n [82039.091021] generic_shutdown_super+0x6c/0x110\n [82039.091427] kill_anon_super+0xe/0x30\n [82039.091832] btrfs_kill_super+0x12/0xa0 [btrfs]\n [82039.092233] deactivate_locked_super+0x3a/0x70\n [82039.092636] cleanup_mnt+0x3b/0x80\n [82039.093039] task_work_run+0x93/0xc0\n [82039.093457] exit_to_usermode_loop+0xfa/0x100\n [82039.093856] do_syscall_64+0x162/0x1d0\n [82039.094244] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n [82039.094634] RIP: 0033:0x7f8db8fbab37\n (...)\n [82039.095876] RSP: 002b:00007ffdce35b468 EFLAGS: 00000246 ORIG_RAX: 00000000000000a6\n [82039.096290] RAX: 0000000000000000 RBX: 0000560d20b00060 RCX: 00007f8db8fbab37\n [82039.096700] RDX: 0000000000000001 RSI: 0000000000000000 RDI: 0000560d20b00240\n [82039.097110] RBP: 0000560d20b00240 R08: 0000560d20b00270 R09: 0000000000000015\n [82039.097522] R10: 00000000000006b4 R11: 0000000000000246 R12: 00007f8db94bce64\n [82039.097937] R13: 0000000000000000 R14: 0000000000000000 R15: 00007ffdce35b6f0\n [82039.098350] irq event stamp: 0\n [82039.098750] hardirqs last enabled at (0): [\u003c0000000000000000\u003e] 0x0\n [82039.099150] hardirqs last disabled at (0): [\u003cffffffffb7884ff2\u003e] copy_process.part.33+0x7f2/0x1f00\n [82039.099545] softirqs last enabled at (0): [\u003cffffffffb7884ff2\u003e] copy_process.part.33+0x7f2/0x1f00\n [82039.099925] softirqs last disabled at (0): [\u003c0000000000000000\u003e] 0x0\n [82039.100292] ---[ end trace f2521afa616ddccc ]---\n [82039.100707] WARNING: CPU: 2 PID: 13167 at fs/btrfs/inode.c:9277 btrfs_destroy_inode+0x1ac/0x270 [btrfs]\n (...)\n [82039.103050] CPU: 2 PID: 13167 Comm: umount Tainted: G W 5.2.0-rc4-btrfs-next-50 #1\n [82039.103428] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.11.2-0-gf9626ccb91-prebuilt.qemu-project.org 04/01/2014\n [82039.104203] RIP: 0010:btrfs_destroy_inode+0x1ac/0x270 [btrfs]\n (...)\n [82039.105461] RSP: 0018:ffffac0b426a7d30 EFLAGS: 00010206\n [82039.105866] RAX: ffff8ddf77691158 RBX: ffff8dde29b34660 RCX: 0000000000000002\n [82039.106270] RDX: 0000000000000000 RSI: 0000000000000001 RDI: ffff8dde29b34660\n [82039.106673] RBP: ffff8ddf5fbec000 R08: 0000000000000000 R09: 0000000000000000\n [82039.107078] R10: ffffac0b426a7c90 R11: ffffffffb9aad768 R12: ffffac0b426a7db0\n [82039.107487] R13: ffff8ddf5fbec0a0 R14: dead000000000100 R15: 0000000000000000\n [82039.107894] FS: 00007f8db96d12c0(0000) GS:ffff8de036b00000(0000) knlGS:0000000000000000\n [82039.108309] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n [82039.108723] CR2: 0000000001416108 CR3: 00000002315cc001 CR4: 00000000003606e0\n [82039.109146] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\n [82039.109567] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\n [82039.109989] Call Trace:\n [82039.110405] destroy_inode+0x3b/0x70\n [82039.110830] btrfs_free_fs_root+0x16/0xa0 [btrfs]\n [82039.111257] btrfs_free_fs_roots+0xd8/0x160 [btrfs]\n [82039.111675] ? wait_for_completion+0x65/0x1a0\n [82039.112101] close_ctree+0x172/0x370 [btrfs]\n [82039.112519] generic_shutdown_super+0x6c/0x110\n [82039.112988] kill_anon_super+0xe/0x30\n [82039.113439] btrfs_kill_super+0x12/0xa0 [btrfs]\n [82039.113861] deactivate_locked_super+0x3a/0x70\n [82039.114278] cleanup_mnt+0x3b/0x80\n [82039.114685] task_work_run+0x93/0xc0\n [82039.115083] exit_to_usermode_loop+0xfa/0x100\n [82039.115476] do_syscall_64+0x162/0x1d0\n [82039.115863] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n [82039.116254] RIP: 0033:0x7f8db8fbab37\n (...)\n [82039.117463] RSP: 002b:00007ffdce35b468 EFLAGS: 00000246 ORIG_RAX: 00000000000000a6\n [82039.117882] RAX: 0000000000000000 RBX: 0000560d20b00060 RCX: 00007f8db8fbab37\n [82039.118330] RDX: 0000000000000001 RSI: 0000000000000000 RDI: 0000560d20b00240\n [82039.118743] RBP: 0000560d20b00240 R08: 0000560d20b00270 R09: 0000000000000015\n [82039.119159] R10: 00000000000006b4 R11: 0000000000000246 R12: 00007f8db94bce64\n [82039.119574] R13: 0000000000000000 R14: 0000000000000000 R15: 00007ffdce35b6f0\n [82039.119987] irq event stamp: 0\n [82039.120387] hardirqs last enabled at (0): [\u003c0000000000000000\u003e] 0x0\n [82039.120787] hardirqs last disabled at (0): [\u003cffffffffb7884ff2\u003e] copy_process.part.33+0x7f2/0x1f00\n [82039.121182] softirqs last enabled at (0): [\u003cffffffffb7884ff2\u003e] copy_process.part.33+0x7f2/0x1f00\n [82039.121563] softirqs last disabled at (0): [\u003c0000000000000000\u003e] 0x0\n [82039.121933] ---[ end trace f2521afa616ddccd ]---\n [82039.122353] WARNING: CPU: 2 PID: 13167 at fs/btrfs/inode.c:9278 btrfs_destroy_inode+0x1bc/0x270 [btrfs]\n (...)\n [82039.124606] CPU: 2 PID: 13167 Comm: umount Tainted: G W 5.2.0-rc4-btrfs-next-50 #1\n [82039.125008] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.11.2-0-gf9626ccb91-prebuilt.qemu-project.org 04/01/2014\n [82039.125801] RIP: 0010:btrfs_destroy_inode+0x1bc/0x270 [btrfs]\n (...)\n [82039.126998] RSP: 0018:ffffac0b426a7d30 EFLAGS: 00010202\n [82039.127399] RAX: ffff8ddf77691158 RBX: ffff8dde29b34660 RCX: 0000000000000002\n [82039.127803] RDX: 0000000000000001 RSI: 0000000000000001 RDI: ffff8dde29b34660\n [82039.128206] RBP: ffff8ddf5fbec000 R08: 0000000000000000 R09: 0000000000000000\n [82039.128611] R10: ffffac0b426a7c90 R11: ffffffffb9aad768 R12: ffffac0b426a7db0\n [82039.129020] R13: ffff8ddf5fbec0a0 R14: dead000000000100 R15: 0000000000000000\n [82039.129428] FS: 00007f8db96d12c0(0000) GS:ffff8de036b00000(0000) knlGS:0000000000000000\n [82039.129846] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n [82039.130261] CR2: 0000000001416108 CR3: 00000002315cc001 CR4: 00000000003606e0\n [82039.130684] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\n [82039.131142] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\n [82039.131561] Call Trace:\n [82039.131990] destroy_inode+0x3b/0x70\n [82039.132417] btrfs_free_fs_root+0x16/0xa0 [btrfs]\n [82039.132844] btrfs_free_fs_roots+0xd8/0x160 [btrfs]\n [82039.133262] ? wait_for_completion+0x65/0x1a0\n [82039.133688] close_ctree+0x172/0x370 [btrfs]\n [82039.134157] generic_shutdown_super+0x6c/0x110\n [82039.134575] kill_anon_super+0xe/0x30\n [82039.134997] btrfs_kill_super+0x12/0xa0 [btrfs]\n [82039.135415] deactivate_locked_super+0x3a/0x70\n [82039.135832] cleanup_mnt+0x3b/0x80\n [82039.136239] task_work_run+0x93/0xc0\n [82039.136637] exit_to_usermode_loop+0xfa/0x100\n [82039.137029] do_syscall_64+0x162/0x1d0\n [82039.137418] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n [82039.137812] RIP: 0033:0x7f8db8fbab37\n (...)\n [82039.139059] RSP: 002b:00007ffdce35b468 EFLAGS: 00000246 ORIG_RAX: 00000000000000a6\n [82039.139475] RAX: 0000000000000000 RBX: 0000560d20b00060 RCX: 00007f8db8fbab37\n [82039.139890] RDX: 0000000000000001 RSI: 0000000000000000 RDI: 0000560d20b00240\n [82039.140302] RBP: 0000560d20b00240 R08: 0000560d20b00270 R09: 0000000000000015\n [82039.140719] R10: 00000000000006b4 R11: 0000000000000246 R12: 00007f8db94bce64\n [82039.141138] R13: 0000000000000000 R14: 0000000000000000 R15: 00007ffdce35b6f0\n [82039.141597] irq event stamp: 0\n [82039.142043] hardirqs last enabled at (0): [\u003c0000000000000000\u003e] 0x0\n [82039.142443] hardirqs last disabled at (0): [\u003cffffffffb7884ff2\u003e] copy_process.part.33+0x7f2/0x1f00\n [82039.142839] softirqs last enabled at (0): [\u003cffffffffb7884ff2\u003e] copy_process.part.33+0x7f2/0x1f00\n [82039.143220] softirqs last disabled at (0): [\u003c0000000000000000\u003e] 0x0\n [82039.143588] ---[ end trace f2521afa616ddcce ]---\n [82039.167472] WARNING: CPU: 3 PID: 13167 at fs/btrfs/extent-tree.c:10120 btrfs_free_block_groups+0x30d/0x460 [btrfs]\n (...)\n [82039.173800] CPU: 3 PID: 13167 Comm: umount Tainted: G W 5.2.0-rc4-btrfs-next-50 #1\n [82039.174847] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.11.2-0-gf9626ccb91-prebuilt.qemu-project.org 04/01/2014\n [82039.177031] RIP: 0010:btrfs_free_block_groups+0x30d/0x460 [btrfs]\n (...)\n [82039.180397] RSP: 0018:ffffac0b426a7dd8 EFLAGS: 00010206\n [82039.181574] RAX: ffff8de010a1db40 RBX: ffff8de010a1db40 RCX: 0000000000170014\n [82039.182711] RDX: ffff8ddff4380040 RSI: ffff8de010a1da58 RDI: 0000000000000246\n [82039.183817] RBP: ffff8ddf5fbec000 R08: 0000000000000000 R09: 0000000000000000\n [82039.184925] R10: ffff8de036404380 R11: ffffffffb8a5ea00 R12: ffff8de010a1b2b8\n [82039.186090] R13: ffff8de010a1b2b8 R14: 0000000000000000 R15: dead000000000100\n [82039.187208] FS: 00007f8db96d12c0(0000) GS:ffff8de036b80000(0000) knlGS:0000000000000000\n [82039.188345] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n [82039.189481] CR2: 00007fb044005170 CR3: 00000002315cc006 CR4: 00000000003606e0\n [82039.190674] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\n [82039.191829] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\n [82039.192978] Call Trace:\n [82039.194160] close_ctree+0x19a/0x370 [btrfs]\n [82039.195315] generic_shutdown_super+0x6c/0x110\n [82039.196486] kill_anon_super+0xe/0x30\n [82039.197645] btrfs_kill_super+0x12/0xa0 [btrfs]\n [82039.198696] deactivate_locked_super+0x3a/0x70\n [82039.199619] cleanup_mnt+0x3b/0x80\n [82039.200559] task_work_run+0x93/0xc0\n [82039.201505] exit_to_usermode_loop+0xfa/0x100\n [82039.202436] do_syscall_64+0x162/0x1d0\n [82039.203339] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n [82039.204091] RIP: 0033:0x7f8db8fbab37\n (...)\n [82039.206360] RSP: 002b:00007ffdce35b468 EFLAGS: 00000246 ORIG_RAX: 00000000000000a6\n [82039.207132] RAX: 0000000000000000 RBX: 0000560d20b00060 RCX: 00007f8db8fbab37\n [82039.207906] RDX: 0000000000000001 RSI: 0000000000000000 RDI: 0000560d20b00240\n [82039.208621] RBP: 0000560d20b00240 R08: 0000560d20b00270 R09: 0000000000000015\n [82039.209285] R10: 00000000000006b4 R11: 0000000000000246 R12: 00007f8db94bce64\n [82039.209984] R13: 0000000000000000 R14: 0000000000000000 R15: 00007ffdce35b6f0\n [82039.210642] irq event stamp: 0\n [82039.211306] hardirqs last enabled at (0): [\u003c0000000000000000\u003e] 0x0\n [82039.211971] hardirqs last disabled at (0): [\u003cffffffffb7884ff2\u003e] copy_process.part.33+0x7f2/0x1f00\n [82039.212643] softirqs last enabled at (0): [\u003cffffffffb7884ff2\u003e] copy_process.part.33+0x7f2/0x1f00\n [82039.213304] softirqs last disabled at (0): [\u003c0000000000000000\u003e] 0x0\n [82039.213875] ---[ end trace f2521afa616ddccf ]---\n \n Fix this by releasing the reserved metadata on failure to allocate data\n extent(s) for the inode cache.\n \n Fixes: 69fe2d75dd91d0 (\"btrfs: make the delalloc block rsv per inode\")\n Signed-off-by: Filipe Manana \u003cfdmanana@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit c53dc79788fdb95fc9ae114929753b08207ce1d4\nAuthor: Filipe Manana \u003cfdmanana@suse.com\u003e\nDate: Thu Jul 4 16:24:09 2019 +0100\n\n Btrfs: fix hang when loading existing inode cache off disk\n \n If we are able to load an existing inode cache off disk, we set the state\n of the cache to BTRFS_CACHE_FINISHED, but we don\u0027t wake up any one waiting\n for the cache to be available. This means that anyone waiting for the\n cache to be available, waiting on the condition that either its state is\n BTRFS_CACHE_FINISHED or its available free space is greather than zero,\n can hang forever.\n \n This could be observed running fstests with MOUNT_OPTIONS\u003d\"-o inode_cache\",\n in particular test case generic/161 triggered it very frequently for me,\n producing a trace like the following:\n \n [63795.739712] BTRFS info (device sdc): enabling inode map caching\n [63795.739714] BTRFS info (device sdc): disk space caching is enabled\n [63795.739716] BTRFS info (device sdc): has skinny extents\n [64036.653886] INFO: task btrfs-transacti:3917 blocked for more than 120 seconds.\n [64036.654079] Not tainted 5.2.0-rc4-btrfs-next-50 #1\n [64036.654143] \"echo 0 \u003e /proc/sys/kernel/hung_task_timeout_secs\" disables this message.\n [64036.654232] btrfs-transacti D 0 3917 2 0x80004000\n [64036.654239] Call Trace:\n [64036.654258] ? __schedule+0x3ae/0x7b0\n [64036.654271] schedule+0x3a/0xb0\n [64036.654325] btrfs_commit_transaction+0x978/0xae0 [btrfs]\n [64036.654339] ? remove_wait_queue+0x60/0x60\n [64036.654395] transaction_kthread+0x146/0x180 [btrfs]\n [64036.654450] ? btrfs_cleanup_transaction+0x620/0x620 [btrfs]\n [64036.654456] kthread+0x103/0x140\n [64036.654464] ? kthread_create_worker_on_cpu+0x70/0x70\n [64036.654476] ret_from_fork+0x3a/0x50\n [64036.654504] INFO: task xfs_io:3919 blocked for more than 120 seconds.\n [64036.654568] Not tainted 5.2.0-rc4-btrfs-next-50 #1\n [64036.654617] \"echo 0 \u003e /proc/sys/kernel/hung_task_timeout_secs\" disables this message.\n [64036.654685] xfs_io D 0 3919 3633 0x00000000\n [64036.654691] Call Trace:\n [64036.654703] ? __schedule+0x3ae/0x7b0\n [64036.654716] schedule+0x3a/0xb0\n [64036.654756] btrfs_find_free_ino+0xa9/0x120 [btrfs]\n [64036.654764] ? remove_wait_queue+0x60/0x60\n [64036.654809] btrfs_create+0x72/0x1f0 [btrfs]\n [64036.654822] lookup_open+0x6bc/0x790\n [64036.654849] path_openat+0x3bc/0xc00\n [64036.654854] ? __lock_acquire+0x331/0x1cb0\n [64036.654869] do_filp_open+0x99/0x110\n [64036.654884] ? __alloc_fd+0xee/0x200\n [64036.654895] ? do_raw_spin_unlock+0x49/0xc0\n [64036.654909] ? do_sys_open+0x132/0x220\n [64036.654913] do_sys_open+0x132/0x220\n [64036.654926] do_syscall_64+0x60/0x1d0\n [64036.654933] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n \n Fix this by adding a wake_up() call right after setting the cache state to\n BTRFS_CACHE_FINISHED, at start_caching(), when we are able to load the\n cache from disk.\n \n Fixes: 82d5902d9c681b (\"Btrfs: Support reading/writing on disk free ino cache\")\n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Filipe Manana \u003cfdmanana@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 2f831ba2854962dd58001e134543bf9518f3041d\nAuthor: Qu Wenruo \u003cwqu@suse.com\u003e\nDate: Tue Jul 16 17:00:34 2019 +0800\n\n btrfs: tree-checker: Add ROOT_ITEM check\n \n This patch will introduce ROOT_ITEM check, which includes:\n - Key-\u003eobjectid and key-\u003eoffset check\n Currently only some easy check, e.g. 0 as rootid is invalid.\n \n - Item size check\n Root item size is fixed.\n \n - Generation checks\n Generation, generation_v2 and last_snapshot should not be greater than\n super generation + 1\n \n - Level and alignment check\n Level should be in [0, 7], and bytenr must be aligned to sector size.\n \n - Flags check\n \n Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id\u003d203261\n Reported-by: Jungyeon Yoon \u003cjungyeon.yoon@gmail.com\u003e\n Signed-off-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 306dabbf16e7b5be0c2fe402275fa618aa70c639\nAuthor: Qu Wenruo \u003cwqu@suse.com\u003e\nDate: Tue Jul 16 17:00:33 2019 +0800\n\n btrfs: extent-tree: Make sure we only allocate extents from block groups with the same type\n \n [BUG]\n With fuzzed image and MIXED_GROUPS super flag, we can hit the following\n BUG_ON():\n \n kernel BUG at fs/btrfs/delayed-ref.c:491!\n invalid opcode: 0000 [#1] PREEMPT SMP NOPTI\n CPU: 0 PID: 1849 Comm: sync Tainted: G O 5.2.0-custom #27\n RIP: 0010:update_existing_head_ref.cold+0x44/0x46 [btrfs]\n Call Trace:\n add_delayed_ref_head+0x20c/0x2d0 [btrfs]\n btrfs_add_delayed_tree_ref+0x1fc/0x490 [btrfs]\n btrfs_free_tree_block+0x123/0x380 [btrfs]\n __btrfs_cow_block+0x435/0x500 [btrfs]\n btrfs_cow_block+0x110/0x240 [btrfs]\n btrfs_search_slot+0x230/0xa00 [btrfs]\n ? __lock_acquire+0x105e/0x1e20\n btrfs_insert_empty_items+0x67/0xc0 [btrfs]\n alloc_reserved_file_extent+0x9e/0x340 [btrfs]\n __btrfs_run_delayed_refs+0x78e/0x1240 [btrfs]\n ? kvm_clock_read+0x18/0x30\n ? __sched_clock_gtod_offset+0x21/0x50\n btrfs_run_delayed_refs.part.0+0x4e/0x180 [btrfs]\n btrfs_run_delayed_refs+0x23/0x30 [btrfs]\n btrfs_commit_transaction+0x53/0x9f0 [btrfs]\n btrfs_sync_fs+0x7c/0x1c0 [btrfs]\n ? __ia32_sys_fdatasync+0x20/0x20\n sync_fs_one_sb+0x23/0x30\n iterate_supers+0x95/0x100\n ksys_sync+0x62/0xb0\n __ia32_sys_sync+0xe/0x20\n do_syscall_64+0x65/0x240\n entry_SYSCALL_64_after_hwframe+0x49/0xbe\n \n [CAUSE]\n This situation is caused by several factors:\n - Fuzzed image\n The extent tree of this fs missed one backref for extent tree root.\n So we can allocated space from that slot.\n \n - MIXED_BG feature\n Super block has MIXED_BG flag.\n \n - No mixed block groups exists\n All block groups are just regular ones.\n \n This makes data space_info-\u003eblock_groups[] contains metadata block\n groups. And when we reserve space for data, we can use space in\n metadata block group.\n \n Then we hit the following file operations:\n \n - fallocate\n We need to allocate data extents.\n find_free_extent() choose to use the metadata block to allocate space\n from, and choose the space of extent tree root, since its backref is\n missing.\n \n This generate one delayed ref head with is_data \u003d 1.\n \n - extent tree update\n We need to update extent tree at run_delayed_ref time.\n \n This generate one delayed ref head with is_data \u003d 0, for the same\n bytenr of old extent tree root.\n \n Then we trigger the BUG_ON().\n \n [FIX]\n The quick fix here is to check block_group-\u003eflags before using it.\n \n The problem can only happen for MIXED_GROUPS fs. Regular filesystems\n won\u0027t have space_info with DATA|METADATA flag, and no way to hit the\n bug.\n \n Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id\u003d203255\n Reported-by: Jungyeon Yoon \u003cjungyeon.yoon@gmail.com\u003e\n Signed-off-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 6968f8f997605f5099d87e083e5754b3fa60b02f\nAuthor: Qu Wenruo \u003cwqu@suse.com\u003e\nDate: Tue Jul 16 17:00:32 2019 +0800\n\n btrfs: delayed-inode: Kill the BUG_ON() in btrfs_delete_delayed_dir_index()\n \n There is one report of fuzzed image which leads to BUG_ON() in\n btrfs_delete_delayed_dir_index().\n \n Although that fuzzed image can already be addressed by enhanced\n extent-tree error handler, it\u0027s still better to hunt down more BUG_ON().\n \n This patch will hunt down two BUG_ON()s in\n btrfs_delete_delayed_dir_index():\n - One for error from btrfs_delayed_item_reserve_metadata()\n Instead of BUG_ON(), we output an error message and free the item.\n And return the error.\n All callers of this function handles the error by aborting current\n trasaction.\n \n - One for possible EEXIST from __btrfs_add_delayed_deletion_item()\n That function can return -EEXIST.\n We already have a good enough error message for that, only need to\n clean up the reserved metadata space and allocated item.\n \n To help above cleanup, also modifiy __btrfs_remove_delayed_item() called\n in btrfs_release_delayed_item(), to skip unassociated item.\n \n Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id\u003d203253\n Signed-off-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 32eede1c73752ffdf14f700fc8eba392ee6ab1ce\nAuthor: Qu Wenruo \u003cwqu@suse.com\u003e\nDate: Fri Jul 19 14:51:44 2019 +0800\n\n btrfs: volumes: Remove ENOSPC-prone btrfs_can_relocate()\n \n [BUG]\n Test case btrfs/156 fails since commit 302167c50b32 (\"btrfs: don\u0027t end\n the transaction for delayed refs in throttle\") with ENOSPC.\n \n [CAUSE]\n The ENOSPC is reported from btrfs_can_relocate().\n \n This function will check:\n - If this block group is empty, we can relocate\n - If we can enough free space, we can relocate\n \n Above checks are valid but the following check is vague due to its\n implementation:\n - If and only if we can allocated a new block group to contain all the\n used space, we can relocate\n \n This design itself is OK, but the way to determine if we can allocate a\n new block group is problematic.\n \n btrfs_can_relocate() uses find_free_dev_extent() to find free space on a\n device.\n However find_free_dev_extent() only searches commit root and excludes\n dev extents allocated in current trans, this makes it unable to use dev\n extent just freed in current transaction.\n \n So for the following example, btrfs_can_relocate() will report ENOSPC:\n The example block group layout:\n 1M 129M 257M 385M 513M 550M\n |///////|///////////|//////////| | |\n // \u003d Used bg, consider all bg is 100% used for easy calculation.\n And all block groups are SINGLE, on-disk bytenr is the same as the\n logical bytenr.\n \n 1) Bg in [129M, 257M) get relocated to [385M, 513M), transid\u003d100\n 1M 129M 257M 385M 513M 550M\n |///////| |//////////|/////////|\n In transid 100, bg in [129M, 257M) get relocated to [385M, 513M)\n \n However transid 100 is not committed yet, so in dev commit tree, we\n still have the old dev extents layout:\n 1M 129M 257M 385M 513M 550M\n |///////|///////////|//////////| | |\n \n 2) Try to relocate bg [257M, 385M)\n We goes into btrfs_can_relocate(), no free space in current bgs, so we\n check if we can find large enough free dev extents.\n \n The first slot is [385M, 513M), but that is already used by new bg at\n [385M, 513M), so we continue search.\n \n The remaining slot is [512M, 550M), smaller than the bg\u0027s length 128M.\n So btrfs_can_relocate report ENOSPC.\n \n However this is over killed, in fact if we just skip btrfs_can_relocate()\n check, and go into regular relocation routine, at extent reservation time,\n if we can\u0027t find free extent, then we fallback to commit transaction,\n which will free up the dev extents and allow new block group to be created.\n \n [FIX]\n The fix here is to remove btrfs_can_relocate() completely.\n \n If we hit the false ENOSPC case just like btrfs/156, extent allocator\n will push harder by committing transaction and we will have space for\n new block group, avoiding the false ENOSPC.\n \n If we really ran out of space, we will hit ENOSPC at\n relocate_block_group(), and btrfs will just reports the ENOSPC error as\n usual.\n \n Signed-off-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit e16a699f9dd4ef1ee6ebfc2707ae4b0c765bc415\nAuthor: Qu Wenruo \u003cwqu@suse.com\u003e\nDate: Fri Jul 19 14:51:43 2019 +0800\n\n btrfs: extent-tree: Add comment for inc_block_group_ro()\n \n inc_block_group_ro() is only designed to mark one block group read-only,\n it doesn\u0027t really care if other block groups have enough free space to\n contain the used space in the block group.\n \n However due to the close connection between this function and\n relocation, sometimes we can be confused and think this function is\n responsible for balance space reservation, which is not true.\n \n Add some comment to make the functionality clear.\n \n Reviewed-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Signed-off-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 6584272c13891c31ba5ac3629e2a5a5ecd6d58de\nAuthor: Qu Wenruo \u003cwqu@suse.com\u003e\nDate: Fri Jul 19 14:51:42 2019 +0800\n\n btrfs: volumes: Add comment for find_free_dev_extent_start()\n \n Since commit 6df9a95e6339 (\"Btrfs: make the chunk allocator completely\n tree lockless\") we search commit root of device tree to avoid deadlock.\n \n This introduced a safety feature, find_free_dev_extent_start() won\u0027t\n use dev extents which just get freed in current transaction.\n \n This safety feature makes sure we won\u0027t allocate new block group using\n just freed dev extents to break CoW.\n \n However, this feature also makes find_free_dev_extent_start() not\n reliable reporting free device space. Just add such comment to make\n later viewer careful about this behavior.\n \n This behavior makes one caller, btrfs_can_relocate() unreliable\n determining the device free space.\n \n Reviewed-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Signed-off-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit d8fc5c5a635ec5c625a3b317ef07206f677d9503\nAuthor: Qu Wenruo \u003cwqu@suse.com\u003e\nDate: Fri Jul 19 14:51:41 2019 +0800\n\n btrfs: volumes: Unexport find_free_dev_extent_start()\n \n This function is only used locally in find_free_dev_extent(), no\n external callers.\n \n So unexport it.\n \n Reviewed-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Signed-off-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 0a04df229d04e96c3fb09545925db23d45af9cae\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Wed Mar 27 16:19:55 2019 +0100\n\n btrfs: assert tree mod log lock in __tree_mod_log_insert\n \n The tree is going to be modified so it must be the exclusive lock.\n \n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 0807dbb23805cb49eebf1ac391b16474beff077b\nAuthor: David Sterba \u003cdsterba@suse.com\u003e\nDate: Wed Mar 27 16:19:55 2019 +0100\n\n btrfs: assert extent map tree lock in add_extent_mapping\n \n As add_extent_mapping is called from several functions, let\u0027s add the\n lock annotation. The tree is going to be modified so it must be the\n exclusive lock.\n \n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 550757fc05a5876f64708c33e95e5f1bf298e217\nAuthor: Jia-Ju Bai \u003cbaijiaju1990@gmail.com\u003e\nDate: Sat Jul 27 16:51:13 2019 +0800\n\n btrfs: Add an assertion to warn incorrect case in insert_inline_extent()\n \n In insert_inline_extent(), the case that checks compressed_size \u003e 0\n and compressed_pages \u003d NULL cannot occur, otherwise a null-pointer\n dereference may occur on line 215:\n \n cpage \u003d compressed_pages[i];\n \n To catch this incorrect case, an assertion is added.\n \n Reviewed-by: Qu Wenruo \u003cwqu@suse.com\u003e\n Signed-off-by: Jia-Ju Bai \u003cbaijiaju1990@gmail.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit e0cb6c11ba1219411306505001019697204efbed\nAuthor: Nikolay Borisov \u003cnborisov@suse.com\u003e\nDate: Wed Jul 17 16:18:17 2019 +0300\n\n btrfs: Remove leftover of in-band dedupe\n \n It\u0027s unlikely in-band dedupe is going to land so just remove any\n leftovers - dedupe.h header as well as the \u0027dedupe\u0027 parameter to\n btrfs_set_extent_delalloc.\n \n Signed-off-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit de005a66e05771f95f4cf037950952056286e1dd\nAuthor: Nikolay Borisov \u003cnborisov@suse.com\u003e\nDate: Wed Jul 17 16:18:16 2019 +0300\n\n btrfs: Remove delalloc_end argument from extent_clear_unlock_delalloc\n \n It was added in ba8b04c1d4ad (\"btrfs: extend btrfs_set_extent_delalloc\n and its friends to support in-band dedupe and subpage size patchset\") as\n a preparatory patch for in-band and subapge block size patchsets.\n However neither of those are likely to be merged anytime soon and the\n code has diverged significantly from the last public post of either\n of those patchsets.\n \n It\u0027s unlikely either of the patchests are going to use those preparatory\n steps so just remove the variables. Since cow_file_range also took\n delalloc_end to pass it to extent_clear_unlock_delalloc remove the\n parameter from that function as well.\n \n Signed-off-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit cf3676525f3140785c4ab6c00fa635a4a1434d9e\nAuthor: Nikolay Borisov \u003cnborisov@suse.com\u003e\nDate: Wed Jul 17 14:41:45 2019 +0300\n\n btrfs: Move free_pages_out label in inline extent handling branch in compress_file_range\n \n This label is only executed if compress_file_range fails to create an\n inline extent. So move its code in the semantically related inline\n extent handling branch. No functional changes.\n \n Signed-off-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 0e304a68c4582e083f9fd12e079b7f1d8d51720a\nAuthor: Nikolay Borisov \u003cnborisov@suse.com\u003e\nDate: Wed Jul 17 14:41:44 2019 +0300\n\n btrfs: Return number of compressed extents directly in compress_file_range\n \n compress_file_range returns a void, yet uses a function parameter as a\n return value. Make that more idiomatic by simply returning the number\n of compressed extents directly. Also track such extents in more aptly\n named variables. No functional changes.\n \n Signed-off-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 7cb8caee599b8054c1a9ef4d7116975ed9a29dbd\nAuthor: Eric Sandeen \u003csandeen@redhat.com\u003e\nDate: Wed Jul 17 12:39:20 2019 -0500\n\n btrfs: use common vfs LABEL ioctl definitions\n \n I lifted the btrfs label get/set ioctls to the vfs some time ago, but\n never followed up to use those common definitions directly in btrfs.\n \n This patch does that.\n \n Signed-off-by: Eric Sandeen \u003csandeen@redhat.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 21f53cd5d635f33105000011939d2559746c2c00\nAuthor: Nikolay Borisov \u003cnborisov@suse.com\u003e\nDate: Thu Jul 25 11:27:28 2019 +0300\n\n btrfs: Remove unused locking functions\n \n Those were split out of btrfs_clear_lock_blocking_rw by\n aa12c02778a9 (\"btrfs: split btrfs_clear_lock_blocking_rw to read and write helpers\")\n however at that time this function was unused due to commit\n 523983401644 (\"Btrfs: kill btrfs_clear_path_blocking\"). Put the final\n nail in the coffin of those 2 functions.\n \n Signed-off-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 233d6097071569478e034152fb456eb6c75239c5\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Mon Jul 8 14:40:09 2019 +0200\n\n btrfs: reduce stack usage for btrfsic_process_written_block\n \n btrfsic_process_written_block() cals btrfsic_process_metablock(),\n which has a fairly large stack usage due to the btrfsic_stack_frame\n variable. It also calls btrfsic_test_for_metadata(), which now\n needs several hundreds of bytes for its SHASH_DESC_ON_STACK().\n \n In some configurations, we end up with both functions on the\n same stack, and gcc warns about the excessive stack usage that\n might cause the available stack space to run out:\n \n fs/btrfs/check-integrity.c:1743:13: error: stack frame size of 1152 bytes in function \u0027btrfsic_process_written_block\u0027 [-Werror,-Wframe-larger-than\u003d]\n \n Marking both child functions as noinline_for_stack helps because\n this guarantees that the large variables are not on the same\n stack frame.\n \n Fixes: d5178578bcd4 (\"btrfs: directly call into crypto framework for checksumming\")\n Reviewed-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit d71379f292f0bd70262acdb2d2dc2fc299e5d796\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Tue Jul 2 22:15:21 2019 +0800\n\n btrfs: remove set but not used variable \u0027offset\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n fs/btrfs/volumes.c: In function __btrfs_map_block:\n fs/btrfs/volumes.c:6023:6: warning:\n variable offset set but not used [-Wunused-but-set-variable]\n \n It is not used any more since commit 343abd1c0ca9 (\"btrfs: Use\n btrfs_get_io_geometry appropriately\")\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit c021ffbd69a1055f477955893e01e6c1b086f1a4\nAuthor: Filipe Manana \u003cfdmanana@suse.com\u003e\nDate: Fri Jul 5 11:09:50 2019 +0100\n\n Btrfs: fix ENOSPC errors, leading to transaction aborts, when cloning extents\n \n When cloning extents (or deduplicating) we create a transaction with a\n space reservation that considers we will drop or update a single file\n extent item of the destination inode (that we modify a single leaf). That\n is fine for the vast majority of scenarios, however it might happen that\n we need to drop many file extent items, and adjust at most two file extent\n items, in the destination root, which can span multiple leafs. This will\n lead to either the call to btrfs_drop_extents() to fail with ENOSPC or\n the subsequent calls to btrfs_insert_empty_item() or btrfs_update_inode()\n (called through clone_finish_inode_update()) to fail with ENOSPC. Such\n failure results in a transaction abort, leaving the filesystem in a\n read-only mode.\n \n In order to fix this we need to follow the same approach as the hole\n punching code, where we create a local reservation with 1 unit and keep\n ending and starting transactions, after balancing the btree inode,\n when __btrfs_drop_extents() returns ENOSPC. So fix this by making the\n extent cloning call calls the recently added btrfs_punch_hole_range()\n helper, which is what does the mentioned work for hole punching, and\n make sure whenever we drop extent items in a transaction, we also add a\n replacing file extent item, to avoid corruption (a hole) if after ending\n a transaction and before starting a new one, the old transaction gets\n committed and a power failure happens before we finish cloning.\n \n A test case for fstests follows soon.\n \n Reported-by: David Goodwin \u003cdavid@codepoets.co.uk\u003e\n Link: https://lore.kernel.org/linux-btrfs/a4a4cf31-9cf4-e52c-1f86-c62d336c9cd1@codepoets.co.uk/\n Reported-by: Sam Tygier \u003csam@tygier.co.uk\u003e\n Link: https://lore.kernel.org/linux-btrfs/82aace9f-a1e3-1f0b-055f-3ea75f7a41a0@tygier.co.uk/\n Fixes: b6f3409b2197e8f (\"Btrfs: reserve sufficient space for ioctl clone\")\n Signed-off-by: Filipe Manana \u003cfdmanana@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 7315b805eb617b8dce8db6f0399dd276dc43aee2\nAuthor: Filipe Manana \u003cfdmanana@suse.com\u003e\nDate: Fri Jun 28 23:11:26 2019 +0100\n\n Btrfs: factor out extent dropping code from hole punch handler\n \n Move the code that is responsible for dropping extents in a range out of\n btrfs_punch_hole() into a new helper function, btrfs_punch_hole_range(),\n so that later it can be used by the reflinking (extent cloning and dedup)\n code to fix a ENOSPC bug.\n \n Signed-off-by: Filipe Manana \u003cfdmanana@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 0465814831a926ce2f83e8f606d067d86745234e\nAuthor: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\nDate: Mon Aug 26 21:37:37 2019 +0900\n\n PM / devfreq: passive: fix compiler warning\n \n The recent commit of\n PM / devfreq: passive: Use non-devm notifiers\n had incurred compiler warning, \"unused variable \u0027dev\u0027\".\n \n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 85af79659f455ddbdc5b6b6e2a888a430cb58948\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Thu Aug 22 11:42:52 2019 +0300\n\n MAINTAINERS: Remove stale record for gpio-intel-mid.c\n \n David Cohen seems left Intel few years back. Remove the stale record in\n MAINTAINERS data base. The file is anyway listed under my maintainership.\n \n Reported-by: Dave Hansen \u003cdave.hansen@intel.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 60849562a5db4a1eee2160167e4dce4590d3eafe\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Mon Aug 26 13:55:21 2019 +0200\n\n ALSA: usb-audio: Fix possible NULL dereference at create_yamaha_midi_quirk()\n \n The previous addition of descriptor validation may lead to a NULL\n dereference at create_yamaha_midi_quirk() when either injd or outjd is\n NULL. Add proper non-NULL checks.\n \n Fixes: 57f8770620e9 (\"ALSA: usb-audio: More validations of descriptor units\")\n Reported-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 14ced7e3a1ae9bed7051df3718c8c7b583854a5c\nAuthor: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\nDate: Mon Aug 5 17:42:56 2019 -0700\n\n phy: qcom-qmp: Correct ready status, again\n \n Despite extensive testing of commit 885bd765963b (\"phy: qcom-qmp: Correct\n READY_STATUS poll break condition\") I failed to conclude that the\n PHYSTATUS bit of the PCS_STATUS register used in PCIe and USB3 falls as\n the PHY gets ready. Similar to the prior bug with UFS the code will\n generally get past the check before the transition and thereby\n \"succeed\".\n \n Correct the name of the register used PCIe and USB3 PHYs, replace\n mask_pcs_ready with a constant expression depending on the type of the\n PHY and check for the appropriate ready state.\n \n Cc: stable@vger.kernel.org\n Cc: Vivek Gautam \u003cvivek.gautam@codeaurora.org\u003e\n Cc: Evan Green \u003cevgreen@chromium.org\u003e\n Cc: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n Reported-by: Marc Gonzalez \u003cmarc.w.gonzalez@free.fr\u003e\n Fixes: 885bd765963b (\"phy: qcom-qmp: Correct READY_STATUS poll break condition\")\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Tested-by: Marc Gonzalez \u003cmarc.w.gonzalez@free.fr\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit be0345b2cc1f3e6044409b274c61bc44d59f640d\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 8 12:59:37 2019 +0530\n\n phy: qualcomm: phy-qcom-qmp: Add of_node_put() before return\n \n Each iteration of for_each_available_child_of_node() puts the previous\n node, but in the case of a return from the middle of the loop, there is\n no put, thus causing a memory leak. Hence create a new label,\n err_node_put, that puts the previous node (child) before returning the\n required value. Also include the statement pm_runtime_disable() under\n this label in order to avoid repetition among mid-loop return\n conditions. Edit the mid-loop return statements to instead go to this\n new label err_node_put.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit e6839c31a608e79f2057fab987dd814f5d3477e6\nAuthor: Yoshihiro Shimoda \u003cyoshihiro.shimoda.uh@renesas.com\u003e\nDate: Tue Aug 6 17:51:19 2019 +0900\n\n phy: renesas: rcar-gen3-usb2: Disable clearing VBUS in over-current\n \n The hardware manual should be revised, but the initial value of\n VBCTRL.OCCLREN is set to 1 actually. If the bit is set, the hardware\n clears VBCTRL.VBOUT and ADPCTRL.DRVVBUS registers automatically\n when the hardware detects over-current signal from a USB power switch.\n However, since the hardware doesn\u0027t have any registers which\n indicates over-current, the driver cannot handle it at all. So, if\n \"is_otg_channel\" hardware detects over-current, since ADPCTRL.DRVVBUS\n register is cleared automatically, the channel cannot be used after\n that.\n \n To resolve this behavior, this patch sets the VBCTRL.OCCLREN to 0\n to keep ADPCTRL.DRVVBUS even if the \"is_otg_channel\" hardware\n detects over-current. (We assume a USB power switch itself protects\n over-current and turns the VBUS off.)\n \n This patch is inspired by a BSP patch from Kazuya Mizuguchi.\n \n Fixes: 1114e2d31731 (\"phy: rcar-gen3-usb2: change the mode to OTG on the combined channel\")\n Cc: \u003cstable@vger.kernel.org\u003e # v4.5+\n Signed-off-by: Yoshihiro Shimoda \u003cyoshihiro.shimoda.uh@renesas.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit 3e644828421e31167bedd8dc4d291cdcc822ab20\nAuthor: Wen Yang \u003cwen.yang99@zte.com.cn\u003e\nDate: Mon Jul 8 14:19:05 2019 +0800\n\n phy: ti: am654-serdes: fix an use-after-free in serdes_am654_clk_register()\n \n The regmap_node variable is still being used in the syscon_node_to_regmap()\n call after the of_node_put() call, which may result in use-after-free.\n \n Fixes: 71e2f5c5c224 (\"phy: ti: Add a new SERDES driver for TI\u0027s AM654x SoC\")\n Signed-off-by: Wen Yang \u003cwen.yang99@zte.com.cn\u003e\n Cc: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n Cc: Roger Quadros \u003crogerq@ti.com\u003e\n Reviewed-by: Roger Quadros \u003crogerq@ti.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit 38b7b678fe989f9c403c001d96887939aaa1b68a\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 22 14:35:55 2019 -0300\n\n perf stat: Remove needless headers from stat.h\n \n Just a forward declaration for \u0027struct timespec\u0027 is needed, ditch the\n rest.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-6shdqw801oqe7ax6r307k27r@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 29e331176da121db22daa5dd9d72533b6a1efb33\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 22 14:26:31 2019 -0300\n\n perf cpumap: No need to include perf.h, ditch it\n \n From a quick look this was never needed and just polluted the build,\n needlessly making things including cpumap.h to be rebuild if perf.h or\n anything it includes gets changed.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-x10p8slllqkn3fc3bntjx3n0@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 73345a18d464b1b945b29f54f630ace6873344e2\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Mon Aug 12 08:27:29 2019 +0200\n\n pinctrl: bcm2835: Pass irqchip when adding gpiochip\n \n We need to convert all old gpio irqchips to pass the irqchip\n setup along when adding the gpio_chip. For more info see\n drivers/gpio/TODO.\n \n For chained irqchips this is a pretty straight-forward\n conversion. The BCM2835 has multiple parents so let\u0027s\n exploit the new facility in the GPIO_IRQCHIP to actually\n deal with multiple parents.\n \n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n [Rebased on changes in the pinctrl tree]\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Tested-by: Stefan Wahren \u003cstefan.wahren@i2se.com\u003e\n Link: https://lore.kernel.org/r/20190812062729.1892-1-linus.walleij@linaro.org\n\ncommit 5f597a04c8530d1e56ad884822e201c64bfdf3ba\nAuthor: Jorge Ramirez-Ortiz \u003cjorge.ramirez-ortiz@linaro.org\u003e\nDate: Thu Jul 25 12:41:36 2019 +0200\n\n cpufreq: Add qcs404 to cpufreq-dt-platdev blacklist\n \n Add qcs404 to cpufreq-dt-platdev blacklist.\n \n Signed-off-by: Jorge Ramirez-Ortiz \u003cjorge.ramirez-ortiz@linaro.org\u003e\n Co-developed-by: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n Signed-off-by: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit f67e8e25bda887328aed7fc356c6435d4c29a778\nAuthor: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\nDate: Thu Jul 25 12:41:35 2019 +0200\n\n cpufreq: qcom: Add support for qcs404 on nvmem driver\n \n Add support for qcs404 on nvmem driver.\n \n The qcs404 SoC has support for Core Power Reduction (CPR), which is\n implemented as a power domain provider, therefore add optional support\n in this driver to attach to a genpd power domain.\n \n Co-developed-by: Jorge Ramirez-Ortiz \u003cjorge.ramirez-ortiz@linaro.org\u003e\n Signed-off-by: Jorge Ramirez-Ortiz \u003cjorge.ramirez-ortiz@linaro.org\u003e\n Signed-off-by: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit 21afe0c258dc89a586c96abb3943bbdeaf032d53\nAuthor: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\nDate: Thu Jul 25 12:41:33 2019 +0200\n\n cpufreq: qcom: Refactor the driver to make it easier to extend\n \n Refactor the driver to make it easier to extend in a later commit.\n \n Create a driver struct to collect all common resources, in order to make\n it easier to free up all common resources.\n Create a driver match_data struct to make it easier to extend the driver\n with support for new features that might only be supported on certain SoCs.\n \n Co-developed-by: Jorge Ramirez-Ortiz \u003cjorge.ramirez-ortiz@linaro.org\u003e\n Signed-off-by: Jorge Ramirez-Ortiz \u003cjorge.ramirez-ortiz@linaro.org\u003e\n Signed-off-by: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n Reviewed-by: Ilia Lin \u003cilia.lin@kernel.org\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit 20b8096ae8c6d47b18de27ea479cf23bd3123fd3\nAuthor: Sricharan R \u003csricharan@codeaurora.org\u003e\nDate: Thu Jul 25 12:41:31 2019 +0200\n\n cpufreq: qcom: Re-organise kryo cpufreq to use it for other nvmem based qcom socs\n \n The kryo cpufreq driver reads the nvmem cell and uses that data to\n populate the opps. There are other qcom cpufreq socs like krait which\n does similar thing. Except for the interpretation of the read data,\n rest of the driver is same for both the cases. So pull the common things\n out for reuse.\n \n Signed-off-by: Sricharan R \u003csricharan@codeaurora.org\u003e\n [niklas.cassel@linaro.org: split dt-binding into a separate patch and\n do not rename the compatible string. Update MAINTAINERS file.]\n Signed-off-by: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n Reviewed-by: Ilia Lin \u003cilia.lin@kernel.org\u003e\n Reviewed-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit 021117dd8815a15428a8a81e269db9a6f7f78bbc\nAuthor: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\nDate: Mon Aug 19 12:12:38 2019 +0200\n\n dt-bindings: opp: Add qcom-opp bindings with properties needed for CPR\n \n Add qcom-opp bindings with properties needed for Core Power Reduction\n (CPR).\n \n CPR is included in a great variety of Qualcomm SoCs, e.g. msm8916 and\n msm8996. CPR was first introduced in msm8974.\n \n Co-developed-by: Jorge Ramirez-Ortiz \u003cjorge.ramirez-ortiz@linaro.org\u003e\n Signed-off-by: Jorge Ramirez-Ortiz \u003cjorge.ramirez-ortiz@linaro.org\u003e\n Signed-off-by: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n Reviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit 5ebc6bedc4990104e1f8cdf34cc67d95b7489411\nAuthor: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\nDate: Mon Aug 19 12:09:57 2019 +0200\n\n dt-bindings: opp: qcom-nvmem: Support pstates provided by a power domain\n \n Some Qualcomm SoCs have support for Core Power Reduction (CPR).\n On these platforms, we need to attach to the power domain provider\n providing the performance states, so that the leaky device (the CPU)\n can configure the performance states (which represent different\n CPU clock frequencies).\n \n Signed-off-by: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit 5699ad0aaf1091824f22492a708478912c38c7d2\nAuthor: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\nDate: Tue May 14 10:38:15 2019 -0700\n\n firmware: bcm47xx_nvram: Allow COMPILE_TEST\n \n Allow building building the BCM47xx NVRAM and SPROM drivers using\n COMPILE_TEST.\n \n Signed-off-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-mips@linux-mips.org\n Cc: joe@perches.com\n Cc: Rafał Miłecki \u003czajec5@gmail.com\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n\ncommit 759d4899d905c2a491d34987366421aebca359de\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Mon Aug 19 23:19:00 2019 +0200\n\n s390/kaslr: reserve memory for kasan usage\n \n Sometimes the kernel fails to boot with:\n \"The Linux kernel failed to boot with the KernelAddressSanitizer:\n out of memory during initialisation\" even with big amounts of memory when\n both kaslr and kasan are enabled.\n \n The problem is that kasan initialization code requires 1/8 of physical\n memory plus some for page tables. To keep as much code instrumented\n as possible kasan avoids using memblock for memory allocations. Instead\n kasan uses trivial memory allocator which simply chops off the memory\n from the end of online physical memory. For that reason when kaslr is\n enabled together with kasan avoid positioning kernel into upper memory\n region which would be utilized during kasan initialization.\n \n Reviewed-by: Gerald Schaefer \u003cgerald.schaefer@de.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 9b692102d87d13d24c97461207cae259121ed44b\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Mon Aug 19 23:14:10 2019 +0200\n\n s390/mem_detect: provide single get_mem_detect_end\n \n get_mem_detect_end is already used in couple of places with potential\n to be utilized in more cases. Provide single get_mem_detect_end\n implementation in asm/mem_detect.h to be used by kasan and startup code.\n \n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 5f0917a281c66fb310988d4c38f3c55b7023b8ff\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Mon Aug 19 17:55:44 2019 +0200\n\n s390/cmma: reuse kstrtobool for option value parsing\n \n \"cmma\" option setup already recognises some textual values. Yet kstrtobool\n is a more common way to parse boolean values, reuse it to unify option\n value parsing behavior and simplify code a bit.\n \n While at it, __setup value parsing callbacks are expected to return\n 1 when an option is recognized, and returning any other value won\u0027t\n trigger any error message currently, so simply return 1.\n \n Reviewed-by: Philipp Rudo \u003cprudo@linux.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 3d644364533931df298dc4f026a74731c2f752cb\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Mon Aug 19 17:41:17 2019 +0200\n\n s390/vdso: reuse kstrtobool for option value parsing\n \n \"vdso\" option setup already recognises integer and textual values. Yet\n kstrtobool is a more common way to parse boolean values, reuse it to\n unify option value parsing behavior and simplify code a bit.\n \n While at it, __setup value parsing callbacks are expected to return\n 1 when an option is recognized, and returning any other value won\u0027t\n trigger any error message currently, so simply return 1.\n \n Also don\u0027t change default vdso_enabled value of 1 when \"vdso\" option\n value is invalid.\n \n Reviewed-by: Philipp Rudo \u003cprudo@linux.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 227f52a43a2fa0bb50f07faa2d5e31530a740499\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Mon Aug 19 10:49:53 2019 +0200\n\n s390/startup: correct command line options parsing\n \n Check val is not NULL before accessing it. This might happen if\n corresponding kernel command line options are used without specifying\n values.\n \n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit a45425d8a5a672d3322164cfce10b7b6fb32fcc2\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Mon Aug 19 10:44:54 2019 +0200\n\n s390/vmcp: correct early_param handling\n \n Check \"p\" is not NULL before passing it to memparse, which doesn\u0027t\n handle that case explicitly.\n \n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 915ef7bda76466a5542a76694c08709895383f1c\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Fri Aug 16 15:49:49 2019 +0200\n\n s390/numa: correct early_param handling\n \n When command line options are used without specifying values\n (e.g. \"emu_size\" instead of \"emu_size\u003d\"), the value is NULL. Check that\n before performing string operations and further processing.\n \n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit feb4eb060c3aecc3c5076bebe699cd09f1133c41\nAuthor: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\nDate: Tue May 14 10:38:14 2019 -0700\n\n firmware: bcm47xx_nvram: Correct size_t printf format\n \n When building on a 64-bit host, we will get warnings like those:\n \n drivers/firmware/broadcom/bcm47xx_nvram.c:103:3: note: in expansion of macro \u0027pr_err\u0027\n pr_err(\"nvram on flash (%i bytes) is bigger than the reserved space in memory, will just copy the first %i bytes\\n\",\n ^~~~~~\n drivers/firmware/broadcom/bcm47xx_nvram.c:103:28: note: format string is defined here\n pr_err(\"nvram on flash (%i bytes) is bigger than the reserved space in memory, will just copy the first %i bytes\\n\",\n ~^\n %li\n \n Use %zu instead for that purpose.\n \n Signed-off-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Reviewed-by: Philippe Mathieu-Daudé \u003cf4bug@amsat.org\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-mips@linux-mips.org\n Cc: joe@perches.com\n Cc: Rafał Miłecki \u003czajec5@gmail.com\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n\ncommit d2f965549006acb865c4638f1f030ebcefdc71f6\nAuthor: Jiaxun Yang \u003cjiaxun.yang@flygoat.com\u003e\nDate: Wed May 29 16:42:59 2019 +0800\n\n MIPS: Treat Loongson Extensions as ASEs\n \n Recently, binutils had split Loongson-3 Extensions into four ASEs:\n MMI, CAM, EXT, EXT2. This patch do the samething in kernel and expose\n them in cpuinfo so applications can probe supported ASEs at runtime.\n \n Signed-off-by: Jiaxun Yang \u003cjiaxun.yang@flygoat.com\u003e\n Cc: Huacai Chen \u003cchenhc@lemote.com\u003e\n Cc: Yunqiang Su \u003cysu@wavecomp.com\u003e\n Cc: stable@vger.kernel.org # v4.14+\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-mips@vger.kernel.org\n\ncommit 322e577b02ab096aaf630caae93f18e1e67b775f\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:08 2019 -0700\n\n MIPS: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-kernel@vger.kernel.org\n\ncommit 63ad05e36ec5afc92c6b787c2e419d1ae3d0c8e6\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Fri Aug 23 08:12:59 2019 -0300\n\n media: cec-notifier: clear cec_adap in cec_notifier_unregister\n \n If cec_notifier_cec_adap_unregister() is called before\n cec_unregister_adapter() then everything is OK (and this is the\n case today). But if it is the other way around, then\n cec_notifier_unregister() is called first, and that doesn\u0027t\n set n-\u003ecec_adap to NULL.\n \n So if e.g. cec_notifier_set_phys_addr() is called after\n cec_notifier_unregister() but before cec_unregister_adapter()\n then n-\u003ecec_adap points to an unregistered and likely deleted\n cec adapter. So just set n-\u003ecec_adap-\u003enotifier and n-\u003ecec_adap\n to NULL for rubustness.\n \n Eventually cec_notifier_unregister will disappear and this will\n be simplified substantially.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 1f40706194054a38e75708122096ea5f0bfbb0d1\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Fri Aug 23 08:07:21 2019 -0300\n\n media: cec-adap: return from cec_s_conn_info() if adap is invalid\n \n Check if cec_s_conn_info is called with a valid cec adapter,\n do nothing if it is invalid.\n \n This makes it possible to call this function even if CEC support is\n disabled in the kernel config.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 39152ee51b77851689f9b23fde6f610d13566c39\nAuthor: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\nDate: Wed Aug 21 15:47:27 2019 +0300\n\n perf/x86/intel/pt: Get rid of reverse lookup table for ToPA\n \n In order to quickly find a ToPA entry by its page offset in the buffer,\n we\u0027re using a reverse lookup table. The problem with it is that it\u0027s a\n large array of mostly similar pointers, especially so now that we\u0027re\n using high order allocations from the page allocator. Because its size\n is limited to whatever is the maximum for kmalloc(), it places a limit\n on the number of ToPA entries per buffer, and therefore, on the total\n buffer size, which otherwise doesn\u0027t have to be there.\n \n Replace the reverse lookup table with a simple runtime lookup. With the\n high order AUX allocations in place, the runtime penalty of such a lookup\n is much smaller and in cases where all entries in a ToPA table are of\n the same size, the complexity is O(1).\n \n Signed-off-by: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Stephane Eranian \u003ceranian@google.com\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Vince Weaver \u003cvincent.weaver@maine.edu\u003e\n Link: http://lkml.kernel.org/r/20190821124727.73310-7-alexander.shishkin@linux.intel.com\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 91feca5e2ecc9752894d57c9a72c2645471929c3\nAuthor: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\nDate: Wed Aug 21 15:47:26 2019 +0300\n\n perf/x86/intel/pt: Free up space in a ToPA descriptor\n \n Currently, we\u0027re storing physical address of a ToPA table in its\n descriptor, which is completely unnecessary. Since the descriptor\n and the table itself share the same page, reducing the descriptor\n size leaves more space for the table.\n \n Signed-off-by: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Stephane Eranian \u003ceranian@google.com\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Vince Weaver \u003cvincent.weaver@maine.edu\u003e\n Link: http://lkml.kernel.org/r/20190821124727.73310-6-alexander.shishkin@linux.intel.com\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 38bb8d77d0b932a0773b5de2ef42479409314f96\nAuthor: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\nDate: Wed Aug 21 15:47:25 2019 +0300\n\n perf/x86/intel/pt: Split ToPA metadata and page layout\n \n PT uses page sized ToPA tables, where the ToPA table resides at the bottom\n and its driver-specific metadata taking up a few words at the top of the\n page. The split is currently calculated manually and needs to be redone\n every time a field is added to or removed from the metadata structure.\n Also, the 32-bit version can be made smaller.\n \n By splitting the table and metadata into separate structures, we are making\n the compiler figure out the division of the page.\n \n Signed-off-by: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Stephane Eranian \u003ceranian@google.com\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Vince Weaver \u003cvincent.weaver@maine.edu\u003e\n Link: http://lkml.kernel.org/r/20190821124727.73310-5-alexander.shishkin@linux.intel.com\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 539f7c26b41d4ed7d88dd9756de3966ae7ca07b4\nAuthor: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\nDate: Wed Aug 21 15:47:24 2019 +0300\n\n perf/x86/intel/pt: Use pointer arithmetics instead in ToPA entry calculation\n \n Currently, pt_buffer_reset_offsets() calculates the current ToPA entry by\n casting pointers to addresses and performing ungainly subtractions and\n divisions instead of a simpler pointer arithmetic, which would be perfectly\n applicable in that case. Fix that.\n \n Signed-off-by: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Stephane Eranian \u003ceranian@google.com\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Vince Weaver \u003cvincent.weaver@maine.edu\u003e\n Link: http://lkml.kernel.org/r/20190821124727.73310-4-alexander.shishkin@linux.intel.com\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit fffec50f541ace292383c0cbe9a2a97d16d201c6\nAuthor: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\nDate: Wed Aug 21 15:47:23 2019 +0300\n\n perf/x86/intel/pt: Use helpers to obtain ToPA entry size\n \n There are a few places in the PT driver that need to obtain the size of\n a ToPA entry, some of them for the current ToPA entry in the buffer.\n Use helpers for those, to make the lines shorter and more readable.\n \n Signed-off-by: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Stephane Eranian \u003ceranian@google.com\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Vince Weaver \u003cvincent.weaver@maine.edu\u003e\n Link: http://lkml.kernel.org/r/20190821124727.73310-3-alexander.shishkin@linux.intel.com\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 90583af61d0c0d2826f42a297a03645b35c49085\nAuthor: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\nDate: Wed Aug 21 15:47:22 2019 +0300\n\n perf/x86/intel/pt: Clean up ToPA allocation path\n \n Some of the allocation parameters are passed as function arguments,\n while the CPU number for per-cpu allocation is passed via the buffer\n object. There\u0027s no reason for this.\n \n Pass the CPU as a function argument instead.\n \n Signed-off-by: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Stephane Eranian \u003ceranian@google.com\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Vince Weaver \u003cvincent.weaver@maine.edu\u003e\n Link: http://lkml.kernel.org/r/20190821124727.73310-2-alexander.shishkin@linux.intel.com\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 149f3b87840e7d292ad059f5fc23f1fa2fc98b9e\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Fri Aug 23 23:32:27 2019 +0300\n\n device property: Remove duplicate test for NULL\n \n There is no need to check twice for a NULL in fwnode_call_bool_op().\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 78cd4bf53635d3aeb73435bc89eb6eb39450f315\nAuthor: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\nDate: Mon Aug 19 13:07:24 2019 +0300\n\n platform/x86: intel_cht_int33fe: Use new API to gain access to the role switch\n \n The driver for the Intel USB role mux now always supplies\n software node for the role switch, so no longer checking\n that, and never creating separate node for the role switch.\n From now on using software_node_find_by_name() function to\n get the handle to the USB role switch.\n \n Signed-off-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\n Acked-by: Andy Shevchenko \u003candy.shevchenko@gmail.com\u003e\n Reviewed-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Tested-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit d2a90ebb65536a6deeb9daf5aa8e0700e8cbb43a\nAuthor: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\nDate: Mon Aug 19 13:07:23 2019 +0300\n\n usb: roles: intel_xhci: Supplying software node for the role mux\n \n The primary purpose for this node will be to allow linking\n the users of the switch to it. The users will be for example\n USB Type-C connectors. By supplying a reference to this\n node in the software nodes representing the USB Type-C\n controllers or connectors, the drivers for those devices can\n access the switch.\n \n Signed-off-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\n Reviewed-by: Andy Shevchenko \u003candy.shevchenko@gmail.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 8cbd9af9d208b1f015cf8a4645602f0a007270a8\nAuthor: Jan Kara \u003cjack@suse.cz\u003e\nDate: Mon Aug 26 11:36:19 2019 +0200\n\n udf: Use dynamic debug infrastructure\n \n Instead of relying on UDFFS_DEBUG define for debug printing, just use\n standard pr_debug() prints and rely on CONFIG_DYNAMIC_DEBUG\n infrastructure for enabling or disabling prints.\n \n Signed-off-by: Jan Kara \u003cjack@suse.cz\u003e\n\ncommit 1666faedb567d03cde1d656ae24c6cc253e67373\nAuthor: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\nDate: Mon Aug 19 13:07:22 2019 +0300\n\n software node: Add software_node_find_by_name()\n \n Function that searches software nodes by node name.\n \n Signed-off-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\n Reviewed-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Tested-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Reviewed-by: Andy Shevchenko \u003candy.shevchenko@gmail.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit a8ef0488cc592921a917362cca66af4a601987b9\nAuthor: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\nDate: Wed Aug 14 09:24:44 2019 -0300\n\n media: imx: add csc/scaler mem2mem device\n \n Add a single imx-media mem2mem video device that uses the IPU IC PP\n (image converter post processing) task for scaling and colorspace\n conversion.\n On i.MX6Q/DL SoCs with two IPUs currently only the first IPU is used.\n \n The hardware only supports writing to destination buffers up to\n 1024x1024 pixels in a single pass, arbitrary sizes can be achieved\n by rendering multiple tiles per frame.\n \n [slongerbeam@gmail.com: use ipu_image_convert_adjust(), fix\n device_run() error handling, add missing media-device header,\n unregister and remove the mem2mem device in error paths in\n imx_media_probe_complete() and in imx_media_remove(), updated\n for sync subdev registration]\n \n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Signed-off-by: Steve Longerbeam \u003cslongerbeam@gmail.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n [hverkuil-cisco@xs4all.nl: correct two minor checkpatch issues]\n [hverkuil-cisco@xs4all.nl: sparse warning: make imx6_media_probe_complete static]\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 8ccff2843fb4e6d9d26e5ae9ffe9840b38b92638\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Sun Aug 25 17:35:42 2019 +0200\n\n HID: logitech-dj: Fix crash when initial logi_dj_recv_query_paired_devices fails\n \n Before this commit dj_probe would exit with an error if the initial\n logi_dj_recv_query_paired_devices fails. The initial call may fail\n when the receiver is connected through a kvm and the focus is away.\n \n When the call fails this causes 2 problems:\n \n 1) dj_probe calls logi_dj_recv_query_paired_devices after calling\n hid_device_io_start() so a HID report may have been received in between\n and our delayedwork_callback may be running. It seems that the initial\n logi_dj_recv_query_paired_devices failure happening with some KVMs triggers\n this exact scenario, causing the work-queue to run on free-ed memory,\n leading to:\n \n BUG: unable to handle page fault for address: 0000000000001e88\n #PF: supervisor read access in kernel mode\n #PF: error_code(0x0000) - not-present page\n PGD 0 P4D 0\n Oops: 0000 [#1] SMP PTI\n CPU: 3 PID: 257 Comm: kworker/3:3 Tainted: G OE 5.3.0-rc5+ #100\n Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./B150M Pro4S/D3, BIOS P7.10 12/06/2016\n Workqueue: events 0xffffffffc02ba200\n RIP: 0010:0xffffffffc02ba1bd\n Code: e8 e8 13 00 d8 48 89 c5 48 85 c0 74 4c 48 8b 7b 10 48 89 ea b9 07 00 00 00 41 b9 09 00 00 00 41 b8 01 00 00 00 be 10 00 00 00 \u003c48\u003e 8b 87 88 1e 00 00 48 8b 40 40 e8 b3 6b b4 d8 48 89 ef 41 89 c4\n RSP: 0018:ffffb760c046bdb8 EFLAGS: 00010286\n RAX: ffff935038ea4550 RBX: ffff935046778000 RCX: 0000000000000007\n RDX: ffff935038ea4550 RSI: 0000000000000010 RDI: 0000000000000000\n RBP: ffff935038ea4550 R08: 0000000000000001 R09: 0000000000000009\n R10: 000000000000e011 R11: 0000000000000001 R12: ffff9350467780e8\n R13: ffff935046778000 R14: 0000000000000000 R15: ffff935046778070\n FS: 0000000000000000(0000) GS:ffff935054e00000(0000) knlGS:0000000000000000\n CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n CR2: 0000000000001e88 CR3: 000000075a612002 CR4: 00000000003606e0\n DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\n DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\n Call Trace:\n 0xffffffffc02ba2f7\n ? process_one_work+0x1b1/0x560\n process_one_work+0x234/0x560\n worker_thread+0x50/0x3b0\n kthread+0x10a/0x140\n ? process_one_work+0x560/0x560\n ? kthread_park+0x80/0x80\n ret_from_fork+0x3a/0x50\n Modules linked in: vboxpci(O) vboxnetadp(O) vboxnetflt(O) vboxdrv(O) bnep vfat fat btusb btrtl btbcm btintel bluetooth intel_rapl_msr ecdh_generic rfkill ecc snd_usb_audio snd_usbmidi_lib intel_rapl_common snd_rawmidi mc x86_pkg_temp_thermal intel_powerclamp coretemp iTCO_wdt iTCO_vendor_support mei_wdt mei_hdcp ppdev kvm_intel kvm irqbypass crct10dif_pclmul crc32_generic crc32_pclmul snd_hda_codec_hdmi snd_hda_codec_realtek snd_hda_codec_generic ledtrig_audio ghash_clmulni_intel intel_cstate snd_hda_intel snd_hda_codec intel_uncore snd_hda_core snd_hwdep intel_rapl_perf snd_seq snd_seq_device snd_pcm snd_timer intel_wmi_thunderbolt snd e1000e soundcore mxm_wmi i2c_i801 bfq mei_me mei intel_pch_thermal parport_pc parport acpi_pad binfmt_misc hid_lg_g15(E) hid_logitech_dj(E) i915 crc32c_intel i2c_algo_bit drm_kms_helper nvme nvme_core drm wmi video uas usb_storage i2c_dev\n CR2: 0000000000001e88\n ---[ end trace 1d3f8afdcfcbd842 ]---\n \n 2) Even if we were to fix 1. by making sure the work is stopped before\n failing probe, failing probe is the wrong thing to do, we have\n logi_dj_recv_queue_unknown_work to deal with the initial\n logi_dj_recv_query_paired_devices failure.\n \n Rather then error-ing out of the probe, causing the receiver to not work at\n all we should rely on this, so that the attached devices will get properly\n enumerated once the KVM focus is switched back.\n \n Cc: stable@vger.kernel.org\n Fixes: 74808f9115ce (\"HID: logitech-dj: add support for non unifying receivers\")\n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\n\ncommit 14d5511691e5290103bc480998bc322e68f139d4\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Fri Aug 23 08:12:59 2019 -0300\n\n media: cec-notifier: clear cec_adap in cec_notifier_unregister\n \n If cec_notifier_cec_adap_unregister() is called before\n cec_unregister_adapter() then everything is OK (and this is the\n case today). But if it is the other way around, then\n cec_notifier_unregister() is called first, and that doesn\u0027t\n set n-\u003ecec_adap to NULL.\n \n So if e.g. cec_notifier_set_phys_addr() is called after\n cec_notifier_unregister() but before cec_unregister_adapter()\n then n-\u003ecec_adap points to an unregistered and likely deleted\n cec adapter. So just set n-\u003ecec_adap-\u003enotifier and n-\u003ecec_adap\n to NULL for rubustness.\n \n Eventually cec_notifier_unregister will disappear and this will\n be simplified substantially.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 8f4b5b0af7e451a4c6698ad3441dde8d1286c80d\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Fri Aug 23 08:07:21 2019 -0300\n\n media: cec-adap: return from cec_s_conn_info() if adap is invalid\n \n Check if cec_s_conn_info is called with a valid cec adapter,\n do nothing if it is invalid.\n \n This makes it possible to call this function even if CEC support is\n disabled in the kernel config.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit ab9a3a737284b3d9e1d2ba43a0ef31b3ef2e2417\nAuthor: Steven J. Magnani \u003csteve.magnani@digidescorp.com\u003e\nDate: Wed Aug 14 07:50:02 2019 -0500\n\n udf: reduce leakage of blocks related to named streams\n \n Windows is capable of creating UDF files having named streams.\n One example is the \"Zone.Identifier\" stream attached automatically\n to files downloaded from a network. See:\n https://msdn.microsoft.com/en-us/library/dn392609.aspx\n \n Modification of a file having one or more named streams in Linux causes\n the stream directory to become detached from the file, essentially leaking\n all blocks pertaining to the file\u0027s streams.\n \n Fix by saving off information about an inode\u0027s streams when reading it,\n for later use when its on-disk data is updated.\n \n Link: https://lore.kernel.org/r/20190814125002.10869-1-steve@digidescorp.com\n Signed-off-by: Steven J. Magnani \u003csteve@digidescorp.com\u003e\n Signed-off-by: Jan Kara \u003cjack@suse.cz\u003e\n\ncommit dfda204198848b47bdb98ab83b94dbb7c7692b55\nAuthor: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\nDate: Tue May 14 16:56:21 2019 +0300\n\n ACPI / property: Add two new Thunderbolt property GUIDs to the list\n \n Ice Lake Thunderbolt controller includes two new device property\n compatible properties that we need to be able to extract in the driver\n so add them to the growing array of GUIDs.\n \n Signed-off-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n Acked-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 3cdb9446a117d5d63af823bde6fe6babc312e77b\nAuthor: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\nDate: Tue Jan 16 22:19:00 2018 +0200\n\n thunderbolt: Add support for Intel Ice Lake\n \n The Thunderbolt controller is integrated into the Ice Lake CPU itself\n and requires special flows to power it on and off using force power bit\n in NHI VSEC registers. Runtime PM (RTD3) and Sx flows also differ from\n the discrete solutions. Now the firmware notifies the driver whether\n RTD3 entry or exit are possible. The driver is responsible of sending\n Go2Sx command through link controller mailbox when system enters Sx\n states (suspend-to-mem/disk). Rest of the ICM firwmare flows follow\n Titan Ridge.\n \n Signed-off-by: Raanan Avargil \u003craanan.avargil@intel.com\u003e\n Signed-off-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n Reviewed-by: Yehezkel Bernat \u003cYehezkelShB@gmail.com\u003e\n Tested-by: Mario Limonciello \u003cmario.limonciello@dell.com\u003e\n\ncommit 3f415e5ee18b0097755afc3ac3a5640b196a239e\nAuthor: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\nDate: Tue Feb 5 12:51:40 2019 +0300\n\n thunderbolt: Expose active parts of NVM even if upgrade is not supported\n \n Ice Lake Thunderbolt controller NVM firmware is part of the BIOS image\n which means it is not writable through the DMA port anymore. However, we\n can still read it so we can keep nvm_version and active parts of NVM.\n This way users still can find out the active NVM version and other\n potentially useful information directly from Linux.\n \n Signed-off-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n Reviewed-by: Yehezkel Bernat \u003cYehezkelShB@gmail.com\u003e\n Tested-by: Mario Limonciello \u003cmario.limonciello@dell.com\u003e\n\ncommit 58f414fa435cf728a82f435bac4781da86afb623\nAuthor: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\nDate: Tue Sep 11 15:34:23 2018 +0300\n\n thunderbolt: Hide switch attributes that are not set\n \n Thunderbolt host routers may not always contain DROM that includes\n device identification information. This is mostly needed for Ice Lake\n systems but some Falcon Ridge controllers on PCs also do not have DROM.\n \n In that case hide the identification attributes.\n \n Signed-off-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n Reviewed-by: Yehezkel Bernat \u003cYehezkelShB@gmail.com\u003e\n Tested-by: Mario Limonciello \u003cmario.limonciello@dell.com\u003e\n\ncommit d94dcbb10183f3b384c84e65724d2b753aa53c4d\nAuthor: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\nDate: Wed Jul 4 08:50:01 2018 +0300\n\n thunderbolt: Do not fail adding switch if some port is not implemented\n \n There are two ways to mark a port as unimplemented. Typical way is to\n return port type as TB_TYPE_INACTIVE when its config space is read.\n Alternatively if the port is not physically present (such as ports 10\n and 11 in ICL) reading from port config space returns\n TB_CFG_ERROR_INVALID_CONFIG_SPACE instead. Currently the driver bails\n out from adding the switch if it receives any error during port\n inititialization which is wrong.\n \n Handle this properly and just leave the port as TB_TYPE_INACTIVE before\n continuing to the next port.\n \n This also allows us to get rid of special casing for Light Ridge port 5\n in eeprom.c.\n \n Signed-off-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n Reviewed-by: Yehezkel Bernat \u003cYehezkelShB@gmail.com\u003e\n Tested-by: Mario Limonciello \u003cmario.limonciello@dell.com\u003e\n\ncommit 943795219d3cb9f8ce6ce51cad3ffe1f61e95c6b\nAuthor: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\nDate: Wed Jul 4 08:46:07 2018 +0300\n\n thunderbolt: Use 32-bit writes when writing ring producer/consumer\n \n The register access should be using 32-bit reads/writes according to the\n datasheet. With the previous generation hardware 16-bit writes have been\n working but starting with ICL this is not the case anymore so fix\n producer/consumer register update to use correct width register address.\n \n Signed-off-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n Reviewed-by: Yehezkel Bernat \u003cYehezkelShB@gmail.com\u003e\n Tested-by: Mario Limonciello \u003cmario.limonciello@dell.com\u003e\n\ncommit f437c24bf694b0293f835dea8c25e3a5c1433d07\nAuthor: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\nDate: Tue Jan 16 22:30:40 2018 +0200\n\n thunderbolt: Move NVM upgrade support flag to struct icm\n \n This is depends on the controller and on the platform/CPU we are\n running. Move it to struct icm so we can set it per controller.\n \n Signed-off-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n Reviewed-by: Yehezkel Bernat \u003cYehezkelShB@gmail.com\u003e\n Tested-by: Mario Limonciello \u003cmario.limonciello@dell.com\u003e\n\ncommit ce19f91eae43e39d5a1da55344756ab5a3c7e8d1\nAuthor: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\nDate: Tue Jun 11 19:31:26 2019 +0300\n\n thunderbolt: Correct path indices for PCIe tunnel\n \n PCIe tunnel path indices got mixed up when we added support for tunnels\n between switches that are not adjacent. This did not affect the\n functionality as it is just an index but fix it now nevertheless to make\n the code easier to understand.\n \n Reported-by: Rajmohan Mani \u003crajmohan.mani@intel.com\u003e\n Fixes: 8c7acaaf020f (\"thunderbolt: Extend tunnel creation to more than 2 adjacent switches\")\n Signed-off-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n Reviewed-by: Yehezkel Bernat \u003cYehezkelShB@gmail.com\u003e\n\ncommit 65af4a10743b766e319fb53812c5926c6d98b100\nAuthor: Michael Braun \u003cmichael-dev@fami-braun.de\u003e\nDate: Tue Aug 20 15:11:46 2019 +0200\n\n netfilter: nfnetlink_log: add support for VLAN information\n \n Currently, there is no vlan information (e.g. when used with a vlan aware\n bridge) passed to userspache, HWHEADER will contain an 08 00 (ip) suffix\n even for tagged ip packets.\n \n Therefore, add an extra netlink attribute that passes the vlan information\n to userspace similarly to 15824ab29f for nfqueue.\n \n Signed-off-by: Michael Braun \u003cmichael-dev@fami-braun.de\u003e\n Reviewed-by: Florian Westphal \u003cfw@strlen.de\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit 63d10e12b00dfc8d8387bea9eaab376881335731\nAuthor: Ander Juaristi \u003ca@juaristi.eus\u003e\nDate: Sat Aug 17 13:17:53 2019 +0200\n\n netfilter: nft_meta: support for time matching\n \n This patch introduces meta matches in the kernel for time (a UNIX timestamp),\n day (a day of week, represented as an integer between 0-6), and\n hour (an hour in the current day, or: number of seconds since midnight).\n \n All values are taken as unsigned 64-bit integers.\n \n The \u0027time\u0027 keyword is internally converted to nanoseconds by nft in\n userspace, and hence the timestamp is taken in nanoseconds as well.\n \n Signed-off-by: Ander Juaristi \u003ca@juaristi.eus\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit a1b840adafcbdc27da2982e7305c342204b8cfd8\nAuthor: Ander Juaristi \u003ca@juaristi.eus\u003e\nDate: Sat Aug 17 13:17:52 2019 +0200\n\n netfilter: nf_tables: Introduce new 64-bit helper register functions\n \n Introduce new helper functions to load/store 64-bit values onto/from\n registers:\n \n - nft_reg_store64\n - nft_reg_load64\n \n This commit also re-orders all these helpers from smallest to largest\n target bit size.\n \n Signed-off-by: Ander Juaristi \u003ca@juaristi.eus\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit cda4569137b90f200bee4922d894ca49d4188681\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Mon Aug 26 09:25:35 2019 +0200\n\n dt-bindings: clk: meson: add sm1 periph clock controller bindings\n \n Update the documentation to support clock driver for the Amlogic SM1 SoC\n and expose the GP1, DSU and the CPU 1, 2 \u0026 3 clocks.\n \n SM1 clock tree is very close, the main differences are :\n - each CPU core can achieve a different frequency, albeit a common PLL\n - a similar tree as the clock tree has been added for the DynamIQ Shared\n Unit\n - has a new GP1 PLL used for the DynamIQ Shared Unit\n - SM1 has additional clocks like for CSI, NanoQ an other components\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n\ncommit 8c7d7b4bb172709f71975c2a6e95c9715171603f\nAuthor: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\nDate: Tue Jul 23 11:44:08 2019 +0530\n\n video: pxafb: Remove cpufreq policy notifier\n \n The cpufreq policy notifier\u0027s CPUFREQ_ADJUST notification is going to\n get removed soon.\n \n The notifier callback pxafb_freq_policy() isn\u0027t doing anything apart\n from printing a debug message on CPUFREQ_ADJUST notification. There is\n no point in keeping an otherwise empty callback and registering the\n notifier.\n \n Remove it.\n \n Acked-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit eb58a4fad3432be6e9dc2e35b78ae99f4a40ce5f\nAuthor: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\nDate: Tue Jul 23 11:44:07 2019 +0530\n\n video: sa1100fb: Remove cpufreq policy notifier\n \n The cpufreq policy notifier\u0027s CPUFREQ_ADJUST notification is going to\n get removed soon.\n \n The notifier callback sa1100fb_freq_policy() isn\u0027t doing anything apart\n from printing a debug message on CPUFREQ_ADJUST notification. There is\n no point in keeping an otherwise empty callback and registering the\n notifier.\n \n Remove it.\n \n Acked-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 40f0fc2a416b343a604a8131247a150588658d32\nAuthor: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\nDate: Tue Jul 23 11:44:06 2019 +0530\n\n arch_topology: Use CPUFREQ_CREATE_POLICY instead of CPUFREQ_NOTIFY\n \n CPUFREQ_NOTIFY is going to get removed soon, lets use\n CPUFREQ_CREATE_POLICY instead of that here. CPUFREQ_CREATE_POLICY is\n called only once (which is exactly what we want here) for each cpufreq\n policy when it is first created.\n \n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Acked-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit afe969074eb7d0f52761ed80da15e9a282d19677\nAuthor: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\nDate: Fri Jul 5 15:49:48 2019 +0530\n\n cpufreq: powerpc_cbe: Switch to QoS requests for freq limits\n \n The cpufreq core now takes the min/max frequency constraints via QoS\n requests and the CPUFREQ_ADJUST notifier shall get removed later on.\n \n Switch over to using the QoS request for maximum frequency constraint\n for ppc_cbe_cpufreq driver.\n \n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n [ rjw: Subject ]\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit dce2e3a8414e31fc286a3f8064df773a3af83a26\nAuthor: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\nDate: Tue Jul 23 11:44:03 2019 +0530\n\n cpufreq: powerpc: macintosh: Switch to QoS requests for freq limits\n \n The cpufreq core now takes the min/max frequency constraints via QoS\n requests and the CPUFREQ_ADJUST notifier shall get removed later on.\n \n Switch over to using the QoS request for maximum frequency constraint\n for windfarm_cpufreq_clamp driver.\n \n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n [ rjw: Subject ]\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 419a7a1f167176d60d036d8002b6e3661fde9707\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 22:27:40 2019 +0200\n\n mtd: rawnand: remove w90x900 driver\n \n The ARM w90x900 platform is getting removed, so this driver is obsolete.\n \n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n\ncommit 0c43125f2778dddb469036caaa14533846ce40bd\nAuthor: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\nDate: Fri Aug 23 15:39:37 2019 +0000\n\n mtd: nand: fix typo, s/erasablocks/eraseblocks\n \n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n\ncommit 3342d2f88ef9519c58d76004afb2c8fd93b3e418\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Tue Aug 20 11:50:57 2019 +0900\n\n mtd: rawnand: sharpsl: add include guard to linux/mtd/sharpsl.h\n \n Add a header include guard just in case.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n\ncommit 220280708a2349bc8dd61ab029483d65440a5aa0\nAuthor: Mason Yang \u003cmasonccyang@mxic.com.tw\u003e\nDate: Mon Aug 19 15:19:09 2019 +0800\n\n dt-bindings: mtd: Document Macronix raw NAND controller bindings\n \n Document the bindings used by the Macronix raw NAND controller.\n \n Signed-off-by: Mason Yang \u003cmasonccyang@mxic.com.tw\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n\ncommit 738b0ca55f4f6ae1035262c2a2a605d2e9085031\nAuthor: Mason Yang \u003cmasonccyang@mxic.com.tw\u003e\nDate: Mon Aug 19 15:19:08 2019 +0800\n\n mtd: rawnand: Add Macronix raw NAND controller driver\n \n Add a driver for Macronix raw NAND controller.\n \n Signed-off-by: Mason Yang \u003cmasonccyang@mxic.com.tw\u003e\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n\ncommit feeb07d0ca5ad0d6edb5bb728e33e2e86118529d\nAuthor: Matthew Ruffell \u003cmatthew.ruffell@canonical.com\u003e\nDate: Thu Aug 15 16:26:40 2019 +1200\n\n drm/hisilicon/hibmc: Make CONFIG_DRM_HISI_HIBMC depend on ARM64\n \n Hisilicon developed hibmc_drm for their arm64 based soc and did not\n intend for this driver to be used on any other architecture than arm64.\n \n Using it on amd64 leads to incorrect video modes being used, making\n the screen unreadable, forcing users to manually blacklist the module\n on the kernel command line to use the d-i server installer or any\n graphical sessions.\n \n Make CONFIG_DRM_HISI_HIBMC firmly depend on ARM64 to ensure it is not\n built for other architectures.\n \n Signed-off-by: Matthew Ruffell \u003cmatthew.ruffell@canonical.com\u003e\n Signed-off-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n\ncommit bda4a850f7dc6dee141a6bdba85976c8cf686626\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 18:38:53 2019 +0800\n\n drm/hisilicon: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n\ncommit 4f1a7007d688a792b6b929220d7f68a25eb493be\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Apr 21 17:51:48 2018 +0800\n\n drm/hisilicon/hibmc: Using module_pci_driver.\n \n Remove boilerplate code by using macro module_pci_driver.\n \n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Reviewed-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Signed-off-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n\ncommit b898e50f9f49f7d90f3bca94ac046145072034a2\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Thu Aug 1 17:14:14 2019 +0300\n\n arm-nommu: remove the unused pgprot_dmacoherent define\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit 5518ea1ad2c0c7f38d067f621d9349e6a11c8879\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Thu Aug 1 17:13:51 2019 +0300\n\n unicore32: remove the unused pgprot_dmacoherent define\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit 0846e1616f0f3365cea732e82e2383932fe644e5\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Mon Aug 26 02:49:15 2019 +0000\n\n cirrus: cs89x0: remove set but not used variable \u0027lp\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/net/ethernet/cirrus/cs89x0.c: In function \u0027cs89x0_platform_probe\u0027:\n drivers/net/ethernet/cirrus/cs89x0.c:1847:20: warning:\n variable \u0027lp\u0027 set but not used [-Wunused-but-set-variable]\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Fixes: 6751edeb8700 (\"cirrus: cs89x0: Use managed interfaces\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 2889456498c635f4e8262a8401b2eddc263953c7\nAuthor: David S. Miller \u003cdavem@davemloft.net\u003e\nDate: Sun Aug 25 19:07:04 2019 -0700\n\n Revert \"net: mediatek: remove set but not used variable \u0027status\u0027\"\n \n This reverts commit ee641b0cdb9486f8212a3da153a46ab3551a97e5.\n \n Actually it is not clear whether this register read is not\n needed for it\u0027s HW side effects or not.\n \n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ee641b0cdb9486f8212a3da153a46ab3551a97e5\nAuthor: Mao Wenan \u003cmaowenan@huawei.com\u003e\nDate: Mon Aug 26 09:31:18 2019 +0800\n\n net: mediatek: remove set but not used variable \u0027status\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n drivers/net/ethernet/mediatek/mtk_eth_soc.c: In function mtk_handle_irq:\n drivers/net/ethernet/mediatek/mtk_eth_soc.c:1951:6: warning: variable status set but not used [-Wunused-but-set-variable]\n \n Fixes: 296c9120752b (\"net: ethernet: mediatek: Add MT7628/88 SoC support\")\n Signed-off-by: Mao Wenan \u003cmaowenan@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6652bf6408895b09d31fd4128a1589a1a0672823\nAuthor: Gustavo Romero \u003cgromero@linux.vnet.ibm.com\u003e\nDate: Tue Aug 20 18:54:11 2019 -0400\n\n selftests/powerpc: Retry on host facility unavailable\n \n TM test tm-unavailable must take into account aborts due to host aborting\n a transactin because of a facility unavailable exception, just like it\n already does for aborts on reschedules (TM_CAUSE_KVM_RESCHED).\n \n Reported-by: Desnes A. Nunes do Rosario \u003cdesnesn@linux.ibm.com\u003e\n Tested-by: Desnes A. Nunes do Rosario \u003cdesnesn@linux.ibm.com\u003e\n Signed-off-by: Gustavo Romero \u003cgromero@linux.vnet.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/1566341651-19747-1-git-send-email-gromero@linux.vnet.ibm.com\n\ncommit 9d535e200f09ce347afc38c81ec7f2901187e5f0\nAuthor: Gustavo Romero \u003cgromero@linux.ibm.com\u003e\nDate: Wed Aug 14 15:52:11 2019 -0500\n\n selftests/powerpc: Fix and enhance TM signal context tests\n \n Currently TM signal context tests for GPR, FPR, VMX, and VSX registers\n print wrong register numbers (wrongly starting from register 0 instead\n of the first register in the non-volatile subset). Besides it the\n output when a mismatch happens is poor giving not much information\n about which context and which register mismatches, because it prints\n both contexts at the same time and not a comparison between the value\n that mismatches and the value expected and, moreover, it stops\n printing on the first mismatch, but it\u0027s important to know if there\n are other mismatches happening beyond the first one.\n \n For instance, this is the current output when a mismatch happens:\n \n test: tm_signal_context_chk_gpr\n tags: git_version:v5.2-8249-g02e970fae465-dirty\n Failed on 0 GPR 1 or 18446744073709551615\n failure: tm_signal_context_chk_gpr\n \n test: tm_signal_context_chk_fpu\n tags: git_version:v5.2-8248-g09c289e3ef80\n Failed on 0 FP -1 or -1\n failure: tm_signal_context_chk_fpu\n \n test: tm_signal_context_chk_vmx\n tags: git_version:v5.2-8248-g09c289e3ef80\n Failed on 0 vmx 0xfffffffffffffffefffffffdfffffffc vs 0xfffffffffffffffefffffffdfffffffc\n failure: tm_signal_context_chk_vmx\n \n test: tm_signal_context_chk_vsx\n tags: git_version:v5.2-8248-g09c289e3ef80\n Failed on 0 vsx 0xfffffffffefffffffdfffffffcffffff vs 0xfffffffffefffffffdfffffffcffffff\n failure: tm_signal_context_chk_vsx\n \n This commit fixes the register numbers printed and enhances the error\n output by providing a full list of mismatching registers separated by\n the context (non-speculative or speculative context), for example:\n \n test: tm_signal_context_chk_gpr\n tags: git_version:v5.2-8249-g02e970fae465-dirty\n GPR14 (1st context) \u003d\u003d 1 instead of -1 (expected)\n GPR15 (1st context) \u003d\u003d 2 instead of -2 (expected)\n GPR14 (2nd context) \u003d\u003d 0 instead of 18446744073709551615 (expected)\n GPR15 (2nd context) \u003d\u003d 0 instead of 18446744073709551614 (expected)\n failure: tm_signal_context_chk_gpr\n \n test: tm_signal_context_chk_fpu\n tags: git_version:v5.2-8249-g02e970fae465-dirty\n FPR14 (1st context) \u003d\u003d -1 instead of 1 (expected)\n FPR15 (1st context) \u003d\u003d -2 instead of 2 (expected)\n failure: tm_signal_context_chk_fpu\n \n test: tm_signal_context_chk_vmx\n tags: git_version:v5.2-8249-g02e970fae465-dirty\n VMX20 (1st context) \u003d\u003d 0xfffffffffffffffefffffffdfffffffc instead of 0x00000001000000020000000300000004 (expected)\n VMX21 (1st context) \u003d\u003d 0xfffffffbfffffffafffffff9fffffff8 instead of 0x00000005000000060000000700000008 (expected)\n failure: tm_signal_context_chk_vmx\n \n test: tm_signal_context_chk_vsx\n tags: git_version:v5.2-8249-g02e970fae465-dirty\n VSX20 (1st context) \u003d\u003d 0xfffffffffefffffffdfffffffcffffff instead of 0x00000001000000020000000300000004 (expected)\n VSX21 (1st context) \u003d\u003d 0xfbfffffffafffffff9fffffff8ffffff instead of 0x00000005000000060000000700000008 (expected)\n failure: tm_signal_context_chk_vsx\n \n Finally, this commit adds comments to the tests in the hope that it\n will help people not so familiar with TM understand the tests.\n \n Signed-off-by: Gustavo Romero \u003cgromero@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190814205211.24840-1-gromero@linux.ibm.com\n\ncommit 284f3a8e0947d8b577565dd644e8fb62836d9018\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Wed Aug 21 09:09:23 2019 +0200\n\n m68k: coldfire: Include the GPIO driver header\n \n The Coldfire GPIO driver needs to explicitly incldue the\n GPIO driver header since it is providing a driver.\n \n Cc: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Greg Ungerer \u003cgerg@linux-m68k.org\u003e\n\ncommit 3109741a8d773b91eec4a1f7764c97a1176ec32d\nAuthor: Eric Auger \u003ceric.auger@redhat.com\u003e\nDate: Fri Aug 23 19:33:30 2019 +0200\n\n KVM: arm/arm64: vgic: Use a single IO device per redistributor\n \n At the moment we use 2 IO devices per GICv3 redistributor: one\n one for the RD_base frame and one for the SGI_base frame.\n \n Instead we can use a single IO device per redistributor (the 2\n frames are contiguous). This saves slots on the KVM_MMIO_BUS\n which is currently limited to NR_IOBUS_DEVS (1000).\n \n This change allows to instantiate up to 512 redistributors and may\n speed the guest boot with a large number of VCPUs.\n \n Signed-off-by: Eric Auger \u003ceric.auger@redhat.com\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 926c61568d0b2e57254e92290369d4539568f8cc\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Sun Aug 25 10:44:17 2019 +0100\n\n KVM: arm/arm64: vgic: Remove spurious semicolons\n \n Detected by Coccinelle (and Will Deacon) using\n scripts/coccinelle/misc/semicolon.cocci.\n \n Reported-by: Will Deacon \u003cwill@kernel.org\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 591f4ea2ed6128de397ff8f2dee6ed009db38174\nAuthor: Ji-Ze Hong (Peter Hong) \u003chpeter@gmail.com\u003e\nDate: Fri Aug 16 13:27:29 2019 +0800\n\n serial: 8250_pci: Add F81504A series Support\n \n Fintek F81504A/508A/512A is PCIE to 4/8/12 UARTs device. It\u0027s support\n IO/MMIO/PCIE conf to access all functions. The old F81504/508/512 is\n only support IO.\n \n Signed-off-by: Ji-Ze Hong (Peter Hong) \u003chpeter+linux_kernel@gmail.com\u003e\n Link: https://lore.kernel.org/r/1565933249-23076-1-git-send-email-hpeter+linux_kernel@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a976556d6705189cee74383bad79960519141a41\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Wed Aug 14 17:07:59 2019 +0300\n\n serial: mctrl_gpio: Use gpiod flags directly\n \n Description of the modem line control GPIOs contain a boolean type to set\n direction of the line. Since GPIO library provides an enumerator type of flags,\n we may utilize it and allow a bit more flexibility on the choice of the type of\n the line parameters. It also removes an additional layer of value conversion.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Stefan Roese \u003csr@denx.de\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Link: https://lore.kernel.org/r/20190814140759.17486-1-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit cadcc6b20d971510f00ea19f3da6d3b992658f0b\nAuthor: Sumera Priyadarsini \u003csylphrenadin@gmail.com\u003e\nDate: Fri Aug 23 01:05:35 2019 +0530\n\n staging: rtl8192u: Add space around binary operators\n \n This patch fixes the file r8180_93cx6.c to avoid the checkpatch.pl warnings:\n \n CHECK: spaces preferred around that \u0027\u003c\u003c\u0027 (ctx:VxV)\n CHECK: spaces preferred around that \u0027-\u0027 (ctx:VxV)\n \n Signed-off-by: Sumera Priyadarsini \u003csylphrenadin@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190822193535.14357-1-sylphrenadin@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 684c215dd06eccc515dfd786347ac556c00f895a\nAuthor: Stephen Brennan \u003cstephen@brennan.io\u003e\nDate: Fri Aug 23 09:24:10 2019 -0700\n\n staging: rtl8192u: remove code under TO_DO_LIST\n \n Several blocks of code are guarded by #ifdef TO_DO_LIST. If this is\n defined, compilation fails. No machinery exists to define this, and no\n documenation on the in-progress feature exists. Since this code is dead,\n let\u0027s delete it.\n \n Signed-off-by: Stephen Brennan \u003cstephen@brennan.io\u003e\n Link: https://lore.kernel.org/r/20190823162410.10038-1-stephen@brennan.io\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 76fe9cfb09e7bf261b814a155f5953fa4d2df89a\nAuthor: Sumera Priyadarsini \u003csylphrenadin@gmail.com\u003e\nDate: Thu Aug 22 20:25:45 2019 +0530\n\n staging: rtl8192u: Remove unnecessary blank lines\n \n This patch fixes the file r8180_93cx6.c to avoid the checkpatch.pl warning:\n \n CHECK: Please don\u0027t use multiple blank lines\n \n Signed-off-by: Sumera Priyadarsini \u003csylphrenadin@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190822145545.19894-1-sylphrenadin@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 5e718a12e1a125045b451678faf6ceffc7591e3b\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Thu Aug 22 09:46:09 2019 +0100\n\n staging: rtl8192u: remove redundant assignment to pointer crypt\n \n The pointer crypt is being set with a value that is never read,\n the assignment is redundant and hence can be removed.\n \n Thanks to Dan Carpenter for sanity checking that this was indeed\n redundant.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Reviewed-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Link: https://lore.kernel.org/r/20190822084609.8971-1-colin.king@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 5ab706bd0bcaea0296426ecca333f69bad60d24e\nAuthor: Saurav Girepunje \u003csaurav.girepunje@gmail.com\u003e\nDate: Wed Aug 21 23:46:34 2019 +0530\n\n staging: rtl8723bs: os_dep: Drop condition with no effect\n \n As the \"else if\" and \"else\" branch body are identical the condition\n has no effect. So drop the \"else if\" condition.\n \n Signed-off-by: Saurav Girepunje \u003csaurav.girepunje@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190821181631.GA11082@saurav\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit eac53016919f2d2a8df607c69dc2b1d47273a61e\nAuthor: Saurav Girepunje \u003csaurav.girepunje@gmail.com\u003e\nDate: Wed Aug 21 23:32:01 2019 +0530\n\n staging: rtl8723bs: hal: Drop condition with no effect\n \n As the \"else if\" and \"else\" branch body are identical the condition\n has no effect. So drop the else if condition.\n \n Signed-off-by: Saurav Girepunje \u003csaurav.girepunje@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190821180153.GA10678@saurav\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit de9defd21539d37816dc869b098ae8eed12857ad\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Wed Aug 21 22:31:10 2019 +0530\n\n staging: rtl8723bs: os_dep: Remove unused declarations\n \n Remove unused Macro declarations\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Link: https://lore.kernel.org/r/1566406901-6700-1-git-send-email-hariprasad.kelam@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit e0ccdef92653f8867e2d1667facfd3c23699f540\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Fri Aug 23 12:38:07 2019 +0200\n\n ALSA: usb-audio: Clean up check_input_term()\n \n The primary changes in this patch are cleanups of __check_input_term()\n and move to a non-nested switch-case block by evaluating the pair of\n UAC version and the unit type, as we\u0027ve done for parse_audio_unit().\n Also each parser is split into the function for readability.\n \n Now, a slight behavior change by this cleanup is the handling of\n processing and extension units. Formerly we\u0027ve dealt with them\n differently between UAC1/2 and UAC3; the latter returns an error if no\n input sources are available, while the former continues to parse.\n \n In this patch, unify the behavior in all cases: when input sources are\n available, it parses recursively, then override the type and the id,\n as well as channel information if not provided yet.\n \n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit b04f537caab4deae1b839438e8fd5ed4bc598b43\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Fri Aug 23 17:27:03 2019 -0700\n\n ARM: dts: vf610-zii-scu4-aib: Configure IRQ line for GPIO expander\n \n Configure IRQ line for SX1503 GPIO expander. We already have\n appropriate pinmux entry and all that is missing is \"interrupt-parent\"\n and \"interrupts\" properties. Add them.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Shawn Guo \u003cshawnguo@kernel.org\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: linux-kernel@vger.kernel.org\n Tested-by: Chris Healy \u003ccphealy@gmail.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 0ef7c7cce43f6ecc2b96d447e69b2900a9655f7c\nAuthor: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\nDate: Thu Aug 8 19:54:08 2019 +0300\n\n PM / devfreq: passive: Use non-devm notifiers\n \n The devfreq passive governor registers and unregisters devfreq\n transition notifiers on DEVFREQ_GOV_START/GOV_STOP using devm wrappers.\n \n If devfreq itself is registered with devm then a warning is triggered on\n rmmod from devm_devfreq_unregister_notifier. Call stack looks like this:\n \n devm_devfreq_unregister_notifier+0x30/0x40\n devfreq_passive_event_handler+0x4c/0x88\n devfreq_remove_device.part.8+0x6c/0x9c\n devm_devfreq_dev_release+0x18/0x20\n release_nodes+0x1b0/0x220\n devres_release_all+0x78/0x84\n device_release_driver_internal+0x100/0x1c0\n driver_detach+0x4c/0x90\n bus_remove_driver+0x7c/0xd0\n driver_unregister+0x2c/0x58\n platform_driver_unregister+0x10/0x18\n imx_devfreq_platdrv_exit+0x14/0xd40 [imx_devfreq]\n \n This happens because devres_release_all will first remove all the nodes\n into a separate todo list so the nested devres_release from\n devm_devfreq_unregister_notifier won\u0027t find anything.\n \n Fix the warning by calling the non-devm APIS for frequency notification.\n Using devm wrappers is not actually useful for a governor anyway: it\n relies on the devfreq core to correctly match the GOV_START/GOV_STOP\n notifications.\n \n Fixes: 996133119f57 (\"PM / devfreq: Add new passive governor\")\n Signed-off-by: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Acked-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 4294a779bd8dff6c65e7e85ffe7a1ea236e92a68\nAuthor: Kamil Konieczny \u003ck.konieczny@partner.samsung.com\u003e\nDate: Wed Aug 7 15:38:36 2019 +0200\n\n PM / devfreq: exynos-bus: Convert to use dev_pm_opp_set_rate()\n \n Reuse opp core code for setting bus clock and voltage. As a side\n effect this allow usage of coupled regulators feature (required\n for boards using Exynos5422/5800 SoCs) because dev_pm_opp_set_rate()\n uses regulator_set_voltage_triplet() for setting regulator voltage\n while the old code used regulator_set_voltage_tol() with fixed\n tolerance. This patch also removes no longer needed parsing of DT\n property \"exynos,voltage-tolerance\" (no Exynos devfreq DT node uses\n it). After applying changes both functions exynos_bus_passive_target()\n and exynos_bus_target() have the same code, so remove\n exynos_bus_passive_target(). In exynos_bus_probe() replace it with\n exynos_bus_target.\n \n Signed-off-by: Kamil Konieczny \u003ck.konieczny@partner.samsung.com\u003e\n Acked-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 2c2b20e0da89c76759ee28c6824413ab2fa3bfc6\nAuthor: Kamil Konieczny \u003ck.konieczny@partner.samsung.com\u003e\nDate: Wed Aug 7 15:38:35 2019 +0200\n\n PM / devfreq: exynos-bus: Correct clock enable sequence\n \n Regulators should be enabled before clocks to avoid h/w hang. This\n require change in exynos_bus_probe() to move exynos_bus_parse_of()\n after exynos_bus_parent_parse_of() and change in error handling.\n Similar change is needed in exynos_bus_exit() where clock should be\n disabled before regulators.\n \n Signed-off-by: Kamil Konieczny \u003ck.konieczny@partner.samsung.com\u003e\n Acked-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit e2fc1677eea749414a7db0058847509566bf64e1\nAuthor: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\nDate: Wed Jul 24 20:59:53 2019 +0200\n\n PM / devfreq: Correct devm_devfreq_remove_device() documentation\n \n Correct the documentation for devm_devfreq_remove_device() argument.\n \n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 1dd62c66d345a4e5dcfa5a4e81999600515b4309\nAuthor: Lukasz Luba \u003cl.luba@partner.samsung.com\u003e\nDate: Wed Jun 5 11:12:34 2019 +0200\n\n PM / devfreq: events: extend events by type of counted data\n \n This patch adds posibility to choose what type of data should be counted\n by the PPMU counter. Now the type comes from DT where the event has been\n defined. When there is no \u0027event-data-type\u0027 the default value is used,\n which is \u0027read+write data in bytes\u0027.\n It is needed when you want to know not only read+write data bytes but\n i.e. only write data in byte, or number of read requests, etc.\n \n Signed-off-by: Lukasz Luba \u003cl.luba@partner.samsung.com\u003e\n Acked-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n [Updated property by MyungJoo. data_type --\u003e event_type]\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 0ae9c3213c2ca34f7e7ebad1436267778d37ba9f\nAuthor: Lukasz Luba \u003cl.luba@partner.samsung.com\u003e\nDate: Wed Jun 5 11:12:33 2019 +0200\n\n PM / devfreq: exynos-events: change matching code during probe\n \n The patch changes the way how the \u0027ops\u0027 gets populated for different\n device versions. The matching function now uses \u0027of_device_id\u0027 in order\n to identify the device type.\n \n Signed-off-by: Lukasz Luba \u003cl.luba@partner.samsung.com\u003e\n Signed-off-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 15c497e97e1aa1c852da91930a04787235a3ab88\nAuthor: zhangyi (F) \u003cyi.zhang@huawei.com\u003e\nDate: Sat Aug 24 23:35:56 2019 -0400\n\n ext4: fix potential use after free after remounting with noblock_validity\n \n Remount process will release system zone which was allocated before if\n \"noblock_validity\" is specified. If we mount an ext4 file system to two\n mountpoints with default mount options, and then remount one of them\n with \"noblock_validity\", it may trigger a use after free problem when\n someone accessing the other one.\n \n # mount /dev/sda foo\n # mount /dev/sda bar\n \n User access mountpoint \"foo\" | Remount mountpoint \"bar\"\n |\n ext4_map_blocks() | ext4_remount()\n check_block_validity() | ext4_setup_system_zone()\n ext4_data_block_valid() | ext4_release_system_zone()\n | free system_blks rb nodes\n access system_blks rb nodes |\n trigger use after free |\n \n This problem can also be reproduced by one mountpint, At the same time,\n add_system_zone() can get called during remount as well so there can be\n racing ext4_data_block_valid() reading the rbtree at the same time.\n \n This patch add RCU to protect system zone from releasing or building\n when doing a remount which inverse current \"noblock_validity\" mount\n option. It assign the rbtree after the whole tree was complete and\n do actual freeing after rcu grace period, avoid any intermediate state.\n \n Reported-by: syzbot+1e470567330b7ad711d5@syzkaller.appspotmail.com\n Signed-off-by: zhangyi (F) \u003cyi.zhang@huawei.com\u003e\n Signed-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Reviewed-by: Jan Kara \u003cjack@suse.cz\u003e\n\ncommit ea4463b8c36b8884c80803d0ba9f8742e0f2bb7b\nAuthor: Xiaoguang Wang \u003cxiaoguang.wang@linux.alibaba.com\u003e\nDate: Sat Aug 24 23:10:17 2019 -0400\n\n jbd2: add missing tracepoint for reserved handle\n \n This issue was found when I use ebpf to trace every jbd2\n handle\u0027s running info in dioread_nolock case.\n \n Signed-off-by: Xiaoguang Wang \u003cxiaoguang.wang@linux.alibaba.com\u003e\n Signed-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n\ncommit 3e4c6948e78b495708a99c0599e50f0faffb2ade\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Wed Aug 21 16:03:48 2019 +0900\n\n kbuild: remove unneeded \u0027+\u0027 marker from kselftest-merge\n \n This line contains $(MAKE), so Make knows that it will invoke sub-make\n without help of the \u0027+\u0027 marker.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit c1236979b4d2cbaef8bb34aec83a44de445e4210\nAuthor: Andrew Lunn \u003candrew@lunn.ch\u003e\nDate: Sun Aug 25 01:04:17 2019 +0200\n\n net: phy: sfp: Add labels to hwmon sensors\n \n SFPs can report two different power values, the transmit power and the\n receive power. Add labels to make it clear which is which. Also add\n labels to the other sensors, VCC power supply, bias and module\n temperature.\n \n sensors(1) now shows:\n \n sff2-isa-0000\n Adapter: ISA adapter\n VCC: +3.23 V\n temperature: +33.4 C\n TX_power: 276.00 uW\n RX_power: 20.00 uW\n bias: +0.01 A\n \n Signed-off-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9549d70a2d71526b8dc41cc0b255219ba46e5bf7\nAuthor: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\nDate: Fri Aug 23 11:48:53 2019 +0200\n\n s390/qeth: add xmit_more support for IQD devices\n \n IQD devices offer limited support for bulking: all frames in a TX buffer\n need to have the same target. qeth_iqd_may_bulk() implements this\n constraint, and allows us to defer the TX doorbell until\n (a) the buffer is full (since each buffer needs its own doorbell), or\n (b) the entire TX queue is full, or\n (b) we reached the BQL limit.\n \n Signed-off-by: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 96bd6c94bdf9de38b0fa0ec679fe40013f1c4576\nAuthor: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\nDate: Fri Aug 23 11:48:52 2019 +0200\n\n s390/qeth: add BQL support for IQD devices\n \n Each TX buffer may contain multiple skbs. So just accumulate the sent\n byte count in the buffer struct, and later use the same count when\n completing the buffer.\n \n Signed-off-by: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 85e537d8f1b6b1201ced628b124b3d08436f5a04\nAuthor: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\nDate: Fri Aug 23 11:48:51 2019 +0200\n\n s390/qeth: when in TX NAPI mode, use napi_consume_skb()\n \n This allows the stack to bulk-free our TX-completed skbs.\n \n Signed-off-by: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit e53edf743d26b39dfd78af43ff97620a4ac13ffc\nAuthor: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\nDate: Fri Aug 23 11:48:50 2019 +0200\n\n s390/qeth: add TX NAPI support for IQD devices\n \n Due to their large MTU and potentially low utilization of TX buffers,\n IQD devices in particular require fast TX recycling. This makes them\n a prime candidate for a TX NAPI path in qeth.\n \n qeth_tx_poll() uses the recently introduced qdio_inspect_queue() helper\n to poll the TX queue for completed buffers. To avoid hogging the CPU for\n too long, we yield to the stack after completing an entire queue\u0027s worth\n of buffers.\n While IQD is expected to transfer its buffers synchronously (and thus\n doesn\u0027t support TX interrupts), a timer covers for the odd case where a\n TX buffer doesn\u0027t complete synchronously. Currently this timer should\n only ever fire for\n (1) the mcast queue,\n (2) the occasional race, where the NAPI poll code observes an update to\n queue-\u003eused_buffers while the TX doorbell hasn\u0027t been issued yet.\n \n Signed-off-by: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit eeac0e20a173dd9407e7092b3ddb45917249d68d\nAuthor: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\nDate: Fri Aug 23 11:48:49 2019 +0200\n\n s390/qeth: collect accurate TX statistics\n \n This consolidates the SW statistics code, and improves it to\n (1) account for the header overhead of each segment on a TSO skb,\n (2) count dangling packets as in-error (during eg. shutdown), and\n (3) only count offloads when the skb was successfully transmitted.\n \n We also count each segment of an TSO skb as one packet - except for\n tx_dropped, to be consistent with dev-\u003etx_dropped.\n \n Signed-off-by: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 313dc689b16c08b081939ee9b87dac3736c780e3\nAuthor: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\nDate: Fri Aug 23 11:48:48 2019 +0200\n\n s390/qdio: let drivers opt-out from Output Queue scanning\n \n If a driver wants to use the new Output Queue poll code, then the qdio\n layer must disable its internal Queue scanning. Let the driver select\n this mode by passing a special scan_threshold of 0.\n \n As the scan_threshold is the same for all Output Queues, also move it\n into the main qdio_irq struct. This allows for fast opt-out checking, a\n driver is expected to operate either _all_ or none of its Output Queues\n in polling mode.\n \n Signed-off-by: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\n Acked-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7c47f5afdeef763599f1ae22d29b8c3904c58315\nAuthor: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\nDate: Fri Aug 23 11:48:47 2019 +0200\n\n s390/qdio: enable drivers to poll for Output completions\n \n While commit d36deae75011 (\"qdio: extend API to allow polling\") enhanced\n the qdio layer so that drivers can poll their Input Queues, we don\u0027t\n have the corresponding infrastructure for Output Queues yet.\n \n Factor out a helper that scans a single QDIO Queue, so that qeth can\n implement TX NAPI on top of it.\n While doing so, remove the duplicated tracking of the next-to-scan index\n (q-\u003efirst_to_check vs q-\u003efirst_to_kick) in this code path.\n \n qdio_handle_aobs() needs to move slightly upwards in the code hierarchy,\n so that it\u0027s still called from the polling path.\n \n Signed-off-by: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\n Acked-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit fbbdbc6473070dcb3ee1d69cf1c49ff78677d716\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Fri Aug 23 01:51:41 2019 -0400\n\n bnxt_en: Fix allocation of zero statistics block size regression.\n \n Recent commit added logic to determine the appropriate statistics block\n size to allocate and the size is stored in bp-\u003ehw_ring_stats_size. But\n if the firmware spec is older than 1.6.0, it is 0 and not initialized.\n This causes the allocation to fail with size 0 and bnxt_open() to\n abort. Fix it by always initializing bp-\u003ehw_ring_stats_size to the\n legacy default size value.\n \n Fixes: 4e7485066373 (\"bnxt_en: Allocate the larger per-ring statistics block for 57500 chips.\")\n Reported-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Tested-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit dd016aca28f67603f32c4e666805db519df2120a\nAuthor: Markus Elfring \u003celfring@users.sourceforge.net\u003e\nDate: Thu Aug 22 18:00:40 2019 +0200\n\n net/core/skmsg: Delete an unnecessary check before the function call “consume_skb”\n \n The consume_skb() function performs also input parameter validation.\n Thus the test around the call is not needed.\n \n This issue was detected by using the Coccinelle software.\n \n Signed-off-by: Markus Elfring \u003celfring@users.sourceforge.net\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 2307f4a517c7dbb26fe0c0513ee0ba9d3239362f\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 22 22:49:37 2019 +0800\n\n net: hns3: Fix -Wunused-const-variable warning\n \n drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h:542:30:\n warning: meta_data_key_info defined but not used [-Wunused-const-variable\u003d]\n drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h:553:30:\n warning: tuple_key_info defined but not used [-Wunused-const-variable\u003d]\n \n The two variable is only used in hclge_main.c,\n so just move the definition over there.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 000ec95fbe757401b5da1c9904840085204e5b3d\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Wed Aug 21 16:02:04 2019 +0900\n\n kbuild: pkg: rename scripts/package/Makefile to scripts/Makefile.package\n \n scripts/package/Makefile does not use $(obj) or $(src) at all.\n It actually generates files and directories in the top of $(objtree).\n I do not see much sense in descending into scripts/package/.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 6a4f6a26d32abb5d5f61861ab3cf5880d9556ff5\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Wed Aug 21 16:02:03 2019 +0900\n\n kbuild: pkg: add package targets to PHONY instead of FORCE\n \n These are not real targets. Adding them to PHONY is preferred.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 46a63d4b0d79cf9e8afa3879acf9f6cf74a84a08\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Wed Aug 21 16:02:02 2019 +0900\n\n kbuild: pkg: clean up package files/dirs from the top Makefile\n \n I am not a big fan of the $(objtree)/ hack for clean-files/clean-dirs.\n \n These are created in the top of $(objtree), so let\u0027s clean them up\n from the top Makefile.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit f3acd33d840d3ea3e1233d234605c85cbbf26054\nAuthor: xiaolinkui \u003cxiaolinkui@kylinos.cn\u003e\nDate: Thu Aug 22 14:58:16 2019 +0800\n\n net: use unlikely for dql_avail case\n \n This is an unlikely case, use unlikely() on it seems logical.\n \n Signed-off-by: xiaolinkui \u003cxiaolinkui@kylinos.cn\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 5aff7c46172b270af2a9b10eed3a8215c4ba9a87\nAuthor: Jacob Huisman \u003cjacobhuisman@kernelthusiast.com\u003e\nDate: Fri Aug 16 13:22:09 2019 +0100\n\n docs: process: fix broken link\n \n http://linux.yyz.us/patch-format.html seems to be down since\n approximately September 2018. There is a working archive copy on\n arhive.org. Replaced the links in documenation + translations.\n \n Signed-off-by: Jacob Huisman \u003cjacobhuisman@kernelthusiast.com\u003e\n Reviewed-by: Federico Vaga \u003cfederico.vaga@vaga.pv.it\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 215701807e53a379e1c9b2e6401676779ef85ea0\nAuthor: Guido Günther \u003cagx@sigxcpu.org\u003e\nDate: Thu Aug 22 13:10:23 2019 +0200\n\n arm64: dts: imx8mq: Add mux controller to iomuxc_gpr\n \n The only mux controls the MIPI DSI input selection.\n \n Signed-off-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 1ea4b76cdfde49a2421523769cb99d4be6aedea3\nAuthor: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\nDate: Thu Aug 22 08:02:38 2019 +0200\n\n ARM: dts: imx6ul-kontron-n6310: Add Kontron i.MX6UL N6310 SoM and boards\n \n Add support for i.MX6UL modules from Kontron Electronics GmbH (before\n acquisition: Exceet Electronics) and evalkit boards based on it:\n \n 1. N6310 SOM: i.MX6 UL System-on-Module, a 25x25 mm solderable module\n (LGA pads and pin castellations) with 256 MB RAM, 1 MB NOR-Flash,\n 256 MB NAND and other interfaces,\n 2. N6310 S: evalkit, w/wo eMMC, without display,\n 3. N6310 S 43: evalkit with 4.3\" display,\n \n The work is based on Exceet/Kontron source code (GPLv2) with numerous\n changes:\n 1. Reorganize files,\n 2. Rename Exceet -\u003e Kontron,\n 3. Rename models/compatibles to match newest Kontron product naming,\n 4. Fix coding style errors and adjust to device tree coding guidelines,\n 5. Fix DTC warnings,\n 6. Extend compatibles so eval boards inherit the SoM compatible,\n 7. Use defines instead of GPIO and interrupt flag values,\n 8. Use proper vendor compatible for Macronix SPI NOR,\n 9. Replace deprecated bindings with proper ones,\n 10. Sort nodes alphabetically,\n 11. Remove Admatec display nodes (not yet supported).\n \n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Frieder Schrempf \u003cfrieder.schrempf@kontron.de\u003e\n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 8d5fc0b95fd1542c575ac64457bd17d74a206497\nAuthor: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\nDate: Thu Aug 22 08:02:37 2019 +0200\n\n dt-bindings: arm: fsl: Add Kontron i.MX6UL N6310 compatibles\n \n Add the compatibles for Kontron i.MX6UL N6310 SoM and boards.\n \n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 730fd9d736290b011b9c344d5e26ed29900a2d94\nAuthor: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\nDate: Thu Aug 22 08:02:36 2019 +0200\n\n dt-bindings: eeprom: at25: Add Anvo ANV32E61W\n \n Document the compatible for ANV32E61W 64kb Serial SPI non-volatile SRAM.\n Although it is a SRAM device, it can be accessed through EEPROM\n interface. At least until there is no proper SRAM driver support for\n it.\n \n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 082b1ed591b177cb55f885090e42067b3f3093c6\nAuthor: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\nDate: Thu Aug 22 08:02:35 2019 +0200\n\n dt-bindings: vendor-prefixes: Add Anvo-Systems\n \n Add vendor prefix for Anvo-Systems Dresden GmbH.\n \n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 968c6f4b3c21e9fae9b49cc91214a01b4827625d\nAuthor: Guido Günther \u003cagx@sigxcpu.org\u003e\nDate: Wed Aug 21 18:33:04 2019 +0200\n\n soc: imx: gpcv2: Print the correct error code\n \n The current code prints \u0027ret\u0027 (thus 0) while it should use \u0027err\u0027.\n \n Signed-off-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n Reviewed-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit ffe0f9a0baf9f10135ea272ef4d944179d106d59\nAuthor: Jon Nettleton \u003cjon@solid-run.com\u003e\nDate: Wed Aug 21 11:20:18 2019 +0300\n\n arm64: dts: fsl: add support for Hummingboard Pulse\n \n The SolidRun Hummingboard Pulse carrier board carries the SolidRun\n i.MX8MQ based SOM.\n \n Notably missing is PCIe support that depends on analog PLLOUT clock.\n Current imx clk driver does not support this clock.\n \n Signed-off-by: Jon Nettleton \u003cjon@solid-run.com\u003e\n Signed-off-by: Baruch Siach \u003cbaruch@tkos.co.il\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 0a6a928300df3cc11047b216c609c75e51aa5729\nAuthor: Baruch Siach \u003cbaruch@tkos.co.il\u003e\nDate: Wed Aug 21 11:20:19 2019 +0300\n\n dt-bindings: arm: fsl: add Hummingboard Pulse\n \n Add binding documentation for the SolidRun Hummingboard Pulse board.\n \n Signed-off-by: Baruch Siach \u003cbaruch@tkos.co.il\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 801592402c392acb0f2ead971d1e8f0e0565b60e\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Aug 20 18:39:36 2019 -0700\n\n ARM: dts: vf610-zii-cfu1: Slow I2C0 down to 100 kHz\n \n Fiber-optic modules attached to the bus are only rated to work at\n 100 kHz, so decrease the bus frequency to accommodate that.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Shawn Guo \u003cshawnguo@kernel.org\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: linux-kernel@vger.kernel.org\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit afd3b35fdb0dda18192d1ea1435e609f329c95a8\nAuthor: Song Hui \u003chui.song_1@nxp.com\u003e\nDate: Tue Aug 20 13:54:38 2019 +0800\n\n arm64: dts: ls1088a: update gpio compatible\n \n add ls1088a gpio specify compatible.\n \n Signed-off-by: Song Hui \u003chui.song_1@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 760e548e7f885d89bf2dfab4838df9379edd19fc\nAuthor: Peng Fan \u003cpeng.fan@nxp.com\u003e\nDate: Tue Aug 20 01:55:07 2019 +0000\n\n clk: imx: imx8mn: fix audio pll setting\n \n The AUDIO PLL max support 650M, so the original clk settings violate\n spec. This patch makes the output 786432000 -\u003e 393216000,\n and 722534400 -\u003e 361267200 to aligned with NXP vendor kernel without any\n impact on audio functionality and go within 650MHz PLL limit.\n \n Signed-off-by: Peng Fan \u003cpeng.fan@nxp.com\u003e\n Reviewed-by: Shengjiu Wang \u003cshengjiu.wang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 90771e50d5bd020a46102275e2bfcc0f60f86958\nAuthor: Gary Bisson \u003cgary.bisson@boundarydevices.com\u003e\nDate: Mon Aug 19 19:26:06 2019 +0200\n\n arm64: dts: imx: Add i.mx8mq nitrogen8m basic dts support\n \n Add basic dts support for i.MX8MQ NITROGEN8M.\n \n Signed-off-by: Gary Bisson \u003cgary.bisson@boundarydevices.com\u003e\n Signed-off-by: Troy Kisky \u003ctroy.kisky@boundarydevices.com\u003e\n [Dafna: porting vendor\u0027s code to mainline]\n Signed-off-by: Dafna Hirschfeld \u003cdafna.hirschfeld@collabora.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit e0496bfd4887b7a046f8fda096fb1f881c00270e\nAuthor: Gary Bisson \u003cgary.bisson@boundarydevices.com\u003e\nDate: Mon Aug 19 19:26:05 2019 +0200\n\n dt-bindings: arm: imx: add imx8mq nitrogen support\n \n The Nitrogen8M is an ARM based single board computer (SBC)\n designed to leverage the full capabilities of NXP’s i.MX8M\n Quad processor.\n \n Signed-off-by: Gary Bisson \u003cgary.bisson@boundarydevices.com\u003e\n Signed-off-by: Troy Kisky \u003ctroy.kisky@boundarydevices.com\u003e\n [Dafna: porting vendor\u0027s code to mainline]\n Signed-off-by: Dafna Hirschfeld \u003cdafna.hirschfeld@collabora.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit e58915179f3f4a839ea3b9aeae1c1e13e98e33b1\nAuthor: Thor Thayer \u003cthor.thayer@linux.intel.com\u003e\nDate: Mon Aug 19 15:48:08 2019 -0500\n\n fpga: altera-cvp: Add Stratix10 (V2) Support\n \n Add Stratix10 specific functions that use a credit mechanism\n to throttle data to the CvP FIFOs. Add a private structure\n with function pointers for V1 vs V2 functions.\n \n Signed-off-by: Thor Thayer \u003cthor.thayer@linux.intel.com\u003e\n Signed-off-by: Moritz Fischer \u003cmdf@kernel.org\u003e\n\ncommit d2083d040a95b923a217377dd51c27d57eecf5e6\nAuthor: Thor Thayer \u003cthor.thayer@linux.intel.com\u003e\nDate: Mon Aug 19 15:48:07 2019 -0500\n\n fpga: altera-cvp: Preparation for V2 parts.\n \n In preparation for adding newer V2 parts that use a FIFO,\n reorganize altera_cvp_chk_error() and change the write\n function to block based.\n V2 parts have a block size matching the FIFO while older\n V1 parts write a 32 bit word at a time.\n \n Signed-off-by: Thor Thayer \u003cthor.thayer@linux.intel.com\u003e\n Signed-off-by: Moritz Fischer \u003cmdf@kernel.org\u003e\n\ncommit eb12511f0d47b4da58cc9fc1e93362081fa3331b\nAuthor: Thor Thayer \u003cthor.thayer@linux.intel.com\u003e\nDate: Mon Aug 19 15:48:06 2019 -0500\n\n fpga: altera-cvp: Discover Vendor Specific offset\n \n Newer Intel FPGAs have different Vendor Specific offsets than\n legacy parts. Use PCI discovery to find the CvP registers.\n Since the register positions remain the same, change the hard\n coded address to a more flexible way of indexing registers\n from the offset.\n Adding new PCI read and write abstraction functions to\n handle the offset (altera_read_config_dword() and\n altera_write_config_dword()).\n \n Signed-off-by: Thor Thayer \u003cthor.thayer@linux.intel.com\u003e\n Signed-off-by: Moritz Fischer \u003cmdf@kernel.org\u003e\n\ncommit 34a33ba609dc19703ab478ab9cb94518e6eb6cc5\nAuthor: Chuanhua Han \u003cchuanhua.han@nxp.com\u003e\nDate: Thu Aug 1 16:31:05 2019 +0800\n\n arm64: dts: ls1088a-qds: Add the spi-flash nodes under the DSPI controller\n \n This patch adds the spi-flash nodes under the DSPI controller for\n ls1088a-qds boards.\n \n Signed-off-by: Chuanhua Han \u003cchuanhua.han@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 60ca9248e47011f86e13e945c2673af0fa24a074\nAuthor: Chuanhua Han \u003cchuanhua.han@nxp.com\u003e\nDate: Thu Aug 1 16:31:04 2019 +0800\n\n arm64: dts: ls1088a: Add the DSPI controller node\n \n This patch adds the DSPI controller node for ls1088a boards.\n \n Signed-off-by: Chuanhua Han \u003cchuanhua.han@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 1e58b6f0cc8fbfb2bc0575aec7a5dea6fc6d01b0\nAuthor: Chuanhua Han \u003cchuanhua.han@nxp.com\u003e\nDate: Thu Aug 1 16:31:03 2019 +0800\n\n dt-bindings: fsl: dspi: Add fsl,ls1088a-dspi compatible string\n \n new compatible string: \"fsl,ls1088a-dspi\".\n \n Signed-off-by: Chuanhua Han \u003cchuanhua.han@nxp.com\u003e\n Acked-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit acf993a0f22215dbc1d39f974e3053d6b4f4e594\nAuthor: André Draszik \u003cgit@andred.net\u003e\nDate: Fri Aug 9 04:12:56 2019 +0100\n\n ARM: imx: stop adjusting ar8031 phy tx delay\n \n Recent changes to the Atheros at803x driver cause\n the approach taken here to stop working because\n commit 6d4cd041f0af\n (\"net: phy: at803x: disable delay only for RGMII mode\")\n and commit cd28d1d6e52e\n (\"net: phy: at803x: Disable phy delay for RGMII mode\")\n fix the AR8031 driver to configure the phy\u0027s (RX/TX)\n delays as per the \u0027phy-mode\u0027 in the device tree.\n \n In particular, the phy tx (and rx) delays are updated\n again as per the \u0027phy-mode\u0027 *after* the code in here\n runs.\n \n Things worked before above commits, because the AR8031\n comes out of reset with RX delay enabled, and the\n at803x driver didn\u0027t touch the delay configuration at\n all when \"rgmii\" mode was selected.\n \n It appears the code in here tries to make device\n trees work that incorrectly specify \"rgmii\", but\n that can\u0027t work any more and it is imperative since\n above commits to have the phy-mode configured\n correctly in the device tree.\n \n I suspect there are a few imx7d based boards using\n the ar8031 phy and phy-mode \u003d \"rgmii\", but given I\n don\u0027t know which ones exactly, I am not in a\n position to update the respective device trees.\n \n Hence this patch is simply removing the superfluous\n code from the imx7d initialisation. An alternative\n could be to add a warning instead, but that would\n penalize all boards that have been updated already.\n \n Signed-off-by: André Draszik \u003cgit@andred.net\u003e\n CC: Russell King \u003clinux@armlinux.org.uk\u003e\n CC: Shawn Guo \u003cshawnguo@kernel.org\u003e\n CC: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\n CC: Pengutronix Kernel Team \u003ckernel@pengutronix.de\u003e\n CC: Fabio Estevam \u003cfestevam@gmail.com\u003e\n CC: NXP Linux Team \u003clinux-imx@nxp.com\u003e\n CC: Kate Stewart \u003ckstewart@linuxfoundation.org\u003e\n CC: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n CC: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n CC: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n CC: linux-arm-kernel@lists.infradead.org\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 929c11a55b9ec08fc21fe47e1f64245c333d9a15\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Thu Aug 22 23:35:03 2019 +0200\n\n ARM: dts: pbab01: correct rtc vendor\n \n The rtc8564 is made by Epson but is similar to the NXP pcf8563. Use the\n correct vendor name.\n \n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 49435c4783868d6252db58f14aedb8a3bca7f3bd\nAuthor: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\nDate: Fri Aug 23 18:34:25 2019 -0700\n\n MAINTAINERS: add proc sysctl KUnit test to PROC SYSCTL section\n \n Add entry for the new proc sysctl KUnit test to the PROC SYSCTL section,\n and add Iurii as a maintainer.\n \n Signed-off-by: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\n Cc: Iurii Zaikin \u003cyzaikin@google.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Acked-by: Luis Chamberlain \u003cmcgrof@kernel.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit 103ebb60074e4c09ebbdd9fd95e85002d85926f5\nAuthor: Iurii Zaikin \u003cyzaikin@google.com\u003e\nDate: Fri Aug 23 18:34:24 2019 -0700\n\n kernel/sysctl-test: Add null pointer test for sysctl.c:proc_dointvec()\n \n KUnit tests for initialized data behavior of proc_dointvec that is\n explicitly checked in the code. Includes basic parsing tests including\n int min/max overflow.\n \n Signed-off-by: Iurii Zaikin \u003cyzaikin@google.com\u003e\n Signed-off-by: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Acked-by: Luis Chamberlain \u003cmcgrof@kernel.org\u003e\n Reviewed-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit ce7526d565568ea9032bccf873835d922297b52e\nAuthor: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\nDate: Fri Aug 23 18:34:23 2019 -0700\n\n MAINTAINERS: add entry for KUnit the unit testing framework\n \n Add myself as maintainer of KUnit, the Linux kernel\u0027s unit testing\n framework.\n \n Signed-off-by: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Reviewed-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit f3abcc8c57a757cc07d89852b3e92a6be40ed0df\nAuthor: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\nDate: Fri Aug 23 18:34:22 2019 -0700\n\n Documentation: kunit: add documentation for KUnit\n \n Add documentation for KUnit, the Linux kernel unit testing framework.\n - Add intro and usage guide for KUnit\n - Add API reference\n \n Signed-off-by: Felix Guo \u003cfelixguoxiuping@gmail.com\u003e\n Signed-off-by: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\n Cc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Reviewed-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit 3c08fc5424988c2e8981679039c86e05793ed47b\nAuthor: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\nDate: Fri Aug 23 18:34:21 2019 -0700\n\n kunit: defconfig: add defconfigs for building KUnit tests\n \n Add defconfig for UML and a fragment that can be used to configure other\n architectures for building KUnit tests. Add option to kunit_tool to use\n a defconfig to create the kunitconfig.\n \n Signed-off-by: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Reviewed-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit 0a53f35d6e5de2fa986a59b13c7ef457f5b6145a\nAuthor: Felix Guo \u003cfelixguoxiuping@gmail.com\u003e\nDate: Fri Aug 23 18:34:20 2019 -0700\n\n kunit: tool: add Python wrappers for running KUnit tests\n \n The ultimate goal is to create minimal isolated test binaries; in the\n meantime we are using UML to provide the infrastructure to run tests, so\n define an abstract way to configure and run tests that allow us to\n change the context in which tests are built without affecting the user.\n This also makes pretty and dynamic error reporting, and a lot of other\n nice features easier.\n \n kunit_config.py:\n - parse .config and Kconfig files.\n \n kunit_kernel.py: provides helper functions to:\n - configure the kernel using kunitconfig.\n - build the kernel with the appropriate configuration.\n - provide function to invoke the kernel and stream the output back.\n \n kunit_parser.py: parses raw logs returned out by kunit_kernel and\n displays them in a user friendly way.\n \n test_data/*: samples of test data for testing kunit.py, kunit_config.py,\n etc.\n \n Signed-off-by: Felix Guo \u003cfelixguoxiuping@gmail.com\u003e\n Signed-off-by: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Reviewed-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit b12cc8c409320ab0411a178ef36ad1cf2e17acd7\nAuthor: Avinash Kondareddy \u003cakndr41@gmail.com\u003e\nDate: Fri Aug 23 18:34:19 2019 -0700\n\n kunit: test: add tests for KUnit managed resources\n \n Add unit tests for KUnit managed resources. KUnit managed resources\n (struct kunit_resource) are resources that are automatically cleaned up\n at the end of a KUnit test, similar to the concept of devm_* managed\n resources.\n \n Signed-off-by: Avinash Kondareddy \u003cakndr41@gmail.com\u003e\n Signed-off-by: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Reviewed-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit a088934bd5167d6724630379043be1e1661fb4c0\nAuthor: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\nDate: Fri Aug 23 18:34:18 2019 -0700\n\n kunit: test: add the concept of assertions\n \n Add support for assertions which are like expectations except the test\n terminates if the assertion is not satisfied.\n \n The idea with assertions is that you use them to state all the\n preconditions for your test. Logically speaking, these are the premises\n of the test case, so if a premise isn\u0027t true, there is no point in\n continuing the test case because there are no conclusions that can be\n drawn without the premises. Whereas, the expectation is the thing you\n are trying to prove. It is not used universally in x-unit style test\n frameworks, but I really like it as a convention. You could still\n express the idea of a premise using the above idiom, but I think\n KUNIT_ASSERT_* states the intended idea perfectly.\n \n Signed-off-by: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Reviewed-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit 6852cbf2064d37a6373709d21a9a31be4623fe69\nAuthor: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\nDate: Fri Aug 23 18:34:17 2019 -0700\n\n kunit: test: add tests for kunit test abort\n \n Add KUnit tests for the KUnit test abort mechanism (see preceding\n commit). Add tests both for general try catch mechanism as well as\n non-architecture specific mechanism.\n \n Signed-off-by: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Reviewed-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit 0461d027c83d08509d60d20bdb00abdb8c6654c6\nAuthor: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\nDate: Fri Aug 23 18:34:16 2019 -0700\n\n kunit: test: add support for test abort\n \n Add support for aborting/bailing out of test cases, which is needed for\n implementing assertions.\n \n An assertion is like an expectation, but bails out of the test case\n early if the assertion is not met. The idea with assertions is that you\n use them to state all the preconditions for your test. Logically\n speaking, these are the premises of the test case, so if a premise isn\u0027t\n true, there is no point in continuing the test case because there are no\n conclusions that can be drawn without the premises. Whereas, the\n expectation is the thing you are trying to prove.\n \n Signed-off-by: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Reviewed-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit 451efcfcc09c0e124f6b39fc584ce2afc124828e\nAuthor: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\nDate: Fri Aug 23 18:34:15 2019 -0700\n\n objtool: add kunit_try_catch_throw to the noreturn list\n \n Fix the following warning seen on GCC 7.3:\n kunit/test-test.o: warning: objtool: kunit_test_unsuccessful_try() falls through to next function kunit_test_catch()\n \n kunit_try_catch_throw is a function added in the following patch in this\n series; it allows KUnit, a unit testing framework for the kernel, to\n bail out of a broken test. As a consequence, it is a new __noreturn\n function that objtool thinks is broken (as seen above). So fix this\n warning by adding kunit_try_catch_throw to objtool\u0027s noreturn list.\n \n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\n Acked-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Link: https://www.spinics.net/lists/linux-kbuild/msg21708.html\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit 8faf220835e89fc65a643e1fa8e318efb8880a53\nAuthor: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\nDate: Fri Aug 23 18:34:14 2019 -0700\n\n kunit: test: add initial tests\n \n Add a test for string stream along with a simpler example.\n \n Signed-off-by: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Reviewed-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit fe539dde39e8f7cb8dd3f31dc3a355449e7ec0a5\nAuthor: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\nDate: Fri Aug 23 18:34:13 2019 -0700\n\n kbuild: enable building KUnit\n \n KUnit is a new unit testing framework for the kernel and when used is\n built into the kernel as a part of it. Add KUnit to the root Kconfig and\n Makefile to allow it to be actually built.\n \n Signed-off-by: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\n Acked-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Cc: Michal Marek \u003cmichal.lkml@markovi.net\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Reviewed-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit f8c6f4eb850dbd87a1f568176bbf200a2cba25c9\nAuthor: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\nDate: Fri Aug 23 18:34:12 2019 -0700\n\n kunit: test: add the concept of expectations\n \n Add support for expectations, which allow properties to be specified and\n then verified in tests.\n \n Signed-off-by: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Reviewed-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit 630a7c17149a86067f4c7b1272c2a23298f3fbab\nAuthor: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\nDate: Fri Aug 23 18:34:11 2019 -0700\n\n kunit: test: add assertion printing library\n \n Add `struct kunit_assert` and friends which provide a structured way to\n capture data from an expectation or an assertion (introduced later in\n the series) so that it may be printed out in the event of a failure.\n \n Signed-off-by: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\n Reviewed-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit 3c49770597ffb8c010da7710efac833429fc7f2b\nAuthor: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\nDate: Fri Aug 23 18:34:10 2019 -0700\n\n kunit: test: add string_stream a std::stream like string builder\n \n A number of test features need to do pretty complicated string printing\n where it may not be possible to rely on a single preallocated string\n with parameters.\n \n So provide a library for constructing the string as you go similar to\n C++\u0027s std::string. string_stream is really just a string builder,\n nothing more.\n \n Signed-off-by: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Reviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit fe33c162065bdf5e5934f08a660b13dac893f256\nAuthor: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\nDate: Fri Aug 23 18:34:09 2019 -0700\n\n kunit: test: add test resource management API\n \n Create a common API for test managed resources like memory and test\n objects. A lot of times a test will want to set up infrastructure to be\n used in test cases; this could be anything from just wanting to allocate\n some memory to setting up a driver stack; this defines facilities for\n creating \"test resources\" which are managed by the test infrastructure\n and are automatically cleaned up at the conclusion of the test.\n \n Signed-off-by: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Reviewed-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit 772b408fdc942d58b6545eda701f8126d95df679\nAuthor: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\nDate: Fri Aug 23 18:34:08 2019 -0700\n\n kunit: test: add KUnit test runner core\n \n Add core facilities for defining unit tests; this provides a common way\n to define test cases, functions that execute code which is under test\n and determine whether the code under test behaves as expected; this also\n provides a way to group together related test cases in test suites (here\n we call them test_modules).\n \n Just define test cases and how to execute them for now; setting\n expectations on code will be defined later.\n \n Signed-off-by: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Reviewed-by: Luis Chamberlain \u003cmcgrof@kernel.org\u003e\n Reviewed-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit b4742e6682d5809ddf4d0a63cb57e629e815ec63\nAuthor: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\nDate: Wed Jul 24 10:17:11 2019 +0200\n\n MIPS: dts: mscc: describe the PTP ready interrupt\n \n This patch adds a description of the PTP ready interrupt, which can be\n triggered when a PTP timestamp is available on an hardware FIFO.\n \n Signed-off-by: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: davem@davemloft.net\n Cc: richardcochran@gmail.com\n Cc: alexandre.belloni@bootlin.com\n Cc: UNGLinuxDriver@microchip.com\n Cc: ralf@linux-mips.org\n Cc: jhogan@kernel.org\n Cc: netdev@vger.kernel.org\n Cc: linux-mips@vger.kernel.org\n Cc: thomas.petazzoni@bootlin.com\n Cc: allan.nielsen@microchip.com\n\ncommit 048dc3abe82738567435d9caa106a20eb417b28a\nAuthor: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\nDate: Wed Jul 24 10:17:09 2019 +0200\n\n MIPS: dts: mscc: describe the PTP register range\n \n This patch adds one register range within the mscc,vsc7514-switch node,\n to describe the PTP registers.\n \n Signed-off-by: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: davem@davemloft.net\n Cc: richardcochran@gmail.com\n Cc: alexandre.belloni@bootlin.com\n Cc: UNGLinuxDriver@microchip.com\n Cc: ralf@linux-mips.org\n Cc: jhogan@kernel.org\n Cc: netdev@vger.kernel.org\n Cc: linux-mips@vger.kernel.org\n Cc: thomas.petazzoni@bootlin.com\n Cc: allan.nielsen@microchip.com\n\ncommit ed90302be64a53d9031c8ce05428c358b16a5d96\nAuthor: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\nDate: Sat Jul 27 14:04:15 2019 +0200\n\n MIPS: lantiq: update the clock alias\u0027 for the mainline PCIe PHY driver\n \n The mainline PCIe PHY driver has it\u0027s own devicetree node. Update the\n clock alias so the mainline driver finds the clocks.\n \n The first PCIe PHY is located at 0x1f106800 and exists on VRX200, ARX300\n and GRX390.\n The second PCIe PHY is located at 0x1f700400 and exists on ARX300 and\n GRX390.\n The third PCIe PHY is located at 0x1f106a00 and exists onl on GRX390.\n Lantiq\u0027s board support package (called \"UGW\") names these registers\n \"PDI\".\n \n Signed-off-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: devicetree@vger.kernel.org\n Cc: john@phrozen.org\n Cc: kishon@ti.com\n Cc: ralf@linux-mips.org\n Cc: robh+dt@kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: hauke@hauke-m.de\n Cc: mark.rutland@arm.com\n Cc: ms@dev.tdt.de\n\ncommit 12051b318bc3ce5b42d6d786191008284b067d83\nAuthor: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nDate: Mon Jul 29 14:10:12 2019 -0700\n\n mips: avoid explicit UB in assignment of mips_io_port_base\n \n The code in question is modifying a variable declared const through\n pointer manipulation. Such code is explicitly undefined behavior, and\n is the lone issue preventing malta_defconfig from booting when built\n with Clang:\n \n If an attempt is made to modify an object defined with a const-qualified\n type through use of an lvalue with non-const-qualified type, the\n behavior is undefined.\n \n LLVM is removing such assignments. A simple fix is to not declare\n variables const that you plan on modifying. Limiting the scope would be\n a better method of preventing unwanted writes to such a variable.\n \n Further, the code in question mentions \"compiler bugs\" without any links\n to bug reports, so it is difficult to know if the issue is resolved in\n GCC. The patch was authored in 2006, which would have been GCC 4.0.3 or\n 4.1.1. The minimal supported version of GCC in the Linux kernel is\n currently 4.6.\n \n For what its worth, there was UB before the commit in question, it just\n added a barrier and got lucky IRT codegen. I don\u0027t think there\u0027s any\n actual compiler bugs related, just runtime bugs due to UB.\n \n Link: https://github.com/ClangBuiltLinux/linux/issues/610\n Fixes: 966f4406d903 (\"[MIPS] Work around bad code generation for \u003casm/io.h\u003e.\")\n Reported-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Debugged-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Suggested-by: Eli Friedman \u003cefriedma@quicinc.com\u003e\n Signed-off-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Reviewed-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Tested-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: ralf@linux-mips.org\n Cc: jhogan@kernel.org\n Cc: Maciej W. Rozycki \u003cmacro@linux-mips.org\u003e\n Cc: Hassan Naveed \u003chnaveed@wavecomp.com\u003e\n Cc: Stephen Kitt \u003csteve@sk2.org\u003e\n Cc: Serge Semin \u003cfancer.lancer@gmail.com\u003e\n Cc: Mike Rapoport \u003crppt@linux.ibm.com\u003e\n Cc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n Cc: Michal Hocko \u003cmhocko@suse.com\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: clang-built-linux@googlegroups.com\n\ncommit 47e4937a4a7ca4184fd282791dfee76c6799966a\nAuthor: Gao Xiang \u003chsiangkao@aol.com\u003e\nDate: Fri Aug 23 05:36:59 2019 +0800\n\n erofs: move erofs out of staging\n \n EROFS filesystem has been merged into linux-staging for a year.\n \n EROFS is designed to be a better solution of saving extra storage\n space with guaranteed end-to-end performance for read-only files\n with the help of reduced metadata, fixed-sized output compression\n and decompression inplace technologies.\n \n In the past year, EROFS was greatly improved by many people as\n a staging driver, self-tested, betaed by a large number of our\n internal users, successfully applied to almost all in-service\n HUAWEI smartphones as the part of EMUI 9.1 and proven to be stable\n enough to be moved out of staging.\n \n EROFS is a self-contained filesystem driver. Although there are\n still some TODOs to be more generic, we have a dedicated team\n actively keeping on working on EROFS in order to make it better\n with the evolution of Linux kernel as the other in-kernel filesystems.\n \n As Pavel suggested, it\u0027s better to do as one commit since git\n can do moves and all histories will be saved in this way.\n \n Let\u0027s promote it from staging and enhance it more actively as\n a \"real\" part of kernel for more wider scenarios!\n \n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Alexander Viro \u003cviro@zeniv.linux.org.uk\u003e\n Cc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n Cc: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Cc: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Cc: Pavel Machek \u003cpavel@denx.de\u003e\n Cc: David Sterba \u003cdsterba@suse.cz\u003e\n Cc: Amir Goldstein \u003camir73il@gmail.com\u003e\n Cc: Christoph Hellwig \u003chch@infradead.org\u003e\n Cc: Darrick J . Wong \u003cdarrick.wong@oracle.com\u003e\n Cc: Dave Chinner \u003cdavid@fromorbit.com\u003e\n Cc: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n Cc: Jan Kara \u003cjack@suse.cz\u003e\n Cc: Richard Weinberger \u003crichard@nod.at\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Chao Yu \u003cyuchao0@huawei.com\u003e\n Cc: Miao Xie \u003cmiaoxie@huawei.com\u003e\n Cc: Li Guifu \u003cbluce.liguifu@huawei.com\u003e\n Cc: Fang Wei \u003cfangwei1@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190822213659.5501-1-hsiangkao@aol.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 86aa04f4c2215912fcca6728f2dcf174f7e31fc4\nAuthor: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nDate: Sun Aug 18 21:46:04 2019 -0500\n\n mtd: rawnand: Fix a memory leak bug\n \n In nand_scan_bbt(), a temporary buffer \u0027buf\u0027 is allocated through\n vmalloc(). However, if check_create() fails, \u0027buf\u0027 is not deallocated,\n leading to a memory leak bug. To fix this issue, free \u0027buf\u0027 before\n returning the error.\n \n Signed-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n\ncommit d83aef09aaa50bdafbb32981859128299abf32eb\nAuthor: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nDate: Sun Aug 18 10:52:49 2019 -0500\n\n mtd: onenand_base: Fix a memory leak bug\n \n In onenand_scan(), if CONFIG_MTD_ONENAND_VERIFY_WRITE is defined,\n \u0027this-\u003everify_buf\u0027 is allocated through kzalloc(). However, it is not\n deallocated in the following execution, if the allocation for\n \u0027this-\u003eoob_buf\u0027 fails, leading to a memory leak bug. To fix this issue,\n free \u0027this-\u003everify_buf\u0027 before returning the error.\n \n Signed-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n\ncommit 52988009843160c5b366b4082ed6df48041c655c\nAuthor: Xiaolin Zhang \u003cxiaolin.zhang@intel.com\u003e\nDate: Fri Aug 23 14:57:31 2019 +0800\n\n drm/i915: to make vgpu ppgtt notificaiton as atomic operation\n \n vgpu ppgtt notification was split into 2 steps, the first step is to\n update PVINFO\u0027s pdp register and then write PVINFO\u0027s g2v_notify register\n with action code to tirgger ppgtt notification to GVT side.\n \n currently these steps were not atomic operations due to no any protection,\n so it is easy to enter race condition state during the MTBF, stress and\n IGT test to cause GPU hang.\n \n the solution is to add a lock to make vgpu ppgtt notication as atomic\n operation.\n \n Cc: stable@vger.kernel.org\n Signed-off-by: Xiaolin Zhang \u003cxiaolin.zhang@intel.com\u003e\n Acked-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/1566543451-13955-1-git-send-email-xiaolin.zhang@intel.com\n\ncommit 23ca7d2503d895c80b65e4321fc6cc678c7642f2\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Jun 28 12:32:20 2019 +0200\n\n PM / devfreq: tegra20: add COMMON_CLK dependency\n \n Compile-testing the new driver on platforms without CONFIG_COMMON_CLK\n leads to a link error:\n \n drivers/devfreq/tegra20-devfreq.o: In function `tegra_devfreq_target\u0027:\n tegra20-devfreq.c:(.text+0x288): undefined reference to `clk_set_min_rate\u0027\n \n Add a dependency on COMMON_CLK to avoid this.\n \n Fixes: 1d39ee8dad6d (\"PM / devfreq: Introduce driver for NVIDIA Tegra20\")\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Reviewed-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 5f8669639f3ec6723ebebd83aaccfbea4b5c362b\nAuthor: Lukasz Luba \u003cl.luba@partner.samsung.com\u003e\nDate: Fri Jun 14 11:53:05 2019 +0200\n\n PM / devfreq: events: add Exynos PPMU new events\n \n Define new performance events supported by Exynos5422 SoC counters.\n The counters are built-in in Dynamic Memory Controller and provide\n information regarding memory utilization.\n \n Signed-off-by: Lukasz Luba \u003cl.luba@partner.samsung.com\u003e\n Acked-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 7544fd7f384591038646d3cd9efb311ab4509e24\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Fri Jun 21 18:39:49 2019 -0300\n\n PM / devfreq: Fix kernel oops on governor module load\n \n A bit unexpectedly (but still documented), request_module may\n return a positive value, in case of a modprobe error.\n This is currently causing issues in the devfreq framework.\n \n When a request_module exits with a positive value, we currently\n return that via ERR_PTR. However, because the value is positive,\n it\u0027s not a ERR_VALUE proper, and is therefore treated as a\n valid struct devfreq_governor pointer, leading to a kernel oops.\n \n Fix this by returning -EINVAL if request_module returns a positive\n value.\n \n Fixes: b53b0128052ff (\"PM / devfreq: Fix static checker warning in try_then_request_governor\")\n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 0a453aca94bd7886bca86cc78ce2771f55152232\nAuthor: Gaël PORTAY \u003cgael.portay@collabora.com\u003e\nDate: Thu May 2 14:40:13 2019 -0400\n\n PM / devfreq: rk3399_dmc: Fix spelling typo\n \n Reorder \u0027i\u0027 and \u0027v\u0027 in \"drvier\".\n \n Signed-off-by: Gaël PORTAY \u003cgael.portay@collabora.com\u003e\n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 2c09083293f46eb1a17be8dd7e92efbdcab43fc8\nAuthor: Gaël PORTAY \u003cgael.portay@collabora.com\u003e\nDate: Thu May 2 14:27:36 2019 -0400\n\n PM / devfreq: Fix spelling typo\n \n Add missing \u0027r\u0027 in \"monitoing\".\n \n Signed-off-by: Gaël PORTAY \u003cgael.portay@collabora.com\u003e\n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit d196175ed8f45248b54bf5c2e7c05ac0e1e97d70\nAuthor: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nDate: Thu May 2 02:38:15 2019 +0300\n\n PM / devfreq: Introduce driver for NVIDIA Tegra20\n \n Add devfreq driver for NVIDIA Tegra20 SoC\u0027s. The driver periodically\n reads out Memory Controller counters and adjusts memory frequency based\n on the memory clients activity.\n \n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n [Removed MAINTAINERS updates by MyungJoo so that it can be sent elsewhere.]\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 23601752911b5dac91207859e0ab77bd8c77545c\nAuthor: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nDate: Thu May 2 02:38:14 2019 +0300\n\n PM / devfreq: tegra: Rename tegra-devfreq.c to tegra30-devfreq.c\n \n In order to reflect that driver serves NVIDIA Tegra30 and later SoC\n generations, let\u0027s rename the driver\u0027s source file to \"tegra30-devfreq.c\".\n This will make driver files to look more consistent after addition of a\n driver for Tegra20.\n \n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 35f8dbc727212bc5a49b90961d98f9ef596e2072\nAuthor: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nDate: Thu May 2 02:38:13 2019 +0300\n\n PM / devfreq: tegra: Enable COMPILE_TEST for the driver\n \n The driver\u0027s compilation doesn\u0027t have any specific dependencies, hence\n the COMPILE_TEST option can be supported in Kconfig.\n \n Signed-off-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 1ac347488529939d74f5e2d17948825b5bc6f5e6\nAuthor: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nDate: Thu May 2 02:38:12 2019 +0300\n\n PM / devfreq: tegra: Support Tegra30\n \n The devfreq driver can be used on Tegra30 without any code change and\n it works perfectly fine, the default Tegra124 parameters are good enough\n for Tegra30.\n \n Signed-off-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n [Modified by MyungJoo to depends on Tegra30/114/124/210 only]\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 546ff09381370015ffaf4d139aae671a0bb66af2\nAuthor: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nDate: Thu May 2 02:38:11 2019 +0300\n\n PM / devfreq: tegra: Reconfigure hardware on governor\u0027s restart\n \n Move hardware configuration to governor\u0027s start/resume methods.\n This allows to re-initialize hardware counters and reconfigure\n cleanly if governor was stopped/paused. That is needed because we\n are not aware of all hardware changes that happened while governor\n was stopped and the paused state may get out of sync with reality,\n hence it\u0027s better to start with a clean slate after the pause. In\n a result there is no memory bandwidth starvation after resume from\n suspend-to-ram that results in display controller underflowing that\n happens on resume because of improper decision made by devfreq about\n the required memory frequency. This change also cleans up code a tad\n by moving hardware-configuration code into a single location.\n \n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 5a7e10c89501ec69adc5c39b24d7d69a223bf275\nAuthor: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nDate: Thu May 2 02:38:10 2019 +0300\n\n PM / devfreq: tegra: Move governor registration to driver\u0027s probe\n \n There is no need to register the ACTMON\u0027s governor separately from\n the driver, hence let\u0027s move the registration into the driver\u0027s probe\n function for consistency and to make code cleaner a tad.\n \n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 386789ebbdd3a204ebe16b7a6277f7a45823549d\nAuthor: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nDate: Thu May 2 02:38:09 2019 +0300\n\n PM / devfreq: tegra: Mark ACTMON\u0027s governor as immutable\n \n The ACTMON\u0027s governor supports only the Tegra\u0027s devfreq device and there\n is no need to use any other governor, hence let\u0027s mark Tegra governor as\n immutable to permanently stick it with Tegra\u0027s devfreq device.\n \n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 151531f79a2daa30141c9fdb23e0cff00195b17a\nAuthor: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nDate: Thu May 2 02:38:08 2019 +0300\n\n PM / devfreq: tegra: Avoid inconsistency of current frequency value\n \n The frequency value potentially could change in-between. It doesn\u0027t\n cause any real problem at all right now, but that could change in the\n future. Hence let\u0027s avoid the inconsistency.\n \n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 8fda5c1fab4f8df22bad43ba93a5ba83594ac256\nAuthor: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nDate: Thu May 2 02:38:07 2019 +0300\n\n PM / devfreq: tegra: Clean up driver\u0027s probe / remove\n \n Reset hardware, disable ACTMON clock, release OPP\u0027s and handle all\n possible error cases correctly, maintaining the correct tear down\n order. Also use devm_platform_ioremap_resource() which is now available\n in the kernel.\n \n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 7514dd05eaae5b8cf0b2af34f75a048ee023fe2c\nAuthor: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nDate: Thu May 2 02:38:06 2019 +0300\n\n PM / devfreq: tegra: Properly disable interrupts\n \n There is no guarantee that interrupt handling isn\u0027t running in parallel\n with tegra_actmon_disable_interrupts(), hence it is necessary to protect\n DEV_CTRL register accesses and clear IRQ status with ACTMON\u0027s IRQ being\n disabled in the Interrupt Controller in order to ensure that device\n interrupt is indeed being disabled.\n \n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit dd3f2616bb99f1d2bfd0536074824722e953f6be\nAuthor: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nDate: Thu May 2 02:38:05 2019 +0300\n\n PM / devfreq: tegra: Drop primary interrupt handler\n \n There is no real need in the primary interrupt handler, hence move\n everything to the secondary (threaded) handler. In a result locking\n is consistent now and there are no potential races with the interrupt\n handler because it is protected with the devfreq\u0027s mutex.\n \n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit b061312f4bca20ed60246f7fe2e1ffa247ce5bf7\nAuthor: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nDate: Thu May 2 02:38:04 2019 +0300\n\n PM / devfreq: tegra: Don\u0027t set EMC clock rate to maximum on probe\n \n There is no real benefit from doing so, hence let\u0027s drop that rate setting\n for consistency.\n \n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 30af44fae8bdccfb21a45dfb4029f954f618d2e3\nAuthor: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nDate: Thu May 2 02:38:03 2019 +0300\n\n PM / devfreq: tegra: Don\u0027t ignore clk errors\n \n The clk_set_min_rate() could fail and in this case clk_set_rate() sets\n rate to 0, which may drop EMC rate to minimum and make machine very\n difficult to use.\n \n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit ed2a8dd22a752af60282af38a47d26b227f772a7\nAuthor: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nDate: Thu May 2 02:38:02 2019 +0300\n\n PM / devfreq: tegra: Replace write memory barrier with the read barrier\n \n The write memory barrier isn\u0027t needed because the BUS buffer is flushed\n by read after write that happens after the removed wmb(), we will also\n use readl() instead of the relaxed version to ensure that read is indeed\n completed.\n \n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit efe9043db49649c1f6c21df8a3860689c2035911\nAuthor: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nDate: Thu May 2 02:38:01 2019 +0300\n\n PM / devfreq: tegra: Replace readl-writel with relaxed versions\n \n There is no need to insert memory barrier on each readl/writel\n invocation, hence use the relaxed versions.\n \n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 62bacb06b9f08965c4ef10e17875450490c948c0\nAuthor: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nDate: Thu May 2 02:38:00 2019 +0300\n\n PM / devfreq: tegra: Fix kHz to Hz conversion\n \n The kHz to Hz is incorrectly converted in a few places in the code,\n this results in a wrong frequency being calculated because devfreq core\n uses OPP frequencies that are given in Hz to clamp the rate, while\n tegra-devfreq gives to the core value in kHz and then it also expects to\n receive value in kHz from the core. In a result memory freq is always set\n to a value which is close to ULONG_MAX because of the bug. Hence the EMC\n frequency is always capped to the maximum and the driver doesn\u0027t do\n anything useful. This patch was tested on Tegra30 and Tegra124 SoC\u0027s, EMC\n frequency scaling works properly now.\n \n Cc: \u003cstable@vger.kernel.org\u003e # 4.14+\n Tested-by: Steev Klimaszewski \u003csteev@kali.org\u003e\n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: MyungJoo Ham \u003cmyungjoo.ham@samsung.com\u003e\n\ncommit 75b974a859e5d9a3ceaa3aa03bbc27d404b32231\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Sat Aug 24 00:51:41 2019 +0100\n\n drm/i915/selftests: Teach igt_gpu_fill_dw() to take intel_context\n \n Avoid having to pass around (ctx, engine) everywhere by passing the\n actual intel_context we intend to use. Today we preach this lesson to\n igt_gpu_fill_dw and its callers\u0027 callers.\n \n The immediate benefit for the GEM selftests is that we aim to use the\n GEM context as the control, the source of the engines on which to test\n the GEM context.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823235141.31799-1-chris@chris-wilson.co.uk\n\ncommit 86b1201d7f8a3cf8a0b803fad05a788a85363040\nAuthor: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\nDate: Fri Aug 23 22:38:00 2019 -0400\n\n ext4: fix punch hole for inline_data file systems\n \n If a program attempts to punch a hole on an inline data file, we need\n to convert it to a normal file first.\n \n This was detected using ext4/032 using the adv configuration. Simple\n reproducer:\n \n mke2fs -Fq -t ext4 -O inline_data /dev/vdc\n mount /vdc\n echo \"\" \u003e /vdc/testfile\n xfs_io -c \u0027truncate 33554432\u0027 /vdc/testfile\n xfs_io -c \u0027fpunch 0 1048576\u0027 /vdc/testfile\n umount /vdc\n e2fsck -fy /dev/vdc\n \n Cc: stable@vger.kernel.org\n Signed-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n\ncommit d4ed7463d02aef4b2270ec2a680813cd8b17def7\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Fri Aug 23 20:07:26 2019 +0200\n\n r8169: fix DMA issue on MIPS platform\n \n As reported by Aaro this patch causes network problems on\n MIPS Loongson platform. Therefore revert it.\n \n Fixes: f072218cca5b (\"r8169: remove not needed call to dma_sync_single_for_device\")\n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Reported-by: Aaro Koskinen \u003caaro.koskinen@iki.fi\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit bd1200b79510a68554890af2f48d92be6eb1daf8\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Fri Aug 23 18:47:21 2019 +0300\n\n drop_monitor: Make timestamps y2038 safe\n \n Timestamps are currently communicated to user space as \u0027struct\n timespec\u0027, which is not considered y2038 safe since it uses a 32-bit\n signed value for seconds.\n \n Fix this while the API is still not part of any official kernel release\n by using 64-bit nanoseconds timestamps instead.\n \n Fixes: ca30707dee2b (\"drop_monitor: Add packet alert mode\")\n Fixes: 5e58109b1ea4 (\"drop_monitor: Add support for packet alert mode for hardware drops\")\n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Neil Horman \u003cnhorman@tuxdriver.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit bf1867db9b850fff2dd54a1a117a684a10b8cd90\nAuthor: Dag Moxnes \u003cdag.moxnes@oracle.com\u003e\nDate: Fri Aug 23 16:03:18 2019 +0200\n\n net/rds: Whitelist rdma_cookie and rx_tstamp for usercopy\n \n Add the RDMA cookie and RX timestamp to the usercopy whitelist.\n \n After the introduction of hardened usercopy whitelisting\n (https://lwn.net/Articles/727322/), a warning is displayed when the\n RDMA cookie or RX timestamp is copied to userspace:\n \n kernel: WARNING: CPU: 3 PID: 5750 at\n mm/usercopy.c:81 usercopy_warn+0x8e/0xa6\n [...]\n kernel: Call Trace:\n kernel: __check_heap_object+0xb8/0x11b\n kernel: __check_object_size+0xe3/0x1bc\n kernel: put_cmsg+0x95/0x115\n kernel: rds_recvmsg+0x43d/0x620 [rds]\n kernel: sock_recvmsg+0x43/0x4a\n kernel: ___sys_recvmsg+0xda/0x1e6\n kernel: ? __handle_mm_fault+0xcae/0xf79\n kernel: __sys_recvmsg+0x51/0x8a\n kernel: SyS_recvmsg+0x12/0x1c\n kernel: do_syscall_64+0x79/0x1ae\n \n When the whitelisting feature was introduced, the memory for the RDMA\n cookie and RX timestamp in RDS was not added to the whitelist, causing\n the warning above.\n \n Signed-off-by: Dag Moxnes \u003cdag.moxnes@oracle.com\u003e\n Tested-by: Jenny \u003cjenny.x.xu@oracle.com\u003e\n Acked-by: Santosh Shilimkar \u003csantosh.shilimkar@oracle.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 87cade2997c9210cfeb625957e44b865a89d0c13\nAuthor: Eran Ben Elisha \u003ceranbe@mellanox.com\u003e\nDate: Fri Aug 23 15:34:47 2019 +0300\n\n net/mlx5: Fix return code in case of hyperv wrong size read\n \n Return code value could be non deterministic in case of wrong size read.\n With this patch, if such error occurs, set rc to be -EIO.\n \n In addition, mlx5_hv_config_common() supports reading of\n HV_CONFIG_BLOCK_SIZE_MAX bytes only, fix to early return error with\n bad input.\n \n Fixes: 913d14e86657 (\"net/mlx5: Add wrappers for HyperV PCIe operations\")\n Reported-by: Leon Romanovsky \u003cleon@kernel.org\u003e\n Signed-off-by: Eran Ben Elisha \u003ceranbe@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c7a42eb49212f93a800560662d17d5293960d3c3\nAuthor: Xin Long \u003clucien.xin@gmail.com\u003e\nDate: Fri Aug 23 19:33:03 2019 +0800\n\n net: ipv6: fix listify ip6_rcv_finish in case of forwarding\n \n We need a similar fix for ipv6 as Commit 0761680d5215 (\"net: ipv4: fix\n listify ip_rcv_finish in case of forwarding\") does for ipv4.\n \n This issue can be reprocuded by syzbot since Commit 323ebb61e32b (\"net:\n use listified RX for handling GRO_NORMAL skbs\") on net-next. The call\n trace was:\n \n kernel BUG at include/linux/skbuff.h:2225!\n RIP: 0010:__skb_pull include/linux/skbuff.h:2225 [inline]\n RIP: 0010:skb_pull+0xea/0x110 net/core/skbuff.c:1902\n Call Trace:\n sctp_inq_pop+0x2f1/0xd80 net/sctp/inqueue.c:202\n sctp_endpoint_bh_rcv+0x184/0x8d0 net/sctp/endpointola.c:385\n sctp_inq_push+0x1e4/0x280 net/sctp/inqueue.c:80\n sctp_rcv+0x2807/0x3590 net/sctp/input.c:256\n sctp6_rcv+0x17/0x30 net/sctp/ipv6.c:1049\n ip6_protocol_deliver_rcu+0x2fe/0x1660 net/ipv6/ip6_input.c:397\n ip6_input_finish+0x84/0x170 net/ipv6/ip6_input.c:438\n NF_HOOK include/linux/netfilter.h:305 [inline]\n NF_HOOK include/linux/netfilter.h:299 [inline]\n ip6_input+0xe4/0x3f0 net/ipv6/ip6_input.c:447\n dst_input include/net/dst.h:442 [inline]\n ip6_sublist_rcv_finish+0x98/0x1e0 net/ipv6/ip6_input.c:84\n ip6_list_rcv_finish net/ipv6/ip6_input.c:118 [inline]\n ip6_sublist_rcv+0x80c/0xcf0 net/ipv6/ip6_input.c:282\n ipv6_list_rcv+0x373/0x4b0 net/ipv6/ip6_input.c:316\n __netif_receive_skb_list_ptype net/core/dev.c:5049 [inline]\n __netif_receive_skb_list_core+0x5fc/0x9d0 net/core/dev.c:5097\n __netif_receive_skb_list net/core/dev.c:5149 [inline]\n netif_receive_skb_list_internal+0x7eb/0xe60 net/core/dev.c:5244\n gro_normal_list.part.0+0x1e/0xb0 net/core/dev.c:5757\n gro_normal_list net/core/dev.c:5755 [inline]\n gro_normal_one net/core/dev.c:5769 [inline]\n napi_frags_finish net/core/dev.c:5782 [inline]\n napi_gro_frags+0xa6a/0xea0 net/core/dev.c:5855\n tun_get_user+0x2e98/0x3fa0 drivers/net/tun.c:1974\n tun_chr_write_iter+0xbd/0x156 drivers/net/tun.c:2020\n \n Fixes: d8269e2cbf90 (\"net: ipv6: listify ipv6_rcv() and ip6_rcv_finish()\")\n Fixes: 323ebb61e32b (\"net: use listified RX for handling GRO_NORMAL skbs\")\n Reported-by: syzbot+eb349eeee854e389c36d@syzkaller.appspotmail.com\n Reported-by: syzbot+4a0643a653ac375612d1@syzkaller.appspotmail.com\n Signed-off-by: Xin Long \u003clucien.xin@gmail.com\u003e\n Acked-by: Edward Cree \u003cecree@solarflare.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit fd81d7e946c6bdb86dbf0bd88fee3e1a545e7979\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Thu Aug 22 18:19:28 2019 +0300\n\n leds: Switch to use fwnode instead of be stuck with OF one\n \n There is no need to be stuck with OF node when we may use agnostic\n firmware node instead.\n \n It allows users to get property if needed independently of provider.\n \n Note, some OF parts are left because %pfw [1] is in progress.\n \n [1]: https://lore.kernel.org/patchwork/cover/1054863/\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit ae6c4c70dfd184f69a24342c09678fb161a7cb4c\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Thu Aug 22 18:19:27 2019 +0300\n\n leds: max77650: Switch to fwnode property API\n \n Switch the max77650 from OF to the fwnode property API.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit e7bde56b7446ccda351a216ff55af09a96fea940\nAuthor: Hayes Wang \u003chayeswang@realtek.com\u003e\nDate: Fri Aug 23 15:33:41 2019 +0800\n\n r8152: add a helper function about setting EEE\n \n Add a helper function \"rtl_eee_enable\" for setting EEE. Besides, I\n move r8153_eee_en() and r8153b_eee_en(). And, I remove r8152b_enable_eee(),\n r8153_set_eee(), and r8153b_set_eee().\n \n Signed-off-by: Hayes Wang \u003chayeswang@realtek.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f4a93be689dfc887d73cdaf4245a8a7d031aa912\nAuthor: Hayes Wang \u003chayeswang@realtek.com\u003e\nDate: Fri Aug 23 15:33:40 2019 +0800\n\n r8152: saving the settings of EEE\n \n Saving the settings of EEE to avoid they become the default settings\n after reset_resume().\n \n Signed-off-by: Hayes Wang \u003chayeswang@realtek.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 77715906921251bd9f75bcf4825f176df3f91208\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 23 19:14:55 2019 +0100\n\n drm/i915: Keep drm_i915_file_private around under RCU\n \n Ensure that the drm_i915_file_private continues to exist as we attempt\n to remove a request from its list, which may race with the destruction\n of the file.\n \n \u003c6\u003e [38.380714] [IGT] gem_ctx_create: starting subtest basic-files\n \u003c0\u003e [42.201329] BUG: spinlock bad magic on CPU#0, kworker/u16:0/7\n \u003c4\u003e [42.201356] general protection fault: 0000 [#1] PREEMPT SMP PTI\n \u003c4\u003e [42.201371] CPU: 0 PID: 7 Comm: kworker/u16:0 Tainted: G U 5.3.0-rc5-CI-Patchwork_14169+ #1\n \u003c4\u003e [42.201391] Hardware name: Dell Inc. OptiPlex 745 /0GW726, BIOS 2.3.1 05/21/2007\n \u003c4\u003e [42.201594] Workqueue: i915 retire_work_handler [i915]\n \u003c4\u003e [42.201614] RIP: 0010:spin_dump+0x5a/0x90\n \u003c4\u003e [42.201625] Code: 00 48 8d 88 c0 06 00 00 48 c7 c7 00 71 09 82 e8 35 ef 00 00 48 85 db 44 8b 4d 08 41 b8 ff ff ff ff 48 c7 c1 0b cd 0f 82 74 0e \u003c44\u003e 8b 83 e0 04 00 00 48 8d 8b c0 06 00 00 8b 55 04 48 89 ee 48 c7\n \u003c4\u003e [42.201660] RSP: 0018:ffffc9000004bd80 EFLAGS: 00010202\n \u003c4\u003e [42.201673] RAX: 0000000000000031 RBX: 6b6b6b6b6b6b6b6b RCX: ffffffff820fcd0b\n \u003c4\u003e [42.201688] RDX: 0000000000000000 RSI: ffff88803de266f8 RDI: 00000000ffffffff\n \u003c4\u003e [42.201703] RBP: ffff888038381ff8 R08: 00000000ffffffff R09: 000000006b6b6b6b\n \u003c4\u003e [42.201718] R10: 0000000041cb0b89 R11: 646162206b636f6c R12: ffff88802a618500\n \u003c4\u003e [42.201733] R13: ffff88802b32c288 R14: ffff888038381ff8 R15: ffff88802b32c250\n \u003c4\u003e [42.201748] FS: 0000000000000000(0000) GS:ffff88803de00000(0000) knlGS:0000000000000000\n \u003c4\u003e [42.201765] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n \u003c4\u003e [42.201778] CR2: 00007f2cefc6d180 CR3: 00000000381ee000 CR4: 00000000000006f0\n \u003c4\u003e [42.201793] Call Trace:\n \u003c4\u003e [42.201805] do_raw_spin_lock+0x66/0xb0\n \u003c4\u003e [42.201898] i915_request_retire+0x548/0x7c0 [i915]\n \u003c4\u003e [42.201989] retire_requests+0x4d/0x60 [i915]\n \u003c4\u003e [42.202078] i915_retire_requests+0x144/0x2e0 [i915]\n \u003c4\u003e [42.202169] retire_work_handler+0x10/0x40 [i915]\n \n Recently, in commit 44c22f3f1a0a (\"drm/i915: Serialize insertion into the\n file-\u003emm.request_list\"), we fixed a race on insertion. Now, it appears\n we also have a race with destruction!\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823181455.31910-1-chris@chris-wilson.co.uk\n\ncommit 936ad29de8129747bbfeec8a85e38c8a830bd5da\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Mon Aug 19 18:23:27 2019 -0700\n\n drm/i915/uc: define GuC and HuC FWs for EHL\n \n First uc firmware release for EHL.\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Cc: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Reviewed-by: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Tested-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Reviewed-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820012327.36443-1-daniele.ceraolospurio@intel.com\n\ncommit 2cb831e0f152e483ab797b44787a4ff426267fbc\nAuthor: Nicolas Ferre \u003cnicolas.ferre@microchip.com\u003e\nDate: Fri Aug 23 10:31:58 2019 +0200\n\n mailmap: map old company name to new one @microchip.com\n \n Map my old email address @atmel.com to my new company name. It happened 3\n years ago but I realized the existence of this file recently.\n \n Signed-off-by: Nicolas Ferre \u003cnicolas.ferre@microchip.com\u003e\n Link: https://lore.kernel.org/r/20190823083158.2649-3-nicolas.ferre@microchip.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 44015a8181a54ce0376513645a5a8d1652677019\nAuthor: Nicolas Ferre \u003cnicolas.ferre@microchip.com\u003e\nDate: Fri Aug 23 10:31:57 2019 +0200\n\n MAINTAINERS: at91: remove the TC entry\n \n \"MICROCHIP TIMER COUNTER (TC) AND CLOCKSOURCE DRIVERS\" is better\n removed because one file entry is outdated and basically, the\n maintainer\u0027s pool of Alexandre, Ludovic and myself is better suited.\n \n drivers/misc/atmel_tclib.c file is going away in a patch to come and\n drivers/clocksource/tcb_clksrc.c file is actually named timer-atmel-tcb.c.\n This new name matches the AT91 entry regular expression.\n \n Signed-off-by: Nicolas Ferre \u003cnicolas.ferre@microchip.com\u003e\n Link: https://lore.kernel.org/r/20190823083158.2649-2-nicolas.ferre@microchip.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit ec8005339badea37d18eb19f02efcc4a98dafc52\nAuthor: Nicolas Ferre \u003cnicolas.ferre@microchip.com\u003e\nDate: Fri Aug 23 10:31:56 2019 +0200\n\n MAINTAINERS: at91: Collect all pinctrl/gpio drivers in same entry\n \n Andrei\u0027s address is not valid anymore, collect all pinctrl/gpio\n entries in the common \"PIN CONTROLLER - MICROCHIP AT91\" one\n and remove the PIOBU specific one.\n \n Signed-off-by: Nicolas Ferre \u003cnicolas.ferre@microchip.com\u003e\n Link: https://lore.kernel.org/r/20190823083158.2649-1-nicolas.ferre@microchip.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 9fac85a6db8999922f2cd92dfe2e83e063b31a94\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Fri Aug 23 11:43:45 2019 +0900\n\n ARM: at91: move platform-specific asm-offset.h to arch/arm/mach-at91\n \n \u003cgenerated/at91_pm_data-offsets.h\u003e is only generated and included by\n arch/arm/mach-at91/, so it does not need to reside in the globally\n visible include/generated/.\n \n I renamed it to arch/arm/mach-at91/pm_data-offsets.h since the prefix\n \u0027at91_\u0027 is just redundant in mach-at91/.\n \n My main motivation of this change is to avoid the race condition for\n the parallel build (-j) when CONFIG_IKHEADERS is enabled.\n \n When it is enabled, all the headers under include/ are archived into\n kernel/kheaders_data.tar.xz and exposed in the sysfs.\n \n In the parallel build, we have no idea in which order files are built.\n \n - If at91_pm_data-offsets.h is built before kheaders_data.tar.xz,\n the header will be included in the archive. Probably nobody will\n use it, but it is harmless except that it will increase the archive\n size needlessly.\n \n - If kheaders_data.tar.xz is built before at91_pm_data-offsets.h,\n the header will not be included in the archive. However, in the next\n build, the archive will be re-generated to include the newly-found\n at91_pm_data-offsets.h. This is not nice from the build system point\n of view.\n \n - If at91_pm_data-offsets.h and kheaders_data.tar.xz are built at the\n same time, the corrupted header might be included in the archive,\n which does not look nice either.\n \n This commit fixes the race.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Link: https://lore.kernel.org/r/20190823024346.591-1-yamada.masahiro@socionext.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 38b4e09fbccab6457536563823222921c49601bb\nAuthor: Jens Axboe \u003caxboe@kernel.dk\u003e\nDate: Fri Aug 23 12:49:00 2019 -0600\n\n null_blk: fix inline misuse\n \n You can\u0027t magically mark a function inline and expect that to work.\n \n Fixes: fceb5d1b19cb (\"null_blk: create a helper for zoned devices\")\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 636e83f2f208555c3d19d8b454ebdd8d8f4652cc\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 23 16:39:44 2019 +0100\n\n drm/i915: Flush the existing fence before GGTT read/write\n \n Our fence management is lazy, very lazy. If the user marks an object as\n untiled, we do not immediately flush the fence but merely mark it as\n dirty. On the next use we have to remember to check and remove the fence,\n by which time we hope it is idle and we do not have to wait.\n \n v2: Throw away the old fence on the next ggtt_pin.\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111468\n Fixes: 1f7fd484fff1 (\"drm/i915: Replace i915_vma_put_fence()\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823153944.20630-1-chris@chris-wilson.co.uk\n\ncommit 191797a892c91aec6cdffc4e05696b722d779fe3\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 23 15:14:21 2019 +0100\n\n drm/i915/gtt: Preallocate Braswell top-level page directory\n \n In order for the Braswell top-level PD to remain the same from the time\n of request construction to its submission onto HW, as we may be\n asynchronously rewriting the page tables (thus changing the expected\n register state after having already stored the old addresses in the\n request), the top level PD must be preallocated.\n \n So wave goodbye to our lazy allocation of those 4x2 pages.\n \n v2: A little bit of write-flushing required (presumably it always has\n been required, but now we are more susceptible and it is showing up!)\n \n v3: Put back the forced-PD-reload on every batch, we can\u0027t survive\n without it and explicitly marking the context for PD reload makes\n Braswell turn nasty.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823141421.2398-1-chris@chris-wilson.co.uk\n\ncommit 6dcb85a0ad990455ae7c596e3fc966ad9c1ba9c5\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 23 14:26:46 2019 +0100\n\n drm/i915: Hold irq-off for the entire fake lock period\n \n Sadly lockdep records when the irqs are re-enabled and then marks up the\n fake lock as being irq-unsafe. Our hand is forced and so we must mark up\n the entire fake lock critical section as irq-off.\n \n Hopefully this is the last tweak required!\n \n v2: Not quite, we need to mark the timeline spinlock as irqsafe. That\n was a genuine bug being hidden by the earlier lockdep splat.\n \n Fixes: d67739268cf0 (\"drm/i915/gt: Mark up the nested engine-pm timeline lock as irqsafe\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823132700.25286-2-chris@chris-wilson.co.uk\n\ncommit 0b14d96820d1fd05ea610a0f5f952b34e5a24f2c\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Aug 21 20:30:33 2019 +0300\n\n drm/i915: Use hweight8() for 8bit masks\n \n Use hweight8() instead of hweight32() for 8bit masks. Doesn\u0027t actually\n matter for us since the arch code will go for hweight32() anyway, but\n maybe we stil want to do this for documentation purposes?\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190821173033.24123-5-ville.syrjala@linux.intel.com\n Reviewed-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n\ncommit c08e91323920ec4b3a4d889af3988e300b1902f1\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Aug 21 20:30:32 2019 +0300\n\n drm/i915: s/num_active_crtcs/num_active_pipes/\n \n Set a good example and talk about pipes rather than crtcs.\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190821173033.24123-4-ville.syrjala@linux.intel.com\n Reviewed-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n\ncommit d048a2684a413b33a28be277a7e9cee960ec8bb2\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Aug 21 20:30:31 2019 +0300\n\n drm/i915: Use enum pipe consistently\n \n Replace all \"int pipe\"s with \"enum pipe pipe\"s to make it clear\n what we\u0027re dealing with.\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190821173033.24123-3-ville.syrjala@linux.intel.com\n Reviewed-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n\ncommit e8edae54c593bc9aea949fbf6b2a59fe53cd5d83\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Aug 21 20:30:30 2019 +0300\n\n drm/i915: Unconfuse pipe vs. crtc-\u003eindex in i915_get_crtc_scanoutpos()\n \n The \"pipe\" argument passed in by the vblank code is in fact the crtc\n index. Don\u0027t assume that is the same as the pipe.\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190821173033.24123-2-ville.syrjala@linux.intel.com\n Reviewed-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n\ncommit d06a79d33e0f97373a29decb66791758a6d2e968\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Aug 21 20:30:29 2019 +0300\n\n drm/i915: Use enum pipe instead of crtc index to track active pipes\n \n We may need to eliminate the crtc-\u003eindex \u003d\u003d pipe assumptions from\n the code to support arbitrary pipes being fused off. Start that by\n switching some bitmasks over to using pipe instead of the crtc index.\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190821173033.24123-1-ville.syrjala@linux.intel.com\n Reviewed-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n\ncommit 100f5f7fbc3e23f58511a11a3751dfacf1ccb5a7\nAuthor: Stuart Summers \u003cstuart.summers@intel.com\u003e\nDate: Fri Aug 23 09:03:07 2019 -0700\n\n drm/i915: Expand subslice mask\n \n Currently, the subslice_mask runtime parameter is stored as an\n array of subslices per slice. Expand the subslice mask array to\n better match what is presented to userspace through the\n I915_QUERY_TOPOLOGY_INFO ioctl. The index into this array is\n then calculated:\n slice * subslice stride + subslice index / 8\n \n v2: Fix 32-bit build\n v3: Use new helper function in SSEU workaround warning message\n v4: Use GEM_BUG_ON to force developers to use valid SSEU configurations\n per platform (Chris)\n \n Signed-off-by: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823160307.180813-12-stuart.summers@intel.com\n\ncommit 668df17f594d6f57e822ceee680ace3233e97f02\nAuthor: Stuart Summers \u003cstuart.summers@intel.com\u003e\nDate: Fri Aug 23 09:03:06 2019 -0700\n\n drm/i915: Add new function to copy subslices for a slice\n \n Add a new function to copy subslices for a specified slice\n between intel_sseu structures for the purpose of determining\n power-gate status. Note that currently ss_stride has a max\n of 1.\n \n Signed-off-by: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823160307.180813-11-stuart.summers@intel.com\n\ncommit eaef5b3c411337db73a205d0e009c72f7a39f3a3\nAuthor: Stuart Summers \u003cstuart.summers@intel.com\u003e\nDate: Fri Aug 23 09:03:05 2019 -0700\n\n drm/i915: Refactor instdone loops on new subslice functions\n \n Refactor instdone loops to use the new intel_sseu_has_subslice\n function.\n \n Signed-off-by: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823160307.180813-10-stuart.summers@intel.com\n\ncommit e1210bbfb1f8dc2865e8ebd6884afa895333b6e4\nAuthor: Stuart Summers \u003cstuart.summers@intel.com\u003e\nDate: Fri Aug 23 09:03:04 2019 -0700\n\n drm/i915: Add function to determine if a slice has a subslice\n \n Add a new function to determine whether a particular slice\n has a given subslice.\n \n Signed-off-by: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823160307.180813-9-stuart.summers@intel.com\n\ncommit 6db40ec80f544e289fe836319b20fcc19a454e8d\nAuthor: Stuart Summers \u003cstuart.summers@intel.com\u003e\nDate: Fri Aug 23 09:03:03 2019 -0700\n\n drm/i915: Use subslice stride to set subslices for a given slice\n \n Add a subslice stride calculation when setting subslices. This\n aligns more closely with the userspace expectation of the subslice\n mask structure.\n \n v2: Use local variable for subslice_mask on HSW and\n clean up a few other subslice_mask local variable\n changes\n v3: Add GEM_BUG_ON for ss_stride to prevent array overflow (Chris)\n Split main set function and refactors in intel_device_info.c\n into separate patches (Chris)\n v4: Reduce ss_stride size check when setting subslices per slice\n based on actual expected max stride (Chris)\n Move that GEM_BUG_ON check for the ss_stride out to the patch\n which adds the ss_stride\n v5: Use memcpy instead of looping through each stride index\n \n Signed-off-by: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823160307.180813-8-stuart.summers@intel.com\n\ncommit 9e8a135ed5a414ba4333cda22b01ae77291f6b9b\nAuthor: Stuart Summers \u003cstuart.summers@intel.com\u003e\nDate: Fri Aug 23 09:03:02 2019 -0700\n\n drm/i915: Add function to set subslices\n \n Add a new function to set a set of subslices for a given\n slice.\n \n v2: Fix typo in subslice_mask assignment\n \n Signed-off-by: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823160307.180813-7-stuart.summers@intel.com\n\ncommit 33ee9e8680963fbfa298fec08eed7b45db0a1ebb\nAuthor: Stuart Summers \u003cstuart.summers@intel.com\u003e\nDate: Fri Aug 23 09:03:01 2019 -0700\n\n drm/i915: Use local variables for subslice_mask for device info\n \n When setting up subslice_mask, instead of operating on the slice\n array directly, use a local variable to start bits per slice, then\n use this to set the per slice array in one step.\n \n Signed-off-by: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823160307.180813-6-stuart.summers@intel.com\n\ncommit 49610c377be7f32d948f0970705b4ad015923aa0\nAuthor: Stuart Summers \u003cstuart.summers@intel.com\u003e\nDate: Fri Aug 23 09:03:00 2019 -0700\n\n drm/i915: Add EU stride runtime parameter\n \n Add a new SSEU runtime parameter, eu_stride, which is\n used to mirror the userspace concept of a range of EUs\n per subslice.\n \n This patch simply adds the parameter and updates usage\n in the QUERY_TOPOLOGY_INFO handler.\n \n v2: Add GEM_BUG_ON to make sure eu_stride is valid\n \n Signed-off-by: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823160307.180813-5-stuart.summers@intel.com\n\ncommit 7a200aad1127e98cbf4da9685f652899598fe357\nAuthor: Stuart Summers \u003cstuart.summers@intel.com\u003e\nDate: Fri Aug 23 09:02:59 2019 -0700\n\n drm/i915: Add subslice stride runtime parameter\n \n Add a new parameter, ss_stride, to the runtime info\n structure. This is used to mirror the userspace concept\n of subslice stride, which is a range of subslices per slice.\n \n This patch simply adds the definition and updates usage\n in the QUERY_TOPOLOGY_INFO handler.\n \n v2: Add GEM_BUG_ON to make sure ss_stride is valid\n \n Signed-off-by: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823160307.180813-4-stuart.summers@intel.com\n\ncommit 8b355db99cfbdaea848e05c288d6d80c391f0b13\nAuthor: Stuart Summers \u003cstuart.summers@intel.com\u003e\nDate: Fri Aug 23 09:02:58 2019 -0700\n\n drm/i915: Add function to set SSEU info per platform\n \n Add a new function to allow each platform to set maximum\n slice, subslice, and EU information to reduce code duplication.\n \n Signed-off-by: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823160307.180813-3-stuart.summers@intel.com\n\ncommit 4087f873df1fbf9d9590953119ea78dde6a759dd\nAuthor: Stuart Summers \u003cstuart.summers@intel.com\u003e\nDate: Fri Aug 23 09:02:57 2019 -0700\n\n drm/i915: Use variable for debugfs device status\n \n Use a local variable to find SSEU runtime information\n in various debugfs functions.\n \n v2: Remove extra line breaks per feedback from Chris\n \n Signed-off-by: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823160307.180813-2-stuart.summers@intel.com\n\ncommit 8b2c858240aca43c59fc7762f10754354406883d\nAuthor: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\nDate: Thu Jul 25 02:53:52 2019 -0700\n\n ice: Don\u0027t allow VSI to remove unassociated ucast filter\n \n If a VSI is not using a unicast filter or did not configure that\n particular unicast filter, driver should not allow it to be removed\n by the rogue VSI.\n \n Signed-off-by: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit bbb968e8b34c2cf5ff99d0dfdb73639070b3dee1\nAuthor: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\nDate: Thu Jul 25 02:53:51 2019 -0700\n\n ice: Fix issues updating VSI MAC filters\n \n VSI, especially VF could request to add or remove filter for another VSI,\n driver should really guide such request and disallow it.\n However, instead of returning error for such malicious request, driver\n can simply return success.\n \n In addition, we are not tracking number of MAC filters configured per\n VF correctly - and this leads to issue updating VF MAC filters whenever\n they were removed and re-configured via bringing VF interface down and\n up. Also, since VF could send request to update multiple MAC filters at\n once, driver should program those filters individually in the switch, in\n order to determine which action resulted to error, and communicate\n accordingly to the VF.\n \n So, with this changes, we now track number of filters added right from\n when VF resources allocation is done, and could properly add filters for\n both trusted and non_trusted VFs, without MAC filters mis-match issue in\n the switch...\n \n Also refactor code, so that driver can use new function to add or remove\n MAC filters.\n \n Signed-off-by: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 5a4a8673102761fb87c94ee20633bf1f2a6911ca\nAuthor: Bruce Allan \u003cbruce.w.allan@intel.com\u003e\nDate: Thu Jul 25 02:53:50 2019 -0700\n\n ice: update ethtool stats on-demand\n \n Users expect ethtool statistics to be updated on-demand when invoking\n \u0027ethtool -S \u003ciface\u003e\u0027 instead of providing a snapshot of statistics taken\n once a second (the frequency of the watchdog task where stats are currently\n updated). Update stats every time \u0027ethtool -S \u003ciface\u003e\u0027 is run.\n \n Also, fix an indentation style issue and an unnecessary local variable\n initialization in ice_get_ethtool_stats() discovered while investigating\n the subject issue.\n \n Signed-off-by: Bruce Allan \u003cbruce.w.allan@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 3f416961b0a5000bf5556de1a53cc3cf87a6e744\nAuthor: Amruth G.P \u003camruth.gouda.parameshwarappa@intel.com\u003e\nDate: Mon Jul 29 02:04:54 2019 -0700\n\n ice: Add input handlers for virtual channel handlers\n \n Move the assignment to local variables after validation.\n \n Remove unnecessary checks in ice_vc_process_vf_msg() as the respective\n functions are now performing the checks.\n \n Signed-off-by: \"Amruth G.P\" \u003camruth.gouda.parameshwarappa@intel.com\u003e\n Signed-off-by: Nitesh B Venkatesh \u003cnitesh.b.venkatesh@intel.com\u003e\n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 3747f03115c1c7e577e86c7698e061a21f20576c\nAuthor: Chinh T Cao \u003cchinh.t.cao@intel.com\u003e\nDate: Mon Jul 29 02:04:53 2019 -0700\n\n ice: Don\u0027t clear auto_fec bit in ice_cfg_phy_fec()\n \n The driver should never clear the auto_fec_enable bit.\n \n Signed-off-by: Chinh T Cao \u003cchinh.t.cao@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 057911ba9b7939a3395e5c3c497fa2af830123e4\nAuthor: Chinh T Cao \u003cchinh.t.cao@intel.com\u003e\nDate: Mon Jul 29 02:04:52 2019 -0700\n\n ice: Fix flag used for module query\n \n When checking the PHY for status, by specification, the driver\n should be using \"topology\" mode when querying the module type.\n \n Signed-off-by: Chinh T Cao \u003cchinh.t.cao@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 90e477379e92c110a34ca0d108aa97d5b02076ee\nAuthor: Mitch Williams \u003cmitch.a.williams@intel.com\u003e\nDate: Mon Jul 29 02:04:51 2019 -0700\n\n ice: silence some bogus error messages\n \n In some circumstances, VF devices can be deactivated while a message is\n in-flight. In that case, a series of scary error message will be\n printed in the log. Since these are actually harmless, check for this\n case and suppress them. No harm, no foul.\n \n Signed-off-by: Mitch Williams \u003cmitch.a.williams@intel.com\u003e\n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 84a118ab58edddf3d29e9972136a4c4a923ea1fa\nAuthor: Dave Ertman \u003cdavid.m.ertman@intel.com\u003e\nDate: Mon Jul 29 02:04:50 2019 -0700\n\n ice: Rename ethtool private flag for lldp\n \n The current flag name of \"enable-fw-lldp\" is a bit cumbersome.\n \n Change priv-flag name to \"fw-lldp-agent\" with a value of on or\n off. This is more straight-forward in meaning.\n \n Signed-off-by: Dave Ertman \u003cdavid.m.ertman@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit f8af5bf5b45ecbbef2c246865b02c7350db81392\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Mon Jul 29 02:04:49 2019 -0700\n\n ice: reject VF attempts to enable head writeback\n \n The virtchnl interface provides a mechanism for a VF driver to request\n head writeback support. This feature is deprecated as of AVF 1.0, but\n older versions of a VF driver may still attempt to request the mode.\n \n Since the ice hardware does not support head writeback, we should not\n accept Tx queue configuration which attempts to enable it.\n \n Currently, the driver simply assumes that the headwb_enabled bit will\n never be set.\n \n If a VF driver does request head writeback, the configuration will\n return successfully, even though head writeback is not enabled. This\n leaves the VF driver in a non functional state since it is assuming to\n be operating in head writeback mode.\n \n Fix the PF driver to reject any attempt to setup headwb_enabled.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 5d86923060fce88a4dda8d8c9c5d5eb32f37c37b\nAuthor: Michel Thierry \u003cmichel.thierry@intel.com\u003e\nDate: Fri Aug 23 01:20:34 2019 -0700\n\n drm/i915/tgl: Enable VD HCP/MFX sub-pipe power gating\n \n HCP/MFX power gating is disabled by default, turn it on for the vd units\n available. User space will also issue a MI_FORCE_WAKEUP properly to\n wake up proper subwell.\n \n During driver load, init_clock_gating happens after device_info_init_mmio\n read the vdbox disable fuse register, so only present vd units will have\n these enabled.\n \n BSpec: 14214\n HSDES: 1209977827\n Signed-off-by: Michel Thierry \u003cmichel.thierry@intel.com\u003e\n Reviewed-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Tony Ye \u003ctony.ye@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823082055.5992-3-lucas.demarchi@intel.com\n\ncommit b3c0692f36a44bd02ee8ccc5bed972a36fbf2e99\nAuthor: Michel Thierry \u003cmichel.thierry@intel.com\u003e\nDate: Fri Aug 23 01:20:33 2019 -0700\n\n drm/i915/tgl: Move GTCR register to cope with GAM MMIO address remap\n \n GAM registers located in the 0x4xxx range have been relocated to 0xCxxx;\n this is to make space for global MOCS registers.\n \n v2: Rename register and bitfield to its new name (suggested by Mika)\n \n HSD: 399379\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Michel Thierry \u003cmichel.thierry@intel.com\u003e\n Reviewed-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190823082055.5992-2-lucas.demarchi@intel.com\n\ncommit dbb9ced797640cfd4996e879c53fb1134e693d53\nAuthor: Paul Burton \u003cpaul.burton@mips.com\u003e\nDate: Fri Aug 23 18:00:38 2019 +0100\n\n MIPS: Fix build with CONFIG_PROC_VMCORE\u003dy\n \n Commit a94e4f24ec83 (\"MIPS: init: Drop boot_mem_map\") introduced a\n reference to a non-existant \"end\" field in struct memblock_region.\n Replace it with a sum of the base \u0026 size fields to fix builds with\n CONFIG_PROC_VMCORE\u003dy.\n \n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit 42a179c80ddd19955ec90f01858b2993145dc874\nAuthor: Michal Swiatkowski \u003cmichal.swiatkowski@intel.com\u003e\nDate: Mon Jul 29 02:04:48 2019 -0700\n\n ice: Copy dcbx configuration only if mode is correct\n \n In rebuild DCB desired_dcbx_cfg was copy to local_dcbx_cfg, but\n if DCBX mode is IEEE desired_dcbx_cfg is not initialized by DCBX\n config from FW. Change logic to copy config value only if mode is\n set to CEE.\n \n If driver copy desired_dcbx_cfg to local_dcbx_cfg in IEEE mode there\n is problem with globr. System is frozen after two or more globr.\n \n Signed-off-by: Michal Swiatkowski \u003cmichal.swiatkowski@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 64bcaec64284e08430db5f00944443ec41a86df2\nAuthor: Dave Ertman \u003cdavid.m.ertman@intel.com\u003e\nDate: Mon Jul 29 02:04:46 2019 -0700\n\n ice: Treat DCBx state NOT_STARTED as valid\n \n When a port is not cabled, but DCBx is enabled in the\n firmware, the status of DCBx will be NOT_STARTED. This\n is a valid state for FW enabled and should not be\n treated as a is_fw_lldp true automatically.\n \n Add the code to treat NOT_STARTED as another valid state.\n \n Signed-off-by: Dave Ertman \u003cdavid.m.ertman@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 625cfb6f20360e82a9c65d8184df49f1d67bc18e\nAuthor: Paul Burton \u003cpaul.burton@mips.com\u003e\nDate: Fri Aug 23 17:48:58 2019 +0100\n\n MIPS: mm: Fix highmem compile\n \n Commit a5718fe8f70f (\"MIPS: mm: Drop boot_mem_map\") removed the\n definition of a page variable for some reason, but that variable is\n still used. Restore it to fix compilation with CONFIG_HIGHMEM enabled.\n \n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit da4a9e73d8a58eccef93682c9f2d910c346d2c1e\nAuthor: Brett Creeley \u003cbrett.creeley@intel.com\u003e\nDate: Mon Jul 29 02:04:45 2019 -0700\n\n ice: Don\u0027t call synchronize_irq() for VF\u0027s from the host\n \n Currently we will call synchronize_irq() from the host for VF\u0027s. This is\n not correct, so don\u0027t allow it.\n \n Signed-off-by: Brett Creeley \u003cbrett.creeley@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit b4d857ded1c50fb2bd1168d6f80ae81397ae468b\nAuthor: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\nDate: Mon Aug 12 10:05:39 2019 -0400\n\n drm/amd/display: 3.2.48\n \n Signed-off-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Reviewed-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f49cfa279b9a2214172d13e87fe6a11bd3941f67\nAuthor: hersen wu \u003chersenxs.wu@amd.com\u003e\nDate: Fri Aug 2 16:01:37 2019 -0400\n\n drm/amd/display: flicking observed while installing driver on Navi10 CF\n \n [WHY] value of dchub_ref_clock is decided by dchubbub global timer\n settings which is programmed by vbios command table disp_init.\n for multi-GPU case, vbios is posted only for primary GPU. without\n vbios posted for the secondary GPU, value of dchub_ref_clock is not\n set properly. this value will affect dcn bandwidth calcuation and\n cause underflow. user will see screen flicking during driver\n installation for dual GPU case.\n \n [HOW] dc init_hw always call vbios command table disp_init to\n make sure dchubbub global timer is configured and enable.\n \n Signed-off-by: hersen wu \u003chersenxs.wu@amd.com\u003e\n Reviewed-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9a7bfcb6acee07ae891ad28d40a1535aab3134a9\nAuthor: Joshua Aberback \u003cjoshua.aberback@amd.com\u003e\nDate: Thu Aug 8 13:22:36 2019 -0400\n\n drm/amd/display: Properly read LVTMA_PWRSEQ_CNTL\n \n [Why]\n The register LVTMA_PWRSEQ_CNTL is used to determine the power state of the\n embedded display. Currently we do not actually read this register\u0027s values,\n so during power down we think that this display is already off, so we skip\n calling into VBIOS to actually turn it off.\n \n [How]\n - add relevant fields to shift / mask initialization\n \n Signed-off-by: Joshua Aberback \u003cjoshua.aberback@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 617089d5837a3aa4167a9c4afb34929e0d52805d\nAuthor: Jun Lei \u003cJun.Lei@amd.com\u003e\nDate: Tue Jul 23 16:56:03 2019 -0400\n\n drm/amd/display: revert wait in pipelock\n \n [why]\n Previous workaround to prevent a vsync flip to be converted\n to immediate flip is no longer needed, and is risky because\n there are cases where it can result in infinite loop.\n \n [how]\n Remove wait loop (which is potentially infinite) before locking\n pipe\n \n Signed-off-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Reviewed-by: Eric Yang \u003ceric.yang2@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9a00d0ff92f4c718054b5c068e255ecc61ef48bb\nAuthor: Derek Lai \u003cDerek.Lai@amd.com\u003e\nDate: Fri Aug 23 11:44:53 2019 -0500\n\n drm/amd/display: Use res_cap to acquire i2c instead of pipe count\n \n [Why]\n We should be using the ddc_num from res_caps. As the\n pipe count !\u003d number of i2c resources.\n \n [How]\n Use ddc_num from res_cap instead of pipe count.\n \n Signed-off-by: Derek Lai \u003cDerek.Lai@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1b0c3247a092db672bf4599f234f0e90d6e30e8b\nAuthor: Dave Ertman \u003cdavid.m.ertman@intel.com\u003e\nDate: Mon Jul 29 02:04:44 2019 -0700\n\n ice: Account for all states of FW DCBx and LLDP\n \n Currently, only the DCBx status is taken into account to\n determine if FW LLDP is possible. But there are NVM version\n coming out with DCBx enabled, and FW LLDP disabled. This\n is causing errors where the driver sees that DCBx is not\n disabled, and then tries to register for LLDP MIB change\n events, and fails.\n \n Change the logic to detect both DCBx and LLDP states in the\n FW engine.\n \n Signed-off-by: Dave Ertman \u003cdavid.m.ertman@intel.com\u003e\n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 470e2ca569bf780f832e3020ceb5afdef16a9084\nAuthor: Bayan Zabihiyan \u003cbayan.zabihiyan@amd.com\u003e\nDate: Thu Aug 8 11:08:52 2019 -0400\n\n drm/amd/display: Expose OTG_V_TOTAL_MID for HW Diags\n \n [Why]\n Existing HW Features, HW Diags test requested that the\n registers be exposed.\n \n [How]\n Add V_TOTAL_MID to existing DC structures.\n Make sure values are passed down throughout DC\n Add Register definition.\n Program the additional registers\n Add additional Logic for V_TOTAL_CONTROL.\n \n Signed-off-by: Bayan Zabihiyan \u003cbayan.zabihiyan@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b1a47f5338a617663b6b24338a8b682fa51c04b2\nAuthor: Jun Lei \u003cJun.Lei@amd.com\u003e\nDate: Wed Aug 7 16:24:46 2019 -0400\n\n drm/amd/display: fix DML not calculating delivery time\n \n [why]\n Calculating DCFCLK DS time requires calculating\n delivery time for luma/chroma, but this value is\n not calculated in DMLv2, it was inadvertently\n removed when porting DMLv2\n \n [how]\n Add the calculation back\n \n Signed-off-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Reviewed-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 0ec7d06b225663d62e8867951114375a512e56c4\nAuthor: David Francis \u003cDavid.Francis@amd.com\u003e\nDate: Thu Jul 25 15:22:16 2019 -0400\n\n drm/amd/display: MST topology debugfs\n \n DRM provides drm_dp_mst_dump_topology, which prints\n useful information about MST devices\n \n Hook this up to a debugfs file named amdgpu_mst_topology\n \n Signed-off-by: David Francis \u003cDavid.Francis@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 78d9b95ecf8a9b4cb871297359b7b804ee0c6925\nAuthor: Charlene Liu \u003ccharlene.liu@amd.com\u003e\nDate: Wed Aug 7 17:25:49 2019 -0400\n\n drm/amd/display: set av_mute in hw_init for HDMI\n \n [Description]\n OS will reserve HW state in UEFI mode.\n Driver init_hw reset to RGB which caused HDMI green in YCbCr mode.\n read HW blank_color based on acc_mode.\n \n Signed-off-by: Charlene Liu \u003ccharlene.liu@amd.com\u003e\n Reviewed-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4551666f5b6c46defad645f7b7e9b4a486985df6\nAuthor: Wyatt Wood \u003cwyatt.wood@amd.com\u003e\nDate: Wed Aug 7 13:48:24 2019 -0400\n\n drm/amd/display: Add Logging for Gamma Related information\n \n [Why]\n A recent bug showed that logging would be useful in debugging\n various gamma issues.\n \n [How]\n Add logging in dc.\n Fix formatting for easier graphing.\n \n Signed-off-by: Wyatt Wood \u003cwyatt.wood@amd.com\u003e\n Reviewed-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c005762253523044c5afade490129fc0dd6f868b\nAuthor: Jaehyun Chung \u003cjaehyun.chung@amd.com\u003e\nDate: Wed Aug 7 11:20:16 2019 -0400\n\n drm/amd/display: Enable HW rotation\n \n [Why] HW rotation is not enabled. Calculations for cursor rotation\n are wrong for the values passed to set_cursor_position.\n \n [How] Swap Src rect and height and vertically mirror surface for\n the correct surface rotation direction. Cursor position is rotated\n according to angle. Offset calculations are tweaked for non-rotated\n cursor hotspot and width/height.\n \n Signed-off-by: Jaehyun Chung \u003cjaehyun.chung@amd.com\u003e\n Reviewed-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5e1613e21fc1ea5c0888f4c858aa220ff8464d53\nAuthor: Bayan Zabihiyan \u003cbayan.zabihiyan@amd.com\u003e\nDate: Fri Jul 26 11:10:11 2019 -0400\n\n drm/amd/display: add Cursor Degamma logic for DCN2\n \n [Why]\n We need to have the ability to to tell us set degamma on the cursor.\n \n [How]\n Pass a flag down to register programming that tells us if the\n current surface format needs cursor degamma.\n \n Signed-off-by: Bayan Zabihiyan \u003cbayan.zabihiyan@amd.com\u003e\n Reviewed-by: Krunoslav Kovac \u003cKrunoslav.Kovac@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c0358809d49129d9a031de2fec152e4a2b75138e\nAuthor: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\nDate: Tue Aug 6 12:17:57 2019 -0400\n\n drm/amd/display: fix odm validation\n \n Update bw validation to use prev and next odm pipe pointers\n for populating dml inputs.\n \n Signed-off-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 224980366d9d9c10797943cc16631c82a50855f8\nAuthor: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\nDate: Tue Aug 6 16:09:07 2019 -0400\n\n drm/amd/display: fix odm stream release\n \n Need to memset all odm pipes when calling dc_remove_stream_from_ctx\n \n Signed-off-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 285e30049708c40a8b0221438d0300cb19a68eff\nAuthor: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\nDate: Tue Aug 6 15:10:33 2019 -0400\n\n drm/amd/display: fix dcn20 odm dpp programming\n \n dcn20 requires special casing for odm.\n This change treats odm as alternative to mpc tree on dcn20.\n \n This is planned to be fixed in a future refactor\n \n Signed-off-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5165d204acb89d540731d43fd738c3273f44cd01\nAuthor: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\nDate: Tue Aug 6 13:23:17 2019 -0400\n\n drm/amd/display: Fix number of slices not being checked for dsc\n \n [why]\n num_slices_h was not being checked\n \n [How]\n Fix the typo and check num_slices_h\n \n Signed-off-by: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\n Reviewed-by: Harry Wentland \u003cHarry.Wentland@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 24d01c9b32547acf7261585aefb5c953930ed0fd\nAuthor: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\nDate: Fri Aug 2 16:32:13 2019 -0400\n\n drm/amd/display: fix odm pipe copy\n \n ODM next and prev pipe were missing from dc_copy_state\n \n Signed-off-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Reviewed-by: Nikola Cornij \u003cNikola.Cornij@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1e7f100ce8c0640634b794604880d9204480c9f1\nAuthor: Yogesh Mohan Marimuthu \u003cyogesh.mohanmarimuthu@amd.com\u003e\nDate: Fri Aug 2 10:52:49 2019 +0530\n\n drm/amd/display: fix trigger not generated for freesync\n \n [Why]\n In newer hardware MANUAL_FLOW_CONTROL is not a trigger bit. Due to this\n front porch is fixed and in these hardware freesync does not work.\n \n [How]\n Change the programming to generate a pulse so that the event will be\n triggered, front porch will be cut short and freesync will work.\n \n Signed-off-by: Yogesh Mohan Marimuthu \u003cyogesh.mohanmarimuthu@amd.com\u003e\n Reviewed-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 436d96357fa956c4723b3400d88ae47cb0e2f4cd\nAuthor: Zi Yu Liao \u003cziyu.liao@amd.com\u003e\nDate: Tue Aug 6 11:58:09 2019 -0400\n\n drm/amd/display: fix stuck test pattern on right half of display\n \n [why]\n With visual confirm enabled, displays where ODM combine is enabled\n has a test pattern stuck on the right half of the display even\n though the display is unblanked.\n \n [how]\n Add a condition to not show the colour ramp test pattern when the\n display is unblanked.\n \n Signed-off-by: Zi Yu Liao \u003cziyu.liao@amd.com\u003e\n Reviewed-by: Eric Yang \u003ceric.yang2@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b5a41620bb88efb9fb31a4fa5e652e3d5bead7d4\nAuthor: Charlene Liu \u003ccharlene.liu@amd.com\u003e\nDate: Fri Aug 2 14:49:58 2019 -0400\n\n drm/amd/display: support spdif\n \n [Description]\n port spdif fix to staging:\n spdif hardwired to afmt inst 1.\n spdif func pointer\n spdif resource allocation (reserve last audio endpoint for spdif only)\n \n Signed-off-by: Charlene Liu \u003ccharlene.liu@amd.com\u003e\n Reviewed-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a9f54ce3c6032361d2b69910114b6e8baf9bdf49\nAuthor: Ahmad Othman \u003cahmad.othman@amd.com\u003e\nDate: Thu Aug 1 15:05:27 2019 -0400\n\n drm/amd/display: Refactoring VTEM\n \n [Why]\n Video Timing Extended Metadata packet (VTEM) is not\n specific to freesync. So move it out of freesync module\n \n [How]\n - Moved VTEM from freesync module to info_packet module\n - Created new structure for VTEM parameters that can be used for VRR\n and FVA\n \n Signed-off-by: Ahmad Othman \u003cahmad.othman@amd.com\u003e\n Reviewed-by: Chris Park \u003cChris.Park@amd.com\u003e\n Acked-by: Ahmad Othman \u003cAhmad.Othman@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 37fe34d243265f792fe2d122bde6e645ad4c1490\nAuthor: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\nDate: Tue Aug 6 00:48:04 2019 -0400\n\n drm/amd/display: 3.2.47\n \n Signed-off-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Reviewed-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1c0e2f435777cd2a2cbdbd55a79ff1b4e664b673\nAuthor: Qingqing Zhuo \u003cqingqing.zhuo@amd.com\u003e\nDate: Fri Aug 2 14:10:33 2019 -0400\n\n drm/amd/display: remove unused function\n \n [Why]\n This function is not being used, it was left in\n when introducing DCN2\n \n [How]\n Remove the function\n \n Signed-off-by: Qingqing Zhuo \u003cqingqing.zhuo@amd.com\u003e\n Reviewed-by: Eric Bernstein \u003cEric.Bernstein@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 75a543347c6be74b2636a1fec7f4ffd1ee4525b2\nAuthor: Martin Leung \u003cmartin.leung@amd.com\u003e\nDate: Fri Aug 2 11:01:05 2019 -0400\n\n drm/amd/display: cleaned up coding error in init_hw\n \n [why]\n during a refactor a redundant code that has unknown behaviour was added.\n \n [how]\n removed old bad code\n \n Fixes: 8a31820b1218 (\"drm/amd/display: Make init_hw and init_pipes generic for seamless boot\")\n Signed-off-by: Martin Leung \u003cmartin.leung@amd.com\u003e\n Reviewed-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ee80de548b47940d4519dbf9f9388a179e99d91a\nAuthor: Jaehyun Chung \u003cjaehyun.chung@amd.com\u003e\nDate: Mon Jul 29 14:48:32 2019 -0400\n\n drm/amd/display: Add VM page fault handle implementation\n \n [How] Allocate memory for default page and program memory block addr\n into default page addr register.\n \n Signed-off-by: Jaehyun Chung \u003cjaehyun.chung@amd.com\u003e\n Reviewed-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit df3b7e32ed459a5348f7b408a9b8142b7358fde8\nAuthor: Qingqing Zhuo \u003cqingqing.zhuo@amd.com\u003e\nDate: Wed Jul 31 18:11:16 2019 -0400\n\n drm/amd/display: refactor Device ID for external chips\n \n IEEE OUI will now be used while referring to certain vendors.\n instead of normal index\n \n Signed-off-by: Qingqing Zhuo \u003cqingqing.zhuo@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7f6e7186f9f44918972874608022975912f03e00\nAuthor: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\nDate: Thu Aug 1 15:52:58 2019 -0400\n\n drm/amd/display: Zero-out dsc init regs\n \n [why]\n Before a statically allocated PPS data structure, that did\n get zeroed-out at startup, had been re-used for making packed PPS\n SDP. With S3 fix, using a non-initialized PPS data structure was\n introduced, while wrongly assuming it\u0027d get initialized before it\u0027s\n populated. As a consequence \u0027vbr_enable\u0027 and perhaps some other\n fields are left uninitialized when making packed PPS SDP. This can\n affect \u0027simple_422\u0027 as well because of the way PPS SDP packing is\n done (the fields are not masked first, only shifted). The behavior\n will be different, depending on the content of uninitialized data.\n \n [how]\n Zero-out PPS data structure at initialization time before it\u0027s\n populated\n \n Fixes: 1a9e3d4569fc (\"drm/amd/display: Set DSC before DIG front-end is connected to its back-end\")\n Signed-off-by: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\n Reviewed-by: Wenjing Liu \u003cWenjing.Liu@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 387ad34cb7ffa589464c3ff77e5adb49026c15b4\nAuthor: Joseph Gravenor \u003cjoseph.gravenor@amd.com\u003e\nDate: Tue Jul 30 16:37:35 2019 -0400\n\n drm/amd/display: Implement voltage limitation stub\n \n add new function to get the voltage at the end of\n dcn_validate_bandwidth, to check against the\n highest voltage we allow.\n \n Created a stub to allow for optimizations\n \n Signed-off-by: Joseph Gravenor \u003cjoseph.gravenor@amd.com\u003e\n Reviewed-by: Eric Yang \u003ceric.yang2@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Acked-by: Sun peng Li \u003cSunpeng.Li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ddef02de0d71d483ad4398393717cc0d53fc990a\nAuthor: Wyatt Wood \u003cwyatt.wood@amd.com\u003e\nDate: Wed Jul 31 15:52:46 2019 -0400\n\n drm/amd/display: add null checks before logging\n \n Adding NULL checks to various parameters in log_tf, to avoid\n nullptr errors\n \n Signed-off-by: Wyatt Wood \u003cwyatt.wood@amd.com\u003e\n Reviewed-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Acked-by: Nikola Cornij \u003cNikola.Cornij@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 43d4baab525dc85f93d1e470a1b6e8ea1b0a5fdd\nAuthor: Ilya Bakoulin \u003cIlya.Bakoulin@amd.com\u003e\nDate: Tue Jul 30 18:30:40 2019 -0400\n\n drm/amd/display: set Hratio and VRatio in dml\n \n Set the writeback Hratio and Vratio in dml.\n \n Signed-off-by: Ilya Bakoulin \u003cIlya.Bakoulin@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1f3324603e39d6a5a7f9f378c980dab390d9a1d5\nAuthor: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\nDate: Wed Jul 31 14:08:45 2019 -0400\n\n drm/amd/display: fix dp stream enable\n \n A previous odm change broke stream enable by always setting\n n_multiply as if odm was on.\n \n This fixes the check for odm by making sure opp count is \u003e1\n rather than not 0.\n \n Signed-off-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 361c61f47a1cd8df26a08ddbb0cbdad72bb1776d\nAuthor: Josip Pavic \u003cJosip.Pavic@amd.com\u003e\nDate: Tue Jul 30 16:56:14 2019 -0400\n\n drm/amd/display: load iram for abm 2.3\n \n [Why]\n ABM 2.3 firmware expects information in iRAM that differs from previous\n versions of ABM, so a mechanism is required to provide it with that\n information.\n \n [How]\n Extend the existing iRAM definition to include parameters added by\n ABM 2.3, and load it if DMCU is running ABM 2.3.\n \n Signed-off-by: Josip Pavic \u003cJosip.Pavic@amd.com\u003e\n Reviewed-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 89cb5614736b9b5d3b833ca2237d10da6b4b0395\nAuthor: Zi Yu Liao \u003cziyu.liao@amd.com\u003e\nDate: Tue Jul 30 15:36:53 2019 -0400\n\n drm/amd/display: fix MPO HUBP underflow with Scatter Gather\n \n [why]\n With Scatter Gather enabled, HUBP underflows during MPO enabled video\n playback. hubp_init has a register write that fixes this problem, but\n the register is cleared when HUBP gets power gated.\n \n [how]\n Make a call to hubp_init during enable_plane, so that the fix can\n be applied after HUBP powers back on again.\n \n Signed-off-by: Zi Yu Liao \u003cziyu.liao@amd.com\u003e\n Reviewed-by: Tony Cheng \u003cTony.Cheng@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 574304041bb9508a4e73533e8d1702b0e713d321\nAuthor: Su Sung Chung \u003cSu.Chung@amd.com\u003e\nDate: Thu Jul 25 14:43:55 2019 -0400\n\n drm/amd/display: fix audio endpoint not getting disabled issue\n \n [Why]\n Disable_audio_stream gets enum option as a paramenter which will decide\n if we free acquired resources or not. However checks for the option is\n guarded by the other condition which check if audio stream is getting\n diabled more than once. With both conditions combined, if we attempt to\n disable audio stream twice in a row, first with keep and second with\n free as an option, we will never free any resources, which will make\n system think there is audio endpoint connected even after we plug out\n the device\n \n [How]\n Get rid of option as parameter to disable_audio_stream and move the part\n of the code that free acquired resources to outside where to keep or to\n free resources is actually determined\n \n Signed-off-by: Su Sung Chung \u003cSu.Chung@amd.com\u003e\n Reviewed-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 67427d4f10e765a18ac33b4008377ef365589501\nAuthor: Julian Parkin \u003cjulian.parkin@amd.com\u003e\nDate: Fri Jul 26 17:13:46 2019 -0400\n\n drm/amd/display: Delete dead code in command_table_helper\n \n [Why]\n dig_encoder_sel_to_atom will always return zero on any ASIC version\n past DCE80 since programming of the FE selection is handled by\n driver, but the translation code was left in the function, making\n it look like a coding error.\n \n [How]\n Remove code that has no effect, and replace with a comment describing\n why it returns zero.\n \n Signed-off-by: Julian Parkin \u003cjulian.parkin@amd.com\u003e\n Reviewed-by: Wenjing Liu \u003cWenjing.Liu@amd.com\u003e\n Acked-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b1f6d01c4a3b7eeb2eb035e79d425cd6a696fa45\nAuthor: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\nDate: Tue Aug 6 17:17:28 2019 -0400\n\n drm/amd/display: re structure odm to allow 4 to 1 support\n \n Currently odm is handled using top_bottom pipe by special casing\n the differing opps to differentiate from mpc combine.\n \n Since top/bottom pipe list was made to track mpc muxing this creates\n difficulties in adding a 4 pipe odm case support.\n \n Rather than continue using mpc combine list, this change reworks odm\n to use it\u0027s own linked list to keep track of odm combine pipes. This\n also opens up options for using mpo with odm, if a practical use case\n is ever found.\n \n Signed-off-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit bad4c3e66543b94b4abb18b6e49fe147684b7064\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Fri Aug 23 09:42:33 2019 -0500\n\n drm/amdgpu: set adev-\u003enum_vmhubs for gmc6,7,8\n \n So that we properly handle them on older asics.\n \n Fixes: 3ff985485b29 (\"drm/amdgpu: Export function to flush TLB of specific vm hub\")\n Tested-by: Tom St Denis \u003ctom.stdenis@amd.com\u003e\n Reviewed-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d99f38aed1a03f78cb7ae04d86e7fc5ac70411e3\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Thu Aug 22 14:17:57 2019 -0500\n\n drm/amdgpu/display: add flag for multi-display mclk switching\n \n Add a dcfeaturemask flag for mclk switching. Disable by default;\n enable once the feature has seen more testing.\n \n Set amdgpu.dcfeaturemask\u003d2 on the kernel command line in grub\n to enable this.\n \n Acked-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8b2608f1cfece410801a7f5fba374abde4540c0c\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Fri Aug 23 11:32:37 2019 -0500\n\n drm/amd/display: update bw_calcs to take pipe sync into account (v3)\n \n Properly set all_displays_in_sync so that when the data is\n propagated to powerplay, it\u0027s set properly and we can enable\n mclk switching when all monitors are in sync.\n \n v2: fix logic, clean up\n v3: check for blending chains, simplify logic\n \n Acked-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 55b852069d02e406fa493cc9366c523eb9b5d0a9\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Thu Aug 8 00:48:58 2019 -0500\n\n drm/amdgpu/powerplay/vega10: enable mclk switching if monitors are synced\n \n If DC has synced the displays, we can enable mclk switching to\n save power.\n \n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit bb6897f1c57129c5ba696237826d0aace41f7a9b\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Thu Aug 8 00:47:49 2019 -0500\n\n drm/amdgpu/powerplay/smu7: enable mclk switching if monitors are synced\n \n If DC has synced the displays, we can enable mclk switching to\n save power.\n \n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 951e15c2b58436b30e435f3169c97e9f4e265b58\nAuthor: Kent Russell \u003ckent.russell@amd.com\u003e\nDate: Fri Aug 23 09:13:18 2019 -0400\n\n drm/powerplay: Fix Vega20 power reading again\n \n For the 40.46 SMU release, they changed CurrSocketPower to\n AverageSocketPower, but this was changed back in 40.47 so just check if\n it\u0027s 40.46 and make the appropriate change\n \n Tested with 40.45, 40.46 and 40.47 successfully\n \n Signed-off-by: Kent Russell \u003ckent.russell@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b313bbebd79f73c151413a86e205ded32ec71c85\nAuthor: Frank.Min \u003cFrank.Min@amd.com\u003e\nDate: Fri Aug 16 15:08:31 2019 +0800\n\n amd/amdkfd: add Arcturus vf DID support\n \n Add the virtual function PCI device id.\n \n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Frank.Min \u003cFrank.Min@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 706feb26f890e1b8297b5d14975160de361edf4f\nAuthor: Kevin Wang \u003ckevin1.wang@amd.com\u003e\nDate: Wed Aug 21 10:58:19 2019 +0800\n\n drm/amd/powerpaly: fix navi series custom peak level value error\n \n fix other navi asic set peak performance level error.\n because the navi10_ppt.c will handle navi12 14 asic,\n it will use navi10 peak value to set other asic, it is not correct.\n \n after patch:\n only navi10 use custom peak value, other asic will used default value.\n \n Signed-off-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 64cc5414fbf673cf676e257f9d6c999357947488\nAuthor: Guchun Chen \u003cguchun.chen@amd.com\u003e\nDate: Fri Aug 16 15:06:52 2019 +0800\n\n drm/amdgpu: correct ras error count type\n \n Use unsigned long type for the same ras count variable.\n This will avoid overflow on 64 bit system.\n \n Signed-off-by: Guchun Chen \u003cguchun.chen@amd.com\u003e\n Reviewed-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 0c3a6101ff2df98d5d960bda9b159d7c4a952f27\nAuthor: Dave Ertman \u003cdavid.m.ertman@intel.com\u003e\nDate: Mon Jul 29 02:04:43 2019 -0700\n\n ice: Allow egress control packets from PF_VSI\n \n For control packets (i.e. LLDP packets) to be able to egress\n from the main VSI, a bit has to be set in the TX_descriptor.\n This should only be done for the main VSI and only if the\n FW LLDP agent is disabled. A bit to allow this also has to\n be set in the VSI context.\n \n Add the logic to add the necessary bits in the VSI context\n for the PF_VSI and the TX_descriptors for control packets\n egressing the PF_VSI.\n \n Signed-off-by: Dave Ertman \u003cdavid.m.ertman@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 7b07a7a4e169180c7da5078bdf5254515e69c152\nAuthor: Mars Cheng \u003cmars.cheng@mediatek.com\u003e\nDate: Mon Aug 19 17:21:34 2019 +0800\n\n dt-bindings: irq: mtk, sysirq: add support for mt6779\n \n Add binding documentation of mediatek,sysirq for mt6779 SoC.\n \n Signed-off-by: Mars Cheng \u003cmars.cheng@mediatek.com\u003e\n Reviewed-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Signed-off-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n\ncommit 563d4f0fa998000e5657ee7145aa4648ac4bf16c\nAuthor: Mars Cheng \u003cmars.cheng@mediatek.com\u003e\nDate: Mon Aug 19 17:21:33 2019 +0800\n\n dt-bindings: mtk-uart: add mt6779 uart bindings\n \n Add documentation for mt6779 uart dt-bindings\n \n Signed-off-by: Mars Cheng \u003cmars.cheng@mediatek.com\u003e\n Signed-off-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n\ncommit d9bd211dfa9377754c4b81333c376036b3a7164b\nAuthor: Mars Cheng \u003cmars.cheng@mediatek.com\u003e\nDate: Mon Aug 19 17:21:32 2019 +0800\n\n dt-bindings: mediatek: add support for mt6779 reference board\n \n Update binding document for mt6779 reference board\n \n Signed-off-by: Mars Cheng \u003cmars.cheng@mediatek.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n\ncommit 7f591fa7a62d3a3f585fd4ba5c3e7b05f4b931be\nAuthor: Hsin-Yi Wang \u003chsinyi@chromium.org\u003e\nDate: Fri Aug 23 14:24:52 2019 +0800\n\n arm64: kexec_file: add rng-seed support\n \n Adding \"rng-seed\" to dtb. It\u0027s fine to add this property if original\n fdt doesn\u0027t contain it. Since original seed will be wiped after\n read, so use a default size 128 bytes here.\n \n Signed-off-by: Hsin-Yi Wang \u003chsinyi@chromium.org\u003e\n Reviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 428826f5358c922dc378830a1717b682c0823160\nAuthor: Hsin-Yi Wang \u003chsinyi@chromium.org\u003e\nDate: Fri Aug 23 14:24:51 2019 +0800\n\n fdt: add support for rng-seed\n \n Introducing a chosen node, rng-seed, which is an entropy that can be\n passed to kernel called very early to increase initial device\n randomness. Bootloader should provide this entropy and the value is\n read from /chosen/rng-seed in DT.\n \n Obtain of_fdt_crc32 for CRC check after early_init_dt_scan_nodes(),\n since early_init_dt_scan_chosen() would modify fdt to erase rng-seed.\n \n Add a new interface add_bootloader_randomness() for rng-seed use case.\n Depends on whether the seed is trustworthy, rng seed would be passed to\n add_hwgenerator_randomness(). Otherwise it would be passed to\n add_device_randomness(). Decision is controlled by kernel config\n RANDOM_TRUST_BOOTLOADER.\n \n Signed-off-by: Hsin-Yi Wang \u003chsinyi@chromium.org\u003e\n Reviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e # drivers/char/random.c\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit e112b032a72c78f15d0c803c5dc6be444c2e6c66\nAuthor: Hsin-Yi Wang \u003chsinyi@chromium.org\u003e\nDate: Fri Aug 23 14:24:50 2019 +0800\n\n arm64: map FDT as RW for early_init_dt_scan()\n \n Currently in arm64, FDT is mapped to RO before it\u0027s passed to\n early_init_dt_scan(). However, there might be some codes\n (eg. commit \"fdt: add support for rng-seed\") that need to modify FDT\n during init. Map FDT to RO after early fixups are done.\n \n Signed-off-by: Hsin-Yi Wang \u003chsinyi@chromium.org\u003e\n Reviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Reviewed-by: Mike Rapoport \u003crppt@linux.ibm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 3e2d94535adb2df15f3907e4b4c7cd8a5a4c2b5a\nAuthor: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\nDate: Thu Aug 22 10:36:30 2019 +0200\n\n clocksource/drivers/hyperv: Enable TSC page clocksource on 32bit\n \n There is no particular reason to not enable TSC page clocksource on\n 32-bit. mul_u64_u64_shr() is available and despite the increased\n computational complexity (compared to 64bit) TSC page is still a huge win\n compared to MSR-based clocksource.\n \n In-kernel reads:\n MSR based clocksource: 3361 cycles\n TSC page clocksource: 49 cycles\n \n Reads from userspace (utilizing vDSO in case of TSC page):\n MSR based clocksource: 5664 cycles\n TSC page clocksource: 131 cycles\n \n Enabling TSC page on 32bits allows to get rid of CONFIG_HYPERV_TSCPAGE as\n it is now not any different from CONFIG_HYPERV_TIMER.\n \n Signed-off-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Michael Kelley \u003cmikelley@microsoft.com\u003e\n Link: https://lkml.kernel.org/r/20190822083630.17059-1-vkuznets@redhat.com\n\ncommit bd00cd52d5be655a2f217e2ed74b91a71cb2b14f\nAuthor: Tianyu Lan \u003cTianyu.Lan@microsoft.com\u003e\nDate: Wed Aug 14 20:32:16 2019 +0800\n\n clocksource/drivers/hyperv: Add Hyper-V specific sched clock function\n \n Hyper-V guests use the default native_sched_clock() in\n pv_ops.time.sched_clock on x86. But native_sched_clock() directly uses the\n raw TSC value, which can be discontinuous in a Hyper-V VM.\n \n Add the generic hv_setup_sched_clock() to set the sched clock function\n appropriately. On x86, this sets pv_ops.time.sched_clock to read the\n Hyper-V reference TSC value that is scaled and adjusted to be continuous.\n \n Also move the Hyper-V reference TSC initialization much earlier in the boot\n process so no discontinuity is observed when pv_ops.time.sched_clock\n calculates its offset.\n \n [ tglx: Folded build fix ]\n \n Signed-off-by: Tianyu Lan \u003cTianyu.Lan@microsoft.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Michael Kelley \u003cmikelley@microsoft.com\u003e\n Link: https://lkml.kernel.org/r/20190814123216.32245-3-Tianyu.Lan@microsoft.com\n\ncommit adb87ff4f96c9700718e09c97a804124d5cd61ff\nAuthor: Tianyu Lan \u003cTianyu.Lan@microsoft.com\u003e\nDate: Wed Aug 14 20:32:15 2019 +0800\n\n clocksource/drivers/hyperv: Allocate Hyper-V TSC page statically\n \n Prepare to add Hyper-V sched clock callback and move Hyper-V Reference TSC\n initialization much earlier in the boot process. Earlier initialization is\n needed so that it happens while the timestamp value is still 0 and no\n discontinuity in the timestamp will occur when pv_ops.time.sched_clock\n calculates its offset.\n \n The earlier initialization requires that the Hyper-V TSC page be allocated\n statically instead of with vmalloc(), so fixup the references to the TSC\n page and the method of getting its physical address.\n \n Signed-off-by: Tianyu Lan \u003cTianyu.Lan@microsoft.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Link: https://lkml.kernel.org/r/20190814123216.32245-2-Tianyu.Lan@microsoft.com\n\ncommit fe76a166a1e3dd2cc906e5468fda9834029d6258\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Wed Aug 21 23:13:34 2019 +0800\n\n f2fs: introduce f2fs_match_name() for cleanup\n \n This patch introduces f2fs_match_name() for cleanup.\n \n BTW, it avoids to fallback to normal comparison once it doesn\u0027t\n match casefolded name.\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit bbf9f7d90f21e05e31b7cdd95b32f64dd2819dfe\nAuthor: Sahitya Tummala \u003cstummala@codeaurora.org\u003e\nDate: Wed Aug 7 19:10:32 2019 +0530\n\n f2fs: Fix indefinite loop in f2fs_gc()\n \n Policy - Foreground GC, LFS and greedy GC mode.\n \n Under this policy, f2fs_gc() loops forever to GC as it doesn\u0027t have\n enough free segements to proceed and thus it keeps calling gc_more\n for the same victim segment. This can happen if the selected victim\n segment could not be GC\u0027d due to failed blkaddr validity check i.e.\n is_alive() returns false for the blocks set in current validity map.\n \n Fix this by keeping track of such invalid segments and skip those\n segments for selection in get_victim_by_default() to avoid endless\n GC loop under such error scenarios. Currently, add this logic under\n CONFIG_F2FS_CHECK_FS to be able to root cause the issue in debug\n version.\n \n Signed-off-by: Sahitya Tummala \u003cstummala@codeaurora.org\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n [Jaegeuk Kim: fix wrong bitmap size]\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 2fde3dd14efe8ceb21a92c2749394a5df61cb7df\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Fri Jul 26 15:41:20 2019 +0800\n\n f2fs: allocate memory in batch in build_sit_info()\n \n build_sit_info() allocate all bitmaps for each segment one by one,\n it\u0027s quite low efficiency, this pach changes to allocate large\n continuous memory at a time, and divide it and assign for each bitmaps\n of segment. For large size image, it can expect improving its mount\n speed.\n \n Signed-off-by: Chen Gong \u003cgongchen4@huawei.com\u003e\n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 4507847c86bfc64e9bdce941a0f707560d3df98a\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Wed Jul 17 17:06:11 2019 +0800\n\n f2fs: support FS_IOC_{GET,SET}FSLABEL\n \n Support two generic fs ioctls FS_IOC_{GET,SET}FSLABEL, letting\n f2fs pass generic/492 testcase.\n \n Fixes were made by Eric where:\n - f2fs: fix buffer overruns in FS_IOC_{GET, SET}FSLABEL\n utf16s_to_utf8s() and utf8s_to_utf16s() take the number of characters,\n not the number of bytes.\n \n - f2fs: fix copying too many bytes in FS_IOC_SETFSLABEL\n Userspace provides a null-terminated string, so don\u0027t assume that the\n full FSLABEL_MAX bytes can always be copied.\n \n - f2fs: add missing authorization check in FS_IOC_SETFSLABEL\n FS_IOC_SETFSLABEL modifies the filesystem superblock, so it shouldn\u0027t be\n allowed to regular users. Require CAP_SYS_ADMIN, like xfs and btrfs do.\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 899fee36fac07e49bb969e3f214e572eecb14f00\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Fri Aug 16 11:03:34 2019 +0800\n\n f2fs: fix to avoid data corruption by forbidding SSR overwrite\n \n There is one case can cause data corruption.\n \n - write 4k to fileA\n - fsync fileA, 4k data is writebacked to lbaA\n - write 4k to fileA\n - kworker flushs 4k to lbaB; dnode contain lbaB didn\u0027t be persisted yet\n - write 4k to fileB\n - kworker flush 4k to lbaA due to SSR\n - SPOR -\u003e dnode with lbaA will be recovered, however lbaA contains fileB\u0027s\n data\n \n One solution is tracking all fsynced file\u0027s block history, and disallow\n SSR overwrite on newly invalidated block on that file.\n \n However, during recovery, no matter the dnode is flushed or fsynced, all\n previous dnodes until last fsynced one in node chain can be recovered,\n that means we need to record all block change in flushed dnode, which\n will cause heavy cost, so let\u0027s just use simple fix by forbidding SSR\n overwrite directly.\n \n Fixes: 5b6c6be2d878 (\"f2fs: use SSR for warm node as well\")\n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit aabc172b986fd797065a61625c22a27a61f3f43d\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 8 10:02:53 2019 +0800\n\n f2fs: Fix build error while CONFIG_NLS\u003dm\n \n If CONFIG_F2FS_FS\u003dy but CONFIG_NLS\u003dm, building fails:\n \n fs/f2fs/file.o: In function `f2fs_ioctl\u0027:\n file.c:(.text+0xb86f): undefined reference to `utf16s_to_utf8s\u0027\n file.c:(.text+0xe651): undefined reference to `utf8s_to_utf16s\u0027\n \n Select CONFIG_NLS to fix this.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Fixes: 61a3da4d5ef8 (\"f2fs: support FS_IOC_{GET,SET}FSLABEL\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit a37d0862d17411edb67677a580a6f505ec2225f6\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Fri Aug 2 18:15:48 2019 +0800\n\n Revert \"f2fs: avoid out-of-range memory access\"\n \n As Pavel Machek reported:\n \n \"We normally use -EUCLEAN to signal filesystem corruption. Plus, it is\n good idea to report it to the syslog and mark filesystem as \"needing\n fsck\" if filesystem can do that.\"\n \n Still we need improve the original patch with:\n - use unlikely keyword\n - add message print\n - return EUCLEAN\n \n However, after rethink this patch, I don\u0027t think we should add such\n condition check here as below reasons:\n - We have already checked the field in f2fs_sanity_check_ckpt(),\n - If there is fs corrupt or security vulnerability, there is nothing\n to guarantee the field is integrated after the check, unless we do\n the check before each of its use, however no filesystem does that.\n - We only have similar check for bitmap, which was added due to there\n is bitmap corruption happened on f2fs\u0027 runtime in product.\n - There are so many key fields in SB/CP/NAT did have such check\n after f2fs_sanity_check_{sb,cp,..}.\n \n So I propose to revert this unneeded check.\n \n This reverts commit 56f3ce675103e3fb9e631cfb4131fc768bc23e9a.\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 290c30d4454ccf85de372af499e32e86f5d94188\nAuthor: Lihong Kou \u003ckoulihong@huawei.com\u003e\nDate: Mon Aug 5 19:13:52 2019 +0800\n\n f2fs: cleanup the code in build_sit_entries.\n \n We do not need to set the SBI_NEED_FSCK flag in the error paths, if we\n return error here, we will not update the checkpoint flag, so the code\n is useless, just remove it.\n \n Signed-off-by: Lihong Kou \u003ckoulihong@huawei.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 27cae0bcc0510ba009f01806b4af13940ad96f7d\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Mon Aug 5 18:27:25 2019 +0800\n\n f2fs: fix wrong available node count calculation\n \n In mkfs, we have counted quota file\u0027s node number in cp.valid_node_count,\n so we have to avoid wrong substraction of quota node number in\n .available_nid/.avail_node_count calculation.\n \n f2fs_write_check_point_pack()\n {\n ..\n set_cp(valid_node_count, 1 + c.quota_inum + c.lpf_inum);\n \n Fixes: 292c196a3695 (\"f2fs: reserve nid resource for quota sysfile\")\n Fixes: 7b63f72f73af (\"f2fs: fix to do sanity check on valid node/block count\")\n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 0b86f78920919cf36239f4e187ebad7f5d78b5f8\nAuthor: Lihong Kou \u003ckoulihong@huawei.com\u003e\nDate: Mon Aug 5 15:27:24 2019 +0800\n\n f2fs: remove duplicate code in f2fs_file_write_iter\n \n We will do the same check in generic_write_checks.\n if (iocb-\u003eki_flags \u0026 IOCB_NOWAIT) \u0026\u0026 !(iocb-\u003eki_flags \u0026 IOCB_DIRECT)\n return -EINVAL;\n just remove the same check in f2fs_file_write_iter.\n \n Signed-off-by: Lihong Kou \u003ckoulihong@huawei.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit d3a1a0e1bf5d4d2107d6b485c5d33e65acaa7cf6\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Mon Jul 29 23:02:29 2019 +0800\n\n f2fs: fix to migrate blocks correctly during defragment\n \n During defragment, we missed to trigger fragmented blocks migration\n for below condition:\n \n In defragment region:\n - total number of valid blocks is smaller than 512;\n - the tail part of the region are all holes;\n \n In addtion, return zero to user via range-\u003elen if there is no\n fragmented blocks.\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 33ac18a15c880d253565c08e8cec3ee4c8a76657\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Thu Aug 15 19:45:35 2019 +0800\n\n f2fs: use wrapped f2fs_cp_error()\n \n Just cleanup, no logic change.\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit fd114ab22dd14ded1783b7f9b0aaddc14c098fe0\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Thu Aug 15 19:45:36 2019 +0800\n\n f2fs: fix to use more generic EOPNOTSUPP\n \n EOPNOTSUPP is widely used as error number indicating operation is\n not supported in syscall, and ENOTSUPP was defined and only used\n for NFSv3 protocol, so use EOPNOTSUPP instead.\n \n Fixes: 0a2aa8fbb969 (\"f2fs: refactor __exchange_data_block for speed up\")\n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 3ee0c5d3b4e8bb0990a800449509c087f270d4a6\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Thu Aug 15 19:45:34 2019 +0800\n\n f2fs: use wrapped IS_SWAPFILE()\n \n Just cleanup, no logic change.\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 2c2eb7a300cd7c6945dafb077801dca95d7a6c25\nAuthor: Daniel Rosenberg \u003cdrosen@google.com\u003e\nDate: Tue Jul 23 16:05:29 2019 -0700\n\n f2fs: Support case-insensitive file name lookups\n \n Modeled after commit b886ee3e778e (\"ext4: Support case-insensitive file\n name lookups\")\n \n \"\"\"\n This patch implements the actual support for case-insensitive file name\n lookups in f2fs, based on the feature bit and the encoding stored in the\n superblock.\n \n A filesystem that has the casefold feature set is able to configure\n directories with the +F (F2FS_CASEFOLD_FL) attribute, enabling lookups\n to succeed in that directory in a case-insensitive fashion, i.e: match\n a directory entry even if the name used by userspace is not a byte per\n byte match with the disk name, but is an equivalent case-insensitive\n version of the Unicode string. This operation is called a\n case-insensitive file name lookup.\n \n The feature is configured as an inode attribute applied to directories\n and inherited by its children. This attribute can only be enabled on\n empty directories for filesystems that support the encoding feature,\n thus preventing collision of file names that only differ by case.\n \n * dcache handling:\n \n For a +F directory, F2Fs only stores the first equivalent name dentry\n used in the dcache. This is done to prevent unintentional duplication of\n dentries in the dcache, while also allowing the VFS code to quickly find\n the right entry in the cache despite which equivalent string was used in\n a previous lookup, without having to resort to -\u003elookup().\n \n d_hash() of casefolded directories is implemented as the hash of the\n casefolded string, such that we always have a well-known bucket for all\n the equivalencies of the same string. d_compare() uses the\n utf8_strncasecmp() infrastructure, which handles the comparison of\n equivalent, same case, names as well.\n \n For now, negative lookups are not inserted in the dcache, since they\n would need to be invalidated anyway, because we can\u0027t trust missing file\n dentries. This is bad for performance but requires some leveraging of\n the vfs layer to fix. We can live without that for now, and so does\n everyone else.\n \n * on-disk data:\n \n Despite using a specific version of the name as the internal\n representation within the dcache, the name stored and fetched from the\n disk is a byte-per-byte match with what the user requested, making this\n implementation \u0027name-preserving\u0027. i.e. no actual information is lost\n when writing to storage.\n \n DX is supported by modifying the hashes used in +F directories to make\n them case/encoding-aware. The new disk hashes are calculated as the\n hash of the full casefolded string, instead of the string directly.\n This allows us to efficiently search for file names in the htree without\n requiring the user to provide an exact name.\n \n * Dealing with invalid sequences:\n \n By default, when a invalid UTF-8 sequence is identified, ext4 will treat\n it as an opaque byte sequence, ignoring the encoding and reverting to\n the old behavior for that unique file. This means that case-insensitive\n file name lookup will not work only for that file. An optional bit can\n be set in the superblock telling the filesystem code and userspace tools\n to enforce the encoding. When that optional bit is set, any attempt to\n create a file name using an invalid UTF-8 sequence will fail and return\n an error to userspace.\n \n * Normalization algorithm:\n \n The UTF-8 algorithms used to compare strings in f2fs is implemented\n in fs/unicode, and is based on a previous version developed by\n SGI. It implements the Canonical decomposition (NFD) algorithm\n described by the Unicode specification 12.1, or higher, combined with\n the elimination of ignorable code points (NFDi) and full\n case-folding (CF) as documented in fs/unicode/utf8_norm.c.\n \n NFD seems to be the best normalization method for F2FS because:\n \n - It has a lower cost than NFC/NFKC (which requires\n decomposing to NFD as an intermediary step)\n - It doesn\u0027t eliminate important semantic meaning like\n compatibility decompositions.\n \n Although:\n \n - This implementation is not completely linguistic accurate, because\n different languages have conflicting rules, which would require the\n specialization of the filesystem to a given locale, which brings all\n sorts of problems for removable media and for users who use more than\n one language.\n \"\"\"\n \n Signed-off-by: Daniel Rosenberg \u003cdrosen@google.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 5aba54302a46fdd589040b928d5d010e5ace1234\nAuthor: Daniel Rosenberg \u003cdrosen@google.com\u003e\nDate: Tue Jul 23 16:05:28 2019 -0700\n\n f2fs: include charset encoding information in the superblock\n \n Add charset encoding to f2fs to support casefolding. It is modeled after\n the same feature introduced in commit c83ad55eaa91 (\"ext4: include charset\n encoding information in the superblock\")\n \n Currently this is not compatible with encryption, similar to the current\n ext4 imlpementation. This will change in the future.\n \n \u003eFrom the ext4 patch:\n \"\"\"\n The s_encoding field stores a magic number indicating the encoding\n format and version used globally by file and directory names in the\n filesystem. The s_encoding_flags defines policies for using the charset\n encoding, like how to handle invalid sequences. The magic number is\n mapped to the exact charset table, but the mapping is specific to ext4.\n Since we don\u0027t have any commitment to support old encodings, the only\n encoding I am supporting right now is utf8-12.1.0.\n \n The current implementation prevents the user from enabling encoding and\n per-directory encryption on the same filesystem at the same time. The\n incompatibility between these features lies in how we do efficient\n directory searches when we cannot be sure the encryption of the user\n provided fname will match the actual hash stored in the disk without\n decrypting every directory entry, because of normalization cases. My\n quickest solution is to simply block the concurrent use of these\n features for now, and enable it later, once we have a better solution.\n \"\"\"\n \n Signed-off-by: Daniel Rosenberg \u003cdrosen@google.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 71e90b4654a9298f9e2375cc733d57b8bf92ce73\nAuthor: Daniel Rosenberg \u003cdrosen@google.com\u003e\nDate: Tue Jul 23 16:05:27 2019 -0700\n\n fs: Reserve flag for casefolding\n \n In preparation for including the casefold feature within f2fs, elevate\n the EXT4_CASEFOLD_FL flag to FS_CASEFOLD_FL.\n \n Signed-off-by: Daniel Rosenberg \u003cdrosen@google.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 0921835c9544c1dc3d2093df6932f1b207eec487\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Fri Jul 26 15:43:17 2019 +0800\n\n f2fs: fix to avoid call kvfree under spinlock\n \n vfree() don\u0027t wish to be called from interrupt context, move it\n out of spin_lock_irqsave() coverage.\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 280fd422958187ac5f069c08d84dd65f7f87c2e6\nAuthor: Jia-Ju Bai \u003cbaijiaju1990@gmail.com\u003e\nDate: Fri Jul 26 11:45:12 2019 +0800\n\n fs: f2fs: Remove unnecessary checks of SM_I(sbi) in update_general_status()\n \n In fill_super() and put_super(), f2fs_destroy_stats() is called\n in prior to f2fs_destroy_segment_manager(), so if current\n sbi can still be visited in global stat list, SM_I(sbi) should be\n released yet.\n For this reason, SM_I(sbi) does not need to be checked in\n update_general_status().\n Thank Chao Yu for advice.\n \n Signed-off-by: Jia-Ju Bai \u003cbaijiaju1990@gmail.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 038d06984f5c50a101c1cf47bc6419064042716d\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Thu Jul 25 22:39:11 2019 +0800\n\n f2fs: disallow direct IO in atomic write\n \n Atomic write needs page cache to cache data of transaction,\n direct IO should never be allowed in atomic write, detect\n and deny it when open atomic write file.\n \n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit fe973b065bce0e61414c33251afae501a757f8c6\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Thu Jul 25 17:33:37 2019 +0800\n\n f2fs: fix to handle quota_{on,off} correctly\n \n With quota_ino feature on, generic/232 reports an inconsistence issue\n on the image.\n \n The root cause is that the testcase tries to:\n - use quotactl to shutdown journalled quota based on sysfile;\n - and then use quotactl to enable/turn on quota based on specific file\n (aquota.user or aquota.group).\n \n Eventually, quota sysfile will be out-of-update due to following specific\n file creation.\n \n Change as below to fix this issue:\n - deny enabling quota based on specific file if quota sysfile exists.\n - set SBI_QUOTA_NEED_REPAIR once sysfile based quota shutdowns via\n ioctl.\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit a25c2cdcb61ab3d8d99623a9e72cf1747979a1f9\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Mon Jul 22 17:57:06 2019 +0800\n\n f2fs: fix to detect cp error in f2fs_setxattr()\n \n It needs to return -EIO if filesystem has been shutdown, fix the\n miss case in f2fs_setxattr().\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 955ebcd3a910b00de94be8797b20b8cfb2ee0fd8\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Mon Jul 22 17:57:05 2019 +0800\n\n f2fs: fix to spread f2fs_is_checkpoint_ready()\n \n We missed to call f2fs_is_checkpoint_ready() in several places, it may\n allow space allocation even when free space was exhausted during\n checkpoint is disabled, fix to add them.\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 7975f3498dc0403d8177c0775b9514158ec66681\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Mon Jul 22 18:03:50 2019 +0800\n\n f2fs: support fiemap() for directory inode\n \n Adjust f2fs_fiemap() to support fiemap() on directory inode.\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 04f9287ab395a5a279db44fb39de69b23640abb9\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Fri Jul 19 15:18:44 2019 +0800\n\n f2fs: fix to avoid discard command leak\n \n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n BUG discard_cmd (Tainted: G B OE ): Objects remaining in discard_cmd on __kmem_cache_shutdown()\n -----------------------------------------------------------------------------\n \n INFO: Slab 0xffffe1ac481d22c0 objects\u003d36 used\u003d2 fp\u003d0xffff936b4748bf50 flags\u003d0x2ffff0000000100\n Call Trace:\n dump_stack+0x63/0x87\n slab_err+0xa1/0xb0\n __kmem_cache_shutdown+0x183/0x390\n shutdown_cache+0x14/0x110\n kmem_cache_destroy+0x195/0x1c0\n f2fs_destroy_segment_manager_caches+0x21/0x40 [f2fs]\n exit_f2fs_fs+0x35/0x641 [f2fs]\n SyS_delete_module+0x155/0x230\n ? vtime_user_exit+0x29/0x70\n do_syscall_64+0x6e/0x160\n entry_SYSCALL64_slow_path+0x25/0x25\n \n INFO: Object 0xffff936b4748b000 @offset\u003d0\n INFO: Object 0xffff936b4748b070 @offset\u003d112\n kmem_cache_destroy discard_cmd: Slab cache still has objects\n Call Trace:\n dump_stack+0x63/0x87\n kmem_cache_destroy+0x1b4/0x1c0\n f2fs_destroy_segment_manager_caches+0x21/0x40 [f2fs]\n exit_f2fs_fs+0x35/0x641 [f2fs]\n SyS_delete_module+0x155/0x230\n do_syscall_64+0x6e/0x160\n entry_SYSCALL64_slow_path+0x25/0x25\n \n Recovery can cache discard commands, so in error path of fill_super(),\n we need give a chance to handle them, otherwise it will lead to leak\n of discard_cmd slab cache.\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 0f1898f93cdcb9275b7ab9c9931c5c21a8fd3d61\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Fri Jul 19 11:51:11 2019 +0800\n\n f2fs: fix to avoid tagging SBI_QUOTA_NEED_REPAIR incorrectly\n \n On a quota disabled image, with fault injection, SBI_QUOTA_NEED_REPAIR\n will be set incorrectly in error path of f2fs_evict_inode(), fix it.\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit a8933b6b68f775b5774e7b075447fae13f4d01fe\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Thu Jul 18 16:39:59 2019 +0800\n\n f2fs: fix to drop meta/node pages during umount\n \n As reported in bugzilla:\n \n https://bugzilla.kernel.org/show_bug.cgi?id\u003d204193\n \n A null pointer dereference bug is triggered in f2fs under kernel-5.1.3.\n \n kasan_report.cold+0x5/0x32\n f2fs_write_end_io+0x215/0x650\n bio_endio+0x26e/0x320\n blk_update_request+0x209/0x5d0\n blk_mq_end_request+0x2e/0x230\n lo_complete_rq+0x12c/0x190\n blk_done_softirq+0x14a/0x1a0\n __do_softirq+0x119/0x3e5\n irq_exit+0x94/0xe0\n call_function_single_interrupt+0xf/0x20\n \n During umount, we will access NULL sbi-\u003enode_inode pointer in\n f2fs_write_end_io():\n \n f2fs_bug_on(sbi, page-\u003emapping \u003d\u003d NODE_MAPPING(sbi) \u0026\u0026\n page-\u003eindex !\u003d nid_of_node(page));\n \n The reason is if disable_checkpoint mount option is on, meta dirty\n pages can remain during umount, and then be flushed by iput() of\n meta_inode, however node_inode has been iput()ed before\n meta_inode\u0027s iput().\n \n Since checkpoint is disabled, all meta/node datas are useless and\n should be dropped in next mount, so in umount, let\u0027s adjust\n drop_inode() to give a hint to iput_final() to drop all those dirty\n datas correctly.\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 1f78adfab379e53b0bd725b869061d9ba8055943\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Fri Jul 12 16:57:00 2019 +0800\n\n f2fs: disallow switching io_bits option during remount\n \n If IO alignment feature is turned on after remount, we didn\u0027t\n initialize mempool of it, it turns out we will encounter panic\n during IO submission due to access NULL mempool pointer.\n \n This feature should be set only at mount time, so simply deny\n configuring during remount.\n \n This fixes bug reported in bugzilla:\n \n https://bugzilla.kernel.org/show_bug.cgi?id\u003d204135\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit c72db71ed61ff51c2b8189ac9889dd18f22eb612\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Fri Jul 12 16:55:42 2019 +0800\n\n f2fs: fix panic of IO alignment feature\n \n Since 07173c3ec276 (\"block: enable multipage bvecs\"), one bio vector\n can store multi pages, so that we can not calculate max IO size of\n bio as PAGE_SIZE * bio-\u003ebi_max_vecs. However IO alignment feature of\n f2fs always has that assumption, so finally, it may cause panic during\n IO submission as below stack.\n \n kernel BUG at fs/f2fs/data.c:317!\n RIP: 0010:__submit_merged_bio+0x8b0/0x8c0\n Call Trace:\n f2fs_submit_page_write+0x3cd/0xdd0\n do_write_page+0x15d/0x360\n f2fs_outplace_write_data+0xd7/0x210\n f2fs_do_write_data_page+0x43b/0xf30\n __write_data_page+0xcf6/0x1140\n f2fs_write_cache_pages+0x3ba/0xb40\n f2fs_write_data_pages+0x3dd/0x8b0\n do_writepages+0xbb/0x1e0\n __writeback_single_inode+0xb6/0x800\n writeback_sb_inodes+0x441/0x910\n wb_writeback+0x261/0x650\n wb_workfn+0x1f9/0x7a0\n process_one_work+0x503/0x970\n worker_thread+0x7d/0x820\n kthread+0x1ad/0x210\n ret_from_fork+0x35/0x40\n \n This patch adds one extra condition to check left space in bio while\n trying merging page to bio, to avoid panic.\n \n This bug was reported in bugzilla:\n \n https://bugzilla.kernel.org/show_bug.cgi?id\u003d204043\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 8896cbdfed0ca34452252b72d6ee97bcfca9abd2\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Fri Jul 12 16:55:41 2019 +0800\n\n f2fs: introduce {page,io}_is_mergeable() for readability\n \n Wrap merge condition into function for readability, no logic change.\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit a94e4f24ec836c8984f839594bad7454184975b1\nAuthor: Jiaxun Yang \u003cjiaxun.yang@flygoat.com\u003e\nDate: Mon Aug 19 22:23:13 2019 +0800\n\n MIPS: init: Drop boot_mem_map\n \n boot_mem_map was introduced very early and cannot handle memory maps\n with nid. Nowadays, memblock can exactly replace boot_mem_map.\n \n Detect pfn info and setup resources with memblock maps.\n \n Signed-off-by: Jiaxun Yang \u003cjiaxun.yang@flygoat.com\u003e\n [paul.burton@mips.com: Fix size calculation in check_kernel_sections_mem]\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: yasha.che3@gmail.com\n Cc: aurelien@aurel32.net\n Cc: sfr@canb.auug.org.au\n Cc: fancer.lancer@gmail.com\n Cc: matt.redfearn@mips.com\n Cc: chenhc@lemote.com\n\ncommit a5718fe8f70f33b1b5b47a153057cfdd19684598\nAuthor: Jiaxun Yang \u003cjiaxun.yang@flygoat.com\u003e\nDate: Mon Aug 19 22:23:12 2019 +0800\n\n MIPS: mm: Drop boot_mem_map\n \n Initialize maar by resource map and replace page_is_ram\n by memblock_is_memory.\n \n Signed-off-by: Jiaxun Yang \u003cjiaxun.yang@flygoat.com\u003e\n [paul.burton@mips.com:\n - Fix bad MAAR address calculations.\n - Use ALIGN() \u0026 define maar_align to make it clearer what\u0027s going on\n with address manipulations.\n - Drop the new used field from struct maar_config.\n - Rework the RAM walk to avoid iterating over the cfg array needlessly\n to find the first unused entry, then count used entries at the end.\n Instead just keep the count as we go.]\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: yasha.che3@gmail.com\n Cc: aurelien@aurel32.net\n Cc: sfr@canb.auug.org.au\n Cc: fancer.lancer@gmail.com\n Cc: matt.redfearn@mips.com\n Cc: chenhc@lemote.com\n\ncommit 7f43020e3bdb63d65661ed377682702f8b34d3ea\nAuthor: Bruno Thomsen \u003cbruno.thomsen@gmail.com\u003e\nDate: Thu Aug 22 15:19:34 2019 +0200\n\n rtc: pcf2127: bugfix: read rtc disables watchdog\n \n The previous fix listed bulk read of registers as root cause of\n accendential disabling of watchdog, since the watchdog counter\n register (WD_VAL) was zeroed.\n \n Fixes: 3769a375ab83 rtc: pcf2127: bulk read only date and time registers.\n \n Tested with the same PCF2127 chip as Sean reveled root cause\n of WD_VAL register value zeroing was caused by reading CTRL2\n register which is one of the watchdog feature control registers.\n \n So the solution is to not read the first two control registers\n (CTRL1 and CTRL2) in pcf2127_rtc_read_time as they are not\n needed anyway. Size of local buf variable is kept to allow\n easy usage of register defines to improve readability of code.\n \n Debug trace line was updated after CTRL1 and CTRL2 are no longer\n read from the chip. Also replaced magic numbers in buf access\n with register defines.\n \n Signed-off-by: Bruno Thomsen \u003cbruno.thomsen@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190822131936.18772-3-bruno.thomsen@gmail.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit bbfe3a7a1d411888d86d8ab322d14dc93bae42db\nAuthor: Bruno Thomsen \u003cbruno.thomsen@gmail.com\u003e\nDate: Thu Aug 22 15:19:33 2019 +0200\n\n rtc: pcf2127: cleanup register and bit defines\n \n Cleanup of defines to follow kernel coding style and increase code\n readability by using same register and bit define style.\n \n Change PCF2127_REG_RAM_{addr_MSB,wrt_cmd,rd_cmd} to upper case as\n kernel coding guide section 12 \u0027Macros, Enums and RTL\u0027 states\n \"Names of macros defining constants and labels in enums are capitalized\".\n \n Improve readability of RAM register comment by making whole sentences.\n \n Remove parentheses from register defines as they are only used\n for expressions and not constants.\n \n As there are no clear style for name of registers and bits in the\n kernel drivers, I suggest the following for at least this driver,\n but hopefully also other RTC drivers.\n \n Register name should follow this convention:\n [chip]_REG_[reg name] 0xXX\n \n Bit name should follow this convention, so it clearly states which\n chip register it\u0027s part of:\n [chip]_BIT_[reg name]_[bit name] BIT(X)\n \n Additionally I suggest bit defines are always placed right below\n its corresponding register define and using an extra tab indentation\n for the BIT(X) part. This will visually make it easy to see that bit\n defines are part of the complete register definition.\n \n Rename PCF2127_OSF to PCF2127_BIT_SC_OSF and move it right below\n PCF2127_REG_SC. This will improve readability of bit checks as it\u0027s\n easy to verify that it uses the correct register.\n \n Move end of line comments above register defines as it\u0027s more like\n a heading for 1 register define and up to 8 bit defines or a\n collection of registers that are close related like timestamp\n split across 6 registers.\n \n Signed-off-by: Bruno Thomsen \u003cbruno.thomsen@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190822131936.18772-2-bruno.thomsen@gmail.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit e788771cacafa4ae5c4dbfc404cd3579e77a0816\nAuthor: Bruno Thomsen \u003cbruno.thomsen@gmail.com\u003e\nDate: Thu Aug 22 15:19:32 2019 +0200\n\n rtc: pcf2127: convert to devm_rtc_allocate_device\n \n This allows further improvement of the driver.\n \n Signed-off-by: Bruno Thomsen \u003cbruno.thomsen@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190822131936.18772-1-bruno.thomsen@gmail.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit b60ff2cfb598ddf03d8dcb43ed53c35aa563e04c\nAuthor: Ondrej Jirman \u003cmegous@megous.com\u003e\nDate: Tue Aug 20 17:19:33 2019 +0200\n\n rtc: sun6i: Add support for H6 RTC\n \n RTC on H6 is mostly the same as on H5 and H3. It has slight differences\n mostly in features that are not yet supported by this driver.\n \n Some differences are already stated in the comments in existing code.\n One other difference is that H6 has extra bit in LOSC_CTRL_REG, called\n EXT_LOSC_EN to enable/disable external low speed crystal oscillator.\n \n It also has bit EXT_LOSC_STA in LOSC_AUTO_SWT_STA_REG, to check whether\n external low speed oscillator is working correctly.\n \n This patch adds support for enabling LOSC when necessary:\n \n - during reparenting\n - when probing the clock\n \n H6 also has capacbility to automatically reparent RTC clock from\n external crystal oscillator, to internal RC oscillator, if external\n oscillator fails. This is enabled by default. Disable it during\n probe.\n \n Signed-off-by: Ondrej Jirman \u003cmegous@megous.com\u003e\n Reviewed-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\n Link: https://lore.kernel.org/r/20190820151934.3860-3-megous@megous.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 903e259f9caf036cd9046884e5de9aa64447755c\nAuthor: Ondrej Jirman \u003cmegous@megous.com\u003e\nDate: Tue Aug 20 17:19:32 2019 +0200\n\n dt-bindings: rtc: sun6i: Add compatible for H6 RTC\n \n RTC on H6 is similar to the one on H5 SoC, but incompatible in small\n details. See the driver for description of differences. For example\n H6 RTC needs to enable the external low speed oscillator. Add new\n compatible for this RTC.\n \n Signed-off-by: Ondrej Jirman \u003cmegous@megous.com\u003e\n Link: https://lore.kernel.org/r/20190820151934.3860-2-megous@megous.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 7b0b551dbc1e0acb2a3a6a7fbebc3ab9cf68a050\nAuthor: Biwen Li \u003cbiwen.li@nxp.com\u003e\nDate: Tue Aug 13 11:01:56 2019 +0800\n\n rtc: fsl-ftm-alarm: add FTM alarm driver\n \n For the platforms including LS1012A, LS1021A, LS1028A, LS1043A,\n LS1046A, LS1088A, LS208xA that has the FlexTimer\n module, implementing alarm functions within RTC subsystem\n to wakeup the system when system going to sleep (work with RCPM driver).\n \n Signed-off-by: Biwen Li \u003cbiwen.li@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190813030157.48590-1-biwen.li@nxp.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 1d74f0992991ca3b640d5290c2722e9acdb82c15\nAuthor: Biwen Li \u003cbiwen.li@nxp.com\u003e\nDate: Tue Aug 13 11:01:57 2019 +0800\n\n dt-bindings: rtc: add bindings for FlexTimer Module\n \n The patch adds bindings for FlexTimer Module\n \n Signed-off-by: Biwen Li \u003cbiwen.li@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190813030157.48590-2-biwen.li@nxp.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 88cd3e6cfac915f50f7aa7b699bdf053afec866e\nAuthor: Jaskaran Khurana \u003cjaskarankhurana@linux.microsoft.com\u003e\nDate: Wed Jul 17 17:46:15 2019 -0700\n\n dm verity: add root hash pkcs#7 signature verification\n \n The verification is to support cases where the root hash is not secured\n by Trusted Boot, UEFI Secureboot or similar technologies.\n \n One of the use cases for this is for dm-verity volumes mounted after\n boot, the root hash provided during the creation of the dm-verity volume\n has to be secure and thus in-kernel validation implemented here will be\n used before we trust the root hash and allow the block device to be\n created.\n \n The signature being provided for verification must verify the root hash\n and must be trusted by the builtin keyring for verification to succeed.\n \n The hash is added as a key of type \"user\" and the description is passed\n to the kernel so it can look it up and use it for verification.\n \n Adds CONFIG_DM_VERITY_VERIFY_ROOTHASH_SIG which can be turned on if root\n hash verification is needed.\n \n Kernel commandline dm_verity module parameter \u0027require_signatures\u0027 will\n indicate whether to force root hash signature verification (for all dm\n verity volumes).\n \n Signed-off-by: Jaskaran Khurana \u003cjaskarankhurana@linux.microsoft.com\u003e\n Tested-and-Reviewed-by: Milan Broz \u003cgmazyland@gmail.com\u003e\n Signed-off-by: Mike Snitzer \u003csnitzer@redhat.com\u003e\n\ncommit 39d13a1ac41dc8254f484fcd21af7ff05e316fad\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Wed Aug 7 08:50:22 2019 +0300\n\n dm crypt: reuse eboiv skcipher for IV generation\n \n Instead of instantiating a separate cipher to perform the encryption\n needed to produce the IV, reuse the skcipher used for the block data\n and invoke it one additional time for each block to encrypt a zero\n vector and use the output as the IV.\n \n For CBC mode, this is equivalent to using the bare block cipher, but\n without the risk of ending up with a non-time invariant implementation\n of AES when the skcipher itself is time variant (e.g., arm64 without\n Crypto Extensions has a NEON based time invariant implementation of\n cbc(aes) but no time invariant implementation of the core cipher other\n than aes-ti, which is not enabled by default).\n \n This approach is a compromise between dm-crypt API flexibility and\n reducing dependence on parts of the crypto API that should not usually\n be exposed to other subsystems, such as the bare cipher API.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Tested-by: Milan Broz \u003cgmazyland@gmail.com\u003e\n Signed-off-by: Mike Snitzer \u003csnitzer@redhat.com\u003e\n\ncommit 123d87d553e26f67e7be318c97c971b6b5fb1daa\nAuthor: Mikulas Patocka \u003cmpatocka@redhat.com\u003e\nDate: Fri Aug 23 09:55:26 2019 -0400\n\n dm: make dm_table_find_target return NULL\n \n Currently, if we pass too high sector number to dm_table_find_target, it\n returns zeroed dm_target structure and callers test if the structure is\n zeroed with the macro dm_target_is_valid.\n \n However, returning NULL is common practice to indicate errors.\n \n This patch refactors the dm code, so that dm_table_find_target returns\n NULL and its callers test the returned value for NULL. The macro\n dm_target_is_valid is deleted. In alloc_targets, we no longer allocate an\n extra zeroed target.\n \n Signed-off-by: Mikulas Patocka \u003cmpatocka@redhat.com\u003e\n Signed-off-by: Mike Snitzer \u003csnitzer@redhat.com\u003e\n\ncommit 81f5090db843be897414418c24fe472fa6e082b6\nAuthor: Yazen Ghannam \u003cyazen.ghannam@amd.com\u003e\nDate: Thu Aug 22 00:00:02 2019 +0000\n\n EDAC/amd64: Support asymmetric dual-rank DIMMs\n \n Future AMD systems will support asymmetric dual-rank DIMMs. These are\n DIMMs where the ranks are of different sizes.\n \n The even rank will use the Primary Even Chip Select registers and the\n odd rank will use the Secondary Odd Chip Select registers.\n \n Recognize if a Secondary Odd Chip Select is being used. Use the\n Secondary Odd Address Mask when calculating the chip select size.\n \n [ bp: move csrow_sec_enabled() to the header, fix CS_ODD define and\n tone-down the capitalized words spelling. ]\n \n Signed-off-by: Yazen Ghannam \u003cyazen.ghannam@amd.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Cc: \"linux-edac@vger.kernel.org\" \u003clinux-edac@vger.kernel.org\u003e\n Cc: James Morse \u003cjames.morse@arm.com\u003e\n Cc: Mauro Carvalho Chehab \u003cmchehab@kernel.org\u003e\n Cc: Tony Luck \u003ctony.luck@intel.com\u003e\n Link: https://lkml.kernel.org/r/20190821235938.118710-8-Yazen.Ghannam@amd.com\n\ncommit a121d6e0caf0982579a1be7f4ad4dcf45e36052b\nAuthor: Jiaxun Yang \u003cjiaxun.yang@flygoat.com\u003e\nDate: Mon Aug 19 22:23:11 2019 +0800\n\n MIPS: xlp: Drop boot_mem_map\n \n Simply replace with memblock functions.\n \n Signed-off-by: Jiaxun Yang \u003cjiaxun.yang@flygoat.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: yasha.che3@gmail.com\n Cc: aurelien@aurel32.net\n Cc: sfr@canb.auug.org.au\n Cc: fancer.lancer@gmail.com\n Cc: matt.redfearn@mips.com\n Cc: chenhc@lemote.com\n\ncommit aa1edac13e5f7662c38b098b9c91b0d93a0cdf50\nAuthor: Jiaxun Yang \u003cjiaxun.yang@flygoat.com\u003e\nDate: Mon Aug 19 22:23:10 2019 +0800\n\n MIPS: ip22: Drop addr_is_ram\n \n It can be replaced by page_is_ram.\n \n Signed-off-by: Jiaxun Yang \u003cjiaxun.yang@flygoat.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: yasha.che3@gmail.com\n Cc: aurelien@aurel32.net\n Cc: sfr@canb.auug.org.au\n Cc: fancer.lancer@gmail.com\n Cc: matt.redfearn@mips.com\n Cc: chenhc@lemote.com\n\ncommit b3c948e2c00f561b22a9865a9b9b952072973285\nAuthor: Jiaxun Yang \u003cjiaxun.yang@flygoat.com\u003e\nDate: Mon Aug 19 22:23:09 2019 +0800\n\n MIPS: msp: Record prom memory\n \n boot_mem_map is nolonger exist so we need to maintain a list\n of prom memory by ourselves\n \n Signed-off-by: Jiaxun Yang \u003cjiaxun.yang@flygoat.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: yasha.che3@gmail.com\n Cc: aurelien@aurel32.net\n Cc: sfr@canb.auug.org.au\n Cc: fancer.lancer@gmail.com\n Cc: matt.redfearn@mips.com\n Cc: chenhc@lemote.com\n\ncommit 79fd0fe44731188054b6f4c433b5f805217a396c\nAuthor: Jiaxun Yang \u003cjiaxun.yang@flygoat.com\u003e\nDate: Mon Aug 19 22:23:08 2019 +0800\n\n MIPS: malta: Drop prom_free_prom_memory\n \n Current prom_free_prom_memory is freeing maps marked\n as BOOT_MEM_ROM_DATA, however, nobody is exactly setting\n this type for malta.\n \n Signed-off-by: Jiaxun Yang \u003cjiaxun.yang@flygoat.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: yasha.che3@gmail.com\n Cc: aurelien@aurel32.net\n Cc: sfr@canb.auug.org.au\n Cc: fancer.lancer@gmail.com\n Cc: matt.redfearn@mips.com\n Cc: chenhc@lemote.com\n\ncommit 0df1007677d51e533f6738e1b94a92ec2fe371be\nAuthor: Jiaxun Yang \u003cjiaxun.yang@flygoat.com\u003e\nDate: Mon Aug 19 22:23:07 2019 +0800\n\n MIPS: fw: Record prom memory\n \n boot_mem_map is nolonger exist so we need to maintain a list\n of prom memory by ourselves.\n \n Signed-off-by: Jiaxun Yang \u003cjiaxun.yang@flygoat.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: yasha.che3@gmail.com\n Cc: aurelien@aurel32.net\n Cc: sfr@canb.auug.org.au\n Cc: fancer.lancer@gmail.com\n Cc: matt.redfearn@mips.com\n Cc: chenhc@lemote.com\n\ncommit 6cda3a5e002f90e6ccce9ed8fb076fe28ae4ef3b\nAuthor: Jiaxun Yang \u003cjiaxun.yang@flygoat.com\u003e\nDate: Mon Aug 19 22:23:06 2019 +0800\n\n MIPS: OCTEON: Drop boot_mem_map\n \n Replace walk through boot_mem_map with for_each_memblock.\n And remove the check of total boot_mem_map.\n \n Signed-off-by: Jiaxun Yang \u003cjiaxun.yang@flygoat.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: yasha.che3@gmail.com\n Cc: aurelien@aurel32.net\n Cc: sfr@canb.auug.org.au\n Cc: fancer.lancer@gmail.com\n Cc: matt.redfearn@mips.com\n Cc: chenhc@lemote.com\n\ncommit e3fe8e555dd05cf74168d18555c44320ed50a0e1\nAuthor: Yang, Philip \u003cPhilip.Yang@amd.com\u003e\nDate: Thu Aug 15 20:52:56 2019 +0000\n\n mm/hmm: fix hmm_range_fault()\u0027s handling of swapped out pages\n \n hmm_range_fault() may return NULL pages because some of the pfns are equal\n to HMM_PFN_NONE. This happens randomly under memory pressure. The reason\n is during the swapped out page pte path, hmm_vma_handle_pte() doesn\u0027t\n update the fault variable from cpu_flags, so it failed to call\n hmm_vam_do_fault() to swap the page in.\n \n The fix is to call hmm_pte_need_fault() to update fault variable.\n \n Fixes: 74eee180b935 (\"mm/hmm/mirror: device page fault handler\")\n Link: https://lore.kernel.org/r/20190815205227.7949-1-Philip.Yang@amd.com\n Signed-off-by: Philip Yang \u003cPhilip.Yang@amd.com\u003e\n Reviewed-by: \"Jérôme Glisse\" \u003cjglisse@redhat.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 12ce6b0d21c2038ddae9da418aab198e284ad922\nAuthor: Max Staudt \u003cmax@enpas.org\u003e\nDate: Fri Aug 23 12:49:11 2019 +0200\n\n ata/pata_buddha: Probe via modalias instead of initcall\n \n Up until now, the pata_buddha driver would only check for cards on\n initcall time. Now, the kernel will call its probe function as soon\n as a compatible card is detected.\n \n v7: Removed suppress_bind_attrs that slipped in\n \n v6: Only do the drvdata workaround for X-Surf (remove breaks otherwise)\n Style\n \n v5: Remove module_exit(): There\u0027s no good way to handle the X-Surf hack.\n Also include a workaround to save X-Surf\u0027s drvdata in case zorro8390\n is active.\n \n v4: Clean up pata_buddha_probe() by using ent-\u003edriver_data.\n Support X-Surf via late_initcall()\n \n v3: Clean up devm_*, implement device removal.\n \n v2: Rename \u0027zdev\u0027 to \u0027z\u0027 to make the patch easy to analyse with\n git diff --ignore-space-change\n \n Signed-off-by: Max Staudt \u003cmax@enpas.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit a3d7d67403fcff366dafe8a7c04b15deec3d9022\nAuthor: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\nDate: Thu Aug 22 21:45:19 2019 -0700\n\n null_blk: create a helper for req completion\n \n This patch creates a helper function for handling the request\n completion in the null_handle_cmd().\n \n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit fceb5d1b19cbe6263f09dbe8e8138edf02eb6e6b\nAuthor: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\nDate: Thu Aug 22 21:45:18 2019 -0700\n\n null_blk: create a helper for zoned devices\n \n This patch creates a helper function for handling zoned block device\n operations.\n \n This patch also restructured the code for null_blk_zoned.c and uses the\n pattern to return blk_status_t and catch the error in the function\n null_handle_cmd() into cmd-\u003eerror variable instead of setting it up in\n the deeper layer just like the way it is done for flush, badblocks and\n memory backed case in the null_handle_cmd(). We also move\n null_handle_zoned() to the null_blk_zoned.c to keep the zoned code\n separate.\n \n Signed-off-by: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 7ea88e229e9df18ecd624b0d39f3dba87432ba33\nAuthor: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\nDate: Thu Aug 22 21:45:17 2019 -0700\n\n null_blk: create a helper for mem-backed ops\n \n This patch creates a helper for handling requests when null_blk is\n memory backed in the null_handle_cmd(). Although the helper is very\n simple right now, it makes the code flow consistent with the rest of\n code in the null_handle_cmd() and provides a uniform code structure\n for future code.\n \n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 8f94d1c1dc1e1094ef83fcadf1b992cf8ff7869e\nAuthor: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\nDate: Thu Aug 22 21:45:16 2019 -0700\n\n null_blk: create a helper for badblocks\n \n This patch creates a helper for handling badblocks code in the\n null_handle_cmd().\n \n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit adb84284796be4989dc3b48c7eef2784af45431d\nAuthor: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\nDate: Thu Aug 22 21:45:15 2019 -0700\n\n null_blk: create a helper for throttling\n \n This patch creates a helper for handling throttling code in the\n null_handle_cmd().\n \n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit d4b186ed227b80334abf1fe2c918c0ddc4374f38\nAuthor: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\nDate: Thu Aug 22 21:45:14 2019 -0700\n\n null_blk: move duplicate code to callers\n \n This is a preparation patch which moves the duplicate code for sectors\n and nr_sectors calculations for bio vs request mode into their\n respective callers (null_queue_bio(), null_qeueue_req()). Now the core\n function only deals with the respective actions and commands instead of\n having to calculte the bio vs req operations and different sector\n related variables. We also move the flush command handling at the top\n which significantly simplifies the rest of the code.\n \n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 7ac2d56f783d6d546e89a4e44fb46532667aa7a5\nAuthor: Cao Van Dong \u003ccv-dong@jinso.co.jp\u003e\nDate: Thu Apr 11 16:54:22 2019 +0900\n\n dt-bindings: can: rcar_can: document r8a77470 support\n \n Document SoC specific bindings for R-Car RZ/G1C(r8a77470) SoC.\n \n Signed-off-by: Cao Van Dong \u003ccv-dong@jinso.co.jp\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 80bd043dc1ff179deb7007da92abb8d1ae01919c\nAuthor: Ulrich Hecht \u003culrich.hecht+renesas@gmail.com\u003e\nDate: Wed Dec 5 09:17:05 2018 +0800\n\n dt-bindings: can: rcar_canfd: document r8a77995 support\n \n Adds compatible strings for the R-Car CAN FD controller in the D3 SoC.\n \n Signed-off-by: Ulrich Hecht \u003culrich.hecht+renesas@gmail.com\u003e\n Acked-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 81633d7509b3bfeae81a0ee3116739812b685177\nAuthor: Ulrich Hecht \u003culrich.hecht+renesas@gmail.com\u003e\nDate: Wed Dec 5 09:17:04 2018 +0800\n\n dt-bindings: can: rcar_can: document r8a77995 support\n \n Adds compatible strings for the R-Car CAN controller in the D3 SoC.\n \n Signed-off-by: Ulrich Hecht \u003culrich.hecht+renesas@gmail.com\u003e\n Acked-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 3bbda1431b43f65736fa8aa7ccfa8a5d5c98b2b8\nAuthor: Marek Vasut \u003cmarek.vasut+renesas@gmail.com\u003e\nDate: Sun Nov 18 18:30:56 2018 +0100\n\n dt-bindings: can: rcar_can: document r8a77990 support\n \n Document the support for rcar_can on R8A77990 SoC devices.\n Add R8A77990 to the list of SoCs which require the \"assigned-clocks\"\n and \"assigned-clock-rates\" properties.\n \n Signed-off-by: Marek Vasut \u003cmarek.vasut+renesas@gmail.com\u003e\n Cc: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 556030f0604f3cf5f1ea91307c0695541e5c74ca\nAuthor: Bibby Hsieh \u003cbibby.hsieh@mediatek.com\u003e\nDate: Tue Aug 20 16:49:28 2019 +0800\n\n soc: mediatek: cmdq: change the type of input parameter\n \n According to the cmdq hardware design, the subsys is u8,\n the offset is u16 and the event id is u16.\n This patch changes the type of subsys, offset and event id\n to the correct type.\n \n Signed-off-by: Bibby Hsieh \u003cbibby.hsieh@mediatek.com\u003e\n Reviewed-by: CK Hu \u003cck.hu@mediatek.com\u003e\n Signed-off-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n\ncommit 1a92f989126ef0743650d43f75bfc431a5c3a310\nAuthor: Bibby Hsieh \u003cbibby.hsieh@mediatek.com\u003e\nDate: Tue Aug 20 16:49:27 2019 +0800\n\n soc: mediatek: cmdq: reorder the parameter\n \n The order of gce instructions is [subsys offset value]\n so reorder the parameter of cmdq_pkt_write_mask\n and cmdq_pkt_write function.\n \n Signed-off-by: Bibby Hsieh \u003cbibby.hsieh@mediatek.com\u003e\n Reviewed-by: CK Hu \u003cck.hu@mediatek.com\u003e\n Signed-off-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n\ncommit 15ede97054889c0bec09f1f9b71beffecf06fc67\nAuthor: Ondrej Jirman \u003cmegous@megous.com\u003e\nDate: Fri Aug 23 11:42:28 2019 +0200\n\n arm64: dts: allwinner: orange-pi-3: Enable WiFi\n \n Orange Pi 3 has AP6256 WiFi/BT module. WiFi part of the module is called\n bcm43356 and can be used with the brcmfmac driver. The module is powered by\n the two always on regulators (not AXP805).\n \n WiFi uses a PG port with 1.8V voltage level signals. SoC needs to be\n configured so that it sets up an 1.8V input bias on this port. This is done\n by the pio driver by reading the vcc-pg-supply voltage.\n \n You\u0027ll need a fw_bcm43456c5_ag.bin firmware file and nvram.txt\n configuration that can be found in the Xulongs\u0027s repository for H6:\n \n https://github.com/orangepi-xunlong/OrangePiH6_external/tree/master/ap6256\n \n Mainline brcmfmac driver expects the firmware and nvram at the following\n paths relative to the firmware directory:\n \n brcm/brcmfmac43456-sdio.bin\n brcm/brcmfmac43456-sdio.txt\n \n Signed-off-by: Ondrej Jirman \u003cmegous@megous.com\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit c55be305915974db160ce6472722ff74f45b8d4e\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Fri Aug 23 00:15:13 2019 +0300\n\n spi: spi-fsl-dspi: Use poll mode in case the platform IRQ is missing\n \n On platforms like LS1021A which use TCFQ mode, an interrupt needs to be\n processed after each byte is TXed/RXed. I tried to make the DSPI\n implementation on this SoC operate in other, more efficient modes (EOQ,\n DMA) but it looks like it simply isn\u0027t possible.\n \n Therefore allow the driver to operate in poll mode, to ease a bit of\n this absurd amount of IRQ load generated in TCFQ mode. Doing so reduces\n both the net time it takes to transmit a SPI message, as well as the\n inter-frame jitter that occurs while doing so.\n \n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190822211514.19288-5-olteanv@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 1eaeba70738e723be1e5787bdfd9a30f7471d730\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Fri Aug 23 00:15:12 2019 +0300\n\n spi: spi-fsl-dspi: Remove impossible to reach error check\n \n dspi-\u003edevtype_data is under the total control of the driver. Therefore,\n a bad value is a driver bug and checking it at runtime (and during an\n ISR, at that!) is pointless.\n \n The second \"else if\" check is only for clarity (instead of a broader\n \"else\") in case other transfer modes are added in the future. But the\n printing is dead code and can be removed.\n \n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190822211514.19288-4-olteanv@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 37b4100180641968056cb4e034cebc38338e8652\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Fri Aug 23 00:15:11 2019 +0300\n\n spi: spi-fsl-dspi: Exit the ISR with IRQ_NONE when it\u0027s not ours\n \n The DSPI interrupt can be shared between two controllers at least on the\n LX2160A. In that case, the driver for one controller might misbehave and\n consume the other\u0027s interrupt. Fix this by actually checking if any of\n the bits in the status register have been asserted.\n \n Fixes: 13aed2392741 (\"spi: spi-fsl-dspi: use IRQF_SHARED mode to request IRQ\")\n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190822211514.19288-3-olteanv@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 12fb61a973935c63f2580b3b053017cc14b51f42\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Fri Aug 23 00:15:10 2019 +0300\n\n spi: spi-fsl-dspi: Reduce indentation level in dspi_interrupt\n \n If the entire function depends on the SPI status register having the\n interrupt bits asserted, then just check it and exit early if those bits\n aren\u0027t set (such as in the case of the shared IRQ being triggered for\n the other peripheral). Cosmetic patch.\n \n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190822211514.19288-2-olteanv@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 85a5f06773c85f8ec5030ed65e5fa3190b3f16bd\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 22 22:43:46 2019 +0800\n\n ACPI: SBS: remove unused const variable \u0027SMBUS_PEC\u0027\n \n drivers/acpi/sbshc.h:18:17: warning:\n SMBUS_PEC defined but not used [-Wunused-const-variable\u003d]\n \n SMBUS_PEC is never used since introduction in\n commit 91087dfa51a2 (\"ACPI: SBS: Split host\n controller (ACPI0001) from SBS driver (ACPI0002)\"),\n so just remove it.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 7574729e91468d568cc198de438feb35ef04f41a\nAuthor: Yazen Ghannam \u003cyazen.ghannam@amd.com\u003e\nDate: Thu Aug 22 00:00:01 2019 +0000\n\n EDAC/amd64: Cache secondary Chip Select registers\n \n AMD Family 17h systems have a set of secondary Chip Select Base\n Addresses and Address Masks. These do not represent unique Chip\n Selects, rather they are used in conjunction with the primary\n Chip Select registers in certain cases.\n \n Cache these secondary Chip Select registers for future use.\n \n Signed-off-by: Yazen Ghannam \u003cyazen.ghannam@amd.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Cc: \"linux-edac@vger.kernel.org\" \u003clinux-edac@vger.kernel.org\u003e\n Cc: James Morse \u003cjames.morse@arm.com\u003e\n Cc: Mauro Carvalho Chehab \u003cmchehab@kernel.org\u003e\n Cc: Tony Luck \u003ctony.luck@intel.com\u003e\n Link: https://lkml.kernel.org/r/20190821235938.118710-7-Yazen.Ghannam@amd.com\n\ncommit 60e05d1cf0875f0cf73472f7dff71d9933c5b697\nAuthor: Cornelia Huck \u003ccohuck@redhat.com\u003e\nDate: Thu Aug 15 13:53:41 2019 +0200\n\n vfio-ccw: add some logging\n \n Usually, the common I/O layer logs various things into the s390\n cio debug feature, which has been very helpful in the past when\n looking at crash dumps. As vfio-ccw devices unbind from the\n standard I/O subchannel driver, we lose some information there.\n \n Let\u0027s introduce some vfio-ccw debug features and log some things\n there. (Unfortunately we cannot reuse the cio debug feature from\n a module.)\n \n Message-Id: \u003c20190816151505.9853-2-cohuck@redhat.com\u003e\n Reviewed-by: Eric Farman \u003cfarman@linux.ibm.com\u003e\n Signed-off-by: Cornelia Huck \u003ccohuck@redhat.com\u003e\n\ncommit 57b3006492a4c11b2d4a772b5b2905d544a32037\nAuthor: Jarkko Nikula \u003cjarkko.nikula@linux.intel.com\u003e\nDate: Thu Aug 22 11:32:00 2019 +0300\n\n ACPI / LPSS: Save/restore LPSS private registers also on Lynxpoint\n \n My assumption in commit b53548f9d9e4 (\"spi: pxa2xx: Remove LPSS private\n register restoring during resume\") that Intel Lynxpoint and compatible\n based chipsets may not need LPSS private registers saving and restoring\n over suspend/resume cycle turned out to be false on Intel Broadwell.\n \n Curtis Malainey sent a patch bringing above change back and reported the\n LPSS SPI Chip Select control was lost over suspend/resume cycle on\n Broadwell machine.\n \n Instead of reverting above commit lets add LPSS private register\n saving/restoring also for all LPSS SPI, I2C and UART controllers on\n Lynxpoint and compatible chipset to make sure context is not lost in\n case nothing else preserves it like firmware or if LPSS is always on.\n \n Fixes: b53548f9d9e4 (\"spi: pxa2xx: Remove LPSS private register restoring during resume\")\n Reported-by: Curtis Malainey \u003ccujomalainey@chromium.org\u003e\n Tested-by: Curtis Malainey \u003ccujomalainey@chromium.org\u003e\n Cc: 5.0+ \u003cstable@vger.kernel.org\u003e # 5.0+\n Signed-off-by: Jarkko Nikula \u003cjarkko.nikula@linux.intel.com\u003e\n Reviewed-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit d41f36a6464a85c06ad920703d878e4491d2c023\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Fri Aug 23 00:24:50 2019 +0300\n\n spi: spi-fsl-dspi: Exit the ISR with IRQ_NONE when it\u0027s not ours\n \n The DSPI interrupt can be shared between two controllers at least on the\n LX2160A. In that case, the driver for one controller might misbehave and\n consume the other\u0027s interrupt. Fix this by actually checking if any of\n the bits in the status register have been asserted.\n \n Fixes: 13aed2392741 (\"spi: spi-fsl-dspi: use IRQF_SHARED mode to request IRQ\")\n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190822212450.21420-2-olteanv@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n Cc: stable@vger.kernel.org\n\ncommit 834020366da9ab3fb87d1eb9a3160eb22dbed63a\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Thu Aug 8 16:51:00 2019 +0100\n\n ARM: 8898/1: mm: Don\u0027t treat faults reported from cache maintenance as writes\n \n Translation faults arising from cache maintenance instructions are\n rather unhelpfully reported with an FSR value where the WnR field is set\n to 1, indicating that the faulting access was a write. Since cache\n maintenance instructions on 32-bit ARM do not require any particular\n permissions, this can cause our private \u0027cacheflush\u0027 system call to fail\n spuriously if a translation fault is generated due to page aging when\n targetting a read-only VMA.\n \n In this situation, we will return -EFAULT to userspace, although this is\n unfortunately suppressed by the popular \u0027__builtin___clear_cache()\u0027\n intrinsic provided by GCC, which returns void.\n \n Although it\u0027s tempting to write this off as a userspace issue, we can\n actually do a little bit better on CPUs that support LPAE, even if the\n short-descriptor format is in use. On these CPUs, cache maintenance\n faults additionally set the CM field in the FSR, which we can use to\n suppress the write permission checks in the page fault handler and\n succeed in performing cache maintenance to read-only areas even in the\n presence of a translation fault.\n \n Reported-by: Orion Hodson \u003coth@google.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n Signed-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\n\ncommit 3e07590e7248db951fed6a2039403b5a39010be7\nAuthor: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\nDate: Wed Jul 31 12:38:48 2019 +0100\n\n ARM: 8896/1: VDSO: Don\u0027t leak kernel addresses\n \n Since commit ad67b74d2469d9b8 (\"printk: hash addresses printed with\n %p\"), an obfuscated kernel pointer is printed at every boot if\n debugging is enabled:\n \n vdso: 1 text pages at base (____ptrval____)\n \n Remove the print completely, as it\u0027s useless without the address.\n \n Based on commit 0f1bf7e39822476b (\"arm64/vdso: don\u0027t leak kernel\n addresses\").\n \n Fixes: ad67b74d2469d9b8 (\"printk: hash addresses printed with %p\")\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\n\ncommit 2a58e142a67411f51830028f7f07c69e619dec5e\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Thu Jul 18 17:35:23 2019 +0100\n\n ARM: 8895/1: visit mach-* and plat-* directories when cleaning\n \n When you run \"make clean\" for arm, it never visits mach-* or plat-*\n directories because machine-y and plat-y are just empty.\n \n When cleaning, all machine, plat directories are accumulated to\n machine-, plat-, respectively. So, let\u0027s pass them to core- to\n clean up those directories.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\n\ncommit 6583d8298e5981876603b35926b5832f7e99a122\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Tue Jul 16 12:33:14 2019 +0100\n\n ARM: 8894/1: boot: Replace open-coded nop with macro\n \n This open-coded nop as mov r0, r0 is a development history\n artifact.\n \n First commit b11fe38883d1\n (\"ARM: 6663/1: make Thumb2 kernel entry point more similar\n to the ARM one\") moved the code around so that the nops\n would come before the conditional thumb instructions, as it\n turned out that some boot loaders were patching the initial\n nop instructions in the kernel. At this point it is clear\n that all mov r0,r0 are open-coded nops.\n \n Then commit 81a0bc39ea19 (\"ARM: add UEFI stub support\")\n moved things around and defined __nop for EFI support and\n missed this open-coded nop.\n \n commit 06a4b6d009a1\n (\"ARM: 8677/1: boot/compressed: fix decompressor header\n layout for v7-M\") makes all invocations of __nop be wide,\n but that is fine, because this is what we want: the\n mov r0,r0 is inside ifndef CONFIG_THUMB2_KERNEL.\n \n Cc: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Acked-by: Nicolas Pitre \u003cnico@fluxnic.net\u003e\n Acked-by: Roy Franz \u003crfranz@marvell.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\n\ncommit 20699a42c020289e31afe33ff0909acdc5ca1350\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Tue Jul 16 12:32:37 2019 +0100\n\n ARM: 8893/1: boot: Explain the 8 nops\n \n This was unclear to me until Russell explained the obvious\n that 8 nops are added to offset an a.out image. Reading\n git history reveals that thumb kernels first removed the\n nops and then kept 7 of them (the last instruction being\n a switch to thumb mode) as it turns out that some boot\n loaders were using this as a \"patch area\". Also the magic\n numbers after the initial nops and the jump of course\n need to stay in the same offset for kernel file\n detection.\n \n Make the code easier to understand with a comment.\n \n Cc: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Acked-by: Nicolas Pitre \u003cnico@fluxnic.net\u003e\n Acked-by: Roy Franz \u003crfranz@marvell.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\n\ncommit 3c86889b0584be3e0243c22bf0901f12dfebac9a\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Thu Jul 11 04:07:56 2019 +0100\n\n ARM: 8876/1: fix O\u003d building with CONFIG_FPE_FASTFPE\n \n To use Fastfpe, a user is supposed to enable CONFIG_FPE_FASTFPE\n and put downstream source files into arch/arm/fastfpe/.\n \n It is not working for O\u003d build because $(wildcard arch/arm/fastfpe)\n checks if it exists in $(objtree), not in $(srctree).\n \n Add the $(srctree)/ prefix to fix it.\n \n While I was here, I slightly refactored the code.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\n\ncommit a05b9608456e0d4464c6f7ca8572324ace57a3f4\nAuthor: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nDate: Mon Jul 8 20:38:15 2019 +0100\n\n ARM: 8875/1: Kconfig: default to AEABI w/ Clang\n \n Clang produces references to __aeabi_uidivmod and __aeabi_idivmod for\n arm-linux-gnueabi and arm-linux-gnueabihf targets incorrectly when AEABI\n is not selected (such as when OABI_COMPAT is selected).\n \n While this means that OABI userspaces wont be able to upgraded to\n kernels built with Clang, it means that boards that don\u0027t enable AEABI\n like s3c2410_defconfig will stop failing to link in KernelCI when built\n with Clang.\n \n Link: https://github.com/ClangBuiltLinux/linux/issues/482\n Link: https://groups.google.com/forum/#!msg/clang-built-linux/yydsAAux5hk/GxjqJSW-AQAJ\n \n Suggested-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Signed-off-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Reviewed-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\n\ncommit 6f8f3570f27993226862681e3b0d88aaa8d7e5bf\nAuthor: Phong Tran \u003ctranmanphong@gmail.com\u003e\nDate: Thu Jun 27 16:37:24 2019 +0100\n\n ARM: 8873/1: perf: cleanup cppcheck shifting warning\n \n There is error from cppcheck tool.\n \"Shifting signed 32-bit value by 31 bits is undefined behaviour errors\"\n \n This error is false positive.\n change to use BIT() macro for improvement.\n \n Signed-off-by: Phong Tran \u003ctranmanphong@gmail.com\u003e\n Acked-by: Will Deacon \u003cwill.deacon@arm.com\u003e\n Signed-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\n\ncommit fb2eca315d92ab269e76be0595538a9ea0c93bc6\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Wed Jun 26 17:39:18 2019 +0100\n\n ARM: 8872/1: Use common outgoing-CPU-notification code\n \n This commit removes the open-coded CPU-offline notification with new\n common code. In particular, this change avoids calling scheduler code\n using RCU from an offline CPU that RCU is ignoring. This is a minimal\n change. A more intrusive change might invoke the cpu_check_up_prepare()\n and cpu_set_state_online() functions at CPU-online time, which would\n allow onlining throw an error if the CPU did not go offline properly.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.vnet.ibm.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Tested-by: Dietmar Eggemann \u003cdietmar.eggemann@arm.com\u003e\n Signed-off-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\n\ncommit 4af014984273a2577bf36c6ee271ed02bc831a04\nAuthor: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\nDate: Wed Jun 26 05:18:25 2019 +0100\n\n ARM: 8871/1: iop13xx: Simplify iop13xx_atu{e,x}_pci_status checks\n \n clang warns:\n \n arch/arm/mach-iop13xx/pci.c:292:7: warning: logical not is only applied\n to the left hand side of this comparison [-Wlogical-not-parentheses]\n if (!iop13xx_atux_pci_status(1) \u003d\u003d 0)\n ^ ~~\n arch/arm/mach-iop13xx/pci.c:439:7: warning: logical not is only applied\n to the left hand side of this comparison [-Wlogical-not-parentheses]\n if (!iop13xx_atue_pci_status(1) \u003d\u003d 0)\n ^ ~~\n \n !func() \u003d\u003d 0 is equivalent to func(), which clears up this warning and\n makes the code more readable.\n \n Link: https://github.com/ClangBuiltLinux/linux/issues/543\n \n Reported-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Reviewed-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Signed-off-by: Russell King \u003crmk+kernel@armlinux.org.uk\u003e\n\ncommit 542694df7511977c3baa6ba855126a0cce883977\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Aug 23 09:58:32 2019 +0900\n\n ASoC: soc-core: rename soc_post_component_init() to soc_rtd_init()\n \n It is easy to read code if it is cleanly using paired function/naming,\n like start \u003c-\u003e stop, register \u003c-\u003e unregister, etc, etc.\n But, current ALSA SoC code is very random, unbalance, not paired, etc.\n It is easy to create bug at the such code, and it will be difficult to\n debug.\n \n From function name point of view, \"soc_post_component_init()\" sounds\n like \"component initialize function\".\n But in reality it is rtd setup function.\n \n This patch renames soc_post_component_init() to soc_rtd_init()\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87v9uo7lc3.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 577bbf23b758848f0c4a50d346460b690c753024\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Thu Aug 22 05:21:14 2019 -0300\n\n media: sunxi: Add A10 CSI driver\n \n The older CSI drivers have camera capture interface different from the one\n in the newer ones.\n \n This IP is pretty simple. Some variants (one controller out of two\n instances on some SoCs) have an ISP embedded, but there\u0027s no code that make\n use of it, so we ignored that part for now.\n \n [Sakari Ailus: Wrapped a few long lines, set mbus code using a macro.]\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 3f0e6df62213ede83e3b223d36594cbcc953cb08\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Thu Aug 22 05:21:13 2019 -0300\n\n media: sunxi: Refactor the Makefile and Kconfig\n \n The Makefile and Kconfig for the sun6i CSI driver are included in the main\n Makefile / KConfig file. Since we\u0027re going to add a new CSI driver for an\n older chip, and the Cedrus driver eventually, it makes more sense to put\n those in our directory.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit c5e8f4ccd7750487f4ce05d0e801637124f14609\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Thu Aug 22 05:21:12 2019 -0300\n\n media: dt-bindings: media: Add Allwinner A10 CSI binding\n \n The Allwinner A10 CMOS Sensor Interface is a camera capture interface also\n used in later (A10s, A13, A20, R8 and GR8) SoCs.\n \n On some SoCs, like the A10, there\u0027s multiple instances of that controller,\n with one instance supporting more channels and having an ISP.\n \n [Sakari Ailus: Add type: object to the endpoint node.]\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 623fd246bb40234fe68dd4e7c1f1f081f9c45a3d\nAuthor: Alexandre Kroupski \u003calexandre.kroupski@ingenico.com\u003e\nDate: Tue Aug 20 08:37:45 2019 -0300\n\n media: atmel: atmel-isi: fix timeout value for stop streaming\n \n In case of sensor malfunction, stop streaming timeout takes much longer\n than expected. This is due to conversion of time to jiffies: milliseconds\n multiplied with HZ (ticks/second) gives out a value of jiffies with 10^3\n greater. We need to also divide by 10^3 to obtain the right jiffies value.\n In other words FRAME_INTERVAL_MILLI_SEC must be in seconds in order to\n multiply by HZ and get the right jiffies value to add to the current\n jiffies for the timeout expire time.\n \n Fixes: 195ebc43bf76 (\"[media] V4L: at91: add Atmel Image Sensor Interface (ISI) support\")\n Signed-off-by: Alexandre Kroupski \u003calexandre.kroupski@ingenico.com\u003e\n Reviewed-by: Eugen Hristev \u003ceugen.hristev@microchip.com\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit f4e9bcc05f4e8d543afbfc0ca1fd4435a2204776\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Fri Aug 9 16:11:16 2019 +0200\n\n gpio: mt7621: Pass irqchip when adding gpiochip\n \n We need to convert all old gpio irqchips to pass the irqchip\n setup along when adding the gpio_chip. For more info see\n drivers/gpio/TODO.\n \n For chained irqchips this is a pretty straight-forward\n conversion.\n \n This driver requests the IRQ directly in the driver so it\n differs a bit from the others.\n \n Cc: Greg Ungerer \u003cgerg@kernel.org\u003e\n Cc: Nicholas Mc Guire \u003chofrat@osadl.org\u003e\n Cc: Sergio Paracuellos \u003csergio.paracuellos@gmail.com\u003e\n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Tested-by: René van Dorst \u003copensource@vdorst.com\u003e\n Link: https://lore.kernel.org/r/20190809141116.16403-1-linus.walleij@linaro.org\n\ncommit b1d64c7139c1ec673bac8be88b27caf02cf73118\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Thu Aug 22 22:45:38 2019 +0200\n\n gpio: ftgpio: Fix an error handling path in \u0027ftgpio_gpio_probe()\u0027\n \n If \u0027devm_kcalloc()\u0027 fails, we should go through the error handling path,\n should some clean-up be needed.\n \n Fixes: 42d9fc7176eb (\"gpio: ftgpio: Pass irqchip when adding gpiochip\")\n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Link: https://lore.kernel.org/r/20190822204538.4791-1-christophe.jaillet@wanadoo.fr\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 9e1975f0bca9ef9fc5d76f056cab7240e25bca76\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Wed Aug 21 16:38:35 2019 +0200\n\n ARM: dts: sunxi: Add missing watchdog clocks\n \n The watchdog has a clock on all our SoCs, but it wasn\u0027t always listed.\n Add it to the devicetree where it\u0027s missing.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 89d1e51462d4749874ba3505c7de46734fda82f1\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Wed Aug 21 16:38:34 2019 +0200\n\n ARM: dts: sunxi: Add missing watchdog interrupts\n \n The watchdog has an interrupt on all our SoCs, but it wasn\u0027t always listed.\n Add it to the devicetree where it\u0027s missing.\n \n Acked-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit c7663fa2a6631e5dddc1b7b2ba2e905ddffaff90\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 22 11:18:17 2019 +0800\n\n gpio: Move gpiochip_lock/unlock_as_irq to gpio/driver.h\n \n If CONFIG_GPIOLIB is not, gpiochip_lock/unlock_as_irq will\n conflict as this:\n \n In file included from sound/soc/codecs/wm5100.c:18:0:\n ./include/linux/gpio.h:224:19: error: static declaration of gpiochip_lock_as_irq follows non-static declaration\n static inline int gpiochip_lock_as_irq(struct gpio_chip *chip,\n ^~~~~~~~~~~~~~~~~~~~\n In file included from sound/soc/codecs/wm5100.c:17:0:\n ./include/linux/gpio/driver.h:494:5: note: previous declaration of gpiochip_lock_as_irq was here\n int gpiochip_lock_as_irq(struct gpio_chip *chip, unsigned int offset);\n ^~~~~~~~~~~~~~~~~~~~\n In file included from sound/soc/codecs/wm5100.c:18:0:\n ./include/linux/gpio.h:231:20: error: static declaration of gpiochip_unlock_as_irq follows non-static declaration\n static inline void gpiochip_unlock_as_irq(struct gpio_chip *chip,\n ^~~~~~~~~~~~~~~~~~~~~~\n In file included from sound/soc/codecs/wm5100.c:17:0:\n ./include/linux/gpio/driver.h:495:6: note: previous declaration of gpiochip_unlock_as_irq was here\n void gpiochip_unlock_as_irq(struct gpio_chip *chip, unsigned int offset);\n ^~~~~~~~~~~~~~~~~~~~~~\n \n Move them to gpio/driver.h and use CONFIG_GPIOLIB guard this.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Fixes: d74be6dfea1b (\"gpio: remove gpiod_lock/unlock_as_irq()\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190822031817.32888-1-yuehaibing@huawei.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 2ff110bbe98e2beeb6488021fb650b1afc736263\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 15 11:37:18 2019 +0530\n\n pinctrl: meson: meson: Add of_node_put() before return\n \n Each iteration of for_each_child_of_node puts the previous node, but in\n the case of a return from the middle of the loop, there is no put, thus\n causing a memory leak. Hence add an of_node_put before the return.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190815060718.3286-1-nishkadg.linux@gmail.com\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 7db47faae79bbffe2da15af4d5e061a49a56169b\nAuthor: Hongwei Zhang \u003chongweiz@ami.com\u003e\nDate: Tue Aug 20 17:05:28 2019 -0400\n\n gpio: aspeed: Add SGPIO driver\n \n Add SGPIO driver support for Aspeed AST2500 SoC.\n \n Signed-off-by: Hongwei Zhang \u003chongweiz@ami.com\u003e\n Reviewed-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Link: https://lore.kernel.org/r/1566335128-31498-2-git-send-email-hongweiz@ami.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 5722f3386eb9847131573513190796533e8f4999\nAuthor: Filipe Laíns \u003clains@archlinux.org\u003e\nDate: Tue Jul 30 13:24:57 2019 +0100\n\n hid-logitech-dj: add the new Lightspeed receiver\n \n This patchs adds the new Lightspeed receiver. Currently it seems to only\n be used in the G305.\n \n Signed-off-by: Filipe Laíns \u003clains@archlinux.org\u003e\n [bentiss: rebased on top of master]\n Signed-off-by: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\n\ncommit 992ff2cc9ec6e30ccc64c341b052e8ccb370db90\nAuthor: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\nDate: Mon Aug 12 18:08:04 2019 +0200\n\n HID: logitech-dj: add support of the G700(s) receiver\n \n Both the G700 and the G700s are sharing the same receiver.\n Include support for this receiver in hid-logitech-dj so that userspace\n can differentiate both.\n \n Signed-off-by: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\n Reviewed-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n\ncommit 472a61e777fe78cdcb6cb6f25efee0ae9f629aca\nAuthor: Stefan Wahren \u003cstefan.wahren@i2se.com\u003e\nDate: Wed Aug 14 14:00:35 2019 +0300\n\n pinctrl/gpio: Take MUX usage into account\n \n The user space like gpioinfo only see the GPIO usage but not the\n MUX usage (e.g. I2C or SPI usage) of a pin. As a user we want\n to know which pin is free/safe to use. So take the MUX usage of\n strict pinmux controllers into account to get a more realistic\n view for ioctl GPIO_GET_LINEINFO_IOCTL.\n \n Signed-off-by: Stefan Wahren \u003cstefan.wahren@i2se.com\u003e\n Tested-by: Ramon Fried \u003crfried.dev@gmail.com\u003e\n Signed-off-by: Ramon Fried \u003crfried.dev@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190814110035.13451-1-ramon.fried@linux.intel.com\n Acked-by: Stefan Wahren \u003cstefan.wahren@i2se.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 2eced4607a1e6f51f928ae3e521fe02be5cb7d23\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Mon Aug 19 18:54:02 2019 +0200\n\n soc: renesas: Enable ARM_ERRATA_754322 for affected Cortex-A9\n \n ARM Erratum 754322 affects Cortex-A9 revisions r2p* and r3p*.\n \n Automatically enable support code to mitigate the erratum when compiling\n a kernel for any of the affected Renesas SoCs:\n - RZ/A1: r3p0,\n - R-Mobile A1: r2p4,\n - R-Car M1A: r2p2-00rel0,\n - R-Car H1: r3p0,\n - SH-Mobile AG5: r2p2.\n \n EMMA Mobile EV2 (r1p3) and RZ/A2 (r4p1) are not affected.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n\ncommit 435dce2da218b0eb1ff758b77831562845394848\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Mon Aug 19 19:05:58 2019 +0200\n\n soc: renesas: Enable ARM_ERRATA_814220 for affected Cortex-A7\n \n ARM Erratum 814220 affects Cortex-A7 revisions r0p2-r0p5.\n \n Automatically enable support code to mitigate the erratum when compiling\n a kernel for any of the affected Renesas SoCs:\n - R-Mobile APE6: r0p2,\n - RZ/G1E: r0p5,\n - RZ/G1C: r0p5,\n - R-Car H2: r0p3,\n - R-Car E2: r0p5,\n - RZ/N1: r0p5.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n\ncommit 93d051550ee02eaff9a2541d825605a7bd778027\nAuthor: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\nDate: Wed Aug 21 13:10:04 2019 +0800\n\n iommu/amd: Override wrong IVRS IOAPIC on Raven Ridge systems\n \n Raven Ridge systems may have malfunction touchpad or hang at boot if\n incorrect IVRS IOAPIC is provided by BIOS.\n \n Users already found correct \"ivrs_ioapic\u003d\" values, let\u0027s put them inside\n kernel to workaround buggy BIOS.\n \n BugLink: https://bugs.launchpad.net/bugs/1795292\n BugLink: https://bugs.launchpad.net/bugs/1837688\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 4cdc12a3ef424361f81bb30a34a3148b03df640c\nAuthor: Ondrej Jirman \u003cmegous@megous.com\u003e\nDate: Tue Aug 20 17:19:34 2019 +0200\n\n arm64: dts: allwinner: h6: Add support for RTC and fix the clock tree\n \n This patch adds RTC node and fixes the clock properties and nodes\n to reflect the real clock tree.\n \n The device nodes for the internal oscillator and osc32k are removed,\n as these clocks are now provided by the RTC device. Clock references\n are fixed accordingly, too.\n \n Signed-off-by: Ondrej Jirman \u003cmegous@megous.com\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit c8fb436b3b633183a4c2b5a958c01baaf8bc42a7\nAuthor: Joerg Roedel \u003cjroedel@suse.de\u003e\nDate: Mon Aug 19 15:22:56 2019 +0200\n\n Documentation: Update Documentation for iommu.passthrough\n \n This kernel parameter now takes also effect on X86.\n \n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 2cc13bb4f59fa7e66acf5b1b78bdf97d73d3416a\nAuthor: Joerg Roedel \u003cjroedel@suse.de\u003e\nDate: Mon Aug 19 15:22:55 2019 +0200\n\n iommu: Disable passthrough mode when SME is active\n \n Using Passthrough mode when SME is active causes certain\n devices to use the SWIOTLB bounce buffer. The bounce buffer\n code has an upper limit of 256kb for the size of DMA\n allocations, which is too small for certain devices and\n causes them to fail.\n \n With this patch we enable IOMMU by default when SME is\n active in the system, making the default configuration work\n for more systems than it does now.\n \n Users that don\u0027t want IOMMUs to be enabled still can disable\n them with kernel parameters.\n \n Reviewed-by: Tom Lendacky \u003cthomas.lendacky@amd.com\u003e\n Tested-by: Tom Lendacky \u003cthomas.lendacky@amd.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 22bb182c839d8ef6c08cf548feb0451c429216d8\nAuthor: Joerg Roedel \u003cjroedel@suse.de\u003e\nDate: Mon Aug 19 15:22:54 2019 +0200\n\n iommu: Set default domain type at runtime\n \n Set the default domain-type at runtime, not at compile-time.\n This keeps default domain type setting in one place when we\n have to change it at runtime.\n \n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 5fa9e7c5fa50f27afd71f2f6373179739bfa4034\nAuthor: Joerg Roedel \u003cjroedel@suse.de\u003e\nDate: Mon Aug 19 15:22:53 2019 +0200\n\n iommu: Print default domain type on boot\n \n Introduce a subsys_initcall for IOMMU code and use it to\n print the default domain type at boot.\n \n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 74bc0f6b22fd1a3c839f61d8475335a98af53c7a\nAuthor: Joerg Roedel \u003cjroedel@suse.de\u003e\nDate: Mon Aug 19 15:22:52 2019 +0200\n\n ia64: Get rid of iommu_pass_through\n \n This variable has no users anymore so it can be removed.\n \n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit c53c47aac4ff9509e7e1a32666baefbd9793c5a0\nAuthor: Joerg Roedel \u003cjroedel@suse.de\u003e\nDate: Mon Aug 19 15:22:51 2019 +0200\n\n x86/dma: Get rid of iommu_pass_through\n \n This variable has no users anymore. Remove it and tell the\n IOMMU code via its new functions about requested DMA modes.\n \n Reviewed-by: Borislav Petkov \u003cbp@suse.de\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 6b9a7d3a466fffc97c4b133852d3dd7905510dd6\nAuthor: Joerg Roedel \u003cjroedel@suse.de\u003e\nDate: Mon Aug 19 15:22:50 2019 +0200\n\n iommu/vt-d: Request passthrough mode from IOMMU core\n \n Get rid of the iommu_pass_through variable and request\n passthrough mode via the new iommu core function.\n \n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit cc7c8ad9736b5db994b84cbb1347972d28852d3b\nAuthor: Joerg Roedel \u003cjroedel@suse.de\u003e\nDate: Mon Aug 19 15:22:49 2019 +0200\n\n iommu/amd: Request passthrough mode from IOMMU core\n \n Get rid of the iommu_pass_through variable and request\n passthrough mode via the new iommu core function.\n \n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit adab0b07cbbc73f9fc338e4fc1749714dd093a7c\nAuthor: Joerg Roedel \u003cjroedel@suse.de\u003e\nDate: Mon Aug 19 15:22:48 2019 +0200\n\n iommu: Use Functions to set default domain type in iommu_set_def_domain_type()\n \n There are functions now to set the default domain type which\n take care of updating other necessary state. Don\u0027t open-code\n it in iommu_set_def_domain_type() and use those functions\n instead.\n \n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 8a69961c7f7583742ab9064feab5ea533a6b1b97\nAuthor: Joerg Roedel \u003cjroedel@suse.de\u003e\nDate: Mon Aug 19 15:22:47 2019 +0200\n\n iommu: Add helpers to set/get default domain type\n \n Add a couple of functions to allow changing the default\n domain type from architecture code and a function for iommu\n drivers to request whether the default domain is\n passthrough.\n \n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit faf1498993cdf65fd3a624b7653bc91909135a55\nAuthor: Joerg Roedel \u003cjroedel@suse.de\u003e\nDate: Mon Aug 19 15:22:46 2019 +0200\n\n iommu: Remember when default domain type was set on kernel command line\n \n Introduce an extensible concept to remember when certain\n configuration settings for the IOMMU code have been set on\n the kernel command line.\n \n This will be used later to prevent overwriting these\n settings with other defaults.\n \n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 2dc889a8846d16574794669bbdca7401900695b7\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Wed Aug 14 18:05:12 2019 +0530\n\n dt-bindings: pinctrl: qcom-pmic-gpio: Add pm8150l support\n \n Add support for the PM8150l GPIO support to the Qualcomm PMIC GPIO\n binding.\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190814123512.6017-3-vkoul@kernel.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 9bdacadd160687599ded0b796d089a918e4ea3f9\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Wed Aug 14 18:05:11 2019 +0530\n\n dt-bindings: pinctrl: qcom-pmic-gpio: Add pm8150b support\n \n Add support for the PM8150b GPIO support to the Qualcomm PMIC GPIO\n binding.\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190814123512.6017-2-vkoul@kernel.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit af0bc634728c0bc6a3f66f911f227d5c6396db88\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Fri Aug 9 15:43:07 2019 +0200\n\n soc: renesas: rmobile-sysc: Set GENPD_FLAG_ALWAYS_ON for always-on domain\n \n Currently the R-Mobile \"always-on\" PM Domain is implemented by returning\n -EBUSY from the generic_pm_domain.power_off() callback, and doing\n nothing in the generic_pm_domain.power_on() callback. However, this\n means the PM Domain core code is not aware of the semantics of this\n special domain, leading to boot warnings like the following on\n SH/R-Mobile SoCs:\n \n sh_cmt e6130000.timer: PM domain c5 will not be powered off\n \n Fix this by making the always-on nature of the domain explicit instead,\n by setting the GENPD_FLAG_ALWAYS_ON flag. This removes the need for the\n domain to provide power control callbacks.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit d67070c372dda5d15730182d8640998601b6d3f0\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Wed Aug 14 18:05:10 2019 +0530\n\n dt-bindings: pinctrl: qcom-pmic-gpio: Add pm8150 support\n \n Add support for the PM8150 GPIO support to the Qualcomm PMIC GPIO\n binding.\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190814123512.6017-1-vkoul@kernel.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit d21b8adbd475dba19ac2086d3306327b4a297418\nAuthor: Daniel Drake \u003cdrake@endlessm.com\u003e\nDate: Wed Aug 14 17:05:40 2019 +0800\n\n pinctrl: amd: disable spurious-firing GPIO IRQs\n \n When cold-booting Asus X434DA, GPIO 7 is found to be already configured\n as an interrupt, and the GPIO level is found to be in a state that\n causes the interrupt to fire.\n \n As soon as pinctrl-amd probes, this interrupt fires and invokes\n amd_gpio_irq_handler(). The IRQ is acked, but no GPIO-IRQ handler was\n invoked, so the GPIO level being unchanged just causes another interrupt\n to fire again immediately after.\n \n This results in an interrupt storm causing this platform to hang\n during boot, right after pinctrl-amd is probed.\n \n Detect this situation and disable the GPIO interrupt when this happens.\n This enables the affected platform to boot as normal. GPIO 7 actually is\n the I2C touchpad interrupt line, and later on, i2c-multitouch loads and\n re-enables this interrupt when it is ready to handle it.\n \n Instead of this approach, I considered disabling all GPIO interrupts at\n probe time, however that seems a little risky, and I also confirmed that\n Windows does not seem to have this behaviour: the same 41 GPIO IRQs are\n enabled under both Linux and Windows, which is a far larger collection\n than the GPIOs referenced by the DSDT on this platform.\n \n Signed-off-by: Daniel Drake \u003cdrake@endlessm.com\u003e\n Link: https://lore.kernel.org/r/20190814090540.7152-1-drake@endlessm.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit d5d4aa1ec1983911b0e1c656c97da16332f85fc9\nAuthor: Denis Efremov \u003cefremov@linux.com\u003e\nDate: Tue Aug 13 09:15:47 2019 +0300\n\n MAINTAINERS: Remove FMC subsystem\n \n Cleanup MAINTAINERS from FMC record since the subsystem was removed.\n \n Cc: Pat Riehecky \u003criehecky@fnal.gov\u003e\n Fixes: 6a80b30086b8 (\"fmc: Delete the FMC subsystem\")\n Signed-off-by: Denis Efremov \u003cefremov@linux.com\u003e\n Link: https://lore.kernel.org/r/20190813061547.17847-1-efremov@linux.com\n Reviewed-by: Federico Vaga \u003cfederico.vaga@cern.ch\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit d2b9c64443013197fbb71984e85f4c9551ffe26c\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Thu Aug 22 10:21:15 2019 +0200\n\n ARM: dts: sun7i: Add CSI0 controller\n \n The CSI controller embedded in the A20 can be supported by our new driver.\n Let\u0027s add it to our DT.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 74639d66e1ec069d27eebb97b545916a86d983ea\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Fri Aug 9 16:40:45 2019 +0200\n\n gpio: tqmx86: Pass irqchip when adding gpiochip\n \n We need to convert all old gpio irqchips to pass the irqchip\n setup along when adding the gpio_chip. For more info see\n drivers/gpio/TODO.\n \n For chained irqchips this is a pretty straight-forward\n conversion.\n \n Cc: Andrew Lunn \u003candrew@lunn.ch\u003e\n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190809144045.26018-1-linus.walleij@linaro.org\n\ncommit e599256ab78b99614b361b551145f3f95c9b3041\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Fri Aug 9 16:19:16 2019 +0200\n\n gpio: vf610: Pass irqchip when adding gpiochip\n \n We need to convert all old gpio irqchips to pass the irqchip\n setup along when adding the gpio_chip. For more info see\n drivers/gpio/TODO.\n \n For chained irqchips this is a pretty straight-forward\n conversion.\n \n Cc: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Andrew Lunn \u003candrew@lunn.ch\u003e\n Cc: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190809141916.20999-1-linus.walleij@linaro.org\n\ncommit 49751efbf63024816d2791e2a173b35c3112f32d\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Fri Aug 9 15:38:45 2019 +0200\n\n gpio: zx: Pass irqchip when adding gpiochip\n \n We need to convert all old gpio irqchips to pass the irqchip\n setup along when adding the gpio_chip. For more info see\n drivers/gpio/TODO.\n \n For chained irqchips this is a pretty straight-forward\n conversion.\n \n Cc: Jonas Gorski \u003cjogo@openwrt.org\u003e\n Cc: Jun Nie \u003cjun.nie@linaro.org\u003e\n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190809133845.30991-1-linus.walleij@linaro.org\n\ncommit 7b732209eb5d78ab114ae5e2a480d7417f30a2b1\nAuthor: Song Hui \u003chui.song_1@nxp.com\u003e\nDate: Thu Aug 8 18:16:28 2019 +0800\n\n gpio: mpc8xxx: add ls1088a platform special function\n \n ls1028a and ls1088a platform share common special function.\n The gpio hardware what they use is the same version.\n \n Signed-off-by: Song Hui \u003chui.song_1@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190808101628.36782-3-hui.song_1@nxp.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit ff6cd68d6601f9aae3eccd691af68f47cf81ec7d\nAuthor: Song Hui \u003chui.song_1@nxp.com\u003e\nDate: Thu Aug 8 18:16:26 2019 +0800\n\n gpio: mpc8xxx: add ls1088a platform gpio node DT binding description\n \n ls1088a and ls1028a platform share common gpio node.\n \n Signed-off-by: Song Hui \u003chui.song_1@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190808101628.36782-1-hui.song_1@nxp.com\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit e26ae7c0432101a924cf745b07470c8592de64cb\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Thu Jun 13 13:57:49 2019 +0200\n\n omapdrm: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n Cc: Jyri Sarha \u003cjsarha@ti.com\u003e\n Cc: Tony Lindgren \u003ctony@atomide.com\u003e\n Cc: zhong jiang \u003czhongjiang@huawei.com\u003e\n Cc: dri-devel@lists.freedesktop.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190613115749.GC26335@kroah.com\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190704023557.4551-1-huangfq.daxian@gmail.com\n\ncommit 3037e0c5e8af377c33f264f2e146d8b3b82bcb1b\nAuthor: Jean-Jacques Hiblot \u003cjjhiblot@ti.com\u003e\nDate: Thu Jul 11 15:52:19 2019 +0200\n\n drm/omap: Add \u0027alpha\u0027 and \u0027pixel blend mode\u0027 plane properties\n \n Add the following properties for planes:\n * alpha\n * pixel blend mode. Only \"Pre-multiplied\" and \"Coverage\" are supported\n \n Signed-off-by: Jean-Jacques Hiblot \u003cjjhiblot@ti.com\u003e\n Signed-off-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711135219.23402-1-jjhiblot@ti.com\n\ncommit 94f68f3a4b2ae71ef4cd4a4be94d76c98df9625b\nAuthor: Sunil Mohan Adapa \u003csunil@medhas.org\u003e\nDate: Wed Aug 21 12:52:17 2019 -0700\n\n arm64: dts: allwinner: a64: Add A64 OlinuXino board (with eMMC)\n \n A64 OLinuXino board from Olimex has three variants with onboard eMMC:\n A64-OLinuXino-1Ge16GW, A64-OLinuXino-1Ge4GW and A64-OLinuXino-2Ge8G-IND. In\n addition, there are two variants without eMMC. One without eMMC and one with SPI\n flash. This suggests the need for separate device tree for the three eMMC\n variants.\n \n This patch has been tested on A64-OLinuXino-1Ge16GW with Linux 5.0 from Debain.\n Basic benchmarks using Flexible IO Tester show reasonable performance from the\n eMMC.\n \n eMMC - Random Write: 21.3MiB/s\n eMMC - Sequential Write: 68.2MiB/s\n SD Card - Random Write: 1690KiB/s\n SD Card - Sequential Write: 11.0MiB/s\n \n Changes:\n \n v3: Separate dts for eMMC variants\n \n v2: Fix descriptions for VCC and VCCQ\n \n Link: https://github.com/armbian/build/commit/174953de1eb09e6aa1ef7075066b573dba625398\n Signed-off-by: Martin Ayotte \u003cmartinayotte@gmail.com\u003e\n [sunil@medhas.org Fix descriptions for VCC and VCCQ, separate dts for eMMC]\n Signed-off-by: Sunil Mohan Adapa \u003csunil@medhas.org\u003e\n Tested-by: Sunil Mohan Adapa \u003csunil@medhas.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 2063288662f2dbfa25d4bc25526f0543308bc077\nAuthor: Sunil Mohan Adapa \u003csunil@medhas.org\u003e\nDate: Wed Aug 21 12:52:16 2019 -0700\n\n dt-bindings: arm: sunxi: Add compatible for A64 OlinuXino with eMMC\n \n A64 OLinuXino board from Olimex has three variants with onboard eMMC:\n A64-OLinuXino-1Ge16GW, A64-OLinuXino-1Ge4GW and A64-OLinuXino-2Ge8G-IND. In\n addition, there are two variants without eMMC. One without eMMC and one with SPI\n flash. This suggests the need for separate device tree for the three eMMC\n variants.\n \n Add new compatible string to the bindings documentation for the A64 OlinuXino\n board variant with on-board eMMC.\n \n Signed-off-by: Sunil Mohan Adapa \u003csunil@medhas.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 18742b249e37ef13967a34eaf26270a0a00477f6\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jul 22 10:12:29 2019 +0200\n\n ARM: dts: v3s: Change the timers compatible\n \n Unlike the A10 that has 6 timers available, the v3s has only three, with only\n three interrupts. Let\u0027s change the compatible to reflect that, and add the\n missing interrupts.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 19aeb5a80cfac916225aca6143ffb1785c780c51\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jul 22 10:12:28 2019 +0200\n\n ARM: dts: h3: Change the timers compatible\n \n Unlike the A10 that has 6 timers available, the H3 has only two, with only\n two interrupts, just like the A23. Let\u0027s change the compatible to reflect\n that.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit df75eaac4938b0a8a5e4de65526a37e67bd33d94\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jul 22 10:12:27 2019 +0200\n\n ARM: dts: a83t: Change the timers compatible\n \n Unlike the A10 that has 6 timers available, the A83t has only two, with\n only two interrupts, just like the A23. Let\u0027s change the compatible to\n reflect that.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 2b9df83fa6b0d1c393c9aa0dfa0c87892b1ac807\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jul 22 10:12:26 2019 +0200\n\n ARM: dts: a23/a33: Change the timers compatible\n \n Unlike the A10 that has 6 timers available, the A23 and A33 has only two,\n with only two interrupts. Let\u0027s change the compatible to reflect that.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 628f020d13209ffef6d12ad7868d8cf11ef22c95\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jul 22 10:12:25 2019 +0200\n\n ARM: dts: sun6i: Add missing timers interrupts\n \n The timer unit in the A31 has 6 interrupts available. List all of them.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit f49f797c2b4e79df328008bb995adbd180f72ed2\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jul 22 10:12:24 2019 +0200\n\n ARM: dts: sun5i: Add missing timers interrupts\n \n The timer unit in the sun5i die has 6 interrupts available. List all of\n them.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 14c17ed24851746b20480d82eba8a67e6a572ea1\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jul 22 10:12:23 2019 +0200\n\n ARM: dts: sun4i: Add missing timers interrupts\n \n The timer unit in the A10 has 6 interrupts available. List all of them.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 0988e6c2cd6cd64f7272c18fe58f3fca26b9120d\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Tue Aug 13 14:47:40 2019 +0200\n\n dt-bindings: mfd: Convert Allwinner GPADC bindings to a schema\n \n The Allwinner SoCs have an embedded GPADC that is doing thermal reading as\n well, supported in Linux, with a matching Device Tree binding.\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for that controller over to a YAML schemas.\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 652a458eb92018c5126701e721255356fdab94a9\nAuthor: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\nDate: Fri Aug 16 22:53:42 2019 +0200\n\n arm64: dts: allwinner: h6: Introduce Tanix TX6 board\n \n Tanix TX6 is an Allwinner H6 based TV box, which supports:\n - Allwinner H6 Quad-core 64-bit ARM Cortex-A53\n - GPU Mali-T720\n - 4GiB DDR3 RAM (3GiB useable)\n - 100Mbps EMAC via AC200 EPHY\n - Cdtech 47822BS Wifi/BT\n - 2x USB 2.0 Host and 1x USB 3.0 Host\n - HDMI port\n - IR receiver\n - 64GiB eMMC\n - 5V/2A DC power supply\n \n Signed-off-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 8e3efec9b97fc797d4558b72ac2c1587e16cae6c\nAuthor: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\nDate: Fri Aug 16 22:53:41 2019 +0200\n\n dt-bindings: arm: sunxi: Add compatible for Tanix TX6 board\n \n Add new Oranth Tanix TX6 board compatible string to the bindings\n documentation.\n \n Signed-off-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 89336e1f096d6e8d99e74df63652f8e64d788cac\nAuthor: Bhushan Shah \u003cbshah@kde.org\u003e\nDate: Fri Aug 16 14:13:09 2019 +0530\n\n arm64: allwinner: h6: add I2C nodes\n \n Add device-tree nodes for i2c0 to i2c2, and also add relevant pinctrl\n nodes.\n \n Suggested-by: Icenowy Zheng \u003cicenowy@aosc.io\u003e\n Signed-off-by: Bhushan Shah \u003cbshah@kde.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 88dc4e7ea714787642708fb0fc460bb7e262b234\nAuthor: Bhushan Shah \u003cbshah@kde.org\u003e\nDate: Fri Aug 16 14:13:08 2019 +0530\n\n dt-bindings: i2c: mv64xxx: Add compatible for the H6 i2c node.\n \n Allwinner H6 have a mv64xxx i2c interface available to be used.\n \n Signed-off-by: Bhushan Shah \u003cbshah@kde.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 968f2c9169399fce43c11a89cbf4ac697f86de6e\nAuthor: Chen-Yu Tsai \u003cwens@csie.org\u003e\nDate: Wed Aug 14 12:22:08 2019 +0800\n\n ARM: dts: sunxi: Add mdio bus sub-node to GMAC\n \n The DWMAC binding never supported having the Ethernet PHY node as a\n direct child to the controller, nor did it support the \"phy\" property\n as a way to specify which Ethernet PHY to use. What seemed to work\n was simply the implementation ignoring the \"phy\" property and instead\n probing all addresses on the MDIO bus and using the first available\n one.\n \n The recent switch from \"phy\" to \"phy-handle\" breaks the assumptions\n of the implementation, and does not match what the binding requires.\n The binding requires that if an MDIO bus is described, it shall be\n a sub-node with the \"snps,dwmac-mdio\" compatible string.\n \n Add a device node for the MDIO bus, and move the Ethernet PHY node\n under it. Also fix up the #address-cells and #size-cells properties\n where needed.\n \n Fixes: de332de26d19 (\"ARM: dts: sunxi: Switch from phy to phy-handle\")\n Signed-off-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit f46f408c152ac925e56c0f38138ae49ba16bbc23\nAuthor: Clément Péron \u003cperon.clem@gmail.com\u003e\nDate: Mon Aug 12 12:23:55 2019 +0200\n\n arm64: dts: allwinner: Enable DDC regulator for Beelink GS1\n \n Beelink GS1 has a DDC I2C bus voltage shifter. This is actually missing\n and video is limited to 1024x768 due to missing EDID information.\n \n Add the DDC regulator in the device-tree.\n \n Signed-off-by: Clément Péron \u003cperon.clem@gmail.com\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 45dd5cf87291fabcef8ce8817d8f2277ccdd14f8\nAuthor: Clément Péron \u003cperon.clem@gmail.com\u003e\nDate: Mon Aug 12 12:51:15 2019 +0200\n\n arm64: dts: allwinner: h6: Enable SPDIF for Beelink GS1\n \n Beelink GS1 board has a SPDIF out connector, so enable it in\n the device-tree and add a simple SPDIF soundcard.\n \n Signed-off-by: Clément Péron \u003cperon.clem@gmail.com\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit f95b598df4195c7e2a7116ba3eeb3742a4cb3e9b\nAuthor: Clément Péron \u003cperon.clem@gmail.com\u003e\nDate: Mon Aug 12 12:51:14 2019 +0200\n\n arm64: dts: allwinner: Add SPDIF node for Allwinner H6\n \n The Allwinner H6 has a SPDIF controller called OWA (One Wire Audio).\n \n Only one pinmuxing is available so set it as default.\n \n Signed-off-by: Clément Péron \u003cperon.clem@gmail.com\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 6f002c57c74616ab2bfd236f48bf254c30c5f36a\nAuthor: Icenowy Zheng \u003cicenowy@aosc.io\u003e\nDate: Sun Jul 28 11:12:27 2019 +0800\n\n ARM: dts: sun8i: s3: add devicetree for Lichee zero plus w/ S3\n \n Lichee zero plus is a core board made by Sipeed, which includes on-board\n TF slot or SMT SD NAND, and optional SPI NOR or eMMC, a UART debug\n header, a microUSB slot and a gold finger connector for expansion. It\n can use either Sochip S3 or Allwinner S3L SoC.\n \n Add the basic device tree for the core board, w/o optional onboard\n storage, and with S3 SoC.\n \n Signed-off-by: Icenowy Zheng \u003cicenowy@aosc.io\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 11d1bdead79c0df33ba831d307af76730e4c1559\nAuthor: Icenowy Zheng \u003cicenowy@aosc.io\u003e\nDate: Sun Jul 28 11:12:25 2019 +0800\n\n ARM: sunxi: dts: s3/s3l/v3: add DTSI files for S3/S3L/V3 SoCs\n \n The Allwinner S3/S3L/V3 SoCs all share the same die with the V3s SoC,\n but with more GPIO wired out of the package.\n \n Add a DTSI file for these SoCs. It just replaces some compatible strings\n of the V3s DTSI now. As these SoCs share the same feature set on Linux,\n we use the first known chip (V3) as the file\u0027s name.\n \n Signed-off-by: Icenowy Zheng \u003cicenowy@aosc.io\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit d400cc4ad18f55a8253b2694f5594738e5cfacc2\nAuthor: Icenowy Zheng \u003cicenowy@aosc.io\u003e\nDate: Sun Jul 28 11:12:26 2019 +0800\n\n dt-bindings: arm: sunxi: add binding for Lichee Zero Plus core board\n \n The Lichee Zero Plus is a core board made by Sipeed, with a microUSB\n connector on it, TF slot or WSON8 SD chip, optional eMMC or SPI Flash.\n It has a gold finger connector for expansion, and UART is available from\n reserved pins w/ 2.54mm pitch. The board can use either SoChip S3 or\n Allwinner V3L SoCs.\n \n Add the device tree binding of the basic version of the core board --\n w/o eMMC or SPI Flash, w/ TF slot or WSON8 SD, and use S3 SoC.\n \n Signed-off-by: Icenowy Zheng \u003cicenowy@aosc.io\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 802cbe1b46dca72c922f1ed33333e1d19e4c988e\nAuthor: Ondrej Jirman \u003cmegous@megous.com\u003e\nDate: Tue Aug 6 17:57:43 2019 +0200\n\n arm64: dts: allwinner: orange-pi-3: Enable HDMI output\n \n Orange Pi 3 has a DDC_CEC_EN signal connected to PH2, that enables the DDC\n I2C bus voltage shifter. Before EDID can be read, we need to pull PH2 high.\n This is realized by the ddc-en-gpios property.\n \n Signed-off-by: Ondrej Jirman \u003cmegous@megous.com\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 56e7c8e021443d7f07e1399f86720448ee8377c5\nAuthor: Chen-Yu Tsai \u003cwens@csie.org\u003e\nDate: Sun Jul 28 22:59:44 2019 +0800\n\n ARM: dts: sun8i: a83t: Enable HDMI output on Cubietruck Plus\n \n The Cubietruck Plus has an HDMI connector tied to the HDMI output of the\n SoC.\n \n Enables display output via HDMI on the Cubietruck Plus. The connector\n device node is named \"hdmi-connector\" as there is also a display port\n connector, which is tied to the MIPI DSI output of the SoC through a\n MIPI-DSI-to-DP bridge. This part is not supported yet.\n \n Signed-off-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 0a6864274e4166cde21f26193350c7fcd9716ef5\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Tue Aug 20 15:59:55 2019 +0200\n\n pinctrl: rza2: Include the appropriate headers\n \n This driver is implementing a GPIO driver so include\n \u003clinux/gpio/driver.h\u003e and not the legacy API \u003clinux/gpio.h\u003e.\n When testing it turns out it also relies on implicit\n inclusion of \u003clinux/io.h\u003e (readw etc) so make sure to\n include that as well.\n \n Cc: Chris Brandt \u003cchris.brandt@renesas.com\u003e\n Cc: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 5b1d96e029bff720ca1067a1222f4b9387b5c214\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Tue Aug 20 15:59:54 2019 +0200\n\n pinctrl: rza2: Drop driver use of consumer flags\n \n These flags are for consumers of GPIO lines, not for\n drivers.\n \n Cc: Chris Brandt \u003cchris.brandt@renesas.com\u003e\n Cc: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit ab6366ee887a273d2d33a204b7ea6e6fb2fe20e4\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 15 11:35:03 2019 +0530\n\n pinctrl: rza1: Add of_node_put() before return\n \n Each iteration of for_each_child_of_node puts the previous node, but in\n the case of a return from the middle of the loop, there is no put, thus\n causing a memory leak. Hence add an of_node_put before the return in\n three places.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit df62267ddec54d93b6e280ba57052e7ecf3633f4\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Wed Aug 14 09:20:32 2019 +0200\n\n pinctrl: sh-pfc: Include the right header\n \n This is a GPIO driver, use the appropriate header\n \u003clinux/gpio/driver.h\u003e rather than the legacy \u003clinux/gpio.h\u003e\n header.\n \n Cc: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 8a0cc47ccc7c17826c669787f95ab7e5efb37444\nAuthor: Yoshihiro Shimoda \u003cyoshihiro.shimoda.uh@renesas.com\u003e\nDate: Thu Aug 8 15:19:03 2019 +0900\n\n pinctrl: sh-pfc: Rollback to mux if required when the gpio is freed\n \n Some drivers require switching between function and gpio at run-time.\n Allow to roll back from gpio to mux when the gpio is freed.\n \n Signed-off-by: Yoshihiro Shimoda \u003cyoshihiro.shimoda.uh@renesas.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 395713d8ca434967808670000e6baa3505e8e947\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Wed Aug 21 13:58:21 2019 -0500\n\n soundwire: intel: handle disabled links\n \n On most hardware platforms, SoundWire interfaces are pin-muxed with\n other interfaces (typically DMIC or I2S) and the status of each link\n needs to be checked at boot time.\n \n For Intel platforms, the BIOS provides a menu to enable/disable the\n links separately, and the information is provided to the OS with an\n Intel-specific _DSD property. The same capability will be added to\n revisions of the MIPI DisCo specification.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190821185821.12690-5-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 79ee6631264075f85dc22191dad801108fc35c15\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Wed Aug 21 13:58:20 2019 -0500\n\n soundwire: intel: add debugfs register dump\n \n Add debugfs file to dump the Intel SoundWire registers\n \n Credits: this patch is based on an earlier internal contribution by\n Vinod Koul, Sanyog Kale, Shreyas Nc and Hardik Shah.\n \n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Acked-by: Sanyog Kale \u003csanyog.r.kale@intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190821185821.12690-4-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit aa85066e5f26912662668535a400ed705eb4c0e7\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Wed Aug 21 13:58:19 2019 -0500\n\n soundwire: cadence_master: add debugfs register dump\n \n Add debugfs file to dump the Cadence master registers.\n \n Credits: this patch is based on an earlier internal contribution by\n Vinod Koul, Sanyog Kale, Shreyas Nc and Hardik Shah.\n \n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Acked-by: Sanyog Kale \u003csanyog.r.kale@intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190821185821.12690-3-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit bf03473d5bcc85fbe9533fa042f67809d8520c4e\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Wed Aug 21 13:58:18 2019 -0500\n\n soundwire: add debugfs support\n \n Add base debugfs mechanism for SoundWire bus by creating soundwire\n root and master-N and slave-x hierarchy.\n \n Also add SDW Slave SCP, DP0 and DP-N register debug file.\n \n Registers not implemented will print as \"XX\"\n \n Credits: this patch is based on an earlier internal contribution by\n Vinod Koul, Sanyog Kale, Shreyas Nc and Hardik Shah.\n \n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Acked-by: Sanyog Kale \u003csanyog.r.kale@intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190821185821.12690-2-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit d22deab6960a6cb015a36e74a2dcbab6ca9f5544\nAuthor: Suraj Jitindar Singh \u003csjitindarsingh@gmail.com\u003e\nDate: Tue Aug 20 16:13:49 2019 +1000\n\n KVM: PPC: Book3S HV: Define usage types for rmap array in guest memslot\n \n The rmap array in the guest memslot is an array of size number of guest\n pages, allocated at memslot creation time. Each rmap entry in this array\n is used to store information about the guest page to which it\n corresponds. For example for a hpt guest it is used to store a lock bit,\n rc bits, a present bit and the index of a hpt entry in the guest hpt\n which maps this page. For a radix guest which is running nested guests\n it is used to store a pointer to a linked list of nested rmap entries\n which store the nested guest physical address which maps this guest\n address and for which there is a pte in the shadow page table.\n \n As there are currently two uses for the rmap array, and the potential\n for this to expand to more in the future, define a type field (being the\n top 8 bits of the rmap entry) to be used to define the type of the rmap\n entry which is currently present and define two values for this field\n for the two current uses of the rmap array.\n \n Since the nested case uses the rmap entry to store a pointer, define\n this type as having the two high bits set as is expected for a pointer.\n Define the hpt entry type as having bit 56 set (bit 7 IBM bit ordering).\n \n Signed-off-by: Suraj Jitindar Singh \u003csjitindarsingh@gmail.com\u003e\n Signed-off-by: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\n\ncommit ff7240ccf0cdab86762635840c6d62eec7147d18\nAuthor: Paul Menzel \u003cpmenzel@molgen.mpg.de\u003e\nDate: Tue Jul 30 10:53:10 2019 +0200\n\n KVM: PPC: Book3S: Mark expected switch fall-through\n \n Fix the error below triggered by `-Wimplicit-fallthrough`, by tagging\n it as an expected fall-through.\n \n arch/powerpc/kvm/book3s_32_mmu.c: In function ‘kvmppc_mmu_book3s_32_xlate_pte’:\n arch/powerpc/kvm/book3s_32_mmu.c:241:21: error: this statement may fall through [-Werror\u003dimplicit-fallthrough\u003d]\n pte-\u003emay_write \u003d true;\n ~~~~~~~~~~~~~~~^~~~~~\n arch/powerpc/kvm/book3s_32_mmu.c:242:5: note: here\n case 3:\n ^~~~\n \n Signed-off-by: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\n\ncommit 8a2eaab7daf03b23ac902481218034ae2fae5e16\nAuthor: Yazen Ghannam \u003cyazen.ghannam@amd.com\u003e\nDate: Thu Aug 22 00:00:00 2019 +0000\n\n EDAC/amd64: Decode syndrome before translating address\n \n AMD Family 17h systems currently require address translation in order to\n report the system address of a DRAM ECC error. This is currently done\n before decoding the syndrome information. The syndrome information does\n not depend on the address translation, so the proper EDAC csrow/channel\n reporting can function without the address. However, the syndrome\n information will not be decoded if the address translation fails.\n \n Decode the syndrome information before doing the address translation.\n The syndrome information is architecturally defined in MCA_SYND and can\n be considered robust. The address translation is system-specific and may\n fail on newer systems without proper updates to the translation\n algorithm.\n \n Fixes: 713ad54675fd (\"EDAC, amd64: Define and register UMC error decode function\")\n Signed-off-by: Yazen Ghannam \u003cyazen.ghannam@amd.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Cc: \"linux-edac@vger.kernel.org\" \u003clinux-edac@vger.kernel.org\u003e\n Cc: James Morse \u003cjames.morse@arm.com\u003e\n Cc: Mauro Carvalho Chehab \u003cmchehab@kernel.org\u003e\n Cc: Tony Luck \u003ctony.luck@intel.com\u003e\n Link: https://lkml.kernel.org/r/20190821235938.118710-6-Yazen.Ghannam@amd.com\n\ncommit befd04abfbe4b933515dddb5659d0744be9dba6a\nAuthor: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\nDate: Tue Aug 20 21:36:07 2019 +0900\n\n clocksource/drivers/sh_cmt: Document \"cmt-48\" as deprecated\n \n Update the CMT driver to mark \"renesas,cmt-48\" as deprecated.\n \n Instead of documenting a theoretical hardware device based on current software\n support level, define DT bindings top-down based on available data sheet\n information and make use of part numbers in the DT compat string.\n \n In case of the only in-tree users r8a7740 and sh73a0 the compat strings\n \"renesas,r8a7740-cmt1\" and \"renesas,sh73a0-cmt1\" may be used instead.\n \n Signed-off-by: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 81b56da380a671f7077acf5afb5078d9f8e0ade6\nAuthor: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\nDate: Tue Aug 20 21:35:56 2019 +0900\n\n clocksource/drivers/sh_cmt: r8a7740 and sh73a0 SoC-specific match\n \n Add SoC-specific matching for CMT1 on r8a7740 and sh73a0.\n \n This allows us to move away from the old DT bindings such as\n - \"renesas,cmt-48-sh73a0\"\n - \"renesas,cmt-48-r8a7740\"\n - \"renesas,cmt-48\"\n in favour for the now commonly used format \"renesas,\u003csoc\u003e-\u003cdevice\u003e\"\n \n Signed-off-by: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 1257887a2a76c90747b4ee2d502b5c29c0174c84\nAuthor: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\nDate: Tue Aug 20 21:35:46 2019 +0900\n\n dt-bindings: timer: renesas, cmt: Update R-Car Gen3 CMT1 usage\n \n The R-Car Gen3 SoCs so far come with a total for 4 on-chip CMT devices:\n - CMT0\n - CMT1\n - CMT2\n - CMT3\n \n CMT0 includes two rather basic 32-bit timer channels. The rest of the on-chip\n CMT devices support 48-bit counters and have 8 channels each.\n \n Based on the data sheet information \"CMT2/3 are exactly same as CMT1\"\n it seems that CMT2 and CMT3 now use the CMT1 compat string in the DTSI.\n \n Clarify this in the DT binding documentation by describing R-Car Gen3 and\n RZ/G2 CMT1 as \"48-bit CMT devices\".\n \n Signed-off-by: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit b7bcb069e6afdf77f0ad1caff69aa9e48cabf8cc\nAuthor: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\nDate: Tue Aug 20 21:35:36 2019 +0900\n\n dt-bindings: timer: renesas, cmt: Add CMT0 and CMT1 to r8a77995\n \n This patch adds DT binding documentation for the CMT devices on\n the R-Car Gen3 D3 (r8a77995) SoC.\n \n Signed-off-by: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit d3bf92ae4b45ff93d37525aa42d1fe9cd7cadb3d\nAuthor: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\nDate: Tue Aug 20 21:35:25 2019 +0900\n\n dt-bindings: timer: renesas, cmt: Add CMT0 and CMT1 to r8a7792\n \n This patch adds DT binding documentation for the CMT devices on\n the R-Car Gen2 V2H (r8a7792) SoC.\n \n Signed-off-by: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit f7848023e0c8d24a91bb0240b1a8b3e36ab74ac7\nAuthor: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\nDate: Tue Aug 20 21:35:14 2019 +0900\n\n dt-bindings: timer: renesas, cmt: Update CMT1 on sh73a0 and r8a7740\n \n This patch reworks the DT binding documentation for the 6-channel\n 48-bit CMTs known as CMT1 on r8a7740 and sh73a0.\n \n After the update the same style of DT binding as the rest of the upstream\n SoCs will now also be used by r8a7740 and sh73a0. The DT binding \"cmt-48\"\n is removed from the DT binding documentation, however software support for\n this deprecated binding will still remain in the CMT driver for some time.\n \n Signed-off-by: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 48edf1ada4974401515fe869dd13508029527ac2\nAuthor: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\nDate: Tue Aug 20 21:35:03 2019 +0900\n\n dt-bindings: timer: renesas, cmt: Add CMT0234 to sh73a0 and r8a7740\n \n Document the on-chip CMT devices included in r8a7740 and sh73a0.\n \n Included in this patch is DT binding documentation for 32-bit CMTs\n CMT0, CMT2, CMT3 and CMT4. They all contain a single channel and are\n quite similar however some minor differences still exist:\n - \"Counter input clock\" (clock input and on-device divider)\n One example is that RCLK 1/1 is supported by CMT2, CMT3 and CMT4.\n - \"Wakeup request\" (supported by CMT0 and CMT2)\n \n Because of this one unique compat string per CMT device is selected.\n \n Signed-off-by: Magnus Damm \u003cdamm+renesas@opensource.se\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 4d3b79ef95cc7ec1ca2792c9b5d244dede596356\nAuthor: Jon Hunter \u003cjonathanh@nvidia.com\u003e\nDate: Wed Aug 21 16:02:41 2019 +0100\n\n clocksource/drivers: Do not warn on probe defer\n \n Deferred probe is an expected return value on many platforms and so\n there\u0027s no need to output a warning that may potentially confuse users.\n \n Signed-off-by: Jon Hunter \u003cjonathanh@nvidia.com\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 510a98d683cb989ef642b351258b6b1f0379836d\nAuthor: Jon Hunter \u003cjonathanh@nvidia.com\u003e\nDate: Wed Aug 21 16:02:40 2019 +0100\n\n clocksource/drivers/timer-of: Do not warn on deferred probe\n \n Deferred probe is an expected return value for clk_get() on many\n platforms. The driver deals with it properly, so there\u0027s no need\n to output a warning that may potentially confuse users.\n \n Signed-off-by: Jon Hunter \u003cjonathanh@nvidia.com\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit e53a3b267fb0a79db9ca1f1e08b97889b22013e6\nAuthor: Yazen Ghannam \u003cyazen.ghannam@amd.com\u003e\nDate: Wed Aug 21 23:59:59 2019 +0000\n\n EDAC/amd64: Find Chip Select memory size using Address Mask\n \n Chip Select memory size reporting on AMD Family 17h was recently fixed\n in order to account for interleaving. However, the current method is not\n robust.\n \n The Chip Select Address Mask can be used to find the memory size. There\n are a couple of cases.\n \n 1) For single-rank and dual-rank non-interleaved, use the address mask\n plus 1 as the size.\n \n 2) For dual-rank interleaved, do #1 but \"de-interleave\" the address mask\n first.\n \n Always \"de-interleave\" the address mask in order to simplify the code\n flow. Bit mask manipulation is necessary to check for interleaving, so\n just go ahead and do the de-interleaving. In the non-interleaved case,\n the original and de-interleaved address masks will be the same.\n \n To de-interleave the mask, count the number of zero bits in the middle\n of the mask and swap them with the most significant bits.\n \n For example,\n Original\u003d0xFFFF9FE, De-interleaved\u003d0x3FFFFFE\n \n Signed-off-by: Yazen Ghannam \u003cyazen.ghannam@amd.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Cc: \"linux-edac@vger.kernel.org\" \u003clinux-edac@vger.kernel.org\u003e\n Cc: James Morse \u003cjames.morse@arm.com\u003e\n Cc: Mauro Carvalho Chehab \u003cmchehab@kernel.org\u003e\n Cc: Tony Luck \u003ctony.luck@intel.com\u003e\n Link: https://lkml.kernel.org/r/20190821235938.118710-5-Yazen.Ghannam@amd.com\n\ncommit 353a1fcb8f9e5857c0fb720b9e57a86c1fb7c17e\nAuthor: Yazen Ghannam \u003cyazen.ghannam@amd.com\u003e\nDate: Wed Aug 21 23:59:57 2019 +0000\n\n EDAC/amd64: Initialize DIMM info for systems with more than two channels\n \n Currently, the DIMM info for AMD Family 17h systems is initialized in\n init_csrows(). This function is shared with legacy systems, and it has a\n limit of two channel support.\n \n This prevents initialization of the DIMM info for a number of ranks, so\n there will be missing ranks in the EDAC sysfs.\n \n Create a new init_csrows_df() for Family17h+ and revert init_csrows()\n back to pre-Family17h support.\n \n Loop over all channels in the new function in order to support systems\n with more than two channels.\n \n Signed-off-by: Yazen Ghannam \u003cyazen.ghannam@amd.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Cc: \"linux-edac@vger.kernel.org\" \u003clinux-edac@vger.kernel.org\u003e\n Cc: James Morse \u003cjames.morse@arm.com\u003e\n Cc: Mauro Carvalho Chehab \u003cmchehab@kernel.org\u003e\n Cc: Tony Luck \u003ctony.luck@intel.com\u003e\n Link: https://lkml.kernel.org/r/20190821235938.118710-4-Yazen.Ghannam@amd.com\n\ncommit f8be8e5680225ac9caf07d4545f8529b7395327f\nAuthor: Yazen Ghannam \u003cyazen.ghannam@amd.com\u003e\nDate: Wed Aug 21 23:59:56 2019 +0000\n\n EDAC/amd64: Recognize DRAM device type ECC capability\n \n AMD Family 17h systems support x4 and x16 DRAM devices. However, the\n device type is not checked when setting mci.edac_ctl_cap.\n \n Set the appropriate capability flag based on the device type.\n \n Default to x8 DRAM device when neither the x4 or x16 bits are set.\n \n [ bp: reverse cpk_en check to save an indentation level. ]\n \n Fixes: 2d09d8f301f5 (\"EDAC, amd64: Determine EDAC MC capabilities on Fam17h\")\n Signed-off-by: Yazen Ghannam \u003cyazen.ghannam@amd.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Cc: \"linux-edac@vger.kernel.org\" \u003clinux-edac@vger.kernel.org\u003e\n Cc: James Morse \u003cjames.morse@arm.com\u003e\n Cc: Mauro Carvalho Chehab \u003cmchehab@kernel.org\u003e\n Cc: Tony Luck \u003ctony.luck@intel.com\u003e\n Link: https://lkml.kernel.org/r/20190821235938.118710-3-Yazen.Ghannam@amd.com\n\ncommit cacc9539cff118735b11a350d121cb7eeca7e358\nAuthor: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nDate: Wed Jul 31 14:15:13 2019 +0200\n\n phy: mvebu-a3700-comphy: Inform users if their firmware is too old\n \n PHY configuration has been implemented in the firmware and accessed\n through SMC calls. In the past, it worked magically if the bootloader\n was correctly doing the initializations.\n \n With up-to-date bindings, the kernel will need a recent firmware in\n order to do the initializations himself (we assume people must update\n their firmware along with their kernel).\n \n People might not understand why IPs that were working correctly before\n stopped to be probed suddendly. In this case, let\u0027s advise the users\n to update their firmware with a visual warning.\n \n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit 6aeec986f1bc5d8b30bf23b2bf330deed28f37ea\nAuthor: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\nDate: Fri Jul 19 14:25:32 2019 +0200\n\n phy: samsung: disable bind/unbind platform driver feature\n \n Samsung PHY drivers control the power to the SoC core components needed by\n their client devices (USB HCDs, SATA, camera ISP bridge, DP encoder) to\n properly operate. Disabling PHYs in runtime usually causes the client\n device to crash with external abort exception or similar issue due to lack\n of API to notify clients about PHY removal. This patch removes the\n possiblity to unbind Samsung Exynos PHY drivers in runtime.\n \n Signed-off-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit bbae18f0fc621df81d05d614eff31f90495516d3\nAuthor: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\nDate: Mon Jul 22 19:32:23 2019 +0200\n\n phy: core: document phy_calibrate()\n \n Commit 36914111e682 (\"drivers: phy: add calibrate method\") added support\n for generic phy_calibrate() method, but it didn\u0027t explain in detail when\n such method is supposed to be called. Add some more documentation directly\n to the phy.h to make it clean that it is intended to be called after every\n host controller reset.\n \n Signed-off-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit beae796d66165b40d80b743ae301a29c312d7320\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Jul 23 16:21:08 2019 +0530\n\n phy: marvell: phy-mvebu-a3700-comphy: Add of_node_put() before return\n \n Each iteration of for_each_available_child_of_node puts the previous\n node, but in the case of a return from the middle of the loop, there is\n no put, thus causing a memory leak. Hence add an of_node_put before the\n return in two places.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit 5b43a20ac6c4c1b3445618c6e0bed6937709eea8\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Jul 23 16:23:27 2019 +0530\n\n phy: marvell: phy-mvebu-cp110-comphy: Add of_node_put() before return\n \n Each iteration of for_each_available_child_of_node puts the previous\n node, but in the case of a return from the middle of the loop, there is\n no put, thus causing a memory leak. Hence add an of_node_put before the\n return in two places.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit 8d160f6b491ec2bcfba2233121ab81f0424e133f\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Jul 23 16:19:19 2019 +0530\n\n phy: marvell: phy-armada38x-comphy: Add of_node_put() before return\n \n Each iteration of for_each_available_child_of_node puts the previous\n node, but in the case of a return from the middle of the loop, there is\n no put, thus causing a memory leak. Hence add an of_node_put before the\n return.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit 4e99276a6f7cf6cbf1d375e2da59c358ca0fd0a8\nAuthor: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\nDate: Sat Jul 27 14:04:14 2019 +0200\n\n phy: enable compile-testing for the Lantiq PHY drivers\n \n Unconditionally include the lantiq subdirectory in the phy Makefile.\n \n All drivers in there have their dependencies maintained. One of these\n (optional) dependencies is COMPILE_TEST, however this can only be\n evaluated when Kconfig scans the lantiq subdirectory.\n \n Signed-off-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit e52a632195bf43d1a91ae699e7536a6ead736aa7\nAuthor: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\nDate: Sat Jul 27 14:04:13 2019 +0200\n\n phy: lantiq: vrx200-pcie: add a driver for the Lantiq VRX200 PCIe PHY\n \n The Lantiq VRX200 SoCs embed a PCIe PHY in the \"sram\" bus. Unlike most\n other IP blocks on this SoC the register values are only 16-bit wide.\n Like other IP blocks on this SoC the register values are in big endian.\n \n The PHY embeds a PLL which can be configured in various modes. Only the\n 36MHz mode is supported for now, the other modes can be implemented when\n there\u0027s a board which actually needs them. OpenWrt uses the out-of-tree\n vendor driver and all supported boards there only need the 36MHz mode.\n \n There are two input clocks:\n - the \"pdi\" clock enables the register access\n - the \"phy\" clock is the clock input and enables the internal PLL\n \n There are two reset lines:\n - \"phy\" resets the PHY itself\n - the \"pcie\" reset line is shared between the PHY and the PCIe\n controller\n \n While the VRX200 SoC has only one PCIe controller and PHY the ARX300\n uses two identical PCIe controllers and PHYs which are compatible with\n the PCIe controller and PHY on VRX200.\n Add a driver for this PHY so PCIe support can be enabled on these SoCs.\n \n Signed-off-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit 088e88be5a380cc4e81963a9a02815da465d144f\nAuthor: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\nDate: Sat Jul 27 14:04:12 2019 +0200\n\n dt-bindings: phy: add binding for the Lantiq VRX200 and ARX300 PCIe PHYs\n \n Add the bindings for the PCIe PHY on Lantiq VRX200 and ARX300 SoCs.\n The IP block contains settings for the PHY and a PLL.\n The PLL mode is configurable through a dedicated #phy-cell in .dts.\n \n Signed-off-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit b87eabf8902da6bfb697d0ff4f75be32e2106711\nAuthor: Eric Whitney \u003cenwlinux@gmail.com\u003e\nDate: Thu Aug 22 23:22:14 2019 -0400\n\n ext4: rework reserved cluster accounting when invalidating pages\n \n The goal of this patch is to remove two references to the buffer delay\n bit in ext4_da_page_release_reservation() as part of a larger effort\n to remove all such references from ext4. These two references are\n principally used to reduce the reserved block/cluster count when pages\n are invalidated as a result of truncating, punching holes, or\n collapsing a block range in a file. The entire function is removed\n and replaced with code in ext4_es_remove_extent() that reduces the\n reserved count as a side effect of removing a block range from delayed\n and not unwritten extents in the extent status tree as is done when\n truncating, punching holes, or collapsing ranges.\n \n The code is written to minimize the number of searches descending from\n rb tree roots for scalability.\n \n Signed-off-by: Eric Whitney \u003cenwlinux@gmail.com\u003e\n Signed-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n\ncommit 219db95bbe796277c739cca17fe134f6c2ce6676\nAuthor: Ayush Ranjan \u003cayushr2@illinois.edu\u003e\nDate: Thu Aug 22 23:18:33 2019 -0400\n\n ext4: documentation fixes\n \n This commit aims to fix the following issues in ext4 documentation:\n - Flexible block group docs said that the aim was to group block\n metadata together instead of block group metadata.\n - The documentation consistly uses \"location\" instead of \"block number\".\n It is easy to confuse location to be an absolute offset on disk. Added\n a line to clarify all location values are in terms of block numbers.\n - Dirent2 docs said that the rec_len field is shortened instead of the\n name_len field.\n - Typo in bg_checksum description.\n - Inode size is 160 bytes now, and hence i_extra_isize is now 32.\n - Cluster size formula was incorrect, it did not include the +10 to\n s_log_cluster_size value.\n - Typo: there were two s_wtime_hi in the superblock struct.\n - Superblock struct was outdated, added the new fields which were part\n of s_reserved earlier.\n - Multiple mount protection seems to be implemented in fs/ext4/mmp.c.\n \n Signed-off-by: Ayush Ranjan \u003cayushr2@illinois.edu\u003e\n Signed-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Reviewed-by: Andreas Dilger \u003cadilger@dilger.ca\u003e\n\ncommit 7963e5ac901251c7a3b36fe7c987623a3f309393\nAuthor: ZhangXiaoxu \u003czhangxiaoxu5@huawei.com\u003e\nDate: Thu Aug 22 23:00:32 2019 -0400\n\n ext4: treat buffers with write errors as containing valid data\n \n I got some errors when I repair an ext4 volume which stacked by an\n iscsi target:\n Entry \u0027test60\u0027 in / (2) has deleted/unused inode 73750. Clear?\n It can be reproduced when the network not good enough.\n \n When I debug this I found ext4 will read entry buffer from disk and\n the buffer is marked with write_io_error.\n \n If the buffer is marked with write_io_error, it means it already\n wroten to journal, and not checked out to disk. IOW, the journal\n is newer than the data in disk.\n If this journal record \u0027delete test60\u0027, it means the \u0027test60\u0027 still\n on the disk metadata.\n \n In this case, if we read the buffer from disk successfully and create\n file continue, the new journal record will overwrite the journal\n which record \u0027delete test60\u0027, then the entry corruptioned.\n \n So, use the buffer rather than read from disk if the buffer is marked\n with write_io_error.\n \n Signed-off-by: Zhang Xiaoxu \u003czhangxiaoxu5@huawei.com\u003e\n Signed-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n\ncommit a1af2afbd244089560794c260b2d4326a86e39b6\nAuthor: Mark Menzynski \u003cmmenzyns@redhat.com\u003e\nDate: Fri Aug 2 11:21:00 2019 +0200\n\n drm/nouveau/volt: Fix for some cards having 0 maximum voltage\n \n Some, mostly Fermi, vbioses appear to have zero max voltage. That causes Nouveau to not parse voltage entries, thus users not being able to set higher clocks.\n \n When changing this value Nvidia driver still appeared to ignore it, and I wasn\u0027t able to find out why, thus the code is ignoring the value if it is zero.\n \n CC: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Signed-off-by: Mark Menzynski \u003cmmenzyns@redhat.com\u003e\n Reviewed-by: Karol Herbst \u003ckherbst@redhat.com\u003e\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit ed22eb56f2bf2908fef7a108809ffa208ff62c0c\nAuthor: Lyude Paul \u003clyude@redhat.com\u003e\nDate: Wed Aug 7 19:47:06 2019 -0400\n\n drm/nouveau/dispnv50: Fix runtime PM ref tracking for non-blocking modesets\n \n This is something that got noticed a while ago back when I was fixing a\n large number of runtime PM related issues in nouveau, but never got\n fixed:\n \n https://patchwork.freedesktop.org/series/46815/#rev7\n \n It\u0027s not safe to iterate the entire list of CRTCs in\n nv50_disp_atomic_commit(), as we could be doing a non-blocking modeset\n on one CRTC in parallel with one or more other CRTCs. Likewise, this\n means it\u0027s also not safe to do so in order to track runtime PM state.\n While this code is certainly wrong, so far the only issues I\u0027ve seen\n this cause in the wild is the occasional PM ref unbalance after an\n atomic check failure + module reloading (since the PCI device will\n outlive nouveau in such scenarios).\n \n So, do this far more elegantly: grab a runtime PM ref across the modeset\n and commit tail, then grab/put references for each CRTC enable/disable.\n This also ends up being much simpler then the previous broken solution\n we had.\n \n Finally, since we\u0027ve removed all it\u0027s users: get rid of\n nouveau_drm-\u003ehave_disp_power_ref.\n \n Signed-off-by: Lyude Paul \u003clyude@redhat.com\u003e\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 2b7e7bb16812cd3a8f85f1d8e2752ab66130a0fe\nAuthor: Lyude Paul \u003clyude@redhat.com\u003e\nDate: Wed Aug 7 19:47:05 2019 -0400\n\n drm/nouveau/dispnv04: Remove runtime PM\n \n Originally when trying to fix the issue of runtime PM references with\n non-blocking CRTCs on nv50, I ended up stumbling on this code when\n trying to remove nouveau_drm-\u003ehave_disp_power_ref, and attempted to fix\n it to remove the dependency on have_disp_power_ref. However, Ilia Mirkin\n pointed out that this code is actually completely useless, as pre-nv50\n never had runtime PM support in the first place! Go figure.\n \n So, since it\u0027s useless just get rid of it. Note that since the only\n thing nouveau_crtc_set_config() was doing was grabbing a runtime PM ref,\n calling drm_crtc_helper_set_config() then dropping the ref; we can just\n remove the function entirely and just call drm_crtc_helper_set_config()\n directly.\n \n Signed-off-by: Lyude Paul \u003clyude@redhat.com\u003e\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 3c978f73959c295ac737b917795926534a7aea01\nAuthor: Mark Menzynski \u003cmmenzyns@redhat.com\u003e\nDate: Thu Jul 18 10:07:41 2019 +0200\n\n drm/nouveau/gpio: check function 76 in the power check as well\n \n Added GPIO is \"Power Alert\". It\u0027s uncertain if this\n GPIO is set on GPU initialization or only if a change is detected by the\n GPU at runtime.\n \n This GPIO can be found on Tesla and sometimes on Fermi GPUs.\n \n Untested, wrote according to documentation.\n \n Signed-off-by: Mark Menzynski \u003cmmenzyns@redhat.com\u003e\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 940794b3dd343ba86a4c50f7e6db285c0a9c8776\nAuthor: Mark Menzynski \u003cmmenzyns@redhat.com\u003e\nDate: Thu Jul 18 10:07:40 2019 +0200\n\n drm/nouveau/gpio: check the gpio function 16 in the power check as well\n \n Added GPIO is \"Thermal and External Power Detect\". It\u0027s uncertain if this\n GPIO is set on GPU initialization or only if a change is detected by the\n GPU at runtime.\n \n This GPIO can be found in Rankine and Curie and rarely on Tesla GPUs\n VBIOS.\n \n Untested, wrote according to documentation.\n \n Signed-off-by: Mark Menzynski \u003cmmenzyns@redhat.com\u003e\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 72251fac062c0b4fe98670ec9e3db3f0702c50ae\nAuthor: Mark Menzynski \u003cmmenzyns@redhat.com\u003e\nDate: Thu Jul 18 10:07:39 2019 +0200\n\n drm/nouveau/gpio: fail if gpu external power is missing\n \n Currently, nouveau doesn\u0027t check if GPU is missing power. This\n patch makes nouveau fail when this happens on latest GPUs.\n \n It checks GPIO function 121 (External Power Emergency), which\n should detect power problems on GPU initialization.\n \n This can be disabled with nouveau.config\u003dNvPowerChecks\u003d1\n \n Tested on TU104, GP106 and GF100.\n \n v3:\n * Add config override for disabling power checks\n \n Signed-off-by: Mark Menzynski \u003cmmenzyns@redhat.com\u003e\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit e79ef1c0076ad82bab6915ada633a9d0efcb73fa\nAuthor: Mark Menzynski \u003cmmenzyns@redhat.com\u003e\nDate: Thu Jul 18 10:07:38 2019 +0200\n\n drm/nouveau/bios/gpio: sort gpios by values\n \n One gpio was in wrong place, moved it for better readability.\n \n Signed-off-by: Mark Menzynski \u003cmmenzyns@redhat.com\u003e\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 69cbbb7b04ff57c17018b27a86e9c2d758d4366e\nAuthor: Ben Skeggs \u003cbskeggs@redhat.com\u003e\nDate: Tue Jul 2 14:19:12 2019 +1000\n\n drm/nouveau/therm: don\u0027t attempt fan control where PMU is already managing it\n \n There\u0027s already a condition in place which attempts to detect this, but\n since we\u0027ve begun to require a PMU subdev even on boards where we don\u0027t\n load a custom FW, it\u0027s become inaccurate.\n \n This will prevent unnecessarily running a periodic fan update thread on\n GP100 and newer, where we don\u0027t yet override the default PMU FW.\n \n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit f0790cda65a1f1a3818db16e95c04c06cb7f2421\nAuthor: Ben Skeggs \u003cbskeggs@redhat.com\u003e\nDate: Tue Jul 2 12:52:11 2019 +1000\n\n drm/nouveau/therm: skip probing for devices not specified in thermal tables\n \n Saves some time during driver load, as described by the relevant section[1]\n of the DCB 4.x specification.\n \n [1] https://nvidia.github.io/open-gpu-doc/DCB/DCB-4.x-Specification.html#_i2c_device_table\n \n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit edc641b2b2f494ce74a2f657544133faf451a80c\nAuthor: Ben Skeggs \u003cbskeggs@redhat.com\u003e\nDate: Wed Jun 12 17:37:23 2019 +1000\n\n drm/nouveau/kms/gv100-: attach pixel blend mode property to planes\n \n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit a0a76244760d8b72ba1bb9c6998e41f5d286d6aa\nAuthor: Ben Skeggs \u003cbskeggs@redhat.com\u003e\nDate: Tue Jun 11 17:13:04 2019 +1000\n\n drm/nouveau/kms/gv100-: attach alpha property to planes\n \n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 9a99e904cc5b08f8eda2366135404fe72dae16af\nAuthor: Ben Skeggs \u003cbskeggs@redhat.com\u003e\nDate: Tue Jun 11 16:46:13 2019 +1000\n\n drm/nouveau/kms/gv100-: add support for plane zpos property\n \n Has a nice side-effect that we only update HW for this when it changes now,\n rather than every time we do a page flip.\n \n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 7a962f2b7276ccfb844583f0db5680e763d6f6da\nAuthor: Ben Skeggs \u003cbskeggs@redhat.com\u003e\nDate: Tue Jun 11 16:40:31 2019 +1000\n\n drm/nouveau/kms/nv50-: attach immutable zpos property to planes\n \n Defaulting to the fixed layout enforced in HW by EVO, and that we\n currently use by default on NVD.\n \n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 6f78991fbcfe44c762fe783812f08500a8411ea9\nAuthor: Ben Skeggs \u003cbskeggs@redhat.com\u003e\nDate: Tue Jun 11 17:03:21 2019 +1000\n\n drm/nouveau/kms/nv50-: create primary plane before overlay planes\n \n zpos normalisation uses plane id to determine ordering for duplicate zpos\n values, and we likely want to keep primary plane on the bottom here.\n \n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit a4ec848910a32c0a1e35504088d579c6e2a017f4\nAuthor: Ben Skeggs \u003cbskeggs@redhat.com\u003e\nDate: Tue Jun 11 18:04:42 2019 +1000\n\n drm/nouveau/kms/nv50-: use __drm_atomic_helper_plane_reset()\n \n We have some of this open-coded already, use the helper to prevent problems\n when adding (for example) support for the alpha property.\n \n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit dffa487822d1ff8f44817124d8f2927057a57775\nAuthor: Ben Skeggs \u003cbskeggs@redhat.com\u003e\nDate: Tue Jun 11 17:46:39 2019 +1000\n\n drm/nouveau/kms/gv100-: implement csc + enable modern colour managment properties\n \n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 2c7442ea17f32a9d6eb8c6672478e7a6fafd4631\nAuthor: Ben Skeggs \u003cbskeggs@redhat.com\u003e\nDate: Tue Jun 11 14:54:32 2019 +1000\n\n drm/nouveau/kms/gv100-: use premultiplied alpha blending between planes\n \n This is apparently the assumed default behaviour when blend properties\n are absent.\n \n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit af364a447b4ac34c719751fdec5dbdd74d888c3b\nAuthor: Ilia Mirkin \u003cimirkin@alum.mit.edu\u003e\nDate: Mon Jun 10 22:32:45 2019 -0400\n\n drm/nouveau/kms/nv50-: enable modern color management properties\n \n For GF119:GV100, we can enable DEGAMMA/CTM/GAMMA. For earlier GPUs, as\n there is no CTM, having both degamma and gamma is a bit pointless. Later\n GPUs currently lack an implementation.\n \n Signed-off-by: Ilia Mirkin \u003cimirkin@alum.mit.edu\u003e\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 88b703527ba70659365d989f29579f1292ebf9c3\nAuthor: Ilia Mirkin \u003cimirkin@alum.mit.edu\u003e\nDate: Tue Jun 11 22:40:36 2019 -0400\n\n drm/nouveau/kms/gf119-: add ctm property support\n \n This adds support on GF119:GV100 (exclusive) for CTM (aka CSC).\n \n Signed-off-by: Ilia Mirkin \u003cimirkin@alum.mit.edu\u003e\n\ncommit 7c844e9d95fb210b40398516d3d7525e8fa38a5f\nAuthor: Ilia Mirkin \u003cimirkin@alum.mit.edu\u003e\nDate: Mon Jun 3 01:59:42 2019 -0400\n\n drm/nouveau/kms/nv50-: remove overlay alpha formats\n \n The overlay logic can only do colorkey-based selection, not\n alpha-blending.\n \n Signed-off-by: Ilia Mirkin \u003cimirkin@alum.mit.edu\u003e\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit cde540211ab1346a262ad9a5ecc52c9ff8f5975f\nAuthor: Ben Skeggs \u003cbskeggs@redhat.com\u003e\nDate: Thu Jun 13 14:23:04 2019 +1000\n\n drm/nouveau/fifo/gk104-: fix parsing of mmu fault data\n \n Pascal was particularly incorrect, as the register changed to be more in the\n same format as the MMU fault buffers are.\n \n Shouldn\u0027t have impacted much more than confusing MMU fault log messages.\n \n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit cf9518b50a9c68cafb07e05fc54731071228ced3\nAuthor: Ben Skeggs \u003cbskeggs@redhat.com\u003e\nDate: Thu Jun 13 13:58:50 2019 +1000\n\n drm/nouveau/fifo/gf1xx: convert to using nvkm_fault_data\n \n Would like to be able to reuse gf100_fifo_intr_fault() for (some of) the\n later chipsets too, as it\u0027s identical.\n \n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 9f9b450752d38c86f4f830214bb9276ed174d5d3\nAuthor: Rhys Kidd \u003crhyskidd@gmail.com\u003e\nDate: Mon Jun 3 00:13:15 2019 +1000\n\n drm/nouveau/bios/init: handle INIT_RESET_END devinit opcode\n \n Signal that the reset sequence has completed.\n \n This opcode signals that the software reset sequence has completed.\n Ordinarily, no actual operations are performed by the opcode.\n However it allows for possible software work arounds by devinit\n engines in software agents other than the VBIOS, such as the resman,\n FCODE, and EFI driver.\n \n Signed-off-by: Rhys Kidd \u003crhyskidd@gmail.com\u003e\n Reviewed-by: Lyude Paul \u003clyude@redhat.com\u003e\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 66cbcc72ae34711854ae7af8056bfb7169f874fd\nAuthor: Rhys Kidd \u003crhyskidd@gmail.com\u003e\nDate: Mon Jun 3 00:13:14 2019 +1000\n\n drm/nouveau/bios/init: handle INIT_RESET_BEGUN devinit opcode\n \n Signal that the reset sequence has begun.\n \n This opcode signals that the software reset sequence has begun.\n Ordinarily, no actual operations are performed by the opcode.\n However it allows for possible software work arounds by devinit\n engines in software agents other than the VBIOS, such as the resman,\n FCODE, and EFI driver.\n \n Signed-off-by: Rhys Kidd \u003crhyskidd@gmail.com\u003e\n Reviewed-by: Lyude Paul \u003clyude@redhat.com\u003e\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 607db661b7ac338943bb5efd94a7eac0051af440\nAuthor: Rhys Kidd \u003crhyskidd@gmail.com\u003e\nDate: Sun Jun 2 22:07:27 2019 +1000\n\n drm/nouveau/bios: downgrade absence of tmds table to info from an error\n \n Absence of a TMDS Info Table is common on Optimus setups where the NVIDIA\n gpu is not connected directly to any outputs.\n \n Reporting an error in this scenario is too harsh. Accordingly, change the\n error message to an info message.\n \n By default the error message also causes a boot flicker for these sytems.\n \n Signed-off-by: Rhys Kidd \u003crhyskidd@gmail.com\u003e\n Reviewed-by: Lyude Paul \u003clyude@redhat.com\u003e\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit ad45354a631bb05a53e535ac28b608fa6bd85561\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Thu Sep 13 19:31:46 2018 +0300\n\n drm/nouveau: Disable atomic support on a per-device basis\n \n We now have per-device driver_features, so let\u0027s use that\n to disable atomic only for pre-nv50.\n \n Cc: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n Cc: Lyude Paul \u003clyude@redhat.com\u003e\n Cc: nouveau@lists.freedesktop.org\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Suggested-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Reviewed-by: Lyude Paul \u003clyude@redhat.com\u003e\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 38a72243235ecf2c1359ce66ebed29a7dfb680f7\nAuthor: Ilia Mirkin \u003cimirkin@alum.mit.edu\u003e\nDate: Mon May 27 22:58:37 2019 -0400\n\n drm/nouveau/kms/nv50-: add fp16 scanout support\n \n Older hardware seems to want 0..1024 values, while new hardware takes\n 0..1 values. We set the gain to 1024 for the earlier display classes.\n \n Signed-off-by: Ilia Mirkin \u003cimirkin@alum.mit.edu\u003e\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit ebf8ca6b3d6d7310646b092f3d5219abe2858e81\nAuthor: Ben Skeggs \u003cbskeggs@redhat.com\u003e\nDate: Wed May 29 16:39:53 2019 +1000\n\n drm/nouveau/kms/nv50-: disable input lut harder\n \n Under some circumstances, it could be left enabled when it shouldn\u0027t be.\n \n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 1e339ab2ac3c769c1b06b9fb7d532f8495ebc56d\nAuthor: Ben Skeggs \u003cbskeggs@redhat.com\u003e\nDate: Wed May 29 15:44:57 2019 +1000\n\n drm/nouveau/kms/tu102-: disable input lut when input is already FP16\n \n On Turing, an input LUT is required to transform inputs in fixed-point\n formats to FP16 for the internal display pipe. We provide an identity\n mapping whenever a window is enabled for this reason.\n \n HW has error checks to ensure when the input is already FP16, that the\n input LUT is also disabled.\n \n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 0446f48cb53d0fa2e122504ce235fcb25b1e2a19\nAuthor: Ben Skeggs \u003cbskeggs@redhat.com\u003e\nDate: Tue May 28 16:33:59 2019 +1000\n\n drm/nouveau/kms/gv100: allow windows to use PACKED8BPP formats\n \n Required for upcoming FP16 scanout support.\n \n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit eb3f4cfa53bca7c17feefa926a365e80844198d6\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Sun May 26 16:36:25 2019 +0530\n\n drm/nouveau: fix nvif/device.h is included more than once\n \n remove duplicate inclusion of nvif/device.h\n \n Issue identified by includecheck\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit c8b3585d4fbdf48b692c14b41dd1d53c329f7592\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Sun May 26 16:29:36 2019 +0530\n\n drm/nouveau/dispnv04: subdev/bios.h is included more than once\n \n remove duplicate inclusion of subdev/bios.h\n \n Issue identified by includecheck\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 7b97492555b1063417c143d2c22c503f6299c0b0\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Fri May 24 12:15:36 2019 -0500\n\n drm/nouveau/mmu: use struct_size() helper\n \n Make use of the struct_size() helper instead of an open-coded version\n in order to avoid any potential type mistakes, in particular in the\n context in which this code is being used.\n \n So, replace the following form:\n \n sizeof(*kind) + sizeof(*kind-\u003edata) * mmu-\u003ekind_nr;\n \n with:\n \n struct_size(kind, data, mmu-\u003ekind_nr)\n \n This code was detected with the help of Coccinelle.\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 690ae20c0426f8a6f48d2c285a53c465ebcb0c1f\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun May 19 16:00:44 2019 +0200\n\n drm/nouveau: drop use of drmp.h\n \n Drop use of the deprecated drmP.h file from drm/nouveau.\n \n Build tested using allyesconfig and allmodconfig.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n Cc: nouveau@lists.freedesktop.org\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 3a2b92725bc846ee5e4d25cbe7c0dcac61083632\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun May 19 16:00:43 2019 +0200\n\n drm/nouveau: drop drmP.h from all header files\n \n Drop include of the deprecated drmP.h from all nouveau heder files.\n This allows us to remove drmP.h from all .c files without any\n side-effects in a follow-up commit.\n \n Build tested using allyeyconfig and allmodconfig\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n Cc: nouveau@lists.freedesktop.org\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit c587cd6fc8b3ed497d1b478ba789311b3156d560\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun May 19 16:00:42 2019 +0200\n\n drm/nouveau: drop drmP.h from nouveau_drv.h\n \n Drop the deprecated drmP.h header from nouveau_drv.h.\n Fix fallout in other parts of the driver.\n \n Build tested using allmodconfig and allyesconfig.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n Cc: nouveau@lists.freedesktop.org\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 977b7e81df7300b5e79d680ae4e02be9fbab0fc2\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun May 19 16:00:41 2019 +0200\n\n drm/nouveau: drop use of DRM_UDELAY\n \n The DRM_UDELAY is a simple wrapper for udealy() and to be consistent\n call udelay() direct like in may other places.\n This avoids the need to pull in drm_os_linux.h when we later\n drop drmP.h uses in nouveau.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n Cc: nouveau@lists.freedesktop.org\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit fcca420d66046d5ddac55a9ddecb0e387547f7a0\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Tue May 14 21:57:01 2019 +0100\n\n drm/nouveau/bios/init: fix spelling mistake \"CONDITON\" -\u003e \"CONDITION\"\n \n There is a spelling mistake in a warning message. Fix it.\n \n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Reviewed-by: Mukesh Ojha \u003cmojha@codeaurora.org\u003e\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit 863a6670a4f1901d1c9088f357337f3c8a17ab35\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Jul 17 14:56:26 2019 +0800\n\n drm/nouveau/secboot: Make acr_r352_ls_gpccs_func static\n \n Fix sparse warning:\n \n drivers/gpu/drm/nouveau/nvkm/subdev/secboot/acr_r352.c:1092:1:\n warning: symbol \u0027acr_r352_ls_gpccs_func\u0027 was not declared. Should it be static?\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n\ncommit e3d550c2c4f2f3dba469bc3c4b83d9332b4e99e1\nAuthor: Rakesh Pandit \u003crakesh@tuxera.com\u003e\nDate: Thu Aug 22 22:53:46 2019 -0400\n\n ext4: fix warning inside ext4_convert_unwritten_extents_endio\n \n Really enable warning when CONFIG_EXT4_DEBUG is set and fix missing\n first argument. This was introduced in commit ff95ec22cd7f (\"ext4:\n add warning to ext4_convert_unwritten_extents_endio\") and splitting\n extents inside endio would trigger it.\n \n Fixes: ff95ec22cd7f (\"ext4: add warning to ext4_convert_unwritten_extents_endio\")\n Signed-off-by: Rakesh Pandit \u003crakesh@tuxera.com\u003e\n Signed-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Cc: stable@kernel.org\n\ncommit 6d24e14140053febc5ac1ce46baca6a4334c5f6c\nAuthor: Ben Wei \u003cbenwei@fb.com\u003e\nDate: Wed Aug 21 22:08:49 2019 +0000\n\n net/ncsi: update response packet length for GCPS/GNS/GNPTS commands\n \n Update response packet length for the following commands per NC-SI spec\n - Get Controller Packet Statistics\n - Get NC-SI Statistics\n - Get NC-SI Pass-through Statistics command\n \n Signed-off-by: Ben Wei \u003cbenwei@fb.com\u003e\n Reviewed-by: Justin Lee \u003cjustin.lee1@dell.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 0a9a4b611a033f625df02a84575a72c0048d5e5e\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Mon Aug 19 11:12:04 2019 -0500\n\n drm/panfrost: Fix sleeping while atomic in panfrost_gem_open\n \n We can\u0027t hold the mm_lock spinlock as panfrost_mmu_map() can sleep:\n \n BUG: sleeping function called from invalid context at kernel/locking/mutex.c:909\n in_atomic(): 1, irqs_disabled(): 0, pid: 974, name: glmark2-es2-drm\n 1 lock held by glmark2-es2-drm/974:\n CPU: 5 PID: 974 Comm: glmark2-es2-drm Tainted: G W L 5.3.0-rc1+ #94\n Hardware name: 96boards Rock960 (DT)\n Call trace:\n dump_backtrace+0x0/0x130\n show_stack+0x14/0x20\n dump_stack+0xc4/0x10c\n ___might_sleep+0x158/0x228\n __might_sleep+0x50/0x88\n __mutex_lock+0x58/0x800\n mutex_lock_interruptible_nested+0x1c/0x28\n drm_gem_shmem_get_pages+0x24/0xa0\n drm_gem_shmem_get_pages_sgt+0x48/0xd0\n panfrost_mmu_map+0x38/0xf8 [panfrost]\n panfrost_gem_open+0xc0/0xd8 [panfrost]\n drm_gem_handle_create_tail+0xe8/0x198\n drm_gem_handle_create+0x3c/0x50\n panfrost_gem_create_with_handle+0x70/0xa0 [panfrost]\n panfrost_ioctl_create_bo+0x48/0x80 [panfrost]\n drm_ioctl_kernel+0xb8/0x110\n drm_ioctl+0x244/0x3f0\n do_vfs_ioctl+0xbc/0x910\n ksys_ioctl+0x78/0xa8\n __arm64_sys_ioctl+0x1c/0x28\n el0_svc_common.constprop.0+0x90/0x168\n el0_svc_handler+0x28/0x78\n el0_svc+0x8/0xc\n \n Fixes: a5efb4c9a562 (\"drm/panfrost: Restructure the GEM object creation\")\n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190819161204.3106-5-robh@kernel.org\n\ncommit 399e06a517b6f1da5f617d413f6e5489f5054f7a\nAuthor: Markus Elfring \u003celfring@users.sourceforge.net\u003e\nDate: Thu Aug 22 20:02:56 2019 +0200\n\n ethernet: Delete unnecessary checks before the macro call “dev_kfree_skb”\n \n The dev_kfree_skb() function performs also input parameter validation.\n Thus the test around the shown calls is not needed.\n \n This issue was detected by using the Coccinelle software.\n \n Signed-off-by: Markus Elfring \u003celfring@users.sourceforge.net\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 5477fccf9abd7342a987d401fa50c0d93ebd09b9\nAuthor: Markus Elfring \u003celfring@users.sourceforge.net\u003e\nDate: Wed Aug 21 22:16:02 2019 +0200\n\n net: usb: Delete unnecessary checks before the macro call “dev_kfree_skb”\n \n The dev_kfree_skb() function performs also input parameter validation.\n Thus the test around the shown calls is not needed.\n \n This issue was detected by using the Coccinelle software.\n \n Signed-off-by: Markus Elfring \u003celfring@users.sourceforge.net\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b7deac31979bd09c69e0e6e064609fad55df35be\nAuthor: Markus Elfring \u003celfring@users.sourceforge.net\u003e\nDate: Wed Aug 21 21:48:46 2019 +0200\n\n hamradio: Delete unnecessary checks before the macro call “dev_kfree_skb”\n \n The dev_kfree_skb() function performs also input parameter validation.\n Thus the test around the shown calls is not needed.\n \n This issue was detected by using the Coccinelle software.\n \n Signed-off-by: Markus Elfring \u003celfring@users.sourceforge.net\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 038dab7efc38254e151adbbade13b8860055d44a\nAuthor: Markus Elfring \u003celfring@users.sourceforge.net\u003e\nDate: Wed Aug 21 21:16:15 2019 +0200\n\n can: Delete unnecessary checks before the macro call “dev_kfree_skb”\n \n The dev_kfree_skb() function performs also input parameter validation.\n Thus the test around the shown calls is not needed.\n \n This issue was detected by using the Coccinelle software.\n \n Signed-off-by: Markus Elfring \u003celfring@users.sourceforge.net\u003e\n Acked-by: Sean Nyekjaer \u003csean@geanix.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 0f817a5eb9e33cec5baf9dc265bdf240b03cf51b\nAuthor: Markus Elfring \u003celfring@users.sourceforge.net\u003e\nDate: Wed Aug 21 20:10:56 2019 +0200\n\n mISDN: Delete unnecessary checks before the macro call “dev_kfree_skb”\n \n The dev_kfree_skb() function performs also input parameter validation.\n Thus the test around the shown calls is not needed.\n \n This issue was detected by using the Coccinelle software.\n \n Signed-off-by: Markus Elfring \u003celfring@users.sourceforge.net\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d3ee8ec7de83abf4325c11bcd36cbd01d7a66789\nAuthor: Marco Hartmann \u003cmarco.hartmann@nxp.com\u003e\nDate: Wed Aug 21 11:43:49 2019 +0000\n\n net: fec: add C45 MDIO read/write support\n \n IEEE 802.3ae clause 45 defines a modified MDIO protocol that uses a two\n staged access model in order to increase the address space.\n \n This patch adds support for C45 MDIO read and write accesses, which are\n used whenever the MII_ADDR_C45 flag in the regnum argument is set.\n In case it is not set, C22 accesses are used as before.\n \n Signed-off-by: Marco Hartmann \u003cmarco.hartmann@nxp.com\u003e\n Acked-by: Fugang Duan \u003cfugang.duan@nxp.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 789d027ec8e7aa300689f418b522ae21e333eb17\nAuthor: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\nDate: Thu Aug 15 12:35:39 2019 -0400\n\n drm/amd/display: fix calc_pll_max_vco_construct\n \n This was broken by a previous change switching to cached fw_info.\n Fixed by inverting a valid bool check.\n \n Bug: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111432\n Fixes: 9adc8050bf3c (\"drm/amd/display: make firmware info only load once during dc_bios create\")\n Signed-off-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Reviewed-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 854009849235308d1319cf4f8ff52f00a3b07026\nAuthor: Thong Thai \u003cthong.thai@amd.com\u003e\nDate: Thu Aug 15 14:00:30 2019 -0400\n\n drm/amdgpu: enable VCN DPG for Renoir\n \n This will enable indirect SRAM loading for VCN DPG mode initialization.\n \n Signed-off-by: Thong Thai \u003cthong.thai@amd.com\u003e\n Reviewed-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Reviewed-by: Prike Liang \u003cPrike.Liang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 134b1461ea728b578468ec3ced29f5819db172b4\nAuthor: Thong Thai \u003cthong.thai@amd.com\u003e\nDate: Thu Aug 22 17:42:53 2019 -0500\n\n Revert \"drm/amdgpu: use direct loading on renoir vcn for the moment\"\n \n This reverts commit 444a0fea5107e9ad7e3cbbafed78678489e31713.\n \n We are ready to enable it now.\n \n Signed-off-by: Thong Thai \u003cthong.thai@amd.com\u003e\n Reviewed-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f13580a94715c1b783ea2f714cdbefc12b04d1c7\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Wed Aug 7 09:57:08 2019 +0800\n\n drm/amdgpu: update gc/sdma goldensetting for rn\n \n This patch updates gc/sdma goldensetting for renoir\n \n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 51b9121af0946afa4ef9fc164f4b0c1686af7225\nAuthor: Prike Liang \u003cPrike.Liang@amd.com\u003e\nDate: Fri Aug 9 14:17:40 2019 +0800\n\n drm/amd/powerplay: Disable renoir smu feature retrieve for the moment\n \n To avoid the dpm frequence range get failed when DPM enabled and it\n will be enabled later once handle well the feature bit map struct.\n \n Signed-off-by: Prike Liang \u003cPrike.Liang@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d5a57e4e31d1bba5f9603a74660c3dddb173626b\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Tue Aug 20 09:57:42 2019 +0200\n\n dt-bindings: net: meson-dwmac: convert to yaml\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for the Synopsys DWMAC Glue for Amlogic SoCs over to a YAML schemas.\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 57b77df7b71904b00ed70acdf3fde01b5a3e88e2\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Tue Aug 20 09:57:41 2019 +0200\n\n dt-bindings: net: snps, dwmac: update reg minItems maxItems\n \n The Amlogic Meson DWMAC glue bindings needs a second reg cells for the\n glue registers, thus update the reg minItems/maxItems to allow more\n than a single reg cell.\n \n Also update the allwinner,sun7i-a20-gmac.yaml derivative schema to specify\n maxItems to 1.\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Acked-by: Rob Herring \u003crobh@kernel.org\u003e\n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d237e0974b21bebcc15ad1f8e23f51538230dcf5\nAuthor: Prike Liang \u003cPrike.Liang@amd.com\u003e\nDate: Thu Jul 25 13:56:51 2019 +0800\n\n drm/amd/powerplay: enable renoir dpm feature\n \n enable the dpm feature for the renoir.\n \n Signed-off-by: Prike Liang \u003cPrike.Liang@amd.com\u003e\n Reviewed-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9a868d8bbb9e4796373220e002771aa6955df2cc\nAuthor: Prike Liang \u003cPrike.Liang@amd.com\u003e\nDate: Fri Aug 2 15:42:41 2019 +0800\n\n drm/amdgpu: enable SDMA power gating for rn\n \n Enable SDMA PG flag during device ip early init.\n \n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Prike Liang \u003cPrike.Liang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 91c5b6b3261b664296a5608830830c1fbfb63d21\nAuthor: Prike Liang \u003cPrike.Liang@amd.com\u003e\nDate: Mon Aug 12 15:25:18 2019 -0500\n\n drm/amdgpu/sdma4: set sdma clock gating for rn\n \n Add support for SDMA clockgating on RN.\n \n Signed-off-by: Prike Liang \u003cPrike.Liang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2f47d6492b4c9ff05fb0d4216b6c9b987839d771\nAuthor: Prike Liang \u003cPrike.Liang@amd.com\u003e\nDate: Mon Jul 29 15:13:42 2019 +0800\n\n drm/amdgpu/mmhub1: set mmhub clock gating for rn\n \n setup mmhub clockgating.\n \n Signed-off-by: Prike Liang \u003cPrike.Liang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8db63b7c38217df3dc5501ca0371a58bd912c9ea\nAuthor: Prike Liang \u003cPrike.Liang@amd.com\u003e\nDate: Fri Aug 2 15:32:57 2019 +0800\n\n drm/amdgpu: enable DF clock gating for rn\n \n Enable DF clock gating during DF IP early init.\n \n Signed-off-by: Prike Liang \u003cPrike.Liang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e2ef3b70e857ace2a52d542974326a521bc8d467\nAuthor: Prike Liang \u003cPrike.Liang@amd.com\u003e\nDate: Fri Aug 2 15:29:26 2019 +0800\n\n drm/amdgpu: enable athub clock gating for rn\n \n Enable athub MG and LS clock gating.\n \n Signed-off-by: Prike Liang \u003cPrike.Liang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 91ec8bbb88d9b3be51d64efa6a778c978ab82554\nAuthor: Prike Liang \u003cPrike.Liang@amd.com\u003e\nDate: Fri Aug 2 15:27:11 2019 +0800\n\n drm/amdgpu: enable IH clock gating for rn\n \n Enable IH clock gating during IH block initialized.\n \n Signed-off-by: Prike Liang \u003cPrike.Liang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 753c929cc798530ec3a4aeb27b2ea7b06daac525\nAuthor: Prike Liang \u003cPrike.Liang@amd.com\u003e\nDate: Fri Aug 2 15:24:45 2019 +0800\n\n drm/amdgpu: enable vcn clock gating for rn\n \n Enable VCN middle grain clock gating.\n \n Signed-off-by: Prike Liang \u003cPrike.Liang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit de273070c59385e9f6108f0be1e3225ec03eba54\nAuthor: Prike Liang \u003cPrike.Liang@amd.com\u003e\nDate: Fri Aug 2 15:21:46 2019 +0800\n\n drm/amdgpu: enable rom clock gating for rn\n \n Enable rom light sleep clock gating.\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n \n Signed-off-by: Prike Liang \u003cPrike.Liang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9deac0a415e0e4718115201435c930df88f83730\nAuthor: Prike Liang \u003cPrike.Liang@amd.com\u003e\nDate: Fri Aug 2 15:18:24 2019 +0800\n\n drm/amdgpu: enable HDP clock gating for rn\n \n Enable HDP light sleep clock gating.\n \n Signed-off-by: Prike Liang \u003cPrike.Liang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d98930f52e3160e34a36c99ac1f6f4070e9a9c06\nAuthor: Prike Liang \u003cPrike.Liang@amd.com\u003e\nDate: Fri Aug 2 15:14:54 2019 +0800\n\n drm/amdgpu: enable BIF clock gating for rn\n \n Enable BIF light sleep clock gating.\n \n Signed-off-by: Prike Liang \u003cPrike.Liang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ef0e7d08a53248afefe3759df912c9d288baa6c9\nAuthor: Prike Liang \u003cPrike.Liang@amd.com\u003e\nDate: Fri Aug 2 15:10:45 2019 +0800\n\n drm/amdgpu: enable sdma clock gating for rn\n \n Enable sdma middle grain and light sleep clock gating.\n \n Signed-off-by: Prike Liang \u003cPrike.Liang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a2d15255eac3473f09fb70b2da764153e8480b74\nAuthor: Prike Liang \u003cPrike.Liang@amd.com\u003e\nDate: Fri Aug 2 15:04:27 2019 +0800\n\n drm/amdgpu: enable mmhub clock gating for rn\n \n Enable mmhub midle grain and light sleep clock gating.\n \n Signed-off-by: Prike Liang \u003cPrike.Liang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ec3636a53a0c9f5dd2691055b673ff44ff95691d\nAuthor: Prike Liang \u003cPrike.Liang@amd.com\u003e\nDate: Thu Aug 1 16:21:07 2019 +0800\n\n drm/amdgpu: enable gfx clock gating for rn\n \n Enable gfx cg/mg/cp etc clock gating.\n \n Signed-off-by: Prike Liang \u003cPrike.Liang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 723d473537778f3f1bd9a91eb286aa5c349c8f0e\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Tue Jul 30 11:28:27 2019 +0800\n\n drm/amd/powerplay: add DPMCLOCKS table implementation\n \n This patch adds add DPMCLOCKS table implementation\n Rename smu_populate_smc_pptable to smu_populate_smc_tables\n \n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 049284bd5264563965e106f03f152a70df9342ca\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Tue Jul 30 11:04:58 2019 +0800\n\n drm/amd/powerplay: init smu tables for rn\n \n Initialize smu tables for renoir:\n WATERMARKS/DPMCLOCKS/SMU_METRICS\n \n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1405ac8f9280e15c12b76f612ac55096076f4761\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Tue Jul 30 10:50:44 2019 +0800\n\n drm/amd/powerplay: add smu tables for rn\n \n add and map smu tables for renoir\n \n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b5604512088bd096ccb5dc49b7779ed99afdeff0\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Wed Aug 21 16:45:52 2019 -0500\n\n drm/amd/powerplay: using valid mapping check for rn\n \n Check whether the message mapping is valid\n \n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a31b059b0385dae98e95d6074950c5d47de34661\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Fri Jul 26 10:51:32 2019 +0800\n\n drm/amd/powerplay: fix checking gfxoff status for rn\n \n For renoir, it should use mmSMUIO_GFX_MISC_CNTL to check\n gfxoff status. For the first time to enter gfxoff status,\n it maybe takes about one second more. So just set the max\n timeout to 5s.\n \n GFXOFF_STATUS(bits 2:1)\u0027s description is below:\n 0\u003dGFXOFF(default).\n 1\u003dTransition out of GFX State.\n 2\u003dNot in GFXOFF.\n 3\u003dTransition into GFXOFF.\n \n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9f21e9ee7f2ffc64a2f0f1f821fb22ded9d18761\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Tue Jul 16 17:33:47 2019 +0800\n\n drm/amdgpu: add and enable gfxoff feature\n \n This patch updates gfxoff feature.\n \n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1268795511e18bf7d4a39fb943caf9df6294a268\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Tue Jul 16 17:21:17 2019 +0800\n\n drm/amdgpu: add set_gfx_cgpg implement (v2)\n \n add set_gfx_cgpg implement\n \n v2: check if using sw_smu (Alex)\n \n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 0a3c84243d6c11e5e3124c5f557b26b4d7924a04\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Tue Jul 23 10:39:38 2019 +0800\n\n drm/amd/powerplay: udpate smu_v12_0_check_fw_version (v2)\n \n This interface support SMU_MSG_GetDriverIfVersion\n and SMU_MSG_GetSmuVersion checking.\n \n v2: squash in driver_if changes (Alex)\n \n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 47903831ada9d6549a5ad2db66d74166cbc211fa\nAuthor: Huang Rui \u003cray.huang@amd.com\u003e\nDate: Thu Aug 8 15:23:17 2019 -0500\n\n drm/amd/powerplay: powerup sdma/vcn for all apu series\n \n All apu series need powerup sdma and vcn via smu messages.\n \n Signed-off-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4a6296680baea5aa118a7617ef0b23606e3d6b5c\nAuthor: Leo Liu \u003cleo.liu@amd.com\u003e\nDate: Thu Aug 8 15:21:44 2019 -0500\n\n drm/amdgpu/powerplay: add Renoir VCN power management\n \n Thus VCN can be powered up for normal operations\n \n Signed-off-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 51548c0fe3d3487c277608bb6436882d0f052fae\nAuthor: Huang Rui \u003cray.huang@amd.com\u003e\nDate: Wed Jul 24 14:05:11 2019 -0500\n\n drm/amdgpu: skip dpm init for renoir\n \n Renoir DPM is not functional so far, we skip it for the comment.\n Will revert this patch once SMU 12 is functional.\n \n Signed-off-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 97222cfac7740ba93c2c17f15db5d7184c6ab836\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Fri Aug 9 10:34:40 2019 -0500\n\n drm/amdgpu/powerplay: add power up/down SDMA interfaces for renoir\n \n 1.Implement PowerUpSDMA/PowerDownSDMA interfaces in the swSMU for renoir\n 2.adjust smu ip block ahead of gfx\u0026sdma ip block\n \n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5dbbe6a77d189954287b524988b994bfda75efd6\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Wed Jul 24 14:00:01 2019 -0500\n\n drm/amdgpu/powerplay: add smu ip block for renoir (v2)\n \n add swSMU [smu_v12_0] for renoir\n \n v2: whitespace fixes (Alex)\n \n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b925e30cb2ef955170bcdd0e37b4ffec9ca98b0a\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Wed May 22 19:16:27 2019 +0800\n\n drm/amdgpu/powerplay: add smu_v12_0.c \u0026 smu_v12_0.h for renoir\n \n add smu_v12_0.c \u0026 smu_v12_0.h for renoir\n \n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9eb75d62a28c75f15f306fd7cc2772bf82d44f83\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Thu Aug 8 15:18:30 2019 -0500\n\n drm/amdgpu/powerplay: add initial renoir_ppt.c for renoir (v3)\n \n Add renoir_ppt and map ppsmc to amdgpu_smu.h\n \n v2: squash in ppsmc updates (Alex)\n v3: squash in driver_if updates (Alex)\n \n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 039be8df355d47707d926a0a9fa0814cebb187c4\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Mon Jul 15 13:53:32 2019 +0800\n\n drm/amd/powerplay: add smu12_driver_if.h (v3)\n \n This patch adds smu12_driver_if.h\n \n v2: squash in updates (Alex)\n v3: more updates (Alex)\n \n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit bc501346c9fe3bfa9e5576ba89f6397f5b3f9b0e\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Thu Jul 25 15:42:52 2019 -0500\n\n drm/amdgpu/powerplay: Add smu_v12_0_ppsmc.h (v2)\n \n This is the SMU v12 driver message interface.\n \n v2: squash in updates\n \n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a97567a06c1ab0e2736cca1515ee635dad888d37\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Aug 16 23:10:11 2019 +0100\n\n drm/amd/display: fix a potential null pointer dereference\n \n Currently the pointer init_data is dereferenced on the assignment\n of fw_info before init_data is sanity checked to see if it is null.\n Fix te potential null pointer dereference on init_data by only\n performing dereference after it is null checked.\n \n Addresses-Coverity: (\"Dereference before null check\")\n Fixes: 9adc8050bf3c (\"drm/amd/display: make firmware info only load once during dc_bios create\")\n Reviewed-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a13362c1c96d1450d59180631233844cb94a7907\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Thu Aug 22 14:09:48 2019 +0100\n\n drm/amdgpu/powerplay: remove redundant assignment to variable baco_state\n \n Variable baco_state is initialized to a value that is never read and it is\n re-assigned later. The initialization is redundant and can be removed.\n \n Addresses-Coverity: (\"Unused Value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7fd5a6fb9a75cc00479a7ad4305b5ab31ae2cd59\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat May 25 20:51:09 2019 +0800\n\n drm/amdkfd: Make deallocate_hiq_sdma_mqd static\n \n Fix sparse warning:\n \n drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_device_queue_manager.c:1846:6:\n warning: symbol \u0027deallocate_hiq_sdma_mqd\u0027 was not declared. Should it be static?\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9e48495017342c5d445b25eedd86d6fd884a6496\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Wed Aug 21 21:00:29 2019 +0800\n\n drm/amdgpu/sdma5: fix number of sdma5 trap irq types for navi1x\n \n v2: set num_types based on num_instances\n \n navi1x has 2 sdma engines but commit\n \"e7b58d03b678 drm/amdgpu: reorganize sdma v4 code to support more instances\"\n changes the max number of sdma irq types (AMDGPU_SDMA_IRQ_LAST) from 2 to 8\n which causes amdgpu_irq_gpu_reset_resume_helper() to recover irq of sdma\n engines with following logic:\n \n (enable irq for sdma0) * 1 time\n (enable irq for sdma1) * 1 time\n (disable irq for sdma1) * 6 times\n \n as a result, after gpu reset, interrupt for sdma1 is lost.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit df2f10151d64d0c7d51deccdca69ca2c9fdcb5d3\nAuthor: David Francis \u003cDavid.Francis@amd.com\u003e\nDate: Wed Jun 19 14:30:59 2019 -0400\n\n drm/amd/display: Enable SST DSC in DM\n \n In create_stream_for_sink, check for SST DP connectors\n \n Parse DSC caps to DC format, then, if DSC is supported,\n compute the config\n \n DSC hardware will be programmed by dc_commit_state\n \n Tested-by: Mikita Lipski \u003cMikita.Lipski@amd.com\u003e\n Signed-off-by: David Francis \u003cDavid.Francis@amd.com\u003e\n Reviewed-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5cb99d46f7bdf02cacf08ea9d4301aff0cd311a6\nAuthor: David Francis \u003cDavid.Francis@amd.com\u003e\nDate: Thu May 23 14:25:35 2019 -0400\n\n Revert \"drm/amd/display: Fix underscan not using proper scaling\"\n \n This reverts commit 80e80ec817f161560b4159608fb41bd289abede3.\n \n This commit fixed an issue with underscan commits not updating all\n needed timing values, but through various refactors it is no longer\n necessary. It causes corruption on odm combine by\n overwriting the halved h_active in the stream timing\n \n Signed-off-by: David Francis \u003cDavid.Francis@amd.com\u003e\n Reviewed-by: Roman Li \u003cRoman.Li@amd.com\u003e\n Reviewed-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Reviewed-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4c3cfe14c04eac5e96f1aef7e2b84ce56a61e897\nAuthor: David Francis \u003cDavid.Francis@amd.com\u003e\nDate: Thu May 23 14:06:08 2019 -0400\n\n Revert \"drm/amd/display: add global master update lock for DCN2\"\n \n This reverts commit 6c5be4ac630805d3a3b20157a0c6421ef815fe78.\n \n This commit was accidentally promoted twice\n \n Signed-off-by: David Francis \u003cDavid.Francis@amd.com\u003e\n Reviewed-by: Roman Li \u003cRoman.Li@amd.com\u003e\n Reviewed-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Reviewed-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b5d055de2bf566786243f0f3f1e838e901362478\nAuthor: David Francis \u003cDavid.Francis@amd.com\u003e\nDate: Thu Mar 28 09:58:11 2019 -0400\n\n Revert \"drm/amd/display: navi10 bring up skip dsc encoder config\"\n \n This reverts commit ec8763486dd3d2fb585c2e6747520406e7dcef98.\n \n Re-enable enc2_dp_set_dsc_config. This function caused warnings\n due to missing register definitions. With the registers added,\n this now works\n \n Signed-off-by: David Francis \u003cDavid.Francis@amd.com\u003e\n Reviewed-by: Roman Li \u003cRoman.Li@amd.com\u003e\n Reviewed-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Reviewed-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit aadeb6081b6e7ecbf6028b006b50f37c54a4e4fd\nAuthor: David Francis \u003cDavid.Francis@amd.com\u003e\nDate: Thu Mar 28 13:52:00 2019 -0400\n\n Revert \"drm/amd/display: skip dsc config for navi10 bring up\"\n \n This reverts commit 9e14d4f17e23ce46d346a6a22a295b4a65b9d918.\n \n optc dsc config was causing warnings due to missing register\n definitions. With the registers restored, the function can\n be re-enabled\n \n The reverted commit also disabled sanity checks and dsc\n power gating. The sanity check warnings are not associated\n with dsc, and power gating on dsc still has an issue on\n non-dsc monitors where the dsc hardware block is never init\n and so cannot respond to power gating requests. Therefore,\n those are left as is\n \n Signed-off-by: David Francis \u003cDavid.Francis@amd.com\u003e\n Reviewed-by: Roman Li \u003cRoman.Li@amd.com\u003e\n Reviewed-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Reviewed-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 75e1cafde153dd52b660f41bf4866fa22f98da9b\nAuthor: Christian König \u003cchristian.koenig@amd.com\u003e\nDate: Fri Aug 16 14:56:35 2019 +0200\n\n drm/amdgpu: fix dma_fence_wait without reference\n \n We need to grab a reference to the fence we wait for.\n \n Signed-off-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Chunming Zhou \u003cdavid1.zhou@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 32e40ffbced3b14ceac1ae13a1a66c5849a6d2d3\nAuthor: Kent Russell \u003ckent.russell@amd.com\u003e\nDate: Thu Aug 22 08:17:40 2019 -0400\n\n drm/powerplay: Fix Vega20 Average Power value v4\n \n The SMU changed reading from CurrSocketPower to AverageSocketPower, so\n reflect this accordingly. This fixes the issue where Average Power\n Consumption was being reported as 0 from SMU 40.46-onward\n \n v2: Fixed headline prefix\n v3: Add check for SMU version for proper compatibility\n v4: Style fix\n \n Signed-off-by: Kent Russell \u003ckent.russell@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ea207b29ae7718a8717de6ea784e99d41826e642\nAuthor: Frank.Min \u003cFrank.Min@amd.com\u003e\nDate: Thu Aug 22 17:16:36 2019 -0500\n\n amd/amdgpu: add Arcturus vf DID support\n \n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Frank.Min \u003cFrank.Min@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9d4f837aa0cb480616dc141a8b0aebc8e8c60a08\nAuthor: Frank.Min \u003cFrank.Min@amd.com\u003e\nDate: Wed Aug 21 11:05:04 2019 +0800\n\n drm/amdgpu: unity mc base address for arcturus\n \n arcturus for sriov would use the unified mc base address\n \n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Frank.Min \u003cFrank.Min@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 81c274c473a25e7e86540cdeaa181421e7daafff\nAuthor: Frank.Min \u003cFrank.Min@amd.com\u003e\nDate: Wed Aug 21 17:20:34 2019 +0800\n\n drm/amdgpu: disable agp for sriov\n \n Since agp is not used for sriov, just disable it\n \n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Frank.Min \u003cFrank.Min@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ac8bd58d7f4460e11d5e2870f8577fc708f451e6\nAuthor: Yabin Cui \u003cyabinc@google.com\u003e\nDate: Thu Aug 15 11:46:28 2019 -0700\n\n coresight: tmc-etr: Fix perf_data check\n \n When tracing etm data of multiple threads on multiple cpus through\n perf interface, each cpu has a unique etr_perf_buffer while sharing\n the same etr device. There is no guarantee that the last cpu starts\n etm tracing also stops last. This makes perf_data check fail.\n \n Fix it by checking etr_buf instead of etr_perf_buffer.\n Also move the code setting and clearing perf_buf to more suitable\n places.\n \n Fixes: 3147da92a8a8 (\"coresight: tmc-etr: Allocate and free ETR memory buffers for CPU-wide scenarios\")\n Signed-off-by: Yabin Cui \u003cyabinc@google.com\u003e\n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n\ncommit 3b420aeb7525db8e19ee48fdb1a2fc1d430765d5\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Mon Aug 19 10:41:37 2019 +0100\n\n misc: xilinx_sdfec: fix spelling mistake: \"Schdule\" -\u003e \"Schedule\"\n \n There is a spelling mistake in a dev_dbg message, fix it.\n \n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Reviewed-by: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\n Link: https://lore.kernel.org/r/20190819094137.390-1-colin.king@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 6123f1fe53985645992b2ff648b3087b77b3ed16\nAuthor: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nDate: Wed Aug 21 10:11:22 2019 +0300\n\n misc: xilinx_sdfec: Prevent integer overflow in xsdfec_table_write()\n \n The checking here needs to handle integer overflows because \"offset\" and\n \"len\" come from the user.\n \n Fixes: 20ec628e8007 (\"misc: xilinx_sdfec: Add ability to configure LDPC\")\n Signed-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Reviewed-by: Michal Simek \u003cmichal.simek@xilinx.com\u003e\n Reviewed-by: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\n Link: https://lore.kernel.org/r/20190821071122.GD26957@mwanda\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 56a635c0ec14950bd6a5bfb4d9d497897f64179f\nAuthor: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nDate: Wed Aug 21 10:09:53 2019 +0300\n\n misc: xilinx_sdfec: Prevent a divide by zero in xsdfec_reg0_write()\n \n The \"psize\" value comes from the user so we need to verify that it\u0027s\n non-zero before we check if \"n % psize\" or it will crash.\n \n Fixes: 20ec628e8007 (\"misc: xilinx_sdfec: Add ability to configure LDPC\")\n Signed-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Reviewed-by: Michal Simek \u003cmichal.simek@xilinx.com\u003e\n Link: https://lore.kernel.org/r/20190821070953.GC26957@mwanda\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 129c3b082c636087a72a30b17c87c57d197f08f5\nAuthor: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nDate: Thu Aug 22 11:31:05 2019 +0300\n\n misc: xilinx_sdfec: Return -EFAULT if copy_from_user() fails\n \n The copy_from_user() function returns the number of bytes remaining to\n be copied but we want to return -EFAULT to the user.\n \n Fixes: 20ec628e8007 (\"misc: xilinx_sdfec: Add ability to configure LDPC\")\n Signed-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Reviewed-by: Michal Simek \u003cmichal.simek@xilinx.com\u003e\n Reviewed-by: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\n Link: https://lore.kernel.org/r/20190822083105.GI3964@kadam\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit dac4f1964a9cbfbf81df8d31caa04ced0d7613b5\nAuthor: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nDate: Wed Aug 21 10:06:06 2019 +0300\n\n misc: xilinx_sdfec: Fix a couple small information leaks\n \n These structs have holes in them so we end up disclosing a few bytes of\n uninitialized stack data.\n \n drivers/misc/xilinx_sdfec.c:305 xsdfec_get_status() warn: check that \u0027status\u0027 doesn\u0027t leak information (struct has a hole after \u0027activity\u0027)\n drivers/misc/xilinx_sdfec.c:449 xsdfec_get_turbo() warn: check that \u0027turbo_params\u0027 doesn\u0027t leak information (struct has a hole after \u0027scale\u0027)\n \n We need to zero out the holes with memset().\n \n Fixes: 6bd6a690c2e7 (\"misc: xilinx_sdfec: Add stats \u0026 status ioctls\")\n Signed-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Reviewed-by: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\n Reviewed-by: Michal Simek \u003cmichal.simek@xilinx.com\u003e\n Link: https://lore.kernel.org/r/20190821070606.GA26957@mwanda\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 826c7facbc605af3c22e8e70cd01ee2c25be774f\nAuthor: Kevin Hilman \u003ckhilman@baylibre.com\u003e\nDate: Wed Aug 7 20:54:02 2019 -0700\n\n arm64: dts: meson: g12a-common: add VRTC\n \n Tested-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit c5f9f5db027e13f617d2481de496d7e20e47b039\nAuthor: Mao Wenan \u003cmaowenan@huawei.com\u003e\nDate: Tue Aug 20 20:40:15 2019 +0800\n\n tty: serial: add dependence for CONFIG_SERIAL_FSL_LINFLEXUART\n \n When CONFIG_SERIAL_FSL_LINFLEXUART\u003dy and CONFIG_PRINTK is not set,\n one compilation error is found as below:\n drivers/tty/serial/fsl_linflexuart.c: In function linflex_earlycon_putchar:\n drivers/tty/serial/fsl_linflexuart.c:608:31: error: CONFIG_LOG_BUF_SHIFT undeclared\n (first use in this function); did you mean CONFIG_ISA_BUS_API?\n if (earlycon_buf.len \u003e\u003d 1 \u003c\u003c CONFIG_LOG_BUF_SHIFT)\n ^~~~~~~~~~~~~~~~~~~~\n CONFIG_ISA_BUS_API\n This because CONFIG_LOG_BUF_SHIFT is depended on CONFIG_PRINTK, fix this\n by adding dependence for CONFIG_SERIAL_FSL_LINFLEXUART.\n \n Fixes: b953815b819b (\"tty: serial: Add linflexuart driver for S32V234\")\n Signed-off-by: Mao Wenan \u003cmaowenan@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190820124015.28409-1-maowenan@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b88d7260d786e54ca3f224e21157b774941a7e3a\nAuthor: Martin Hundebøll \u003cmartin@geanix.com\u003e\nDate: Mon Aug 12 23:12:43 2019 +0200\n\n tty: n_gsm: add ioctl to map serial device to mux\u0027ed tty\n \n Guessing the first tty for a gsm0710 multiplexed serial device is not\n currently possible, which makes it racy to use with multiple modems.\n \n Add a way to map the physical serial tty to its related mux devices\n using an ioctl.\n \n Signed-off-by: Martin Hundebøll \u003cmartin@geanix.com\u003e\n Link: https://lore.kernel.org/r/20190812211243.98686-1-martin@geanix.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d4c61c4a16decd8ace8660f22c81609a539fccba\nAuthor: Manasi Navare \u003cmanasi.d.navare@intel.com\u003e\nDate: Wed Aug 21 14:59:50 2019 -0700\n\n drm/i915/dp: Fix DSC enable code to use cpu_transcoder instead of encoder-\u003etype\n \n This patch fixes the intel_configure_pps_for_dsc_encoder() function to use\n cpu_transcoder instead of encoder-\u003etype to select the correct DSC registers\n that was wrongly used in the original patch for one DSC register isntance.\n \n Fixes: 7182414e2530 (\"drm/i915/dp: Configure i915 Picture parameter Set registers during DSC enabling\")\n Cc: Ville Syrjala \u003cville.syrjala@linux.intel.com\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@linux.intel.com\u003e\n Cc: Ville Syrjala \u003cville.syrjala@linux.intel.com\u003e\n Cc: \u003cstable@vger.kernel.org\u003e # v5.0+\n Signed-off-by: Manasi Navare \u003cmanasi.d.navare@intel.com\u003e\n Reviewed-by: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190821215950.24223-1-manasi.d.navare@intel.com\n\ncommit 53e5a3be73ab7773677f69ec5296185edcfe9415\nAuthor: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\nDate: Sat Aug 17 00:51:24 2019 +0800\n\n serial: 8250_pci: Merge 8250_moxa to 8250_pci\n \n Moxa serial boards only need a special setup function, we can use\n generic 8250 framework for other parts.\n \n So let\u0027s merge 8250_moxa to 8250_pci.\n \n Signed-off-by: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190816165124.16942-1-kai.heng.feng@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 1f67396849f1e5826de0616dff93cf9f7fd987dd\nAuthor: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\nDate: Wed Aug 21 20:39:09 2019 +0800\n\n serial: sprd: Add loopback function support\n \n Add loopback function support for Spreadtrum serial controller.\n \n Signed-off-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Link: https://lore.kernel.org/r/1275cd9968f1ceb5ac049cc23f1e508025cd552f.1566375260.git.baolin.wang@linaro.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 6ef35398e827dcf321a012c8f7a4d8d739674db6\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Mon Aug 12 16:28:50 2019 -0700\n\n rtc: Add Amlogic Virtual Wake RTC\n \n The Amlogic Meson GX SoCs uses a special register to store the\n time in seconds to wakeup after a system suspend.\n \n In order to be able to reuse the RTC wakealarm feature, this\n driver implements a fake RTC device which uses the system time\n to deduce a suspend delay.\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n [khilman: rebase to v5.3-rc, rework and modernization]\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Link: https://lore.kernel.org/r/20190812232850.8016-3-khilman@kernel.org\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit ed16239637f6656d3e6656ed1753e70d7823d94f\nAuthor: Kevin Hilman \u003ckhilman@baylibre.com\u003e\nDate: Mon Aug 12 16:28:49 2019 -0700\n\n dt-bindings: rtc: new binding for Amlogic VRTC\n \n Add binding fo the new VRTC driver for Amlogic SoCs. The 64-bit\n family of SoCs only has an RTC managed by firmware, and this VRTC\n driver provides the simple, one-register firmware interface.\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Link: https://lore.kernel.org/r/20190812232850.8016-2-khilman@kernel.org\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 92c8026854c25093946e0d7fe536fd9eac440f06\nAuthor: hexin \u003chexin.op@gmail.com\u003e\nDate: Thu Aug 22 11:35:19 2019 +0800\n\n vfio_pci: Restore original state on release\n \n vfio_pci_enable() saves the device\u0027s initial configuration information\n with the intent that it is restored in vfio_pci_disable(). However,\n the commit referenced in Fixes: below replaced the call to\n __pci_reset_function_locked(), which is not wrapped in a state save\n and restore, with pci_try_reset_function(), which overwrites the\n restored device state with the current state before applying it to the\n device. Reinstate use of __pci_reset_function_locked() to return to\n the desired behavior.\n \n Fixes: 890ed578df82 (\"vfio-pci: Use pci \"try\" reset interface\")\n Signed-off-by: hexin \u003chexin15@baidu.com\u003e\n Signed-off-by: Liu Qi \u003cliuqi16@baidu.com\u003e\n Signed-off-by: Zhang Yu \u003czhangyu31@baidu.com\u003e\n Signed-off-by: Alex Williamson \u003calex.williamson@redhat.com\u003e\n\ncommit 1b5f5d388b2b4fcb3324914bc90b3956e50ac2b0\nAuthor: Marcin Formela \u003cmarcin.formela@intel.com\u003e\nDate: Tue Jul 23 06:01:44 2019 -0400\n\n i40e: fix retrying in i40e_aq_get_phy_capabilities\n \n Fixed a bug where driver was breaking out of the loop and\n reporting an error without retrying first.\n \n Signed-off-by: Marcin Formela \u003cmarcin.formela@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 65c275e40164d5ef62f2dd8f2705db29b64978f4\nAuthor: Sylwia Wnuczko \u003csylwia.wnuczko@intel.com\u003e\nDate: Tue Jul 23 06:01:43 2019 -0400\n\n i40e: Persistent LLDP support\n \n This patch adds a function to read NVM module data and uses it to\n read current LLDP agent configuration from NVM API version 1.8.\n \n Signed-off-by: Sylwia Wnuczko \u003csylwia.wnuczko@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit a39f165db5c4f00becb3e50d65ea2321df64cdf3\nAuthor: Piotr Kwapulinski \u003cpiotr.kwapulinski@intel.com\u003e\nDate: Tue Jul 23 06:01:41 2019 -0400\n\n i40e: allow reset in recovery mode\n \n Driver waits after issuing a reset. When a reset takes too long a driver\n gives up. Implemented by invoking PF reset in a loop. After defined\n number of unsuccessful PF reset trials it returns error.\n Without this patch PF reset fails when NIC is in recovery mode.\n \n So make i40e_set_mac_type() public. i40e driver requires i40e_set_mac_type()\n to be public. It is required for recovery mode handling. Without this patch\n recovery mode could not be detected in i40e_probe().\n \n Signed-off-by: Piotr Kwapulinski \u003cpiotr.kwapulinski@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 541d97310ae3c02675578a236674e0bea1ad6483\nAuthor: Grzegorz Siwik \u003cgrzegorz.siwik@intel.com\u003e\nDate: Tue Jul 23 06:01:40 2019 -0400\n\n i40e: Remove function i40e_update_dcb_config()\n \n This patch removes function i40e_update_dcb_config(). Instead of\n i40e_update_dcb_config() we use i40e_init_dcb(), which implements the\n correct NVM read.\n \n Signed-off-by: Grzegorz Siwik \u003cgrzegorz.siwik@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 9889707b06acfe9bb37a6edcaae627d4a5eacc72\nAuthor: Slawomir Laba \u003cslawomirx.laba@intel.com\u003e\nDate: Tue Jul 23 06:01:39 2019 -0400\n\n i40e: Fix crash caused by stress setting of VF MAC addresses\n \n Add update to the VSI pointer passed to the i40e_set_vf_mac function.\n If VF is in reset state the driver waits in i40e_set_vf_mac function\n for the reset to be complete, yet after reset the vsi pointer\n that was passed into this function is no longer valid.\n \n The patch updates local VSI pointer directly from pf-\u003evsi array,\n by using the id stored in VF pointer (lan_vsi_idx).\n \n Without this commit the driver might occasionally invoke general\n protection fault in kernel and disable the OS entirely.\n \n Signed-off-by: Slawomir Laba \u003cslawomirx.laba@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 1e0303fd29428e3e3bbd8edf72f80750f86a7116\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Tue Jul 23 06:01:37 2019 -0400\n\n i40e: reset veb.tc_stats when resetting veb.stats\n \n The stats structure for the VEB switch statistics is reset periodically,\n but the tc_stats are not reset at the same time.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit f93b3fd9a34ba8ac2228e219e05259e503325ab7\nAuthor: Piotr Azarewicz \u003cpiotr.azarewicz@intel.com\u003e\nDate: Tue Jul 23 06:01:36 2019 -0400\n\n i40e: Update FW API version to 1.9\n \n Upcoming FW increment API version to 1.9 due to Extend PHY access AQ\n command support. SW is ready for that support as well.\n \n Signed-off-by: Piotr Azarewicz \u003cpiotr.azarewicz@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit d4256c8e9a7fab6e6c5cc3aa52ef7d2ff543b1ff\nAuthor: Adrian Podlawski \u003cadrian.podlawski@intel.com\u003e\nDate: Tue Jul 23 06:01:35 2019 -0400\n\n i40e: check_recovery_mode had wrong if statement\n \n Function check_recovery_mode had wrong if statement.\n Now we check proper FWS1B register values, which are responsible for\n the recovery mode. Recovery mode has 4 values for x710 and 2 for x722.\n That\u0027s why we need 6 different flags which are defined in the code.\n Now in the if statement, we recognize type of mac address\n and register value.\n Without those changes driver could show wrong state.\n \n Signed-off-by: Adrian Podlawski \u003cadrian.podlawski@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit d802c760ab87bc0d98e7517b766cf1b9f186fe65\nAuthor: Sylwia Wnuczko \u003csylwia.wnuczko@intel.com\u003e\nDate: Tue Jul 23 06:01:34 2019 -0400\n\n i40e: Add drop mode parameter to set mac config\n \n This patch adds \"drop mode\" parameter to set mac config AQ command.\n This bit controls the behavior when a no-drop packet is blocking a TC\n queue.\n 0 – The PF driver is notified.\n 1 – The blocking packet is dropped and then the PF driver is notified.\n \n Signed-off-by: Sylwia Wnuczko \u003csylwia.wnuczko@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit fb59826288a662e0a0c1934d7d068dc91f35764d\nAuthor: Beilei Xing \u003cbeilei.xing@intel.com\u003e\nDate: Tue Jul 23 06:01:33 2019 -0400\n\n i40e: fix shifts of signed values\n \n This patch fixes following error reported by cppcheck:\n (error) Shifting signed 32-bit value by 31 bits is undefined behaviour\n \n Signed-off-by: Beilei Xing \u003cbeilei.xing@intel.com\u003e\n Signed-off-by: Ferruh Yigit \u003cferruh.yigit@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 408bfc382efca14cc02c57ae935d5b7efd72e02a\nAuthor: huhai \u003chuhai@kylinos.cn\u003e\nDate: Mon Jul 22 19:55:59 2019 +0800\n\n i40e: add check on i40e_configure_tx_ring() return value\n \n When i40e_configure_tx_ring(vsi-\u003etx_rings[i]) returns an error, we should\n exit from i40e_vsi_configure_tx and return the error, instead of continuing\n to check whether xdp is enable, and configure the xdp transmit ring.\n \n Signed-off-by: huhai \u003chuhai@kylinos.cn\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit bc6c1eaaedee7bb1bbdef27e155f15510128762e\nAuthor: Mauro S. M. Rodrigues \u003cmaurosr@linux.vnet.ibm.com\u003e\nDate: Thu Jul 18 14:52:02 2019 -0300\n\n i40e: Check if transceiver implements DDM before access\n \n Similar to the ixgbe issue fixed in:\n 655c91414579 (\"ixgbe: Check DDM existence in transceiver before access)\n \n i40e has the same issue when reading eeprom from SFP\u0027s module that comply\n with SFF-8472 but not implement the Digital Diagnostic Monitoring (DDM)\n interface described in it. The existence of such area is specified by bit\n 6 of byte 92, set to 1 if implemented.\n \n Without this patch, due to not checking this bit i40e fails to read SFP\n module\u0027s eeprom with the follow message:\n \n ethtool -m enP51p1s0f0\n Cannot get Module EEPROM data: Input/output error\n \n Because it fails to read the additional 256 bytes in which it was assumed\n to exist the DDM data.\n \n Signed-off-by: \"Mauro S. M. Rodrigues\" \u003cmaurosr@linux.vnet.ibm.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 33b165684ab70867d4545643f550a5d48d3ddc57\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Mon Jul 15 14:35:07 2019 +0200\n\n i40e: reduce stack usage in i40e_set_fc\n \n The functions i40e_aq_get_phy_abilities_resp() and i40e_set_fc() both\n have giant structure on the stack, which makes each one use stack frames\n larger than 500 bytes.\n \n As clang decides one function into the other, we get a warning for\n exceeding the frame size limit on 32-bit architectures:\n \n drivers/net/ethernet/intel/i40e/i40e_common.c:1654:23: error: stack frame size of 1116 bytes in function \u0027i40e_set_fc\u0027 [-Werror,-Wframe-larger-than\u003d]\n \n When building with gcc, the inlining does not happen, but i40e_set_fc()\n calls i40e_aq_get_phy_abilities_resp() anyway, so they add up on the\n kernel stack just as much.\n \n The parts that actually use large stacks don\u0027t overlap, so make sure\n each one is a separate function, and mark them as noinline_for_stack to\n prevent the compilers from combining them again.\n \n Fixes: 0a862b43acc6 (\"i40e/i40evf: Add module_types and update_link_info\")\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit cee508a0bddba1627575de2e41fb8e321fa380a8\nAuthor: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\nDate: Tue Aug 20 15:30:59 2019 -0700\n\n drm/dp/dsc: Add Support for all BPCs supported by TGL\n \n DSC engine on ICL supports only 8 and 10 BPC as the input\n BPC. But DSC engine in TGL supports 8, 10 and 12 BPC.\n Add 12 BPC support for DSC while calculating compression\n configuration.\n \n v2: Remove the separate define TGL_DP_DSC_MAX_SUPPORTED_BPC\n and use the value directly.(More such defines can be removed\n as part of future patches). (Ville)\n \n v3: Use values directly instead of accessing the defines\n everytime for min and max DSC BPC.\n \n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Manasi Navare \u003cmanasi.d.navare@intel.com\u003e\n Signed-off-by: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Reviewed-by: Manasi Navare \u003cmanasi.d.navare@intel.com\u003e\n Signed-off-by: Manasi Navare \u003cmanasi.d.navare@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820223059.18052-1-anusha.srivatsa@intel.com\n\ncommit d9c5c083416500e95da098c01be092b937def7fa\nAuthor: Gerald BAEZA \u003cgerald.baeza@st.com\u003e\nDate: Thu Aug 22 09:07:01 2019 +0000\n\n libperf: Fix alignment trap with xyarray contents in \u0027perf stat\u0027\n \n Following the patch \u0027perf stat: Fix --no-scale\u0027, an alignment trap\n happens in process_counter_values() on ARMv7 platforms due to the\n attempt to copy non 64 bits aligned double words (pointed by \u0027count\u0027)\n via a NEON vectored instruction (\u0027vld1\u0027 with 64 bits alignment\n constraint).\n \n This patch sets a 64 bits alignment constraint on \u0027contents[]\u0027 field in\n \u0027struct xyarray\u0027 since the \u0027count\u0027 pointer used above points to such a\n structure.\n \n Signed-off-by: Gerald Baeza \u003cgerald.baeza@st.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/1566464769-16374-1-git-send-email-gerald.baeza@st.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 1ea770f6c1971bc101b3741f4d88b0b4ea5c4181\nAuthor: Ravi Bangoria \u003cravi.bangoria@linux.ibm.com\u003e\nDate: Thu Aug 22 14:20:45 2019 +0530\n\n perf c2c: Fix report with offline cpus\n \n If c2c is recorded on a machine where any cpus are offline, \u0027perf c2c\n report\u0027 throws an error \"node/cpu topology bugFailed setup nodes\".\n \n It fails because while preparing node-cpu mapping we don\u0027t consider\n offline cpus.\n \n Reported-by: Nageswara R Sastry \u003cnasastry@in.ibm.com\u003e\n Signed-off-by: Ravi Bangoria \u003cravi.bangoria@linux.ibm.com\u003e\n Acked-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Fixes: 1e181b92a2da (\"perf c2c report: Add \u0027node\u0027 sort key\")\n Link: http://lkml.kernel.org/r/20190822085045.25108-1-ravi.bangoria@linux.ibm.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit a2f354e3abb853f9a40048829e1f839e8f7fada5\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Thu Aug 22 13:11:41 2019 +0200\n\n libperf: Add perf_thread_map__nr/perf_thread_map__pid functions\n \n So it\u0027s part of libperf library as basic functions operating on\n perf_thread_map objects.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190822111141.25823-6-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 5e51b0bb245d963f5ce750256c504be95201e38c\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 22 10:48:31 2019 -0300\n\n perf cpumap: Remove needless includes from cpumap.h\n \n The util/cpumap.h file doesn\u0027t use anything in refcount.h not in\n debug.h, it needs just a forward reference to \u0027struct cpu_map_data\u0027,\n that is defined in util/event.h and cpumap.h was getting indirectly via,\n of all things, debug.h\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-mtjww98yptt4ppo6g2blavg5@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit f7004f5990e230710112b89f2f11777e5e64e258\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 21 16:39:29 2019 -0300\n\n perf evsel: Switch to libperf\u0027s cpumap.h\n \n We don\u0027t need what is in perf\u0027s util/cpumap.h, just the struct cpu_map\n that is in libperf\u0027s internal/cpumap.h file to cover this one case:\n \n tools/perf/util/evsel.h:215:27: error: dereferencing pointer to incomplete type ‘struct perf_cpu_map’\n 215 | return evsel__cpus(evsel)-\u003enr;\n \n So switch to libperf\u0027s cpumap.h and add some missing struct foward\n declarations and include sys/types.h to get pid_t.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-ufjkpohijti05ggk69s91ktf@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 1028f96226b1f4419bd5b56c3d234a83329d4f5b\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 21 14:48:57 2019 -0300\n\n perf x86 kvm-stat: Add missing string.h header\n \n It uses strcmp(), strstr() and was getting the required string.h header\n by luck, from evsel.h -\u003e cpumap.h -\u003e debug.h -\u003e string.h, add the\n missing header.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-qrz8hhvrhwnmt5ocfwk4br5d@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 43cc5d5ecbd194a730f3045547afc1fa8e694389\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 21 14:44:25 2019 -0300\n\n perf evsel: util/evsel.h needs stdio.h as it uses FILE\n \n And it was getting it by luck from util/cpumap.h that shouldn\u0027t be\n included in util/evsel.h as it only needs what is in libperf, i.e.\n struct cpu_map, that is in internal/cpumap.h, so add stdio.h before\n we fix that.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-2ywx5sl031tj3zske7c7edgv@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit a06b7f422d6a759b085ab2988d878b70e2dd0064\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 21 14:24:10 2019 -0300\n\n perf evsel: Remove needless stddef.h from util/evsel.h\n \n We added it in 07ac002f2fcc (\"perf evsel: Introduce is_group_member\n method\") but we already ditched that function, and there was nothing\n else left that needed NULL nor anything else from stddef.h, ditch it.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-1zy0xfsy61x81f3fpyx5znco@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit ddee688a83073a9beebc5c86b67c712de5861411\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 21 14:20:54 2019 -0300\n\n perf evsel: Remove needless counts.h header from util/evsel.h\n \n We need only a struct forward declaration, so prune the header\n dependency tree a bit more.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-oqvgf04w4ku8xasrz79zquim@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 69714a4e3959eb051e685c6dce06c6d5a8f27c3c\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 21 14:09:54 2019 -0300\n\n perf evsel: Add missing perf/evsel.h header in util/evsel.h\n \n Since util/evsel.h uses perf_evsel__cpus() that has its prototype in\n libperf\u0027s perf/evsel.h file, we need it explicitely included.\n \n This was working by luck as util/evsel.h includes counts.h, but that is\n not necessary, just some forward declarations, so, before we remove\n counts.h from util/evsel.h, add what is realli needed.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-nfb9e0t4jm9zhvr0q86hc29d@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 430482c2e34ea8c164240ce295b5f57f16ab5621\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 21 14:02:35 2019 -0300\n\n perf scripting python: Add missing counts.h header\n \n It is getting this via evsel.h, that don\u0027t strictly need counts.h, just\n forward declarations for some structs, so add it here before we remove\n it from there.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-6bxk3ltwkw91qcld2ot86bgg@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit bfc49182c64e9dbdea494a577894194701b61e72\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 21 14:02:05 2019 -0300\n\n perf stat: Add missing counts.h\n \n It is getting this via evsel.h, that don\u0027t strictly need counts.h, just\n forward declarations for some structs, so add it here before we remove\n it from there.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-jwcbm9gv9llloe3he5qkdefs@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit e4aec1b1bdad744f3afc3ebf2b337ac1bcfa9be0\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 21 14:01:24 2019 -0300\n\n perf tests: Add missing counts.h\n \n Those are getting counts.h via evsel.h, that don\u0027t strictly need\n counts.h, just forward declarations for some structs, so add it here\n before we remove it from there.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-phldqlfxxu563txja7evd4zt@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 0f31c0195c14cc37b2f2a08e3f4219d69faa1b34\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 21 14:00:17 2019 -0300\n\n perf script: Add missing counts.h\n \n It is getting this via evsel.h, that don\u0027t strictly need counts.h, just\n forward declarations for some structs, so add it here before we remove\n it from there.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-q4shpvlxyjqz7val1hyrdak9@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit e14e5497d5253712e1545f79ad93309499ffb544\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 21 13:58:12 2019 -0300\n\n perf evlist: Add missing xyarray.h header\n \n It gets it very indirectly, via evsel.h -\u003e counts.h, and since counts.h\n doesn\u0027t need xyarray.h at all, add it here before we remove it there.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-hkizv6gojwfklj9ezaiiztll@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 964f384989585bc265fd929b2a7977b83fbe4c3b\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 21 11:57:50 2019 -0300\n\n perf bpf: Add missing xyarray.h header\n \n This was being obtained indirectly via evsel.h -\u003e counts.h, since we\n don\u0027t need xyarray in counts.h, we need to add it here explicitely\n before removing it from counts.h.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-jirmxg527i82yz31bwad9we7@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 2d64ae9b85614dc0fcca68aad5da305dec44a9b1\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 21 11:56:13 2019 -0300\n\n perf counts: Add missing headers needed for types used\n \n We get these by sheer luck, since we\u0027re cleaning unneeded headers use,\n this needs to be done first to avoid breakage down the line.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-p7bncbi53t4p2kobkbmu86a4@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 7646602401e6f45e4013ddb7c41f6bc211032d02\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 21 11:30:29 2019 -0300\n\n perf evsel: Move xyarray.h from evsel.c to evsel.h to reduce include dep tree\n \n All we need in util/evsel.h is the foward declaration of \u0027struct\n xyarray\u0027, not the internal/xyarray.h, that can be moved to util/evsel.c\n and then we reduce the header dependency tree.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-wwqce6ixwcyq6yzx3ljrdm80@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 0b8026e8fb0ea3893caa2f1924a2c15fcf6760b3\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 21 10:54:14 2019 -0300\n\n perf metricgroup: Remove needless includes from metricgroup.h\n \n There we need just some struct forward declarations, do that instead and\n add the includes needed by metricgroup.c.\n \n That should help with needless rebuilds when changing the removed\n headers from metricgroup.h.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-1fkskjws6imir2hhztqhdyb0@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit e740ca86f354038f55978c8ac7bec69b57f0c8e0\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 22 17:15:42 2019 -0300\n\n perf kvm s390: Add missing string.h header\n \n It uses strstr(), needs to include string.h or its not going to build\n when we remove string.h from the place it is getting from indirectly, by\n luck.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-72y0i0uiaqght5b83e3ae7p4@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit df7415bfc06f8ee482a510aea39c12121e409dcc\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Tue Aug 20 15:33:25 2019 -0700\n\n drm/i915: Do not unmask PSR interruption in IRQ postinstall\n \n No need to unmask PSR interrutpion if PSR is not enabled, better move\n the call to intel_psr_enable_source().\n \n v2: Renamed intel_psr_irq_control() to psr_irq_control() (Lucas)\n \n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Dhinakaran Pandiyan \u003cdhinakaran.pandiyan@intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820223325.27490-3-jose.souza@intel.com\n\ncommit 99fc38b12095cd0ca9fc5b8c8e27e827dff6ed34\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Tue Aug 20 15:33:24 2019 -0700\n\n drm/i915: Add transcoder restriction to PSR2\n \n According to PSR2_CTL definition in BSpec there is only one instance\n of PSR2_CTL. Platforms gen \u003c 12 with EDP transcoder only support PSR2\n on TRANSCODER_EDP while on TGL PSR2 is only supported by\n TRANSCODER_A.\n \n Since BDW PSR is allowed on any port, but we need to restrict by\n transcoder.\n \n v8: Renamed _psr2_supported_in_trans() to psr2_supported() (Lucas)\n \n v9: Renamed psr2_supported() to transcoder_has_psr2() (Ville)\n \n BSpec: 7713\n BSpec: 20584\n Cc: Dhinakaran Pandiyan \u003cdhinakaran.pandiyan@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Reviewed-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Anshuman Gupta \u003canshuman.gupta@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820223325.27490-2-jose.souza@intel.com\n\ncommit 4ab4fa1032170b6ec50ffff0767096b3200dde8b\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Tue Aug 20 15:33:23 2019 -0700\n\n drm/i915/psr: Make PSR registers relative to transcoders\n \n PSR registers are a mess, some have the full address while others just\n have the additional offset from psr_mmio_base.\n \n For BDW+ psr_mmio_base is nothing more than TRANSCODER_EDP_OFFSET +\n 0x800 and using it makes more difficult for people with an PSR\n register address or PSR register name from from BSpec as i915 also\n don\u0027t match the BSpec names.\n For HSW psr_mmio_base is _DDI_BUF_CTL_A + 0x800 and PSR registers are\n only available in DDIA.\n \n Other reason to make relative to transcoder is that since BDW every\n transcoder have PSR registers, so in theory it should be possible to\n have PSR enabled in a non-eDP transcoder.\n \n So for BDW+ we can use _TRANS2() to get the register offset of any\n PSR register in any transcoder while for HSW we have _HSW_PSR_ADJ\n that will calculate the register offset for the single PSR instance,\n noting that we are already guarded about trying to enable PSR in other\n port than DDIA on HSW by the \u0027if (dig_port-\u003ebase.port !\u003d PORT_A)\u0027 in\n intel_psr_compute_config(), this check should only be valid for HSW\n and will be changed in future.\n PSR2 registers and PSR_EVENT was added after Haswell so that is why\n _PSR_ADJ() is not used in some macros.\n \n The only registers that can not be relative to transcoder are\n PSR_IMR and PSR_IIR that are not relative to anything, so keeping it\n hardcoded. That changed for TGL but it will be handled in another\n patch.\n \n Also removing BDW_EDP_PSR_BASE from GVT because it is not used as it\n is the only PSR register that GVT have.\n \n v5:\n - Macros changed to be more explicit about HSW (Dhinakaran)\n - Squashed with the patch that added the tran parameter to the\n macros (Dhinakaran)\n \n v6:\n - Checking for interruption errors after module reload in the\n transcoder that will be used (Dhinakaran)\n - Using lowercase to the registers offsets\n \n v7:\n - Removing IS_HASWELL() from registers macros(Jani)\n \n Cc: Dhinakaran Pandiyan \u003cdhinakaran.pandiyan@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@linux.intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Zhi Wang \u003czhi.a.wang@intel.com\u003e\n Reviewed-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820223325.27490-1-jose.souza@intel.com\n\ncommit 7010998c6caf7cf9706f31c1880b7aeac904e874\nAuthor: Matthew Wang \u003cmatthewmwang@chromium.org\u003e\nDate: Thu Aug 22 10:48:06 2019 -0700\n\n nl80211: add NL80211_CMD_UPDATE_FT_IES to supported commands\n \n Add NL80211_CMD_UPDATE_FT_IES to supported commands. In mac80211 drivers,\n this can be implemented via existing NL80211_CMD_AUTHENTICATE and\n NL80211_ATTR_IE, but non-mac80211 drivers have a separate command for\n this. A driver supports FT if it either is mac80211 or supports this\n command.\n \n Signed-off-by: Matthew Wang \u003cmatthewmwang@chromium.org\u003e\n Reviewed-by: Brian Norris \u003cbriannorris@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190822174806.2954-1-matthewmwang@chromium.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit b26af9304467d8e7844397e702b51a4bdc74bedd\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Thu Aug 22 13:20:34 2019 +0100\n\n mac80211: minstrel_ht: fix infinite loop because supported is not being shifted\n \n Currently the for-loop will spin forever if variable supported is\n non-zero because supported is never changed. Fix this by adding in\n the missing right shift of supported.\n \n Addresses-Coverity: (\"Infinite loop\")\n Fixes: 48cb39522a9d (\"mac80211: minstrel_ht: improve rate probing for devices with static fallback\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190822122034.28664-1-colin.king@canonical.com\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 9f72a1d0300b03a9fc6cba93f8cc0ab326185788\nAuthor: Kever Yang \u003ckever.yang@rock-chips.com\u003e\nDate: Wed Aug 21 11:11:24 2019 +0800\n\n dt-bindings: arm: rockchip: remove reference to fennec board\n \n The rk3288 fennec board has been removed, remove the binding document at\n the same time.\n \n Signed-off-by: Kever Yang \u003ckever.yang@rock-chips.com\u003e\n Acked-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit 3bf7ec62f8b37143d1f18bc30f77c8f4093c1e09\nAuthor: Kever Yang \u003ckever.yang@rock-chips.com\u003e\nDate: Wed Aug 21 11:11:23 2019 +0800\n\n ARM: dts: rockchip: remove rk3288 fennec board support\n \n Since there is no one using this board, remove it.\n \n Signed-off-by: Kever Yang \u003ckever.yang@rock-chips.com\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit e8cae2e64202f5acc209e81cb4f24614daa900b7\nAuthor: Jonas Karlman \u003cjonas@kwiboo.se\u003e\nDate: Wed Aug 21 17:54:38 2019 +0000\n\n arm64: dts: rockchip: add rk3328 VPU node\n \n This patch add a VPU device node for rk3328.\n \n Signed-off-by: Jonas Karlman \u003cjonas@kwiboo.se\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit 3b81d8bdd9143449a85166e7987aedea145fce54\nAuthor: Nick Crews \u003cncrews@chromium.org\u003e\nDate: Mon Aug 5 14:22:14 2019 -0600\n\n platform/chrome: wilco_ec: Add batt_ppid_info command to telemetry driver\n \n Add the GET_BATT_PPID_INFO\u003d0x8A command to the allowlist of accepted\n telemetry commands. In addition, since this new command requires\n verifying the contents of some of the arguments, I also restructure\n the request to use a union of the argument structs. Also, zero out the\n request buffer before each request, and change \"whitelist\" to\n \"allowlist\".\n \n Signed-off-by: Nick Crews \u003cncrews@chromium.org\u003e\n Signed-off-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n\ncommit be28f76b7e8ffaf04690e30fd17683d8afea1fd9\nAuthor: Ashish Kumar \u003cAshish.Kumar@nxp.com\u003e\nDate: Tue Aug 13 15:53:09 2019 +0530\n\n spi: spi-fsl-qspi: Add ls2080a compatibility string to bindings\n \n There are 2 version of QSPI-IP, according to which controller registers sets\n can be big endian or little endian.There are some other minor changes like\n RX fifo depth etc.\n \n The big endian version uses driver compatible \"fsl,ls1021a-qspi\" and\n little endian version uses driver compatible \"fsl,ls2080a-qspi\"\n \n Signed-off-by: Kuldeep Singh \u003ckuldeep.singh@nxp.com\u003e\n Signed-off-by: Ashish Kumar \u003cashish.kumar@nxp.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Acked-by: Han Xu \u003chan.xu@nxp.com\u003e\n Link: https://lore.kernel.org/r/1565691791-26167-1-git-send-email-Ashish.Kumar@nxp.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c76c992525245ec1c7b6738bf887c42099abab02\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Thu Aug 22 13:53:40 2019 +0100\n\n nexthops: remove redundant assignment to variable err\n \n Variable err is initialized to a value that is never read and it is\n re-assigned later. The initialization is redundant and can be removed.\n \n Addresses-Coverity: (\"Unused Value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Reviewed-by: David Ahern \u003cdsahern@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 303290e130bfd504a8230f49e545f6b91e120553\nAuthor: Ashish Kumar \u003cAshish.Kumar@nxp.com\u003e\nDate: Tue Aug 13 15:53:11 2019 +0530\n\n spi: fsl-qspi: Enhance binding to extend example for flash entry\n \n Add example for adding flash entry on various boards\u0027 dts\n using flash manufacture spansion/cypress.\n \n Signed-off-by: Ashish Kumar \u003cAshish.Kumar@nxp.com\u003e\n Link: https://lore.kernel.org/r/1565691791-26167-3-git-send-email-Ashish.Kumar@nxp.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c663d542bfb40eeeb6d393ed155c23a4666d65e1\nAuthor: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\nDate: Sun Aug 18 19:27:50 2019 +0200\n\n MAINTAINERS: Extend patterns for Samsung SoC, Security Subsystem and clock drivers\n \n Extend the patterns to cover all related files in respective\n categories:\n 1. Samsung Exynos ARM architecture: add soc drivers headers and make\n directory matches consistent,\n 2. Samsung Security SubSystem driver (crypto): add bindings,\n 3. Samsung SoC clock drivers: add S3C24xx, S3C64xx and S5Pv210 bindings.\n \n Cc: Kukjin Kim \u003ckgene@kernel.org\u003e\n Cc: Vladimir Zapolskiy \u003cvz@mleia.com\u003e\n Cc: Kamil Konieczny \u003ck.konieczny@partner.samsung.com\u003e\n Cc: Sylwester Nawrocki \u003cs.nawrocki@samsung.com\u003e\n Cc: Tomasz Figa \u003ctomasz.figa@gmail.com\u003e\n Cc: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Acked-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Acked-by: Sylwester Nawrocki \u003cs.nawrocki@samsung.com\u003e\n\ncommit 28578825ede61834a2c46e7f9a89012c4c7a667f\nAuthor: Sylwester Nawrocki \u003cs.nawrocki@samsung.com\u003e\nDate: Wed Aug 21 17:39:26 2019 +0200\n\n soc: samsung: chipid: Select missing dependency for EXYNOS_CHIPID\n \n The chipid driver uses the MFD syscon API but it was not covered\n properly in Kconfig.\n \n Signed-off-by: Sylwester Nawrocki \u003cs.nawrocki@samsung.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 823034575c177d1d69673102a994d5d33b05e3de\nAuthor: Sylwester Nawrocki \u003cs.nawrocki@samsung.com\u003e\nDate: Wed Aug 21 17:05:39 2019 +0200\n\n soc: samsung: chipid: Remove the regmap lookup error log\n \n In commit 40d8aff614f7 (\"soc: samsung: chipid: Convert exynos-chipid\n driver to use the regmap API\") of_find_compatible_node() call was\n substituted with syscon_regmap_lookup_by_compatible() but also an error\n log was added for case where lookup fails. On multiplatform the lookup\n will always fail on any non-samsung device so the log is incorrect.\n Remove the error log and just return an error code from\n syscon_regmap_lookup_by_compatible() which internally calls\n of_find_compatible_node().\n \n Reported-by: Jon Hunter \u003cjonathanh@nvidia.com\u003e\n Signed-off-by: Sylwester Nawrocki \u003cs.nawrocki@samsung.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 45a2c0ccf6b96b046ad167a8c90ceaa5e59d07cb\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 22 15:03:37 2019 -0300\n\n perf arm64: Add missing debug.h header\n \n This file uses pr_debug() but isn\u0027t including debug.h, getting it by\n luck, fix it.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-t7pisnsdfh88kclpw52jcwl7@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 495efdb01f89a5fc53f9b2e61f5726d804d4a15d\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Tue Aug 20 14:05:20 2019 +0900\n\n ASoC: soc-core: initialize list at one place\n \n Initialize component related list at random place is very difficult\n to read. This patch initialize it at snd_soc_component_initialize().\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Reviewed-by: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/87y2zozazp.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit b2f3e0c9e4ee2e07fd880cc0c4da20c42d633d4b\nAuthor: Jaska Uimonen \u003cjaska.uimonen@intel.com\u003e\nDate: Thu Aug 22 00:11:38 2019 +0300\n\n ASoC: SOF: topology: fix get control data return type and arguments\n \n sof_get_control_data returns negative values even though the return\n value is defined unsigned (size_t). So change the return value type to\n int and add the data size as pointer argument to sof_get_control_data to\n avoid ambiquity in the meaning of the return type.\n \n Fixes: cac974a51ebb (\"ASoC: SOF: topology: use set_get_data in process load\")\n Reported by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Signed-off-by: Jaska Uimonen \u003cjaska.uimonen@intel.com\u003e\n Link: https://lore.kernel.org/r/20190821211138.14618-1-jaska.uimonen@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 8d92bb516831e80fac916701447ee6e9f0a6f0f2\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Tue Aug 20 14:05:16 2019 +0900\n\n ASoC: soc-core: initialize component list\n \n It might return without initializing in error case.\n In such case, uninitialized variable might be used at error handler.\n This patch initializes all necessary variable before return.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Reviewed-by: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/87zhk4zazt.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 43ca5dab978294eae26a36f8989b6f0769da4256\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Tue Aug 20 14:05:32 2019 +0900\n\n ASoC: soc-topology: use for_each_component_dais() at remove_dai()\n \n commit 52abe6cc1866a (\"ASoC: topology: fix oops/use-after-free case\n with dai driver\") fixups remove_dai() error, but it is using\n list_for_each_entry() for component-\u003edai_list.\n \n We already have for_each_component_dais() macro for it.\n Let\u0027s use exising method.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87tvaczazd.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 6b8ac43c33b912bc5435192ce3b3f051f6f2dc9d\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Tue Aug 20 14:05:28 2019 +0900\n\n ASoC: soc-dai: use bit field for bus_control\n \n .bus_control can be bit field.\n this patch do it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Reviewed-by: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/87v9uszazh.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit a4de83a385670c22c31e9bbb726595a447b32ba4\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Tue Aug 20 14:05:10 2019 +0900\n\n ASoC: soc-core: soc_cleanup_card_resources() become void\n \n There is no need to check return value for\n soc_cleanup_card_resources(). Let\u0027t makes it as void.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Reviewed-by: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/871rxg1lda.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit a33c0d166cc5bcb3b9718649b84974216709acb1\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Tue Aug 20 14:05:02 2019 +0900\n\n ASoC: soc-core: add NOTE to snd_soc_rtdcom_lookup()\n \n We can find specified name component via snd_soc_rtdcom_lookup().\n But, it is not enough under multi CPU/Codec/Platform, because many\n components which have same driver name might be connected to same rtd.\n \n Not using this function as much as possible is best solution,\n but some drivers are already deeply depended to it.\n \n We can expand this function, for example having \"num\" which specifies\n found order at parameter, etc (In such case, it need to have fixed\n probing order).\n Or, use different driver name in such component, etc.\n \n We will have such issue if multi CPU/Codec/Platform were supported.\n To indicate it, this patch adds NOTE to this function.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Reviewed-by: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/874l2c1ldi.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 4168ddabb480bef818c93f378428632fb681b500\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Tue Aug 20 14:04:49 2019 +0900\n\n ASoC: soc-core: use device_register()\n \n It is easy to read code if it is cleanly using paired function/naming,\n like start \u003c-\u003e stop, register \u003c-\u003e unregister, etc, etc.\n But, current ALSA SoC code is very random, unbalance, not paired, etc.\n \n soc-core.c is using device_unregiser(), but there is no its paired\n device_regiser(). We can find its code at soc_post_component_init()\n which is using device_initialize() and device_add().\n Here, device_initialize() + device_add() \u003d device_register().\n \n -- linux/drivers/base/core.c --\n int device_register(struct device *dev)\n {\n device_initialize(dev);\n return device_add(dev);\n }\n \n device_initialize() is doing each dev member\u0027s initialization only,\n not related to device parent/release/groups.\n Thus, we can postpone it.\n let\u0027s use device_register() instead of device_initialize()/device_add().\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Reviewed-by: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/878sro1ldw.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 6191cbde5ff02ef5307e7b2e4f5b8f7607c2b1df\nAuthor: Jiaxin Yu \u003cjiaxin.yu@mediatek.com\u003e\nDate: Thu Aug 22 20:51:00 2019 +0800\n\n ASoC: mediatek: mt8183: switch tdm pins gpio function when playback on or off\n \n Pull TDM pins down when TDM BE shutdown to avoid current leakage.\n \n Signed-off-by: Jiaxin Yu \u003cjiaxin.yu@mediatek.com\u003e\n Link: https://lore.kernel.org/r/1566478261-13464-2-git-send-email-jiaxin.yu@mediatek.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 97aad5ce88164bc5f884b5234f015ac553ea2fe5\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 22 22:37:47 2019 +0800\n\n ASoC: mediatek: mt2701: Fix -Wunused-const-variable warnings\n \n sound/soc/mediatek/mt2701/mt2701-afe-common.h:66:27: warning:\n mt2701_afe_backup_list defined but not used [-Wunused-const-variable\u003d]\n \n mt2701_afe_backup_list is only used in mt2701-afe-pcm.c,\n so just move the definition over there.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190822143747.20944-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c101fb29737f4558bf589d0d66371d9e21040568\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 22 22:36:08 2019 +0800\n\n ASoC: wm8988: fix typo in wm8988_right_line_controls\n \n sound/soc/codecs/wm8988.c:270:30: warning:\n wm8988_rline_enum defined but not used [-Wunused-const-variable\u003d]\n \n wm8988_rline_enum should be used in wm8988_right_line_controls.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Fixes: 5409fb4e327a (\"ASoC: Add WM8988 CODEC driver\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190822143608.59824-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 202acc565a1f050f82fdab646fd9c86bfdb21733\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Wed Aug 21 11:47:30 2019 -0500\n\n ASoC: SOF: imx: Add i.MX8 HW support\n \n Add support for the audio DSP hardware found on NXP i.MX8 platform.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190821164730.7385-4-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit b9132b89933177286fb427bd03affcebc4d649e1\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Wed Aug 21 11:47:29 2019 -0500\n\n ASoC: SOF: Add OF DSP device support\n \n Add support for device tree based SOF DSP devices.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190821164730.7385-3-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 7db2f2dfc701f9f903283e830653b02fb6ce823c\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Wed Aug 21 11:47:28 2019 -0500\n\n dt-bindings: dsp: fsl: Add DSP core binding support\n \n This describes the DSP device tree node.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190821164730.7385-2-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 4035ff36a6e0a07a1f0b0609520e17bd69a74c8b\nAuthor: Hou Zhiqiang \u003cZhiqiang.Hou@nxp.com\u003e\nDate: Tue Aug 20 07:29:01 2019 +0000\n\n arm64: dts: fsl: Remove num-lanes property from PCIe nodes\n \n Remove the num-lanes property to avoid the driver setting the\n link width.\n \n On FSL Layerscape SoCs, the number of lanes assigned to PCIe\n controller is not fixed, it is determined by the selected SerDes\n protocol in the RCW (Reset Configuration Word).\n \n The PCIe link training is completed automatically through the selected\n SerDes protocol - the link width set-up is updated by hardware after\n power on reset, so the num-lanes property is not needed for Layerscape\n PCIe.\n \n The current num-lanes property was added erroneously, which actually\n indicates the maximum lanes the PCIe controller can support up to,\n instead of the lanes assigned to the PCIe controller. The link width set\n by SerDes protocol will be overridden by the num-lanes property, hence\n the subsequent re-training will fail when the assigned lanes do not\n match the value in the num-lanes property.\n \n Remove the property to fix the issue\n \n Signed-off-by: Hou Zhiqiang \u003cZhiqiang.Hou@nxp.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Reviewed-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n\ncommit 0ce6a624473e7e5752a84416f85f73d082308edd\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 22 22:30:07 2019 +0800\n\n ASoC: AMD: Fix Kconfig warning without GPIOLIB\n \n While do rand build without GPIOLIB, we get Kconfig warning:\\\n \n WARNING: unmet direct dependencies detected for SND_SOC_MAX98357A\n Depends on [n]: SOUND [\u003dy] \u0026\u0026 !UML \u0026\u0026 SND [\u003dm] \u0026\u0026 SND_SOC [\u003dm] \u0026\u0026 GPIOLIB [\u003dn]\n Selected by [m]:\n - SND_SOC_AMD_CZ_DA7219MX98357_MACH [\u003dm] \u0026\u0026 SOUND [\u003dy] \u0026\u0026 !UML \u0026\u0026 SND [\u003dm] \u0026\u0026 SND_SOC [\u003dm] \u0026\u0026 SND_SOC_AMD_ACP [\u003dm] \u0026\u0026 I2C [\u003dy]\n \n Add GPIOLIB dependency to fix this.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190822143007.73644-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 1243cb6a676ffcfa72dd25859edddf66cde0b638\nAuthor: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nDate: Thu Aug 22 15:17:43 2019 +0100\n\n arm64: Add tagged-address-abi.rst to index.rst\n \n Documentation/arm64/tagged-address-abi.rst introduces the\n relaxation of the syscall ABI that allows userspace to pass\n certain tagged pointers to kernel syscalls.\n \n Add the document to index.rst for a correct generation of the\n table of content.\n \n Cc: Will Deacon \u003cwill@kernel.org\u003e\n Cc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 568adba9eb201875bc561745e0f19b831b7e2bbc\nAuthor: Hou Zhiqiang \u003cZhiqiang.Hou@nxp.com\u003e\nDate: Tue Aug 20 07:28:55 2019 +0000\n\n ARM: dts: ls1021a: Remove num-lanes property from PCIe nodes\n \n Remove the num-lanes property to avoid the driver setting the\n link width.\n \n On FSL Layerscape SoCs, the number of lanes assigned to PCIe\n controller is not fixed, it is determined by the selected SerDes\n protocol in the RCW (Reset Configuration Word).\n \n The PCIe link training is completed automatically through the selected\n SerDes protocol - the link width set-up is updated by hardware after\n power on reset, so the num-lanes property is not needed for Layerscape\n PCIe.\n \n The current num-lanes property was added erroneously, which actually\n indicates the maximum lanes the PCIe controller can support up to,\n instead of the lanes assigned to the PCIe controller. The link width set\n by SerDes protocol will be overridden by the num-lanes property, hence\n the subsequent re-training will fail when the assigned lanes do not\n match the value in the num-lanes property.\n \n Remove the property to fix the issue.\n \n Signed-off-by: Hou Zhiqiang \u003cZhiqiang.Hou@nxp.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Reviewed-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n\ncommit 66de33f09fd97201847de7e1e2ec8a117242e1d6\nAuthor: Hou Zhiqiang \u003cZhiqiang.Hou@nxp.com\u003e\nDate: Tue Aug 20 07:28:49 2019 +0000\n\n PCI: dwc: Return directly when num-lanes is not found\n \n The num-lanes is optional since it is not needed on some platforms\n that bring up the link in firmware.\n \n The link programming is based on the num-lanes properties (which is\n optional); if it is not present code must return instead of fiddling\n with the lanes value to print an error message.\n \n Signed-off-by: Hou Zhiqiang \u003cZhiqiang.Hou@nxp.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Reviewed-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n\ncommit a91bcc2b65370e7debf1fc26b93a4c2a54433220\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Wed Aug 21 18:41:27 2019 +0100\n\n Revert \"iommu/arm-smmu-v3: Disable detection of ATS and PRI\"\n \n This reverts commit b5e86196b83fd68e065a7c811ab8925fb0dc3893.\n \n Now that ATC invalidation is performed in the correct places and without\n incurring a locking overhead for non-ATS systems, we can re-enable the\n corresponding SMMU feature detection.\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit cdb8a3c3463563b7bdb6f653bf4b0ffa3a95f366\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Aug 20 16:28:54 2019 +0100\n\n iommu/arm-smmu-v3: Avoid locking on invalidation path when not using ATS\n \n When ATS is not in use, we can avoid taking the \u0027devices_lock\u0027 for the\n domain on the invalidation path by simply caching the number of ATS\n masters currently attached. The fiddly part is handling a concurrent\n -\u003eattach() of an ATS-enabled master to a domain that is being\n invalidated, but we can handle this using an \u0027smp_mb()\u0027 to ensure that\n our check of the count is ordered after completion of our prior TLB\n invalidation.\n \n This also makes our -\u003eattach() and -\u003edetach() flows symmetric wrt ATS\n interactions.\n \n Acked-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit ee4ea764ea03253016034603cab352c0798ce173\nAuthor: Hou Zhiqiang \u003cZhiqiang.Hou@nxp.com\u003e\nDate: Tue Aug 20 07:28:43 2019 +0000\n\n dt-bindings: PCI: designware: Remove the num-lanes from Required properties\n \n The num-lanes is not a mandatory property, e.g. on FSL\n Layerscape SoCs, the PCIe link training is completed\n automatically based on the selected SerDes protocol, it\n does not need the num-lanes to set-up the link width.\n \n Currently it is both a Required and Optional property,\n let\u0027s remove it from the Required properties.\n \n Signed-off-by: Hou Zhiqiang \u003cZhiqiang.Hou@nxp.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Reviewed-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n\ncommit d971e28e2ce4696fcc32998c8aced5e47701fffe\nAuthor: Yazen Ghannam \u003cyazen.ghannam@amd.com\u003e\nDate: Wed Aug 21 23:59:55 2019 +0000\n\n EDAC/amd64: Support more than two controllers for chip selects handling\n \n The struct chip_select array that\u0027s used for saving chip select bases\n and masks is fixed at length of two. There should be one struct\n chip_select for each controller, so this array should be increased to\n support systems that may have more than two controllers.\n \n Increase the size of the struct chip_select array to eight, which is the\n largest number of controllers per die currently supported on AMD\n systems.\n \n Fix number of DIMMs and Chip Select bases/masks on Family17h, because\n AMD Family 17h systems support 2 DIMMs, 4 CS bases, and 2 CS masks per\n channel.\n \n Also, carve out the Family 17h+ reading of the bases/masks into a\n separate function. This effectively reverts the original bases/masks\n reading code to before Family 17h support was added.\n \n Signed-off-by: Yazen Ghannam \u003cyazen.ghannam@amd.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Cc: \"linux-edac@vger.kernel.org\" \u003clinux-edac@vger.kernel.org\u003e\n Cc: James Morse \u003cjames.morse@arm.com\u003e\n Cc: Mauro Carvalho Chehab \u003cmchehab@kernel.org\u003e\n Cc: Tony Luck \u003ctony.luck@intel.com\u003e\n Link: https://lkml.kernel.org/r/20190821235938.118710-2-Yazen.Ghannam@amd.com\n\ncommit cdfee5623290bc893f595636b44fa28e8207c5b3\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Fri Aug 16 08:24:35 2019 +0200\n\n driver core: initialize a default DMA mask for platform device\n \n We still treat devices without a DMA mask as defaulting to 32-bits for\n both mask, but a few releases ago we\u0027ve started warning about such\n cases, as they require special cases to work around this sloppyness.\n Add a dma_mask field to struct platform_device so that we can initialize\n the dma_mask pointer in struct device and initialize both masks to\n 32-bits by default, replacing similar functionality in m68k and\n powerpc. The arch_setup_pdev_archdata hooks is now unused and removed.\n \n Note that the code looks a little odd with the various conditionals\n because we have to support platform_device structures that are\n statically allocated.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Acked-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n Link: https://lore.kernel.org/r/20190816062435.881-7-hch@lst.de\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 383035211c79d4d98481a09ad429b31c7dbf22bd\nAuthor: Tony Camuso \u003ctcamuso@redhat.com\u003e\nDate: Thu Aug 22 08:24:53 2019 -0400\n\n ipmi: move message error checking to avoid deadlock\n \n V1-\u003eV2: in handle_one_rcv_msg, if data_size \u003e 2, set requeue to zero and\n goto out instead of calling ipmi_free_msg.\n Kosuke Tatsukawa \u003ctatsu@ab.jp.nec.com\u003e\n \n In the source stack trace below, function set_need_watch tries to\n take out the same si_lock that was taken earlier by ipmi_thread.\n \n ipmi_thread() [drivers/char/ipmi/ipmi_si_intf.c:995]\n smi_event_handler() [drivers/char/ipmi/ipmi_si_intf.c:765]\n handle_transaction_done() [drivers/char/ipmi/ipmi_si_intf.c:555]\n deliver_recv_msg() [drivers/char/ipmi/ipmi_si_intf.c:283]\n ipmi_smi_msg_received() [drivers/char/ipmi/ipmi_msghandler.c:4503]\n intf_err_seq() [drivers/char/ipmi/ipmi_msghandler.c:1149]\n smi_remove_watch() [drivers/char/ipmi/ipmi_msghandler.c:999]\n set_need_watch() [drivers/char/ipmi/ipmi_si_intf.c:1066]\n \n Upstream commit e1891cffd4c4896a899337a243273f0e23c028df adds code to\n ipmi_smi_msg_received() to call smi_remove_watch() via intf_err_seq()\n and this seems to be causing the deadlock.\n \n commit e1891cffd4c4896a899337a243273f0e23c028df\n Author: Corey Minyard \u003ccminyard@mvista.com\u003e\n Date: Wed Oct 24 15:17:04 2018 -0500\n ipmi: Make the smi watcher be disabled immediately when not needed\n \n The fix is to put all messages in the queue and move the message\n checking code out of ipmi_smi_msg_received and into handle_one_recv_msg,\n which processes the message checking after ipmi_thread releases its\n locks.\n \n Additionally,Kosuke Tatsukawa \u003ctatsu@ab.jp.nec.com\u003e reported that\n handle_new_recv_msgs calls ipmi_free_msg when handle_one_rcv_msg returns\n zero, so that the call to ipmi_free_msg in handle_one_rcv_msg introduced\n another panic when \"ipmitool sensor list\" was run in a loop. He\n submitted this part of the patch.\n \n +free_msg:\n + requeue \u003d 0;\n + goto out;\n \n Reported by: Osamu Samukawa \u003cosa-samukawa@tg.jp.nec.com\u003e\n Characterized by: Kosuke Tatsukawa \u003ctatsu@ab.jp.nec.com\u003e\n Signed-off-by: Tony Camuso \u003ctcamuso@redhat.com\u003e\n Fixes: e1891cffd4c4 (\"ipmi: Make the smi watcher be disabled immediately when not needed\")\n Cc: stable@vger.kernel.org # 5.1\n Signed-off-by: Corey Minyard \u003ccminyard@mvista.com\u003e\n\ncommit c4436c9149c5d2bc0c49ab57ec85c75ea1c4d61c\nAuthor: Kamlakant Patel \u003ckamlakantp@marvell.com\u003e\nDate: Wed Aug 21 12:04:33 2019 +0000\n\n ipmi_ssif: avoid registering duplicate ssif interface\n \n It is possible that SSIF interface entry is present in both DMI and ACPI\n tables. In SMP systems, in such cases it is possible that ssif_probe could\n be called simultaneously from i2c interface (from ACPI) and from DMI on\n different CPUs at kernel boot. Both try to register same SSIF interface\n simultaneously and result in race.\n \n In such cases where ACPI and SMBIOS both IPMI entries are available, we\n need to prefer ACPI over SMBIOS so that ACPI functions work properly if\n they use IPMI.\n So, if we get an ACPI interface and have already registered an SMBIOS\n at the same address, we need to remove the SMBIOS one and add the ACPI.\n \n Log:\n [ 38.774743] ipmi device interface\n [ 38.805006] ipmi_ssif: IPMI SSIF Interface driver\n [ 38.861979] ipmi_ssif i2c-IPI0001:06: ssif_probe CPU 99 ***\n [ 38.863655] ipmi_ssif 0-000e: ssif_probe CPU 14 ***\n [ 38.863658] ipmi_ssif: Trying SMBIOS-specified SSIF interface at i2c address 0xe, adapter xlp9xx-i2c, slave address 0x0\n [ 38.869500] ipmi_ssif: Trying ACPI-specified SSIF interface at i2c address 0xe, adapter xlp9xx-i2c, slave address 0x0\n [ 38.914530] ipmi_ssif: Unable to clear message flags: -22 7 c7\n [ 38.952429] ipmi_ssif: Unable to clear message flags: -22 7 00\n [ 38.994734] ipmi_ssif: Error getting global enables: -22 7 00\n [ 39.015877] ipmi_ssif 0-000e: IPMI message handler: Found new BMC (man_id: 0x00b3d1, prod_id: 0x0001, dev_id: 0x20)\n [ 39.377645] ipmi_ssif i2c-IPI0001:06: IPMI message handler: Found new BMC (man_id: 0x00b3d1, prod_id: 0x0001, dev_id: 0x20)\n [ 39.387863] ipmi_ssif 0-000e: IPMI message handler: BMC returned incorrect response, expected netfn 7 cmd 42, got netfn 7 cmd 1\n ...\n [NOTE] : Added custom prints to explain the problem.\n \n In the above log, ssif_probe is executed simultaneously on two different\n CPUs.\n \n This patch fixes this issue in following way:\n - Adds ACPI entry also to the \u0027ssif_infos\u0027 list.\n - Checks the list if SMBIOS is already registered, removes it and adds\n ACPI.\n - If ACPI is already registered, it ignores SMBIOS.\n - Adds mutex lock throughout the probe process to avoid race.\n \n Signed-off-by: Kamlakant Patel \u003ckamlakantp@marvell.com\u003e\n Message-Id: \u003c1566389064-27356-1-git-send-email-kamlakantp@marvell.com\u003e\n Signed-off-by: Corey Minyard \u003ccminyard@mvista.com\u003e\n\ncommit 528a25d040bc2123fe5a9f8697a60f8bff40af0a\nAuthor: Julien Masson \u003cjmasson@baylibre.com\u003e\nDate: Thu Aug 22 16:43:41 2019 +0200\n\n drm: meson: use match data to detect vpu compatibility\n \n This patch introduce new enum which contains all VPU family (GXBB,\n GXL, GXM and G12A).\n This enum is used to detect the VPU compatible with the device.\n \n We only need to set .data to the corresponding enum in the device\n table, no need to check .compatible string anymore.\n \n Signed-off-by: Julien Masson \u003cjmasson@baylibre.com\u003e\n Tested-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Acked-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/87imqpz21w.fsf@masson.i-did-not-set--mail-host-address--so-tickle-me\n\ncommit 69ecd44d68a7bf4caeda39825af720362db69233\nAuthor: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\nDate: Mon Aug 12 18:23:26 2019 +0200\n\n HID: multitouch: add support for the Smart Tech panel\n \n This panel is not very friendly to us:\n it exposes multiple multitouch collections, some of them being of\n logical application stylus.\n \n Usually, a device has only one report per application, and that is\n what I assumed in commit 8dfe14b3b47f (\"HID: multitouch: ditch mt_report_id\")\n \n To avoid breaking all working device, add a new class and a new quirk\n for that situation.\n \n Reported-and-tested-by: Matthias Fend \u003cMatthias.Fend@wolfvision.net\u003e\n Signed-off-by: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\n\ncommit c23e2043d5f7177c3f40da70663b2f3a1de82f5c\nAuthor: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\nDate: Mon Aug 12 18:23:25 2019 +0200\n\n HID: multitouch: do not filter mice nodes\n \n It was a good idea at the time to not create a mouse node for the\n multitouch touchscreens, but:\n - touchscreens following the Win 8 protocol should not have this\n disturbing mouse node anymore, or if they have, it should be\n used for something else (like a joystick attached to the screen)\n - touchpads have it, and they should not use it unless there is a bug,\n but when the laptop has a trackstick, the data are reported through this\n mouse node.\n \n So instead of whitelisting all of the devices that have a need for the\n mouse node, just export it.\n hid-input.c will append a suffix to it (\u0027Mouse\u0027), so users will eventually\n see if something goes wrong.\n \n Signed-off-by: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\n\ncommit 87fcb6a69e54527ccaeb5878168cccdba9f6b6ae\nAuthor: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\nDate: Mon Aug 12 18:27:39 2019 +0200\n\n HID: do not call hid_set_drvdata(hdev, NULL) in drivers\n \n This is a common pattern in the HID drivers to reset the drvdata. Some\n do it properly, some do it only in case of failure.\n \n But, this is actually already handled by driver core, so there is no need\n to do it manually.\n \n [for hid-sensor-hub.c]\n Acked-by: Srinivas Pandruvada \u003csrinivas.pandruvada@linux.intel.com\u003e\n [For hid-picolcd_core.c]\n Acked-by: Bruno Prémont \u003cbonbons@linux-vserver.org\u003e\n Signed-off-by: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\n\ncommit 362c571b92e2fa734c1f2c1e303d1dc27f12ec95\nAuthor: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\nDate: Mon Aug 12 18:27:40 2019 +0200\n\n HID: wacom: do not call hid_set_drvdata(hdev, NULL)\n \n This is a common pattern in the HID drivers to reset the drvdata.\n However, this is actually already handled by driver core, so there\n is no need to do it manually.\n \n Signed-off-by: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\n Acked-by: Jason Gerecke \u003cjason.gerecke@wacom.com\u003e\n\ncommit fe59fbd5dba9ae3bc94fc3d7d8556fe1e8707b61\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Thu Aug 22 11:33:47 2019 +0530\n\n arm64: defconfig: Enable SM8150 GCC and pinctrl driver\n \n Enable GCC config CONFIG_SM_GCC_8150 and pinctrl config\n CONFIG_PINCTRL_SM8150 to make it possible to boot the SM8150 MTP.\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit b4df75de3b3930703415aa053a269ae07c78d9b2\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Thu Aug 22 13:11:40 2019 +0200\n\n libperf: Move perf\u0027s cpu_map__idx() to perf_cpu_map__idx()\n \n As an internal function that will be used by both perf and libperf, but\n is not exported at this point.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190822111141.25823-5-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 315c0a1f0ccdd44c65f80ccbc62202fed8a23050\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Thu Aug 22 13:11:39 2019 +0200\n\n libperf: Move perf\u0027s cpu_map__empty() to perf_cpu_map__empty()\n \n So it\u0027s part of the libperf library as one of basic functions operating\n on the perf_cpu_map class.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190822111141.25823-4-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 6549cd8f2cc2cdf7e107fbbc3a68ecefb774bb2f\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Thu Aug 22 13:11:38 2019 +0200\n\n perf tools: Use perf_cpu_map__nr instead of cpu_map__nr\n \n Switch the rest of the perf code to use libperf\u0027s perf_cpu_map__nr(),\n which is the same as current cpu_map__nr() and remove the cpu_map__nr()\n function.\n \n Link: http://lkml.kernel.org/n/tip-6e0guy75clis7nm0xpuz9fga@git.kernel.org\n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190822111141.25823-3-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 3bfa00ef4aef7c62320ee3f0a691c9ac8c6e3339\nAuthor: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nDate: Fri Aug 9 17:07:39 2019 +0300\n\n btrfs: fix errror pointer check in __btrfs_map_block()\n \n The btrfs_get_chunk_map() never returns NULL, it returns error pointers.\n \n Fixes: 89b798ad1b42 (\"btrfs: Use btrfs_get_io_geometry appropriately\")\n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Signed-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit db9a5fd02a06113848fd3eabe302f56059d27366\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Thu Aug 22 13:11:37 2019 +0200\n\n tools headers: Add missing perf_event.h include\n \n We need perf_event.h include for \u0027struct perf_event_mmap_page\u0027.\n \n Link: http://lkml.kernel.org/n/tip-bolqkmqajexhccjb0ib0an8w@git.kernel.org\n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190822111141.25823-2-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit f836b27ecad98770bc53600d1ebd6533c921a4e7\nAuthor: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\nDate: Mon Aug 19 10:18:28 2019 -0400\n\n NFS: Have nfs4_proc_get_lease_time() call nfs4_call_sync_custom()\n \n This removes some code duplication, since both functions were doing the\n same thing.\n \n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit cc15e24a3af249455158bdb2ad02b9685eb4d6e9\nAuthor: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\nDate: Wed Aug 14 16:22:31 2019 -0400\n\n NFS: Have nfs41_proc_secinfo_no_name() call nfs4_call_sync_custom()\n \n We need to use the custom rpc_task_setup here to set the\n RPC_TASK_NO_ROUND_ROBIN flag on the RPC call.\n \n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit 4c952e3d1b0dc1835fda7f191b6f92e28b2fa435\nAuthor: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\nDate: Wed Aug 14 15:46:48 2019 -0400\n\n NFS: Have nfs41_proc_reclaim_complete() call nfs4_call_sync_custom()\n \n An async call followed by an rpc_wait_for_completion() is basically the\n same as a synchronous call, so we can use nfs4_call_sync_custom() to\n keep our custom callback ops and the RPC_TASK_NO_ROUND_ROBIN flag.\n \n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit 50493364e784b6e4077e41ab10eba4f798d13d9a\nAuthor: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\nDate: Wed Aug 14 15:30:16 2019 -0400\n\n NFS: Have _nfs4_proc_secinfo() call nfs4_call_sync_custom()\n \n We do this to set the RPC_TASK_NO_ROUND_ROBIN flag in the task_setup\n structure\n \n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit dae40965d51e563603222aa8d4fad7ab3cec0e2d\nAuthor: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\nDate: Wed Aug 14 15:28:28 2019 -0400\n\n NFS: Have nfs4_proc_setclientid() call nfs4_call_sync_custom()\n \n Rather than running the task manually\n \n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit 48c058543cbbb6b34114bbf8f0967e86dcd06b14\nAuthor: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\nDate: Wed Aug 14 15:27:00 2019 -0400\n\n NFS: Add an nfs4_call_sync_custom() function\n \n There are a few cases where we need to manually configure the\n rpc_task_setup structure to get the behavior we want.\n \n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit 251137b8fa97ef37ed7dadbed9de48475021fd8b\nAuthor: Qii Wang \u003cqii.wang@mediatek.com\u003e\nDate: Thu Aug 22 20:35:16 2019 +0800\n\n arm64: dts: mt8183: add I2C nodes\n \n Add i2c nodes to mt8183 and mt8183-evb.\n \n Signed-off-by: Qii Wang \u003cqii.wang@mediatek.com\u003e\n Signed-off-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n\ncommit c4e1da5e7d5fd32d00628ed0afd190e5cf891567\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Wed Aug 21 22:23:18 2019 -0500\n\n drm/amdgpu/powerplay: silence a warning in smu_v11_0_setup_pptable\n \n I think gcc is confused as I don\u0027t see how size could be used\n uninitialized, but go ahead and silence the warning.\n \n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d1916c86ccdcb67996278a850a22762102702d85\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Aug 12 17:39:58 2019 +0200\n\n block: move same page handling from __bio_add_pc_page to the callers\n \n Hiding page refcount manipulation inside a low-level bio helper is\n somewhat awkward. Instead return the same page information to the\n callers, where it fits in much better.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 384209cd5b93a926321fafe880ed05b1bca97260\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Aug 12 17:39:57 2019 +0200\n\n block: create a bio_try_merge_pc_page helper\n \n Passsthrough bio handling should be the same as normal bio handling,\n except that we need to take hardware limitations into account. Thus\n use the common try_merge implementation after checking the hardware\n limits. This changes behavior in that we now also check segment\n and dma boundary settings for same page merges, which is a little\n more work but has no effect as those need to be larger than the\n page size.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 320ea869a12cec206756207c6ca5f817ec45c7f2\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Aug 12 17:39:56 2019 +0200\n\n block: improve the gap check in __bio_add_pc_page\n \n If we can add more data into an existing segment we do not create a gap\n per definition, so move the check for a gap after the attempt to merge\n into the segment.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 7e04a46d84f73ef0f21d2257d6ba2a194c0f1511\nAuthor: Daniel Axtens \u003cdja@axtens.net\u003e\nDate: Thu Jun 27 15:30:08 2019 +1000\n\n powerpc/configs: Disable /dev/port in skiroot defconfig\n \n While reviewing lockdown patches, I discovered that we still enable\n /dev/port (CONFIG_DEVPORT) in skiroot.\n \n /dev/port is used for old x86 style IO accesses. It\u0027s set up in\n drivers/char/mem.c, and is only created if arch_has_dev_port() returns\n true. Per arch/powerpc/include/asm/io.h, on PPC64 with PCI, this is\n only true if there\u0027s a legacy ISA bridge.\n \n Even if a system has a legacy ISA bridge installed, we have no\n business accessing it in skiroot.\n \n Deselect CONFIG_DEVPORT for skiroot.\n \n Signed-off-by: Daniel Axtens \u003cdja@axtens.net\u003e\n [mpe: Incorporate emailed comments into the change log]\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190627053008.29315-1-dja@axtens.net\n\ncommit b8baa05a0e505f8311fb03f5c72e75fa0e13fd95\nAuthor: Gustavo Romero \u003cgromero@linux.ibm.com\u003e\nDate: Wed Aug 14 15:56:38 2019 -0500\n\n selftests/powerpc: Ignore generated files\n \n Currently some binary files which are generated when tests are compiled\n are not ignored by git, so \u0027git status\u0027 catch them.\n \n For copyloops test, fix wrong binary names already in .gitignore. For\n ptrace, security, and stringloops tests add missing binary names to the\n .gitignore file.\n \n Signed-off-by: Gustavo Romero \u003cgromero@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190814205638.25322-2-gromero@linux.ibm.com\n\ncommit 6278f55ba5ed20f486ea2049da0cf9a9c1d9a5d5\nAuthor: Gustavo Romero \u003cgromero@linux.ibm.com\u003e\nDate: Wed Aug 14 15:56:37 2019 -0500\n\n powerpc: Document xmon options\n \n Document all options currently supported by xmon debugger.\n \n Signed-off-by: Gustavo Romero \u003cgromero@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190814205638.25322-1-gromero@linux.ibm.com\n\ncommit 27d4396ed5a1c70f894557fd5732725eb1d94542\nAuthor: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\nDate: Fri Aug 16 14:48:16 2019 +1000\n\n powerpc/eeh: Slightly simplify eeh_add_to_parent_pe()\n \n Simplify some needlessly complicated boolean logic in\n eeh_add_to_parent_pe().\n \n Signed-off-by: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/09259a50308f10aa764695912bc87dc1d1cf654c.1565930772.git.sbobroff@linux.ibm.com\n\ncommit cef50c67c1d511bbbc974cead2bebeb6f83730ce\nAuthor: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\nDate: Fri Aug 16 14:48:15 2019 +1000\n\n powerpc/eeh: Remove unused return path from eeh_pe_dev_traverse()\n \n There are no users of the early-out return value from\n eeh_pe_dev_traverse(), so remove it.\n \n Signed-off-by: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/c648070f5b28fe8ca1880b48e64b267959ffd369.1565930772.git.sbobroff@linux.ibm.com\n\ncommit 2e25505147b8acf6510b9d5d951fd4c75f2e9bf2\nAuthor: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\nDate: Fri Aug 16 14:48:14 2019 +1000\n\n powerpc/eeh: Fix crash when edev-\u003epdev changes\n \n If a PCI device is removed during eeh_pe_report_edev(), between the\n calls to device_lock() and device_unlock(), edev-\u003epdev will change and\n cause a crash as the wrong mutex is released.\n \n To correct this, hold the PCI rescan/remove lock while taking a copy\n of edev-\u003epdev and performing a get_device() on it. Use this value to\n release the mutex, but also pass it through to the device driver\u0027s EEH\n handlers so that they always see the same device.\n \n Signed-off-by: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/3c590579a0faa24d20c826dcd26c739eb4d454e6.1565930772.git.sbobroff@linux.ibm.com\n\ncommit 1ff8f36fc770dd2b3eb294312f270db8cf94cc13\nAuthor: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\nDate: Fri Aug 16 14:48:13 2019 +1000\n\n powerpc/eeh: Convert log messages to eeh_edev_* macros\n \n Convert existing messages, where appropriate, to use the eeh_edev_*\n logging macros.\n \n The only effect should be minor adjustments to the log messages, apart\n from:\n \n - A new message in pseries_eeh_probe() \"Probing device\" to match the\n powernv case.\n - The \"Probing device\" message in pnv_eeh_probe() is now generated\n slightly later, which will mean that it is no longer emitted for\n devices that aren\u0027t probed due to the initial checks.\n \n Signed-off-by: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/ce505a0a7a4a5b0367f0f40f8b26e7c0a9cf4cb7.1565930772.git.sbobroff@linux.ibm.com\n\ncommit b093f2cbedfbaba6bf1c520fbfcb46403f3c7802\nAuthor: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\nDate: Fri Aug 16 14:48:12 2019 +1000\n\n powerpc/eeh: Introduce EEH edev logging macros\n \n Now that struct eeh_dev includes the BDFN of it\u0027s PCI device, make use\n of it to replace eeh_edev_info() with a set of dev_dbg()-style macros\n that only need a struct edev.\n \n With the BDFN available without the struct pci_dev, eeh_pci_name() is\n now unnecessary, so remove it.\n \n While only the \"info\" level function is used here, the others will be\n used in followup work.\n \n Signed-off-by: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/f90ae9a53d762be7b0ccbad79e62b5a1b4f4996e.1565930772.git.sbobroff@linux.ibm.com\n\ncommit 7c33a994d32d89937b23673e7507e8ec1daad893\nAuthor: Oliver O\u0027Halloran \u003coohall@gmail.com\u003e\nDate: Fri Aug 16 14:48:11 2019 +1000\n\n powerpc/eeh: Add bdfn field to eeh_dev\n \n Preparation for removing pci_dn from the powernv EEH code. The only\n thing we really use pci_dn for is to get the bdfn of the device for\n config space accesses, so adding that information to eeh_dev reduces\n the need to carry around the pci_dn.\n \n Signed-off-by: Oliver O\u0027Halloran \u003coohall@gmail.com\u003e\n [SB: Re-wrapped commit message, fixed whitespace damage.]\n Signed-off-by: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/e458eb69a1f591d8a120782f23a8506b15d3c654.1565930772.git.sbobroff@linux.ibm.com\n\ncommit c44e4ccadaca5884ac82b6dfffbd693bec3b583e\nAuthor: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\nDate: Fri Aug 16 14:48:10 2019 +1000\n\n powerpc/eeh: Refactor around eeh_probe_devices()\n \n Now that EEH support for all devices (on PowerNV and pSeries) is\n provided by the pcibios bus add device hooks, eeh_probe_devices() and\n eeh_addr_cache_build() are redundant and can be removed.\n \n Move the EEH enabled message into it\u0027s own function so that it can be\n called from multiple places.\n \n Note that previously on pSeries, useless EEH sysfs files were created\n for some devices that did not have EEH support and this change\n prevents them from being created.\n \n Signed-off-by: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/33b0a6339d5ac88693de092d6fba984f2a5add66.1565930772.git.sbobroff@linux.ibm.com\n\ncommit b905f8cdca7725e750a84f7188ea6821750124c3\nAuthor: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\nDate: Fri Aug 16 14:48:09 2019 +1000\n\n powerpc/eeh: EEH for pSeries hot plug\n \n On PowerNV and pSeries, devices currently acquire EEH support from\n several different places: Boot-time devices from eeh_probe_devices()\n and eeh_addr_cache_build(), Virtual Function devices from the pcibios\n bus add device hooks and hot plugged devices from pci_hp_add_devices()\n (with other platforms using other methods as well). Unfortunately,\n pSeries machines currently discover hot plugged devices using\n pci_rescan_bus(), not pci_hp_add_devices(), and so those devices do\n not receive EEH support.\n \n Rather than adding another case for pci_rescan_bus(), this change\n widens the scope of the pcibios bus add device hooks so that they can\n handle all devices. As a side effect this also supports devices\n discovered after manually rescanning via /sys/bus/pci/rescan.\n \n Note that on PowerNV, this change allows the EEH subsystem to become\n enabled after boot as long as it has not been forced off, which was\n not previously possible (it was already possible on pSeries).\n \n Signed-off-by: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/72ae8ae9c54097158894a52de23690448de38ea9.1565930772.git.sbobroff@linux.ibm.com\n\ncommit 685a0bc00abcf1d40d160eaafab9989f565ab2b5\nAuthor: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\nDate: Fri Aug 16 14:48:08 2019 +1000\n\n powerpc/eeh: Initialize EEH address cache earlier\n \n The EEH address cache is currently initialized and populated by a\n single function: eeh_addr_cache_build(). While the initial population\n of the cache can only be done once resources are allocated,\n initialization (just setting up a spinlock) could be done much\n earlier.\n \n So move the initialization step into a separate function and call it\n from a core_initcall (rather than a subsys initcall).\n \n This will allow future work to make use of the cache during boot time\n PCI scanning.\n \n Signed-off-by: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\n Reviewed-by: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/0557206741bffee76cdfff042f65321f6f7a5b41.1565930772.git.sbobroff@linux.ibm.com\n\ncommit 617082a4817a4354fa3de05c80b5f6088e2083b7\nAuthor: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\nDate: Fri Aug 16 14:48:07 2019 +1000\n\n powerpc/eeh: Improve debug messages around device addition\n \n Also remove useless comment.\n \n Signed-off-by: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\n Reviewed-by: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/59db84f4bf94718a12f206bc923ac797d47e4cc1.1565930772.git.sbobroff@linux.ibm.com\n\ncommit aa06e3d60e245284d1e55497eb3108828092818d\nAuthor: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\nDate: Fri Aug 16 14:48:06 2019 +1000\n\n powerpc/eeh: Clear stale EEH_DEV_NO_HANDLER flag\n \n The EEH_DEV_NO_HANDLER flag is used by the EEH system to prevent the\n use of driver callbacks in drivers that have been bound part way\n through the recovery process. This is necessary to prevent later stage\n handlers from being called when the earlier stage handlers haven\u0027t,\n which can be confusing for drivers.\n \n However, the flag is set for all devices that are added after boot\n time and only cleared at the end of the EEH recovery process. This\n results in hot plugged devices erroneously having the flag set during\n the first recovery after they are added (causing their driver\u0027s\n handlers to be incorrectly ignored).\n \n To remedy this, clear the flag at the beginning of recovery\n processing. The flag is still cleared at the end of recovery\n processing, although it is no longer really necessary.\n \n Also clear the flag during eeh_handle_special_event(), for the same\n reasons.\n \n Signed-off-by: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/b8ca5629d27de74c957d4f4b250177d1b6fc4bbd.1565930772.git.sbobroff@linux.ibm.com\n\ncommit 3f068aae7a958555533847af88705b5629f31600\nAuthor: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\nDate: Fri Aug 16 14:48:05 2019 +1000\n\n powerpc/64: Adjust order in pcibios_init()\n \n The pcibios_init() function for PowerPC 64 currently calls\n pci_bus_add_devices() before pcibios_resource_survey(). This means\n that at boot time, when the pcibios_bus_add_device() hooks are called\n by pci_bus_add_devices(), device resources have not been allocated and\n they are unable to perform EEH setup, so a separate pass is needed.\n \n This patch adjusts that order so that it will become possible to\n consolidate the EEH setup work into a single location.\n \n The only functional change is to execute pcibios_resource_survey()\n (excepting ppc_md.pcibios_fixup(), see below) before\n pci_bus_add_devices() instead of after it.\n \n Because pcibios_scan_phb() and pci_bus_add_devices() are called\n together in a loop, this must be broken into one loop for each call.\n Then the call to pcibios_resource_survey() is moved up in between\n them. This changes the ordering but because pcibios_resource_survey()\n also calls ppc_md.pcibios_fixup(), that call is extracted out into\n pcibios_init() to where pcibios_resource_survey() was, so that it is\n not moved.\n \n The only other caller of pcibios_resource_survey() is the PowerPC 32\n version of pcibios_init(), and therefore, that is modified to call\n ppc_md.pcibios_fixup() right after pcibios_resource_survey() so that\n there is no functional change there at all.\n \n The re-arrangement will cause very few side-effects because at this\n stage in the boot, pci_bus_add_devices() does very little:\n - pci_create_sysfs_dev_files() does nothing (no sysfs yet)\n - pci_proc_attach_device() does nothing (no proc yet)\n - device_attach() does nothing (no drivers yet)\n This leaves only the pci_final_fixup calls, D3 support, and marking\n the device as added. Of those, only the pci_final_fixup calls have the\n potential to be affected by resource allocation.\n \n The only pci_final_fixup handlers that touch resources seem to be one\n for x86 (pci_amd_enable_64bit_bar()), and a PowerPC 32 platform driver\n (quirk_final_uli1575()), neither of which use this pcibios_init()\n function. Even if they did, it would almost certainly be a bug, under\n the current ordering, to rely on or make changes to resources before\n they were allocated.\n \n Signed-off-by: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\n Reviewed-by: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/4506b0489eabd0921a3587d90bd44c7683f3472d.1565930772.git.sbobroff@linux.ibm.com\n\ncommit 56347074c5307d7bca5db38eb2c764c64ae57514\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sat Jul 13 12:21:06 2019 +0900\n\n powerpc: remove meaningless KBUILD_ARFLAGS addition\n \n The KBUILD_ARFLAGS addition in arch/powerpc/Makefile has never worked\n in a useful way because it is always overridden by the following code\n in the top Makefile:\n \n # use the deterministic mode of AR if available\n KBUILD_ARFLAGS :\u003d $(call ar-option,D)\n \n The code in the top Makefile was added in 2011, by commit 40df759e2b9e\n (\"kbuild: Fix build with binutils \u003c\u003d 2.19\").\n \n The KBUILD_ARFLAGS addition for ppc has always been dead code from the\n beginning.\n \n Nobody has reported a problem since 43c9127d94d6 (\"powerpc: Add option\n to use thin archives\"), so this code was unneeded.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190713032106.8509-1-yamada.masahiro@socionext.com\n\ncommit be91233b1053d9725d3a437e3da44ccff2639477\nAuthor: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\nDate: Thu Aug 22 05:46:28 2019 -0700\n\n drm/i915: Update DRIVER_DATE to 20190822\n \n Signed-off-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n\ncommit e31b791fae32d01e8d0ee2e731b74d5471293cba\nAuthor: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nDate: Thu Aug 22 22:09:15 2019 +1000\n\n crypto: talitos - Fix build warning in aead_des3_setkey\n \n This patch removes the variable flags which is now unused thanks\n to the new DES helpers.\n \n Fixes: 9d574ae8ebc1 (\"crypto: talitos/des - switch to new...\")\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit c963050259f2daccb7141ee58bc0fc0b760836ba\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 22 14:49:15 2019 +0300\n\n crypto: n2/des - fix build breakage after DES updates\n \n Fix build breakage caused by the DES library refactor.\n \n Fixes: d4b90dbc8578 (\"crypto: n2/des - switch to new verification routines\")\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 5a74362c9a8b122366dd8f2197fc6b784300d6a2\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 22 13:24:54 2019 +0300\n\n crypto: s390/aes - fix typo in XTS_BLOCK_SIZE identifier\n \n Fix a typo XTS_BLOCKSIZE -\u003e XTS_BLOCK_SIZE, causing the build to\n break.\n \n Fixes: ce68acbcb6a5 (\"crypto: s390/xts-aes - invoke fallback for...\")\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 170ae7ae4c9a19fd94b7e5c8de70bfc7c00facef\nAuthor: Chaotian Jing \u003cchaotian.jing@mediatek.com\u003e\nDate: Thu Aug 22 14:47:41 2019 +0800\n\n mmc: mediatek: support 24bits segment size\n \n MSDC IP which support 64G DRAM will support 24bits BDMA buffer length,\n so add support it.\n \n Signed-off-by: Chaotian Jing \u003cchaotian.jing@mediatek.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit dd7678ec164f6bb3676bfbac3afd9a267378443b\nAuthor: Chaotian Jing \u003cchaotian.jing@mediatek.com\u003e\nDate: Thu Aug 22 14:47:40 2019 +0800\n\n mmc: mediatek: fix controller busy when plug out SD\n \n when plug out SD card, may get data CRC error, the MMC core will issue\n CMD13 to get card status, then CMD13 timeout(due to card plug out) will\n trigger CMD19 tuning, first CMD19 timeout has not call msdc_reset_hw()\n and cause the next CMD19 gets controller busy.\n \n Signed-off-by: Chaotian Jing \u003cchaotian.jing@mediatek.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit 8e8d7127e547f20127cb2d0fc3041e19f36e44f7\nAuthor: Yoshihiro Shimoda \u003cyoshihiro.shimoda.uh@renesas.com\u003e\nDate: Tue Aug 20 17:43:10 2019 +0900\n\n mmc: renesas_sdhi_sys_dmac: Remove all R-Car Gen3 SoCs\n \n All R-Car Gen3 SoCs with any ES version cannot use this DMAC actually.\n So, this patch removes the compatibles of R-Car Gen3 SoCs from\n renesas_sdhi_sys_dmac_of_match. Since the previous code has\n an empty whitelist to prevent probing of R-Car Gen3 SoCs,\n no behavior changes.\n \n Signed-off-by: Yoshihiro Shimoda \u003cyoshihiro.shimoda.uh@renesas.com\u003e\n Reviewed-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Tested-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit 6615262c1c88532dde21b7004378a740d15da175\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Sat Aug 10 14:16:08 2019 +0200\n\n mmc: jz4740: Drop dependency on arch header\n \n We don\u0027t need to set the \u0027slave_id\u0027 anymore - that field is never read\n by the DMA driver.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit d2b0ed87fd14e48fa050a14b9404837910862f17\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Sat Aug 10 14:16:07 2019 +0200\n\n mmc: jz4740: Code cleanup\n \n Fix wrong code indentation which made the code hard to read, and fix\n return with value in void function.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit bde197cd538f924049f0e68a6cb83d62082a28b4\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Thu Aug 8 11:53:01 2019 -0500\n\n mmc: sdhci-cadence: use struct_size() helper\n \n One of the more common cases of allocation size calculations is finding\n the size of a structure that has a zero-sized array at the end, along\n with memory for some number of elements for that array. For example:\n \n struct sdhci_cdns_priv {\n ...\n struct sdhci_cdns_phy_param phy_params[0];\n };\n \n Make use of the struct_size() helper instead of an open-coded version\n in order to avoid any potential type mistakes.\n \n So, replace the following form:\n \n sizeof(*priv) + sizeof(priv-\u003ephy_params[0]) * nr_phy_params\n \n with:\n \n struct_size(priv, phy_params, nr_phy_params)\n \n Also, notice that, in this case, variable priv_size is not necessary,\n hence it is removed.\n \n This code was detected with the help of Coccinelle.\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Acked-by: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit d98873dedd4182f3fede5839fe68809aa28d62c7\nAuthor: Yinbo Zhu \u003cyinbo.zhu@nxp.com\u003e\nDate: Wed Aug 14 15:26:49 2019 +0800\n\n mmc: sdhci-of-esdhc: add erratum A011334 support in ls1028a 1.0 SoC\n \n This patch is to add erratum A011334 support in ls1028a 1.0 SoC\n \n Signed-off-by: Yinbo Zhu \u003cyinbo.zhu@nxp.com\u003e\n Acked-by: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit e1b832503e8f29ea6e20c30db9c3176576c0fc78\nAuthor: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nDate: Wed Aug 21 17:47:29 2019 +0100\n\n arm64: Define Documentation/arm64/tagged-address-abi.rst\n \n On AArch64 the TCR_EL1.TBI0 bit is set by default, allowing userspace\n (EL0) to perform memory accesses through 64-bit pointers with a non-zero\n top byte. Introduce the document describing the relaxation of the\n syscall ABI that allows userspace to pass certain tagged pointers to\n kernel syscalls.\n \n Cc: Will Deacon \u003cwill.deacon@arm.com\u003e\n Cc: Szabolcs Nagy \u003cszabolcs.nagy@arm.com\u003e\n Acked-by: Kevin Brodsky \u003ckevin.brodsky@arm.com\u003e\n Acked-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\n Signed-off-by: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\n Co-developed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 3b2bf49e602e5411a3634cd89c0e2770ee1ee12e\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 22 07:59:17 2019 +0100\n\n drm/i915: Kill the undead i915_gem_batch_pool.c\n \n You have to cut it off at the neck, otherwise it just reappears in the\n next merge, like it did in commit 3f866026f0ce (\"Merge drm/drm-next\n into drm-intel-next-queued\")\n \n References: 3f866026f0ce (\"Merge drm/drm-next into drm-intel-next-queued\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Acked-by: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190822065917.18988-1-chris@chris-wilson.co.uk\n\ncommit cac33c104a1007fb679011807c57553ac1288efe\nAuthor: Hsin-Hsiung Wang \u003chsin-hsiung.wang@mediatek.com\u003e\nDate: Thu Aug 22 16:55:40 2019 +0800\n\n arm64: dts: mt8183: fix pwrap gic number\n \n The correct gic number of pwrap is 185 instead of 209. This patch fixes\n it to avoid triggering error interrupt.\n \n Fixes: e526c9bc11f8 (\"arm64: dts: Add Mediatek SoC MT8183 and evaluation board dts and Makefile\")\n \n Signed-off-by: Hsin-Hsiung Wang \u003chsin-hsiung.wang@mediatek.com\u003e\n Signed-off-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n\ncommit cc212241df0b8975bb0e6d7f9028405a9c664e49\nAuthor: Ryder Lee \u003cryder.lee@mediatek.com\u003e\nDate: Thu Jul 11 10:14:23 2019 +0800\n\n arm: dts: mediatek: add basic support for MT7629 SoC\n \n This adds basic support for MT7629 reference board.\n \n Tested-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Ryder Lee \u003cryder.lee@mediatek.com\u003e\n Signed-off-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n\ncommit b8e4f1fdfa422398c2d6c47bfb7d1feb3046d70a\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Thu Aug 22 09:25:27 2019 +0200\n\n ALSA: usb-audio: Remove superfluous bLength checks\n \n Now that we got the more comprehensive validation code for USB-audio\n descriptors, the check of overflow in each descriptor unit parser\n became superfluous. Drop some of the obvious cases.\n \n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 52c3e317a857091fd746e15179a637f32be4d337\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Thu Aug 22 08:23:10 2019 +0200\n\n ALSA: usb-audio: Unify the release of usb_mixer_elem_info objects\n \n Instead of the direct kfree() calls, introduce a new local helper to\n release the usb_mixer_elem_info object. This will be extended to do\n more than a single kfree() in the later patches.\n \n Also, use the standard goto instead of multiple calls in\n parse_audio_selector_unit() error paths.\n \n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 68e9fde245591d18200f8a9054cac22339437adb\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Thu Aug 15 16:30:39 2019 +0200\n\n ALSA: usb-audio: Simplify parse_audio_unit()\n \n Minor code refactoring by combining the UAC version and the type in\n the switch-case flow, so that we reduce the indentation and\n redundancy. One good bonus is that the duplicated definition of the\n same type value (e.g. UAC2_EFFECT_UNIT) can be handled more cleanly.\n \n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 57f8770620e9b51c61089751f0b5ad3dbe376ff2\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Tue Aug 20 17:17:09 2019 +0200\n\n ALSA: usb-audio: More validations of descriptor units\n \n Introduce a new helper to validate each audio descriptor unit before\n and check the unit before actually accessing it. This should harden\n against the OOB access cases with malformed descriptors that have been\n recently frequently reported by fuzzers.\n \n The existing descriptor checks are still kept although they become\n superfluous after this patch. They\u0027ll be cleaned up eventually\n later.\n \n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit b6852ae75d4be5d3398a4a501443de221151fa9b\nAuthor: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\nDate: Thu Aug 15 13:09:31 2019 -0700\n\n KVM: VMX: Fix and tweak the comments for VM-Enter\n \n Fix an incorrect/stale comment regarding the vmx_vcpu pointer, as guest\n registers are now loaded using a direct pointer to the start of the\n register array.\n \n Opportunistically add a comment to document why the vmx_vcpu pointer is\n needed, its consumption via \u0027call vmx_update_host_rsp\u0027 is rather subtle.\n \n Signed-off-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\n Reviewed-by: Jim Mattson \u003cjmattson@google.com\u003e\n Signed-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n\ncommit 12b58f4ed2a1b93f0eae84f308f6e0a143a533d4\nAuthor: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\nDate: Thu Aug 15 10:22:37 2019 -0700\n\n KVM: Assert that struct kvm_vcpu is always as offset zero\n \n KVM implementations that wrap struct kvm_vcpu with a vendor specific\n struct, e.g. struct vcpu_vmx, must place the vcpu member at offset 0,\n otherwise the usercopy region intended to encompass struct kvm_vcpu_arch\n will instead overlap random chunks of the vendor specific struct.\n E.g. padding a large number of bytes before struct kvm_vcpu triggers\n a usercopy warn when running with CONFIG_HARDENED_USERCOPY\u003dy.\n \n Signed-off-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\n Reviewed-by: Jim Mattson \u003cjmattson@google.com\u003e\n Signed-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n\ncommit b382f44e98506bcb00acada0e30151a73e782a93\nAuthor: Wanpeng Li \u003cwanpengli@tencent.com\u003e\nDate: Mon Aug 5 10:03:24 2019 +0800\n\n KVM: X86: Add pv tlb shootdown tracepoint\n \n Add pv tlb shootdown tracepoint.\n \n Cc: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n Cc: Radim Krčmář \u003crkrcmar@redhat.com\u003e\n Signed-off-by: Wanpeng Li \u003cwanpengli@tencent.com\u003e\n Signed-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n\ncommit 92735b1b3313c97d24b54492a21083c75206e6ed\nAuthor: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\nDate: Fri Aug 2 15:06:17 2019 -0700\n\n KVM: x86: Unconditionally call x86 ops that are always implemented\n \n Remove a few stale checks for non-NULL ops now that the ops in question\n are implemented by both VMX and SVM.\n \n Note, this is **not** stable material, the Fixes tags are there purely\n to show when a particular op was first supported by both VMX and SVM.\n \n Fixes: 74f169090b6f (\"kvm/svm: Setup MCG_CAP on AMD properly\")\n Fixes: b31c114b82b2 (\"KVM: X86: Provide a capability to disable PAUSE intercepts\")\n Fixes: 411b44ba80ab (\"svm: Implements update_pi_irte hook to setup posted interrupt\")\n Cc: Krish Sadhukhan \u003ckrish.sadhukhan@oracle.com\u003e\n Signed-off-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\n Signed-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n\ncommit 26c44a63a291893e0a00f01e96b6e1d0310a79a9\nAuthor: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\nDate: Thu Aug 1 13:35:23 2019 -0700\n\n KVM: x86/mmu: Consolidate \"is MMIO SPTE\" code\n \n Replace the open-coded \"is MMIO SPTE\" checks in the MMU warnings\n related to software-based access/dirty tracking to make the code\n slightly more self-documenting.\n \n No functional change intended.\n \n Signed-off-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\n Signed-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n\ncommit 4af7715110a2617fc40ac2c1232f664019269f3a\nAuthor: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\nDate: Thu Aug 1 13:35:22 2019 -0700\n\n KVM: x86/mmu: Add explicit access mask for MMIO SPTEs\n \n When shadow paging is enabled, KVM tracks the allowed access type for\n MMIO SPTEs so that it can do a permission check on a MMIO GVA cache hit\n without having to walk the guest\u0027s page tables. The tracking is done\n by retaining the WRITE and USER bits of the access when inserting the\n MMIO SPTE (read access is implicitly allowed), which allows the MMIO\n page fault handler to retrieve and cache the WRITE/USER bits from the\n SPTE.\n \n Unfortunately for EPT, the mask used to retain the WRITE/USER bits is\n hardcoded using the x86 paging versions of the bits. This funkiness\n happens to work because KVM uses a completely different mask/value for\n MMIO SPTEs when EPT is enabled, and the EPT mask/value just happens to\n overlap exactly with the x86 WRITE/USER bits[*].\n \n Explicitly define the access mask for MMIO SPTEs to accurately reflect\n that EPT does not want to incorporate any access bits into the SPTE, and\n so that KVM isn\u0027t subtly relying on EPT\u0027s WX bits always being set in\n MMIO SPTEs, e.g. attempting to use other bits for experimentation breaks\n horribly.\n \n Note, vcpu_match_mmio_gva() explicits prevents matching GVA\u003d\u003d0, and all\n TDP flows explicit set mmio_gva to 0, i.e. zeroing vcpu-\u003earch.access for\n EPT has no (known) functional impact.\n \n [*] Using WX to generate EPT misconfigurations (equivalent to reserved\n bit page fault) ensures KVM can employ its MMIO page fault tricks\n even platforms without reserved address bits.\n \n Fixes: ce88decffd17 (\"KVM: MMU: mmio page fault support\")\n Signed-off-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\n Signed-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n\ncommit 871bd0346018df53055141f09754cb5ffb334c7b\nAuthor: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\nDate: Thu Aug 1 13:35:21 2019 -0700\n\n KVM: x86: Rename access permissions cache member in struct kvm_vcpu_arch\n \n Rename \"access\" to \"mmio_access\" to match the other MMIO cache members\n and to make it more obvious that it\u0027s tracking the access permissions\n for the MMIO cache.\n \n Signed-off-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\n Signed-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n\ncommit c8e16b78c6142afea428dc316c900644951cf7f3\nAuthor: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\nDate: Tue Aug 13 15:53:35 2019 +0200\n\n x86: KVM: svm: eliminate hardcoded RIP advancement from vmrun_interception()\n \n Just like we do with other intercepts, in vmrun_interception() we should be\n doing kvm_skip_emulated_instruction() and not just RIP +\u003d 3. Also, it is\n wrong to increment RIP before nested_svm_vmrun() as it can result in\n kvm_inject_gp().\n \n We can\u0027t call kvm_skip_emulated_instruction() after nested_svm_vmrun() so\n move it inside.\n \n Suggested-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\n Signed-off-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\n Reviewed-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\n Signed-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n\ncommit e7134c1bb5af1ca6dcbd983e036e67b82e6a2de4\nAuthor: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\nDate: Tue Aug 13 15:53:34 2019 +0200\n\n x86: KVM: svm: eliminate weird goto from vmrun_interception()\n \n Regardless of whether or not nested_svm_vmrun_msrpm() fails, we return 1\n from vmrun_interception() so there\u0027s no point in doing goto. Also,\n nested_svm_vmrun_msrpm() call can be made from nested_svm_vmrun() where\n other nested launch issues are handled.\n \n nested_svm_vmrun() returns a bool, however, its result is ignored in\n vmrun_interception() as we always return \u00271\u0027. As a preparatory change\n to putting kvm_skip_emulated_instruction() inside nested_svm_vmrun()\n make nested_svm_vmrun() return an int (always \u00271\u0027 for now).\n \n Suggested-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\n Signed-off-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\n Reviewed-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\n Signed-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n\ncommit c4762fdab55cb1ad715c427f875a549ef257dbb9\nAuthor: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\nDate: Tue Aug 13 15:53:33 2019 +0200\n\n x86: KVM: svm: remove hardcoded instruction length from intercepts\n \n Various intercepts hard-code the respective instruction lengths to optimize\n skip_emulated_instruction(): when next_rip is pre-set we skip\n kvm_emulate_instruction(vcpu, EMULTYPE_SKIP). The optimization is, however,\n incorrect: different (redundant) prefixes could be used to enlarge the\n instruction. We can\u0027t really avoid decoding.\n \n svm-\u003enext_rip is not used when CPU supports \u0027nrips\u0027 (X86_FEATURE_NRIPS)\n feature: next RIP is provided in VMCB. The feature is not really new\n (Opteron G3s had it already) and the change should have zero affect.\n \n Remove manual svm-\u003enext_rip setting with hard-coded instruction lengths.\n The only case where we now use svm-\u003enext_rip is EXIT_IOIO: the instruction\n length is provided to us by hardware.\n \n Hardcoded RIP advancement remains in vmrun_interception(), this is going to\n be taken care of separately.\n \n Reported-by: Jim Mattson \u003cjmattson@google.com\u003e\n Reviewed-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\n Signed-off-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\n Signed-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n\ncommit 02d4160fbd765110e836c84c30c7babb5af84a44\nAuthor: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\nDate: Tue Aug 13 15:53:32 2019 +0200\n\n x86: KVM: add xsetbv to the emulator\n \n To avoid hardcoding xsetbv length to \u00273\u0027 we need to support decoding it in\n the emulator.\n \n Signed-off-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\n Signed-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n\ncommit 97413d2981bb3619a9d39f0b611318664d7231f5\nAuthor: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\nDate: Tue Aug 13 15:53:31 2019 +0200\n\n x86: KVM: clear interrupt shadow on EMULTYPE_SKIP\n \n When doing x86_emulate_instruction(EMULTYPE_SKIP) interrupt shadow has to\n be cleared if and only if the skipping is successful.\n \n There are two immediate issues:\n - In SVM skip_emulated_instruction() we are not zapping interrupt shadow\n in case kvm_emulate_instruction(EMULTYPE_SKIP) is used to advance RIP\n (!nrpip_save).\n - In VMX handle_ept_misconfig() when running as a nested hypervisor we\n (static_cpu_has(X86_FEATURE_HYPERVISOR) case) forget to clear interrupt\n shadow.\n \n Note that we intentionally don\u0027t handle the case when the skipped\n instruction is supposed to prolong the interrupt shadow (\"MOV/POP SS\") as\n skip-emulation of those instructions should not happen under normal\n circumstances.\n \n Suggested-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\n Reviewed-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\n Signed-off-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\n Signed-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n\ncommit f8ea7c6049d5d0766b84c8107583ed0094773d06\nAuthor: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\nDate: Tue Aug 13 15:53:30 2019 +0200\n\n x86: kvm: svm: propagate errors from skip_emulated_instruction()\n \n On AMD, kvm_x86_ops-\u003eskip_emulated_instruction(vcpu) can, in theory,\n fail: in !nrips case we call kvm_emulate_instruction(EMULTYPE_SKIP).\n Currently, we only do printk(KERN_DEBUG) when this happens and this\n is not ideal. Propagate the error up the stack.\n \n On VMX, skip_emulated_instruction() doesn\u0027t fail, we have two call\n sites calling it explicitly: handle_exception_nmi() and\n handle_task_switch(), we can just ignore the result.\n \n On SVM, we also have two explicit call sites:\n svm_queue_exception() and it seems we don\u0027t need to do anything there as\n we check if RIP was advanced or not. In task_switch_interception(),\n however, we are better off not proceeding to kvm_task_switch() in case\n skip_emulated_instruction() failed.\n \n Suggested-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\n Signed-off-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\n Signed-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n\ncommit 05402f64544f054edbe41f45f6ec52dbe0c1a324\nAuthor: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\nDate: Tue Aug 13 15:53:29 2019 +0200\n\n x86: KVM: svm: don\u0027t pretend to advance RIP in case wrmsr_interception() results in #GP\n \n svm-\u003enext_rip is only used by skip_emulated_instruction() and in case\n kvm_set_msr() fails we rightfully don\u0027t do that. Move svm-\u003enext_rip\n advancement to \u0027else\u0027 branch to avoid creating false impression that\n it\u0027s always advanced (and make it look like rdmsr_interception()).\n \n This is a preparatory change to removing hardcoded RIP advancement\n from instruction intercepts, no functional change.\n \n Signed-off-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\n Reviewed-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\n Signed-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n\ncommit 016cd759642c5d313cb36e817362ade13f04ca94\nAuthor: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\nDate: Thu Aug 15 09:20:32 2019 -0700\n\n KVM: x86: Fix x86_decode_insn() return when fetching insn bytes fails\n \n Jump to the common error handling in x86_decode_insn() if\n __do_insn_fetch_bytes() fails so that its error code is converted to the\n appropriate return type. Although the various helpers used by\n x86_decode_insn() return X86EMUL_* values, x86_decode_insn() itself\n returns EMULATION_FAILED or EMULATION_OK.\n \n This doesn\u0027t cause a functional issue as the sole caller,\n x86_emulate_instruction(), currently only cares about success vs.\n failure, and success is indicated by \u00270\u0027 for both types\n (X86EMUL_CONTINUE and EMULATION_OK).\n \n Fixes: 285ca9e948fa (\"KVM: emulate: speed up do_insn_fetch\")\n Signed-off-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\n Signed-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n\ncommit 0c54914d0c52a15db9954a76ce80fee32cf318f4\nAuthor: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\nDate: Mon Aug 19 17:24:07 2019 +0200\n\n KVM: x86: use Intel speculation bugs and features as derived in generic x86 code\n \n Similar to AMD bits, set the Intel bits from the vendor-independent\n feature and bug flags, because KVM_GET_SUPPORTED_CPUID does not care\n about the vendor and they should be set on AMD processors as well.\n \n Suggested-by: Jim Mattson \u003cjmattson@google.com\u003e\n Reviewed-by: Jim Mattson \u003cjmattson@google.com\u003e\n Signed-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n\ncommit 50896de4be7e08fc47f857f6e0c0e9ab2654b941\nAuthor: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\nDate: Thu Aug 15 09:17:31 2019 +0200\n\n KVM: x86: always expose VIRT_SSBD to guests\n \n Even though it is preferrable to use SPEC_CTRL (represented by\n X86_FEATURE_AMD_SSBD) instead of VIRT_SPEC, VIRT_SPEC is always\n supported anyway because otherwise it would be impossible to\n migrate from old to new CPUs. Make this apparent in the\n result of KVM_GET_SUPPORTED_CPUID as well.\n \n However, we need to hide the bit on Intel processors, so move\n the setting to svm_set_supported_cpuid.\n \n Cc: Konrad Rzeszutek Wilk \u003ckonrad.wilk@oracle.com\u003e\n Reported-by: Eduardo Habkost \u003cehabkost@redhat.com\u003e\n Reviewed-by: Jim Mattson \u003cjmattson@google.com\u003e\n Signed-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n\ncommit 4c6903a0f9d76be8a907d8238a5e86c829c8954d\nAuthor: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\nDate: Wed Aug 14 12:07:34 2019 -0400\n\n KVM: x86: fix reporting of AMD speculation bug CPUID leaf\n \n The AMD_* bits have to be set from the vendor-independent\n feature and bug flags, because KVM_GET_SUPPORTED_CPUID does not care\n about the vendor and they should be set on Intel processors as well.\n On top of this, SSBD, STIBP and AMD_SSB_NO bit were not set, and\n VIRT_SSBD does not have to be added manually because it is a\n cpufeature that comes directly from the host\u0027s CPUID bit.\n \n Reviewed-by: Jim Mattson \u003cjmattson@google.com\u003e\n Signed-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n\ncommit 051c78af14fcd74a22b5af45548ad9d588247cc7\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Thu Aug 22 09:58:07 2019 +0200\n\n ALSA: hda/realtek - Blacklist PC beep for Lenovo ThinkCentre M73/93\n \n Lenovo ThinkCentre M73 and M93 don\u0027t seem to have a proper beep\n although the driver tries to probe and set up blindly.\n Blacklist these machines for suppressing the beep creation.\n \n BugLink: https://bugzilla.kernel.org/show_bug.cgi?id\u003d204635\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 2722b535337b35a9c0be50de687cccda4ab9468d\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Thu Aug 22 09:55:37 2019 +0200\n\n ALSA: hda/realtek - Check beep whitelist before assigning in all codecs\n \n Some Realtek codec parsers didn\u0027t check the availability of PC beep.\n Add has_cdefine_beep() check appropriately.\n \n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 1f7fd484fff1b432373f0d0682fb0e2015113161\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 22 07:15:57 2019 +0100\n\n drm/i915: Replace i915_vma_put_fence()\n \n Avoid calling i915_vma_put_fence() by using our alternate paths that\n bind a secondary vma avoiding the original fenced vma. For the few\n instances where we need to release the fence (i.e. on binding when the\n GGTT range becomes invalid), replace the put_fence with a revoke_fence.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190822061557.18402-1-chris@chris-wilson.co.uk\n\ncommit b7d151ba4b295abdc9736eed3fb85cd475f2c883\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 22 07:09:13 2019 +0100\n\n drm/i915: Pull obj-\u003euserfault tracking under the ggtt-\u003emutex\n \n Since we want to revoke the ggtt vma from only under the ggtt-\u003emutex, we\n need to move protection of the userfault tracking from the struct_mutex\n to the ggtt-\u003emutex.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190822060914.2671-2-chris@chris-wilson.co.uk\n\ncommit e2ccc50a3a6c90cace6a9aef40072e558dd13863\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 22 07:09:12 2019 +0100\n\n drm/i915: Track ggtt fence reservations under its own mutex\n \n We can reduce the locking for fence registers from the dev-\u003estruct_mutex\n to a local mutex. We could introduce a mutex for the sole purpose of\n tracking the fence acquisition, except there is a little bit of overlap\n with the fault tracking, so use the i915_ggtt.mutex as it covers both.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190822060914.2671-1-chris@chris-wilson.co.uk\n\ncommit 5f9242775bb61f390f0885f23fc16397262c7538\nAuthor: Alan Stern \u003cstern@rowland.harvard.edu\u003e\nDate: Tue Aug 20 16:00:21 2019 -0400\n\n HID: logitech: Fix general protection fault caused by Logitech driver\n \n The syzbot fuzzer found a general protection fault in the HID subsystem:\n \n kasan: CONFIG_KASAN_INLINE enabled\n kasan: GPF could be caused by NULL-ptr deref or user memory access\n general protection fault: 0000 [#1] SMP KASAN\n CPU: 0 PID: 3715 Comm: syz-executor.3 Not tainted 5.2.0-rc6+ #15\n Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS\n Google 01/01/2011\n RIP: 0010:__pm_runtime_resume+0x49/0x180 drivers/base/power/runtime.c:1069\n Code: ed 74 d5 fe 45 85 ed 0f 85 9a 00 00 00 e8 6f 73 d5 fe 48 8d bd c1 02\n 00 00 48 b8 00 00 00 00 00 fc ff df 48 89 fa 48 c1 ea 03 \u003c0f\u003e b6 04 02 48\n 89 fa 83 e2 07 38 d0 7f 08 84 c0 0f 85 fe 00 00 00\n RSP: 0018:ffff8881d99d78e0 EFLAGS: 00010202\n RAX: dffffc0000000000 RBX: 0000000000000020 RCX: ffffc90003f3f000\n RDX: 0000000416d8686d RSI: ffffffff82676841 RDI: 00000020b6c3436a\n RBP: 00000020b6c340a9 R08: ffff8881c6d64800 R09: fffffbfff0e84c25\n R10: ffff8881d99d7940 R11: ffffffff87426127 R12: 0000000000000004\n R13: 0000000000000000 R14: ffff8881d9b94000 R15: ffffffff897f9048\n FS: 00007f047f542700(0000) GS:ffff8881db200000(0000) knlGS:0000000000000000\n CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n CR2: 0000001b30f21000 CR3: 00000001ca032000 CR4: 00000000001406f0\n DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\n DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\n Call Trace:\n pm_runtime_get_sync include/linux/pm_runtime.h:226 [inline]\n usb_autopm_get_interface+0x1b/0x50 drivers/usb/core/driver.c:1707\n usbhid_power+0x7c/0xe0 drivers/hid/usbhid/hid-core.c:1234\n hid_hw_power include/linux/hid.h:1038 [inline]\n hidraw_open+0x20d/0x740 drivers/hid/hidraw.c:282\n chrdev_open+0x219/0x5c0 fs/char_dev.c:413\n do_dentry_open+0x497/0x1040 fs/open.c:778\n do_last fs/namei.c:3416 [inline]\n path_openat+0x1430/0x3ff0 fs/namei.c:3533\n do_filp_open+0x1a1/0x280 fs/namei.c:3563\n do_sys_open+0x3c0/0x580 fs/open.c:1070\n do_syscall_64+0xb7/0x560 arch/x86/entry/common.c:301\n entry_SYSCALL_64_after_hwframe+0x49/0xbe\n \n It turns out the fault was caused by a bug in the HID Logitech driver,\n which violates the requirement that every pathway calling\n hid_hw_start() must also call hid_hw_stop(). This patch fixes the bug\n by making sure the requirement is met.\n \n Reported-and-tested-by: syzbot+3cbe5cd105d2ad56a1df@syzkaller.appspotmail.com\n Signed-off-by: Alan Stern \u003cstern@rowland.harvard.edu\u003e\n CC: \u003cstable@vger.kernel.org\u003e\n Signed-off-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n\ncommit 416dacb819f59180e4d86a5550052033ebb6d72c\nAuthor: Alan Stern \u003cstern@rowland.harvard.edu\u003e\nDate: Wed Aug 21 13:27:12 2019 -0400\n\n HID: hidraw: Fix invalid read in hidraw_ioctl\n \n The syzbot fuzzer has reported a pair of problems in the\n hidraw_ioctl() function: slab-out-of-bounds read and use-after-free\n read. An example of the first:\n \n BUG: KASAN: slab-out-of-bounds in strlen+0x79/0x90 lib/string.c:525\n Read of size 1 at addr ffff8881c8035f38 by task syz-executor.4/2833\n \n CPU: 1 PID: 2833 Comm: syz-executor.4 Not tainted 5.3.0-rc2+ #1\n Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS\n Google 01/01/2011\n Call Trace:\n __dump_stack lib/dump_stack.c:77 [inline]\n dump_stack+0xca/0x13e lib/dump_stack.c:113\n print_address_description+0x6a/0x32c mm/kasan/report.c:351\n __kasan_report.cold+0x1a/0x33 mm/kasan/report.c:482\n kasan_report+0xe/0x12 mm/kasan/common.c:612\n strlen+0x79/0x90 lib/string.c:525\n strlen include/linux/string.h:281 [inline]\n hidraw_ioctl+0x245/0xae0 drivers/hid/hidraw.c:446\n vfs_ioctl fs/ioctl.c:46 [inline]\n file_ioctl fs/ioctl.c:509 [inline]\n do_vfs_ioctl+0xd2d/0x1330 fs/ioctl.c:696\n ksys_ioctl+0x9b/0xc0 fs/ioctl.c:713\n __do_sys_ioctl fs/ioctl.c:720 [inline]\n __se_sys_ioctl fs/ioctl.c:718 [inline]\n __x64_sys_ioctl+0x6f/0xb0 fs/ioctl.c:718\n do_syscall_64+0xb7/0x580 arch/x86/entry/common.c:296\n entry_SYSCALL_64_after_hwframe+0x49/0xbe\n RIP: 0033:0x459829\n Code: fd b7 fb ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 48 89 f8 48 89 f7\n 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 \u003c48\u003e 3d 01 f0 ff\n ff 0f 83 cb b7 fb ff c3 66 2e 0f 1f 84 00 00 00 00\n RSP: 002b:00007f7a68f6dc78 EFLAGS: 00000246 ORIG_RAX: 0000000000000010\n RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 0000000000459829\n RDX: 0000000000000000 RSI: 0000000080404805 RDI: 0000000000000004\n RBP: 000000000075bf20 R08: 0000000000000000 R09: 0000000000000000\n R10: 0000000000000000 R11: 0000000000000246 R12: 00007f7a68f6e6d4\n R13: 00000000004c21de R14: 00000000004d5620 R15: 00000000ffffffff\n \n The two problems have the same cause: hidraw_ioctl() fails to test\n whether the device has been removed. This patch adds the missing test.\n \n Reported-and-tested-by: syzbot+5a6c4ec678a0c6ee84ba@syzkaller.appspotmail.com\n Signed-off-by: Alan Stern \u003cstern@rowland.harvard.edu\u003e\n CC: \u003cstable@vger.kernel.org\u003e\n Signed-off-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n\ncommit ade925995b172f1d7410d1c665b2f47c5e99bef0\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Wed Aug 21 12:43:12 2019 +0300\n\n drm: fix module name in edid_firmware log message\n \n The module is drm_kms_helper, not drm_kms_firmware.\n \n Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id\u003d204549\n Reported-by: Göran Uddeborg \u003cgoeran@uddeborg.se\u003e\n Fixes: ac6c35a4d8c7 (\"drm: add backwards compatibility support for drm_kms_helper.edid_firmware\")\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190821094312.5514-1-jani.nikula@intel.com\n\ncommit 8e458fe2ee05cf3fb1fb6bf9857e71e15c4d7441\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Aug 21 20:16:06 2019 +0100\n\n drm/i915: Generalise the clflush dma-worker\n \n Extract the dma-fence worker used by clflush for wider use, as we\n anticipate using workers coupled to dma-fences more frequently.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190821191606.17001-1-chris@chris-wilson.co.uk\n\ncommit e9a7cc1d97f13f11970f7c484bdfc0502efc4223\nAuthor: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\nDate: Wed Aug 21 16:16:32 2019 -0700\n\n cpufreq: Print driver name if cpufreq_suspend() fails\n \n Instead of printing the policy, which is incidentally a kernel pointer,\n so with limited interest, print the cpufreq driver name that failed to\n be suspend, which is more useful for debugging.\n \n Signed-off-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Acked-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit cef35af34d6dc3792333075115c7deb7062b6e18\nAuthor: Eran Ben Elisha \u003ceranbe@mellanox.com\u003e\nDate: Thu Aug 22 05:06:00 2019 +0000\n\n net/mlx5e: Add mlx5e HV VHCA stats agent\n \n HV VHCA stats agent is responsible on running a preiodic rx/tx\n packets/bytes stats update. Currently the supported format is version\n MLX5_HV_VHCA_STATS_VERSION. Block ID 1 is dedicated for statistics data\n transfer from the VF to the PF.\n \n The reporter fetch the statistics data from all opened channels, fill it\n in a buffer and send it to mlx5_hv_vhca_write_agent.\n \n As the stats layer should include some metadata per block (sequence and\n offset), the HV VHCA layer shall modify the buffer before actually send it\n over block 1.\n \n Signed-off-by: Eran Ben Elisha \u003ceranbe@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n Signed-off-by: Haiyang Zhang \u003chaiyangz@microsoft.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 29ddad43166db232955e686758e6fc23d97e3a6a\nAuthor: Eran Ben Elisha \u003ceranbe@mellanox.com\u003e\nDate: Thu Aug 22 05:05:56 2019 +0000\n\n net/mlx5: Add HV VHCA control agent\n \n Control agent is responsible over of the control block (ID 0). It should\n update the PF via this block about every capability change. In addition,\n upon block 0 invalidate, it should activate all other supported agents\n with data requests from the PF.\n \n Upon agent create/destroy, the invalidate callback of the control agent\n is being called in order to update the PF driver about this change.\n \n The control agent is an integral part of HV VHCA and will be created\n and destroy as part of the HV VHCA init/cleanup flow.\n \n Signed-off-by: Eran Ben Elisha \u003ceranbe@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n Signed-off-by: Haiyang Zhang \u003chaiyangz@microsoft.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 87175120defd2907d42592653c35feea9de0437a\nAuthor: Eran Ben Elisha \u003ceranbe@mellanox.com\u003e\nDate: Thu Aug 22 05:05:51 2019 +0000\n\n net/mlx5: Add HV VHCA infrastructure\n \n HV VHCA is a layer which provides PF to VF communication channel based on\n HyperV PCI config channel. It implements Mellanox\u0027s Inter VHCA control\n communication protocol. The protocol contains control block in order to\n pass messages between the PF and VF drivers, and data blocks in order to\n pass actual data.\n \n The infrastructure is agent based. Each agent will be responsible of\n contiguous buffer blocks in the VHCA config space. This infrastructure will\n bind agents to their blocks, and those agents can only access read/write\n the buffer blocks assigned to them. Each agent will provide three\n callbacks (control, invalidate, cleanup). Control will be invoked when\n block-0 is invalidated with a command that concerns this agent. Invalidate\n callback will be invoked if one of the blocks assigned to this agent was\n invalidated. Cleanup will be invoked before the agent is being freed in\n order to clean all of its open resources or deferred works.\n \n Block-0 serves as the control block. All execution commands from the PF\n will be written by the PF over this block. VF will ack on those by\n writing on block-0 as well. Its format is described by struct\n mlx5_hv_vhca_control_block layout.\n \n Signed-off-by: Eran Ben Elisha \u003ceranbe@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n Signed-off-by: Haiyang Zhang \u003chaiyangz@microsoft.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 913d14e866573350de3adede3c90cefb81944b0c\nAuthor: Eran Ben Elisha \u003ceranbe@mellanox.com\u003e\nDate: Thu Aug 22 05:05:47 2019 +0000\n\n net/mlx5: Add wrappers for HyperV PCIe operations\n \n Add wrapper functions for HyperV PCIe read / write /\n block_invalidate_register operations. This will be used as an\n infrastructure in the downstream patch for software communication.\n \n This will be enabled by default if CONFIG_PCI_HYPERV_INTERFACE is set.\n \n Signed-off-by: Eran Ben Elisha \u003ceranbe@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n Signed-off-by: Haiyang Zhang \u003chaiyangz@microsoft.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 348dd93e40c112afda3cd07daa6f470e474d29dc\nAuthor: Haiyang Zhang \u003chaiyangz@microsoft.com\u003e\nDate: Thu Aug 22 05:05:41 2019 +0000\n\n PCI: hv: Add a Hyper-V PCI interface driver for software backchannel interface\n \n This interface driver is a helper driver allows other drivers to\n have a common interface with the Hyper-V PCI frontend driver.\n \n Signed-off-by: Haiyang Zhang \u003chaiyangz@microsoft.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit e5d2f910cfeca852f6e2dc19dfa8dab264ce0cde\nAuthor: Dexuan Cui \u003cdecui@microsoft.com\u003e\nDate: Thu Aug 22 05:05:37 2019 +0000\n\n PCI: hv: Add a paravirtual backchannel in software\n \n Windows SR-IOV provides a backchannel mechanism in software for communication\n between a VF driver and a PF driver. These \"configuration blocks\" are\n similar in concept to PCI configuration space, but instead of doing reads and\n writes in 32-bit chunks through a very slow path, packets of up to 128 bytes\n can be sent or received asynchronously.\n \n Nearly every SR-IOV device contains just such a communications channel in\n hardware, so using this one in software is usually optional. Using the\n software channel, however, allows driver implementers to leverage software\n tools that fuzz the communications channel looking for vulnerabilities.\n \n The usage model for these packets puts the responsibility for reading or\n writing on the VF driver. The VF driver sends a read or a write packet,\n indicating which \"block\" is being referred to by number.\n \n If the PF driver wishes to initiate communication, it can \"invalidate\" one or\n more of the first 64 blocks. This invalidation is delivered via a callback\n supplied by the VF driver by this driver.\n \n No protocol is implied, except that supplied by the PF and VF drivers.\n \n Signed-off-by: Jake Oshins \u003cjakeo@microsoft.com\u003e\n Signed-off-by: Dexuan Cui \u003cdecui@microsoft.com\u003e\n Cc: Haiyang Zhang \u003chaiyangz@microsoft.com\u003e\n Cc: K. Y. Srinivasan \u003ckys@microsoft.com\u003e\n Cc: Stephen Hemminger \u003csthemmin@microsoft.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n Signed-off-by: Haiyang Zhang \u003chaiyangz@microsoft.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit e87cf35835d933d8c295cc946dae8547d1a6bd63\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Thu Aug 22 11:51:46 2019 +0530\n\n soundwire: intel: remove unused variables\n \n Variables \u0027nval\u0027 and \u0027i\u0027 are no longer used sdw_master_read_intel_prop()\n so remove them.\n \n drivers/soundwire/intel.c: In function \u0027sdw_master_read_intel_prop\u0027:\n drivers/soundwire/intel.c:829:12: warning: unused variable \u0027i\u0027 [-Wunused-variable]\n int nval, i;\n ^\n drivers/soundwire/intel.c:829:6: warning: unused variable \u0027nval\u0027 [-Wunused-variable]\n int nval, i;\n ^~~~\n \n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Fixes: 085f4ace103d (\"soundwire: intel: read mclk_freq property from firmware\")\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 9176b425bf8ff98fd12bffeea8e39373345cd745\nAuthor: Andrew-sh.Cheng \u003candrew-sh.cheng@mediatek.com\u003e\nDate: Tue Aug 13 21:31:48 2019 +0800\n\n cpufreq: mediatek: Add support for mt8183\n \n Add compatible string for mediatek mt8183\n \n Signed-off-by: Andrew-sh.Cheng \u003candrew-sh.cheng@mediatek.com\u003e\n [ VK: Fixed rebase conflict ]\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit dce0bb84a73bac8a5bf3ad54e922215b61350481\nAuthor: Andrew-sh.Cheng \u003candrew-sh.cheng@mediatek.com\u003e\nDate: Tue Aug 13 21:31:46 2019 +0800\n\n cpufreq: mediatek: change to regulator_get_optional\n \n For new mediatek chip mt8183,\n cci and little cluster share the same buck,\n so need to modify the attribute of regulator from exclusive to optional\n \n Signed-off-by: Andrew-sh.Cheng \u003candrew-sh.cheng@mediatek.com\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit 75c000c4bcbe2b0eb82baf90c7dd75c7380cc3fd\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Sun Aug 18 02:32:22 2019 -0400\n\n cpufreq: imx-cpufreq-dt: Add i.MX8MN support\n \n i.MX8MN has different speed grading definition as below, it has 4 bits\n to define speed grading, add support for it.\n \n SPEED_GRADE[3:0] MHz\n 0000 2300\n 0001 2200\n 0010 2100\n 0011 2000\n 0100 1900\n 0101 1800\n 0110 1700\n 0111 1600\n 1000 1500\n 1001 1400\n 1010 1300\n 1011 1200\n 1100 1100\n 1101 1000\n 1110 900\n 1111 800\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit 8ec5035039c743888c5ecd04a680676ef1537fe8\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Sun Aug 18 02:32:21 2019 -0400\n\n cpufreq: Use imx-cpufreq-dt for i.MX8MN\u0027s speed grading\n \n Add i.MX8MN to blacklist, so that imx-cpufreq-dt driver can handle\n speed grading bits just like other i.MX8M SoCs.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit ada54f35b2270942f45250fa1c553a426b579f9e\nAuthor: Douglas RAILLARD \u003cdouglas.raillard@arm.com\u003e\nDate: Thu Aug 8 14:18:57 2019 +0100\n\n cpufreq: qcom-hw: invoke frequency-invariance setter function\n \n Add calls to arch_set_freq_scale() in qcom-cpufreq-hw driver to enable\n frequency invariance.\n \n Signed-off-by: Douglas RAILLARD \u003cdouglas.raillard@arm.com\u003e\n Reviewed-by: Quentin Perret \u003cquentin.perret@arm.com\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit 0eae1e37db8acef1e511cdbc63634cb700188644\nAuthor: Sibi Sankar \u003csibis@codeaurora.org\u003e\nDate: Wed Aug 7 17:15:43 2019 +0530\n\n cpufreq: qcom-hw: Update logic to detect turbo frequency\n \n The core count read back from the each domain\u0027s look up table serves\n as an indicator for the onset of the turbo frequency and not accurate\n representation of number of cores in a paticular domain. Update turbo\n detection logic accordingly to add support for SM8150 SoCs.\n \n Signed-off-by: Sibi Sankar \u003csibis@codeaurora.org\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit 08c327f63f355fce190ac3e1ac62e19d2c5f004d\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Sat Aug 17 16:24:35 2019 +0200\n\n crypto: sha256_generic - Switch to the generic lib/crypto/sha256.c lib code\n \n Drop the duplicate generic sha256 (and sha224) implementation from\n crypto/sha256_generic.c and use the implementation from\n lib/crypto/sha256.c instead.\n \n \"diff -u lib/crypto/sha256.c sha256_generic.c\" shows that the core\n sha256_transform function from both implementations is identical and\n the other code is functionally identical too.\n \n Suggested-by: Eric Biggers \u003cebiggers@kernel.org\u003e\n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 7d2f5b0c43e0bb346fbf78daefd68cd0bfc56ca3\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Sat Aug 17 16:24:34 2019 +0200\n\n crypto: sha256 - Add sha224 support to sha256 library code\n \n Add sha224 support to the lib/crypto/sha256 library code. This will allow\n us to replace both the sha256 and sha224 parts of crypto/sha256_generic.c\n when we remove the code duplication in further patches in this series.\n \n Suggested-by: Eric Biggers \u003cebiggers@kernel.org\u003e\n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 01d3aee86625bd798a5e69afb92517d5530c7ed1\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Sat Aug 17 16:24:33 2019 +0200\n\n crypto: sha256 - Make lib/crypto/sha256.c suitable for generic use\n \n Before this commit lib/crypto/sha256.c has only been used in the s390 and\n x86 purgatory code, make it suitable for generic use:\n \n * Export interesting symbols\n * Add -D__DISABLE_EXPORTS to CFLAGS_sha256.o for purgatory builds to\n avoid the exports for the purgatory builds\n * Add to lib/crypto/Makefile and crypto/Kconfig\n \n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 906a4bb97f5d0389cae9cc3634e8059fed5334b5\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Sat Aug 17 16:24:32 2019 +0200\n\n crypto: sha256 - Use get/put_unaligned_be32 to get input, memzero_explicit\n \n Use get/put_unaligned_be32 in lib/crypto/sha256.c to load / store data\n so that it can be used with unaligned buffers too, making it more generic.\n \n And use memzero_explicit for better clearing of sensitive data.\n \n Note unlike other patches in this series this commit actually makes\n functional changes to the sha256 code as used by the purgatory code.\n \n This fully aligns the lib/crypto/sha256.c sha256 implementation with the\n one from crypto/sha256_generic.c allowing us to remove the latter in\n further patches in this series.\n \n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit ad767ee858b38af634c957a792cb001d54a7b981\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Sat Aug 17 16:24:31 2019 +0200\n\n crypto: sha256 - Move lib/sha256.c to lib/crypto\n \n Generic crypto implementations belong under lib/crypto not directly in\n lib, likewise the header should be in include/crypto, not include/linux.\n \n Note that the code in lib/crypto/sha256.c is not yet available for\n generic use after this commit, it is still only used by the s390 and x86\n purgatory code. Making it suitable for generic use is done in further\n patches in this series.\n \n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 1a01333d313447f0cdfcc2882ea3afe387a24be0\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Sat Aug 17 16:24:30 2019 +0200\n\n crypto: sha256_generic - Fix some coding style issues\n \n Add a bunch of missing spaces after commas and arround operators.\n \n Note the main goal of this is to make sha256_transform and its helpers\n identical in formatting too the duplcate implementation in lib/sha256.c,\n so that \"diff -u\" can be used to compare them to prove that no functional\n changes are made when further patches in this series consolidate the 2\n implementations into 1.\n \n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit aca1111965d78477f9169c0df54d0ea06173572f\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Sat Aug 17 16:24:29 2019 +0200\n\n crypto: sha256 - Fix some coding style issues\n \n For some reason after the first 15 steps the last statement of each\n step ends with \"t1+t2\", missing spaces around the \"+\". This commit\n fixes this. This was done with a \u0027s/\u003d t1+t2/\u003d t1 + t2/\u0027 to make sure\n no functional changes are introduced.\n \n Note the main goal of this is to make lib/sha256.c\u0027s sha256_transform\n and its helpers identical in formatting too the duplcate implementation\n in crypto/sha256_generic.c so that \"diff -u\" can be used to compare them\n to prove that no functional changes are made when further patches in\n this series consolidate the 2 implementations into 1.\n \n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 23966841934908ad4ef997231f1fdd1f9a9d0f42\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Fri Aug 16 17:06:24 2019 +0300\n\n crypto: vmx/xts - use fallback for ciphertext stealing\n \n For correctness and compliance with the XTS-AES specification, we are\n adding support for ciphertext stealing to XTS implementations, even\n though no use cases are known that will be enabled by this.\n \n Since the Power8 implementation already has a fallback skcipher standby\n for other purposes, let\u0027s use it for this purpose as well. If ciphertext\n stealing use cases ever become a bottleneck, we can always revisit this.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit ce68acbcb6a5d5dbaa9e76df924e1c191e8c7516\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Fri Aug 16 15:35:45 2019 +0300\n\n crypto: s390/xts-aes - invoke fallback for ciphertext stealing\n \n For correctness and compliance with the XTS-AES specification, we are\n adding support for ciphertext stealing to XTS implementations, even\n though no use cases are known that will be enabled by this.\n \n Since the s390 implementation already has a fallback skcipher standby\n for other purposes, let\u0027s use it for this purpose as well. If ciphertext\n stealing use cases ever become a bottleneck, we can always revisit this.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 8ce5fac2dc1bf64e1e6d2371e4ff9a9bfe8fd49f\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Fri Aug 16 15:21:50 2019 +0300\n\n crypto: x86/xts - implement support for ciphertext stealing\n \n Align the x86 code with the generic XTS template, which now supports\n ciphertext stealing as described by the IEEE XTS-AES spec P1619.\n \n Tested-by: Stephan Mueller \u003csmueller@chronox.de\u003e\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit b395ed4f948a21c27391ed7981e7e210d19c9366\nAuthor: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nDate: Fri Aug 16 08:47:43 2019 +1000\n\n crypto: hisilicon - Fix warning on printing %p with dma_addr_t\n \n This patch fixes a printk format warning by replacing %p with %#llx\n for dma_addr_t.\n \n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 9a394d1208147715b8a365f44b4e7bfbb2094748\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:01:12 2019 +0300\n\n fs: cifs: move from the crypto cipher API to the new DES library interface\n \n Some legacy code in the CIFS driver uses single DES to calculate\n some password hash, and uses the crypto cipher API to do so. Given\n that there is no point in invoking an accelerated cipher for doing\n 56-bit symmetric encryption on a single 8-byte block of input, the\n flexibility of the crypto cipher API does not add much value here,\n and so we\u0027re much better off using a library call into the generic\n C implementation.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 18fbe0da8e98fe167fbfe1757003e2a2a74d24f3\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:01:11 2019 +0300\n\n crypto: des - remove now unused __des3_ede_setkey()\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit cc1d24b980ded2124f6f80b1321427031fd2beb3\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:01:10 2019 +0300\n\n crypto: x86/des - switch to library interface\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 04007b0e6cbbab5836ac891626e91edf10d46341\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:01:09 2019 +0300\n\n crypto: des - split off DES library from generic DES cipher driver\n \n Another one for the cipher museum: split off DES core processing into\n a separate module so other drivers (mostly for crypto accelerators)\n can reuse the code without pulling in the generic DES cipher itself.\n This will also permit the cipher interface to be made private to the\n crypto API itself once we move the only user in the kernel (CIFS) to\n this library interface.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 6b5c4818621fb0d3eda1d4059634b38e09bd7243\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:01:08 2019 +0300\n\n crypto: des - remove unused function\n \n Remove the old DES3 verification functions that are no longer used.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 4fd4be0576166f0b360d22744e143cd7847b0c29\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:01:07 2019 +0300\n\n crypto: 3des - move verification out of exported routine\n \n In preparation of moving the shared key expansion routine into the\n DES library, move the verification done by __des3_ede_setkey() into\n its callers.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit dcb15794ec571c10243c24eaff8512b789dfef2a\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:01:06 2019 +0300\n\n crypto: ux500/des - switch to new verification routines\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 9d574ae8ebc1e6e485ef949f086157601008a9d3\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:01:05 2019 +0300\n\n crypto: talitos/des - switch to new verification routines\n \n Cc: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 08d4b408e8f51562676fa47b728abd6ddf9828c2\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:01:04 2019 +0300\n\n crypto: sun4i/des - switch to new verification routines\n \n Cc: Corentin Labbe \u003cclabbe.montjoie@gmail.com\u003e\n Acked-by: Corentin Labbe \u003cclabbe.montjoie@gmail.com\u003e\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit b5d0ba83a0fb09b8dfd8bfdd9072d532da4f3091\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:01:03 2019 +0300\n\n crypto: stm32/des - switch to new verification routines\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 4eaef05210cbe7052adf53d030fd44b37b69c5a3\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:01:02 2019 +0300\n\n crypto: rk3288/des - switch to new verification routines\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit f96c897c855cfbde74d92f5cd65dc1108799bd8a\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:01:01 2019 +0300\n\n crypto: qce/des - switch to new verification routines\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 0157fb268893cddeb79c3cb6fc05d13a132aa66c\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:01:00 2019 +0300\n\n crypto: picoxcell/des - switch to new verification routines\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 304daa4f0ee54e42dea9e167e162caba17d48543\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:00:59 2019 +0300\n\n crypto: omap/des - switch to new verification routines\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit d4b90dbc8578c567827901e625018ead7561f741\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:00:58 2019 +0300\n\n crypto: n2/des - switch to new verification routines\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 9bfa85ebd5eec6c357ab348b8fbca73529111bd7\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:00:57 2019 +0300\n\n crypto: cesa/des - switch to new verification routines\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 3ca20b65cbfd1acd733c2d12848261b00b2d55e1\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:00:56 2019 +0300\n\n crypto: ixp4xx/des - switch to new verification routines\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 21f5a15e0f26c7ea15bee9ed3c5fd1daa40b0b77\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:00:55 2019 +0300\n\n crypto: safexcel/des - switch to new verification routines\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 894b68d8be4b305615ab3be52667478fe5f6e0b5\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:00:54 2019 +0300\n\n crypto: hisilicon/des - switch to new verification routines\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit f4ed68609798211f6503e66d7d371230ea18cf14\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:00:53 2019 +0300\n\n crypto: hifn/des - switch to new verification routines\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 00cd6b233fc89463d4f4eddeb9abf1e009cc09c0\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:00:52 2019 +0300\n\n crypto: ccree/des - switch to new verification routines\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit b525041633145828bd6744a4d1b79dbc084315b3\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:00:51 2019 +0300\n\n crypto: ccp/des - switch to new verification routines\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 43d81a8b258e1ed894af0f5d81921365004b6d90\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:00:50 2019 +0300\n\n crypto: nitrox/des - switch to new verification routines\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 0e1cbe9795ace1ea5e1621e52b9d918d9382fcfd\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:00:49 2019 +0300\n\n crypto: cpt/des - switch to new verification routines\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit a628c5a11d8b87a16dbc0d5bbc9b6ff233865622\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:00:48 2019 +0300\n\n crypto: caam/des - switch to new verification routines\n \n Cc: Horia Geanta \u003choria.geanta@nxp.com\u003e\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 05a7238d085f6cfaff9ad41e56f74d9923afa091\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:00:47 2019 +0300\n\n crypto: bcm/des - switch to new verification routines\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 92c203e2dca3c816c7001e429134f412d9d89389\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:00:46 2019 +0300\n\n crypto: atmel/des - switch to new verification routines\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 8f467cf29f192f9ebd89dcc4a18901e62c17e37a\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:00:45 2019 +0300\n\n crypto: sparc/des - switch to new verification routines\n \n Switch to the refactored DES key verification routines. While at it,\n rename the DES encrypt/decrypt routines so they will not conflict with\n the DES library later on.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 4e2c820897d93a4dbe76865d860087f809963308\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:00:44 2019 +0300\n\n crypto: s390/des - switch to new verification routines\n \n Switch to the refactored DES key verification routines. While at it,\n rename the DES encrypt/decrypt routines so they will not conflict with\n the DES library later on.\n \n Reviewed-by: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 6ee41e5420d0afa8cddf09aa7384dabe570f8dc7\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Thu Aug 15 12:00:43 2019 +0300\n\n crypto: des/3des_ede - add new helpers to verify keys\n \n The recently added helper routine to perform key strength validation\n of triple DES keys is slightly inadequate, since it comes in two versions,\n neither of which are highly useful for anything other than skciphers (and\n many drivers still use the older blkcipher interfaces).\n \n So let\u0027s add a new helper and, considering that this is a helper function\n that is only intended to be used by crypto code itself, put it in a new\n des.h header under crypto/internal.\n \n While at it, implement a similar helper for single DES, so that we can\n start replacing the pattern of calling des_ekey() into a temp buffer\n that occurs in many drivers in drivers/crypto.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 220f67917ae649336b5fef4dec8024fed4b177f3\nAuthor: Denis Efremov \u003cefremov@linux.com\u003e\nDate: Tue Aug 13 09:06:10 2019 +0300\n\n MAINTAINERS: nx crypto: Fix typo in a filepath\n \n Fix typo in nx_debugfs.c filepath. File extension changed from .h to .c\n The file nx_debugfs.h never existed.\n \n Cc: Breno Leitão \u003cleitao@debian.org\u003e\n Cc: Nayna Jain \u003cnayna@linux.ibm.com\u003e\n Cc: Paulo Flabiano Smorigo \u003cpfsmorigo@gmail.com\u003e\n Cc: Dan Streetman \u003cddstreet@ieee.org\u003e\n Cc: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n Cc: linux-crypto@vger.kernel.org\n Signed-off-by: Denis Efremov \u003cefremov@linux.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit b0c091ae04f6746f541b9be91809e1f4f43e9a65\nAuthor: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\nDate: Mon Aug 12 12:32:57 2019 -0700\n\n lib/mpi: Eliminate unused umul_ppmm definitions for MIPS\n \n Clang errors out when building this macro:\n \n lib/mpi/generic_mpih-mul1.c:37:24: error: invalid use of a cast in a\n inline asm context requiring an l-value: remove the cast or build with\n -fheinous-gnu-extensions\n umul_ppmm(prod_high, prod_low, s1_ptr[j], s2_limb);\n ~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n lib/mpi/longlong.h:652:20: note: expanded from macro \u0027umul_ppmm\u0027\n : \"\u003dl\" ((USItype)(w0)), \\\n ~~~~~~~~~~^~~\n lib/mpi/generic_mpih-mul1.c:37:3: error: invalid output constraint \u0027\u003dh\u0027\n in asm\n umul_ppmm(prod_high, prod_low, s1_ptr[j], s2_limb);\n ^\n lib/mpi/longlong.h:653:7: note: expanded from macro \u0027umul_ppmm\u0027\n \"\u003dh\" ((USItype)(w1)) \\\n ^\n 2 errors generated.\n \n The C version that is used for GCC 4.4 and up works well with clang;\n however, it is not currently being used because Clang masks itself\n as GCC 4.2.1 for compatibility reasons. As Nick points out, we require\n GCC 4.6 and newer in the kernel so we can eliminate all of the\n versioning checks and just use the C version of umul_ppmm for all\n supported compilers.\n \n Link: https://github.com/ClangBuiltLinux/linux/issues/605\n Suggested-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Reviewed-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit ff296293b3538d19278a7f7cd1f3aa600ad9164c\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Mon Aug 19 08:02:45 2019 -0700\n\n random: Support freezable kthreads in add_hwgenerator_randomness()\n \n The kthread calling this function is freezable after commit 03a3bb7ae631\n (\"hwrng: core - Freeze khwrng thread during suspend\") is applied.\n Unfortunately, this function uses wait_event_interruptible() but doesn\u0027t\n check for the kthread being woken up by the fake freezer signal. When a\n user suspends the system, this kthread will wake up and if it fails the\n entropy size check it will immediately go back to sleep and not go into\n the freezer. Eventually, suspend will fail because the task never froze\n and a warning message like this may appear:\n \n PM: suspend entry (deep)\n Filesystems sync: 0.000 seconds\n Freezing user space processes ... (elapsed 0.001 seconds) done.\n OOM killer disabled.\n Freezing remaining freezable tasks ...\n Freezing of tasks failed after 20.003 seconds (1 tasks refusing to freeze, wq_busy\u003d0):\n hwrng R running task 0 289 2 0x00000020\n [\u003cc08c64c4\u003e] (__schedule) from [\u003cc08c6a10\u003e] (schedule+0x3c/0xc0)\n [\u003cc08c6a10\u003e] (schedule) from [\u003cc05dbd8c\u003e] (add_hwgenerator_randomness+0xb0/0x100)\n [\u003cc05dbd8c\u003e] (add_hwgenerator_randomness) from [\u003cbf1803c8\u003e] (hwrng_fillfn+0xc0/0x14c [rng_core])\n [\u003cbf1803c8\u003e] (hwrng_fillfn [rng_core]) from [\u003cc015abec\u003e] (kthread+0x134/0x148)\n [\u003cc015abec\u003e] (kthread) from [\u003cc01010e8\u003e] (ret_from_fork+0x14/0x2c)\n \n Check for a freezer signal here and skip adding any randomness if the\n task wakes up because it was frozen. This should make the kthread freeze\n properly and suspend work again.\n \n Fixes: 03a3bb7ae631 (\"hwrng: core - Freeze khwrng thread during suspend\")\n Reported-by: Keerthy \u003cj-keerthy@ti.com\u003e\n Tested-by: Keerthy \u003cj-keerthy@ti.com\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 4aa6cf8b6bd12ebb137710164bb0e40589bdd3a1\nAuthor: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\nDate: Wed Aug 21 14:38:31 2019 +0530\n\n MAINTAINERS: Add Amit Kucheria as reviewer for thermal\n \n Add Amit Kucheria as the reviewer for thermal as he would like to\n participate in the review process effort for the thermal framework.\n \n Signed-off-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Signed-off-by: Zhang Rui \u003crui.zhang@intel.com\u003e\n\ncommit a52c26f1d768b6185ed56444a1e5a1629ffd0c4c\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Jul 26 22:00:54 2019 +0800\n\n drm/amdkfd: remove set but not used variable \u0027pdd\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_process.c: In function restore_process_worker:\n drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_process.c:949:29: warning:\n variable pdd set but not used [-Wunused-but-set-variable]\n \n It is not used since\n commit 5b87245faf57 (\"drm/amdkfd: Simplify kfd2kgd interface\")\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 192d56d2437f690bce20644f3a5eb2aaf420dc87\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Jul 26 15:51:31 2019 +0800\n\n drm/amd/display: remove duplicated include from dc_link.c\n \n Remove duplicated include.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 252d2a5246363c36b66bb13d73978b8b4b71b1ef\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Jul 10 01:57:20 2019 +0000\n\n drm/amdgpu: remove duplicated include from gfx_v9_0.c\n \n Remove duplicated include.\n \n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6892c1f866bf17f9c945f04716610e4240c35f54\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Jun 27 07:53:50 2019 +0000\n\n drm/amdgpu: remove set but not used variable \u0027psp_enabled\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/gpu/drm/amd/amdgpu/nv.c: In function \u0027nv_common_early_init\u0027:\n drivers/gpu/drm/amd/amdgpu/nv.c:471:7: warning:\n variable \u0027psp_enabled\u0027 set but not used [-Wunused-but-set-variable]\n \n It\u0027s not used since inroduction in\n commit c6b6a42175f5 (\"drm/amdgpu: add navi10 common ip block (v3)\")\n \n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b859c579cdff00a6458d69442a19f85a403fd71b\nAuthor: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\nDate: Tue Aug 20 16:57:13 2019 -0700\n\n drm/amd/display: Fix 32-bit divide error in wait_for_alt_mode\n \n When building arm32 allyesconfig:\n \n ld.lld: error: undefined symbol: __aeabi_uldivmod\n \u003e\u003e\u003e referenced by dc_link.c\n \u003e\u003e\u003e gpu/drm/amd/display/dc/core/dc_link.o:(wait_for_alt_mode) in archive drivers/built-in.a\n \u003e\u003e\u003e referenced by dc_link.c\n \u003e\u003e\u003e gpu/drm/amd/display/dc/core/dc_link.o:(wait_for_alt_mode) in archive drivers/built-in.a\n \n time_taken_in_ns is of type unsigned long long so we need to use div_u64\n to avoid this error.\n \n Fixes: b5b1f4554904 (\"drm/amd/display: Enable type C hotplug\")\n Reported-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\n Acked-by: Randy Dunlap \u003crdunlap@infradead.org\u003e # build-tested\n Reviewed-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5a6a4c9d1b6743268480b5b5661a35c38d4df424\nAuthor: Nicolai Hähnle \u003cnicolai.haehnle@amd.com\u003e\nDate: Tue Aug 20 15:39:53 2019 +0200\n\n drm/amdgpu: prevent memory leaks in AMDGPU_CS ioctl\n \n Error out if the AMDGPU_CS ioctl is called with multiple SYNCOBJ_OUT and/or\n TIMELINE_SIGNAL chunks, since otherwise the last chunk wins while the\n allocated array as well as the reference counts of sync objects are leaked.\n \n Signed-off-by: Nicolai Hähnle \u003cnicolai.haehnle@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 452575c5703cd7d0cf285ef7a97de52af9a3d70d\nAuthor: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\nDate: Tue Aug 20 13:15:25 2019 -0400\n\n drm/amd/display: Lock the CRTC when setting CRC source\n \n [Why]\n We need to ensure that we\u0027re holding the lock on the CRTC when setting\n the CRC source since we\u0027re modifying the CRTC state directly.\n \n We also need to wait for any outstanding non-blocking commits to finish\n so they aren\u0027t reading state that\u0027s potentially being modified -\n non-blocking commits don\u0027t hold the CRTC lock while doing commit tail\n work.\n \n [How]\n Lock the CRTC using its mutex. While holding the lock check if there\u0027s\n any commit active on the CRTC - if there is, it\u0027s non-blocking and\n we should wait until it\u0027s finished by waiting for hw_done to be\n signaled since that\u0027s the last point where we touch CRTC state.\n \n Signed-off-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Reviewed-by: David Francis \u003cDavid.Francis@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 57638021b711de052df781001bd6fd35804c3646\nAuthor: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\nDate: Tue Aug 20 10:16:14 2019 -0400\n\n drm/amd/display: Split out DC programming for CRC capture\n \n [Why]\n Calling amdgpu_dm_crtc_set_crc_source in amdgpu_dm directly has the\n consequence of adding additional vblank references or starting DPRX\n CRC capture more than once without calling stop first.\n \n Vblank references for CRC capture should be managed entirely by opening\n and closing the CRC file from userspace.\n \n Stream state also shouldn\u0027t be required on the CRC so we can close the\n file after the CRTC has been disabled.\n \n [How]\n Do DC programming required for configuring CRC capture separately from\n setting the source. Whenever we re-enable or reset a CRC this\n programming should be reapplied.\n \n CRC vblank reference handling in amdgpu_dm can be entirely dropped after\n this.\n \n Stream state also no longer needs to be required since we can just defer\n the programming to when the stream is actually enabled.\n \n Signed-off-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Reviewed-by: David Francis \u003cDavid.Francis@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit df61eae4b87ac4dfb9c4fceea2f223a29466507b\nAuthor: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\nDate: Tue Aug 20 10:10:08 2019 -0400\n\n drm/amd/display: Use connector list for finding DPRX CRC aux\n \n [Why]\n This change is a refactor in preparation for adding locking and removing\n the requirement for a stream state on the CRTC for enabling CRC capture\n to fix igt@kms_plane_multiple@* warnings.\n \n [How]\n We can get the aux by finding the matching connector for the CRTC\n with the assumption that we\u0027re not doing cloning.\n \n Signed-off-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Reviewed-by: David Francis \u003cDavid.Francis@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7a235125959cffa28c20059ff1cb8725571fcf9f\nAuthor: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\nDate: Tue Aug 20 12:58:37 2019 -0400\n\n drm/amd/display: Check return code for CRC drm_crtc_vblank_get\n \n [Why]\n The call to drm_crtc_vblank_get can fail if vblank is disabled and\n we try to increment the reference.\n \n Since drm_crtc_vblank_get internally drops the reference when it fails\n it means the subsequent drm_crtc_vblank_put(...) when closing the file\n drops a zero reference.\n \n This was found via igt@kms_plane@pixel-format-pipe-A-planes.\n \n [How]\n Check the return code and return it on failure.\n \n We wouldn\u0027t have been able to enable CRC reading anyway since vblank\n wasn\u0027t enabled.\n \n Signed-off-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Reviewed-by: David Francis \u003cDavid.Francis@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6da6c2792808d767012389d6d83cf9d11a2c8256\nAuthor: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\nDate: Tue Aug 20 15:11:37 2019 +0800\n\n drm/amd/amdgpu: disable MMHUB PG for navi10\n \n Disable MMHUB PG for navi10 according to the production requirement.\n \n Signed-off-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6169631236da7fc7d8e9db8c339b64fd0831e298\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Tue Aug 20 12:22:16 2019 +0800\n\n drm/amd/powerplay: correct typo\n \n \"COMPUTE\" was wrongly spelled as \"CUSTOM\".\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a056ddce9b15a117949e59ad44bea5b3d12f17b5\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Fri Aug 16 17:11:46 2019 +0800\n\n drm/amd/powerplay: correct SW smu11 thermal range settings\n \n Problems with current settings:\n 1. The min value was overrided to 0 on Vega20 \u0026 Navi10. While\n the expected should be -273.15 C.\n 2. The thermal min/max threshold was output in wrong unit on\n Navi10 \u0026 Arcturus. As TEMP_RANGE_MIN/MAX is already in\n millicelsius. And \"*1000\" in smu_v11_0_start_thermal_control\n makes the output wrongly.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 69174eebc90287c38b27b98cc2c4beb6ae8564c8\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Fri Aug 16 13:52:26 2019 +0800\n\n drm/amd/powerplay: set Arcturus default fclk as bootup value on dpm disabled\n \n On fclk dpm disabled, the default dpm table will be setup with only one\n level and clock frequency as bootup value.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 88810f907bc632c161f8172546b6b5495be42ec3\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Fri Aug 16 13:47:01 2019 +0800\n\n drm/amd/powerplay: get bootup fclk value\n \n This is available with firmwareinfo table v3.2 or later.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9aef809b5cc0ab27b1cdebb3f800a4bacf7f603e\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Fri Aug 16 11:34:12 2019 +0800\n\n drm/amd/powerplay: expose supported clock domains only through sysfs\n \n Do not expose those unsupported clock domains through sysfs on\n Arcturus.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d80ead63a8f554c7c945205918eb75ceb58063ba\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Fri Aug 16 10:55:21 2019 +0800\n\n drm/amd/powerplay: update Arcturus smc fw and driver interface header\n \n Update smc fw and driver interface header.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 03dce35deb8575c85490b90b3c3a5f6e7613c1b9\nAuthor: Kevin Wang \u003ckevin1.wang@amd.com\u003e\nDate: Tue Aug 20 13:28:51 2019 +0800\n\n drm/amd/powerplay: remove duplicate macro smu_get_uclk_dpm_states in amdgpu_smu.h\n \n remove duplicate macro smu_get_uclk_dpm_states in amdgpu_smu.h\n \n \"\n #define smu_get_uclk_dpm_states(smu, clocks_in_khz, num_states) \\\n ((smu)-\u003eppt_funcs-\u003eget_uclk_dpm_states ? (smu)-\u003eppt_funcs-\u003eget_uclk_dpm_states((smu), (clocks_in_khz), (num_states)) : 0)\n #define smu_get_max_sustainable_clocks_by_dc(smu, max_clocks) \\\n ((smu)-\u003efuncs-\u003eget_max_sustainable_clocks_by_dc ? (smu)-\u003efuncs-\u003eget_max_sustainable_clocks_by_dc((smu), (max_clocks)) : 0)\n #define smu_get_uclk_dpm_states(smu, clocks_in_khz, num_states) \\\n ((smu)-\u003eppt_funcs-\u003eget_uclk_dpm_states ? (smu)-\u003eppt_funcs-\u003eget_uclk_dpm_states((smu), (clocks_in_khz), (num_states)) : 0)\n \"\n \n Signed-off-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ebecc6c48f39b3c549bee1e4ecb9be01bf341a0f\nAuthor: Kevin Wang \u003ckevin1.wang@amd.com\u003e\nDate: Mon Aug 19 23:38:02 2019 +0800\n\n drm/amd/powerplay: fix variable type errors in smu_v11_0_setup_pptable\n \n fix size type errors, from uint32_t to uint16_t.\n it will cause only initializes the highest 16 bits in\n smu_get_atom_data_table function.\n \n bug report:\n This fixes the following static checker warning.\n drivers/gpu/drm/amd/amdgpu/../powerplay/smu_v11_0.c:390 smu_v11_0_setup_pptable()\n warn: passing casted pointer \u0027\u0026size\u0027 to \u0027smu_get_atom_data_table()\u0027 32 vs 16.\n \n Signed-off-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reported-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9744214073bf85178765d58895ff33199cd31a7b\nAuthor: Kevin Wang \u003ckevin1.wang@amd.com\u003e\nDate: Fri Aug 16 15:17:42 2019 +0800\n\n drm/amd/powerplay: add smu_smc_read_sensor support for arcturus\n \n the bellow patch refine the sensor read sequence,\n but missed to add arcuturs support. (arcuturs_ppt.c)\n it will cause some sensor is not supported in arcturus.\n \n drm/amd/powerplay: change smu_read_sensor sequence in smu\n \n Signed-off-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6a9d8de7e9c75334949c005f9b1c50c6f857606d\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Sun Aug 18 17:59:57 2019 +0200\n\n drm/amdgpu: Fix a typo in the include header guard of \u0027navi12_ip_offset.h\u0027\n \n \u0027_navi10_ip_offset_HEADER\u0027 is already used in \u0027navi10_ip_offset.h\u0027, so use\n \u0027_navi12_ip_offset_HEADER\u0027 instead here.\n \n Reviewed-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 828d6fde7f574d74b0a6a591345d3c42b62d5e21\nAuthor: Tianci.Yin \u003ctianci.yin@amd.com\u003e\nDate: Mon Aug 19 15:30:22 2019 +0800\n\n drm/amdgpu/psp: move TMR to cpu invisible vram region\n \n so that more visible vram can be available for umd.\n \n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e.\n Signed-off-by: Tianci.Yin \u003ctianci.yin@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 50e275e88044c8dfb952606a01b284dfe3ad36ab\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Fri Aug 16 18:00:54 2019 +0800\n\n drm/amdgpu: remove redundant argument for psp_funcs::cmd_submit callback\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c181159a5b1e1e75474764c6e45b1273a8834728\nAuthor: Yong Zhao \u003cYong.Zhao@amd.com\u003e\nDate: Thu Aug 1 22:55:50 2019 -0400\n\n drm/amdkfd: Fill the name field in node topology with asic name v2\n \n The name field in node topology has not been used. We re-purpose it to\n hold the asic name, which can be queried by user space applications\n through sysfs.\n \n Signed-off-by: Yong Zhao \u003cYong.Zhao@amd.com\u003e\n Reviewed-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 51bfac71cade386966791a8db87a5912781d249f\nAuthor: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\nDate: Fri Aug 16 11:08:52 2019 +0800\n\n drm/amdgpu: Set no-retry as default.\n \n This is to improve performance.\n \n Signed-off-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Tested-by: Candice Li \u003ccandice.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c5fb912653dae3f8785d7aa0ae156b0eb56fc76f\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu Aug 15 17:44:22 2019 +0800\n\n drm/amdgpu: add firmware header printing for psp fw loading (v2)\n \n firmware header information is printed for direct fw loading but not\n added for psp fw loading yet\n \n v2: squash in warning fix (Alex)\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6c2243efa056e7bcb1cce84c1b8ecb6667852905\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu Aug 15 17:43:36 2019 +0800\n\n drm/amdgpu: fix debug level for ppt offset/size\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit cc216214ac84d2217d52965231fa5d041d46ee99\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu Aug 15 15:08:02 2019 +0800\n\n drm/amdgpu: remove special autoload handling for navi12\n \n s/r list in rlc firmware is ready, so remove the special autoload handling\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 01c22997bed059ee562c1c92f27c36806869a8ff\nAuthor: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\nDate: Wed Aug 21 11:27:13 2019 -0400\n\n drm/amd/display: Calculate bpc based on max_requested_bpc\n \n [Why]\n The only place where state-\u003emax_bpc is updated on the connector is\n at the start of atomic check during drm_atomic_connector_check. It\n isn\u0027t updated when adding the connectors to the atomic state after\n the fact. It also doesn\u0027t necessarily reflect the right value when\n called in amdgpu during mode validation outside of atomic check.\n \n This can cause the wrong bpc to be used even if the max_requested_bpc\n is the correct value.\n \n [How]\n Don\u0027t rely on state-\u003emax_bpc reflecting the real bpc value and just\n do the min(...) based on display info bpc and max_requested_bpc.\n \n Fixes: 01933ba42d3d (\"drm/amd/display: Use current connector state if NULL when checking bpc\")\n Signed-off-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Reviewed-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b05f65d7720b172b6fde3abfa49ed66837071d45\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Thu Aug 15 08:27:09 2019 -0500\n\n drm/amdgpu/gfx9: update pg_flags after determining if gfx off is possible\n \n We need to set certain power gating flags after we determine\n if the firmware version is sufficient to support gfxoff.\n Previously we set the pg flags in early init, but we later\n we might have disabled gfxoff if the firmware versions didn\u0027t\n support it. Move adding the additional pg flags after we\n determine whether or not to support gfxoff.\n \n Fixes: 005440066f92 (\"drm/amdgpu: enable gfxoff again on raven series (v2)\")\n Tested-by: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\n Tested-by: Tom St Denis \u003ctom.stdenis@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Cc: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\n\ncommit 4ad35c1f56386c8e7019c921bba1af109fde9693\nAuthor: Guo Ren \u003cren_guo@c-sky.com\u003e\nDate: Wed Aug 21 19:15:52 2019 +0800\n\n csky: Fixup 610 vipt cache flush mechanism\n \n 610 has vipt aliasing issue, so we need to finish the cache flush\n apis mentioned in cachetlb.rst to avoid data corruption.\n \n Here is the list of modified apis in the patch:\n \n - flush_kernel_dcache_page (new add)\n - flush_dcache_mmap_lock (new add)\n - flush_dcache_mmap_unlock (new add)\n - flush_kernel_vmap_range (new add)\n - invalidate_kernel_vmap_range (new add)\n - flush_anon_page (new add)\n - flush_cache_range (new add)\n - flush_cache_vmap (flush all)\n - flush_cache_vunmap (flush all)\n - flush_cache_mm (only dcache flush)\n - flush_icache_page (just nop)\n - copy_from_user_page (remove no need flush)\n - copy_to_user_page (remove no need flush)\n \n Change to V2:\n - Fixup compile error with xa_lock*(\u0026mapping-\u003ei_pages)\n \n Signed-off-by: Guo Ren \u003cren_guo@c-sky.com\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: Christoph Hellwig \u003chch@infradead.org\u003e\n\ncommit c96245148c1ec7af086da322481bf4119d1141d3\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Tue Aug 6 20:15:48 2019 -0300\n\n mm/mmu_notifiers: remove unregister_no_release\n \n mmu_notifier_unregister_no_release() and mmu_notifier_call_srcu() no\n longer have any users, they have all been converted to use\n mmu_notifier_put().\n \n So delete this difficult to use interface.\n \n Link: https://lore.kernel.org/r/20190806231548.25242-12-jgg@ziepe.ca\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Tested-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 47f725ee7b5f5cae1f83512961bcf8b41a7a5794\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Tue Aug 6 20:15:44 2019 -0300\n\n RDMA/odp: remove ib_ucontext from ib_umem\n \n At this point the ucontext is only being stored to access the ib_device,\n so just store the ib_device directly instead. This is more natural and\n logical as the umem has nothing to do with the ucontext.\n \n Link: https://lore.kernel.org/r/20190806231548.25242-8-jgg@ziepe.ca\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit c571feca2dc972dc5afeba9036d08239f1c51af1\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Tue Aug 6 20:15:43 2019 -0300\n\n RDMA/odp: use mmu_notifier_get/put for \u0027struct ib_ucontext_per_mm\u0027\n \n This is a significant simplification, no extra list is kept per FD, and\n the interval tree is now shared between all the ucontexts, reducing\n overhead if there are multiple ucontexts active.\n \n Link: https://lore.kernel.org/r/20190806231548.25242-7-jgg@ziepe.ca\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 1709510221c57fd566479c228434ff9edd6435be\nAuthor: Sibi Sankar \u003csibis@codeaurora.org\u003e\nDate: Wed Aug 7 12:39:57 2019 +0530\n\n soc: qcom: aoss: Add AOSS QMP support\n \n Add AOSS QMP support for SM8150 and SC7180 SoCs.\n \n Reviewed-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Tested-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Sibi Sankar \u003csibis@codeaurora.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit a1c4743e990d9a4a3254b4d13be2921f59bee019\nAuthor: Sibi Sankar \u003csibis@codeaurora.org\u003e\nDate: Wed Aug 7 12:39:56 2019 +0530\n\n dt-bindings: soc: qcom: aoss: Add SM8150 and SC7180 support\n \n Add SM8150 and SC7180 AOSS QMP to the list of possible bindings.\n \n Tested-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Sibi Sankar \u003csibis@codeaurora.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit e6fdc7ad66f115a24e00058299297c33ba34343a\nAuthor: Sibi Sankar \u003csibis@codeaurora.org\u003e\nDate: Wed Aug 7 12:39:53 2019 +0530\n\n dt-bindings: firmware: scm: Add SM8150 and SC7180 support\n \n Add compatible for SM8150 and SC7180 SoCs.\n \n Tested-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Sibi Sankar \u003csibis@codeaurora.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 2aabdf3b50eabad59107f6adfc24436d9bda4fb9\nAuthor: Sibi Sankar \u003csibis@codeaurora.org\u003e\nDate: Wed Aug 7 12:39:52 2019 +0530\n\n dt-bindings: firmware: scm: re-order compatible list\n \n re-order compatible list to maintain sort order.\n \n Tested-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Sibi Sankar \u003csibis@codeaurora.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit f117249e4b7831248404ff1fb47bfe44c355caea\nAuthor: Sibi Sankar \u003csibis@codeaurora.org\u003e\nDate: Wed Aug 7 12:39:51 2019 +0530\n\n soc: qcom: smem: Update max processor count\n \n Update max processor count to reflect the number of\n co-processors on SC7180 SoCs.\n \n Reviewed-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Tested-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Sibi Sankar \u003csibis@codeaurora.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 5970882a2510e8bffaef518a82ea207798187a93\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Tue Jun 25 22:40:20 2019 +0300\n\n net/mlx5e: Add trace point for neigh update\n \n Allow tracing neigh state during neigh update task that is executed on\n workqueue and is scheduled by neigh state change event.\n \n Usage example:\n \u003e# cd /sys/kernel/debug/tracing\n \u003e# echo mlx5:mlx5e_rep_neigh_update \u003e\u003e set_event\n \u003e# cat trace\n ...\n kworker/u48:7-2221 [009] ...1 1475.387435: mlx5e_rep_neigh_update:\n netdev: ens1f0 MAC: 24:8a:07:9a:17:9a IPv4: 1.1.1.10 IPv6: ::ffff:1.1.1.10 neigh_connected\u003d1\n \n Added corresponding documentation in\n Documentation/networking/device-driver/mellanox/mlx5.rst\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Dmytro Linkin \u003cdmitrolin@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit c786fe596bede275f887f212eebee74490043b84\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Tue Jun 25 22:33:15 2019 +0300\n\n net/mlx5e: Add trace point for neigh used value update\n \n Allow tracing result of neigh used value update task that is executed\n periodically on workqueue.\n \n Usage example:\n \u003e# cd /sys/kernel/debug/tracing\n \u003e# echo mlx5:mlx5e_tc_update_neigh_used_value \u003e\u003e set_event\n \u003e# cat trace\n ...\n kworker/u48:4-8806 [009] ...1 55117.882428: mlx5e_tc_update_neigh_used_value:\n netdev: ens1f0 IPv4: 1.1.1.10 IPv6: ::ffff:1.1.1.10 neigh_used\u003d1\n \n Added corresponding documentation in\n Documentation/networking/device-driver/mellanox/mlx5.rst\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Dmytro Linkin \u003cdmitrolin@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 7a978759b4e0e7a2ad3f10cbf9077915a85ec956\nAuthor: Dmytro Linkin \u003cdmitrolin@mellanox.com\u003e\nDate: Thu Jun 27 10:55:02 2019 +0000\n\n net/mlx5e: Add tc flower tracepoints\n \n Implemented following tracepoints:\n 1. Configure flower (mlx5e_configure_flower)\n 2. Delete flower (mlx5e_delete_flower)\n 3. Stats flower (mlx5e_stats_flower)\n \n Usage example:\n \u003e# cd /sys/kernel/debug/tracing\n \u003e# echo mlx5:mlx5e_configure_flower \u003e\u003e set_event\n \u003e# cat trace\n ...\n tc-6535 [019] ...1 2672.404466: mlx5e_configure_flower: cookie\u003d0000000067874a55 actions\u003d REDIRECT\n \n Added corresponding documentation in\n Documentation/networking/device-driver/mellanox/mlx5.rst\n \n Signed-off-by: Dmytro Linkin \u003cdmitrolin@mellanox.com\u003e\n Reviewed-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 95435ad7999b1218367f0667ed5fe98d042ffe78\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Sun Aug 4 13:52:31 2019 +0300\n\n net/mlx5e: Only access fully initialized flows in neigh update\n \n To remove dependency on rtnl lock and prevent neigh update code from\n accessing uninitialized flows when executing concurrently with tc, extend\n mlx5e_tc_flow with \u0027init_done\u0027 completion. Modify helper\n mlx5e_take_all_encap_flows() to wait for flow completion after obtaining\n reference to it. Modify mlx5e_tc_encap_flows_del() and\n mlx5e_tc_encap_flows_add() to skip flows that don\u0027t have OFFLOADED flag\n set, which can happen if concurrent flow initialization failed.\n \n This commit finishes neigh update refactoring for concurrent execution\n started in previous change in this series.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 2a1f1768fa17805ca2e937e2e034a7c3433d3bdc\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Sun Aug 4 12:25:57 2019 +0300\n\n net/mlx5e: Refactor neigh update for concurrent execution\n \n In order to remove dependency on rtnl lock and allow neigh update workqueue\n task to execute concurrently with tc, refactor mlx5e_rep_neigh_update() for\n concurrent execution:\n \n - Lock encap table when accessing encap entry to prevent concurrent\n changes. To do this properly, the initial encap state check is moved from\n mlx5e_rep_neigh_update() into mlx5e_rep_update_flows() to be performed\n under encap_tbl_lock protection.\n \n - Wait for encap to be fully initialized before accessing it by means of\n \u0027res_ready\u0027 completion.\n \n - Add mlx5e_take_all_encap_flows() helper which is used to construct a\n temporary list of flows and efi indexes that is used to access current\n encap data in flow which can be attached to multiple encaps\n simultaneously. Release the flows from temporary list after\n encap_tbl_lock critical section. This is necessary because\n mlx5e_flow_put() can\u0027t be called while holding encap_tbl_lock.\n \n - Modify mlx5e_tc_encap_flows_add() and mlx5e_tc_encap_flows_del() to work\n with user-provided list of flows built by mlx5e_take_all_encap_flows(),\n instead of traversing encap flow list directly.\n \n This is first step in complex neigh update refactoring, which is finished\n by following commit in this series.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 6a06c2f7843d85b43ccea6e89de8e432834c089b\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Sat Aug 3 21:43:06 2019 +0300\n\n net/mlx5e: Refactor neigh used value update for concurrent execution\n \n In order to remove dependency on rtnl lock and allow neigh used value\n update workqueue task to execute concurrently with tc, refactor\n mlx5e_tc_update_neigh_used_value() for concurrent execution:\n \n - Lock encap table when accessing encap entry to prevent concurrent\n changes.\n \n - Save offloaded encap flows to temporary list and release them after encap\n entry is updated. Add mlx5e_put_encap_flow_list() helper which is\n intended to be shared with neigh update code in following patch in this\n series. This is necessary because mlx5e_flow_put() can\u0027t be called while\n holding encap_tbl_lock.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit ac0d917632cf7fbbe953f2ec82c2c979ab1b4a06\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Mon Jun 11 14:16:14 2018 +0300\n\n net/mlx5e: Protect neigh hash encap list with spinlock and rcu\n \n Rcu-ify mlx5e_neigh_hash_entry-\u003eencap_list by changing operations on encap\n list to their rcu counterparts and extending encap structure with rcu_head\n to free the encap instances after rcu grace period. Use rcu read lock when\n traversing encap list. Implement helper mlx5e_get_next_valid_encap()\n function that is used by mlx5e_tc_update_neigh_used_value() to safely\n iterate over valid entries of nhe-\u003eencap_list.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Jianbo Liu \u003cjianbol@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 70e83bd3b0e434ee9fa8e0010f76080aa4320725\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Tue Jun 12 11:59:43 2018 +0300\n\n net/mlx5e: Refactor mlx5e_neigh_update_table-\u003eencap_lock\n \n To remove dependency on rtnl lock, always take neigh update encap lock when\n modifying neigh update hash table and list. Originally, this lock was only\n used to synchronize with netevent handler function, which is called from bh\n context and cannot use rtnl lock for synchronization. Take lock in encap\n entry attach function to prevent concurrent modifications of neigh update\n hash table and list.\n \n Taking the encap lock when creating new nhe introduces a problem that we\n need to allocate new entry with sleeping GFP_KERNEL flag while holding a\n spinlock. However, since previous patch in this series has already\n converted lookup in netevent handler function to user rcu read lock instead\n of encap lock, we can safely convert the lock type to mutex.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Jianbo Liu \u003cjianbol@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 1216ce9d4a740bed88393177174fef275069a560\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Fri Jun 8 11:49:28 2018 +0300\n\n net/mlx5e: Extend neigh hash entry with rcu\n \n To remove dependency on rtnl lock and to allow unlocked iteration over list\n of neigh hash entries, extend nhe with rcu. Change operations on neigh list\n to their rcu counterparts and free neigh hash entry with rcu timeout.\n \n Introduce mlx5e_get_next_nhe() helper that is used to iterate over rcu\n neigh list with reference to nhe taken.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Jianbo Liu \u003cjianbol@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 61081f9c09df33fe0123d234e840e3b4d3c17269\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Fri Jun 8 11:31:28 2018 +0300\n\n net/mlx5e: Always take reference to neigh entry\n \n Neigh entry has reference counter, however it is only used when scheduling\n neigh update event. In all other cases reference to neigh entry is not\n taken while working with it. Neigh code relies on synchronization provided\n by rtnl lock and uses encap list size as implicit reference counter.\n \n To remove dependency on rtnl lock, always take reference to neigh entry\n while using it. Remove neigh entry from hash table and delete it only when\n reference counter reaches zero. This can result spurious neigh update\n events, when there is an event on entry that has zero encaps attached.\n However, such events are rare and properly handled by neigh update handler.\n \n Extend encap entry with reference to neigh hash entry in order to be able\n to directly release it when encap is detached, instead of lookup nhe by key\n through hash table. Extend nhe with reference to device priv structure to\n guarantee correctness when nhe is used with stack devices, bond setup, in\n which case it is non-trivial to determine correct device when releasing the\n nhe.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Jianbo Liu \u003cjianbol@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 93415e45d30cbcb24ea0e45d2efd0a3511866f26\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Wed Jun 13 14:49:46 2018 +0300\n\n net/mlx5e: Extract code that queues neigh update work into function\n \n As a preparation for following refactoring that removes rtnl lock\n dependency from neigh hash entry handlers, extract code that enqueues neigh\n update work into standalone function. This commit doesn\u0027t change\n functionality.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Jianbo Liu \u003cjianbol@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit d8cf9372b654406c92cbe1d342359dc58368bfe7\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Thu Aug 22 00:12:39 2019 +0530\n\n arm64: dts: qcom: sm8150: Add apps shared nodes\n \n Add hwlock, pmu, smem, tcsr_mutex_regs, apss_shared mailbox, apps_rsc\n including the rpmhcc child nodes to the SM8150 DTSI\n \n Co-developed-by: Sibi Sankar \u003csibis@codeaurora.org\u003e\n Signed-off-by: Sibi Sankar \u003csibis@codeaurora.org\u003e\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n Reviewed-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 912c373a764ffcf81e80c2550b9d95cbec5b5438\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Thu Aug 22 00:12:38 2019 +0530\n\n arm64: dts: qcom: sm8150: Add reserved-memory regions\n \n Add the reserved memory regions in SM8150\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n Reviewed-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 22076f3bbb06e86a4907e825ac9f305413a5bbeb\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Thu Aug 22 00:12:37 2019 +0530\n\n arm64: dts: qcom: sm8150-mtp: Add regulators\n \n Add the regulators found in the mtp platform. This platform consists of\n pmic PM8150, PM8150L and PM8009.\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n Reviewed-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 63db27709e48d9981b02511bc73fcc5112b6b038\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Thu Aug 22 00:12:36 2019 +0530\n\n arm64: dts: qcom: sm8150-mtp: Add base dts file\n \n This add base DTS file for sm8150-mtp and enables boot to console, adds\n tlmm reserved range, resin node, volume down key and also includes pmic\n file.\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n Reviewed-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit e92b61c8e77596e5cbe0b5e4a96f651ed7fb97fc\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Thu Aug 22 00:12:35 2019 +0530\n\n arm64: dts: qcom: pm8150l: Add base dts file\n \n PMIC pm8150l is a slave pmic and this adds base DTS file for pm8150l\n with power-on, adc and gpio nodes\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n Reviewed-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 229d5bcad0d038c836dd036aa8dd315a88b48df0\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Thu Aug 22 00:12:34 2019 +0530\n\n arm64: dts: qcom: pm8150b: Add base dts file\n \n PMIC pm8150b is a slave pmic and this adds base DTS file for pm8150b\n with power-on, adc, and gpio nodes\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n Reviewed-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 5101f22a5c37c01677785f61a3ede97737e7c832\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Thu Aug 22 00:12:33 2019 +0530\n\n arm64: dts: qcom: pm8150: Add base dts file\n \n Add base DTS file for pm8150 along with GPIOs, power-on, rtc and vadc\n nodes\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n Reviewed-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit e13c6d144fa0fd378c4062459736b4da28555414\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Thu Aug 22 00:12:32 2019 +0530\n\n arm64: dts: qcom: sm8150: Add base dts file\n \n This add base DTS file with cpu, psci, firmware, clock node tlmm and\n spmi and enables boot to console\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n Reviewed-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit b90ff3554aa3e123bb7e6d08789f6fd92d86ddde\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Wed Aug 21 11:40:19 2019 +0200\n\n ACPI: PM: s2idle: Always set up EC GPE for system wakeup\n \n Commit 10a08fd65ec1 (\"ACPI: PM: Set up EC GPE for system wakeup from\n drivers that need it\") assumed that the EC GPE would only need to be\n set up for system wakeup if either the intel-hid or the intel-vbtn\n driver was in use, but that turns out to be incorrect. In particular,\n on ASUS Zenbook UX430UNR/i7-8550U, if the EC GPE is not enabled while\n suspended, the system cannot be woken up by opening the lid or\n pressing a key, and that machine doesn\u0027t use any of the drivers\n mentioned above.\n \n For this reason, always set up the EC GPE for system wakeup from\n suspend-to-idle by setting and clearing its wake mask in the ACPI\n suspend-to-idle callbacks.\n \n Fixes: 10a08fd65ec1 (\"ACPI: PM: Set up EC GPE for system wakeup from drivers that need it\")\n Reported-by: Kristian Klausen \u003ckristian@klausen.dk\u003e\n Tested-by: Kristian Klausen \u003ckristian@klausen.dk\u003e\n Acked-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 45dc1576e4575ba621cb6d017faf41531d8c1073\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Mon Aug 19 12:35:03 2019 +0200\n\n ACPI: PM: s2idle: Avoid rearming SCI for wakeup unnecessarily\n \n It is only necessary to rearm the ACPI SCI for wakeup if\n pm_system_cancel_wakeup() has been called, so invoke\n rearm_wake_irq() only in that case.\n \n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 01faced6f65d0224bf6ce1262a4137771e28519f\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 15:26:19 2019 -0300\n\n media: dt-bindings: media: Convert Allwinner A10 IR to a schema\n \n The older Allwinner SoCs have a IR receiver supported in Linux, with a\n matching Device Tree binding.\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for that controller over to a YAML schemas.\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 7c31b9d6734241480578de0b2c6bfa6d7e85dc96\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 15:26:18 2019 -0300\n\n media: dt-bindings: media: Add YAML schemas for the generic RC bindings\n \n The RC controllers have a bunch of generic properties that are needed in a\n device tree. Add a YAML schemas for those.\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit a10feaf8c464c3f9cfdd3a8a7ce17e1c0d498da1\nAuthor: Tomas Bortoli \u003ctomasbortoli@gmail.com\u003e\nDate: Wed Jul 31 12:19:05 2019 -0300\n\n media: ttusb-dec: Fix info-leak in ttusb_dec_send_command()\n \n The function at issue does not always initialize each byte allocated\n for \u0027b\u0027 and can therefore leak uninitialized memory to a USB device in\n the call to usb_bulk_msg()\n \n Use kzalloc() instead of kmalloc()\n \n Signed-off-by: Tomas Bortoli \u003ctomasbortoli@gmail.com\u003e\n Reported-by: syzbot+0522702e9d67142379f1@syzkaller.appspotmail.com\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 6f005abb00848556a6bff1d3bc8a76f7dc2f9c24\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Aug 19 04:47:01 2019 -0300\n\n media: dvb-frontends/cxd2099: Make en_templ constant\n \n Static structure en_templ, of type dvb_ca_en50221, is not used except to\n be copied into a local variable. Hence make it const to prevent\n unintended modification of the original fields.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 05fa782e2547e2fa2d9233182abdb406605346e0\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Aug 19 04:38:18 2019 -0300\n\n media: ec168: Make structure ec168_props constant\n \n The static structure ec168_props, of type dvb_usb_device_properties, is\n used only once, when it is assigned to field props of the constant\n structure ec168_driver_info. As this means that ec168_props will never be\n modified, make it const as well in order to prevent unintended\n modification of its fields.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit a37c2efc087ce849e75ee135e1a52fd1fb2f81eb\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Aug 19 04:36:54 2019 -0300\n\n media: firewire: Make structure fdtv_ca constant\n \n Static structure fdtv_ca, of type dvb_device, is only used when it is\n passed as the third argument to function dvb_register_device(). However,\n in the definition of dvb_register_device, the third parameter is\n declared as constant. Hence make fdtv_ca const as well.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit b501b09ae8c02bc38479c18b9ccb9946c6f252dc\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Aug 19 04:35:42 2019 -0300\n\n media: dvb-bt8xx: Make variable dst_config constant\n \n Static structure dst_config, of type dst_config, is not used except to\n be assigned as the value of field state of a variable having type\n dst_state *. In the definition of dst_state, field config is declared as\n const. Hence dst_config, when assigned to config, cannot be modified.\n Therefore, make dst_config const as well.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit a9547e520b423a4eb89186b86e7afa937405f3a2\nAuthor: Jan Pieter van Woerkom \u003cjp@jpvw.nl\u003e\nDate: Thu Aug 15 13:37:33 2019 -0300\n\n media: dvb-usb: add T230 to dvbsky\n \n commit 5fa88151ecdb (\"[media] dvb-usb-cxusb: Geniatech T230 - resync TS\n FIFO after lock\") does not solve the problem for all devices. This is the\n same issue reported on the T230C, which was moved to the dvbsky.c driver\n to work around this issue in commit 5742240577b6 (\"media: dvbsky: MyGica\n T230C support\").\n \n In addition, the cxusb driver causes an oops when unplugging the device\n while streaming; this problem does not happen any more.\n \n Signed-off-by: Jan Pieter van Woerkom \u003cjp@jpvw.nl\u003e\n Tested-by: James Hutchinson \u003cjahutchinson99@googlemail.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit fe087322f289718a39b4c947adddd1ce0eab801c\nAuthor: Jan Pieter van Woerkom \u003cjp@jpvw.nl\u003e\nDate: Thu Aug 15 13:41:58 2019 -0300\n\n media: dvb-usb: remove T230 from cxusb\n \n Remove this device from the cxusb driver so it can be added to the dvbsky\n driver.\n \n Signed-off-by: Jan Pieter van Woerkom \u003cjp@jpvw.nl\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit fcd5ce4b3936242e6679875a4d3c3acfc8743e15\nAuthor: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nDate: Sun Aug 18 00:45:40 2019 -0300\n\n media: dvb-core: fix a memory leak bug\n \n In dvb_create_media_entity(), \u0027dvbdev-\u003eentity\u0027 is allocated through\n kzalloc(). Then, \u0027dvbdev-\u003epads\u0027 is allocated through kcalloc(). However, if\n kcalloc() fails, the allocated \u0027dvbdev-\u003eentity\u0027 is not deallocated, leading\n to a memory leak bug. To fix this issue, free \u0027dvbdev-\u003eentity\u0027 before\n returning -ENOMEM.\n \n Signed-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit b1da86fce4299ab1060a5e878696d29ce29fed6e\nAuthor: Sean Young \u003csean@mess.org\u003e\nDate: Sat Aug 17 09:02:40 2019 -0300\n\n media: ttpci: unknown protocol is rc-mm-32\n \n This protocol responds to a real philips rc-mm remote; it does not respond\n to IR encoded with the encoder in ir-rcmm-decoder.c.\n \n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 2f6451ed8b5223a28eab81116dd2f901c0dc9f4a\nAuthor: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nDate: Sat Aug 17 04:04:04 2019 -0300\n\n media: dvb-frontends: fix a memory leak bug\n \n In cx24117_load_firmware(), \u0027buf\u0027 is allocated through kmalloc() to hold\n the firmware. However, if i2c_transfer() fails, it is not deallocated,\n leading to a memory leak bug.\n \n Signed-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 8c3d3cdbd5dde9bc87d556d2daa9d771ac5254dc\nAuthor: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nDate: Sat Aug 17 03:42:24 2019 -0300\n\n media: dvb-frontends: fix memory leaks\n \n In dib7000pc_detection(), \u0027tx\u0027 and \u0027rx\u0027 are allocated through kzalloc()\n respectively. However, if DiB7000PC is detected, they are not deallocated,\n leading to memory leaks. To fix this issue, create a label to free \u0027tx\u0027 and\n \u0027rx\u0027 before returning from the function.\n \n Signed-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 9fc3ce31f5bde660197f35135e90a1cced58aa2c\nAuthor: A Sun \u003cas1033x@comcast.net\u003e\nDate: Thu Aug 15 13:41:19 2019 -0300\n\n media: mceusb: fix (eliminate) TX IR signal length limit\n \n Fix and eliminate mceusb\u0027s IR length limit for IR signals transmitted to\n the MCE IR blaster ports.\n \n An IR signal TX exceeding 306 pulse/space samples presently causes -EINVAL\n return error. There\u0027s no such limitation nor error with the MCE device\n hardware. And valid IR signals exist with more than 400 pulse/space for the\n control of certain appliances (eg Panasonic ACXA75C00600 air conditioner).\n \n The scope of this patch is limited to the mceusb driver. There are still\n IR signal TX length and time constraints that related modules of rc core\n (eg LIRC) impose, further up the driver stack.\n \n Changes for mceusb_tx_ir():\n \n Converts and sends LIRC IR pulse/space sequence to MCE device IR\n pulse/space format.\n \n Break long length LIRC sequence into multiple (unlimited number of) parts\n for sending to the MCE device.\n Reduce kernel stack IR buffer size: 128 (was 384)\n Increase MCE IR data packet size: 31 (was 5)\n Zero time LIRC pulse/space no longer copied to MCE IR data.\n Eliminate overwriting the source/input LIRC IR data in txbuf[].\n Eliminate -EINVAL return; return number of IR samples sent (\u003e0) or\n MCE write error code (\u003c0).\n \n New mce_write() and mce_write_callback():\n \n Implements synchronous blocking I/O, with timeout, for writing/sending\n data to the MCE device.\n \n An unlimited multipart IR signal sent to the MCE device faster than real\n time requires flow control absent with the original mce_request_packet()\n and mce_async_callback() asynchronous I/O implementation. Also absent is\n TX error feedback.\n \n mce_write() combines and replaces mce_request_packet() and\n mce_async_callback() with conversion to synchronous I/O.\n mce_write() returns bytes sent (\u003e0) or MCE device write error (\u003c0).\n Debug hex dump TX data before processing.\n \n Rename mce_async_out() -\u003e mce_command_out():\n \n The original name is misleading with underlying synchronous I/O\n implementation. Function renamed to mce_command_out().\n \n Changes in mceusb_handle_command():\n \n Add support for MCE device error case MCE_RSP_TX_TIMEOUT\n \"IR TX timeout (TX buffer underrun)\"\n \n Changes in mceusb_dev_printdata():\n \n Changes support test and debug of multipart TX IR.\n \n Add buffer boundary information (offset and buffer size) to TX hex dump.\n Correct TX trace bug \"Raw IR data, 0 pulse/space samples\"\n Add trace for MCE_RSP_TX_TIMEOUT \"IR TX timeout (TX buffer underrun)\"\n \n Other changes:\n \n The driver\u0027s write to USB device architecture change (async to sync I/O)\n is significant so we bump DRIVER_VERSION to \"1.95\" (from \"1.94\").\n \n Tests:\n \n $ cat -n irdata1 | head -3\n 1 carrier 36000\n 2 pulse 6350\n 3 space 6350\n $ cat -n irdata1 | tail -3\n 76 pulse 6350\n 77 space 6350\n 78 pulse 6350\n $ ir-ctl -s irdata1\n \n [1549021.073612] mceusb 1-1.3:1.0: requesting 36000 HZ carrier\n [1549021.073635] mceusb 1-1.3:1.0: tx data[0]: 9f 06 01 45 (len\u003d4 sz\u003d4)\n [1549021.073649] mceusb 1-1.3:1.0: Request carrier of 35714 Hz (period 28us)\n [1549021.073848] mceusb 1-1.3:1.0: tx done status \u003d 4 (wait \u003d 100, expire \u003d 100 (1000ms), urb-\u003eactual_length \u003d 4, urb-\u003estatus \u003d 0)\n [1549021.074689] mceusb 1-1.3:1.0: rx data[0]: 9f 06 01 45 (len\u003d4 sz\u003d4)\n [1549021.074701] mceusb 1-1.3:1.0: Got carrier of 35714 Hz (period 28us)\n [1549021.102023] mceusb 1-1.3:1.0: tx data[0]: 9f 08 03 (len\u003d3 sz\u003d3)\n [1549021.102036] mceusb 1-1.3:1.0: Request transmit blaster mask of 0x03\n [1549021.102219] mceusb 1-1.3:1.0: tx done status \u003d 3 (wait \u003d 100, expire \u003d 100 (1000ms), urb-\u003eactual_length \u003d 3, urb-\u003estatus \u003d 0)\n [1549021.131979] mceusb 1-1.3:1.0: tx data[0]: 9e ff 7f ff 7f ff 7f ff 7f ff 7f ff 7f ff 7f ff 7f ff 7f ff 7f ff 7f ff 7f ff 7f ff 7f ff 7f 9e ff 7f ff 7f ff 7f ff 7f ff 7f ff 7f ff 7f ff 7f ff 7f ff 7f ff 7f ff 7f ff 7f ff 7f ff 7f 91 ff (len\u003d81 sz\u003d81)\n [1549021.131992] mceusb 1-1.3:1.0: Raw IR data, 30 pulse/space samples\n [1549021.133592] mceusb 1-1.3:1.0: tx done status \u003d 81 (wait \u003d 100, expire \u003d 100 (1000ms), urb-\u003eactual_length \u003d 81, urb-\u003estatus \u003d 0)\n \n Hex dumps limited to 64 bytes.\n 0xff is MCE maximum time pulse, 0x7f is MCE maximum time space.\n \n $ cat -n irdata2 | head -3\n 1 carrier 36000\n 2 pulse 50\n 3 space 50\n $ cat -n irdata2 | tail -3\n 254 pulse 50\n 255 space 50\n 256 pulse 50\n $ ir-ctl -s irdata2\n \n [1549306.586998] mceusb 1-1.3:1.0: tx data[0]: 9f 08 03 (len\u003d3 sz\u003d3)\n [1549306.587015] mceusb 1-1.3:1.0: Request transmit blaster mask of 0x03\n [1549306.587252] mceusb 1-1.3:1.0: tx done status \u003d 3 (wait \u003d 100, expire \u003d 100 (1000ms), urb-\u003eactual_length \u003d 3, urb-\u003estatus \u003d 0)\n [1549306.613275] mceusb 1-1.3:1.0: tx data[0]: 9e 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 9e 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 9e 81 (len\u003d128 sz\u003d128)\n [1549306.613291] mceusb 1-1.3:1.0: Raw IR data, 30 pulse/space samples\n [1549306.614837] mceusb 1-1.3:1.0: tx done status \u003d 128 (wait \u003d 100, expire \u003d 100 (1000ms), urb-\u003eactual_length \u003d 128, urb-\u003estatus \u003d 0)\n [1549306.614861] mceusb 1-1.3:1.0: tx data[0]: 9e 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 9e 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 01 81 9e 01 (len\u003d128 sz\u003d128)\n [1549306.614869] mceusb 1-1.3:1.0: Raw IR data, 30 pulse/space samples\n [1549306.620199] mceusb 1-1.3:1.0: tx done status \u003d 128 (wait \u003d 100, expire \u003d 100 (1000ms), urb-\u003eactual_length \u003d 128, urb-\u003estatus \u003d 0)\n [1549306.620212] mceusb 1-1.3:1.0: tx data[0]: 89 81 01 81 01 81 01 81 01 81 80 (len\u003d11 sz\u003d11)\n [1549306.620221] mceusb 1-1.3:1.0: Raw IR data, 9 pulse/space samples\n [1549306.633294] mceusb 1-1.3:1.0: tx done status \u003d 11 (wait \u003d 98, expire \u003d 100 (1000ms), urb-\u003eactual_length \u003d 11, urb-\u003estatus \u003d 0)\n \n Hex dumps limited to 64 bytes.\n 0x81 is MCE minimum time pulse, 0x01 is MCE minimum time space.\n TX IR part 3 sz\u003d11 shows 20msec I/O blocking delay\n (100expire - 98wait \u003d 2jiffies)\n \n Signed-off-by: A Sun \u003cas1033x@comcast.net\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 8f5f33f71caefd85befddcf902df96de362c9d18\nAuthor: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\nDate: Thu Aug 15 11:59:17 2019 -0300\n\n media: rc: add keymap for HardKernel ODROID remote\n \n This is a simple NEC remote control device shipped with the HardKernel\n ODROID range of SBC devices.\n \n Signed-off-by: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit fa992b335aa778db112e91aac19c45e4d914a1f4\nAuthor: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\nDate: Thu Aug 15 11:59:16 2019 -0300\n\n media: rc: add keymap for WeTeK Play 2 remote\n \n The WeTek Play 2 Android STB ships with an unusual remote where the\n main up/down/left/right/enter controls are surrounded with an outer\n ring of additional keys which are listed in clockwise order.\n \n Signed-off-by: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 373078971272e9dfcb5f80e309f148a88e36dba8\nAuthor: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\nDate: Thu Aug 15 11:59:15 2019 -0300\n\n media: rc: add keymap for WeTek Hub remote\n \n The WeTek Hub Android STB ships with a simple NEC remote.\n \n Signed-off-by: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 7bb53f361c59b68e521a05fce579ccfa8021c3a0\nAuthor: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\nDate: Thu Aug 15 11:59:14 2019 -0300\n\n media: rc: add keymap for Tanix TX5 max remote\n \n The Tanix TX5 max Android STB ships with a simple NEC remote.\n \n Signed-off-by: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit e30399e1bd6e215ec20981612646ec73a4385c33\nAuthor: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\nDate: Thu Aug 15 11:59:13 2019 -0300\n\n media: rc: add keymap for Tanix TX3 mini remote\n \n The Tanix TX3 mini Android STB ships with a simple NEC remote.\n \n Signed-off-by: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit a53dee7b70384840b13cbe92588fa68dce2c0b62\nAuthor: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\nDate: Thu Aug 15 11:59:12 2019 -0300\n\n media: rc: add keymap for Khadas VIM/EDGE remote\n \n Khadas VIM and Edge SBC devices use the same NEC remote device. The\n remote includes a mouse button for Android use. This has been mapped\n to KEY_MUTE.\n \n Signed-off-by: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 9b9e9e5b461789668ce0271f968d8eb5c5d3a332\nAuthor: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\nDate: Thu Aug 15 11:59:11 2019 -0300\n\n media: rc: add keymap for Amediatech X96-MAX remote\n \n The X96-Max Android STB ships with a simple NEC remote. It includes\n a TV section with preset buttons for controlling a TV. These are not\n configurable, but are noted to aid visual recognition of the device.\n \n Signed-off-by: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 619f6fc390909fce3247c3e07e6882b868b645da\nAuthor: Marc Gonzalez \u003cmarc.w.gonzalez@free.fr\u003e\nDate: Mon Jul 15 06:50:29 2019 -0300\n\n media: si2168: Refactor command setup code\n \n Use cmd_init() to fill a struct si2168_cmd command.\n \n Signed-off-by: Marc Gonzalez \u003cmarc.w.gonzalez@free.fr\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit c268e7adea52be0093de1164c425f3c8d8927770\nAuthor: Sean Young \u003csean@mess.org\u003e\nDate: Sat Aug 3 00:12:03 2019 -0300\n\n media: dvb-frontends: use ida for pll number\n \n KASAN: global-out-of-bounds Read in dvb_pll_attach\n \n Syzbot reported global-out-of-bounds Read in dvb_pll_attach, while\n accessing id[dvb_pll_devcount], because dvb_pll_devcount was 65,\n that is more than size of \u0027id\u0027 which is DVB_PLL_MAX(64).\n \n Rather than increasing dvb_pll_devcount every time, use ida so that\n numbers are allocated correctly. This does mean that no more than\n 64 devices can be attached at the same time, but this is more than\n sufficient.\n \n usb 1-1: dvb_usb_v2: will pass the complete MPEG2 transport stream to the\n software demuxer\n dvbdev: DVB: registering new adapter (774 Friio White ISDB-T USB2.0)\n usb 1-1: media controller created\n dvbdev: dvb_create_media_entity: media entity \u0027dvb-demux\u0027 registered.\n tc90522 0-0018: Toshiba TC90522 attached.\n usb 1-1: DVB: registering adapter 0 frontend 0 (Toshiba TC90522 ISDB-T\n module)...\n dvbdev: dvb_create_media_entity: media entity \u0027Toshiba TC90522 ISDB-T\n module\u0027 registered.\n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n BUG: KASAN: global-out-of-bounds in dvb_pll_attach+0x6c5/0x830\n drivers/media/dvb-frontends/dvb-pll.c:798\n Read of size 4 at addr ffffffff89c9e5e0 by task kworker/0:1/12\n \n CPU: 0 PID: 12 Comm: kworker/0:1 Not tainted 5.2.0-rc6+ #13\n Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS\n Google 01/01/2011\n Workqueue: usb_hub_wq hub_event\n Call Trace:\n __dump_stack lib/dump_stack.c:77 [inline]\n dump_stack+0xca/0x13e lib/dump_stack.c:113\n print_address_description+0x67/0x231 mm/kasan/report.c:188\n __kasan_report.cold+0x1a/0x32 mm/kasan/report.c:317\n kasan_report+0xe/0x20 mm/kasan/common.c:614\n dvb_pll_attach+0x6c5/0x830 drivers/media/dvb-frontends/dvb-pll.c:798\n dvb_pll_probe+0xfe/0x174 drivers/media/dvb-frontends/dvb-pll.c:877\n i2c_device_probe+0x790/0xaa0 drivers/i2c/i2c-core-base.c:389\n really_probe+0x281/0x660 drivers/base/dd.c:509\n driver_probe_device+0x104/0x210 drivers/base/dd.c:670\n __device_attach_driver+0x1c2/0x220 drivers/base/dd.c:777\n bus_for_each_drv+0x15c/0x1e0 drivers/base/bus.c:454\n __device_attach+0x217/0x360 drivers/base/dd.c:843\n bus_probe_device+0x1e4/0x290 drivers/base/bus.c:514\n device_add+0xae6/0x16f0 drivers/base/core.c:2111\n i2c_new_client_device+0x5b3/0xc40 drivers/i2c/i2c-core-base.c:778\n i2c_new_device+0x19/0x50 drivers/i2c/i2c-core-base.c:821\n dvb_module_probe+0xf9/0x220 drivers/media/dvb-core/dvbdev.c:985\n friio_tuner_attach+0x125/0x1d0 drivers/media/usb/dvb-usb-v2/gl861.c:536\n dvb_usbv2_adapter_frontend_init\n drivers/media/usb/dvb-usb-v2/dvb_usb_core.c:675 [inline]\n dvb_usbv2_adapter_init drivers/media/usb/dvb-usb-v2/dvb_usb_core.c:804\n [inline]\n dvb_usbv2_init drivers/media/usb/dvb-usb-v2/dvb_usb_core.c:865 [inline]\n dvb_usbv2_probe.cold+0x24dc/0x255d\n drivers/media/usb/dvb-usb-v2/dvb_usb_core.c:980\n usb_probe_interface+0x305/0x7a0 drivers/usb/core/driver.c:361\n really_probe+0x281/0x660 drivers/base/dd.c:509\n driver_probe_device+0x104/0x210 drivers/base/dd.c:670\n __device_attach_driver+0x1c2/0x220 drivers/base/dd.c:777\n bus_for_each_drv+0x15c/0x1e0 drivers/base/bus.c:454\n __device_attach+0x217/0x360 drivers/base/dd.c:843\n bus_probe_device+0x1e4/0x290 drivers/base/bus.c:514\n device_add+0xae6/0x16f0 drivers/base/core.c:2111\n usb_set_configuration+0xdf6/0x1670 drivers/usb/core/message.c:2023\n generic_probe+0x9d/0xd5 drivers/usb/core/generic.c:210\n usb_probe_device+0x99/0x100 drivers/usb/core/driver.c:266\n really_probe+0x281/0x660 drivers/base/dd.c:509\n driver_probe_device+0x104/0x210 drivers/base/dd.c:670\n __device_attach_driver+0x1c2/0x220 drivers/base/dd.c:777\n bus_for_each_drv+0x15c/0x1e0 drivers/base/bus.c:454\n __device_attach+0x217/0x360 drivers/base/dd.c:843\n bus_probe_device+0x1e4/0x290 drivers/base/bus.c:514\n device_add+0xae6/0x16f0 drivers/base/core.c:2111\n usb_new_device.cold+0x8c1/0x1016 drivers/usb/core/hub.c:2534\n hub_port_connect drivers/usb/core/hub.c:5089 [inline]\n hub_port_connect_change drivers/usb/core/hub.c:5204 [inline]\n port_event drivers/usb/core/hub.c:5350 [inline]\n hub_event+0x1ada/0x3590 drivers/usb/core/hub.c:5432\n process_one_work+0x905/0x1570 kernel/workqueue.c:2269\n process_scheduled_works kernel/workqueue.c:2331 [inline]\n worker_thread+0x7ab/0xe20 kernel/workqueue.c:2417\n kthread+0x30b/0x410 kernel/kthread.c:255\n ret_from_fork+0x24/0x30 arch/x86/entry/entry_64.S:352\n \n The buggy address belongs to the variable:\n id+0x100/0x120\n \n Memory state around the buggy address:\n ffffffff89c9e480: fa fa fa fa 00 00 fa fa fa fa fa fa 00 00 00 00\n ffffffff89c9e500: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00\n \u003e ffffffff89c9e580: 00 00 00 00 00 00 00 00 00 00 00 00 fa fa fa fa\n ^\n ffffffff89c9e600: 04 fa fa fa fa fa fa fa 04 fa fa fa fa fa fa fa\n ffffffff89c9e680: 04 fa fa fa fa fa fa fa 04 fa fa fa fa fa fa fa\n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n \n Reported-by: syzbot+8a8f48672560c8ca59dd@syzkaller.appspotmail.com\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 8e40983dec63300013ff5755c36e3ac455e351a5\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Aug 21 20:38:51 2019 +0100\n\n drm/i915/selftests: Fixup a couple of missing serialisation with vma\n \n In commit 70d6894d1456 (\"drm/i915: Serialize against vma moves\")\n I managed to miss a couple of i915_vma_move_to_active() that had not\n serialised against an async vma pinning. Add the missing\n i915_request_await.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190821193851.18232-1-chris@chris-wilson.co.uk\n\ncommit 2b9b5e74507fe8e6146b048c0dadbe2fe7b298e5\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 21 21:57:01 2019 +0800\n\n net: stmmac: dwc-qos: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ad124aa34e514393218f46f9701fdeaf35cbf5d3\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 21 21:55:50 2019 +0800\n\n net: stmmac: dwmac-anarion: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f33bf6b00f20c9d26c42dfdaf8b83c2b0c1e6f71\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 21 21:54:06 2019 +0800\n\n net: stmmac: dwmac-meson: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 999232a38ff10209df1ce40cabe8cf8982c530e7\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 21 21:51:30 2019 +0800\n\n net: stmmac: dwmac-meson8b: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 913919e51ee69df7a3250056c522d683c937bf46\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 21 21:46:13 2019 +0800\n\n net: systemport: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Reviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4ca3348dff9ac63a571288019f6030a3d7e1e940\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 21 21:41:31 2019 +0800\n\n net: bcmgenet: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Reviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 531fd23b00610fb066da7776ea25db8aa383a4b8\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 21 21:38:54 2019 +0800\n\n pxa168_eth: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 37f76049b095da153efa047538f73b1af8dfbc6a\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 21 21:33:02 2019 +0800\n\n net: mvneta: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4f830a5af7b5b3614a03ef784cd2f4a83ceebf27\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 21 21:29:45 2019 +0800\n\n net: fec: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b6df983076e517af660e14a3ed6d3b1a986dfc2c\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 21 21:05:09 2019 +0800\n\n ezchip: nps_enet: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1a1ba7118807e82fab7d44ccf0910d3bb0015b0b\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 21 21:02:41 2019 +0800\n\n cirrus: cs89x0: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ffb36a10c656c0b106401c6792b55b6506078949\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 21 20:59:59 2019 +0800\n\n net: sxgbe: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4865695c0f1aa302db2b817be3ef6b0cba53763b\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 21 20:58:11 2019 +0800\n\n net: ks8851-ml: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 5bd5b56457b0415c82f359ad67dc38d1b7eb381d\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 21 20:53:57 2019 +0800\n\n net: socionext: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 8a54d4c21919022e9fc59e71ca0f86d4431a8e2a\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 21 20:50:50 2019 +0800\n\n via-rhine: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c8ace62ff3a98bdc52c6df23c23465cfd9e3a847\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 21 20:48:50 2019 +0800\n\n net: ethernet: ti: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Reviewed-by: Grygorii Strashko \u003cgrygorii.strashko@ti.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit eba39fd6fe310710f011130e46e6d13e5c87d6f9\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 21 20:32:03 2019 +0800\n\n amd-xgbe: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Acked-by: Tom Lendacky \u003cthomas.lendacky@amd.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1e672e3644940d83bd94e7cb46bac6bb3627de02\nAuthor: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nDate: Tue Aug 20 22:21:21 2019 -0500\n\n NFSv4: Fix a memory leak bug\n \n In nfs4_try_migration(), if nfs4_begin_drain_session() fails, the\n previously allocated \u0027page\u0027 and \u0027locations\u0027 are not deallocated, leading to\n memory leaks. To fix this issue, go to the \u0027out\u0027 label to free \u0027page\u0027 and\n \u0027locations\u0027 before returning the error.\n \n Signed-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit 1455865a040a0f632dce5d8f7cb6604517f56c6d\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Wed Aug 21 10:19:37 2019 +0300\n\n selftests: mlxsw: Add a test case for devlink-trap\n \n Test generic devlink-trap functionality over mlxsw. These tests are not\n specific to a single trap, but do not check the devlink-trap common\n infrastructure either.\n \n Currently, the only test case is device deletion (by reloading the\n driver) while packets are being trapped.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 170270329b1b5f854e8a98aea16bfa4af6922146\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Wed Aug 21 10:19:36 2019 +0300\n\n selftests: mlxsw: Add test cases for devlink-trap L2 drops\n \n Test that each supported packet trap is triggered under the right\n conditions and that packets are indeed dropped and not forwarded.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b5ce611fd96e18711d1b580170676fba9cafe258\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Wed Aug 21 10:19:35 2019 +0300\n\n mlxsw: spectrum: Add devlink-trap support\n \n Register supported packet traps (layer 2 drops only, currently) and\n associated trap group with devlink during driver initialization.\n \n The amount of traffic generated by these packet drop traps is capped at\n 10Kpps to ensure the CPU is not overwhelmed by incoming packets.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9e6290c75a5078d0a8fac89d908dc59db9c27a15\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Wed Aug 21 10:19:34 2019 +0300\n\n mlxsw: Add trap group for layer 2 discards\n \n Discard trap groups are defined in a different enum so that they could\n all share the same policer ID: MLXSW_REG_HTGT_TRAP_GROUP_MAX + 1.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a812cedb8e3594833986fe690b0224d5577644ce\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Wed Aug 21 10:19:33 2019 +0300\n\n mlxsw: Add layer 2 discard trap IDs\n \n Add the trap IDs used to report layer 2 drops.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6a44bae3b2b9b03c03712c0460f674efaf684131\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Wed Aug 21 10:19:32 2019 +0300\n\n mlxsw: reg: Add new trap actions\n \n Subsequent patches will add discard traps support in mlxsw. The driver\n cannot configure such traps with a normal trap action, but needs to use\n exception trap action, which also increments an error counter.\n \n On the other hand, when these traps are initialized or set to drop\n action, they should use the default drop action set by the firmware.\n This guarantees that when the feature is disabled we get the exact same\n behavior as before the feature was introduced.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b7bf027087cb81b4673515ea8a0e3850ab2f9055\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Wed Aug 21 10:19:31 2019 +0300\n\n mlxsw: core: Add API to set trap action\n \n Up until now the action of a trap was never changed during its lifetime.\n This is going to change by subsequent patches that will allow devlink to\n control the action of certain traps.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 0bb52b0dfc88a155688f492aba8e686147600278\nAuthor: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\nDate: Wed Aug 21 09:52:19 2019 +0100\n\n tools: bpftool: add \"bpftool map freeze\" subcommand\n \n Add a new subcommand to freeze maps from user space.\n \n Signed-off-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit c354ff2ef233a694d657b03a499c3c3fddbec599\nAuthor: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\nDate: Wed Aug 21 09:52:18 2019 +0100\n\n tools: bpftool: show frozen status for maps\n \n When listing maps, read their \"frozen\" status from procfs, and tell if\n maps are frozen.\n \n As commit log for map freezing command mentions that the feature might\n be extended with flags (e.g. for write-only instead of read-only) in the\n future, use an integer and not a boolean for JSON output.\n \n Signed-off-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 435eba4ae0692e2f3d62988f8648efd65c935b6a\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:51:49 2019 -0400\n\n xprtrdma: Optimize rpcrdma_post_recvs()\n \n Micro-optimization: In rpcrdma_post_recvs, since commit e340c2d6ef2a\n (\"xprtrdma: Reduce the doorbell rate (Receive)\"), the common case is\n to return without doing anything. Found with perf.\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit f981322804ca39bb836cf3fbfdebbc47d6584731\nAuthor: Fabien Parent \u003cfparent@baylibre.com\u003e\nDate: Mon Aug 5 15:02:15 2019 +0200\n\n dt-bindings: rng: mtk-rng: Add documentation for MT8516\n \n This commit adds the device-tree documentation for the RNG IP on the\n MediaTek MT8516 SoC.\n \n Signed-off-by: Fabien Parent \u003cfparent@baylibre.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 1738de336ebc9f8d4bb1b3126c5417a5923a660a\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:51:03 2019 -0400\n\n xprtrdma: Inline XDR chunk encoder functions\n \n Micro-optimization: Save the cost of three function calls during\n transport header encoding.\n \n These were \"noinline\" before to generate more meaningful call stacks\n during debugging, but this code is now pretty stable.\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit 17d47f93bc69cc629aa6a20c0ac7b20c9965c116\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:50:16 2019 -0400\n\n xprtrdma: Fix bc_max_slots return value\n \n For the moment the returned value just happens to be correct because\n the current backchannel server implementation does not vary the\n number of credits it offers. The spec does permit this value to\n change during the lifetime of a connection, however.\n \n The actual maximum is fixed for all RPC/RDMA transports, because\n each transport instance has to pre-allocate the resources for\n processing BC requests. That\u0027s the value that should be returned.\n \n Fixes: 7402a4fedc2b (\"SUNRPC: Fix up backchannel slot table ... \")\n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit 3ffce7b6100cbfc1dcbec8be2b824b2d75866455\nAuthor: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nDate: Wed Jul 31 14:25:35 2019 +0200\n\n dt-bindings: ata: fix typo in Allwinner R40 reset specific paragraph\n \n \"Property\" is misspelled, correct it.\n \n Fixes: caf8975b97c1 (\"dt-bindings: ata: update ahci_sunxi bindings\")\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit dce3e8fd039cc1b62760b3ad6822cf04c262cd0e\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Aug 19 16:31:47 2019 +0200\n\n posix-cpu-timers: Remove tsk argument from run_posix_cpu_timers()\n \n It\u0027s always current. Don\u0027t give people wrong ideas.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190819143801.945469967@linutronix.de\n\ncommit 692117c1f7a6770ed41dd8f277cd9fed1dfb16f1\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Aug 19 16:31:46 2019 +0200\n\n posix-cpu-timers: Sanitize bogus WARNONS\n \n Warning when p \u003d\u003d NULL and then proceeding and dereferencing p does not\n make any sense as the kernel will crash with a NULL pointer dereference\n right away.\n \n Bailing out when p \u003d\u003d NULL and returning an error code does not cure the\n underlying problem which caused p to be NULL. Though it might allow to\n do proper debugging.\n \n Same applies to the clock id check in set_process_cpu_timer().\n \n Clean them up and make them return without trying to do further damage.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190819143801.846497772@linutronix.de\n\ncommit 956aa03b52e9c07e6ba769afd4e6f8c43436d02c\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Jul 31 13:42:01 2019 +0200\n\n dt-bindings: arm-boards: Update pointer to ARM CPU bindings\n \n The ARM CPU DT bindings were converted from plain text to YAML, but not\n all referrers were updated.\n \n Fixes: 672951cbd1b70a9e (\"dt-bindings: arm: Convert cpu binding to json-schema\")\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Acked-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit d1dec5ca5fb87d64d1e7405bd6006728915ea5c6\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Tue Aug 20 00:56:02 2019 +0900\n\n ARM: s3c64xx: squash samsung_usb_phy.h into setup-usb-phy.c\n \n This is only used by arch/arm/mach-s3c64xx/setup-usb-phy.c\n \n $ git grep samsung_usb_phy_type\n include/linux/usb/samsung_usb_phy.h:enum samsung_usb_phy_type {\n $ git grep USB_PHY_TYPE_DEVICE\n arch/arm/mach-s3c64xx/setup-usb-phy.c: if (type \u003d\u003d USB_PHY_TYPE_DEVICE)\n arch/arm/mach-s3c64xx/setup-usb-phy.c: if (type \u003d\u003d USB_PHY_TYPE_DEVICE)\n include/linux/usb/samsung_usb_phy.h: USB_PHY_TYPE_DEVICE,\n $ git grep USB_PHY_TYPE_HOST\n include/linux/usb/samsung_usb_phy.h: USB_PHY_TYPE_HOST,\n \n Actually, \u0027enum samsung_usb_phy_type\u0027 is unused; the \u0027type\u0027 parameter\n has \u0027int\u0027 type. Anyway, there is no need to declare this enum in the\n globally visible header. Squash the header.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit b4d08173bfef9b138e55022f5a3eb90c6246e7e9\nAuthor: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\nDate: Wed Aug 21 13:50:35 2019 +0100\n\n arm64: sdm845: add adsp and cdsp fastrpc nodes\n \n Add fastrpc compute context bank nodes to both cdsp and adsp.\n \n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 6bfa3134bd3a185e98031c8a8dee18e82153df2d\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Wed Aug 21 18:11:17 2019 +0900\n\n arm64: add arch/arm64/Kbuild\n \n Use the standard obj-y form to specify the sub-directories under\n arch/arm64/. No functional change intended.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 2671828c3ff4ffadf777f793a1f3232d6e51394a\nAuthor: James Morse \u003cjames.morse@arm.com\u003e\nDate: Tue Aug 20 18:45:57 2019 +0100\n\n arm64: entry: Move ct_user_exit before any other exception\n \n When taking an SError or Debug exception from EL0, we run the C\n handler for these exceptions before updating the context tracking\n code and unmasking lower priority interrupts.\n \n When booting with nohz_full lockdep tells us we got this wrong:\n | \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n | WARNING: suspicious RCU usage\n | 5.3.0-rc2-00010-gb4b5e9dcb11b-dirty #11271 Not tainted\n | -----------------------------\n | include/linux/rcupdate.h:643 rcu_read_unlock() used illegally wh!\n |\n | other info that might help us debug this:\n |\n |\n | RCU used illegally from idle CPU!\n | rcu_scheduler_active \u003d 2, debug_locks \u003d 1\n | RCU used illegally from extended quiescent state!\n | 1 lock held by a.out/432:\n | #0: 00000000c7a79515 (rcu_read_lock){....}, at: brk_handler+0x00\n |\n | stack backtrace:\n | CPU: 1 PID: 432 Comm: a.out Not tainted 5.3.0-rc2-00010-gb4b5e9d1\n | Hardware name: ARM LTD ARM Juno Development Platform/ARM Juno De8\n | Call trace:\n | dump_backtrace+0x0/0x140\n | show_stack+0x14/0x20\n | dump_stack+0xbc/0x104\n | lockdep_rcu_suspicious+0xf8/0x108\n | brk_handler+0x164/0x1b0\n | do_debug_exception+0x11c/0x278\n | el0_dbg+0x14/0x20\n \n Moving the ct_user_exit calls to be before do_debug_exception() means\n they are also before trace_hardirqs_off() has been updated. Add a new\n ct_user_exit_irqoff macro to avoid the context-tracking code using\n irqsave/restore before we\u0027ve updated trace_hardirqs_off(). To be\n consistent, do this everywhere.\n \n The C helper is called enter_from_user_mode() to match x86 in the hope\n we can merge them into kernel/context_tracking.c later.\n \n Cc: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\n Fixes: 6c81fe7925cc4c42 (\"arm64: enable context tracking\")\n Signed-off-by: James Morse \u003cjames.morse@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 1f8919b170318e7e13e303eedac363d44057995f\nAuthor: Peter Wu \u003cpeter@lekensteyn.nl\u003e\nDate: Wed Aug 21 00:09:00 2019 +0100\n\n bpf: sync bpf.h to tools/\n \n Fix a \u0027struct pt_reg\u0027 typo and clarify when bpf_trace_printk discards\n lines. Affects documentation only.\n \n Signed-off-by: Peter Wu \u003cpeter@lekensteyn.nl\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 55c33dfbeb831eb3ab7cc1a3e295b0d4d57f23a3\nAuthor: Peter Wu \u003cpeter@lekensteyn.nl\u003e\nDate: Wed Aug 21 00:08:59 2019 +0100\n\n bpf: clarify when bpf_trace_printk discards lines\n \n I opened /sys/kernel/tracing/trace once and kept reading from it.\n bpf_trace_printk somehow did not seem to work, no entries were appended\n to that trace file. It turns out that tracing is disabled when that file\n is open. Save the next person some time and document this.\n \n The trace file is described in Documentation/trace/ftrace.rst, however\n the implication \"tracing is disabled\" did not immediate translate to\n \"bpf_trace_printk silently discards entries\".\n \n Signed-off-by: Peter Wu \u003cpeter@lekensteyn.nl\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 8050a395112db5bedb7caa6cb673e711a3c04cd9\nAuthor: Peter Wu \u003cpeter@lekensteyn.nl\u003e\nDate: Wed Aug 21 00:08:58 2019 +0100\n\n bpf: fix \u0027struct pt_reg\u0027 typo in documentation\n \n There is no \u0027struct pt_reg\u0027.\n \n Signed-off-by: Peter Wu \u003cpeter@lekensteyn.nl\u003e\n Reviewed-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 5cbd22c179012114099fe3996999b54fd2a092c5\nAuthor: Peter Wu \u003cpeter@lekensteyn.nl\u003e\nDate: Wed Aug 21 00:08:57 2019 +0100\n\n bpf: clarify description for CONFIG_BPF_EVENTS\n \n PERF_EVENT_IOC_SET_BPF supports uprobes since v4.3, and tracepoints\n since v4.7 via commit 04a22fae4cbc (\"tracing, perf: Implement BPF\n programs attached to uprobes\"), and commit 98b5c2c65c29 (\"perf, bpf:\n allow bpf programs attach to tracepoints\") respectively.\n \n Signed-off-by: Peter Wu \u003cpeter@lekensteyn.nl\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit d3c0444b7ea81ca75a7d111b07793926c010359f\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Wed Aug 21 18:16:05 2019 +0100\n\n security: Add hooks to rule on setting a watch\n \n Add security hooks that will allow an LSM to rule on whether or not a watch\n may be set. More than one hook is required as the watches watch different\n types of object.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n cc: Casey Schaufler \u003ccasey@schaufler-ca.com\u003e\n cc: Stephen Smalley \u003csds@tycho.nsa.gov\u003e\n cc: linux-security-module@vger.kernel.org\n\ncommit f4e8f69b757164e817a71705dc36d6f11574c3bd\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Wed Aug 21 18:16:05 2019 +0100\n\n uapi: General notification ring definitions\n \n Add UAPI definitions for the general notification ring, including the\n following pieces:\n \n (1) struct watch_notification.\n \n This is the metadata header for each entry in the ring. It includes a\n type and subtype that indicate the source of the message\n (eg. WATCH_TYPE_MOUNT_NOTIFY) and the kind of the message\n (eg. NOTIFY_MOUNT_NEW_MOUNT).\n \n The header also contains an information field that conveys the\n following information:\n \n - WATCH_INFO_LENGTH. The size of the entry (entries are variable\n length).\n \n - WATCH_INFO_ID. The watch ID specified when the watchpoint was\n set.\n \n - WATCH_INFO_TYPE_INFO. (Sub)type-specific information.\n \n - WATCH_INFO_FLAG_*. Flag bits overlain on the type-specific\n information. For use by the type.\n \n All the information in the header can be used in filtering messages at\n the point of writing into the buffer.\n \n (2) struct watch_queue_buffer.\n \n This describes the layout of the ring. Note that the first slots in\n the ring contain a special metadata entry that contains the ring\n pointers. The producer in the kernel knows to skip this and it has a\n proper header (WATCH_TYPE_META, WATCH_META_SKIP_NOTIFICATION) that\n indicates the size so that the ring consumer can handle it the same as\n any other record and just skip it.\n \n Note that this means that ring entries can never be split over the end\n of the ring, so if an entry would need to be split, a skip record is\n inserted to wrap the ring first; this is also WATCH_TYPE_META,\n WATCH_META_SKIP_NOTIFICATION.\n \n (3) WATCH_INFO_NOTIFICATIONS_LOST.\n \n This is a flag that can be set in the metadata header by the kernel to\n indicate that at least one message was lost since it was last cleared\n by userspace.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 89a565dba1a0ade82f90008589e5a7c22656b149\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:26 2019 +0000\n\n drm: kirin: Move ade drm init to kirin drm drv\n \n As part of refactoring the kirin driver to better support\n different hardware revisions, this patch renames ade_data to\n kirin_drm_private, and moves crtc_init and plane_init to\n kirin drm drv too. Now that they are generic the functions\n can be shared between the kirin620 and (to be added later)\n kirin960 specific support code.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n [jstultz: Reworded commit message]\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-26-john.stultz@linaro.org\n\ncommit 28cd05eee4097e7c94dac8e20d3c4af8b450447b\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:25 2019 +0000\n\n drm: kirin: Pass driver data to crtc init and plane init\n \n As part of refactoring the kirin driver to better support\n different hardware revisions, this patch changes the code\n via a passed in driver_data pointer, rather than hardcoding\n them via ade_driver_data variable.\n \n This will allow those funcitons to be later moved to the\n generic kirin_drm_drv.c using alternative driver_data structures\n that support other hardware.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n [jstultz: Reworded commit message]\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-25-john.stultz@linaro.org\n\ncommit 2e89b4fbca8f8e6bdb674ae9778e7a0fd22b28ff\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:24 2019 +0000\n\n drm: kirin: Add alloc_hw_ctx/clean_hw_ctx ops in driver data\n \n As part of refactoring the kirin driver to better support\n different hardware revisions, this patch changes the\n alloc/clean_hw_ctx functions to be called via driver_data\n specific funciton pointers.\n \n This will allow the ade_drm_init to later be made generic and\n moved to kirin_drm_drv.c\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n [jstultz: Reworded commit message]\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-24-john.stultz@linaro.org\n\ncommit 27dd04a59158f2c52974c3691163b8014c8195f1\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:23 2019 +0000\n\n drm: kirin: Make driver_data variable non-global\n \n As part of refactoring the kirin driver to better support\n different hardware revisions, this patch changes the driver_data\n value to not be a global variable. Instead the driver_data value\n is accessed via the of_device_get_match_data() when needed.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n [jstultz: Reworded commit message]\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-23-john.stultz@linaro.org\n\ncommit a8c8af59e428f88db721e1b357ec14365822da1a\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:22 2019 +0000\n\n drm: kirin: Fix dev-\u003edriver_data setting\n \n As part of refactoring the kirin driver to better support\n different hardware revisions, this patch changes the\n dev-\u003edriver_data to point to a drm_device, not ade_data.\n \n Thus we set the driver data to drm device after alloc.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n [jstultz: Reworded commit message]\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-22-john.stultz@linaro.org\n\ncommit a51735befeca43239020fc68a44cb8490035b836\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:21 2019 +0000\n\n drm: kirin: Rename plane_init and crtc_init\n \n As part of refactoring the kirin driver to better support\n different hardware revisions, this patch renames\n ade_crtc/plane_init kirin_plane/crtc_init, as they will later be\n moved to kirin drm drv and shared with the kirin960 hardware\n support.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n [jstultz: reworded commit message]\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-21-john.stultz@linaro.org\n\ncommit d606dc9a6323741b74c4e815a8ee2c158186e82f\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:20 2019 +0000\n\n drm: kirin: Add register connect helper functions in drm init\n \n As part of refactoring the kirin driver to better support\n different hardware revisions, this patch adds a flag to the\n device specific driver data so that we can conditionally\n register the connectors at init.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n [jstultz: reworded commit message]\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-20-john.stultz@linaro.org\n\ncommit 1053d01864937897c90103cb118acf729a7f2df4\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:19 2019 +0000\n\n drm: kirin: Move drm driver to driver data\n \n As part of refactoring the kirin driver to better support\n different hardware revisions, this patch moves the drm_driver\n structure to be under device specific driver data.\n \n This will allow us to more easily add support for kirin960\n hardware with later patches.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n [jstultz: Reworded commit message]\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-19-john.stultz@linaro.org\n\ncommit 49af4611399a8345971bfb25dfca2b1ec51ecc0b\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:18 2019 +0000\n\n drm: kirin: Move config max_width and max_height to driver data\n \n As part of refactoring the kirin driver to better support\n different hardware revisions, this patch moves the max_width\n and max_height values used in kirin_drm_mode_config_inita to\n hardware specific driver data.\n \n This will make it easier to add support for new devices\n via a new kirin_drm_data structure.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n [jstultz: reworded commit message]\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-18-john.stultz@linaro.org\n\ncommit 48fa7c17d11a66a0a6202c3ac90c77907515304b\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:17 2019 +0000\n\n drm: kirin: Move plane number and primay plane in driver data\n \n As part of refactoring the kirin driver to better support\n different hardware revisions, this patch moves the number of\n planes and the primary plane value to the kirin_drm_data\n structure\n \n This will make it easier to add support for new devices\n via a new kirin_drm_data structure.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-17-john.stultz@linaro.org\n\ncommit bdaf419efed49ca0e3ea4f1a0037e48c2f3b8ddf\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:16 2019 +0000\n\n drm: kirin: Move mode config function to driver_data\n \n As part of refactoring the kirin driver to better support\n different hardware revisions, this patch moves the mode config\n initialization values into the kirin_drm_data structure.\n \n This will make it easier to add support for new devices\n via a new kirin_drm_data structure.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-16-john.stultz@linaro.org\n\ncommit e200d8eb390951fa10b87299ed084428b401fe85\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:15 2019 +0000\n\n drm: kirin: Move channel formats to driver data\n \n As part of refactoring the kirin driver to better support\n different hardware revisions, this patch moves the channel\n format arrays into the kirin_drm_data structure.\n \n This will make it easier to add support for new devices\n via a new kirin_drm_data structure.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n [jstultz: reworded commit message]\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-15-john.stultz@linaro.org\n\ncommit 5fb2e411be93524070658716fcba6d50a6c71943\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:14 2019 +0000\n\n drm: kirin: Move ade crtc/plane help functions to driver_data\n \n As part of refactoring the kirin driver to better support\n different hardware revisions, this patch moves the crtc\n and plane funcs/helper_funcs to the struct kirin_drm_data.\n \n This will make it easier to add support for new devices\n via a new kirin_drm_data structure.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n [jstultz: reworded commit message]\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-14-john.stultz@linaro.org\n\ncommit 7903ba41e93197f8b13c04a8ad6be9701ffa4412\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:13 2019 +0000\n\n drm: kirin: Reanme dc_ops to kirin_drm_data\n \n As part of refactoring the kirin driver to better support\n different hardware revisions, this patch renames the\n struct kirin_dc_ops to struct kirin_drm_data and cleans\n up the related variable names.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n [jstultz: reworded commit message]\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-13-john.stultz@linaro.org\n\ncommit c11a03f6eb5ff525c3253f693e31813ba4eabcbe\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:12 2019 +0000\n\n drm: kirin: Move kirin_crtc, kirin_plane, kirin_format to kirin_drm_drv.h\n \n As part of refactoring the kirin driver to better support\n different hardware revisions, this patch moves some shared\n structures and helpers to the common kirin_drm_drv.h\n \n These structures will later used by both kirin620 and\n future kirin960 driver\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n [jstultz: reworded commit message]\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-12-john.stultz@linaro.org\n\ncommit 529277121bba76f01fa75093b742b8429a067471\nAuthor: John Stultz \u003cjohn.stultz@linaro.org\u003e\nDate: Tue Aug 20 23:06:11 2019 +0000\n\n drm: kirin: Move workqueue to ade_hw_ctx structure\n \n The workqueue used to reset the display when we hit an LDI\n underflow error is ADE specific, so since this patch series\n works to make the kirin_crtc structure more generic, move the\n workqueue to the ade_hw_ctx structure instead.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-11-john.stultz@linaro.org\n\ncommit 36f8d22dbcd9bdbe0a6751defaaccbe97843e140\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:10 2019 +0000\n\n drm: kirin: Move request irq handle in ade hw ctx alloc\n \n As part of refactoring the kirin driver to better support\n different hardware revisions, this patch modifies the\n initialization routines so the devm_request_irq() function\n is called as part of the allocation function.\n \n This will be needed in the future when we will have different\n allocation functions to allocate hardware specific hw_ctx\n structures, which will setup the vblank irq differently.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n [jstultz: reworded commit message]\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-10-john.stultz@linaro.org\n\ncommit 43774b0eed5612a8830b82cb7122beb6d05b5ac5\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:09 2019 +0000\n\n drm: kirin: Dynamically allocate the hw_ctx\n \n As part of refactoring the kirin driver to better support\n different hardware revisions, this patch modifies the\n initialization function to dynamically allocate the ade_hw_ctx\n structure previously kept as part of struct ade_data.\n \n This is done so that later we can have the hw_ctx point to\n hardware revision specific ctx structures.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n [jstultz: reworded commit message]\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-9-john.stultz@linaro.org\n\ncommit ada7f67da99def9fcbbd99dd03f323bc0182d21c\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:08 2019 +0000\n\n drm: kirin: Rename ade_crtc to kirin_crtc\n \n As part of refactoring the kirin driver to better support\n different hardware revisions, this patch renames the\n struct ade_crtc to kirin_crtc.\n \n The struct kirin_crtc will later used by both kirin620 and\n future kirin960 driver, and will be moved to a common\n kirin_drm_drv.h in a future patch\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n [jstultz: reworded commit message]\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-8-john.stultz@linaro.org\n\ncommit 0ae622c5327bf98d005cc6e3db534e14d742194c\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:07 2019 +0000\n\n drm: kirin: Rename ade_plane to kirin_plane\n \n As part of refactoring the kirin driver to better support\n different hardware revisions, this patch renames the\n struct ade_plane to kirin_plane.\n \n The struct kirin_plane will later used by both kirin620 and\n future kirin960 driver, and will be moved to a common\n kirin_drm_drv.h in a future patch\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n [jstultz: reworded commit message]\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-7-john.stultz@linaro.org\n\ncommit a202da164321c762c8592b129e5162b38446d7a9\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:06 2019 +0000\n\n drm: kirin: Remove out_format from ade_crtc\n \n As part of refactoring the kirin driver to better support\n different hardware revisions, this patch removes the out_format\n field in the struct ade_crtc, which was only ever set to\n LDI_OUT_RGB_888.\n \n Thus this patch removes the field and instead directly uses\n LDI_OUT_RGB_888.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n [jstultz: reworded commit message]\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-6-john.stultz@linaro.org\n\ncommit e0d8eba5dd3373595f74f8a92abf4347877acd9d\nAuthor: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:05 2019 +0000\n\n drm: kirin: Remove uncessary parameter indirection\n \n In a few functions, we pass in a struct ade_crtc, which we only\n use to get to the underlying struct ade_hw_ctx.\n \n Thus this patch refactors the functions to just take the\n struct ade_hw_ctx directly.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n [jstultz: reworded commit message]\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-5-john.stultz@linaro.org\n\ncommit 9ec16cee89dc479145439966c537edd8117a9bb0\nAuthor: John Stultz \u003cjohn.stultz@linaro.org\u003e\nDate: Tue Aug 20 23:06:04 2019 +0000\n\n drm: kirin: Remove unreachable return\n \n The \u0027return 0\u0027 in kirin_drm_platform_probe() is unreachable\n code, so remove it.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Suggested by: Xu YiPing \u003cxuyiping@hisilicon.com\u003e\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-4-john.stultz@linaro.org\n\ncommit 46aa549fd1a909a265db3f381fcf739bf43dd95b\nAuthor: John Stultz \u003cjohn.stultz@linaro.org\u003e\nDate: Tue Aug 20 23:06:03 2019 +0000\n\n drm: kirin: Remove HISI_KIRIN_DW_DSI config option\n \n The CONFIG_HISI_KIRIN_DW_DSI option is only used w/ kirin\n driver, so cut out the middleman and condense the config\n logic down.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-3-john.stultz@linaro.org\n\ncommit ff57c6513820efe945b61863cf4a51b79f18b592\nAuthor: Da Lv \u003clvda3@hisilicon.com\u003e\nDate: Tue Aug 20 23:06:02 2019 +0000\n\n drm: kirin: Fix for hikey620 display offset problem\n \n The original HiKey (620) board has had a long running issue\n where when using a 1080p montior, the display would occasionally\n blink and come come back with a horizontal offset (usually also\n shifting the colors, depending on the value of the offset%4).\n \n After lots of analysis by HiSi developers, they found the issue\n was due to when running at 1080p, it was possible to hit the\n device memory bandwidth limits, which could cause the DSI signal\n to get out of sync.\n \n Unfortunately the DSI logic doesn\u0027t have the ability to\n automatically recover from this situation, but we can get a an\n LDI underflow interrupt when it happens.\n \n To then correct the issue, when we get an LDI underflow irq, we\n we can simply suspend and resume the display, which resets the\n hardware.\n \n Thus, this patch enables the ldi underflow interrupt, and\n initializes a workqueue that is used to suspend/resume the\n display to recover. Then when the irq occurs we clear it and\n schedule the workqueue to reset display engine.\n \n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel \u003cdri-devel@lists.freedesktop.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Da Lv \u003clvda3@hisilicon.com\u003e\n Signed-off-by: Yidong Lin \u003clinyidong@huawei.com\u003e\n [jstultz: Reworded the commit message, checkpatch cleanups]\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820230626.23253-2-john.stultz@linaro.org\n\ncommit fba0e448a2c5b297a4ddc1ec4e48f4aa6600a1c9\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Mon Aug 19 14:17:10 2019 +0300\n\n RDMA/mlx5: Use odp instead of mr-\u003eumem in pagefault_mr\n \n These are the same thing since mr always comes from odp-\u003eprivate. It is\n confusing to reference the same memory via two names.\n \n Link: https://lore.kernel.org/r/20190819111710.18440-13-leon@kernel.org\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit a705f3e3a1a8cbe0064730024398b2320ae1ce74\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Mon Aug 19 14:17:09 2019 +0300\n\n RDMA/mlx5: Use ib_umem_start instead of umem.address\n \n These are subtly different, the address is the original VA requested\n during umem_get, while ib_umem_start() is the version that is rounded to\n the proper page size, ie is the true start of the umem\u0027s dma map.\n \n Link: https://lore.kernel.org/r/20190819111710.18440-12-leon@kernel.org\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit ce51346feede2ea41de0ad58af2b514223e11dad\nAuthor: Moni Shoua \u003cmonis@mellanox.com\u003e\nDate: Mon Aug 19 14:17:08 2019 +0300\n\n RDMA/core: Make invalidate_range a device operation\n \n The callback function \u0027invalidate_range\u0027 is implemented in a driver so the\n place for it is in the ib_device_ops structure and not in ib_ucontext.\n \n Signed-off-by: Moni Shoua \u003cmonis@mellanox.com\u003e\n Reviewed-by: Guy Levi \u003cguyle@mellanox.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Link: https://lore.kernel.org/r/20190819111710.18440-11-leon@kernel.org\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 37824952dc8fcd96e5c5a1ce9abf3f0ba09b1e5e\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Mon Aug 19 14:17:07 2019 +0300\n\n RDMA/odp: Use kvcalloc for the dma_list and page_list\n \n There is no specific need for these to be in the valloc space, let the\n system decide automatically how to do the allocation.\n \n Link: https://lore.kernel.org/r/20190819111710.18440-10-leon@kernel.org\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 204e3e5630c5d41948fc11d8419c07da8f3e5a4d\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Mon Aug 19 14:17:06 2019 +0300\n\n RDMA/odp: Check for overflow when computing the umem_odp end\n \n Since the page size can be extended in the ODP case by IB_ACCESS_HUGETLB\n the existing overflow checks done by ib_umem_get() are not\n sufficient. Check for overflow again.\n \n Further, remove the unchecked math from the inlines and just use the\n precomputed value stored in the interval_tree_node.\n \n Link: https://lore.kernel.org/r/20190819111710.18440-9-leon@kernel.org\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 0446cad9ca385c3b8d6e5a1184e59650fa7a7a6d\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Mon Aug 19 14:17:05 2019 +0300\n\n RDMA/odp: Provide ib_umem_odp_release() to undo the allocs\n \n Now that there are allocator APIs that return the ib_umem_odp directly\n it should be freed through a umem_odp free\u0027er as well.\n \n Link: https://lore.kernel.org/r/20190819111710.18440-8-leon@kernel.org\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 261dc53f8ee037bf2fbf68f90c319b04062a126c\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Mon Aug 19 14:17:04 2019 +0300\n\n RDMA/odp: Split creating a umem_odp from ib_umem_get\n \n This is the last creation API that is overloaded for both, there is very\n little code sharing and a driver has to be specifically ready for a\n umem_odp to be created to use the odp version.\n \n Link: https://lore.kernel.org/r/20190819111710.18440-7-leon@kernel.org\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit f20bef6a951b6ef619655ed846113f706d0824d7\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Mon Aug 19 14:17:03 2019 +0300\n\n RDMA/odp: Make the three ways to create a umem_odp clear\n \n The three paths to build the umem_odps are kind of muddled, they are:\n - As a normal ib_mr umem\n - As a child in an implicit ODP umem tree\n - As the root of an implicit ODP umem tree\n \n Only the first two are actually umem\u0027s, the last is an abuse.\n \n The implicit case can only be triggered by explicit driver request, it\n should never be co-mingled with the normal case. While we are here, make\n sensible function names and add some comments to make this clearer.\n \n Link: https://lore.kernel.org/r/20190819111710.18440-6-leon@kernel.org\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 22d79c9a912c9d56258a0adf8a9c8cbfb6f2fd1a\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Mon Aug 19 14:17:02 2019 +0300\n\n RMDA/odp: Consolidate umem_odp initialization\n \n This is done in two different places, consolidate all the post-allocation\n initialization into a single function.\n \n Link: https://lore.kernel.org/r/20190819111710.18440-5-leon@kernel.org\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit fd7dbf035edcfb035977423e2a5102832c1427f4\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Mon Aug 19 14:17:01 2019 +0300\n\n RDMA/odp: Make it clearer when a umem is an implicit ODP umem\n \n Implicit ODP umems are special, they don\u0027t have any page lists, they don\u0027t\n exist in the interval tree and they are never DMA mapped.\n \n Instead of trying to guess this based on a zero length use an explicit\n flag.\n \n Further, do not allow non-implicit umems to be 0 size.\n \n Link: https://lore.kernel.org/r/20190819111710.18440-4-leon@kernel.org\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit bd5defaee872da9b81e3c72045eb6794445cd2e6\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Fri Aug 16 08:24:34 2019 +0200\n\n dma-mapping: remove is_device_dma_capable\n \n No users left.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lore.kernel.org/r/20190816062435.881-6-hch@lst.de\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f993de88a55f2dd56f17248d06c0dfda9a9799db\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Mon Aug 19 14:17:00 2019 +0300\n\n RDMA/odp: Iterate over the whole rbtree directly\n \n Instead of intersecting a full interval, just iterate over every element\n directly. This is faster and clearer.\n \n Link: https://lore.kernel.org/r/20190819111710.18440-3-leon@kernel.org\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 5d6ff300f0114e64f4ad911d256ac1587204ac5d\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Fri Aug 16 08:24:33 2019 +0200\n\n usb/max3421: remove the dummy {un,}map_urb_for_dma methods\n \n Now that we have an explicit HCD_DMA flag, there is not need to override\n these methods.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lore.kernel.org/r/20190816062435.881-5-hch@lst.de\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7b81cb6bddd2c4f2489506771070924bd0ae9902\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Fri Aug 16 08:24:32 2019 +0200\n\n usb: add a HCD_DMA flag instead of guestimating DMA capabilities\n \n The usb core is the only major place in the kernel that checks for\n a non-NULL device dma_mask to see if a device is DMA capable. This\n is generally a bad idea, as all major busses always set up a DMA mask,\n even if the device is not DMA capable - in fact bus layers like PCI\n can\u0027t even know if a device is DMA capable at enumeration time. This\n leads to lots of workaround in HCD drivers, and also prevented us from\n setting up a DMA mask for platform devices by default last time we\n tried.\n \n Replace this guess with an explicit HCD_DMA that is set by drivers that\n appear to have DMA support.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lore.kernel.org/r/20190816062435.881-4-hch@lst.de\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 4776f3529d6b1e47f02904ad1d264d25ea22b27b\nAuthor: Xiong Zhang \u003cxiong.y.zhang@intel.com\u003e\nDate: Tue Aug 20 13:46:17 2019 +0800\n\n drm/i915: Don\u0027t deballoon unused ggtt drm_mm_node in linux guest\n \n The following call trace may exist in linux guest dmesg when guest i915\n driver is unloaded.\n [ 90.776610] [drm:vgt_deballoon_space.isra.0 [i915]] deballoon space: range [0x0 - 0x0] 0 KiB.\n [ 90.776621] BUG: unable to handle kernel NULL pointer dereference at 00000000000000c0\n [ 90.776691] IP: drm_mm_remove_node+0x4d/0x320 [drm]\n [ 90.776718] PGD 800000012c7d0067 P4D 800000012c7d0067 PUD 138e4c067 PMD 0\n [ 90.777091] task: ffff9adab60f2f00 task.stack: ffffaf39c0fe0000\n [ 90.777142] RIP: 0010:drm_mm_remove_node+0x4d/0x320 [drm]\n [ 90.777573] Call Trace:\n [ 90.777653] intel_vgt_deballoon+0x4c/0x60 [i915]\n [ 90.777729] i915_ggtt_cleanup_hw+0x121/0x190 [i915]\n [ 90.777792] i915_driver_unload+0x145/0x180 [i915]\n [ 90.777856] i915_pci_remove+0x15/0x20 [i915]\n [ 90.777890] pci_device_remove+0x3b/0xc0\n [ 90.777916] device_release_driver_internal+0x157/0x220\n [ 90.777945] driver_detach+0x39/0x70\n [ 90.777967] bus_remove_driver+0x51/0xd0\n [ 90.777990] pci_unregister_driver+0x23/0x90\n [ 90.778019] SyS_delete_module+0x1da/0x240\n [ 90.778045] entry_SYSCALL_64_fastpath+0x24/0x87\n [ 90.778072] RIP: 0033:0x7f34312af067\n [ 90.778092] RSP: 002b:00007ffdea3da0d8 EFLAGS: 00000206\n [ 90.778297] RIP: drm_mm_remove_node+0x4d/0x320 [drm] RSP: ffffaf39c0fe3dc0\n [ 90.778344] ---[ end trace f4b1bc8305fc59dd ]---\n \n Four drm_mm_node are used to reserve guest ggtt space, but some of them\n may be skipped and not initialised due to space constraints in\n intel_vgt_balloon(). If drm_mm_remove_node() is called with\n uninitialized drm_mm_node, the above call trace occurs.\n \n This patch check drm_mm_node\u0027s validity before calling\n drm_mm_remove_node().\n \n Fixes: ff8f797557c7(\"drm/i915: return the correct usable aperture size under gvt environment\")\n Cc: stable@vger.kernel.org\n Signed-off-by: Xiong Zhang \u003cxiong.y.zhang@intel.com\u003e\n Acked-by: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/1566279978-9659-1-git-send-email-xiong.y.zhang@intel.com\n\ncommit 0709831a50d31b3caf2237e8d7fe89e15b0d919d\nAuthor: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nDate: Sat Aug 17 09:55:20 2019 +0300\n\n usb: host: ohci-pxa27x: Fix and \u0026 vs | typo\n \n The code is supposed to clear the RH_A_NPS and RH_A_PSM bits, but it\u0027s\n a no-op because of the \u0026 vs | typo. This bug predates git and it was\n only discovered using static analysis so it must not affect too many\n people in real life.\n \n Signed-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Acked-by: Alan Stern \u003cstern@rowland.harvard.edu\u003e\n Link: https://lore.kernel.org/r/20190817065520.GA29951@mwanda\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 353e3cf8590cf182a9f42e67993de3aca91e8090\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Aug 20 15:12:12 2019 +0100\n\n iommu/arm-smmu-v3: Fix ATC invalidation ordering wrt main TLBs\n \n When invalidating the ATC for an PCIe endpoint using ATS, we must take\n care to complete invalidation of the main SMMU TLBs beforehand, otherwise\n the device could immediately repopulate its ATC with stale translations.\n \n Hooking the ATC invalidation into -\u003eunmap() as we currently do does the\n exact opposite: it ensures that the ATC is invalidated *before* the\n main TLBs, which is bogus.\n \n Move ATC invalidation into the actual (leaf) invalidation routines so\n that it is always called after completing main TLB invalidation.\n \n Reviewed-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit c71ccbe26376e81de6582c68c6783de488fe9e24\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Aug 21 16:57:25 2019 +0100\n\n drm/i915/gtt: Add some range asserts\n \n These should have been validated in the upper layers, but for sanity\u0027s\n sake, repeat them.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190821155728.2839-2-chris@chris-wilson.co.uk\n\ncommit bfff88ec1afefb76669d29da9fc90d5ad7c53ec3\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Aug 20 14:28:59 2019 +0100\n\n iommu/arm-smmu-v3: Rework enabling/disabling of ATS for PCI masters\n \n To prevent any potential issues arising from speculative Address\n Translation Requests from an ATS-enabled PCIe endpoint, rework our ATS\n enabling/disabling logic so that we enable ATS at the SMMU before we\n enable it at the endpoint, and disable things in the opposite order.\n \n Reviewed-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 7314ca8699e1defd6f2883f203c0e828a4f51f9f\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Wed Aug 21 12:38:15 2019 +0100\n\n iommu/arm-smmu-v3: Don\u0027t issue CMD_SYNC for zero-length invalidations\n \n Calling arm_smmu_tlb_inv_range() with a size of zero, perhaps due to\n an empty \u0027iommu_iotlb_gather\u0027 structure, should be a NOP. Elide the\n CMD_SYNC when there is no invalidation to be performed.\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit f75d8e33df91ee66fc12820d0da1454b76ebcff9\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Aug 20 17:32:18 2019 +0100\n\n iommu/arm-smmu-v3: Remove boolean bitfield for \u0027ats_enabled\u0027 flag\n \n There\u0027s really no need for this to be a bitfield, particularly as we\n don\u0027t have bitwise addressing on arm64.\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit b5e86196b83fd68e065a7c811ab8925fb0dc3893\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Wed Aug 21 14:17:00 2019 +0100\n\n iommu/arm-smmu-v3: Disable detection of ATS and PRI\n \n Detecting the ATS capability of the SMMU at probe time introduces a\n spinlock into the -\u003eunmap() fast path, even when ATS is not actually\n in use. Furthermore, the ATC invalidation that exists is broken, as it\n occurs before invalidation of the main SMMU TLB which leaves a window\n where the ATC can be repopulated with stale entries.\n \n Given that ATS is both a new feature and a specialist sport, disable it\n for now whilst we fix it properly in subsequent patches. Since PRI\n requires ATS, disable that too.\n \n Cc: \u003cstable@vger.kernel.org\u003e\n Fixes: 9ce27afc0830 (\"iommu/arm-smmu-v3: Add support for PCI ATS\")\n Acked-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit b08a6259a141baba98c57524ee4283fd91445bb8\nAuthor: Gavin Li \u003cgit@thegavinli.com\u003e\nDate: Wed Aug 14 14:29:24 2019 -0700\n\n usb: usbfs: only account once for mmap()\u0027ed usb memory usage\n \n Memory usage for USB memory allocated via mmap() is already accounted\n for at mmap() time; no need to account for it again at submiturb time.\n \n Signed-off-by: Gavin Li \u003cgit@thegavinli.com\u003e\n Acked-by: Alan Stern \u003cstern@rowland.harvard.edu\u003e\n Link: https://lore.kernel.org/r/20190814212924.10381-1-gavinli@thegavinli.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a599e48662b4b505bef45d4831061f9d50703e17\nAuthor: Jacob Huisman \u003cjacobhuisman@kernelthusiast.com\u003e\nDate: Thu Aug 15 18:42:10 2019 +0100\n\n usb: usb-skeleton: make comment block in line with coding style\n \n Comment block was not in accordance with coding style.\n Fixes two checkpatch warnings:\n WARNING: Block comments use * on subsequent lines\n WARNING: Block comments use a trailing */ on a separate line\n \n Signed-off-by: Jacob Huisman \u003cjacobhuisman@kernelthusiast.com\u003e\n Link: https://lore.kernel.org/r/20190815174210.580-1-jacobhuisman@kernelthusiast.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit bb3e9c767c6134a5761470038e8c75cdb6f04867\nAuthor: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\nDate: Mon Aug 12 23:27:57 2019 +0200\n\n ARM: dts: at91: at91sam9x5dm.dtsi: Style cleanup\n \n - use tags from included dtsi instead of duplicating the hierarchy\n \n There are no differences in the generated .dtbs\n \n Signed-off-by: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\n Link: https://lore.kernel.org/r/20190812212757.23432-9-uwe@kleine-koenig.org\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 532173b404a8d6b39d17c4db457915bd793ec918\nAuthor: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\nDate: Mon Aug 12 23:27:56 2019 +0200\n\n ARM: dts: at91: at91sam9x5_lcd.dtsi: Style cleanup\n \n - use tags from included dtsi instead of duplicating the hierarchy\n \n There are no differences in the generated .dtbs\n \n Signed-off-by: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\n Link: https://lore.kernel.org/r/20190812212757.23432-8-uwe@kleine-koenig.org\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit b222de0242dfb7efaa6e35b68ce844878667ca82\nAuthor: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\nDate: Mon Aug 12 23:27:55 2019 +0200\n\n ARM: dts: at91: at91sam9xx5ek: Style cleanup\n \n - newline between properties and sub-nodes\n - use tags from included dtsi instead of duplicating the hierarchy\n \n There are no differences in the generated .dtbs\n \n Signed-off-by: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\n Link: https://lore.kernel.org/r/20190812212757.23432-7-uwe@kleine-koenig.org\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit c1ad3ffb182e31eb445b7afbb0c02c70a3eef48f\nAuthor: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\nDate: Mon Aug 12 23:27:54 2019 +0200\n\n ARM: dts: at91: at91sam9g15: Style cleanup\n \n - use tags from included dtsi instead of duplicating the hierarchy\n \n There are no differences in the generated .dtbs\n \n Signed-off-by: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\n Link: https://lore.kernel.org/r/20190812212757.23432-6-uwe@kleine-koenig.org\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit a77eb442f98747a5d567ea0b1a7b95949415fce6\nAuthor: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\nDate: Mon Aug 12 23:27:53 2019 +0200\n\n ARM: dts: at91: kizboxmini: Style cleanup\n \n - use tags from included dtsi instead of duplicating the hierarchy\n \n There are no differences in the generated .dtb\n \n Signed-off-by: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\n Link: https://lore.kernel.org/r/20190812212757.23432-5-uwe@kleine-koenig.org\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit fd117bba8c1e6a2726e52662aceb1076d4b4280f\nAuthor: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\nDate: Mon Aug 12 23:27:52 2019 +0200\n\n ARM: dts: at91: cosino: Style cleanup\n \n - newline between properties and sub-nodes\n - use tags from included dtsi instead of duplicating the hierarchy\n - status should be the last property\n \n There are no differences in the generated .dtb\n \n Signed-off-by: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\n Link: https://lore.kernel.org/r/20190812212757.23432-4-uwe@kleine-koenig.org\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit c756f614577e7c725efdae106fb8e329b3b12005\nAuthor: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\nDate: Mon Aug 12 23:27:51 2019 +0200\n\n ARM: dts: at91: ariettag25: style cleanup\n \n - newline between properties and sub-nodes\n - use tags from included dtsi instead of duplicating the hierarchy\n - status should be the last property\n - drop duplicated alias\n \n There are no differences in the generated .dtb\n \n Signed-off-by: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\n Link: https://lore.kernel.org/r/20190812212757.23432-3-uwe@kleine-koenig.org\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 60839c530853312a0a32e7588966524ac6030a18\nAuthor: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\nDate: Mon Aug 12 23:27:50 2019 +0200\n\n ARM: dts: at91: ariag25: Style cleanup\n \n - newline between properties and sub-nodes\n - use tags from included dtsi instead of duplicating the hierarchy\n \n There are no differences in the generated .dtb\n \n Signed-off-by: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\n Link: https://lore.kernel.org/r/20190812212757.23432-2-uwe@kleine-koenig.org\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 1fef2bf578f084daaebc84435803da3b44f7ac77\nAuthor: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\nDate: Mon Aug 12 23:27:49 2019 +0200\n\n ARM: dts: at91: Add label for sam9x5\u0027s internal RTC\n \n This allows to simplify several machine device trees using this label\n instead of duplicating the SoC\u0027s hierarchy.\n \n Signed-off-by: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\n Link: https://lore.kernel.org/r/20190812212757.23432-1-uwe@kleine-koenig.org\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 1b9bd09630d4db4827cc04d358a41a16a6bc2cb0\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Tue Aug 20 19:16:57 2019 +0300\n\n drm/i915: Do not create a new max_bpc prop for MST connectors\n \n We\u0027re not allowed to create new properties after device registration\n so for MST connectors we need to either create the max_bpc property\n earlier, or we reuse one we already have. Let\u0027s do the latter apporach\n since the corresponding SST connector already has the prop and its\n min/max are correct also for the MST connector.\n \n The problem was highlighted by commit 4f5368b5541a (\"drm/kms:\n Catch mode_object lifetime errors\") which results in the following\n spew:\n [ 1330.878941] WARNING: CPU: 2 PID: 1554 at drivers/gpu/drm/drm_mode_object.c:45 __drm_mode_object_add+0xa0/0xb0 [drm]\n ...\n [ 1330.879008] Call Trace:\n [ 1330.879023] drm_property_create+0xba/0x180 [drm]\n [ 1330.879036] drm_property_create_range+0x15/0x30 [drm]\n [ 1330.879048] drm_connector_attach_max_bpc_property+0x62/0x80 [drm]\n [ 1330.879086] intel_dp_add_mst_connector+0x11f/0x140 [i915]\n [ 1330.879094] drm_dp_add_port.isra.20+0x20b/0x440 [drm_kms_helper]\n ...\n \n Cc: stable@vger.kernel.org\n Cc: Lyude Paul \u003clyude@redhat.com\u003e\n Cc: sunpeng.li@amd.com\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Fixes: 5ca0ef8a56b8 (\"drm/i915: Add max_bpc property for DP MST\")\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820161657.9658-1-ville.syrjala@linux.intel.com\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Reviewed-by: Lyude Paul \u003clyude@redhat.com\u003e\n\ncommit df901c85cc28b538c62f6bc20b16a8bd05fcb756\nAuthor: Hou Zhiqiang \u003cZhiqiang.Hou@nxp.com\u003e\nDate: Sat Jul 13 22:11:29 2019 +0800\n\n PCI: mobiveil: Fix the CPU base address setup in inbound window\n \n Current code erroneously sets-up the CPU base address through the\n parameter \u0027pci_addr\u0027, which is passed to initialize the CPU (AXI) base\n address of the inbound window where the controller maps the PCI address\n space into CPU physical address space; furthermore, it also truncates it\n by programming only the lower 32-bit value into the inbound CPU address\n register.\n \n Fix both issues by introducing a new parameter \u0027u64 cpu_addr\u0027 to\n initialize both lower 32-bit and upper 32-bit of the CPU physical\n base address mapping PCI inbound transactions into CPU (AXI) ones.\n \n Fixes: 9af6bcb11e12 (\"PCI: mobiveil: Add Mobiveil PCIe Host Bridge IP driver\")\n Signed-off-by: Hou Zhiqiang \u003cZhiqiang.Hou@nxp.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Reviewed-by: Minghuan Lian \u003cMinghuan.Lian@nxp.com\u003e\n Reviewed-by: Subrahmanya Lingappa \u003cl.subrahmanya@mobiveil.co.in\u003e\n\ncommit 7cc2e18f21008f4093b49099264ca4d65b9aa223\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Mon Aug 19 14:16:59 2019 +0300\n\n RDMA/odp: Use the common interval tree library instead of generic\n \n ODP is working with userspace VA\u0027s in the interval tree which always fit\n into an unsigned long, so we can use the common code.\n \n This comes at a cost of a 16 byte increase in ib_umem_odp struct size due\n to storing the interval tree start/last in addition to the umem\n addr/length. However these values were computed and are performance\n critical for the interval lookup, so this seems like a worthwhile trade\n off.\n \n Removes 2k of .text from the kernel.\n \n Link: https://lore.kernel.org/r/20190819111710.18440-2-leon@kernel.org\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit a20ab592d1a87218229109d109b8e2feae6f598d\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Aug 21 15:23:36 2019 +0100\n\n drm/i915/execlists: Set priority hint prior to submission\n \n Since we now run process_csb() outside of the engine-\u003eactive.lock, we\n can process a CS-event immediately upon our ELSP write. As we currently\n inspect the pending queue *after* the ELSP write, there is an\n opportunity for a CS-event to update the pending queue before we can\n read it, making ourselves chases an invalid pointer.\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111427\n Fixes: df403069029d (\"drm/i915/execlists: Lift process_csb() out of the irq-off spinlock\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190821142336.21609-1-chris@chris-wilson.co.uk\n\ncommit f401441deda68326852560bf70d59e95f585bbb3\nAuthor: Stephen Brennan \u003cstephen@brennan.io\u003e\nDate: Wed Aug 21 07:35:40 2019 -0700\n\n staging: rtl8192u: fix spacing in ieee80211\n \n Checkpatch emits several errors, warnings, and checks about spacing.\n Apply checkpatch\u0027s suggested spacing rules.\n \n Signed-off-by: Stephen Brennan \u003cstephen@brennan.io\u003e\n Link: https://lore.kernel.org/r/20190821143540.4501-4-stephen@brennan.io\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 2b1d4c425ae548f3f8da11238abf59a08f1626b7\nAuthor: Stephen Brennan \u003cstephen@brennan.io\u003e\nDate: Wed Aug 21 07:35:39 2019 -0700\n\n staging: rtl8192u: fix macro alignment in ieee80211\n \n Several macros display unaligned, due to mixes of tabs and spaces. These\n can be fixed by making spacing consistent, do this.\n \n Signed-off-by: Stephen Brennan \u003cstephen@brennan.io\u003e\n Link: https://lore.kernel.org/r/20190821143540.4501-3-stephen@brennan.io\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit ee16d7d3e0128b347210b27214043e245d03c256\nAuthor: Stephen Brennan \u003cstephen@brennan.io\u003e\nDate: Wed Aug 21 07:35:38 2019 -0700\n\n staging: rtl8192u: fix OPEN_BRACE errors in ieee80211\n \n Checkpatch emits several errors regarding braces being on the incorrect\n line. These can be fixed by moving the brace, do this. In a few cases,\n some comments were moved to facilitate this.\n \n Signed-off-by: Stephen Brennan \u003cstephen@brennan.io\u003e\n Link: https://lore.kernel.org/r/20190821143540.4501-2-stephen@brennan.io\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 6808f8cf57f10c1491074572eb247e7fc2a60f47\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 21 20:28:02 2019 +0800\n\n staging: rtl8192e: remove set but not used variable \u0027data_len\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n In function ieee80211_ccmp_encrypt:\n drivers/staging/rtl8192u/ieee80211/ieee80211_crypt_ccmp.c:162:6:\n warning: variable data_len set but not used [-Wunused-but-set-variable]\n \n It is not used since commit 5ee5265674ce (\"staging:\n rtl8192e: rtllib_crypt_ccmp.c: Use crypto API ccm(aes)\")\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190821122802.44028-1-yuehaibing@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 4a67342ea9c7c34903f4068f21c98385f9f62b8d\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 21 20:25:56 2019 +0800\n\n staging: rtl8192e: remove two set but not used variables\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n In function \u0027_rtl92e_dm_tx_power_tracking_callback_tssi\u0027:\n drivers/staging/rtl8192e/rtl8192e/rtl_dm.c:621:7:\n warning: variable \u0027bHighpowerstate\u0027 set but not used [-Wunused-but-set-variable]\n In function \u0027_rtl92e_dm_rx_path_sel_byrssi\u0027:\n drivers/staging/rtl8192e/rtl8192e/rtl_dm.c:1904:32:\n warning: variable \u0027cck_rx_ver2_min_index\u0027 set but not used [-Wunused-but-set-variable]\n \n They are never used, so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190821122556.37636-1-yuehaibing@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 68772c84f7e0380033ce560802b0d179d76be392\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 21 20:22:50 2019 +0800\n\n staging: rtl8192u: ieee80211: remove set but not used variable \u0027data_len\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/staging/rtl8192u/ieee80211/ieee80211_crypt_ccmp.c:\n In function ieee80211_ccmp_encrypt:\n drivers/staging/rtl8192u/ieee80211/ieee80211_crypt_ccmp.c:162:6:\n warning: variable data_len set but not used [-Wunused-but-set-variable]\n \n It is not used since commit eb0e7bf3ca94 (\"staging:\n rtl8192u: ieee80211: ieee80211_crypt_ccmp.c: Use crypto API ccm(aes)\")\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190821122250.71404-1-yuehaibing@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit cdfca821571dfad27434b71cb43e0654667c0fd1\nAuthor: Mark Brown \u003cbroonie@kernel.org\u003e\nDate: Mon Aug 19 21:06:50 2019 +0100\n\n merge_config.sh: Check error codes from make\n \n When we execute make after merging the configurations we ignore any\n errors it produces causing whatever is running merge_config.sh to be\n unaware of any failures. This issue was noticed by Guillaume Tucker\n while looking at problems with testing of clang only builds in KernelCI\n which caused Kbuild to be unable to find a working host compiler.\n \n This implementation was suggested by Yamada-san.\n \n Suggested-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Reported-by: Guillaume Tucker \u003cguillaume.tucker@collabora.com\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit eb27ea5ce7f367b185953cc9bc6e606004cfd8c4\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Mon Aug 19 17:58:43 2019 +0900\n\n kbuild: move modkern_{c,a}flags to Makefile.lib from Makefile.build\n \n Makefile.lib is included by Makefile.modfinal as well as Makefile.build.\n \n Move modkern_cflags to Makefile.lib in order to simplify cmd_cc_o_c\n in Makefile.modfinal. Move modkern_cflags as well for consistency.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 2ff2b7ec65dcea2261bdf404654975f44ad6323d\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Mon Aug 19 14:54:20 2019 +0900\n\n kbuild: add CONFIG_ASM_MODVERSIONS\n \n Add CONFIG_ASM_MODVERSIONS. This allows to remove one if-conditional\n nesting in scripts/Makefile.build.\n \n scripts/Makefile.build is run every time Kbuild descends into a\n sub-directory. So, I want to avoid $(wildcard ...) evaluation\n where possible although computing $(wildcard ...) is so cheap that\n it may not make measurable performance difference.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Acked-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n\ncommit a564bdeb5e0d79dbcebb9a138e7b9f2d7ed97df9\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Mon Aug 19 13:18:07 2019 +0900\n\n .gitignore: ignore modules.order explicitly\n \n The pattern \u0027*.order\u0027 was added by commit c6025f4c8bbe (\"kbuild: ignore\n *.order files\") to ignore modules.order files.\n \n I do not see any other user of the \u0027.order\u0027 extension.\n \n Ignore \u0027modules.order\u0027 explicitly instead of \u0027*.order\u0027.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit e12a0ce2fa69798194f3a8628baf6edfbd5c548f\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Aug 21 22:01:52 2019 +0800\n\n staging: erofs: detect potential multiref due to corrupted images\n \n As reported by erofs-utils fuzzer, currently, multiref\n (ondisk deduplication) hasn\u0027t been supported for now,\n we should forbid it properly.\n \n Fixes: 3883a79abd02 (\"staging: erofs: introduce VLE decompression support\")\n Cc: \u003cstable@vger.kernel.org\u003e # 4.19+\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190821140152.229648-1-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 9b9a3f20cbe0ba9269cde6fff9f9c69907e150cf\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Thu Aug 15 01:06:23 2019 +0900\n\n kbuild: split final module linking out into Makefile.modfinal\n \n I think splitting the modpost and linking modules into separate\n Makefiles will be useful especially when more complex build steps\n come in. The main motivation of this commit is to integrate the\n proposed klp-convert feature cleanly.\n \n I moved the logging \u0027Building modules, stage 2.\u0027 to Makefile.modpost\n to avoid the code duplication although I do not know whether or not\n this message is needed in the first place.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 2a7f77c7be1b5adb69712e440e97e0c6fa7ebecb\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:49:30 2019 -0400\n\n xprtrdma: Clean up xprt_rdma_set_connect_timeout()\n \n Clean up: The function name should match the documenting comment.\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit b0b227f071a06d0ce5ef803538899299933d2931\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:48:43 2019 -0400\n\n xprtrdma: Use an llist to manage free rpcrdma_reps\n \n rpcrdma_rep objects are removed from their free list by only a\n single thread: the Receive completion handler. Thus that free list\n can be converted to an llist, where a single-threaded consumer and\n a multi-threaded producer (rpcrdma_buffer_put) can both access the\n llist without the need for any serialization.\n \n This eliminates spin lock contention between the Receive completion\n handler and rpcrdma_buffer_get, and makes the rep consumer wait-\n free.\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit 4d6b8890ddb16120c5dd25edc8d215cfd8222b63\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:47:57 2019 -0400\n\n xprtrdma: Remove rpcrdma_buffer::rb_mrlock\n \n Clean up: Now that the free list is used sparingly, get rid of the\n separate spin lock protecting it.\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit b38795385c9be10f905505b9fa9c4af8dd4f737d\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:29 2019 -0700\n\n mmc: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n Cc: linux-mmc@vger.kernel.org\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit 371d0723a46737c1a472daa238977356177998aa\nAuthor: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\nDate: Tue Jul 30 09:07:23 2019 +0300\n\n mmc: sdhci-pci: Add another Id for Intel CML\n \n Add another PCI Id for Intel CML.\n \n Signed-off-by: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit 7b7bea9fb05be520d2775e8223d988227935882e\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Sun Jul 28 19:02:59 2019 -0500\n\n mmc: sdhci-s3c: Mark expected switch fall-through\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warnings:\n \n drivers/mmc/host/sdhci-s3c.c: In function \u0027sdhci_s3c_probe\u0027:\n drivers/mmc/host/sdhci-s3c.c:613:19: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n host-\u003emmc-\u003ecaps |\u003d MMC_CAP_8_BIT_DATA;\n drivers/mmc/host/sdhci-s3c.c:614:2: note: here\n case 4:\n ^~~~\n \n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Acked-by: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit 627405f30452d2d55c91c9db2d55b6490ca64c1b\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Sun Jul 28 19:01:23 2019 -0500\n\n mmc: atmel-mci: Mark expected switch fall-throughs\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warnings:\n \n drivers/mmc/host/atmel-mci.c: In function \u0027atmci_get_cap\u0027:\n drivers/mmc/host/atmel-mci.c:2415:30: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n host-\u003ecaps.has_odd_clk_div \u003d 1;\n ~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~\n drivers/mmc/host/atmel-mci.c:2416:2: note: here\n case 0x400:\n ^~~~\n drivers/mmc/host/atmel-mci.c:2422:28: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n host-\u003ecaps.has_highspeed \u003d 1;\n ~~~~~~~~~~~~~~~~~~~~~~~~~^~~\n drivers/mmc/host/atmel-mci.c:2423:2: note: here\n case 0x200:\n ^~~~\n drivers/mmc/host/atmel-mci.c:2426:40: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n host-\u003ecaps.need_notbusy_for_read_ops \u003d 1;\n ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~\n drivers/mmc/host/atmel-mci.c:2427:2: note: here\n case 0x100:\n ^~~~\n \n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Acked-by: Ludovic Desroches \u003cludovic.desroches@microchip.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit e9cb9207c4748ee246f202812b31832a9db0eb99\nAuthor: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\nDate: Thu Jul 25 07:28:22 2019 +0000\n\n mmc: sdhci-pltfm: Use devm_platform_ioremap_resource() to simplify code\n \n devm_platform_ioremap_resource() wraps platform_get_resource() and\n devm_ioremap_resource() in a single helper, let\u0027s use that helper to\n simplify the code.\n \n Signed-off-by: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit c6bfc259aee8af92c7783eec07894ccc44784e52\nAuthor: Ulf Hansson \u003culf.hansson@linaro.org\u003e\nDate: Tue Jul 23 14:28:09 2019 +0200\n\n mmc: mmci: Clarify comments and some code for busy detection\n \n The code dealing with busy detection is somewhat complicated. In a way to\n make it a bit clearer, let\u0027s try to clarify the comments in the code about\n it.\n \n Additionally, move the part for clearing the so called busy start IRQ, to\n the place where the IRQ is actually delivered. Ideally, this should make\n the code a bit more robust.\n \n Finally, to improve understanding of the code and the sequence of the busy\n detection, move the corresponding code around a bit in mmci_cmd_irq().\n \n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n Tested-by: Jean Nicolas GRAUX \u003cjean-nicolas.graux@st.com\u003e\n Reviewed-by: Jean Nicolas GRAUX \u003cjean-nicolas.graux@st.com\u003e\n\ncommit 90761a28bb38fc41798e765aff11739ca81f02b7\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 19:50:44 2019 +0800\n\n memstick: r592: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit d97de5d6463fe18619e21fa17474c0bf6a21785f\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 20:47:07 2019 +0800\n\n mmc: sdhci-pci: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Acked-by: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit ffe87671ed01cc85eab7982c379d4b4e16fc0050\nAuthor: Stefan Wahren \u003cwahrenst@gmx.net\u003e\nDate: Mon Jul 22 08:23:23 2019 +0200\n\n mmc: sdhci-iproc: Add support for emmc2 of the BCM2711\n \n The additional emmc2 interface of the BCM2711 is an improved version\n of the old emmc controller, which is able to provide DDR50 mode on the\n Raspberry Pi 4. Except 32 bit only register access no other quirks are\n known yet.\n \n Signed-off-by: Stefan Wahren \u003cwahrenst@gmx.net\u003e\n Reviewed-by: Matthias Brugger \u003cmbrugger@suse.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit 1131a75eb3520647a901de502ce07fa929326eba\nAuthor: Stefan Wahren \u003cwahrenst@gmx.net\u003e\nDate: Mon Jul 22 07:54:40 2019 +0200\n\n dt-bindings: mmc: sdhci-iproc: Add brcm,bcm2711-emmc2\n \n Add a new compatible for the additional emmc2 controller\n on BCM2711 and clearify usage.\n \n Signed-off-by: Stefan Wahren \u003cwahrenst@gmx.net\u003e\n Reviewed-by: Matthias Brugger \u003cmbrugger@suse.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit c35a0c9121c90749794ffe914e2318bd996c0da7\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jul 22 14:07:40 2019 +0200\n\n dt-bindings: mmc: sunxi: Add H5 compatibles\n \n The conversion to the YAML binding left out two compatibles. Add them.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit 7b739fcea70d8f14e8a392791ec3b599e4a5ca99\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jul 22 14:07:39 2019 +0200\n\n dt-bindings: mmc: sunxi: reorder MMC compatibles\n \n The MMC compatibles in the bindings are not ordered as they should, so\n let\u0027s make sure it\u0027s sorted.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit 5e5704c4fd5a11d3ff36ab9dec7491366992c9f6\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jul 17 16:52:59 2019 +0800\n\n mmc: mxs: use devm_platform_ioremap_resource() to simplify code\n \n Use the new helper devm_platform_ioremap_resource() which wraps the\n platform_get_resource() and devm_ioremap_resource() together, to\n simplify the code.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit fafa06569b8d736d26bcebba1c138da7fa7253d6\nAuthor: Lars Persson \u003clars.persson@axis.com\u003e\nDate: Tue Jul 9 15:53:51 2019 +0200\n\n mmc: usdhi6rol0: Add maintainers\n \n The usdhi6rol0 driver is exclusively used for the ARTPEC family of\n SoCs. Other SoCs with the same IP of Panasonic origin use the tmio_mmc\n driver. Therefore we assigner maintainer responsibility to us at Axis\n until the two drivers become unified.\n \n Signed-off-by: Lars Persson \u003clarper@axis.com\u003e\n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n\ncommit 6dc6ec9e04c468d994bff6eb660f3146f94cbfd9\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:47:10 2019 -0400\n\n xprtrdma: Cache free MRs in each rpcrdma_req\n \n Instead of a globally-contended MR free list, cache MRs in each\n rpcrdma_req as they are released. This means acquiring and releasing\n an MR will be lock-free in the common case, even outside the\n transport send lock.\n \n The original idea of per-rpcrdma_req MR free lists was suggested by\n Shirley Ma \u003cshirley.ma@oracle.com\u003e several years ago. I just now\n figured out how to make that idea work with on-demand MR allocation.\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit 68b2c8c1e421096f4b46ac2ac502d25ca067a2a6\nAuthor: Julien Grall \u003cjulien.grall@arm.com\u003e\nDate: Wed Aug 21 10:24:09 2019 +0100\n\n hrtimer: Don\u0027t take expiry_lock when timer is currently migrated\n \n migration_base is used as a placeholder when an hrtimer is migrated to a\n different CPU. In the case that hrtimer_cancel_wait_running() hits a timer\n which is currently migrated it would pointlessly acquire the expiry lock of\n the migration base, which is even not initialized.\n \n Surely it could be initialized, but there is absolutely no point in\n acquiring this lock because the timer is guaranteed not to run it\u0027s\n callback for which the caller waits to finish on that base. So it would\n just do the inc/lock/dec/unlock dance for nothing.\n \n As the base switch is short and non-preemptible, there is no issue when the\n wait function returns immediately.\n \n The timer base and base-\u003ecpu_base cannot be NULL in the code path which is\n invoking that, so just replace those checks with a check whether base is\n migration base.\n \n [ tglx: Updated from RT patch. Massaged changelog. Added comment. ]\n \n Signed-off-by: Julien Grall \u003cjulien.grall@arm.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190821092409.13225-4-julien.grall@arm.com\n\ncommit dd2261ed45aaeddeb77768f291d604179bcab096\nAuthor: Julien Grall \u003cjulien.grall@arm.com\u003e\nDate: Wed Aug 21 10:24:07 2019 +0100\n\n hrtimer: Protect lockless access to timer-\u003ebase\n \n The update to timer-\u003ebase is protected by the base-\u003ecpu_base-\u003elock().\n However, hrtimer_cancel_wait_running() does access it lockless. So the\n compiler is allowed to refetch timer-\u003ebase which can cause havoc when the\n timer base is changed concurrently.\n \n Use READ_ONCE() to prevent this.\n \n [ tglx: Adapted from a RT patch ]\n \n Signed-off-by: Julien Grall \u003cjulien.grall@arm.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190821092409.13225-2-julien.grall@arm.com\n\ncommit 05cbaf4ddd02b69d78d43481e3813b4579572f71\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Aug 20 13:25:36 2019 +0100\n\n iommu/arm-smmu-v3: Document ordering guarantees of command insertion\n \n It turns out that we\u0027ve always relied on some subtle ordering guarantees\n when inserting commands into the SMMUv3 command queue. With the recent\n changes to elide locking when possible, these guarantees become more\n subtle and even more important.\n \n Add a comment documented the barrier semantics of command insertion so\n that we don\u0027t have to derive the behaviour from scratch each time it\n comes up on the list.\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit e6806e9a63a759e445383915bb9d2ec85a90aebf\nAuthor: Mark Zhang \u003cmarkz@mellanox.com\u003e\nDate: Mon Aug 19 14:36:25 2019 +0300\n\n net/mlx5: Create bypass and loopback flow steering namespaces for RDMA RX\n \n Use different namespaces for bypass and switchdev loopback because they\n have different priorities and default table miss action requirement:\n 1. bypass: with multiple priorities support, and\n MLX5_FLOW_TABLE_MISS_ACTION_DEF as the default table miss action;\n 2. switchdev loopback: with single priority support, and\n MLX5_FLOW_TABLE_MISS_ACTION_SWITCH_DOMAIN as the default table miss\n action.\n \n Signed-off-by: Mark Zhang \u003cmarkz@mellanox.com\u003e\n Reviewed-by: Mark Bloch \u003cmarkb@mellanox.com\u003e\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n\ncommit f66ad830b11406cdff84e7d8656a0a9e34b0b606\nAuthor: Mark Zhang \u003cmarkz@mellanox.com\u003e\nDate: Mon Aug 19 14:36:24 2019 +0300\n\n net/mlx5: Add per-namespace flow table default miss action support\n \n Currently all the namespaces under the same steering domain share the same\n default table miss action, however in some situations (e.g., RDMA RX)\n different actions are required. This patch adds a per-namespace default\n table miss action instead of using the miss action of the steering domain.\n \n Signed-off-by: Mark Zhang \u003cmarkz@mellanox.com\u003e\n Reviewed-by: Mark Bloch \u003cmarkb@mellanox.com\u003e\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n\ncommit 7ae7834ec446e5f7fed9bb990d16354853a206d0\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Wed Aug 21 15:06:56 2019 +0200\n\n ASoC: sun4i-i2s: Add support for DSP formats\n \n In addition to the I2S format, the controller also supports the DSP_*\n formats.\n \n This requires some extra care on the LRCK period calculation, since the\n controller, with the PCM formats, require that the value set is no longer\n the periods of LRCK for a single channel, but for all of them.\n \n Let\u0027s add the code to deal with this, and support the DSP_A and DSP_B\n formats.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/5562db1ac8759f12b1b87c3258223eed629ef771.1566392800.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 84884c7ad5e8794aa19e48eaa8de93f4e1d26af9\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Wed Aug 21 15:06:55 2019 +0200\n\n ASoC: sun4i-i2s: Replace call to params_width by local variable\n \n The sun4i_i2s_hw_params function already has a variable holding the value\n returned by params_width, so let\u0027s just use that variable instead of\n calling params_width multiple times.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/f85a1c1e014080a4bbc3abd19bc8fdcb86f0981a.1566392800.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9e8a93ac27d101e0ace024196a4bc3386568cc00\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Wed Aug 21 15:06:54 2019 +0200\n\n ASoC: sun4i-i2s: Use the actual format width instead of an hardcoded one\n \n The LRCK period field in the FMT0 register holds the number of LRCK period\n for one channel in I2S mode.\n \n This has been hardcoded to 32, while it really should be the physical width\n of the format, which creates an improper clock when using a 16bit format,\n with the i2s controller as LRCK master.\n \n Fixes: 7d2993811a1e (\"ASoC: sun4i-i2s: Add support for H3\")\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/f08a0c3605cd1d79752b38d704690190183f7865.1566392800.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 5389f4765789e4ecf9831bc968562befdd2f3bee\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Wed Aug 21 15:06:53 2019 +0200\n\n ASoC: sun4i-i2s: Use the physical / slot width for the clocks\n \n The clock dividers function has been using the word size to compute the\n clock rate at which it\u0027s supposed to be running, but the proper formula\n would be to use the physical width and / or slot width in TDM.\n \n It doesn\u0027t make any difference at the moment since all the formats\n supported have the same sample width and physical width, but it\u0027s not going\n to last forever.\n \n Fixes: 7d2993811a1e (\"ASoC: sun4i-i2s: Add support for H3\")\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/41a359d9885f397e066816961e5e3236afcbe0a1.1566392800.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 6846895fde05558a39ffcee7631a802a923d9b16\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Aug 21 13:32:34 2019 +0100\n\n drm/i915: Replace PIN_NONFAULT with calls to PIN_NOEVICT\n \n When under severe stress for GTT mappable space, the LRU eviction model\n falls off a cliff. We spend all our time scanning the much larger\n non-mappable area searching for something within the mappable zone we can\n evict. Turn this on its head by only using the full vma for the object if\n it is already pinned in the mappable zone or there is sufficient *free*\n space to accommodate it (prioritizing speedy reuse). If there is not,\n immediately fall back to using small chunks (tilerow for GTT mmap, single\n pages for pwrite/relocation) and using random eviction before doing a full\n search.\n \n Testcase: igt/gem_concurrent_blt\n References: https://bugs.freedesktop.org/show_bug.cgi?id\u003d110848\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190821123234.19194-1-chris@chris-wilson.co.uk\n\ncommit 540f62d26f023a6e9c21aacb10b2f880ab2d896a\nAuthor: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nDate: Mon Jun 24 14:32:19 2019 +1000\n\n PCI: Protect pci_reassign_bridge_resources() against addition/removal\n \n pci_reassign_bridge_resources() can be called by pci_resize_resource() at\n runtime, and it walks the PCI tree up and down with no protection against\n hotplug or similar tree changes.\n \n Hold the pci_bus_sem while walking the tree.\n \n Link: https://lore.kernel.org/r/02ca29627597445442bb14c069678e549429dace.camel@kernel.crashing.org\n Signed-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit fdf37037664041cfdbcab6cf54ddb2c1c219034b\nAuthor: Ilya Leoshkevich \u003ciii@linux.ibm.com\u003e\nDate: Tue Aug 20 13:29:39 2019 +0200\n\n btf: do not use CONFIG_OUTPUT_FORMAT\n \n Building s390 kernel with CONFIG_DEBUG_INFO_BTF fails, because\n CONFIG_OUTPUT_FORMAT is not defined. As a matter of fact, this variable\n appears to be x86-only, so other arches might be affected as well.\n \n Fix by obtaining this value from objdump output, just like it\u0027s already\n done for bin_arch. The exact objdump invocation is \"inspired\" by\n arch/powerpc/boot/wrapper.\n \n Also, use LANG\u003dC for the existing bin_arch objdump invocation to avoid\n potential build issues on systems with non-English locale.\n \n Fixes: 341dfcf8d78e (\"btf: expose BTF info through sysfs\")\n Signed-off-by: Ilya Leoshkevich \u003ciii@linux.ibm.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 8467c3b456b57fcb4b3179c34c20cbc9c88956c5\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Sun Aug 18 19:46:49 2019 +0100\n\n staging: rts5208: remove redundant assignment to retval\n \n Variable retval is initialized to a value that is never read and it\n is re-assigned later. The initialization is redundant and can be\n removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190818184649.13828-1-colin.king@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 1327d35ac3c187b3c1ae650d90400f8fe5e41b94\nAuthor: Eduardo Barretto \u003cedusbarretto@gmail.com\u003e\nDate: Sun Aug 18 15:35:55 2019 -0300\n\n staging: kpc2000: kpc2000_i2c: Fix different address spaces warnings\n \n This patch fixes the following sparse warnings:\n \n kpc2000_i2c.c:137: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:137: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:137: got void *\n kpc2000_i2c.c:146: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:146: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:146: got void *\n kpc2000_i2c.c:147: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:147: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:147: got void *\n kpc2000_i2c.c:166: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:166: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:166: got void *\n kpc2000_i2c.c:166: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:166: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:166: got void *\n kpc2000_i2c.c:168: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:168: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:168: got void *\n kpc2000_i2c.c:168: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:168: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:168: got void *\n kpc2000_i2c.c:171: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:171: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:171: got void *\n kpc2000_i2c.c:174: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:174: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:174: got void *\n kpc2000_i2c.c:193: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:193: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:193: got void *\n kpc2000_i2c.c:194: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:194: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:194: got void *\n kpc2000_i2c.c:214: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:214: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:214: got void *\n kpc2000_i2c.c:219: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:219: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:219: got void *\n kpc2000_i2c.c:226: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:226: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:226: got void *\n kpc2000_i2c.c:238: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:238: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:238: got void *\n kpc2000_i2c.c:244: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:244: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:244: got void *\n kpc2000_i2c.c:252: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:252: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:252: got void *\n kpc2000_i2c.c:257: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:257: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:257: got void *\n kpc2000_i2c.c:259: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:259: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:259: got void *\n kpc2000_i2c.c:267: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:267: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:267: got void *\n kpc2000_i2c.c:273: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:273: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:273: got void *\n kpc2000_i2c.c:293: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:293: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:293: got void *\n kpc2000_i2c.c:294: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:294: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:294: got void *\n kpc2000_i2c.c:309: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:309: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:309: got void *\n kpc2000_i2c.c:312: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:312: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:312: got void *\n kpc2000_i2c.c:317: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:317: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:317: got void *\n kpc2000_i2c.c:324: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:324: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:324: got void *\n kpc2000_i2c.c:328: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:328: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:328: got void *\n kpc2000_i2c.c:329: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:329: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:329: got void *\n kpc2000_i2c.c:330: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:330: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:330: got void *\n kpc2000_i2c.c:338: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:338: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:338: got void *\n kpc2000_i2c.c:340: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:340: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:340: got void *\n kpc2000_i2c.c:342: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:342: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:342: got void *\n kpc2000_i2c.c:350: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:350: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:350: got void *\n kpc2000_i2c.c:350: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:350: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:350: got void *\n kpc2000_i2c.c:351: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:351: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:351: got void *\n kpc2000_i2c.c:414: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:414: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:414: got void *\n kpc2000_i2c.c:420: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:420: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:420: got void *\n kpc2000_i2c.c:422: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:422: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:422: got void *\n kpc2000_i2c.c:427: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:427: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:427: got void *\n kpc2000_i2c.c:428: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:428: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:428: got void *\n kpc2000_i2c.c:430: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:430: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:430: got void *\n kpc2000_i2c.c:435: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:435: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:435: got void *\n kpc2000_i2c.c:436: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:436: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:436: got void *\n kpc2000_i2c.c:438: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:438: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:438: got void *\n kpc2000_i2c.c:439: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:439: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:439: got void *\n kpc2000_i2c.c:445: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:445: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:445: got void *\n kpc2000_i2c.c:446: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:446: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:446: got void *\n kpc2000_i2c.c:454: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:454: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:454: got void *\n kpc2000_i2c.c:459: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:459: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:459: got void *\n kpc2000_i2c.c:461: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:461: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:461: got void *\n kpc2000_i2c.c:472: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:472: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:472: got void *\n kpc2000_i2c.c:472: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:472: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:472: got void *\n kpc2000_i2c.c:475: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:475: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:475: got void *\n kpc2000_i2c.c:475: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:475: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:475: got void *\n kpc2000_i2c.c:493: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:493: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:493: got void *\n kpc2000_i2c.c:493: warning: incorrect type in argument 2\n (different address spaces)\n kpc2000_i2c.c:493: expected void volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:493: got void *\n kpc2000_i2c.c:512: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:512: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:512: got void *\n kpc2000_i2c.c:516: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:516: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:516: got void *\n kpc2000_i2c.c:516: warning: incorrect type in argument 1\n (different address spaces)\n kpc2000_i2c.c:516: expected void const volatile [noderef] \u003casn:2\u003e *addr\n kpc2000_i2c.c:516: got void *\n \n Signed-off-by: Eduardo Barretto \u003cedusbarretto@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190818183555.7167-1-edusbarretto@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit be237202ee052ef2885285b4550d9638470ce38d\nAuthor: Donald Yandt \u003cdonald.yandt@gmail.com\u003e\nDate: Sun Aug 18 11:20:23 2019 -0400\n\n staging: android: Remove ion device tree bindings from the TODO\n \n Commit 23a4388f24f5 (\"staging: android: ion: Remove file ion_chunk_heap.c\")\n and eadbf7a34e44 (\"staging: android: ion: Remove file ion_carveout_heap.c\")\n removed the chunk and carveout heaps from ion but left behind the device\n tree bindings for them in the TODO, this patch removes it.\n \n Signed-off-by: Donald Yandt \u003cdonald.yandt@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190818152023.891-1-donald.yandt@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f55ef00e0ea2664e1a2a7bd4e8670d27a0bf2b86\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Sun Aug 18 17:06:09 2019 +0200\n\n staging: rtl8712: Improve naming of include hearder guards\n \n Choose a better name for the include hearder guard used in rtl871x_io.h.\n \u0027_IO_H_\u0027 is to generic and does not match the comment after the #endif.\n Use \u0027_RTL871X_IO_H_\u0027 instead.\n \n Also make the comments in the #endif /* XXX */ match the name used in\n #ifndef.\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Link: https://lore.kernel.org/r/20190818150609.3376-1-christophe.jaillet@wanadoo.fr\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit bfc4ccb1584129d3ef051f8bf6155835374f132f\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Aug 21 11:09:08 2019 +0800\n\n staging: erofs: avoid loop in submit chains\n \n As reported by erofs-utils fuzzer, 2 conditions\n can happen in corrupted images, which can cause\n unexpected behaviors.\n - access the same pcluster one more time;\n - access the tail end pcluster again, e.g.\n _ access again (will trigger tail merging)\n |\n 1 2 3 1 2 -\u003e 1 2 3 1\n |_ tail end of the chain \\___/ (unexpected behavior)\n Let\u0027s detect and avoid them now.\n \n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190821030908.40282-1-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 2cb05d8788f77ed14b99a0fbb2748346071f3a33\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Mon Aug 19 22:27:05 2019 +0530\n\n staging: rtl8723bs: core: Remove unneeded declaration WFD_OUI\n \n Remove unneeded declaration \"extern unsigned char WFD_OUI\"\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190819165705.GA5782@hari-Inspiron-1545\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 88db7fc592559226cda373245770d90af0af3b0a\nAuthor: Masanari Iida \u003cstandby24x7@gmail.com\u003e\nDate: Wed Aug 21 00:33:56 2019 +0900\n\n staging: comedi: ni_mio_common: Fix a typo in ni_mio_common.c\n \n This patch fix a spelling typo in ni_mio_common.c\n \n Signed-off-by: Masanari Iida \u003cstandby24x7@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190820153356.25189-1-standby24x7@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 598bb8913d015150b7734b55443c0e53e7189fc7\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Mon Aug 19 18:34:26 2019 +0800\n\n staging: erofs: avoid endless loop of invalid lookback distance 0\n \n As reported by erofs-utils fuzzer, Lookback distance should\n be a positive number, so it should be actually looked back\n rather than spinning.\n \n Fixes: 02827e1796b3 (\"staging: erofs: add erofs_map_blocks_iter\")\n Cc: \u003cstable@vger.kernel.org\u003e # 4.19+\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190819103426.87579-7-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit bb4b5c08a8e0f854a313184498b22fa37dd087db\nAuthor: Ivan Khoronzhuk \u003civan.khoronzhuk@linaro.org\u003e\nDate: Thu Aug 15 15:13:56 2019 +0300\n\n samples: bpf: syscall_nrs: use mmap2 if defined\n \n For arm32 xdp sockets mmap2 is preferred, so use it if it\u0027s defined.\n Declaration of __NR_mmap can be skipped and it breaks build.\n \n Signed-off-by: Ivan Khoronzhuk \u003civan.khoronzhuk@linaro.org\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 624676e788992e4da8de1fad93bfe51563a7f9a0\nAuthor: Ivan Khoronzhuk \u003civan.khoronzhuk@linaro.org\u003e\nDate: Thu Aug 15 15:13:55 2019 +0300\n\n xdp: xdp_umem: replace kmap on vmap for umem map\n \n For 64-bit there is no reason to use vmap/vunmap, so use page_address\n as it was initially. For 32 bits, in some apps, like in samples\n xdpsock_user.c when number of pgs in use is quite big, the kmap\n memory can be not enough, despite on this, kmap looks like is\n deprecated in such cases as it can block and should be used rather\n for dynamic mm.\n \n Signed-off-by: Ivan Khoronzhuk \u003civan.khoronzhuk@linaro.org\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 71dd77fd4bf7d1675a95dfe04a99669ce15b58f8\nAuthor: Ivan Khoronzhuk \u003civan.khoronzhuk@linaro.org\u003e\nDate: Thu Aug 15 15:13:54 2019 +0300\n\n libbpf: use LFS (_FILE_OFFSET_BITS) instead of direct mmap2 syscall\n \n Drop __NR_mmap2 fork in flavor of LFS, that is _FILE_OFFSET_BITS\u003d64\n (glibc \u0026 bionic) / LARGEFILE64_SOURCE (for musl) decision. It allows\n mmap() to use 64bit offset that is passed to mmap2 syscall. As result\n pgoff is not truncated and no need to use direct access to mmap2 for\n 32 bits systems.\n \n Signed-off-by: Ivan Khoronzhuk \u003civan.khoronzhuk@linaro.org\u003e\n Acked-by: Yonghong Song \u003cyhs@fb.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 42ac26d253ebe7a5f409443f2fbd239d66b65f57\nAuthor: Santosh Sivaraj \u003csantosh@fossix.org\u003e\nDate: Tue Aug 20 13:43:52 2019 +0530\n\n powerpc: add machine check safe copy_to_user\n \n Use memcpy_mcsafe() implementation to define copy_to_user_mcsafe()\n \n Signed-off-by: Santosh Sivaraj \u003csantosh@fossix.org\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820081352.8641-8-santosh@fossix.org\n\ncommit 4d4a273854c9eac1a1b163830658dc825e183ad5\nAuthor: Balbir Singh \u003cbsingharora@gmail.com\u003e\nDate: Tue Aug 20 13:43:51 2019 +0530\n\n powerpc/memcpy: Add memcpy_mcsafe for pmem\n \n The pmem infrastructure uses memcpy_mcsafe in the pmem layer so as to\n convert machine check exceptions into a return value on failure in case\n a machine check exception is encountered during the memcpy. The return\n value is the number of bytes remaining to be copied.\n \n This patch largely borrows from the copyuser_power7 logic and does not add\n the VMX optimizations, largely to keep the patch simple. If needed those\n optimizations can be folded in.\n \n Signed-off-by: Balbir Singh \u003cbsingharora@gmail.com\u003e\n [arbab@linux.ibm.com: Added symbol export]\n Co-developed-by: Santosh Sivaraj \u003csantosh@fossix.org\u003e\n Signed-off-by: Santosh Sivaraj \u003csantosh@fossix.org\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820081352.8641-7-santosh@fossix.org\n\ncommit 895e3dceeb97855dc9990136cbb80a842fe581aa\nAuthor: Balbir Singh \u003cbsingharora@gmail.com\u003e\nDate: Tue Aug 20 13:43:50 2019 +0530\n\n powerpc/mce: Handle UE event for memcpy_mcsafe\n \n If we take a UE on one of the instructions with a fixup entry, set nip\n to continue execution at the fixup entry. Stop processing the event\n further or print it.\n \n Co-developed-by: Reza Arbab \u003carbab@linux.ibm.com\u003e\n Signed-off-by: Reza Arbab \u003carbab@linux.ibm.com\u003e\n Signed-off-by: Balbir Singh \u003cbsingharora@gmail.com\u003e\n Reviewed-by: Mahesh Salgaonkar \u003cmahesh@linux.vnet.ibm.com\u003e\n Reviewed-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Santosh Sivaraj \u003csantosh@fossix.org\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820081352.8641-6-santosh@fossix.org\n\ncommit 49ec9177b8ec9c081850744468811e708a7c9841\nAuthor: Santosh Sivaraj \u003csantosh@fossix.org\u003e\nDate: Tue Aug 20 13:43:49 2019 +0530\n\n extable: Add function to search only kernel exception table\n \n Certain architecture specific operating modes (e.g., in powerpc machine\n check handler that is unable to access vmalloc memory), the\n search_exception_tables cannot be called because it also searches the\n module exception tables if entry is not found in the kernel exception\n table.\n \n Signed-off-by: Santosh Sivaraj \u003csantosh@fossix.org\u003e\n Reviewed-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820081352.8641-5-santosh@fossix.org\n\ncommit 1a1715f516fd7fcfedffee5978ec4c07c5164470\nAuthor: Reza Arbab \u003carbab@linux.ibm.com\u003e\nDate: Tue Aug 20 13:43:48 2019 +0530\n\n powerpc/mce: Make machine_check_ue_event() static\n \n The function doesn\u0027t get used outside this file, so make it static.\n \n Signed-off-by: Reza Arbab \u003carbab@linux.ibm.com\u003e\n Signed-off-by: Santosh Sivaraj \u003csantosh@fossix.org\u003e\n Reviewed-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820081352.8641-4-santosh@fossix.org\n\ncommit 99ead78afd1128bfcebe7f88f3b102fb2da09aee\nAuthor: Balbir Singh \u003cbsingharora@gmail.com\u003e\nDate: Tue Aug 20 13:43:47 2019 +0530\n\n powerpc/mce: Fix MCE handling for huge pages\n \n The current code would fail on huge pages addresses, since the shift would\n be incorrect. Use the correct page shift value returned by\n __find_linux_pte() to get the correct physical address. The code is more\n generic and can handle both regular and compound pages.\n \n Fixes: ba41e1e1ccb9 (\"powerpc/mce: Hookup derror (load/store) UE errors\")\n Signed-off-by: Balbir Singh \u003cbsingharora@gmail.com\u003e\n [arbab@linux.ibm.com: Fixup pseries_do_memory_failure()]\n Signed-off-by: Reza Arbab \u003carbab@linux.ibm.com\u003e\n Tested-by: Mahesh Salgaonkar \u003cmahesh@linux.vnet.ibm.com\u003e\n Signed-off-by: Santosh Sivaraj \u003csantosh@fossix.org\u003e\n Cc: stable@vger.kernel.org # v4.15+\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820081352.8641-3-santosh@fossix.org\n\ncommit b5bda6263cad9a927e1a4edb7493d542da0c1410\nAuthor: Santosh Sivaraj \u003csantosh@fossix.org\u003e\nDate: Tue Aug 20 13:43:46 2019 +0530\n\n powerpc/mce: Schedule work from irq_work\n \n schedule_work() cannot be called from MCE exception context as MCE can\n interrupt even in interrupt disabled context.\n \n Fixes: 733e4a4c4467 (\"powerpc/mce: hookup memory_failure for UE errors\")\n Cc: stable@vger.kernel.org # v4.15+\n Reviewed-by: Mahesh Salgaonkar \u003cmahesh@linux.vnet.ibm.com\u003e\n Reviewed-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Acked-by: Balbir Singh \u003cbsingharora@gmail.com\u003e\n Signed-off-by: Santosh Sivaraj \u003csantosh@fossix.org\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190820081352.8641-2-santosh@fossix.org\n\ncommit 38290431d56d7d3928ac89e9f8d3d6b3c8df4c6e\nAuthor: Jacopo Mondi \u003cjacopo+renesas@jmondi.org\u003e\nDate: Sat Jul 6 16:07:30 2019 +0200\n\n arm64: dts: renesas: Update \u0027vsps\u0027 properties for readability\n \n Update the \u0027vsps\u0027 property in the R-Car Gen3 SoC device tree files to\n match what\u0027s in the documentation example.\n \n Signed-off-by: Jacopo Mondi \u003cjacopo+renesas@jmondi.org\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit fe9841595c604ce942c3729a72411096c3f4b675\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Wed Feb 27 14:02:54 2019 +0900\n\n arm64: dts: renesas: r8a77965-salvator-x(s): Enable HDMI sound\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Tested-by: Jacopo Mondi \u003cjacopo+renesas@jmondi.org\u003e # Salvator-XS\n [geert: Preserve sort order]\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 137befe19f310400a8b20fd8a4ce8c4141aafde0\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 21:25:27 2019 +0200\n\n ASoC: sun4i-i2s: Add support for TDM slots\n \n The i2s controller supports TDM, for up to 8 slots. Let\u0027s support the TDM\n API.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/26392af30b3e7b31ee48d5b867d45be8675db046.1566242458.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c26a8841119826badc8d358a4266880f83359f26\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Tue Aug 20 14:04:58 2019 +0900\n\n ASoC: soc-core: remove unneeded dai_link check from snd_soc_remove_dai_link()\n \n snd_soc_remove_dai_link() has card connected dai_link check. but\n 1) we need to call list_del() anyway,\n because it is \"remove\" function,\n 2) It is doing many thing for this card / dai_link already\n before checking dai_link.\n \n This patch removes poinless dai_link check\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Reviewed-by: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/875zms1ldm.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit b03bfaec1d52123d5d941488f71e06964535e471\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Tue Aug 20 14:04:54 2019 +0900\n\n ASoC: soc-core: merge snd_soc_initialize_card_lists()\n \n snd_soc_initialize_card_lists() is doing card related\n INIT_LIST_HEAD(), but, it is already doing at\n snd_soc_register_card(). We don\u0027t need to do it separately.\n This patch merges these.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Reviewed-by: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/877e781ldq.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c1fe93581ae9d85c3a783b5fad21912bb88a0f34\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Wed Aug 14 11:29:11 2019 +0300\n\n ASoC: dt-bindings: Introduce compatible string for imx8qm\n \n Register map for i.MX8QM is similar with i.MX6 series. Integration\n of SAI IP into i.MX8QM SOC features a FIFO size of 64 X 32 bits samples.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190814082911.665-3-daniel.baluta@nxp.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 6eeb60be5ebb73b2e5911e26fb1aed02940b7d09\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Wed Aug 14 11:29:10 2019 +0300\n\n ASoC: fsl_sai: Add support for imx8qm\n \n SAI module on imx8qm features a register map similar with imx6 series\n (it doesn\u0027t have VERID and PARAM registers at the beginning\n of address spece).\n \n Also, it has one FIFO which can help up to 64 * 32 bit samples.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190814082911.665-2-daniel.baluta@nxp.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 10df063855822fcea3c0f51dbf534ad643d3cb1b\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Thu Aug 15 01:06:22 2019 +0900\n\n kbuild: rebuild modules when module linker scripts are updated\n \n Currently, the timestamp of module linker scripts are not checked.\n Add them to the dependency of modules so they are correctly rebuilt.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 888f0c346ff01d544d1cb9da6395a7b3def7fe87\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Thu Aug 15 01:06:21 2019 +0900\n\n kbuild: move KBUILD_LDS, KBUILD_VMLINUX_{OBJS,LIBS} to makefiles.rst\n \n These three variables are not intended to be tweaked by users.\n Move them from kbuild.rst to makefiles.rst.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit cbdf59ad65ebb2d022663805b787aafaedb93f1d\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Thu Aug 15 00:19:19 2019 +0900\n\n treewide: remove dummy Makefiles for single targets\n \n Now that the single target build descends into sub-directories in the\n same way as the normal build, these dummy Makefiles are not needed\n any more.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 394053f4a4b3e3eeeaa67b67fc886a9a75bd9e4d\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Thu Aug 15 00:19:18 2019 +0900\n\n kbuild: make single targets work more correctly\n \n Currently, the single target build directly descends into the directory\n of the target. For example,\n \n $ make foo/bar/baz.o\n \n ... directly descends into foo/bar/.\n \n On the other hand, the normal build usually descends one directory at\n a time, i.e. descends into foo/, and then foo/bar/.\n \n This difference causes some problems.\n \n [1] miss subdir-asflags-y, subdir-ccflags-y in upper Makefiles\n \n The options in subdir-{as,cc}flags-y take effect in the current\n and its sub-directories. In other words, they are inherited\n downward. In the example above, the single target will miss\n subdir-{as,cc}flags-y if they are defined in foo/Makefile.\n \n [2] could be built in a different directory\n \n As Documentation/kbuild/modules.rst section 4.3 says, Kbuild can\n handle files that are spread over several sub-directories.\n \n The build rule of foo/bar/baz.o may not necessarily be specified in\n foo/bar/Makefile. It might be specifies in foo/Makefile as follows:\n \n [foo/Makefile]\n obj-y :\u003d bar/baz.o\n \n This often happens when a module is so big that its source files\n are divided into sub-directories.\n \n In this case, there is no Makefile in the foo/bar/ directory, yet\n the single target descends into foo/bar/, then fails due to the\n missing Makefile. You can still do \u0027make foo/bar/\u0027 for partial\n building, but cannot do \u0027make foo/bar/baz.s\u0027. I believe the single\n target \u0027%.s\u0027 is a useful feature for inspecting the compiler output.\n \n Some modules work around this issue by putting an empty Makefile\n in every sub-directory.\n \n This commit fixes those problems by making the single target build\n descend in the same way as the normal build does.\n \n Another change is the single target build will observe the CONFIG\n options. Previously, it allowed users to build the foo.o even when\n the corresponding CONFIG_FOO is disabled:\n \n obj-$(CONFIG_FOO) +\u003d foo.o\n \n In the new behavior, the single target build will just fail and show\n \"No rule to make target ...\" (or \"Nothing to be done for ...\" if the\n stale object already exists, but cannot be updated).\n \n The disadvantage of this commit is the build speed. Now that the\n single target build visits every directory and parses lots of\n Makefiles, it is slower than before. (But, I hope it will not be\n too slow.)\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 8959e39272d6e625da1cd62f2e7622d79e04447d\nAuthor: Kees Cook \u003ckeescook@chromium.org\u003e\nDate: Tue Aug 13 08:15:32 2019 -0700\n\n kbuild: Parameterize kallsyms generation and correct reporting\n \n When kallsyms generation happens, temporary vmlinux outputs are linked\n but the quiet make output didn\u0027t report it, giving the impression that\n the prior command is taking longer than expected.\n \n Instead, report the linking step explicitly. While at it, this\n consolidates the repeated \"kallsyms generation step\" into a single\n function and removes the existing copy/pasting.\n \n Signed-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit c7c0eecf894c3e353499fffd6f2ddaefc89bbd4a\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sun Aug 11 02:01:35 2019 +0900\n\n kbuild: re-implement detection of CONFIG options leaked to user-space\n \n scripts/headers_check.pl can detect references to CONFIG options in\n exported headers, but it has been disabled for more than a decade.\n \n Reverting commit 7e3fa5614117 (\"kbuild: drop check for CONFIG_ in\n headers_check\") would emit the following warnings for headers_check\n on x86:\n \n usr/include/mtd/ubi-user.h:283: leaks CONFIG_MTD_UBI_BEB_LIMIT to userspace where it is not valid\n usr/include/linux/cm4000_cs.h:26: leaks CONFIG_COMPAT to userspace where it is not valid\n usr/include/linux/pkt_cls.h:301: leaks CONFIG_NET_CLS_ACT to userspace where it is not valid\n usr/include/linux/videodev2.h:2465: leaks CONFIG_VIDEO_ADV_DEBUG to userspace where it is not valid\n usr/include/linux/bpf.h:249: leaks CONFIG_EFFICIENT_UNALIGNED_ACCESS to userspace where it is not valid\n usr/include/linux/bpf.h:819: leaks CONFIG_CGROUP_NET_CLASSID to userspace where it is not valid\n usr/include/linux/bpf.h:1011: leaks CONFIG_IP_ROUTE_CLASSID to userspace where it is not valid\n usr/include/linux/bpf.h:1742: leaks CONFIG_BPF_KPROBE_OVERRIDE to userspace where it is not valid\n usr/include/linux/bpf.h:1747: leaks CONFIG_FUNCTION_ERROR_INJECTION to userspace where it is not valid\n usr/include/linux/bpf.h:1936: leaks CONFIG_XFRM to userspace where it is not valid\n usr/include/linux/bpf.h:2184: leaks CONFIG_BPF_LIRC_MODE2 to userspace where it is not valid\n usr/include/linux/bpf.h:2210: leaks CONFIG_BPF_LIRC_MODE2 to userspace where it is not valid\n usr/include/linux/bpf.h:2227: leaks CONFIG_SOCK_CGROUP_DATA to userspace where it is not valid\n usr/include/linux/bpf.h:2311: leaks CONFIG_NET to userspace where it is not valid\n usr/include/linux/bpf.h:2348: leaks CONFIG_NET to userspace where it is not valid\n usr/include/linux/bpf.h:2422: leaks CONFIG_BPF_LIRC_MODE2 to userspace where it is not valid\n usr/include/linux/bpf.h:2528: leaks CONFIG_NET to userspace where it is not valid\n usr/include/linux/pktcdvd.h:37: leaks CONFIG_CDROM_PKTCDVD_WCACHE to userspace where it is not valid\n usr/include/linux/hw_breakpoint.h:27: leaks CONFIG_HAVE_MIXED_BREAKPOINTS_REGS to userspace where it is not valid\n usr/include/linux/raw.h:17: leaks CONFIG_MAX_RAW_DEVS to userspace where it is not valid\n usr/include/linux/elfcore.h:62: leaks CONFIG_BINFMT_ELF_FDPIC to userspace where it is not valid\n usr/include/linux/eventpoll.h:82: leaks CONFIG_PM_SLEEP to userspace where it is not valid\n usr/include/linux/atmdev.h:104: leaks CONFIG_COMPAT to userspace where it is not valid\n usr/include/asm-generic/unistd.h:651: leaks CONFIG_MMU to userspace where it is not valid\n usr/include/asm-generic/bitsperlong.h:9: leaks CONFIG_64BIT to userspace where it is not valid\n usr/include/asm-generic/fcntl.h:119: leaks CONFIG_64BIT to userspace where it is not valid\n usr/include/asm/auxvec.h:14: leaks CONFIG_IA32_EMULATION to userspace where it is not valid\n usr/include/asm/e820.h:14: leaks CONFIG_NODES_SHIFT to userspace where it is not valid\n usr/include/asm/e820.h:39: leaks CONFIG_X86_PMEM_LEGACY to userspace where it is not valid\n usr/include/asm/e820.h:49: leaks CONFIG_INTEL_TXT to userspace where it is not valid\n usr/include/asm/mman.h:7: leaks CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS to userspace where it is not valid\n \n Most of these are false positives because scripts/headers_check.pl\n parses comment lines.\n \n It is also false negative. arch/x86/include/uapi/asm/auxvec.h contains\n CONFIG_IA32_EMULATION and CONFIG_X86_64, but the only former is reported.\n \n It would be possible to fix scripts/headers_check.pl, of course.\n However, we already have some duplicated checks between headers_check\n and CONFIG_UAPI_HEADER_TEST. At this moment of time, there are still\n dozens of headers excluded from the header test (usr/include/Makefile),\n but we might be able to remove headers_check eventually.\n \n I re-implemented it in scripts/headers_install.sh by using sed because\n the most of code in scripts/headers_install.sh is written in sed.\n \n This patch works like this:\n \n [1] Run scripts/unifdef first because we need to drop the code\n surrounded by #ifdef __KERNEL__ ... #endif\n \n [2] Remove all C style comments. The sed code is somewhat complicated\n since we need to deal with both single and multi line comments.\n \n Precisely speaking, a comment block is replaced with a space just\n in case.\n \n CONFIG_FOO/* this is a comment */CONFIG_BAR\n \n should be converted into:\n \n CONFIG_FOO CONFIG_BAR\n \n instead of:\n \n CONFIG_FOOCONFIG_BAR\n \n [3] Match CONFIG_... pattern. It correctly matches to all CONFIG\n options that appear in a single line.\n \n After this commit, this would detect the following warnings, all of\n which are real ones.\n \n warning: include/uapi/linux/pktcdvd.h: leak CONFIG_CDROM_PKTCDVD_WCACHE to user-space\n warning: include/uapi/linux/hw_breakpoint.h: leak CONFIG_HAVE_MIXED_BREAKPOINTS_REGS to user-space\n warning: include/uapi/linux/raw.h: leak CONFIG_MAX_RAW_DEVS to user-space\n warning: include/uapi/linux/elfcore.h: leak CONFIG_BINFMT_ELF_FDPIC to user-space\n warning: include/uapi/linux/eventpoll.h: leak CONFIG_PM_SLEEP to user-space\n warning: include/uapi/linux/atmdev.h: leak CONFIG_COMPAT to user-space\n warning: include/uapi/asm-generic/fcntl.h: leak CONFIG_64BIT to user-space\n warning: arch/x86/include/uapi/asm/auxvec.h: leak CONFIG_IA32_EMULATION to user-space\n warning: arch/x86/include/uapi/asm/auxvec.h: leak CONFIG_X86_64 to user-space\n warning: arch/x86/include/uapi/asm/mman.h: leak CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS to user-space\n \n However, it is not nice to show them right now. I created a list of\n existing leakages. They are not warned, but a new leakage will be\n blocked by the 0-day bot.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n\ncommit 76cd306d797923aabcf373289b231bf8d9a9340a\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sun Aug 11 00:53:05 2019 +0900\n\n kbuild: unify clean-dirs rule for in-kernel and external module\n \n Factor out the duplicated code for in-kernel and external module\n cleaning.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit c99f3918cf0a64b99ac145591c7984891c64baa4\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sun Aug 11 00:53:04 2019 +0900\n\n kbuild: unify vmlinux-dirs and module-dirs rules\n \n The in-kernel build and external module build have similar code\n for descending into sub-directories.\n \n Factor out the code into the common place.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 2042b5486bd311db67b85915ee6291905b72e270\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sun Aug 11 00:53:03 2019 +0900\n\n kbuild: unset variables in top Makefile instead of setting 0\n \n There is no need to set 0 to variables such as config-targets,\n mixed-targets, etc.\n \n Unset instead of setting 0 in order to use \u0027ifdef\u0027 to test them.\n \n I also renamed:\n \n config-targets -\u003e config-build\n mixed-targets -\u003e mixed-build\n dot-config -\u003e need-config\n \n to clarify what we are doing.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 125d059b624180b2c441181c797e41354bfe0649\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sun Aug 11 00:53:02 2019 +0900\n\n kbuild: do not descend to ./Kbuild when cleaning\n \n \u0027make clean\u0027 descends into ./Kbuild, but does not clean anything\n since everything is added to no-clean-files.\n \n There is no need to descend to ./Kbuild in the first place.\n We can drop the no-clean-files assignment.\n \n With this, there is no more user of no-clean-files. I will keep it\n for a while to see whether a new user will appear.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 23ad2b4672a7572e0f091cfe90aac1cd9bdca28a\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Aug 21 11:52:19 2019 +0200\n\n arm64: dts: renesas: r8a774c0: Fix register range of display node\n \n Since the R8A774C0 SoC uses DU{0,1} only, the register block length\n should be 0x40000.\n \n Based on commit 06585ed38b6698bc (\"arm64: dts: renesas: r8a77990: Fix\n register range of display node\") for R-Car E3.\n \n Fixes: 8ed3a6b223159df3 (\"arm64: dts: renesas: r8a774c0: Add display output support\")\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n\ncommit 019cbd4a4feb3aa3a917d78e7110e3011bbff6d5\nAuthor: Thierry Reding \u003ctreding@nvidia.com\u003e\nDate: Wed Aug 14 11:00:48 2019 +0200\n\n drm/nouveau: Initialize GEM object before TTM object\n \n TTM assumes that drivers initialize the embedded GEM object before\n calling the ttm_bo_init() function. This is not currently the case\n in the Nouveau driver. Fix this by splitting up nouveau_bo_new()\n into nouveau_bo_alloc() and nouveau_bo_init() so that the GEM can\n be initialized before TTM BO initialization when necessary.\n \n Fixes: b96f3e7c8069 (\"drm/ttm: use gem vma_node\")\n Acked-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Acked-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n Signed-off-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190814093524.GA31345@ulmo\n\ncommit ac1213b7d79580ca52d7517e61e1ec4120134f0c\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Wed Jul 31 16:13:42 2019 +0100\n\n gpio/ixp4xx: Register the base PA instead of its VA in fwnode\n \n Do not expose the base VA (it appears in debugfs). Instead,\n record the PA, which at least can be used to precisely identify\n the associated irqchip and domain.\n \n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Acked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 416f79c23dbe47e0e223efc06d3487e1d90a92ee\nAuthor: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\nDate: Fri Jul 19 15:22:26 2019 +0200\n\n s390/paes: Prepare paes functions for large key blobs\n \n The context used to store the key blob used a fixed 80 bytes\n buffer. And all the set_key functions did not even check the given key\n size. With CCA variable length AES cipher keys there come key blobs\n with about 136 bytes and maybe in the future there will arise the need\n to store even bigger key blobs.\n \n This patch reworks the paes set_key functions and the context\n buffers to work with small key blobs (\u003c\u003d 128 bytes) directly in the\n context buffer and larger blobs by allocating additional memory and\n storing the pointer in the context buffer. If there has been memory\n allocated for storing a key blob, it also needs to be freed on release\n of the tfm. So all the paes ciphers now have a init and exit function\n implemented for this job.\n \n Signed-off-by: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\n Reviewed-by: Ingo Franzki \u003cifranzki@linux.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit f2bbc96e7cfad3891b7bf9bd3e566b9b7ab4553d\nAuthor: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\nDate: Wed Jun 19 14:26:05 2019 +0200\n\n s390/pkey: add CCA AES cipher key support\n \n Introduce new ioctls and structs to be used with these new ioctls\n which are able to handle CCA AES secure keys and CCA AES cipher keys:\n \n PKEY_GENSECK2: Generate secure key, version 2.\n Generate either a CCA AES secure key or a CCA AES cipher key.\n \n PKEY_CLR2SECK2: Generate secure key from clear key value, version 2.\n Construct a CCA AES secure key or CCA AES cipher key from a given\n clear key value.\n \n PKEY_VERIFYKEY2: Verify the given secure key, version 2.\n Check for correct key type. If cardnr and domain are given, also\n check if this apqn is able to handle this type of key. If cardnr and\n domain are 0xFFFF, on return these values are filled with an apqn\n able to handle this key. The function also checks for the master key\n verification patterns of the key matching to the current or\n alternate mkvp of the apqn. CCA AES cipher keys are also checked\n for CPACF export allowed (CPRTCPAC flag). Currently CCA AES secure\n keys and CCA AES cipher keys are supported (may get extended in the\n future).\n \n PKEY_KBLOB2PROTK2: Transform a key blob (of any type) into\n a protected key, version 2. Difference to version 1 is only that\n this new ioctl has additional parameters to provide a list of\n apqns to be used for the transformation.\n \n PKEY_APQNS4K: Generate a list of APQNs based on the key blob given.\n Is able to find out which type of secure key is given (CCA AES\n secure key or CCA AES cipher key) and tries to find all matching\n crypto cards based on the MKVP and maybe other criterias (like CCA\n AES cipher keys need a CEX6C or higher). The list of APQNs is\n further filtered by the key\u0027s mkvp which needs to match to either\n the current mkvp or the alternate mkvp (which is the old mkvp on CCA\n adapters) of the apqns. The flags argument may be used to limit the\n matching apqns. If the PKEY_FLAGS_MATCH_CUR_MKVP is given, only the\n current mkvp of each apqn is compared. Likewise with the\n PKEY_FLAGS_MATCH_ALT_MKVP. If both are given it is assumed to return\n apqns where either the current or the alternate mkvp matches. If no\n matching APQN is found, the ioctl returns with 0 but the\n apqn_entries value is 0.\n \n PKEY_APQNS4KT: Generate a list of APQNs based on the key type given.\n Build a list of APQNs based on the given key type and maybe further\n restrict the list by given master key verification patterns.\n For different key types there may be different ways to match the\n master key verification patterns. For CCA keys (CCA data key and CCA\n cipher key) the first 8 bytes of cur_mkvp refer to the current mkvp\n value of the apqn and the first 8 bytes of the alt_mkvp refer to the\n old mkvp. The flags argument controls if the apqns current and/or\n alternate mkvp should match. If the PKEY_FLAGS_MATCH_CUR_MKVP is\n given, only the current mkvp of each apqn is compared. Likewise with\n the PKEY_FLAGS_MATCH_ALT_MKVP. If both are given, it is assumed to\n return apqns where either the current or the alternate mkvp\n matches. If no matching APQN is found, the ioctl returns with 0 but\n the apqn_entries value is 0.\n \n These new ioctls are now prepared for another new type of secure key\n blob which may come in the future. They all use a pointer to the key\n blob and a key blob length information instead of some hardcoded byte\n array. They all use the new enums pkey_key_type, pkey_key_size and\n pkey_key_info for getting/setting key type, key size and additional\n info about the key. All but the PKEY_VERIFY2 ioctl now work based on a\n list of apqns. This list is walked through trying to perform the\n operation on exactly this apqn without any further checking (like card\n type or online state). If the apqn fails, simple the next one in the\n list is tried until success (return 0) or the end of the list is\n reached (return -1 with errno ENODEV). All apqns in the list need to\n be exact apqns (0xFFFF as any card or domain is not allowed). There\n are two new ioctls which can be used to build a list of apqns based on\n a key or key type and maybe restricted by match to a current or\n alternate master key verifcation pattern.\n \n Signed-off-by: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\n Reviewed-by: Ingo Franzki \u003cifranzki@linux.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 4bc123b18ce6ae6c42c69d0456b5acbd2f7bc8bd\nAuthor: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\nDate: Wed Jul 3 13:16:51 2019 +0200\n\n s390/zcrypt: Add low level functions for CCA AES cipher keys\n \n This patch adds low level functions, structs and defines to support\n CCA AES cipher keys:\n - struct cipherkeytoken can be used for an inside view of the CCA AES\n cipher key token blob.\n - function cca_cipher2protkey() derives an CPACF protected key from an\n CCA AES cipher key.\n - function cca_gencipherkey() generates an CCA AES cipher key with\n random value.\n - function cca_findcard2() constructs a list of apqns based on input\n constrains like min hardware type, mkvp values.\n - cca_check_secaescipherkey() does a check on the given CCA AES cipher\n key blob.\n - cca_clr2cipherkey() generates an CCA AES cipher key from a given\n clear key value.\n \n Signed-off-by: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\n Reviewed-by: Ingo Franzki \u003cifranzki@linux.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 4da57a2fea064f662c29e77da043baebb8d6cdc8\nAuthor: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\nDate: Tue Jun 18 15:53:12 2019 +0200\n\n s390/zcrypt: extend cca_findcard function and helper\n \n Rework and extension of the cca_findcard function to be prepared for\n other types of secure key blobs. Split the function and extract an\n internal function which has no awareness of key blobs any\n more. Improve this function and the helper code around to be able to\n check for a minimal crypto card hardware level (Background: the newer\n AES cipher keys need to match to the master key verification pattern\n and need to have a crypto card CEX6 or higher).\n \n No API change, neither for the in-kernel API nor the ioctl interface.\n \n Signed-off-by: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\n Reviewed-by: Ingo Franzki \u003cifranzki@linux.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 183cb46954dd204e3578a25ad1284aab3debec52\nAuthor: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\nDate: Wed Jul 3 13:09:03 2019 +0200\n\n s390/pkey: pkey cleanup: narrow in-kernel API, fix some variable types\n \n There are a lot of pkey functions exported as in-kernel callable\n API functions but not used at all. This patch narrows down the\n pkey in-kernel API to what is currently only used and exploited.\n \n Within the kernel just use u32 without any leading __u32. Also\n functions declared in a header file in arch/s390/include/asm\n don\u0027t need a comment \u0027In-kernel API\u0027, this is by definition,\n otherwise the header file would be in arch/s390/include/uapi/asm.\n \n Signed-off-by: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\n Reviewed-by: Ingo Franzki \u003cifranzki@linux.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 22a33c7e4ed6aae8c47de72461162021dad87473\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Fri Aug 16 10:43:55 2019 +0200\n\n s390/startup: round down \"mem\" option to page boundary\n \n Make a usable value out of \"mem\" option once and for all. Kasan memory\n allocator just takes memory_end or online memory size as allocation\n base. If memory_end is not aligned paging structures allocated in kasan\n end up unaligned as well. So this change fixes potential kasan crash\n as well.\n \n Fixes: 78333d1f908a (\"s390/kasan: add support for mem\u003d kernel parameter\")\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 80ef517b018257220f14410004ede094c577c276\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Thu Aug 15 14:49:02 2019 +0200\n\n s390/startup: adjust _sdma and _edma to page boundaries\n \n Move .dma.text section alignment out of section description, otherwise\n zeros used to align the section are included in the section itself (and\n section is not really aligned by itself).\n $ objdump -h arch/s390/boot/compressed/vmlinux\n 5 .dma.text 00001e38 000000000001b1c8 000000000001b1c8 0001c1c8 2**2\n CONTENTS, ALLOC, LOAD, READONLY, CODE\n 6 .dma.ex_table 00000018 000000000001d000 000000000001d000 0001e000 2**2\n CONTENTS, ALLOC, LOAD, READONLY, DATA\n 7 .dma.data 00000240 000000000001d080 000000000001d080 0001e080 2**7\n CONTENTS, ALLOC, LOAD, DATA\n $ cat /sys/kernel/debug/memblock/reserved\n 0: 0x0000000000000000..0x0000000000011fff\n 1: 0x000000000001b1c8..0x000000000001d2bf\n ...\n \n Also add alignment before _edma linker symbol definition, so that entire\n .dma* region is rounded up to page boundaries.\n $ objdump -h arch/s390/boot/compressed/vmlinux\n 5 .dma.text 00001000 000000000001c000 000000000001c000 0001d000 2**2\n CONTENTS, ALLOC, LOAD, READONLY, CODE\n 6 .dma.ex_table 00000018 000000000001d000 000000000001d000 0001e000 2**2\n CONTENTS, ALLOC, LOAD, READONLY, DATA\n 7 .dma.data 00000240 000000000001d080 000000000001d080 0001e080 2**7\n CONTENTS, ALLOC, LOAD, DATA\n $ cat /sys/kernel/debug/memblock/reserved\n 0: 0x0000000000000000..0x0000000000011fff\n 1: 0x000000000001c000..0x000000000001dfff\n ...\n $ cat /sys/kernel/debug/kernel_page_tables\n ---[ Identity Mapping ]---\n 0x0000000000000000-0x000000000001c000 112K PTE RW NX\n 0x000000000001c000-0x000000000001d000 4K PTE RO X\n 0x000000000001d000-0x0000000000100000 908K PTE RW NX\n ...\n \n Reviewed-by: Gerald Schaefer \u003cgerald.schaefer@de.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit e991e5bb11d6eacf8a49867ff9d4ec6e1cde3718\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Wed Aug 14 14:27:44 2019 +0200\n\n s390/stacktrace: use common arch_stack_walk infrastructure\n \n Use common arch_stack_walk infrastructure to avoid duplicated code and\n avoid taking care of the stack storage and filtering.\n \n Common code also uses try_get_task_stack/put_task_stack when needed which\n have been missing in our code, which also solves potential problem for us.\n \n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 2c7fa8a11cc528e49e88352fce8cf083104b3797\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Tue Aug 13 19:23:51 2019 +0200\n\n s390/kasan: avoid report in get_wchan\n \n Reading other running task\u0027s stack can be a dangerous endeavor. Kasan\n stack memory access instrumentation includes special prologue and epilogue\n to mark/remove red zones in shadow memory between stack variables. For\n that reason there is always a race between a task reading value in other\n task\u0027s stack and that other task returning from a function and entering\n another one generating different red zones pattern.\n \n To avoid kasan reports simply perform uninstrumented memory reads.\n \n Acked-by: Ilya Leoshkevich \u003ciii@linux.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 8769f610fe6d473e5e8e221709c3ac402037da6c\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Tue Aug 13 20:11:08 2019 +0200\n\n s390/process: avoid potential reading of freed stack\n \n With THREAD_INFO_IN_TASK (which is selected on s390) task\u0027s stack usage\n is refcounted and should always be protected by get/put when touching\n other task\u0027s stack to avoid race conditions with task\u0027s destruction code.\n \n Fixes: d5c352cdd022 (\"s390: move thread_info into task_struct\")\n Cc: stable@vger.kernel.org # v4.10+\n Acked-by: Ilya Leoshkevich \u003ciii@linux.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit f45f7b5bdaa4828ce871cf03f7c01599a0de57a5\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Mon Aug 12 15:52:07 2019 +0200\n\n s390/kasan: provide uninstrumented __strlen\n \n s390 kasan code uses sclp_early_printk to report initialization\n failures. The code doing that should not be instrumented, because kasan\n shadow memory has not been set up yet. Even though sclp_early_core.c is\n compiled with instrumentation disabled it uses strlen function, which\n is instrumented and would produce shadow memory access if used. To\n avoid that, introduce uninstrumented __strlen function to be used\n instead.\n \n Before commit 7e0d92f00246 (\"s390/kasan: improve string/memory functions\n checks\") few string functions (including strlen) were escaping kasan\n instrumentation due to usage of platform specific versions which are\n implemented in inline assembly.\n \n Fixes: 7e0d92f00246 (\"s390/kasan: improve string/memory functions checks\")\n Acked-by: Ilya Leoshkevich \u003ciii@linux.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 2e83e0eb85ca62985406920f97ece36d822d421f\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Sun Aug 11 20:55:18 2019 +0200\n\n s390: clean .bss before running uncompressed kernel\n \n Clean uncompressed kernel .bss section in the startup code before\n the uncompressed kernel is executed. At this point of time initrd and\n certificates have been already rescued. Uncompressed kernel .bss size\n is known from vmlinux_info. It is also taken into consideration during\n uncompressed kernel positioning by kaslr (so it is safe to clean it).\n \n With that uncompressed kernel is starting with .bss section zeroed and\n no .bss section usage restrictions apply. Which makes chkbss checks for\n uncompressed kernel objects obsolete and they can be removed.\n \n early_nobss.c is also not needed anymore. Parts of it which are still\n relevant are moved to early.c. Kasan initialization code is now called\n directly from head64 (early.c is instrumented and should not be\n executed before kasan shadow memory is set up).\n \n Reviewed-by: Philipp Rudo \u003cprudo@linux.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 5ae6393e6d41c7cc38c1ce679bc91c475412c624\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Aug 19 13:09:46 2019 +0530\n\n PCI: kirin: Make structure kirin_dw_pcie_ops constant\n \n Static variable kirin_dw_pcie_ops, of type dw_pcie_ops, is used only\n once, when it is assigned to the constant field ops of variable pci\n (having type dw_pcie) so kirin_dw_pcie_ops is never modified.\n \n Make it constant to protect it from unintended modification.\n \n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Reviewed-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n\ncommit 19413fe04f49b32c63b470079afef8d1c7995cc1\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Sun Aug 11 22:04:54 2019 +0200\n\n s390/startup: purge obsolete .gitignore patterns\n \n sizes.h and vmlinux.scr.lds are not generated since commit 369f91c37451\n (\"s390/decompressor: rework uncompressed image info collection\").\n \n vmlinux.bin.full is not generated since commit 183ab05ff285 (\"s390: get\n rid of the first mb of uncompressed image\").\n \n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit da9ed30d29c48233cec4a19e2bd794bcaa230341\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Thu Aug 8 13:50:02 2019 +0200\n\n s390/startup: add initial pgm check handler\n \n The startup code is getting more complicated with features like kaslr and\n secure boot in place. In a potential unexpected startup code crash case\n the system would end up in a pgm check loop at address 0, overwriting\n pgm check old psw value and just making debugging more complicated. To\n avoid that introduce startup program check handler which is active\n immediately after kernel start and until early_pgm_check_handler is set\n in kernel/early.c. So it covers kernel relocation phase and transition\n to it. This pgm check handler simply saves general/control registers and\n psw in the save area which should guarantee that we still have something\n to look at when standalone dumper is called without saving registers. And\n it does disabled wait with a faulty address in the end.\n \n Acked-by: Ilya Leoshkevich \u003ciii@linux.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 40e90656c13587c0826e8afeef3cebe585db2d2c\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Thu Aug 8 15:18:26 2019 +0800\n\n s390/mm: use refcount_t for refcount\n \n Reference counters are preferred to use refcount_t instead of\n atomic_t.\n This is because the implementation of refcount_t can prevent\n overflows and detect possible use-after-free.\n So convert atomic_t ref counters to refcount_t.\n \n Link: http://lkml.kernel.org/r/20190808071826.6649-1-hslester96@gmail.com\n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Reviewed-by: David Hildenbrand \u003cdavid@redhat.com\u003e\n Reviewed-by: Cornelia Huck \u003ccohuck@redhat.com\u003e\n Signed-off-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 3434caec5b70519ead87e7b66dbdc18e8dbaa506\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Thu Aug 8 15:18:17 2019 +0800\n\n s390/extmem: use refcount_t for refcount\n \n Reference counters are preferred to use refcount_t instead of\n atomic_t.\n This is because the implementation of refcount_t can prevent\n overflows and detect possible use-after-free.\n So convert atomic_t ref counters to refcount_t.\n \n Link: http://lkml.kernel.org/r/20190808071817.6595-1-hslester96@gmail.com\n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit c4c3772384bc93e3d1eccd88bd7f5d74de74e0d7\nAuthor: Denis Efremov \u003cefremov@linux.com\u003e\nDate: Tue Aug 6 19:01:37 2019 +0300\n\n s390/pci: PCI_IOV_RESOURCES loop refactoring in zpci_map_resources\n \n This patch alters the for loop iteration scheme in zpci_map_resources\n to make it more usual. Thus, the patch generalizes the style for\n PCI_IOV_RESOURCES iteration and improves readability.\n \n Link: http://lkml.kernel.org/r/20190806160137.29275-1-efremov@linux.com\n Signed-off-by: Denis Efremov \u003cefremov@linux.com\u003e\n Signed-off-by: Sebastian Ott \u003csebott@linux.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 59793c5ab93fed9661deaf6d8ba054af681c0a6a\nAuthor: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nDate: Fri Aug 2 12:28:20 2019 +0200\n\n s390: move vmalloc option parsing to startup code\n \n Few other crucial memory setup options are already handled in\n the startup code. Those values are needed by kaslr and kasan\n implementations. \"vmalloc\" is the last piece required for future\n improvements such as early decision on kernel page levels depth required\n for actual memory setup, as well as vmalloc memory area access monitoring\n in kasan.\n \n Reviewed-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit f62f7dcbf023160ca47eb4bc7228ece8207f8a8e\nAuthor: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nDate: Thu Jul 25 09:23:39 2019 +0200\n\n Documentation/s390: remove outdated debugging390 documentation\n \n This file would need a lot of work to make sense again. Thomas Huth\n started working on that four years ago, but that wasn\u0027t finished.\n Therefore remove this.\n \n Signed-off-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 8c72e0c85212df4e7c77fca55556e423fe17e801\nAuthor: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nDate: Thu Jul 25 09:21:47 2019 +0200\n\n Documentation/s390: remove outdated dasd documentation\n \n The contents of the file is completely outdated - just remove it.\n \n Signed-off-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 783877458f1fb5c3ac05d9c89f9506337c2a6f98\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Aug 21 10:39:05 2019 +0100\n\n drm/i915/gtt: Include asm/smp.h\n \n We need asm/smp.h for wbinvd_on_all_cpus()\n \n Reported-by: kbuild-all@01.org\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190821093905.7693-1-chris@chris-wilson.co.uk\n\ncommit e63eacf76852cf1631bbb1325e125d5e44c7df12\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Tue Aug 20 16:40:19 2019 +0300\n\n drm/i915/hdmi: make hdcp2_msg_data const\n \n It\u0027s static const data, make it so.\n \n Cc: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Reviewed-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820134019.13229-5-jani.nikula@intel.com\n\ncommit 032048db141065bed54d627eee019216f110f531\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Tue Aug 20 16:40:18 2019 +0300\n\n drm/i915/hdmi: stylistic cleanup around hdcp2_msg_data\n \n Split struct declaration and array definition. Fix indents and\n whitespace. No functional changes.\n \n Cc: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Reviewed-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820134019.13229-4-jani.nikula@intel.com\n\ncommit e8465e1c6e26b3df03fd53a43dce5f58eaac17e8\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Tue Aug 20 16:40:17 2019 +0300\n\n drm/i915/dp: make hdcp2_dp_msg_data const\n \n It\u0027s static const data, make it so.\n \n Cc: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Reviewed-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820134019.13229-3-jani.nikula@intel.com\n\ncommit 3be3a877cc22ace214e994e347aac29c8bdf7f24\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Tue Aug 20 16:40:16 2019 +0300\n\n drm/i915/dp: avoid shadowing variables\n \n Everything seems to be all right, but shadowing is to be avoided.\n \n Cc: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Reviewed-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820134019.13229-2-jani.nikula@intel.com\n\ncommit 57bf7f43ba027d72740a12c13e1b7bc56429afbb\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Tue Aug 20 16:40:15 2019 +0300\n\n drm/i915/dp: stylistic cleanup around hdcp2_msg_data\n \n Split struct declaration and array definition. Fix indents and\n whitespace. No functional changes.\n \n Cc: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Reviewed-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820134019.13229-1-jani.nikula@intel.com\n\ncommit 19d17d935b29f30510874fd34e93357eea0daba2\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 15 11:36:09 2019 +0530\n\n pinctrl: st: Add of_node_put() before return\n \n Each iteration of for_each_child_of_node puts the previous node, but in\n the case of a return from the middle of the loop, there is no put, thus\n causing a memory leak. Hence add an of_node_put before the return in\n three places.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190815060609.3056-1-nishkadg.linux@gmail.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit a5f6679fc81e42fcbef0184770d8a3b04c0f153e\nAuthor: Avi Fishman \u003cavifishman70@gmail.com\u003e\nDate: Mon Jul 29 20:03:54 2019 +0300\n\n clocksource/drivers/npcm: Fix GENMASK and timer operation\n \n NPCM7XX_Tx_OPER GENMASK bits are wrong, fix them.\n \n Hopefully the NPCM7XX_REG_TICR0 register reset value of those bits was 0,\n so it did not cause an issue.\n \n The function npcm7xx_timer_oneshot() reads the register\n NPCM7XX_REG_TCSR0, modifies it and then reads it again overwriting the\n previous changes. Remove the extra read which is pointless.\n \n The function npcm7xx_timer_periodic() is correct but the code writes\n to the NPCM7XX_REG_TICR0 register while it is dealing with the\n NPCM7XX_REG_TCSR0 register, that is confusing. Separate the write to\n the registers in the code for the sake of clarity.\n \n Fixes: 1c00289ecd12 (\"clocksource/drivers/npcm: Add NPCM7xx timer driver\")\n Signed-off-by: Avi Fishman \u003cavifishman70@gmail.com\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 65818ad0815f3a2ba6a41327cce8b600ee04be32\nAuthor: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\nDate: Wed Aug 14 08:08:48 2019 +0200\n\n clk: sunxi-ng: h6: Allow I2S to change parent rate\n \n I2S doesn\u0027t work if parent rate couldn\u0027t be change. Difference between\n wanted and actual rate is too big.\n \n Fix this by adding CLK_SET_RATE_PARENT flag to I2S clocks.\n \n Signed-off-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Signed-off-by: Marcus Cooper \u003ccodekipper@gmail.com\u003e\n Signed-off-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\n\ncommit e9670ccb39dbcfc934ae9533163e81a769d2cb82\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Mon Jun 3 09:37:19 2019 +0200\n\n ARM: debug-ll: Add support for r7s9210\n \n Enable low-level debugging support for RZ/A2M (r7s9210).\n \n The RZA2MEVB board uses either SCIF2 (SDRAM enabled) or SCIF4 (HyperRAM\n only) for the serial console.\n \n Note that \"SCIFA\" serial ports on RZ/A2 SoCs use a compressed register\n layout, hence add support for that to renesas-scif.S.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n\ncommit 48cb39522a9d4d4680865e40a88f975a1cee6abc\nAuthor: Felix Fietkau \u003cnbd@nbd.name\u003e\nDate: Tue Aug 20 11:54:49 2019 +0200\n\n mac80211: minstrel_ht: improve rate probing for devices with static fallback\n \n On some devices that only support static rate fallback tables sending rate\n control probing packets can be really expensive.\n Probing lower rates can already hurt throughput quite a bit. What hurts even\n more is the fact that on mt76x0/mt76x2, single probing packets can only be\n forced by directing packets at a different internal hardware queue, which\n causes some heavy reordering and extra latency.\n The reordering issue is mainly problematic while pushing lots of packets to\n a particular station. If there is little activity, the overhead of probing is\n neglegible.\n \n The static fallback behavior is designed to pretty much only handle rate\n control algorithms that use only a very limited set of rates on which the\n algorithm switches up/down based on packet error rate.\n \n In order to better support that kind of hardware, this patch implements a\n different approach to rate probing where it switches to a slightly higher rate,\n waits for tx status feedback, then updates the stats and switches back to\n the new max throughput rate. This only triggers above a packet rate of 100\n per stats interval (~50ms).\n For that kind of probing, the code has to reduce the set of probing rates\n a lot more compared to single packet probing, so it uses only one packet\n per MCS group which is either slightly faster, or as close as possible to\n the max throughput rate.\n This allows switching between similar rates with different numbers of\n streams. The algorithm assumes that the hardware will work its way lower\n within an MCS group in case of retransmissions, so that lower rates don\u0027t\n have to be probed by the high packets per second rate probing code.\n \n To further reduce the search space, it also does not probe rates with lower\n channel bandwidth than the max throughput rate.\n \n At the moment, these changes will only affect mt76x0/mt76x2.\n \n Signed-off-by: Felix Fietkau \u003cnbd@nbd.name\u003e\n Link: https://lore.kernel.org/r/20190820095449.45255-4-nbd@nbd.name\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 21f7981b4bd904871c6bbd67333cf0f69ff7c06a\nAuthor: Felix Fietkau \u003cnbd@nbd.name\u003e\nDate: Tue Aug 20 11:54:48 2019 +0200\n\n mac80211: minstrel_ht: fix default max throughput rate indexes\n \n Use the first supported rate instead of 0 (which can be invalid)\n \n Signed-off-by: Felix Fietkau \u003cnbd@nbd.name\u003e\n Link: https://lore.kernel.org/r/20190820095449.45255-3-nbd@nbd.name\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit f793c7eedd94fa41e8b969b18c4907335065a376\nAuthor: Felix Fietkau \u003cnbd@nbd.name\u003e\nDate: Tue Aug 20 11:54:47 2019 +0200\n\n mac80211: minstrel_ht: reduce unnecessary rate probing attempts\n \n On hardware with static fallback tables (e.g. mt76x2), rate probing attempts\n can be very expensive.\n On such devices, avoid sampling rates slower than the per-group max throughput\n rate, based on the assumption that the fallback table will take care of probing\n lower rates within that group if the higher rates fail.\n \n To further reduce unnecessary probing attempts, skip duplicate attempts on\n rates slower than the max throughput rate.\n \n Signed-off-by: Felix Fietkau \u003cnbd@nbd.name\u003e\n Link: https://lore.kernel.org/r/20190820095449.45255-2-nbd@nbd.name\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 56dd918ff06e3ee24d8067e93ed12b2a39e71394\nAuthor: Felix Fietkau \u003cnbd@nbd.name\u003e\nDate: Tue Aug 20 11:54:46 2019 +0200\n\n mac80211: minstrel_ht: fix per-group max throughput rate initialization\n \n The group number needs to be multiplied by the number of rates per group\n to get the full rate index\n \n Fixes: 5935839ad735 (\"mac80211: improve minstrel_ht rate sorting by throughput \u0026 probability\")\n Signed-off-by: Felix Fietkau \u003cnbd@nbd.name\u003e\n Link: https://lore.kernel.org/r/20190820095449.45255-1-nbd@nbd.name\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 2a38075cd0beefa4da326380cf54c7b365ddc035\nAuthor: Alexei Avshalom Lazar \u003cailizaro@codeaurora.org\u003e\nDate: Sun Aug 18 17:35:17 2019 +0300\n\n nl80211: Add support for EDMG channels\n \n 802.11ay specification defines Enhanced Directional Multi-Gigabit\n (EDMG) STA and AP which allow channel bonding of 2 channels and more.\n \n Introduce new NL attributes that are needed for enabling and\n configuring EDMG support.\n \n Two new attributes are used by kernel to publish driver\u0027s EDMG\n capabilities to the userspace:\n NL80211_BAND_ATTR_EDMG_CHANNELS - bitmap field that indicates the 2.16\n GHz channel(s) that are supported by the driver.\n When this attribute is not set it means driver does not support EDMG.\n NL80211_BAND_ATTR_EDMG_BW_CONFIG - represent the channel bandwidth\n configurations supported by the driver.\n \n Additional two new attributes are used by the userspace for connect\n command and for AP configuration:\n NL80211_ATTR_WIPHY_EDMG_CHANNELS\n NL80211_ATTR_WIPHY_EDMG_BW_CONFIG\n \n New rate info flag - RATE_INFO_FLAGS_EDMG, can be reported from driver\n and used for bitrate calculation that will take into account EDMG\n according to the 802.11ay specification.\n \n Signed-off-by: Alexei Avshalom Lazar \u003cailizaro@codeaurora.org\u003e\n Link: https://lore.kernel.org/r/1566138918-3823-2-git-send-email-ailizaro@codeaurora.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 183c7687802e4132eb782808a8bf80689a9219c1\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Aug 5 19:55:22 2019 -0500\n\n soundwire: intel: move shutdown() callback and don\u0027t export symbol\n \n All DAI callbacks are in intel.c except for shutdown. Move and remove\n export symbol\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806005522.22642-18-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 04592dced55a3c55eb29bd49f4fcd3e1b2ce96d6\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Aug 5 19:55:21 2019 -0500\n\n soundwire: cadence_master: add kernel parameter to override interrupt mask\n \n The code has a set of defaults which may not be relevant in all cases,\n add kernel parameter as a helper - mostly for early board bring-up.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806005522.22642-17-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 50302fc7b936c1b00e39d4a4f118fb43d3f5ef2c\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Aug 5 19:55:20 2019 -0500\n\n soundwire: intel_init: add kernel module parameter to filter out links\n \n The hardware and ACPI info may report the presence of links that are\n not physically enabled (e.g. due to pin-muxing or hardware reworks),\n which in turn can result in errors being thrown. This shouldn\u0027t be the\n case for production devices but will happen a lot on development\n devices - even more so when they expose a connector.\n \n Even when the ACPI information is correct, it\u0027s useful to be able to\n only enable the links that need attention - mostly to filter out\n dynamic debug messages.\n \n Add a module parameter to filter out such links, e.g. adding the\n following config to a file in /etc/modprobe.d will select the second\n and third links only.\n \n options soundwire_intel_init sdw_link_mask\u003d0x6\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806005522.22642-16-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit a50954e211d9ad4f01de73c5bfcc340903b085f1\nAuthor: Rander Wang \u003crander.wang@linux.intel.com\u003e\nDate: Mon Aug 5 19:55:18 2019 -0500\n\n soundwire: cadence_master: fix divider setting in clock register\n \n The existing code uses an OR operation which would mix the original\n divider setting with the new one, resulting in an invalid\n configuration that can make codecs hang.\n \n Add the mask definition and use cdns_updatel to update divider\n \n Signed-off-by: Rander Wang \u003crander.wang@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806005522.22642-14-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 3859872f47c4b7d7a767df9e60ca4b44bbbcb0e2\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Aug 5 19:55:17 2019 -0500\n\n soundwire: cadence_master: make use of mclk_freq property\n \n Now that the prototype and Intel implementation are enabled, use this\n property to avoid hard-coded values.\n \n For example for ICL the mclk_freq value is 38.4 MHz while on CNL/CML\n it\u0027s 24 MHz. The mclk_freq should not be confused with the\n max_clk_freq, which si the maximum bus clock. The mclk_freq is\n typically tied to the oscillator frequency and does not change between\n platforms. The max_clk_freq value is linked to the maximum bandwidth\n needed and topology/trace length.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806005522.22642-13-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 085f4ace103dc81163d5fa6298b94e87c7ec2082\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Aug 5 19:55:16 2019 -0500\n\n soundwire: intel: read mclk_freq property from firmware\n \n The BIOS provides an Intel-specific property, let\u0027s use it to avoid\n hard-coded clock dividers.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806005522.22642-12-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 82fc8d06501a0694bfb1ada8ba69372cbc297fb9\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Aug 5 19:55:15 2019 -0500\n\n soundwire: add new mclk_freq field for properties\n \n To help pass platform-specific values, add a new field that can either\n be set by the Master driver or read from firmware (BIOS/DT).\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806005522.22642-11-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 3a0be1a65b50126b525d222bb59719eecda7ee3e\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Aug 5 19:55:14 2019 -0500\n\n soundwire: stream: remove unnecessary variable initializations\n \n A number of variables don\u0027t need to be initialized.\n \n In a couple of cases where we loop on a list of runtimes, the code\n handling of the \u0027bus\u0027 variable leads to warnings that it may not be\n initialized. Add a specific error case to trap such cases.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806005522.22642-10-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit e0279b6b5522f86680a9e16a12dbb2e40b1e2698\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Aug 5 19:55:13 2019 -0500\n\n soundwire: stream: fix disable sequence\n \n When we disable the stream and then call hw_free, two bank switches\n will be handled and as a result we re-enable the stream on hw_free.\n \n Make sure the stream is disabled on both banks.\n \n TODO: we need to completely revisit all this and make sure we have a\n mirroring mechanism between current and alternate banks.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806005522.22642-9-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit ce3304d8da8fa8e20001ed6128c7d04f703be305\nAuthor: Bard liao \u003cyung-chuan.liao@linux.intel.com\u003e\nDate: Mon Aug 5 19:55:12 2019 -0500\n\n soundwire: include mod_devicetable.h to avoid compiling warnings\n \n When integrating SoundWire, kbuild throws this warning with randconfig:\n \n \u003e\u003e include/linux/soundwire/sdw.h:571:17: warning: \u0027struct\n sdw_device_id\u0027 declared inside parameter list will not be visible\n outside of this definition or declaration\n \n const struct sdw_device_id *id);\n ^~~~~~~~~~~~~\n \n Fix by adding the relevant include\n \n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Bard liao \u003cyung-chuan.liao@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806005522.22642-8-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 05be59ac41b2b4487170d495f20a411698290607\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Aug 5 19:55:11 2019 -0500\n\n soundwire: cadence_master: use firmware defaults for frame shape\n \n Remove hard-coding and use firmware (BIOS/DT) values. If they are\n wrong use default 48x2 frame shape.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806005522.22642-7-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit fe4b70f2ce27e961198605ac5ddc288f066f7150\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Aug 5 19:55:10 2019 -0500\n\n soundwire: export helpers to find row and column values\n \n Add a prefix for common tables and export 2 helpers to set the frame\n shapes based on row/col values.\n \n These changes simplify bandwidth allocation algorithms as well as the\n Cadence parts which all need to convert from frame shape to indices\n used by the standard. These helpers are used in the following patch.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806005522.22642-6-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 6e0ac6a663b834b9ea16b2340e66aa3717c465b2\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Aug 5 19:55:09 2019 -0500\n\n soundwire: bus: improve dynamic debug comments for enumeration\n \n update comments to provide better understanding of enumeration flows.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806005522.22642-5-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 9b5884a0c60b7194fd62d48f62f893f353f41eb4\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Aug 5 19:55:08 2019 -0500\n\n soundwire: cadence_master: revisit interrupt settings\n \n Adding missing interrupt masks (parity, etc) and missing checks.\n Clarify which masks are for which usage.\n \n Signed-off-by: Bard Liao \u003cyung-chuan.liao@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806005522.22642-4-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 18046335643de6d21327f5ae034c8fb8463f6715\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Aug 5 19:55:07 2019 -0500\n\n soundwire: intel: fix channel number reported by hardware\n \n On all released Intel controllers (CNL/CML/ICL), PDI2 reports an\n invalid count, force the correct hardware-supported value\n \n This may have to be revisited with platform-specific values if the\n hardware changes, but for now this is good enough.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806005522.22642-3-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 05c8afe42559c98f89022431e457bfbdaf01c806\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Aug 5 19:55:06 2019 -0500\n\n soundwire: intel: prevent possible dereference in hw_params\n \n This should not happen in production systems but we should test for\n all callback arguments before invoking the config_stream callback.\n \n Update the prototype to clarify that the first argument is mandatory.\n \n Also use local variable instead of multiple dereferences to improve\n readability.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806005522.22642-2-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 5eb624ebc700592d0c9e9e1b3ad59ffc6108c683\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Mon Aug 19 21:56:33 2019 +0900\n\n arm64: dts: renesas: r8a77970: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 13a398a17afed664d3539d9d2c85fcb489b494e5\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Mon Aug 19 21:56:18 2019 +0900\n\n arm64: dts: renesas: r8a7796: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 5db16ba82f38849a78ae932c0b7eada4cd2eb919\nAuthor: John Crispin \u003cjohn@phrozen.org\u003e\nDate: Tue Aug 13 09:07:12 2019 +0200\n\n mac80211: fix possible NULL pointerderef in obss pd code\n \n he_spr_ie_elem is dereferenced before the NULL check. fix this by moving\n the assignment after the check.\n \n fixes commit 697f6c507c74 (\"mac80211: propagate HE operation info into\n bss_conf\")\n \n This was reported by the static code checker.\n \n Reported-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Signed-off-by: John Crispin \u003cjohn@phrozen.org\u003e\n Link: https://lore.kernel.org/r/20190813070712.25509-1-john@phrozen.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 9cf02338880dfc5928e98e3a1200d5aacfa6d1c6\nAuthor: Ben Greear \u003cgreearb@candelatech.com\u003e\nDate: Fri Aug 9 11:00:01 2019 -0700\n\n mac80211: add assoc-at support\n \n Report timestamp for when sta becomes associated.\n \n Signed-off-by: Ben Greear \u003cgreearb@candelatech.com\u003e\n Link: https://lore.kernel.org/r/20190809180001.26393-2-greearb@candelatech.com\n [fix ktime_get_boot_ns() to ktime_get_boottime_ns(), assoc_at type to u64]\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit e376df94d3f75161fa7a0f114fd2dff190d0b6af\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Mon Aug 19 21:56:01 2019 +0900\n\n arm64: dts: renesas: r8a774c0: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 6c7a00339e2a64b068c986301f37bd31eb83d7e9\nAuthor: Ben Greear \u003cgreearb@candelatech.com\u003e\nDate: Fri Aug 9 11:00:00 2019 -0700\n\n cfg80211: Support assoc-at timer in sta-info\n \n Report timestamp of when sta became associated.\n \n This is the boottime clock, units are nano-seconds.\n \n Signed-off-by: Ben Greear \u003cgreearb@candelatech.com\u003e\n Link: https://lore.kernel.org/r/20190809180001.26393-1-greearb@candelatech.com\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 62524a5857d2356b53e6e75fdce95dfd3454ab9e\nAuthor: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\nDate: Fri Aug 2 13:31:05 2019 +0200\n\n cfg80211: apply same mandatory rate flags for 5GHz and 6GHz\n \n For the new 6GHz band the same rules apply for mandatory rates so\n add it to set_mandatory_flags_band() function.\n \n Reviewed-by: Pieter-Paul Giesberts \u003cpieter-paul.giesberts@broadcom.com\u003e\n Reviewed-by: Leon Zegers \u003cleon.zegers@broadcom.com\u003e\n Signed-off-by: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\n Link: https://lore.kernel.org/r/1564745465-21234-9-git-send-email-arend.vanspriel@broadcom.com\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 5ea4e7802c43144f5529d1b60c01853dc5c24797\nAuthor: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\nDate: Fri Aug 2 13:31:04 2019 +0200\n\n cfg80211: ibss: use 11a mandatory rates for 6GHz band operation\n \n The default mandatory rates, ie. when not specified by user-space, is\n determined by the band. Select 11a rateset for 6GHz band.\n \n Reviewed-by: Pieter-Paul Giesberts \u003cpieter-paul.giesberts@broadcom.com\u003e\n Reviewed-by: Leon Zegers \u003cleon.zegers@broadcom.com\u003e\n Signed-off-by: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\n Link: https://lore.kernel.org/r/1564745465-21234-8-git-send-email-arend.vanspriel@broadcom.com\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 0816e6b1177adb4f120767434c67441c30de10d2\nAuthor: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\nDate: Fri Aug 2 13:31:03 2019 +0200\n\n cfg80211: use same IR permissive rules for 6GHz band\n \n The function cfg80211_ir_permissive_chan() is applicable for\n 6GHz band as well so make sure it is handled.\n \n Reviewed-by: Pieter-Paul Giesberts \u003cpieter-paul.giesberts@broadcom.com\u003e\n Reviewed-by: Leon Zegers \u003cleon.zegers@broadcom.com\u003e\n Signed-off-by: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\n Link: https://lore.kernel.org/r/1564745465-21234-7-git-send-email-arend.vanspriel@broadcom.com\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit e548a1c36b11ccf56627e5a2581409e2f27a6ac4\nAuthor: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\nDate: Fri Aug 2 13:31:02 2019 +0200\n\n cfg80211: add 6GHz in code handling array with NUM_NL80211_BANDS entries\n \n In nl80211.c there is a policy for all bands in NUM_NL80211_BANDS and\n in trace.h there is a callback trace for multicast rates which is per\n band in NUM_NL80211_BANDS. Both need to be extended for the new\n NL80211_BAND_6GHZ.\n \n Reviewed-by: Pieter-Paul Giesberts \u003cpieter-paul.giesberts@broadcom.com\u003e\n Reviewed-by: Leon Zegers \u003cleon.zegers@broadcom.com\u003e\n Signed-off-by: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\n Link: https://lore.kernel.org/r/1564745465-21234-6-git-send-email-arend.vanspriel@broadcom.com\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 852f04620e5b7c27eadbf81d086d04f61431c9dc\nAuthor: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\nDate: Fri Aug 2 13:31:01 2019 +0200\n\n cfg80211: extend ieee80211_operating_class_to_band() for 6GHz\n \n Add 6GHz operating class range as defined in 802.11ax D4.1 Annex E.\n \n Reviewed-by: Pieter-Paul Giesberts \u003cpieter-paul.giesberts@broadcom.com\u003e\n Reviewed-by: Leon Zegers \u003cleon.zegers@broadcom.com\u003e\n Signed-off-by: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\n Link: https://lore.kernel.org/r/1564745465-21234-5-git-send-email-arend.vanspriel@broadcom.com\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit fa1f1085bc063da5a44f779c9b655b7026c52d68\nAuthor: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\nDate: Fri Aug 2 13:31:00 2019 +0200\n\n cfg80211: util: add 6GHz channel to freq conversion and vice versa\n \n Extend the functions ieee80211_channel_to_frequency() and\n ieee80211_frequency_to_channel() to support 6GHz band according\n specification in 802.11ax D4.1 27.3.22.2.\n \n Reviewed-by: Pieter-Paul Giesberts \u003cpieter-paul.giesberts@broadcom.com\u003e\n Reviewed-by: Leon Zegers \u003cleon.zegers@broadcom.com\u003e\n Signed-off-by: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\n Link: https://lore.kernel.org/r/1564745465-21234-4-git-send-email-arend.vanspriel@broadcom.com\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 63a0f811558b9a5587a8c52978a57a8b528b074a\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Mon Aug 19 21:55:41 2019 +0900\n\n arm64: dts: renesas: r8a774c0: cat874: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit fadbdd069376833015f56dac27f9ed3b5d5caeb4\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Mon Aug 19 21:54:55 2019 +0900\n\n arm64: dts: renesas: r8a774a1: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit f89769cfdd5a469c9d5791a06a670d424e847477\nAuthor: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\nDate: Fri Aug 2 13:30:59 2019 +0200\n\n cfg80211: add 6GHz UNII band definitions\n \n For the new 6GHz there are new UNII band definitions as listed\n in the FCC notice [1].\n \n [1] https://docs.fcc.gov/public/attachments/FCC-18-147A1_Rcd.pdf\n \n Reviewed-by: Pieter-Paul Giesberts \u003cpieter-paul.giesberts@broadcom.com\u003e\n Reviewed-by: Leon Zegers \u003cleon.zegers@broadcom.com\u003e\n Signed-off-by: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\n Link: https://lore.kernel.org/r/1564745465-21234-3-git-send-email-arend.vanspriel@broadcom.com\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit c5b9a7f826735228a38fab4a7b2707f032468c88\nAuthor: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\nDate: Fri Aug 2 13:30:58 2019 +0200\n\n nl80211: add 6GHz band definition to enum nl80211_band\n \n In the 802.11ax specification a new band is introduced, which\n is also proposed by FCC for unlicensed use. This band is referred\n to as 6GHz spanning frequency range from 5925 to 7125 MHz.\n \n Reviewed-by: Pieter-Paul Giesberts \u003cpieter-paul.giesberts@broadcom.com\u003e\n Reviewed-by: Leon Zegers \u003cleon.zegers@broadcom.com\u003e\n Signed-off-by: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\n Link: https://lore.kernel.org/r/1564745465-21234-2-git-send-email-arend.vanspriel@broadcom.com\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 3a00f08140646a54e0eff2ce8938bf248dad6153\nAuthor: John Crispin \u003cjohn@phrozen.org\u003e\nDate: Wed Aug 7 09:59:49 2019 +0200\n\n mac80211: add missing length field increment when generating Radiotap header\n \n The code generating the Tx Radiotap header when using tx_status_ext was\n missing a field increment after setting the VHT bandwidth.\n \n Fixes: 3d07ffcaf320 (\"mac80211: add struct ieee80211_tx_status support to ieee80211_add_tx_radiotap_header\")\n Signed-off-by: John Crispin \u003cjohn@phrozen.org\u003e\n Link: https://lore.kernel.org/r/20190807075949.32414-4-john@phrozen.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit a027c3041fef93da93a2c49eb22b05891d2350b6\nAuthor: John Crispin \u003cjohn@phrozen.org\u003e\nDate: Wed Aug 7 09:59:48 2019 +0200\n\n mac80211: 80Mhz was not reported properly when using tx_status_ext\n \n When reporting 80MHz, we need to set 4 and not 2 inside the corresponding\n field inside the Tx Radiotap header.\n \n Fixes: 3d07ffcaf320 (\"mac80211: add struct ieee80211_tx_status support to ieee80211_add_tx_radiotap_header\")\n Signed-off-by: John Crispin \u003cjohn@phrozen.org\u003e\n Link: https://lore.kernel.org/r/20190807075949.32414-3-john@phrozen.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 8db6e7367d3ca78dd85c5e9fec2bd6d69981a980\nAuthor: John Crispin \u003cjohn@phrozen.org\u003e\nDate: Wed Aug 7 09:59:47 2019 +0200\n\n mac80211: fix bad guard when reporting legacy rates\n \n When reporting legacy rates inside the TX Radiotap header we need to split\n the check between \"uses tx_statua_ext\" and \"is legacy rate\". Not doing so\n would make the code drop into the !tx_status_ext path.\n \n Fixes: 3d07ffcaf320 (\"mac80211: add struct ieee80211_tx_status support to ieee80211_add_tx_radiotap_header\")\n Signed-off-by: John Crispin \u003cjohn@phrozen.org\u003e\n Link: https://lore.kernel.org/r/20190807075949.32414-2-john@phrozen.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit dc649d649a5e07eaf8f4aefb8f0e35cc178dd11f\nAuthor: John Crispin \u003cjohn@phrozen.org\u003e\nDate: Wed Aug 7 09:59:46 2019 +0200\n\n mac80211: fix TX legacy rate reporting when tx_status_ext is used\n \n The RX Radiotap header length was not calculated properly when reporting\n legacy rates using tx_status_ext.\n \n Fixes: 3d07ffcaf320 (\"mac80211: add struct ieee80211_tx_status support to ieee80211_add_tx_radiotap_header\")\n Signed-off-by: John Crispin \u003cjohn@phrozen.org\u003e\n Link: https://lore.kernel.org/r/20190807075949.32414-1-john@phrozen.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 57012813f1ba82412cdcc3665b04d4353b357378\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Mon May 7 11:16:48 2018 +0200\n\n ARM: dts: r8a77470: Add PMU device node\n \n Enable support for the ARM Performance Monitor Units in the Cortex-A7\n CPU cores on RZ/G1C by adding a device node for the PMU.\n \n New Linux output:\n \n hw perfevents: enabled with armv7_cortex_a7 PMU driver, 5 counters available\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Tested-by: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\n Reviewed-by: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\n\ncommit ff1750101351a225754ca4a14b83275c766cf540\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Aug 21 05:20:44 2019 +0100\n\n drm/i915/gtt: Relax assertion for pt_used\n \n When inserting the final level PTE, we check that we are not overflowing\n the page table (checking that pt_used does not exceed the size of the\n table). However, we have to allow for every other PTE to be pinned by a\n simultaneous removal thread (as on remove we bump the pt_used counter\n before adjusting the table).\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190821042044.7354-1-chris@chris-wilson.co.uk\n\ncommit 913787ca40b9b5981bea1d1157cdd8cb23e12b54\nAuthor: Alexander Sverdlin \u003calexander.sverdlin@nokia.com\u003e\nDate: Fri Jul 12 12:14:39 2019 +0000\n\n mtd: spi-nor: intel-spi: Whitelist 4B read commands\n \n spi-nor.c issues 4B commands for some Flash chips bigger than 16Mbytes.\n Xeon(R) D-1500 documentation mentions its Integrated PCH Logic supports\n Flash chips up to 64Mbytes.\n D-1500 Integrated PCH documenation however has inconsistencies regarding\n FADDR register width and says nothing about particular commands issued\n to support 64Mbytes of Flash.\n \n Nevetheless the tests on Xeon(R) CPU D-1548 with 512Mbit Flash chips\n Macronix MX25L51245G and Micron MT25QL512A showed that erase, write and\n read operations work just fine after SPINOR_OP_READ_4B and\n SPINOR_OP_READ_FAST_4B are white-listed (currently only\n SPINOR_OP_READ_FAST_4B is used and only for Macronix).\n \n Signed-off-by: Alexander Sverdlin \u003calexander.sverdlin@nokia.com\u003e\n Reviewed-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n\ncommit 313aca5a9c781a19537e6cc882511c13eff3bdba\nAuthor: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nDate: Mon Aug 19 12:16:00 2019 -0500\n\n mtd: spi-nor: fix a memory leak bug\n \n In spi_nor_parse_4bait(), \u0027dwords\u0027 is allocated through kmalloc(). However,\n it is not deallocated in the following execution if spi_nor_read_sfdp()\n fails, leading to a memory leak. To fix this issue, free \u0027dwords\u0027 before\n returning the error.\n \n Fixes: 816873eaeec6 (\"mtd: spi-nor: parse SFDP 4-byte Address Instruction Table\")\n Signed-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n\ncommit ac3b5c464a344f005a3534d0f9fb99f6753c3a55\nAuthor: Cao Van Dong \u003ccv-dong@jinso.co.jp\u003e\nDate: Thu Apr 11 16:54:22 2019 +0900\n\n dt-bindings: can: rcar_can: document r8a77470 support\n \n Document SoC specific bindings for R-Car RZ/G1C(r8a77470) SoC.\n \n Signed-off-by: Cao Van Dong \u003ccv-dong@jinso.co.jp\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 3e9e38d918bd01068b5ccba17d69a8ae9bf56142\nAuthor: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nDate: Thu Aug 15 11:32:52 2019 +0300\n\n mtd: spi-nor: Fix an error code in spi_nor_read_raw()\n \n The problem is that if \"ret\" is negative then when we check if\n \"ret \u003e len\", that condition is going to be true because of type\n promotion. So this patch re-orders the code to check for negatives\n first and preserve those error codes.\n \n Fixes: f384b352cbf0 (\"mtd: spi-nor: parse Serial Flash Discoverable Parameters (SFDP) tables\")\n Signed-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n\ncommit 1993de62ea605c4792f9556dd2ed950aaa7deedc\nAuthor: Ulrich Hecht \u003culrich.hecht+renesas@gmail.com\u003e\nDate: Wed Dec 5 09:17:05 2018 +0800\n\n dt-bindings: can: rcar_canfd: document r8a77995 support\n \n Adds compatible strings for the R-Car CAN FD controller in the D3 SoC.\n \n Signed-off-by: Ulrich Hecht \u003culrich.hecht+renesas@gmail.com\u003e\n Acked-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 635a77d66dfc650bd853b752a7f2ce268e23958e\nAuthor: Ulrich Hecht \u003culrich.hecht+renesas@gmail.com\u003e\nDate: Wed Dec 5 09:17:04 2018 +0800\n\n dt-bindings: can: rcar_can: document r8a77995 support\n \n Adds compatible strings for the R-Car CAN controller in the D3 SoC.\n \n Signed-off-by: Ulrich Hecht \u003culrich.hecht+renesas@gmail.com\u003e\n Acked-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 3ea529753a54d8d31bddc286fe5133b044f9fb2e\nAuthor: Marek Vasut \u003cmarek.vasut@gmail.com\u003e\nDate: Sun Nov 18 18:30:56 2018 +0100\n\n dt-bindings: can: rcar_can: document r8a77990 support\n \n Document the support for rcar_can on R8A77990 SoC devices.\n Add R8A77990 to the list of SoCs which require the \"assigned-clocks\"\n and \"assigned-clock-rates\" properties.\n \n Signed-off-by: Marek Vasut \u003cmarek.vasut+renesas@gmail.com\u003e\n Cc: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit a6f26606ddd03c5eab8b2132f1bfaa768c06158f\nAuthor: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nDate: Sat Aug 17 09:56:04 2019 +0300\n\n rtc: rv3029: revert error handling patch to rv3029_eeprom_write()\n \n My error handling \"cleanup\" was totally wrong. Both the \"err\" and \"ret\"\n variables are required. The \"err\" variable holds the error codes for\n rv3029_eeprom_enter/exit() and the \"ret\" variable holds the error codes\n for if actual write fails. In my patch if the write failed, the\n function probably still returned success.\n \n Reported-by: Tom Evans \u003ctom.evans@motec.com.au\u003e\n Fixes: 97f5b0379c38 (\"rtc: rv3029: Clean up error handling in rv3029_eeprom_write()\")\n Signed-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Link: https://lore.kernel.org/r/20190817065604.GB29951@mwanda\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 80ba93639b5dfde6b7897499b69fefc733a94a59\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Sun Aug 18 23:52:32 2019 +0200\n\n rtc: ds1672: remove unnecessary check\n \n The rtc pointer is already checked earlier, it is not necessary to check it\n again.\n \n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190818215232.17320-1-alexandre.belloni@bootlin.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 44c638ce4ec6fb9d33978348ebb90847b9aa274a\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Mon Aug 19 00:00:41 2019 +0200\n\n rtc: remove superfluous error message\n \n The RTC core now has error messages in case of registration failure, there\n is no need to have other messages in the drivers.\n \n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190818220041.17833-2-alexandre.belloni@bootlin.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 87b20a08dcc265959f5c59f18603ea0487fe609b\nAuthor: Haiyang Zhang \u003chaiyangz@microsoft.com\u003e\nDate: Thu Aug 15 17:01:45 2019 +0000\n\n PCI: hv: Detect and fix Hyper-V PCI domain number collision\n \n Currently in Azure cloud, for passthrough devices, the host sets the\n device instance ID\u0027s bytes 8 - 15 to a value derived from the host HWID,\n which is the same on all devices in a VM. So, the device instance ID\u0027s\n bytes 8 and 9 provided by the host are no longer unique. This affects\n all Azure hosts since July 2018, and can cause device passthrough to VMs\n to fail because the bytes 8 and 9 are used as PCI domain number.\n Collision of domain numbers will cause the second device with the same\n domain number fail to load.\n \n In the cases of collision, we will detect and find another number that is\n not in use.\n \n Suggested-by: Michael Kelley \u003cmikelley@microsoft.com\u003e\n Signed-off-by: Haiyang Zhang \u003chaiyangz@microsoft.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Acked-by: Sasha Levin \u003csashal@kernel.org\u003e\n\ncommit be700103efd1050808db1cf00e52c3a2837bf802\nAuthor: Haiyang Zhang \u003chaiyangz@microsoft.com\u003e\nDate: Thu Aug 15 17:01:37 2019 +0000\n\n PCI: hv: Detect and fix Hyper-V PCI domain number collision\n \n Currently in Azure cloud, for passthrough devices, the host sets the\n device instance ID\u0027s bytes 8 - 15 to a value derived from the host HWID,\n which is the same on all devices in a VM. So, the device instance ID\u0027s\n bytes 8 and 9 provided by the host are no longer unique. This affects\n all Azure hosts since July 2018, and can cause device passthrough to VMs\n to fail because the bytes 8 and 9 are used as PCI domain number.\n Collision of domain numbers will cause the second device with the same\n domain number fail to load.\n \n In the cases of collision, we will detect and find another number that is\n not in use.\n \n Suggested-by: Michael Kelley \u003cmikelley@microsoft.com\u003e\n Signed-off-by: Haiyang Zhang \u003chaiyangz@microsoft.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Acked-by: Sasha Levin \u003csashal@kernel.org\u003e\n\ncommit 4262ee88f07f6771eafcc5efa1050a7df1b88cb7\nAuthor: Thor Thayer \u003cthor.thayer@linux.intel.com\u003e\nDate: Thu Aug 15 17:55:36 2019 -0500\n\n mtd: spi-nor: Fix Cadence QSPI RCU Schedule Stall\n \n The current Cadence QSPI driver sometimes caused a\n \"rcu_sched self-detected stall\" while writing large files.\n \n Stall Report:\n \u0027# mtd_debug write /dev/mtd1 0 48816464 blob.img\n [ 1815.454227] rcu: INFO: rcu_sched self-detected stall on CPU\n [ 1815.459789] rcu: 0-....: (2099 ticks this GP) idle\u003d8c6/1/0x40000002\n softirq\u003d6492/6492 fqs\u003d935\n [ 1815.468442] rcu: (t\u003d2100 jiffies g\u003d8749 q\u003d247)\n \u003csnip\u003e (abbreviated backtrace)\n [ 1815.772086] [\u003cc05a3ea0\u003e] (cqspi_exec_flash_cmd) (cqspi_read_reg)\n [ 1815.786203] [\u003cc05a5488\u003e] (cqspi_read_reg) from (read_sr)\n [ 1815.803790] [\u003cc05a0330\u003e] (read_sr) from\n (spi_nor_wait_till_ready_with_timeout)\n [ 1815.816610] [\u003cc05a182c\u003e] (spi_nor_wait_till_ready_with_timeout) from\n (spi_nor_write+0x104/0x1d0)\n [ 1815.836791] [\u003cc05a1a44\u003e] (spi_nor_write) from (part_write+0x50/0x58)\n \u003csnip\u003e\n [ 1815.997961] cadence-qspi ff809000.spi: Flash command execution timed out.\n [ 1816.004733] error -110 reading SR\n file_to_flash: write, size 0x2e8e150, n 0x2e8e150\n write(): Connection timed out\n \n This was caused by a tight loop in cqspi_wait_for_bit(). Fix by using\n readl_relaxed_poll_timeout() which sleeps 10us while polling a register.\n \n Fit onto 80 character line by truncating the bool clear parameter\n \n Fixes: 140623410536 (\"mtd: spi-nor: Add driver for Cadence Quad SPI Flash Controller\")\n Signed-off-by: Thor Thayer \u003cthor.thayer@linux.intel.com\u003e\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n\ncommit 954fc5d20454b6d1c14ea5177c72a9b7d22b8e13\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Tue Aug 20 13:11:35 2019 +0200\n\n pinctrl: st: Include the right header\n \n The ST pinctrl driver wants to provode a gpio_chip but is not\n including the header for this, fix the inclusion to use the right\n header. \u003clinux/of_gpio.h\u003e has to remain as the driver is calling\n of_get_named_gpio().\n \n Cc: Patrice Chotard \u003cpatrice.chotard@st.com\u003e\n Cc: Maxime Coquelin \u003cmaxime.coquelin@st.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190820111135.10701-1-linus.walleij@linaro.org\n\ncommit 6e28aaab07c19fb97cedea9840ef796f1a772350\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Aug 19 13:27:57 2019 +0530\n\n pinctrl: rk805: Make structures constant\n \n Static structures rk805_pinctrl_desc and rk805_gpio_chip, of types\n gpio_chip and pinctrl_desc respectively, are not used except to be\n copied into the fields of a different variable. Hence make\n rk805_pinctrl_desc and rk805_gpio_chip both constant to protect them\n from unintended modification.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190819075757.1753-1-nishkadg.linux@gmail.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit a4873accca0529b6a32e1ceb458be843cdea9ec6\nAuthor: Sowjanya Komatineni \u003cskomatineni@nvidia.com\u003e\nDate: Fri Aug 16 12:41:47 2019 -0700\n\n pinctrl: tegra: Flush pinctrl writes during resume\n \n This patch adds pinctrl register read to flush all the prior pinctrl\n writes and then adds barrier for pinctrl register read to complete\n during resume to make sure all pinctrl changes are effective.\n \n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Reviewed-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Signed-off-by: Sowjanya Komatineni \u003cskomatineni@nvidia.com\u003e\n Link: https://lore.kernel.org/r/1565984527-5272-3-git-send-email-skomatineni@nvidia.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit c2cf351eba2ff6002ce8eb178452219d2521e38e\nAuthor: Sowjanya Komatineni \u003cskomatineni@nvidia.com\u003e\nDate: Fri Aug 16 12:41:46 2019 -0700\n\n pinctrl: tegra: Fix write barrier placement in pmx_writel\n \n pmx_writel uses writel which inserts write barrier before the\n register write.\n \n This patch has fix to replace writel with writel_relaxed followed\n by a readback and memory barrier to ensure write operation is\n completed for successful pinctrl change.\n \n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Reviewed-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Signed-off-by: Sowjanya Komatineni \u003cskomatineni@nvidia.com\u003e\n Link: https://lore.kernel.org/r/1565984527-5272-2-git-send-email-skomatineni@nvidia.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 6f265e5d4da7a7de326ced5a7ef8f982dfd44390\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Mon Aug 12 15:25:54 2019 +0200\n\n pinctrl: bcm-iproc: Pass irqchip when adding gpiochip\n \n We need to convert all old gpio irqchips to pass the irqchip\n setup along when adding the gpio_chip. For more info see\n drivers/gpio/TODO.\n \n For chained irqchips this is a pretty straight-forward\n conversion.\n \n Cc: Pramod Kumar \u003cpramodku@broadcom.com\u003e\n Cc: Ray Jui \u003crjui@broadcom.com\u003e\n Cc: Scott Branden \u003csbranden@broadcom.com\u003e\n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190812132554.18313-1-linus.walleij@linaro.org\n Acked-by: Scott Branden \u003cscott.branden@broadcom.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 799e064cc79e7a9e6be2d4721c882d4fb92aad65\nAuthor: Eddie James \u003ceajames@linux.ibm.com\u003e\nDate: Thu Aug 15 14:11:03 2019 -0500\n\n fsi: scom: Don\u0027t abort operations for minor errors\n \n The scom driver currently fails out of operations if certain system\n errors are flagged in the status register; system checkstop, special\n attention, or recoverable error. These errors won\u0027t impact the ability\n of the scom engine to perform operations, so the driver should continue\n under these conditions.\n Also, don\u0027t do a PIB reset for these conditions, since it won\u0027t help.\n \n Fixes: 6b293258cded (\"fsi: scom: Major overhaul\")\n Signed-off-by: Eddie James \u003ceajames@linux.ibm.com\u003e\n Acked-by: Jeremy Kerr \u003cjk@ozlabs.org\u003e\n Acked-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n Signed-off-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n\ncommit 150a6a931a328fa043522bce86db1501f2f1cb6f\nAuthor: Andrew Jeffery \u003candrew@aj.id.au\u003e\nDate: Fri Jul 12 13:07:26 2019 +0930\n\n ARM: dts: aspeed: Enable first MMC slot on AST2500 EVB\n \n The EVB contains two slots. Enable one of them for testing purposes.\n \n Signed-off-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Signed-off-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n\ncommit f5e84eae7956c694d27ddaba7113fe7d1174eff7\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Aug 20 16:15:46 2019 +0300\n\n dmaengine: dw: platform: Split OF helpers to separate module\n \n For better maintenance split OF helpers to the separate module.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190820131546.75744-11-andriy.shevchenko@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit b685fe26e9af7318d73bf50af659a282d188a3e5\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Aug 20 16:15:45 2019 +0300\n\n dmaengine: dw: platform: Split ACPI helpers to separate module\n \n For better maintenance split ACPI helpers to the separate module.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190820131546.75744-10-andriy.shevchenko@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 84da042e702347c186b84d9cc2e766f83e1814d6\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Aug 20 16:15:44 2019 +0300\n\n dmaengine: dw: platform: Move handle check to dw_dma_acpi_controller_register()\n \n Move ACPI handle check to the dw_dma_acpi_controller_register().\n \n While here, convert it to has_acpi_companion() which is recommended way.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190820131546.75744-9-andriy.shevchenko@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit e7b8514e4d68bec21fc6385fa0a66797ddc34ac9\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Aug 20 16:15:43 2019 +0300\n\n dmaengine: dw: platform: Switch to acpi_dma_controller_register()\n \n There is a possibility to have registered ACPI DMA controller\n while it has been gone already.\n \n To avoid the potential crash, move to non-managed\n acpi_dma_controller_register().\n \n Fixes: 42c91ee71d6d (\"dw_dmac: add ACPI support\")\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190820131546.75744-8-andriy.shevchenko@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit a9c56721d6ae99b22e983d0722e6b1b53a11dd59\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Aug 20 16:15:42 2019 +0300\n\n dmaengine: dw: platform: Use devm_platform_ioremap_resource()\n \n Use the new helper that wraps the calls to platform_get_resource()\n and devm_ioremap_resource() together.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190820131546.75744-7-andriy.shevchenko@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit f8d9ddbc28518a0f7d39d2ce51f778759b49f73e\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Aug 20 16:15:41 2019 +0300\n\n dmaengine: dw: platform: Enable iDMA 32-bit on Intel Elkhart Lake\n \n Intel® PSE (Programmable Services Engine) provides few DMA controllers\n to the host on Intel Elkhart Lake. Enable them in the ACPI glue driver.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190820131546.75744-6-andriy.shevchenko@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit b3757413b91ecf4a227c5f075446afe8126d882f\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Aug 20 16:15:40 2019 +0300\n\n dmaengine: dw: platform: Use struct dw_dma_chip_pdata\n \n Now, when we have a generic structure for the chip and platform data,\n use it in the platform glue driver.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190820131546.75744-5-andriy.shevchenko@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit ae923c91aa3b4f7e20b764744188c0d859243d81\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Aug 20 16:15:39 2019 +0300\n\n dmaengine: dw: Export struct dw_dma_chip_pdata for wider use\n \n We are expecting some devices can be enumerated either as PCI or ACPI.\n Nevertheless, they will share same information, thus, provide a generic\n struct dw_dma_chip_pdata for all glue drivers.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190820131546.75744-4-andriy.shevchenko@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 4b8584bac0407367f8157244c8952ed80bdb12fb\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Aug 20 16:15:38 2019 +0300\n\n dmaengine: acpi: Add kernel doc parameter descriptions\n \n Kernel documentation script is not happy about absence of function parameter\n descriptions:\n \n drivers/dma/acpi-dma.c:163: warning: Function parameter or member \u0027data\u0027 not described in \u0027acpi_dma_controller_register\u0027\n drivers/dma/acpi-dma.c:247: warning: Function parameter or member \u0027data\u0027 not described in \u0027devm_acpi_dma_controller_register\u0027\n drivers/dma/acpi-dma.c:274: warning: Function parameter or member \u0027dev\u0027 not described in \u0027devm_acpi_dma_controller_free\u0027\n \n Append the descriptions of above mentioned function parameters.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190820131546.75744-3-andriy.shevchenko@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit a6bc332373e5176c2d085657bc643a28950b20b0\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Aug 20 16:15:37 2019 +0300\n\n dmaengine: acpi: Set up DMA mask based on CSRT\n \n CSRT has an information about address width, which is supported by\n the certain DMA controller.\n \n Use information from CSRT to set up DMA mask for shared controller.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190820131546.75744-2-andriy.shevchenko@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit c35227958d7ecceae42a7d958380eae906f7b8e6\nAuthor: Andrew Jeffery \u003candrew@aj.id.au\u003e\nDate: Fri Jul 12 13:07:25 2019 +0930\n\n ARM: dts: aspeed: Describe SD controllers\n \n The AST2400 and AST2500 both share the same SD controller, at the same\n location in the physical address space and the same hardware interrupt,\n with the same clock configurations.\n \n Signed-off-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Signed-off-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n\ncommit 138e1a0990e80db486ab9f6c06bd5c01f9a97999\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Mon Aug 19 18:34:23 2019 +0800\n\n staging: erofs: add two missing erofs_workgroup_put for corrupted images\n \n As reported by erofs-utils fuzzer, these error handling\n path will be entered to handle corrupted images.\n \n Lack of erofs_workgroup_puts will cause unmounting\n unsuccessfully.\n \n Fix these return values to EFSCORRUPTED as well.\n \n Fixes: 3883a79abd02 (\"staging: erofs: introduce VLE decompression support\")\n Cc: \u003cstable@vger.kernel.org\u003e # 4.19+\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190819103426.87579-4-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 3407a4198faf01c9d7596c45b8606834b8dfc2b7\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Mon Aug 19 18:34:22 2019 +0800\n\n staging: erofs: cannot set EROFS_V_Z_INITED_BIT if fill_inode_lazy fails\n \n As reported by erofs-utils fuzzer, unsupported compressed\n clustersize will make fill_inode_lazy fail, for such case\n we cannot set EROFS_V_Z_INITED_BIT since we need return\n failure for each z_erofs_map_blocks_iter().\n \n Fixes: 152a333a5895 (\"staging: erofs: add compacted compression indexes support\")\n Cc: \u003cstable@vger.kernel.org\u003e # 5.3+\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190819103426.87579-3-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit ee45197c807895e156b2be0abcaebdfc116487c8\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Mon Aug 19 18:34:21 2019 +0800\n\n staging: erofs: some compressed cluster should be submitted for corrupted images\n \n As reported by erofs_utils fuzzer, a logical page can belong\n to at most 2 compressed clusters, if one compressed cluster\n is corrupted, but the other has been ready in submitting chain.\n \n The chain needs to submit anyway in order to keep the page\n working properly (page unlocked with PG_error set, PG_uptodate\n not set).\n \n Let\u0027s fix it now.\n \n Fixes: 3883a79abd02 (\"staging: erofs: introduce VLE decompression support\")\n Cc: \u003cstable@vger.kernel.org\u003e # 4.19+\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190819103426.87579-2-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit acb383f1dcb4f1e79b66d4be3a0b6f519a957b0d\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Sun Aug 18 20:54:57 2019 +0800\n\n staging: erofs: fix an error handling in erofs_readdir()\n \n Richard observed a forever loop of erofs_read_raw_page() [1]\n which can be generated by forcely setting -\u003eu.i_blkaddr\n to 0xdeadbeef (as my understanding block layer can\n handle access beyond end of device correctly).\n \n After digging into that, it seems the problem is highly\n related with directories and then I found the root cause\n is an improper error handling in erofs_readdir().\n \n Let\u0027s fix it now.\n \n [1] https://lore.kernel.org/r/1163995781.68824.1566084358245.JavaMail.zimbra@nod.at/\n \n Reported-by: Richard Weinberger \u003crichard@nod.at\u003e\n Fixes: 3aa8ec716e52 (\"staging: erofs: add directory operations\")\n Cc: \u003cstable@vger.kernel.org\u003e # 4.19+\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190818125457.25906-1-hsiangkao@aol.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a64d9493f587f8ec101d92ec19c7d88d6d3b5567\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Sun Aug 18 18:28:24 2019 +0800\n\n staging: erofs: refuse to mount images with malformed volume name\n \n As Richard reminder [1], A valid volume name should be\n ended in NIL terminator within the length of volume_name.\n \n Since this field currently isn\u0027t really used, let\u0027s fix\n it to avoid potential bugs in the future.\n \n [1] https://lore.kernel.org/r/1133002215.69049.1566119033047.JavaMail.zimbra@nod.at/\n \n Reported-by: Richard Weinberger \u003crichard@nod.at\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190818102824.22330-1-hsiangkao@aol.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit c3082a674f46fe49383b157882c41dfabaa37113\nAuthor: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\nDate: Thu Jul 11 19:51:25 2019 +0530\n\n PM: QoS: Get rid of unused flags\n \n The network_latency and network_throughput flags for PM-QoS have not\n found much use in drivers or in userspace since they were introduced.\n \n Commit 4a733ef1bea7 (\"mac80211: remove PM-QoS listener\") removed the\n only user PM_QOS_NETWORK_LATENCY in the kernel a while ago and there\n don\u0027t seem to be any userspace tools using the character device files\n either.\n \n PM_QOS_MEMORY_BANDWIDTH was never even added to the trace events.\n \n Remove all the flags except cpu_dma_latency.\n \n Signed-off-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit da5c504c7aae96db68c4b38e2564a88e91842d89\nAuthor: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\nDate: Fri Aug 9 07:52:49 2019 +0530\n\n cpufreq: intel_pstate: Implement QoS supported freq constraints\n \n Intel pstate driver exposes min_perf_pct and max_perf_pct sysfs files,\n which can be used to force a limit on the min/max P state of the driver.\n Though these files eventually control the min/max frequencies that the\n CPUs will run at, they don\u0027t make a change to policy-\u003emin/max values.\n \n When the values of these files are changed (in passive mode of the\n driver), it leads to calling -\u003elimits() callback of the cpufreq\n governors, like schedutil. On a call to it the governors shall\n forcefully update the frequency to come within the limits. Since the\n limits, i.e. policy-\u003emin/max, aren\u0027t updated by the driver, the\n governors fails to get the target freq within limit and sometimes aborts\n the update believing that the frequency is already set to the target\n value.\n \n This patch implements the QoS supported frequency constraints to update\n policy-\u003emin/max values whenever min_perf_pct or max_perf_pct files are\n updated. This is only done for the passive mode as of now, as the driver\n is already working fine in active mode.\n \n Fixes: ecd288429126 (\"cpufreq: schedutil: Don\u0027t set next_freq to UINT_MAX\")\n Reported-by: Doug Smythies \u003cdsmythies@telus.net\u003e\n Tested-by: Doug Smythies \u003cdsmythies@telus.net\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 1446794a89c13816bd526dcbff4051cac9743db7\nAuthor: Todd Brandt \u003ctodd.e.brandt@linux.intel.com\u003e\nDate: Mon Aug 12 14:08:44 2019 -0700\n\n pm-graph v5.5\n \n Upgrade bootgraph/sleepgraph to be able to run on python2 and python3.\n Both now simply require python, the system can choose which to use.\n \n bootgraph python3 update:\n - add floor function to handle integer arithmetic\n - change argument loop to use next() instead of args.next()\n - open dmesg log and popen in binary, use decode(ascii, ignore)\n - sort all html data to allow diff between python versions\n - change exception handler to use python3 as instead of comma\n \n sleepgraph python3 update:\n - import configparser not ConfigParser (p2 needs python-configparser)\n - add floor function to handle integer arithmetic\n - change argument loop to use next() instead of args.next()\n - handle popen output in binary, use decode(ascii, ignore)\n - sort all html/output data to allow diff between python versions\n - force gzip open to use text mode, same for file open\n - ensure no binary data is written to logs (ascii convert devprops info)\n - use codecs library to handle zlib encoding for mcelog data\n - remove all uses of python3.7 keyword \"async\" as members or vars\n - assume all FPDT and DMI data is in binary string form\n \n sleepgraph:\n - turbostat will be used by default if it\u0027s found \u0026 the mode is freeze\n - a new option \"-noturbostat\" will disable its use\n - fix bug where two callgraphs with the same start time overwrite.\n - fix s2idle processing where two suspend/resume_machines occur back2back\n - update getexec function to use which first (assuming PATH exists)\n - new platforminfo data in log with: lspci, gpe counts, /proc/interrupts\n - new data is zipped, b64 encoded, and tacked on the end of ftrace\n \n Signed-off-by: Todd Brandt \u003ctodd.e.brandt@linux.intel.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 62c23a89fd269f0308226b63ac4a427339b6f945\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Tue Aug 13 13:21:21 2019 +0100\n\n cpufreq: remove redundant assignment to ret\n \n Variable ret is initialized to a value that is never read and it is\n re-assigned later. The initialization is redundant and can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Acked-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 78c0f050847cac4c43dbfc1916a63d1557c74ac0\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Mon Aug 19 15:41:58 2019 -0700\n\n PM / wakeup: Unexport wakeup_source_sysfs_{add,remove}()\n \n These functions are just used by the PM core, and that isn\u0027t modular so\n these functions don\u0027t need to be exported. Drop the exports.\n \n Fixes: c8377adfa781 (\"PM / wakeup: Show wakeup sources stats in sysfs\")\n Reviewed-by: Tri Vo \u003ctrong@android.com\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 2ca3d1ecb8c432ee212d80fa7615cdd5d1df62e3\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Mon Aug 19 15:41:57 2019 -0700\n\n PM / wakeup: Register wakeup class kobj after device is added\n \n The device_set_wakeup_enable() function can be called on a device that\n hasn\u0027t been registered with device_add() yet. This allows the device to\n be in a state where wakeup is enabled for it but the device isn\u0027t\n published to userspace in sysfs yet.\n \n After commit c8377adfa781 (\"PM / wakeup: Show wakeup sources stats in\n sysfs\"), calling device_set_wakeup_enable() will fail for a device that\n hasn\u0027t been registered with the driver core via device_add(). This is\n because we try to create sysfs entries for the device and associate a\n wakeup class kobject with it before the device has been registered.\n Let\u0027s follow a similar approach that device_set_wakeup_capable() takes\n here and register the wakeup class either from\n device_set_wakeup_enable() when the device is already registered, or\n from dpm_sysfs_add() when the device is being registered with the driver\n core via device_add().\n \n Fixes: c8377adfa781 (\"PM / wakeup: Show wakeup sources stats in sysfs\")\n Reported-by: Qian Cai \u003ccai@lca.pw\u003e\n Reviewed-by: Tri Vo \u003ctrong@android.com\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit ae367b7936408444afc76a8a3e141abede9ccbe4\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Aug 13 17:40:53 2019 -0700\n\n PM / wakeup: Fix sysfs registration error path\n \n We shouldn\u0027t call wakeup_source_destroy() from the error path in\n wakeup_source_register() because that calls __pm_relax() which takes a\n lock that isn\u0027t initialized until wakeup_source_add() is called. Add a\n new function, wakeup_source_free(), that just does the bare minimum to\n free a wakeup source that was created but hasn\u0027t been added yet and use\n it from the two places it\u0027s needed. This fixes the following problem\n seen on various x86 server boxes:\n \n INFO: trying to register non-static key.\n the code is fine but needs lockdep annotation.\n turning off the locking correctness validator.\n CPU: 12 PID: 1 Comm: swapper/0 Not tainted 5.3.0-rc4-\n Hardware name: HP ProLiant XL420 Gen9/ProLiant XL420 Gen9, BIOS U19 12/27/2015\n Call Trace:\n   dump_stack+0x62/0x9a\n   register_lock_class+0x95a/0x960\n   ? __platform_driver_probe+0xcd/0x230\n   ? __platform_create_bundle+0xc0/0xe0\n   ? i8042_init+0x4ec/0x578\n   ? do_one_initcall+0xfe/0x45a\n   ? kernel_init_freeable+0x614/0x6a7\n   ? kernel_init+0x11/0x138\n   ? ret_from_fork+0x35/0x40\n   ? is_dynamic_key+0xf0/0xf0\n   ? rwlock_bug.part.0+0x60/0x60\n   ? __debug_check_no_obj_freed+0x8e/0x250\n   __lock_acquire.isra.13+0x5f/0x830\n   ? __debug_check_no_obj_freed+0x152/0x250\n   lock_acquire+0x107/0x220\n   ? __pm_relax.part.2+0x21/0xa0\n   _raw_spin_lock_irqsave+0x35/0x50\n   ? __pm_relax.part.2+0x21/0xa0\n   __pm_relax.part.2+0x21/0xa0\n   wakeup_source_destroy.part.3+0x18/0x190\n   wakeup_source_register+0x43/0x50\n \n Fixes: c8377adfa781 (\"PM / wakeup: Show wakeup sources stats in sysfs\")\n Reported-by: Qian Cai \u003ccai@lca.pw\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit c8377adfa78103be5380200eb9dab764d7ca890e\nAuthor: Tri Vo \u003ctrong@android.com\u003e\nDate: Tue Aug 6 18:48:46 2019 -0700\n\n PM / wakeup: Show wakeup sources stats in sysfs\n \n Add an ID and a device pointer to \u0027struct wakeup_source\u0027. Use them to to\n expose wakeup sources statistics in sysfs under\n /sys/class/wakeup/wakeup\u003cID\u003e/*.\n \n Co-developed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Co-developed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Tri Vo \u003ctrong@android.com\u003e\n Tested-by: Kalesh Singh \u003ckaleshsingh@google.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 2434aea58e652a9fe114181ac90aa60e2f2e1b25\nAuthor: Tri Vo \u003ctrong@android.com\u003e\nDate: Tue Aug 6 18:48:45 2019 -0700\n\n PM / wakeup: Use wakeup_source_register() in wakelock.c\n \n kernel/power/wakelock.c duplicates wakeup source creation and\n registration code from drivers/base/power/wakeup.c.\n \n Change struct wakelock\u0027s wakeup source to a pointer and use\n wakeup_source_register() function to create and register said wakeup\n source. Use wakeup_source_unregister() on cleanup path.\n \n Signed-off-by: Tri Vo \u003ctrong@android.com\u003e\n Reviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 0d105d0f25386ee5b74603db6d249ebed7590cbc\nAuthor: Tri Vo \u003ctrong@android.com\u003e\nDate: Tue Aug 6 18:48:44 2019 -0700\n\n PM / wakeup: Drop wakeup_source_init(), wakeup_source_prepare()\n \n wakeup_source_init() has no users. Remove it.\n \n As a result, wakeup_source_prepare() is only called from\n wakeup_source_create(). Merge wakeup_source_prepare() into\n wakeup_source_create() and remove it.\n \n Change wakeup_source_create() behavior so that assigning NULL to wakeup\n source\u0027s name throws an error.\n \n Signed-off-by: Tri Vo \u003ctrong@android.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 8698fab1c69a7265282d6b2ac081122e94d72cf8\nAuthor: Krzysztof Wilczynski \u003ckw@linux.com\u003e\nDate: Mon Aug 19 15:53:24 2019 +0200\n\n ACPI/PCI: Remove surplus parentheses from a return statement\n \n Remove unnecessary parentheses enclosing the value in a return\n statement in the drivers/acpi/pci_link.c.\n \n Signed-off-by: Krzysztof Wilczynski \u003ckw@linux.com\u003e\n Reviewed-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 71bb4d9a4085ec6e2a90ca62c6afcafb945b55aa\nAuthor: Bob Moore \u003crobert.moore@intel.com\u003e\nDate: Fri Aug 16 14:43:28 2019 -0700\n\n ACPICA: Update version to 20190816.\n \n ACPICA commit db29ce57ced39ec610667c4b946bc3bb38bc8efa\n \n Version 20190816.\n \n Link: https://github.com/acpica/acpica/commit/db29ce57\n Signed-off-by: Bob Moore \u003crobert.moore@intel.com\u003e\n Signed-off-by: Erik Schmauss \u003cerik.schmauss@intel.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 8696beed34d1c895f54279be19de82becf7a869b\nAuthor: Jung-uk Kim \u003cjkim@free_BSD.org\u003e\nDate: Fri Aug 16 14:43:27 2019 -0700\n\n ACPICA: Add \"Windows 2019\" string to _OSI support.\n \n ACPICA commit 32fffb242800b0202986e86d9b0e16f88a23de66\n \n Link: https://github.com/acpica/acpica/commit/32fffb24\n Signed-off-by: Jung-uk Kim \u003cjkim@free_BSD.org\u003e\n Signed-off-by: Bob Moore \u003crobert.moore@intel.com\u003e\n Signed-off-by: Erik Schmauss \u003cerik.schmauss@intel.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit be0381cf2d57c21d06ebf8d8f8a3d5b8bfb2e4ab\nAuthor: Jung-uk Kim \u003cjkim@free_BSD.org\u003e\nDate: Fri Aug 16 14:43:26 2019 -0700\n\n ACPICA: Differentiate Windows 8.1 from Windows 8.\n \n ACPICA commit 66db7b38f61e63f11e48a0ea993d92b12e0a17ca\n \n Link: https://github.com/acpica/acpica/commit/66db7b38\n Signed-off-by: Jung-uk Kim \u003cjkim@free_BSD.org\u003e\n Signed-off-by: Bob Moore \u003crobert.moore@intel.com\u003e\n Signed-off-by: Erik Schmauss \u003cerik.schmauss@intel.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 36056d0cd677bbde0dc75457be71119a630ebfc0\nAuthor: Bob Moore \u003crobert.moore@intel.com\u003e\nDate: Fri Aug 16 14:43:25 2019 -0700\n\n ACPICA: Fully deploy ACPI_PRINTF_LIKE macro\n \n ACPICA commit d06def132a8852d02c9c7fee60f17b2011066e8e\n \n Macro was not being used across all \"printf-like\" functions.\n \n Also, clean up all calls to such functions now that they are\n analyzed by the compiler (gcc). Both in 32-bit mode and 64-bit\n mode.\n \n Link: https://github.com/acpica/acpica/commit/d06def13\n Signed-off-by: Bob Moore \u003crobert.moore@intel.com\u003e\n Signed-off-by: Erik Schmauss \u003cerik.schmauss@intel.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 09d2c01ba9e73254e9e6c4fda59d5cf6bd3c89ed\nAuthor: Bob Moore \u003crobert.moore@intel.com\u003e\nDate: Fri Aug 16 14:43:24 2019 -0700\n\n ACPICA: iASL,acpi_dump: Improve y/n query\n \n The y/n query is used for file overwrite. Use fgetc, check for\n standalone newline.\n \n ACPICA commit f9eb60ead76e5b2b6e578b553f592452ccfca47a\n \n Link: https://github.com/acpica/acpica/commit/f9eb60ea\n Signed-off-by: Bob Moore \u003crobert.moore@intel.com\u003e\n Signed-off-by: Erik Schmauss \u003cerik.schmauss@intel.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 231ec06e793891071e491e574fe351835aeb9c5c\nAuthor: Bob Moore \u003crobert.moore@intel.com\u003e\nDate: Fri Aug 16 14:43:23 2019 -0700\n\n ACPICA: Fix issues with arg types within printf format strings\n \n ACPICA commit db2638ccaac84b61e92f34d60c3630ff5217f852\n \n Link: https://github.com/acpica/acpica/commit/db2638cc\n Link: https://github.com/acpica/acpica/commit/634c3085\n Signed-off-by: Bob Moore \u003crobert.moore@intel.com\u003e\n Signed-off-by: Erik Schmauss \u003cerik.schmauss@intel.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 779cc7ce3dc519b05f274b79fb5a6c1703f373b5\nAuthor: Bob Moore \u003crobert.moore@intel.com\u003e\nDate: Fri Aug 16 14:43:22 2019 -0700\n\n ACPICA: Macros: remove pointer math on a null pointer\n \n ACPICA commit 02bbca5070e42d298c9b824300aa0eb8a082d797\n \n Causes warnings on some compilers and/or tools.\n Changed ACPI_TO_POINTER to use ACPI_CAST_PTR instead of using\n arithmetic.\n \n Link: https://github.com/acpica/acpica/commit/02bbca50\n Reported-by: Qian Cai \u003ccai@lca.pw\u003e\n Signed-off-by: Bob Moore \u003crobert.moore@intel.com\u003e\n Signed-off-by: Erik Schmauss \u003cerik.schmauss@intel.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 67a72420a326b45514deb3f212085fb2cd1595b5\nAuthor: Bob Moore \u003crobert.moore@intel.com\u003e\nDate: Fri Aug 16 14:43:21 2019 -0700\n\n ACPICA: Increase total number of possible Owner IDs\n \n ACPICA commit 1f1652dad88b9d767767bc1f7eb4f7d99e6b5324\n \n From 255 to 4095 possible IDs.\n \n Link: https://github.com/acpica/acpica/commit/1f1652da\n Reported-by: Hedi Berriche \u003chedi.berriche @hpe.com\u003e\n Signed-off-by: Bob Moore \u003crobert.moore@intel.com\u003e\n Signed-off-by: Erik Schmauss \u003cerik.schmauss@intel.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 189c7213f4c5decddfe5969e28a1ce0bcae28cb1\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Aug 16 14:43:20 2019 -0700\n\n ACPICA: Debugger: remove redundant assignment on obj_desc\n \n ACPICA commit f530f1acb3128136ad97c715fdaebbbeff283ee2\n \n Pointer obj_desc is being initialized with a value that is never\n read and it is being updated later with a new value. The initialization\n is redundant and can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n \n Link: https://github.com/acpica/acpica/commit/f530f1ac\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Bob Moore \u003crobert.moore@intel.com\u003e\n Signed-off-by: Erik Schmauss \u003cerik.schmauss@intel.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 2b2d4247e445b97113c5e61df8fbc0d756482ef8\nAuthor: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\nDate: Wed Aug 14 15:58:58 2019 +0300\n\n Documentation: ACPI: DSD: Convert LED documentation to ReST\n \n Convert the LED documentation in text format into ReST.\n \n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 49ed86f503be80aac158a567c4cfd31cf1cd181e\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Aug 19 12:57:06 2019 +0530\n\n soc: amlogic: meson-gx-socinfo: Add of_node_put() before return\n \n The variable np in function meson_gx_socinfo_init takes the return value\n of of_find_compatible_node, which gets a node but does not put it. If\n this node is not put it may cause a memory leak. Hence put np after its\n usefulness has been exhausted.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Fixes: a9daaba2965e (\"soc: Add Amlogic SoC Information driver\")\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 6abc7622271dc520f241462e2474c71723638851\nAuthor: Liguang Zhang \u003czhangliguang@linux.alibaba.com\u003e\nDate: Mon Jul 15 14:58:44 2019 +0800\n\n ACPI / APEI: Release resources if gen_pool_add() fails\n \n Destroy ghes_estatus_pool and release memory allocated via vmalloc() on\n errors in ghes_estatus_pool_init() in order to avoid memory leaks.\n \n [ bp: do the labels properly and with descriptive names and massage. ]\n \n Signed-off-by: Liguang Zhang \u003czhangliguang@linux.alibaba.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Link: https://lkml.kernel.org/r/1563173924-47479-1-git-send-email-zhangliguang@linux.alibaba.com\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit be6f7ef69cf0d8cfeb6af5ad62c558fe35e13cce\nAuthor: Brett Creeley \u003cbrett.creeley@intel.com\u003e\nDate: Thu Jul 25 01:55:41 2019 -0700\n\n ice: improve print for VF\u0027s when adding/deleting MAC filters\n \n When we fail to add/delete MAC filters in the VF, the print doesn\u0027t\n distinguish between the two. Fix that by printing whether or not we\n failed to add/delete the MAC filter respectively.\n \n Signed-off-by: Brett Creeley \u003cbrett.creeley@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit cbfe31b5d74dcfba46329813855e28e53d080d57\nAuthor: Pawel Kaminski \u003cpawel.kaminski@intel.com\u003e\nDate: Thu Jul 25 01:55:40 2019 -0700\n\n ice: Change type for queue counts\n \n These queue variables are being assigned values that are type u16.\n Change the local variables to match these types. Since these\n represent queue counts, they should never be negative.\n \n Signed-off-by: Pawel Kaminski \u003cpawel.kaminski@intel.com\u003e\n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit c275684b9250f66e83b1769d76120e03ac964ec4\nAuthor: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\nDate: Thu Jul 25 01:55:39 2019 -0700\n\n ice: Move VF resources definition to SR-IOV specific file\n \n In order to use some of the VF resources definition in the SR-IOV specific\n virtchnl header file, this patch moves applicable code to\n ice_virtchnl_pf.h file accordingly... and they should have been defined in\n the destination file originally.\n \n Signed-off-by: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 11836214d5b7fbfe9df8aeb5dc2b23a6bedf185c\nAuthor: Brett Creeley \u003cbrett.creeley@intel.com\u003e\nDate: Thu Jul 25 01:55:38 2019 -0700\n\n ice: Increase size of Mailbox receive queue for many VFs\n \n Currently we use the ICE_MBXQ_LEN for both the Mailbox send and receive\n queues that are used to communicate with VFs. This is fine for the send\n queue because the PF driver will lock the queue for every single send,\n but for the Mailbox receive queue every VF is posting to its Mailbox\n send queue and the hardware is then handing the message to the PF on its\n Mailbox receive queue. This becomes a problem with many VFs because it\n seems to overburden the Mailbox receive queue on the PF. Fix this by\n increasing the Mailbox receive queue for the PF to 512 entries.\n \n The number 512 was determined based on the number of VFs supported by\n the device. We can have a total of 256 VFs so in the worst case this\n allows the VFs to put 2 messages in the PFs Mailbox receive queue at the\n same time.\n \n Signed-off-by: Brett Creeley \u003cbrett.creeley@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 60d628ea27d29499d2f71817a6caf5b20b6e293a\nAuthor: Brett Creeley \u003cbrett.creeley@intel.com\u003e\nDate: Thu Jul 25 01:55:37 2019 -0700\n\n ice: Reduce wait times during VF bringup/reset\n \n Currently there are a couple places where the VF is waiting too long when\n checking the status of registers. This is causing the AVF driver to\n spin for longer than necessary in the __IAVF_STARTUP state. Sometimes\n it causes the AVF to go into the __IAVF_COMM_FAILED, which may retrigger\n the __IAVF_STARTUP state. Try to reduce the chance of this happening by\n removing unnecessary wait times in VF bringup/resets.\n \n Signed-off-by: Brett Creeley \u003cbrett.creeley@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 1337175deca72c3a1c846344d5743667a4dbedef\nAuthor: Paul Greenwalt \u003cpaul.greenwalt@intel.com\u003e\nDate: Thu Jul 25 01:55:36 2019 -0700\n\n ice: update GLINT_DYN_CTL and GLINT_VECT2FUNC register access\n \n Register access for GLINT_DYN_CTL and GLINT_VECT2FUNC should be within\n the PF space and not the absolute device space.\n \n Signed-off-by: Paul Greenwalt \u003cpaul.greenwalt@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit e6c45149b88e9dee8a97e6c365fe967c196fd500\nAuthor: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\nDate: Thu Jul 25 01:55:35 2019 -0700\n\n ice: Do not always bring up PF VSI in ice_ena_vsi()\n \n During rebuild ice_ena_vsi() is called to recover the VSI state.\n This function assumes the PF VSI is always to be enabled, however,\n it\u0027s possible that during reset/rebuild the interface can be\n brought down. If this occurs, we can attempt to bring up the PF\n VSI on a downed interface which can lead to various crashes. If\n the interface is not running, do not bring up the associated VSI.\n \n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit ac6f733a7bd5e23ce9d58ef995f51fbd1ad2fa97\nAuthor: Mitch Williams \u003cmitch.a.williams@intel.com\u003e\nDate: Thu Jul 25 01:55:34 2019 -0700\n\n ice: allow empty Rx descriptors\n \n In some circumstances, the hardware will hand us a receive descriptor\n which has no data attached, but is otherwise valid. The receive code was\n improperly ignoring these descriptors, which result in an infinite loop.\n \n To fix this, change the receive code to process all descriptors,\n regardless of the size of the associated data. Add checks to the\n memory-handling functions to allow for zero size.\n \n Signed-off-by: Mitch Williams \u003cmitch.a.williams@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 7829570e287d938fc49b8ae151d9af26436967a8\nAuthor: Usha Ketineni \u003cusha.k.ketineni@intel.com\u003e\nDate: Thu Jul 25 01:55:33 2019 -0700\n\n ice: Fix kernel hang with DCB reset in CEE mode\n \n This patch fixes the set local MIB AQ call failures in the DCB rebuild path\n by setting the defaults for the ETS recommended DCB configuration. Also,\n willing bits for the DCB configuration needs to be set correctly. Resets\n works fine in IEEE mode as the ETS recommended DCB configuration is\n populated but not in CEE mode.\n Without this patch, PFR causes the kernel hang in CEE mode.\n \n Signed-off-by: Usha Ketineni \u003cusha.k.ketineni@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 2ab28bb04ce6432f15cd5086e7c7384f3deab639\nAuthor: Brett Creeley \u003cbrett.creeley@intel.com\u003e\nDate: Thu Jul 25 01:55:32 2019 -0700\n\n ice: Set WB_ON_ITR when we don\u0027t re-enable interrupts\n \n Currently when busy polling is enabled we aren\u0027t setting/enabling\n WB_ON_ITR in the driver. This doesn\u0027t break the driver, but it does\n cause issues. If we don\u0027t enable WB_ON_ITR mode we will still get\n write-backs from hardware during polling when a cache line has been\n filled, but if a cache line is not filled we will not get the\n write-back because WB_ON_ITR is not set. Fix this by enabling\n WB_ON_ITR in the driver when interrupts are disabled.\n \n Signed-off-by: Brett Creeley \u003cbrett.creeley@intel.com\u003e\n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 53119e513478b3b5c2b81a9a049f52cf17fcfff3\nAuthor: Enrico Weigelt \u003cinfo@metux.net\u003e\nDate: Tue Aug 20 12:08:43 2019 -0700\n\n Input: soc_button_array - use platform_device_register_resndata()\n \n The registration of gpio-keys device can be written much shorter\n by using the platform_device_register_resndata() helper.\n \n Signed-off-by: Enrico Weigelt \u003cinfo@metux.net\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit cfd8579ddc02d8c5024bae7106ba37079653b3d2\nAuthor: Hui Wang \u003chui.wang@canonical.com\u003e\nDate: Tue Aug 20 12:07:01 2019 -0700\n\n Input: psmouse - drop all unneeded functions from mouse headers\n \n Recently we had a building error if we enable the MOUSE_PS2_ALPS while\n disable the MOUSE_PS2_TRACKPOINT, and was fixed by 49e6979e7e92\n (\"Input: psmouse - fix build error of multiple definition\").\n \n We could improve that fix by dropping all unneeded functions and\n CONFIG_MOUSE_ guards from the header, it is safe to do that since\n those functions are not directly called by psmouse-base.c anymore.\n \n Signed-off-by: Hui Wang \u003chui.wang@canonical.com\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit f1a4a66d23102e7681beb8725e7d00f359ef5c87\nAuthor: Paul Greenwalt \u003cpaul.greenwalt@intel.com\u003e\nDate: Thu Jul 25 01:55:31 2019 -0700\n\n ice: fix set pause param autoneg check\n \n When ETHTOOL_GLINKSETTINGS is defined get pause param pause-\u003eautoneg\n reports SW configured setting, however when not defined get pause param\n pause-\u003eautoneg reports the link status. Set pause param needs to compare\n pause-\u003eautoneg with the same source as get pause param to block the user\n from changing autoneg with the set pause param option, or the user\n may be incorrectly blocked from changing Rx|Tx pause settings.\n \n Signed-off-by: Paul Greenwalt \u003cpaul.greenwalt@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 1eba383f4e36637ba859cb82b40e198c415db802\nAuthor: Maxim Mikityanskiy \u003cmaximmi@mellanox.com\u003e\nDate: Thu Aug 15 19:46:16 2019 +0000\n\n net/mlx5: Add lag_tx_port_affinity capability bit\n \n Add the lag_tx_port_affinity HCA capability bit that indicates that\n setting port affinity of TISes is supported.\n \n Signed-off-by: Maxim Mikityanskiy \u003cmaximmi@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit caa1854735449d7afac6781679621fb9142fe810\nAuthor: Aya Levin \u003cayal@mellanox.com\u003e\nDate: Thu Aug 15 19:46:14 2019 +0000\n\n net/mlx5: Expose IP-in-IP capability bit\n \n Expose Fw indication that it supports Stateless Offloads for IP over IP\n tunneled packets. The following offloads are supported for the inner\n packets: RSS, RX \u0026 TX Checksum Offloads, LSO and Flow Steering.\n \n Signed-off-by: Aya Levin \u003cayal@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 30b10e89f2ae2c7b7b32548e04f57ab7eb030dfb\nAuthor: Moshe Shemesh \u003cmoshe@mellanox.com\u003e\nDate: Thu Aug 15 19:46:11 2019 +0000\n\n net/mlx5: Add support for VNIC_ENV internal rq counter\n \n Add mlx5 interface support for reading internal rq out of buffer counter\n as part of QUERY_VNIC_ENV command. The command is used by the driver to\n query vnic diagnostic statistics from FW.\n \n Signed-off-by: Moshe Shemesh \u003cmoshe@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 866ff8f22380a49d665ed72521704844bba6de08\nAuthor: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\nDate: Thu Aug 15 19:46:09 2019 +0000\n\n net/mlx5: Improve functions documentation\n \n Fix documentation of mlx5_eq_enable/disable to cleanup compiler warnings.\n \n drivers/net/ethernet/mellanox/mlx5/core//eq.c:334:\n warning: Function parameter or member \u0027dev\u0027 not described in \u0027mlx5_eq_enable\u0027\n warning: Function parameter or member \u0027eq\u0027 not described in \u0027mlx5_eq_enable\u0027\n warning: Function parameter or member \u0027nb\u0027 not described in \u0027mlx5_eq_enable\u0027\n \n drivers/net/ethernet/mellanox/mlx5/core//eq.c:355:\n warning: Function parameter or member \u0027dev\u0027 not described in \u0027mlx5_eq_disable\u0027\n warning: Function parameter or member \u0027eq\u0027 not described in \u0027mlx5_eq_disable\u0027\n warning: Function parameter or member \u0027nb\u0027 not described in \u0027mlx5_eq_disable\u0027\n \n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit eed6f7dc28d35bbed74115fc246b9a29af0918de\nAuthor: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\nDate: Thu Aug 15 19:46:08 2019 +0000\n\n net/mlx5: Add missing include file to lib/crypto.c\n \n Add missing include file to avoid compiler warnings:\n drivers/net/ethernet/mellanox/mlx5/core//lib/crypto.c:6:5:\n warning: no previous prototype for ‘mlx5_create_encryption_key’\n 6 | int mlx5_create_encryption_key(struct mlx5_core_dev *mdev,\n | ^~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/net/ethernet/mellanox/mlx5/core//lib/crypto.c:60:6:\n warning: no previous prototype for ‘mlx5_destroy_encryption_key’\n 60 | void mlx5_destroy_encryption_key(struct mlx5_core_dev *mdev, ...\n \n Fixes: 45d3b55dc665 (\"net/mlx5: Add crypto library to support create/destroy encryption key\")\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 0506d5ba68766a5d3fcc322a03005b825652de3f\nAuthor: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\nDate: Tue Aug 20 16:46:43 2019 +0200\n\n s390/lcs: don\u0027t use intparm for channel IO\n \n lcs passes an intparm when calling ccw_device_*(), even though lcs_irq()\n later makes no use of this.\n \n To reduce the confusion, consistently pass 0 as intparm instead.\n \n Signed-off-by: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\n Reviewed-by: Sebastian Ott \u003csebott@linux.ibm.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4f6e01f3d6ff9ca56692a29df2f6c2a97daf0a95\nAuthor: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\nDate: Tue Aug 20 16:46:42 2019 +0200\n\n s390/ctcm: don\u0027t use intparm for channel IO\n \n ctcm passes an intparm when calling ccw_device_*(), even though\n ctcm_irq_handler() later makes no use of this.\n \n To reduce the confusion, consistently pass 0 as intparm instead.\n \n Signed-off-by: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\n Reviewed-by: Sebastian Ott \u003csebott@linux.ibm.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 59b757a9bf2bd30173029fa7bd6821239d6a7242\nAuthor: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\nDate: Tue Aug 20 16:46:41 2019 +0200\n\n s390/qeth: streamline control code for promisc mode\n \n We have logic to determine the desired promisc mode in _each_ code path.\n Change things around so that there is a clean split between\n (a) high-level code that selects the new mode, and (b) implementations\n of the various mechanisms to program this mode.\n \n This also keeps qeth_promisc_to_bridge() from polluting the debug logs\n on each RX modeset.\n \n Signed-off-by: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\n Reviewed-by: Alexandra Winter \u003cwintera@linux.ibm.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 2744d811690364c8586b19f3dd9f365e2bfe6837\nAuthor: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\nDate: Tue Aug 20 16:46:40 2019 +0200\n\n s390/qeth: get vnicc sub-cmd type from reply data\n \n When processing the reply for a vnicc cmd, there\u0027s no need to remember\n which specific sub-cmd type we initially sent. The reply itself contains\n all the needed information.\n \n Signed-off-by: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 308946b07430728016851d11051dcf1c78887973\nAuthor: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\nDate: Tue Aug 20 16:46:39 2019 +0200\n\n s390/qeth: merge qeth_reply struct into qeth_cmd_buffer\n \n Except for card-\u003eread_cmd, every cmd we issue now passes through\n qeth_send_control_data() and allocates a qeth_reply struct. The way we\n use this struct requires additional refcounting, and pointer tracking.\n \n Clean up things by moving most of qeth_reply\u0027s content into the main\n cmd struct. This keeps things in one place, saves us the additional\n refcounting and simplifies the overall code flow.\n A nice little benefit is that we can now match incoming replies against\n the pending requests themselves, without caching the requests\u0027 seqnos.\n \n The qeth_reply struct stays around for a little bit longer in a shrunk\n form, to avoid touching every single callback.\n \n Signed-off-by: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 32e85a0d83eed96ec2f2a6a2d527ef927e90ea2e\nAuthor: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\nDate: Tue Aug 20 16:46:38 2019 +0200\n\n s390/qeth: keep cmd alive after IO completion\n \n Current code releases the cmd struct after its initial IO has completed.\n Any reply processing is done independently, using a separate qeth_reply\n struct.\n In preparation for merging the cmd and reply structs together, take an\n additional reference on the cmd object so that it stays around all the\n way until qeth_send_control_data() returns.\n \n Signed-off-by: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7c5f8ffb335747a8aabd8fb504a66b39aeaf4d21\nAuthor: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\nDate: Tue Aug 20 16:46:37 2019 +0200\n\n s390/qeth: use correct length field in SNMP cmd callback\n \n qeth_snmp_command_cb() is the only cmd callback that pulls the reply\u0027s\n data length from a low-level transport header field. This requires\n additional complexity (ie. reply-\u003eoffset) to make the header accessible\n to what is supposed to be a pure IPA cmd callback.\n \n Adapter cmds have a length field in their sub-cmd header, get the data\n length from there instead.\n \n Signed-off-by: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 12fc286f84b19ba61a8f64c80dc683cfb19f0251\nAuthor: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\nDate: Tue Aug 20 16:46:36 2019 +0200\n\n s390/qeth: propagate length of processed cmd IO data to callback\n \n When an cmd IO completes in qeth_irq(), calculate how much data was\n processed by the device and pass this value to the cmd\u0027s callback.\n \n This allows cmds that retrieve data from the device to check whether\n sufficient data was received, so we do that in qeth_read_conf_data_cb().\n \n Suggested-by: Jens Remus \u003cjremus@linux.ibm.com\u003e\n Signed-off-by: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit afc1f67b99d59c4ddd29013fbb404032b4b6d08e\nAuthor: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\nDate: Tue Aug 20 16:46:35 2019 +0200\n\n s390/qeth: use node_descriptor struct\n \n Rather than fumbling with hard-coded offsets, use the proper struct to\n access the retrieved RCD information.\n \n Signed-off-by: Julian Wiedmann \u003cjwi@linux.ibm.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d9bd6d279236a1c90c1e73220c2ae52996c8ff52\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Tue Aug 20 22:14:46 2019 +0800\n\n netdevsim: Fix build error without CONFIG_INET\n \n If CONFIG_INET is not set, building fails:\n \n drivers/net/netdevsim/dev.o: In function `nsim_dev_trap_report_work\u0027:\n dev.c:(.text+0x67b): undefined reference to `ip_send_check\u0027\n \n Use ip_fast_csum instead of ip_send_check to avoid\n dependencies on CONFIG_INET.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Fixes: da58f90f11f5 (\"netdevsim: Add devlink-trap support\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Acked-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Reviewed-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Tested-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 805a1f620ba38c5f6de8b9697f35dcb38d8112b5\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:46:24 2019 -0400\n\n xprtrdma: Ensure creating an MR does not trigger FS writeback\n \n Probably would be good to also pass GFP flags to ib_alloc_mr.\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit e9a12e14322d7ddafeed6aec0d3fb02c0b5dc03c\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Tue Aug 20 16:40:52 2019 +0200\n\n arm64: dts: add support for SM1 based SEI Robotics SEI610\n \n Add support for the Amlogic SM1 Based SEI610 board.\n \n The SM1 SoC is a derivative of the G12A SoC Family with :\n - Cortex-A55 core instead of A53\n - more power domains, including USB \u0026 PCIe\n - a neural network co-processor (NNA)\n - a CSI input and image processor\n - some changes in the audio complex, thus not yet enabled\n \n The SEI610 board is a derivative of the SEI510 board with :\n - removed ADC based touch button, replaced with 3x GPIO buttons\n - physical switch disabling on-board MICs\n - USB-C port for USB 2.0 OTG\n - On-board FTDI USB2SERIAL port for Linux console\n \n Audio, Display and USB support will be added later when support of the\n corresponding power domains will be added, for now they are kept disabled.\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Acked-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n [khilman: fix minor typo regultor -\u003e regulator]\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit ece644858c253528c1917c9f2751ef4be32408cc\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Tue Aug 20 22:40:20 2019 +0530\n\n dt-bindings: arm: Add kryo485 compatible\n \n Kryo485 is found in SM8150, so add it it list of cpu compatibles\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 3b39f52a02d4b3322744a0a32d59142e01afa435\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:45:37 2019 -0400\n\n xprtrdma: Move rpcrdma_mr_get out of frwr_map\n \n Refactor: Retrieve an MR and handle error recovery entirely in\n rpc_rdma.c, as this is not a device-specific function.\n \n Note that since commit 89f90fe1ad8b (\"SUNRPC: Allow calls to\n xprt_transmit() to drain the entire transmit queue\"), the\n xprt_transmit function handles the cond_resched. The transport no\n longer has to do this itself.\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit 1ca3f4c054a4e3765bdeb62c849d940b5bc8002d\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:44:50 2019 -0400\n\n xprtrdma: Combine rpcrdma_mr_put and rpcrdma_mr_unmap_and_put\n \n Clean up. There is only one remaining rpcrdma_mr_put call site, and\n it can be directly replaced with unmap_and_put because mr-\u003emr_dir is\n set to DMA_NONE just before the call.\n \n Now all the call sites do a DMA unmap, and we can just rename\n mr_unmap_and_put to mr_put, which nicely matches mr_get.\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit 7cb9a94c158b956f46cf093ed966d0c1e996dddb\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Aug 19 16:31:45 2019 +0200\n\n posix-cpu-timers: Fixup stale comment\n \n The comment above cleanup_timers() is outdated. The timers are only removed\n from the task/process list heads but not modified in any other way.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190819143801.747233612@linutronix.de\n\ncommit 3758b0f86ef502e2f342055caef6d2232c2558b7\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Aug 19 16:31:43 2019 +0200\n\n alarmtimers: Avoid rtc.h include\n \n rtc.h is not needed in alarmtimers when a forward declaration of struct\n rtc_device is provided. That allows to include posix-timers.h without\n adding more includes to alarmtimer.h or creating circular include\n dependencies.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190819143801.565389536@linutronix.de\n\ncommit ce03f613461642669d6150c405dd28f4bfd54bbb\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Aug 19 16:31:42 2019 +0200\n\n posix-timers: Cleanup forward declarations and includes\n \n - Rename struct siginfo to kernel_siginfo as that is used and required\n - Add a forward declaration for task_struct and remove sched.h include\n - Remove timex.h include as it is not needed\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190819143801.472005793@linutronix.de\n\ncommit b1b9f97a0937211dbca638c476ac47ee39875661\nAuthor: Gavi Teitz \u003cgavi@mellanox.com\u003e\nDate: Sun Aug 11 09:21:20 2019 +0300\n\n net/mlx5: Fix the order of fc_stats cleanup\n \n Previously, mlx5_cleanup_fc_stats() would cleanup the flow counter\n pool beofre releasing all the counters to it, which would result in\n flow counter bulks not getting freed. Resolve this by changing the\n order in which elements of fc_stats are cleaned up, so that the flow\n counter pool is cleaned up after all the counters are released.\n \n Also move cleanup actions for freeing the bulk query memory and\n destroying the idr to the end of mlx5_cleanup_fc_stats().\n \n Signed-off-by: Gavi Teitz \u003cgavi@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Reviewed-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 3c140dd54f37693b48609819cbaf2e294a025b92\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Mon Aug 12 15:38:38 2019 +0300\n\n net/mlx5e: Fix deallocation of non-fully init encap entries\n \n Recent rtnl lock dependency refactoring changed encap entry attach code to\n insert encap entry to hash table before it was fully initialized in order\n to allow concurrent tc users to wait on completion for encap entry to\n finish initialization. That change required all the users of encap entry to\n obtain reference to it first and for caller that creates encap to put\n reference to it on error, instead of freeing the entry memory directly.\n However, releasing reference to such encap entry that wasn\u0027t fully\n initialized causes NULL pointer dereference in\n mlx5e_rep_encap_entry_detach() which expects e-\u003eout_dev to be set and encap\n to be attached to nhe:\n \n [ 1092.454517] BUG: unable to handle page fault for address: 00000000000420e8\n [ 1092.454571] #PF: supervisor read access in kernel mode\n [ 1092.454602] #PF: error_code(0x0000) - not-present page\n [ 1092.454632] PGD 800000083032c067 P4D 800000083032c067 PUD 84107d067 PMD 0\n [ 1092.454673] Oops: 0000 [#1] SMP PTI\n [ 1092.454697] CPU: 20 PID: 22393 Comm: tc Not tainted 5.3.0-rc3+ #589\n [ 1092.454733] Hardware name: Supermicro SYS-2028TP-DECR/X10DRT-P, BIOS 2.0b 03/30/2017\n [ 1092.454806] RIP: 0010:mlx5e_rep_encap_entry_detach+0x1c/0x630 [mlx5_core]\n [ 1092.454845] Code: be f4 ff ff ff e9 11 ff ff ff 0f 1f 40 00 0f 1f 44 00 00 55 48 89 e5 41 57 41 56 41 55 41 54 49 89 fc 53 48 89 f3 48 83 ec 30 \u003c48\u003e 8b 87 28 16 04 00 48 89 f7 48 05 d0 03 00 00 48 89\n 45 c8 e8 cb\n [ 1092.454942] RSP: 0018:ffffb6f08421f5a0 EFLAGS: 00010286\n [ 1092.454974] RAX: 0000000000000000 RBX: ffff8ab668644e00 RCX: ffffb6f08421f56c\n [ 1092.455013] RDX: ffff8ab668644e40 RSI: ffff8ab668644e00 RDI: 0000000000000ac0\n [ 1092.455053] RBP: ffffb6f08421f5f8 R08: 0000000000000001 R09: 0000000000000000\n [ 1092.455092] R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000ac0\n [ 1092.455131] R13: 00000000ffffff9b R14: ffff8ab63f200ac0 R15: ffff8ab668644e40\n [ 1092.455171] FS: 00007fa195bdc480(0000) GS:ffff8ab66fa00000(0000) knlGS:0000000000000000\n [ 1092.455216] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n [ 1092.455249] CR2: 00000000000420e8 CR3: 0000000867522001 CR4: 00000000001606e0\n [ 1092.455288] Call Trace:\n [ 1092.455315] ? __mutex_unlock_slowpath+0x4d/0x2a0\n [ 1092.455365] mlx5e_encap_dealloc.isra.0+0x31/0x60 [mlx5_core]\n [ 1092.455424] mlx5e_tc_add_fdb_flow+0x596/0x750 [mlx5_core]\n [ 1092.455484] __mlx5e_add_fdb_flow+0x152/0x210 [mlx5_core]\n [ 1092.455534] mlx5e_configure_flower+0x4d5/0xe30 [mlx5_core]\n [ 1092.455574] tc_setup_cb_call+0x67/0xb0\n [ 1092.455601] fl_hw_replace_filter+0x142/0x300 [cls_flower]\n [ 1092.455639] fl_change+0xd24/0x1bdb [cls_flower]\n [ 1092.455675] tc_new_tfilter+0x3e0/0x970\n [ 1092.455709] ? tc_del_tfilter+0x720/0x720\n [ 1092.455735] rtnetlink_rcv_msg+0x389/0x4b0\n [ 1092.455763] ? netlink_deliver_tap+0x95/0x400\n [ 1092.455791] ? rtnl_dellink+0x2d0/0x2d0\n [ 1092.455817] netlink_rcv_skb+0x49/0x110\n [ 1092.455844] netlink_unicast+0x171/0x200\n [ 1092.455872] netlink_sendmsg+0x224/0x3f0\n [ 1092.455901] sock_sendmsg+0x5e/0x60\n [ 1092.455924] ___sys_sendmsg+0x2ae/0x330\n [ 1092.455950] ? task_work_add+0x43/0x50\n [ 1092.455976] ? fput_many+0x45/0x80\n [ 1092.456004] ? __lock_acquire+0x248/0x18e0\n [ 1092.456033] ? find_held_lock+0x2b/0x80\n [ 1092.456058] ? task_work_run+0x7b/0xd0\n [ 1092.456085] __sys_sendmsg+0x59/0xa0\n [ 1092.457013] do_syscall_64+0x5c/0xb0\n [ 1092.457924] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n [ 1092.458842] RIP: 0033:0x7fa195da27b8\n [ 1092.459918] Code: 89 02 48 c7 c0 ff ff ff ff eb bb 0f 1f 80 00 00 00 00 f3 0f 1e fa 48 8d 05 65 8f 0c 00 8b 00 85 c0 75 17 b8 2e 00 00 00 0f 05 \u003c48\u003e 3d 00 f0 ff ff 77 58 c3 0f 1f 80 00 00 00 00 48 83\n ec 28 89 54\n [ 1092.462634] RSP: 002b:00007fff94409298 EFLAGS: 00000246 ORIG_RAX: 000000000000002e\n [ 1092.464011] RAX: ffffffffffffffda RBX: 000000005d515b0e RCX: 00007fa195da27b8\n [ 1092.465391] RDX: 0000000000000000 RSI: 00007fff94409300 RDI: 0000000000000003\n [ 1092.466761] RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000006\n [ 1092.468121] R10: 0000000000404ec2 R11: 0000000000000246 R12: 0000000000000001\n [ 1092.469456] R13: 0000000000480640 R14: 0000000000000016 R15: 0000000000000001\n [ 1092.470766] Modules linked in: act_mirred act_tunnel_key cls_flower dummy vxlan ip6_udp_tunnel udp_tunnel sch_ingress nfsv3 nfs_acl nfs lockd grace fscache tun bridge stp llc sunrpc rdma_ucm rdma_cm\n iw_cm ib_cm mlx5_ib ib_uverbs ib_core intel_rapl_msr intel_rapl_common sb_edac x86_pkg_temp_thermal intel_powerclamp coretemp mlx5_core kvm_intel kvm irqbypass crct10dif_pclmul mei_me crc32_pclmul crc32\n c_intel igb iTCO_wdt ghash_clmulni_intel ses mlxfw intel_cstate iTCO_vendor_support ptp intel_uncore lpc_ich pps_core mei i2c_i801 joydev intel_rapl_perf ioatdma enclosure ipmi_ssif pcspkr dca wmi ipmi_\n si ipmi_devintf ipmi_msghandler acpi_pad acpi_power_meter ast i2c_algo_bit drm_vram_helper ttm drm_kms_helper drm mpt3sas raid_class scsi_transport_sas\n [ 1092.479618] CR2: 00000000000420e8\n [ 1092.481214] ---[ end trace ce2e0f4d9a67f604 ]---\n \n To fix the issue, set e-\u003ecompl_result to positive value after encap was\n initialized successfully. Check e-\u003ecompl_result value in\n mlx5e_encap_dealloc() and only detach and dealloc encap if the value is\n positive.\n \n Fixes: d589e785baf5 (\"net/mlx5e: Allow concurrent creation of encap entries\")\n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 26aa7ab10f15ca24a9564d802ed63006775126d5\nAuthor: Aya Levin \u003cayal@mellanox.com\u003e\nDate: Tue Jul 23 11:46:02 2019 +0300\n\n Documentation: net: mlx5: Devlink health documentation updates\n \n Add documentation for devlink health rx reporter supported by mlx5.\n Update tx reporter documentation.\n \n Signed-off-by: Aya Levin \u003cayal@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 8276ea1353a4968a212f04ddf16659223e5408d9\nAuthor: Aya Levin \u003cayal@mellanox.com\u003e\nDate: Wed Jun 26 23:21:40 2019 +0300\n\n net/mlx5e: Report and recover from CQE with error on RQ\n \n Add support for report and recovery from error on completion on RQ by\n setting the queue back to ready state. Handle only errors with a\n syndrome indicating the RQ might enter error state and could be\n recovered.\n \n Signed-off-by: Aya Levin \u003cayal@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 0a35ab3e138296cfe192628520e4d5f3ff23e730\nAuthor: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\nDate: Fri Jun 14 15:21:15 2019 -0700\n\n net/mlx5e: RX, Handle CQE with error at the earliest stage\n \n Just to be aligned with the MPWQE handlers, handle RX WQE with error\n for legacy RQs in the top RX handlers, just before calling skb_from_cqe().\n \n CQE error handling will now be called at the same stage regardless of\n the RQ type or netdev mode NIC, Representor, IPoIB, etc ..\n \n This will be useful for down stream patch to improve error CQE\n handling.\n \n Signed-off-by: Aya Levin \u003cayal@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 32c57fb26863b48982e33aa95f3b5b23f24b1feb\nAuthor: Aya Levin \u003cayal@mellanox.com\u003e\nDate: Tue Jun 25 21:42:27 2019 +0300\n\n net/mlx5e: Report and recover from rx timeout\n \n Add support for report and recovery from rx timeout. On driver open we\n post NOP work request on the rx channels to trigger napi in order to\n fillup the rx rings. In case napi wasn\u0027t scheduled due to a lost\n interrupt, perform EQ recovery.\n \n Signed-off-by: Aya Levin \u003cayal@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit be5323c8379f488f1de53206edeaf80fc20d7686\nAuthor: Aya Levin \u003cayal@mellanox.com\u003e\nDate: Tue Jun 25 17:44:28 2019 +0300\n\n net/mlx5e: Report and recover from CQE error on ICOSQ\n \n Add support for report and recovery from error on completion on ICOSQ.\n Deactivate RQ and flush, then deactivate ICOSQ. Set the queue back to\n ready state (firmware) and reset the ICOSQ and the RQ (software\n resources). Finally, activate the ICOSQ and the RQ.\n \n Signed-off-by: Aya Levin \u003cayal@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 9d18b5144a0a850e722e7c3d7b700eb1fba7b7e2\nAuthor: Aya Levin \u003cayal@mellanox.com\u003e\nDate: Tue Jul 2 15:47:29 2019 +0300\n\n net/mlx5e: Split open/close ICOSQ into stages\n \n Align ICOSQ open/close behaviour with RQ and SQ. Split open flow into\n open and activate where open handles creation and activate enables the\n queue. Do a symmetric thing in close flow: split into close and\n deactivate.\n \n Signed-off-by: Aya Levin \u003cayal@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 9032e7192eac8e657b52cf1c89fe730308b72c2a\nAuthor: Aya Levin \u003cayal@mellanox.com\u003e\nDate: Tue Jun 25 16:26:46 2019 +0300\n\n net/mlx5e: Add support to rx reporter diagnose\n \n Add rx reporter, which supports diagnose call-back. Diagnostics output\n include: information common to all RQs: RQ type, RQ size, RQ stride\n size, CQ size and CQ stride size. In addition advertise information per\n RQ and its related icosq and attached CQ.\n \n $ devlink health diagnose pci/0000:00:0b.0 reporter rx\n Common config:\n RQ:\n type: 2 stride size: 2048 size: 8\n CQ:\n stride size: 64 size: 1024\n RQs:\n channel ix: 0 rqn: 4308 HW state: 1 SW state: 3 posted WQEs: 7 cc: 7 ICOSQ HW state: 1\n CQ:\n cqn: 1032 HW status: 0\n channel ix: 1 rqn: 4313 HW state: 1 SW state: 3 posted WQEs: 7 cc: 7 ICOSQ HW state: 1\n CQ:\n cqn: 1036 HW status: 0\n channel ix: 2 rqn: 4318 HW state: 1 SW state: 3 posted WQEs: 7 cc: 7 ICOSQ HW state: 1\n CQ:\n cqn: 1040 HW status: 0\n channel ix: 3 rqn: 4323 HW state: 1 SW state: 3 posted WQEs: 7 cc: 7 ICOSQ HW state: 1\n CQ:\n cqn: 1044 HW status: 0\n \n $ devlink health diagnose pci/0000:00:0b.0 reporter rx -jp\n {\n \"Common config\": {\n \"RQ\": {\n \"type\": 2,\n \"stride size\": 2048,\n \"size\": 8\n },\n \"CQ\": {\n \"stride size\": 64,\n \"size\": 1024\n }\n },\n \"RQs\": [ {\n \"channel ix\": 0,\n \"rqn\": 4308,\n \"HW state\": 1,\n \"SW state\": 3,\n \"posted WQEs\": 7,\n \"cc\": 7,\n \"ICOSQ HW state\": 1,\n \"CQ\": {\n \"cqn\": 1032,\n \"HW status\": 0\n }\n },{\n \"channel ix\": 1,\n \"rqn\": 4313,\n \"HW state\": 1,\n \"SW state\": 3,\n \"posted WQEs\": 7,\n \"cc\": 7,\n \"ICOSQ HW state\": 1,\n \"CQ\": {\n \"cqn\": 1036,\n \"HW status\": 0\n }\n },{\n \"channel ix\": 2,\n \"rqn\": 4318,\n \"HW state\": 1,\n \"SW state\": 3,\n \"posted WQEs\": 7,\n \"cc\": 7,\n \"ICOSQ HW state\": 1,\n \"CQ\": {\n \"cqn\": 1040,\n \"HW status\": 0\n }\n },{\n \"channel ix\": 3,\n \"rqn\": 4323,\n \"HW state\": 1,\n \"SW state\": 3,\n \"posted WQEs\": 7,\n \"cc\": 7,\n \"ICOSQ HW state\": 1,\n \"CQ\": {\n \"cqn\": 1044,\n \"HW status\": 0\n }\n } ]\n }\n \n Signed-off-by: Aya Levin \u003cayal@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 11af6a6d09e9a90e05f4a21564232b30c6c25d69\nAuthor: Aya Levin \u003cayal@mellanox.com\u003e\nDate: Thu Jul 11 17:17:36 2019 +0300\n\n net/mlx5e: Add helper functions for reporter\u0027s basics\n \n Introduce helper functions for create and destroy reporters and update\n channels. In the following patch, rx reporter is added and it will use\n these helpers too.\n \n Signed-off-by: Aya Levin \u003cayal@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 2bf09e60ae5ef68c2282f97baf37b7dbd9cc1d48\nAuthor: Aya Levin \u003cayal@mellanox.com\u003e\nDate: Sun Jun 30 15:08:00 2019 +0300\n\n net/mlx5e: Add cq info to tx reporter diagnose\n \n Add cq information to general diagnose output: CQ size and stride size.\n Per SQ add information about the related CQ: cqn and CQ\u0027s HW status.\n \n $ devlink health diagnose pci/0000:00:0b.0 reporter tx\n Common Config:\n SQ:\n stride size: 64 size: 1024\n CQ:\n stride size: 64 size: 1024\n SQs:\n channel ix: 0 tc: 0 txq ix: 0 sqn: 4307 HW state: 1 stopped: false cc: 0 pc: 0\n CQ:\n cqn: 1030 HW status: 0\n channel ix: 1 tc: 0 txq ix: 1 sqn: 4312 HW state: 1 stopped: false cc: 0 pc: 0\n CQ:\n cqn: 1034 HW status: 0\n channel ix: 2 tc: 0 txq ix: 2 sqn: 4317 HW state: 1 stopped: false cc: 0 pc: 0\n CQ:\n cqn: 1038 HW status: 0\n channel ix: 3 tc: 0 txq ix: 3 sqn: 4322 HW state: 1 stopped: false cc: 0 pc: 0\n CQ:\n cqn: 1042 HW status: 0\n \n $ devlink health diagnose pci/0000:00:0b.0 reporter tx -jp\n {\n \"Common Config\": {\n \"SQ\": {\n \"stride size\": 64,\n \"size\": 1024\n },\n \"CQ\": {\n \"stride size\": 64,\n \"size\": 1024\n }\n },\n \"SQs\": [ {\n \"channel ix\": 0,\n \"tc\": 0,\n \"txq ix\": 0,\n \"sqn\": 4307,\n \"HW state\": 1,\n \"stopped\": false,\n \"cc\": 0,\n \"pc\": 0,\n \"CQ\": {\n \"cqn\": 1030,\n \"HW status\": 0\n }\n },{\n \"channel ix\": 1,\n \"tc\": 0,\n \"txq ix\": 1,\n \"sqn\": 4312,\n \"HW state\": 1,\n \"stopped\": false,\n \"cc\": 0,\n \"pc\": 0,\n \"CQ\": {\n \"cqn\": 1034,\n \"HW status\": 0\n }\n },{\n \"channel ix\": 2,\n \"tc\": 0,\n \"txq ix\": 2,\n \"sqn\": 4317,\n \"HW state\": 1,\n \"stopped\": false,\n \"cc\": 0,\n \"pc\": 0,\n \"CQ\": {\n \"cqn\": 1038,\n \"HW status\": 0\n }\n },{\n \"channel ix\": 3,\n \"tc\": 0,\n \"txq ix\": 3,\n \"sqn\": 4322,\n \"HW state\": 1,\n \"stopped\": false,\n \"cc\": 0,\n \"pc\": 0,\n \"CQ\": {\n \"cqn\": 1042,\n \"HW status\": 0\n } ]\n }\n \n Signed-off-by: Aya Levin \u003cayal@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 2d708887a4b1cb142c3179b3b1030dab047467b6\nAuthor: Aya Levin \u003cayal@mellanox.com\u003e\nDate: Sun Jun 30 11:34:15 2019 +0300\n\n net/mlx5e: Extend tx reporter diagnostics output\n \n Enhance tx reporter\u0027s diagnostics output to include: information common\n to all SQs: SQ size, SQ stride size.\n In addition add channel ix, tc, txq ix, cc and pc.\n \n $ devlink health diagnose pci/0000:00:0b.0 reporter tx\n Common config:\n SQ:\n stride size: 64 size: 1024\n SQs:\n channel ix: 0 tc: 0 txq ix: 0 sqn: 4307 HW state: 1 stopped: false cc: 0 pc: 0\n channel ix: 1 tc: 0 txq ix: 1 sqn: 4312 HW state: 1 stopped: false cc: 0 pc: 0\n channel ix: 2 tc: 0 txq ix: 2 sqn: 4317 HW state: 1 stopped: false cc: 0 pc: 0\n channel ix: 3 tc: 0 txq ix: 3 sqn: 4322 HW state: 1 stopped: false cc: 0 pc: 0\n \n $ devlink health diagnose pci/0000:00:0b.0 reporter tx -jp\n {\n \"Common config\": {\n \"SQ\": {\n \"stride size\": 64,\n \"size\": 1024\n }\n },\n \"SQs\": [ {\n \"channel ix\": 0,\n \"tc\": 0,\n \"txq ix\": 0,\n \"sqn\": 4307,\n \"HW state\": 1,\n \"stopped\": false,\n \"cc\": 0,\n \"pc\": 0\n },{\n \"channel ix\": 1,\n \"tc\": 0,\n \"txq ix\": 1,\n \"sqn\": 4312,\n \"HW state\": 1,\n \"stopped\": false,\n \"cc\": 0,\n \"pc\": 0\n },{\n \"channel ix\": 2,\n \"tc\": 0,\n \"txq ix\": 2,\n \"sqn\": 4317,\n \"HW state\": 1,\n \"stopped\": false,\n \"cc\": 0,\n \"pc\": 0\n },{\n \"channel ix\": 3,\n \"tc\": 0,\n \"txq ix\": 3,\n \"sqn\": 4322,\n \"HW state\": 1,\n \"stopped\": false,\n \"cc\": 0,\n \"pc\": 0\n } ]\n }\n \n Signed-off-by: Aya Levin \u003cayal@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit dd921fd24179e51fc8d8d7bd7978f369da5ba34a\nAuthor: Aya Levin \u003cayal@mellanox.com\u003e\nDate: Mon Jun 24 21:41:21 2019 +0300\n\n net/mlx5e: Extend tx diagnose function\n \n The following patches in the set enhance the diagnostics info of tx\n reporter. Therefore, it is better to pass a pointer to the SQ for\n further data extraction.\n \n Signed-off-by: Aya Levin \u003cayal@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit c50de4af1d635fab3a5c8bd358f55623c01f7ee5\nAuthor: Aya Levin \u003cayal@mellanox.com\u003e\nDate: Mon Jul 1 15:08:13 2019 +0300\n\n net/mlx5e: Generalize tx reporter\u0027s functionality\n \n Prepare for code sharing with rx reporter, which is added in the\n following patches in the set. Introduce a generic error_ctx for\n agnostic recovery despatch.\n \n Signed-off-by: Aya Levin \u003cayal@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 06293ae4fa0a1b62bf3bb8add8f9bbe8815b0aba\nAuthor: Aya Levin \u003cayal@mellanox.com\u003e\nDate: Mon Jul 1 15:51:51 2019 +0300\n\n net/mlx5e: Change naming convention for reporter\u0027s functions\n \n Change from mlx5e_tx_reporter_* to mlx5e_reporter_tx_*. In the following\n patches in the set rx reporter is added, the new naming convention is\n more uniformed.\n \n Signed-off-by: Aya Levin \u003cayal@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 4edc17fdfdf15c2971d15cbfa4d6f2f5f537ee5e\nAuthor: Aya Levin \u003cayal@mellanox.com\u003e\nDate: Mon Jul 1 14:53:34 2019 +0300\n\n net/mlx5e: Rename reporter header file\n \n Rename reporter.h -\u003e health.h so patches in the set can use it for\n health related functionality.\n \n Signed-off-by: Aya Levin \u003cayal@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 0bee3b601b77dbe7981b5474ae8758d6bf60177a\nAuthor: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\nDate: Tue Aug 20 15:12:23 2019 +0200\n\n hrtimer: Improve comments on handling priority inversion against softirq kthread\n \n The handling of a priority inversion between timer cancelling and a a not\n well defined possible preemption of softirq kthread is not very clear.\n \n Especially in the posix timers side it\u0027s unclear why there is a specific RT\n wait callback.\n \n All the nice explanations can be found in the initial changelog of\n f61eff83cec9 (hrtimer: Prepare support for PREEMPT_RT\").\n \n Extract the detailed informations from there and put it into comments.\n \n Signed-off-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190820132656.GC2093@lenoir\n\ncommit ec8f954a40da8cd3d159713b608e901f0cd909a9\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Fri Aug 2 07:35:59 2019 +0200\n\n posix-timers: Use a callback for cancel synchronization on PREEMPT_RT\n \n Posix timer delete retry loops are affected by the same priority inversion\n and live lock issues as the other timers.\n \n Provide a RT specific synchronization function which keeps a reference to\n the timer by holding rcu read lock to prevent the timer from being freed,\n dropping the timer lock and invoking the timer specific wait function via a\n new callback.\n \n This does not yet cover posix CPU timers because they need more special\n treatment on PREEMPT_RT.\n \n [ This is folded into the original attempt which did not use a callback. ]\n \n Originally-by: Anna-Maria Gleixenr \u003canna-maria@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190819143801.656864506@linutronix.de\n\ncommit 265a38d4611360ae3d5bb612d586a3126507a954\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:44:04 2019 -0400\n\n xprtrdma: Simplify rpcrdma_mr_pop\n \n Clean up: rpcrdma_mr_pop call sites check if the list is empty\n first. Let\u0027s replace the list_empty with less costly logic.\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit 4eaceea3a00f8e936a7f48dcd0c975a57f88930f\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Sat Aug 17 02:38:37 2019 -0700\n\n drm/i915: Fix DP-MST crtc_mask\n \n Each fake MST encoder is tied to a specific pipe. Fix the encoder\u0027s\n crtc_mask to reflect that fact.\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Reviewed-by: Dhinakaran Pandiyan \u003cdhinakaran.pandiyan@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190817093902.2171-16-lucas.demarchi@intel.com\n\ncommit d318cedf81e21579992104adce9c9c7e97ed7a1d\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Sat Aug 17 02:38:26 2019 -0700\n\n drm/i915/tgl: update DMC firmware to 2.04\n \n 2 important fixes:\n - vblank counter is now working\n - PSR1 is working\n \n Cc: Jose Souza \u003cjose.souza@intel.com\u003e\n Cc: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190817093902.2171-5-lucas.demarchi@intel.com\n\ncommit 9288d74f3f58bdf778606a3eee67df89305752f4\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Sat Aug 17 02:38:25 2019 -0700\n\n drm/i915/tgl: Move transcoders to pipes\u0027 powerwells\n \n When trying to read registers from transcoder C and D while PG3 is ON it\n causes unclaimed access warnings. Adding the powerwells for the pipes\n fixes the issue, but doesn\u0027t match the spec.\n \n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190817093902.2171-4-lucas.demarchi@intel.com\n\ncommit 465242ee36ebd5a4f31df64692c634f402c8f2d9\nAuthor: Michel Thierry \u003cmichel.thierry@intel.com\u003e\nDate: Sat Aug 17 02:38:24 2019 -0700\n\n drm/i915/tgl: add support for reading the timestamp frequency\n \n There are no changes with respect to GEN11, which Paulo wrote.\n \n This gets rid of the \"Missing switch case in read_timestamp_frequency\"\n message at boot for Tiger Lake.\n \n [ Lucas: BSpec: 10742 and 9024, but there\u0027s a mismatch on the values.\n Let\u0027s say a glitch in the spec. Tested locally and it works. ]\n \n Cc: Paulo Zanoni \u003cpaulo.r.zanoni@intel.com\u003e\n Cc: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Signed-off-by: Michel Thierry \u003cmichel.thierry@intel.com\u003e\n Reviewed-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190817093902.2171-3-lucas.demarchi@intel.com\n\ncommit 924068e50a6ca426b8fda1d2c06d6d2bde212136\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Mon Aug 19 00:00:40 2019 +0200\n\n rtc: class: add debug message when registration fails\n \n Add a message when __rtc_register_device fails because rtc-\u003eops is NULL.\n This may only happen when developing a new driver so use dev_dbg to avoid\n compiling it in by default.\n \n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190818220041.17833-1-alexandre.belloni@bootlin.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit cb36cf803f3f468baa22b165bc4e67ae88811ad1\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Mon Aug 19 20:26:56 2019 +0200\n\n rtc: pcf2123: add proper compatible string\n \n nxp,rtc-pcf2123 is not a proper compatible strong for this RTC. The part\n name is only pcf2123 and is less confusing.\n \n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190819182656.29744-9-alexandre.belloni@bootlin.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit d5b626e1350344db7fb1bf68f6089aaceb7df79a\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Mon Aug 19 20:26:55 2019 +0200\n\n rtc: pcf2123: let the core handle range offsetting\n \n Set the RTC range properly and use the core windowing and offsetting to\n (unfortunately) map back to a 1970-2069 range.\n \n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190819182656.29744-8-alexandre.belloni@bootlin.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 935a7f45979067e98268546854bd7ef874207a60\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Mon Aug 19 20:26:54 2019 +0200\n\n rtc: pcf2123: convert to devm_rtc_allocate_device\n \n This allows further improvement of the driver. Also remove the unecessary\n error string as the core will already display error messages.\n \n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190819182656.29744-7-alexandre.belloni@bootlin.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 9a5aeaad73ece38c77bc4c955ab3c9a4c5e3277f\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Mon Aug 19 20:26:53 2019 +0200\n\n rtc: pcf2123: remove useless error path goto\n \n kfree_exit only returns ret, remove it. This also fixes the\n devm_regmap_init_spi error case where the probe wouldn\u0027t actually fail\n because ret is initialized to 0.\n \n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190819182656.29744-6-alexandre.belloni@bootlin.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 9126a2b16b678fd14fa2171e2f05395ecb229a2c\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Mon Aug 19 20:26:52 2019 +0200\n\n rtc: pcf2123: rename struct and variables\n \n Rename struct pcf2123_plat_data to struct pcf2123_data and pdata to\n pcf2123 to make the driver use a more common subsystem style.\n \n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190819182656.29744-5-alexandre.belloni@bootlin.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit d3bad6026f0bd34df11aedb1b41c432df8005eda\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Mon Aug 19 20:26:51 2019 +0200\n\n rtc: pcf2123: stop using dev.platform_data\n \n platform_data is for platform specific data, use driver_data instead.\n \n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190819182656.29744-4-alexandre.belloni@bootlin.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 577f648207e0ede8153997c7195d4cb31268f410\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Mon Aug 19 20:26:50 2019 +0200\n\n rtc: pcf2123: implement .alarm_irq_enable\n \n Implement .alarm_irq_enable so it is possible to use RTC_ALM_SET,\n RTC_AIE_ON and RTC_AIE_OFF.\n \n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190819182656.29744-3-alexandre.belloni@bootlin.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit d0ce6ef7146692a1540b4dd3a68970da6bcd5207\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Mon Aug 19 20:26:49 2019 +0200\n\n rtc; pcf2123: fix possible alarm race condition\n \n Clear the flag after disabling the alarm to ensure the alarm doesn\u0027t fire\n between clearing the flag and disabling the alarm.\n \n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190819182656.29744-2-alexandre.belloni@bootlin.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 5bdf40dab622d301757f81325d9b80902b3bbae7\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Mon Aug 19 20:26:48 2019 +0200\n\n rtc: pcf2123: don\u0027t use weekday alarm\n \n The week day may not be set properly by userspace. This would result is\n missed alarms. Disable alarm matching on weekday.\n \n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190819182656.29744-1-alexandre.belloni@bootlin.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit ea6591b4dadba6f4cac8a7964ce8181770f035e1\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Wed Aug 14 16:55:17 2019 -0700\n\n drm/i915/tgl: disable DDIC\n \n The current SKUs added for Tiger Lake don\u0027t have DDIC hooked up, even\n though it is supported by the SoC. The current state for these SKUs is\n problematic since while enabling the combo phy, PORT_COMP_DW* return\n 0xFFFFFFFF, which is invalid per register definition.\n \n During initialization we check what phys are not yet enabled by reading\n PHY_MISC_C and try to enable it by toggling the \"DE to IO Comp Pwr Down\"\n bit. But after that any read to the PORT_COMP_DW* returns invalid\n results. This removes the following warning\n \n [56997.634353] Missing case (val \u003d\u003d 4294967295)\n [56997.639241] WARNING: CPU: 5 PID: 768 at drivers/gpu/drm/i915/display/intel_combo_phy.c:54 cnl_get_procmon_ref_values+0xc9/0xf0 [i915]\n [56997.639808] Modules linked in: i915(+) prime_numbers x86_pkg_temp_thermal coretemp kvm_intel kvm irqbypass crct10dif_pclmul crc32_pclmul ghash_clmulni_intel e1000e [last unloaded: prime_numbers]\n [56997.639808] CPU: 5 PID: 768 Comm: insmod Tainted: G U W 5.2.0-demarchi+ #65\n [56997.639808] Hardware name: Intel Corporation Tiger Lake Client Platform/TigerLake U DDR4 SODIMM RVP, BIOS TGLSFWI1.R00.2252.A03.1906270154 06/27/2019\n [56997.639808] RIP: 0010:cnl_get_procmon_ref_values+0xc9/0xf0 [i915]\n [56997.639808] Code: 2c a0 85 c9 74 e0 81 f9 00 00 00 01 75 09 48 c7 c0 0c a4 2c a0 eb cf 48 c7 c6 3c 3a 31 a0 48 c7 c7 40 3a 31 a0 e8 6b 4d ea e0 \u003c0f\u003e 0b 48 c7 c0 00 a4 2c a0 eb b1 48 c7 c0 24 a4 2\n c a0 eb a8 e8 be\n [56997.639808] RSP: 0018:ffffc9000068f8a8 EFLAGS: 00010286\n [56997.639808] RAX: 0000000000000000 RBX: ffff88848fa90000 RCX: 0000000000000000\n [56997.639808] RDX: ffff8884a08b5ef8 RSI: ffff8884a08a6658 RDI: 00000000ffffffff\n [56997.639808] RBP: 0000000000000002 R08: 0000000000000000 R09: 0000000000000000\n [56997.639808] R10: 0000000000000000 R11: 0000000000000000 R12: ffff88848fa90000\n [56997.639808] R13: 0000000000000000 R14: 0000000000000002 R15: 0006c00000162000\n [56997.639808] FS: 00007f61ca3d12c0(0000) GS:ffff8884a0880000(0000) knlGS:0000000000000000\n [56997.639808] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n [56997.639808] CR2: 00007f71be6a92c0 CR3: 0000000494750006 CR4: 0000000000760ee0\n [56997.639808] PKRU: 55555554\n [56997.639808] Call Trace:\n [56997.639808] cnl_verify_procmon_ref_values+0x36/0xf0 [i915]\n [56997.639808] ? rcu_read_lock_sched_held+0x6f/0x80\n [56997.639808] ? gen11_fwtable_read32+0x257/0x290 [i915]\n [56997.639808] icl_combo_phy_verify_state.part.0+0x22/0xa0 [i915]\n [56997.639808] intel_combo_phy_init+0x17e/0x3e0 [i915]\n [56997.639808] ? icl_display_core_init+0x2c/0x1a0 [i915]\n [56997.639808] ? _raw_spin_unlock_irqrestore+0x4c/0x60\n [56997.639808] icl_display_core_init+0x34/0x1a0 [i915]\n [56997.639808] intel_power_domains_init_hw+0x200/0x570 [i915]\n [56997.639808] i915_driver_probe+0x103b/0x17e0 [i915]\n [56997.639808] ? printk+0x53/0x6a\n [56997.639808] i915_pci_probe+0x3b/0x190 [i915]\n \n We may or may not need to change the implementation to account for DDIC\n being available on other SKUs. For now I think the best thing to do is\n to just disable the port.\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190814235517.10032-1-lucas.demarchi@intel.com\n\ncommit fc0bc0190bc5ea1e44317c84c9f92f9196a7441b\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Mon Aug 19 16:00:53 2019 -0400\n\n net: dsa: mv88e6xxx: wrap SERDES IRQ in power function\n \n Now that mv88e6xxx_serdes_power is only called after driver setup,\n we can wrap the SERDES IRQ code directly within it for clarity.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b759f528ca3dea889ab985265be46715c3586eef\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Mon Aug 19 16:00:52 2019 -0400\n\n net: dsa: mv88e6xxx: enable SERDES after setup\n \n SERDES is powered on for CPU and DSA ports and powered down for unused\n ports at setup time. But now that DSA calls mv88e6xxx_port_enable\n and mv88e6xxx_port_disable for all ports, the SERDES power can now\n be handled after setup inconditionally for all ports.\n \n Using the port enable and disable callbacks also have the benefit to\n handle the SERDES IRQ for non user ports as well.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 3903f315165dff72796e46d177aaf1cbd67aa07d\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Mon Aug 19 16:00:51 2019 -0400\n\n net: dsa: mv88e6xxx: do not change STP state on port disabling\n \n When disabling a port, that is not for the driver to decide what to\n do with the STP state. This is already handled by the DSA layer.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 0394a63acfe2a6e1c08af0eb1a9133ee8650d7bd\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Mon Aug 19 16:00:50 2019 -0400\n\n net: dsa: enable and disable all ports\n \n Call the .port_enable and .port_disable functions for all ports,\n not only the user ports, so that drivers may optimize the power\n consumption of all ports after a successful setup.\n \n Unused ports are now disabled on setup. CPU and DSA ports are now\n enabled on setup and disabled on teardown. User ports were already\n enabled at slave creation and disabled at slave destruction.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Reviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 74be4babe72fd1ed1bba6b52d0bdc0d1e13f7af8\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Mon Aug 19 16:00:49 2019 -0400\n\n net: dsa: do not enable or disable non user ports\n \n The .port_enable and .port_disable operations are currently only\n called for user ports, hence assuming they have a slave device. In\n preparation for using these operations for other port types as well,\n simply guard all implementations against non user ports and return\n directly in such case.\n \n Note that bcm_sf2_sw_suspend() currently calls bcm_sf2_port_disable()\n (and thus b53_disable_port()) against the user and CPU ports, so do\n not guards those functions. They will be called for unused ports in\n the future, but that was expected by those drivers anyway.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 955222ca5281ee7ded6b899605c055b147a15c73\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Mon Aug 19 16:00:48 2019 -0400\n\n net: dsa: use a single switch statement for port setup\n \n It is currently difficult to read the different steps involved in the\n setup and teardown of ports in the DSA code. Keep it simple with a\n single switch statement for each port type: UNUSED, CPU, DSA, or USER.\n \n Also no need to call devlink_port_unregister from within dsa_port_setup\n as this step is inconditionally handled by dsa_port_teardown on error.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d82dd83df27e31da3f2a838b32d0b82940dcd504\nAuthor: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\nDate: Thu Jul 25 01:55:30 2019 -0700\n\n ice: Restructure VFs initialization flows\n \n This patch restructures how VFs are configured, and resources allocated.\n Instead of freeing resources that were never allocated, and resetting\n empty VFs that have never been created - the new flow will just allocate\n resources for number of requested VFs based on the availability.\n \n During VFs initialization process, global interrupt is disabled, and\n rearmed after getting MSIX vectors for VFs. This allows immediate mailbox\n communications, instead of delaying it till later and VFs.\n PF communications resulted to using polling instead of actual interrupt.\n The issue manifested when creating higher number of VFs (128 VFs) per PF.\n \n Signed-off-by: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 9118fcd5255f6d3891c90e01edd98dfc402ac663\nAuthor: Brett Creeley \u003cbrett.creeley@intel.com\u003e\nDate: Thu Jul 25 01:55:29 2019 -0700\n\n ice: Assume that more than one Rx queue is rare in ice_napi_poll\n \n Currently we divide budget by the number of Rx queues per Rx ring\n container in ice_napi_poll even if there is only 1. This is an\n unnecessary divide for the normal case of 1 Rx ring per Rx ring\n container. Fix this by using an unlikely() call in the case where we\n actually need to divide.\n \n Also, we will always set budget_per_ring even if there are no Rx rings\n in the Rx ring container so we don\u0027t need to initialize it to 0.\n \n Signed-off-by: Brett Creeley \u003cbrett.creeley@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit c1ddf1f5c4adc1e885dce8533b2245d979f38280\nAuthor: Brett Creeley \u003cbrett.creeley@intel.com\u003e\nDate: Thu Jul 25 01:55:28 2019 -0700\n\n ice: Use the software based tail when checking for hung Tx ring\n \n Currently in ice_get_tx_pending we try to read a Tx ring\u0027s tail. This is\n then compared with the software based head (next_to_clean) to determine\n if we have pending work. This will never work because reading of the Tx\n ring\u0027s tail is no longer supported. Fix this by using the software based\n tail (next_to_use) to determine if there is pending work.\n \n Signed-off-by: Brett Creeley \u003cbrett.creeley@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit effdd8a6b6ecb30b088f36ed0a1fb67cdc7610a9\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Tue Aug 20 16:40:51 2019 +0200\n\n dt-bindings: arm: amlogic: add SEI Robotics SEI610 bindings\n \n Add the compatible for the Amlogic SM1 Based SEI610 board.\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 2fd10e6d4eb77796e21aa1b1493b24198f660227\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Tue Aug 20 16:40:50 2019 +0200\n\n dt-bindings: arm: amlogic: add SM1 bindings\n \n Add bindings for the new Amlogic SM1 SoC Family.\n \n It a derivative of the G12A SoC Family with :\n - Cortex-A55 core instead of A53\n - more power domains\n - a neural network co-processor\n - a CSI input and image processor\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 623699711a39134a2b2cd6bdba81c5167deceec8\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Tue Aug 20 16:40:49 2019 +0200\n\n soc: amlogic: clk-measure: Add support for SM1\n \n Add the clk-measurer clocks IDs for the Amlogic SM1 SoC family.\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit b555cafb766a10846d228346aa71e517a5393c95\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Tue Aug 20 16:40:48 2019 +0200\n\n dt-bindings: soc: amlogic: clk-measure: Add SM1 compatible\n \n Add the Amlogic SM1 Compatible for the clk-measurer IP.\n \n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit d2187f8e445403b7aeb08e64c1528761154e9ab3\nAuthor: Hayes Wang \u003chayeswang@realtek.com\u003e\nDate: Mon Aug 19 14:40:36 2019 +0800\n\n r8152: divide the tx and rx bottom functions\n \n Move the tx bottom function from NAPI to a new tasklet. Then, for\n multi-cores, the bottom functions of tx and rx may be run at same\n time with different cores. This is used to improve performance.\n \n On x86, Tx/Rx 943/943 Mbits/sec -\u003e 945/944.\n For arm platform, Tx/Rx: 917/917 Mbits/sec -\u003e 933/933.\n \n Signed-off-by: Hayes Wang \u003chayeswang@realtek.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit aaee0c1ffd6399d291b030b49d622b81dd5071c5\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Tue Aug 13 14:45:13 2019 -0600\n\n PCI/IOV: Move sysfs SR-IOV functions to iov.c\n \n The sysfs SR-IOV functions are only needed when the kernel is built with\n SR-IOV support. Rather than put them in pci-sysfs.c under #ifdef\n CONFIG_PCI_IOV, move them to iov.c, which is only compiled when\n CONFIG_PCI_IOV\u003dy.\n \n Link: https://lore.kernel.org/r/20190813204513.4790-4-skunberg.kelsey@gmail.com\n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Donald Dutile \u003cddutile@redhat.com\u003e\n Reviewed-by: Kuppuswamy Sathyanarayanan \u003csathyanarayanan.kuppuswamy@linux.intel.com\u003e\n\ncommit e2154044dd4168bc25c70170dfa6179b57f63914\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Tue Aug 13 14:45:12 2019 -0600\n\n PCI: sysfs: Change permissions from symbolic to octal\n \n We prefer octal permissions over symbolic permissions such as \"(S_IWUSR |\n S_IWGRP)\". Change all symbolic permissions to octal permissions, e.g.,\n \n - (S_IWUSR | S_IWGRP)\n + 0220\n \n Link: https://lore.kernel.org/r/20190813204513.4790-3-skunberg.kelsey@gmail.com\n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Donald Dutile \u003cddutile@redhat.com\u003e\n\ncommit 4e2b79436e4f22b225ee445832705bb7c4675807\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Thu Aug 15 09:33:52 2019 -0600\n\n PCI: sysfs: Change DEVICE_ATTR() to DEVICE_ATTR_WO()\n \n DEVICE_ATTR() should only be used when files have unusual permissions.\n Change DEVICE_ATTR() with \u00270220\u0027 write-only permissions to\n DEVICE_ATTR_WO(), e.g.,\n \n - static DEVICE_ATTR(_name, (S_IWUSR | S_IWGRP), NULL, _store);\n + static DEVICE_ATTR_WO(_name);\n \n Since _store is no longer passed, make the _name passed by DEVICE_ATTR_WO()\n and the related _name##_store() name match with each other, e.g.,\n \n DEVICE_ATTR_WO(bus_rescan) must be able to call bus_rescan_store()\n \n Link: https://lore.kernel.org/r/20190815153352.86143-4-skunberg.kelsey@gmail.com\n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Donald Dutile \u003cddutile@redhat.com\u003e\n\ncommit 8bdfa145f58220796abddc93b5038efa6e7e90fc\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Tue Aug 13 14:45:11 2019 -0600\n\n PCI: sysfs: Define device attributes with DEVICE_ATTR*()\n \n Device attributes should be defined using DEVICE_ATTR*(_name, _mode, _show,\n _store). Convert them all from __ATTR*() to DEVICE_ATTR*(), e.g.,\n \n - struct device_attribute dev_attr_##_name \u003d __ATTR(_name, _mode, _show,\n _store)\n + static DEVICE_ATTR(foo, S_IWUSR | S_IRUGO, show_foo, store_foo)\n \n Link: https://lore.kernel.org/r/20190813204513.4790-2-skunberg.kelsey@gmail.com\n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Donald Dutile \u003cddutile@redhat.com\u003e\n\ncommit e95656ea15e54d4e6a192d560d84008b53fc1eb5\nAuthor: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nDate: Tue Apr 18 17:28:30 2017 -0700\n\n Input: add support for polling to input devices\n \n Separating \"normal\" and \"polled\" input devices was a mistake, as often we\n want to allow the very same device work on both interrupt-driven and\n polled mode, depending on the board on which the device is used.\n \n This introduces new APIs:\n \n - input_setup_polling\n - input_set_poll_interval\n - input_set_min_poll_interval\n - input_set_max_poll_interval\n \n These new APIs allow switching an input device into polled mode with sysfs\n attributes matching drivers using input_polled_dev APIs that will be\n eventually removed.\n \n Tested-by: Michal Vokáč \u003cmichal.vokac@ysoft.com\u003e\n Acked-by: Benjamin Tissoires \u003cbenjamin.tissoires@redhat.com\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 2da9d2b5b9d64cf5228dd3fb4d1220ff3fec8440\nAuthor: Jason Gerecke \u003ckillertofu@gmail.com\u003e\nDate: Fri Aug 16 15:07:15 2019 -0700\n\n Input: wacom_w8001 - allocate additional space for \u0027phys\u0027\n \n GCC warns that the output of our call to \u0027snprintf\u0027 in \u0027w8001_connect\u0027\n may be truncated since both \u0027serio-\u003ephys\u0027 and \u0027w8001-\u003ephys\u0027 are 32 bytes\n in length. Increase the amount of space allocated for the latter to\n compensate.\n \n Signed-off-by: Jason Gerecke \u003cjason.gerecke@wacom.com\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit e9e006f5fcf2bab59149cb38a48a4817c1b538b4\nAuthor: Mike Christie \u003cmchristi@redhat.com\u003e\nDate: Sun Aug 4 14:10:06 2019 -0500\n\n nbd: fix max number of supported devs\n \n This fixes a bug added in 4.10 with commit:\n \n commit 9561a7ade0c205bc2ee035a2ac880478dcc1a024\n Author: Josef Bacik \u003cjbacik@fb.com\u003e\n Date: Tue Nov 22 14:04:40 2016 -0500\n \n nbd: add multi-connection support\n \n that limited the number of devices to 256. Before the patch we could\n create 1000s of devices, but the patch switched us from using our\n own thread to using a work queue which has a default limit of 256\n active works.\n \n The problem is that our recv_work function sits in a loop until\n disconnection but only handles IO for one connection. The work is\n started when the connection is started/restarted, but if we end up\n creating 257 or more connections, the queue_work call just queues\n connection257+\u0027s recv_work and that waits for connection 1 - 256\u0027s\n recv_work to be disconnected and that work instance completing.\n \n Instead of reverting back to kthreads, this has us allocate a\n workqueue_struct per device, so we can block in the work.\n \n Cc: stable@vger.kernel.org\n Reviewed-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: Mike Christie \u003cmchristi@redhat.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 2da22da573481cc4837e246d0eee4d518b3f715e\nAuthor: Mike Christie \u003cmchristi@redhat.com\u003e\nDate: Tue Aug 13 11:39:52 2019 -0500\n\n nbd: fix zero cmd timeout handling v2\n \n This fixes a regression added in 4.9 with commit:\n \n commit 0eadf37afc2500e1162c9040ec26a705b9af8d47\n Author: Josef Bacik \u003cjbacik@fb.com\u003e\n Date: Thu Sep 8 12:33:40 2016 -0700\n \n nbd: allow block mq to deal with timeouts\n \n where before the patch userspace would set the timeout to 0 to disable\n it. With the above patch, a zero timeout tells the block layer to use\n the default value of 30 seconds. For setups where commands can take a\n long time or experience transient issues like network disruptions this\n then results in IO errors being sent to the application.\n \n To fix this, the patch still uses the common block layer timeout\n framework, but if zero is set, nbd just logs a message and then resets\n the timer when it expires.\n \n Reviewed-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: Mike Christie \u003cmchristi@redhat.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 887e975c4172d0d5670c39ead2f18ba1e4ec8133\nAuthor: Mike Christie \u003cmchristi@redhat.com\u003e\nDate: Tue Aug 13 11:39:51 2019 -0500\n\n nbd: add missing config put\n \n Fix bug added with the patch:\n \n commit 8f3ea35929a0806ad1397db99a89ffee0140822a\n Author: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Date: Mon Jul 16 12:11:35 2018 -0400\n \n nbd: handle unexpected replies better\n \n where if the timeout handler runs when the completion path is and we fail\n to grab the mutex in the timeout handler we will leave a config reference\n and cannot free the config later.\n \n Reviewed-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: Mike Christie \u003cmchristi@redhat.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 00514677ed829efa2fc688ca60a72e52695529cc\nAuthor: Mike Christie \u003cmchristi@redhat.com\u003e\nDate: Tue Aug 13 11:39:50 2019 -0500\n\n nbd: add function to convert blk req op to nbd cmd\n \n This adds a helper function to convert a block req op to a nbd cmd type.\n It will be used in the last patch to log the type in the timeout\n handler.\n \n Reviewed-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: Mike Christie \u003cmchristi@redhat.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 55313e92bd17a872b99cfb5f21dd6d3ff1999bce\nAuthor: Mike Christie \u003cmchristi@redhat.com\u003e\nDate: Tue Aug 13 11:39:49 2019 -0500\n\n nbd: add set cmd timeout helper\n \n Add a helper to set the cmd timeout. It does not really do a lot now,\n but will be more useful in the next patches.\n \n Reviewed-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\n Signed-off-by: Mike Christie \u003cmchristi@redhat.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 60e2dde1e91ae0addb21ac380cc36ebee7534e49\nAuthor: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nDate: Mon Aug 19 15:41:42 2019 -0500\n\n led: triggers: Fix a memory leak bug\n \n In led_trigger_set(), \u0027event\u0027 is allocated in kasprintf(). However, it is\n not deallocated in the following execution if the label \u0027err_activate\u0027 or\n \u0027err_add_groups\u0027 is entered, leading to memory leaks. To fix this issue,\n free \u0027event\u0027 before returning the error.\n \n Fixes: 52c47742f79d (\"leds: triggers: send uevent when changing triggers\")\n Signed-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 395790566eec37706dedeb94779045adc3a7581e\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:43:17 2019 -0400\n\n xprtrdma: Toggle XPRT_CONGESTED in xprtrdma\u0027s slot methods\n \n Commit 48be539dd44a (\"xprtrdma: Introduce -\u003ealloc_slot call-out for\n xprtrdma\") added a separate alloc_slot and free_slot to the RPC/RDMA\n transport. Later, commit 75891f502f5f (\"SUNRPC: Support for\n congestion control when queuing is enabled\") modified the generic\n alloc/free_slot methods, but neglected the methods in xprtrdma.\n \n Found via code review.\n \n Fixes: 75891f502f5f (\"SUNRPC: Support for congestion control ... \")\n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit eed48a9c161588544999ee8d451392921d846ee8\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:42:31 2019 -0400\n\n xprtrdma: Rename rpcrdma_buffer::rb_all\n \n Clean up: There are other \"all\" list heads. For code clarity\n distinguish this one as for use only for MRs by renaming it.\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit 2dfdcd88cf0ea66eec0478de82283ef20eb6f421\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:41:44 2019 -0400\n\n xprtrdma: Rename CQE field in Receive trace points\n \n Make the field name the same for all trace points that handle\n pointers to struct rpcrdma_rep. That makes it easy to grep for\n matching rep points in trace output.\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit c9cff337eab394c4dc8b128dde7308a1dd2e653a\nAuthor: Codrin Ciubotariu \u003ccodrin.ciubotariu@microchip.com\u003e\nDate: Tue Aug 20 19:24:11 2019 +0300\n\n ASoC: mchp-i2s-mcc: Fix simultaneous capture and playback in master mode\n \n This controller supports capture and playback running at the same time,\n with the limitation that both capture and playback must be configured the\n same way (sample rate, sample format, number of channels, etc). For this,\n we have to assure that the configuration registers look the same when\n capture and playback are initiated.\n This patch fixes a bug in which the controller is in master mode and the\n hw_params() callback fails for the second audio stream. The fail occurs\n because the divisors are calculated after comparing the configuration\n registers for capture and playback. The fix consists in calculating the\n divisors before comparing the configuration registers. BCLK and LRC are\n then configured and started only if the controller is not already running.\n \n Fixes: 7e0cdf545a55 (\"ASoC: mchp-i2s-mcc: add driver for I2SC Multi-Channel Controller\")\n Signed-off-by: Codrin Ciubotariu \u003ccodrin.ciubotariu@microchip.com\u003e\n Link: https://lore.kernel.org/r/20190820162411.24836-4-codrin.ciubotariu@microchip.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 0f6fc97501b790c971b11b52a654009d21c45238\nAuthor: Codrin Ciubotariu \u003ccodrin.ciubotariu@microchip.com\u003e\nDate: Tue Aug 20 19:24:10 2019 +0300\n\n ASoC: mchp-i2s-mcc: Wait for RX/TX RDY only if controller is running\n \n Since hw_free() can be called multiple times and not just after a stop\n trigger command, we should check whether the RX or TX ready interrupt was\n truly enabled previously. For this, we assure that the condition of the\n wait event is always true, except when RX/TX interrupts are enabled.\n \n Fixes: 7e0cdf545a55 (\"ASoC: mchp-i2s-mcc: add driver for I2SC Multi-Channel Controller\")\n Signed-off-by: Codrin Ciubotariu \u003ccodrin.ciubotariu@microchip.com\u003e\n Link: https://lore.kernel.org/r/20190820162411.24836-3-codrin.ciubotariu@microchip.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 988b59467b2b14523a266957affbe9eca3e99fc9\nAuthor: Codrin Ciubotariu \u003ccodrin.ciubotariu@microchip.com\u003e\nDate: Tue Aug 20 19:24:09 2019 +0300\n\n ASoC: mchp-i2s-mcc: Fix unprepare of GCLK\n \n If hw_free() gets called after hw_params(), GCLK remains prepared,\n preventing further use of it. This patch fixes this by unpreparing the\n clock in hw_free() or if hw_params() gets an error.\n \n Fixes: 7e0cdf545a55 (\"ASoC: mchp-i2s-mcc: add driver for I2SC Multi-Channel Controller\")\n Signed-off-by: Codrin Ciubotariu \u003ccodrin.ciubotariu@microchip.com\u003e\n Link: https://lore.kernel.org/r/20190820162411.24836-2-codrin.ciubotariu@microchip.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit bbf9a127abca4aac5cc75f882bc7efcc398e86ae\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 21:25:26 2019 +0200\n\n ASoC: sun4i-i2s: Support more channels\n \n We\u0027ve been limited to 2 channels in the driver while the controller\n supports from 1 to 8 channels, in both capture and playback. let\u0027s remove\n the hardcoded checks and numbers, and extend the range of channel numbers\n we can use.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/27d9de5cd56f3a544851b8cd8af08bf836d19637.1566242458.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 0083a507a78fdfa868acc0709408b59e72488a61\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 21:25:25 2019 +0200\n\n ASoC: sun4i-i2s: Pass the channels number as an argument\n \n The channels number have been hardcoded to 2 so far, while the controller\n supports more than that.\n \n Remove the instance where it has been hardcoded to compute the BCLK\n divider, and pass it through as an argument to ease further support of more\n channels.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/48887cf7abfaab6597db233b24d7a088a913e48a.1566242458.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 3e9acd7ac6933cdc20c441bbf9a38ed9e42e1490\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 21:25:24 2019 +0200\n\n ASoC: sun4i-i2s: Remove duplicated quirks structure\n \n The A83t and H3 have the same quirks, so it doesn\u0027t make sense to duplicate\n the quirks structure.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/5ade5de27d23918c5ef30387c23aead951d5ad64.1566242458.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 69e450e50ca6dde566f3ac3f2c329fb0492441ef\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 21:25:23 2019 +0200\n\n ASoC: sun4i-i2s: Fix the LRCK period on A83t\n \n Unlike the previous SoCs, the A83t, like the newer ones, need the LRCK\n bitfield to be set. Let\u0027s add it.\n \n Fixes: 21faaea1343f (\"ASoC: sun4i-i2s: Add support for A83T\")\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/6a0ee0bc1375bcb53840d3fb2d2f3d9732b8e57e.1566242458.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit bf943d527987c38f6fb11f9515e0cf2839286eb8\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 21:25:22 2019 +0200\n\n ASoC: sun4i-i2s: Fix MCLK Enable bit offset on A83t\n \n The A83t, unlike previous SoCs, has the MCLK enable bit at the 8th bit of\n the CLK_DIV register, unlike what is declared in the driver.\n \n Fixes: 21faaea1343f (\"ASoC: sun4i-i2s: Add support for A83T\")\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/43b07f8cd8e0e280c64ce61d57c307678c923e9b.1566242458.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 2e04fc4dbf50195262aa5a2ae6d35baa5b598cae\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 21:25:21 2019 +0200\n\n ASoC: sun4i-i2s: Fix WSS and SR fields for the A83t\n \n The A83t has the same bit fields offsets than the A10 and A31, while this\n was the first device with the new layout, fix that.\n \n Fixes: 21faaea1343f (\"ASoC: sun4i-i2s: Add support for A83T\")\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/d93f0943cc39d880750daf459a0eeab34c63518e.1566242458.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit dd657eae8164f7e4bafe8b875031a7c6c50646a9\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 21:25:20 2019 +0200\n\n ASoC: sun4i-i2s: Fix the LRCK polarity\n \n The LRCK polarity \"normal\" polarity in the I2S/TDM specs and in the\n Allwinner datasheet are not the same. In the case where the i2s controller\n is being used as the LRCK master, it\u0027s pretty clear when looked at under a\n scope.\n \n Let\u0027s fix this, and add a comment to clear up as much the confusion as\n possible.\n \n Fixes: 7d2993811a1e (\"ASoC: sun4i-i2s: Add support for H3\")\n Fixes: 21faaea1343f (\"ASoC: sun4i-i2s: Add support for A83T\")\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/e03fb6b2a916223070b9f18405b0ef117a452ff4.1566242458.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 515fcfbc773632e160f4b94e8df8d278a8d704f7\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 21:25:19 2019 +0200\n\n ASoC: sun4i-i2s: Fix LRCK and BCLK polarity offsets on newer SoCs\n \n The LRCK and BCLK polarity offsets on newer SoCs has been\n changed, yet the driver didn\u0027t take it into account for all of them.\n \n This was taken into account for the H3, but not the A83t. This was handled\n using a reg_field for the H3.\n \n However, the value in that field will not be the same, so reg_field is not\n adapted in that case. Let\u0027s change for proper calls with the regular\n values.\n \n Fixes: 21faaea1343f (\"ASoC: sun4i-i2s: Add support for A83T\")\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/9cbdde80a299288878e58225df4d7884e0301348.1566242458.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit cf2c0e1ce9544df42170fb921f12da82dc0cc8d6\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 21:25:17 2019 +0200\n\n ASoC: sun4i-i2s: RX and TX counter registers are swapped\n \n The RX and TX counters registers offset have been swapped, fix that.\n \n Fixes: fa7c0d13cb26 (\"ASoC: sunxi: Add Allwinner A10 Digital Audio driver\")\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/8b26477560ad5fd8f69e037b167c5e61de5c26a3.1566242458.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c1d3a921d72bd21f266ca28c15213fbe78160a4b\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 21:25:16 2019 +0200\n\n ASoC: sun4i-i2s: Fix the MCLK and BCLK dividers on newer SoCs\n \n From: Marcus Cooper \u003ccodekipper@gmail.com\u003e\n \n The clock division dividers have changed between the older (A10/A31) and\n newer (H3, A64, etc) SoCs.\n \n While this was addressed through an offset on some SoCs, it was missing\n some dividers as well, so the support wasn\u0027t perfect. Let\u0027s introduce a\n pointer in the quirk structure for the divider calculation functions to use\n so we can have the proper range now.\n \n Signed-off-by: Marcus Cooper \u003ccodekipper@gmail.com\u003e\n [Maxime: Fix the commit log, use a field in the quirk structure]\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/0e5b4abf06cd3202354315201c6af44caeb20236.1566242458.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit fb19739d7f688142b61d0fca476188c4fd9e937a\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 21:25:15 2019 +0200\n\n ASoC: sun4i-i2s: Use module clock as BCLK parent on newer SoCs\n \n On the first generation of Allwinner SoCs (A10-A31), the i2s controller was\n using the MCLK as BCLK parent. However, this changed since the introduction\n of the A83t and BCLK now uses the module clock as its parent.\n \n Let\u0027s introduce a hook to get the parent rate and use that in our divider\n calculations.\n \n Fixes: 7d2993811a1e (\"ASoC: sun4i-i2s: Add support for H3\")\n Fixes: 21faaea1343f (\"ASoC: sun4i-i2s: Add support for A83T\")\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/0b6665be216b3bd0e7bc43724818f05f3f8ee881.1566242458.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 71137bcd0a9a778f9407a3bee46c62fcccee4f83\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 21:25:12 2019 +0200\n\n ASoC: sun4i-i2s: Move the format configuration to a callback\n \n The two main generations of our I2S controller require a slightly different\n format configuration, mostly because of a quite different register layout\n and some additional registers being needed on the newer generation.\n \n This used to be controlled through a bunch of booleans, however this proved\n to be quite impractical, especially since a bunch of SoCs forgot to set\n those parameters and therefore were broken from that point of view.\n \n Fixes: 21faaea1343f (\"ASoC: sun4i-i2s: Add support for A83T\")\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/dc818644c3e40734e7a97247c994b1fca1c3c047.1566242458.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit d70be625f25af7a2bc91b7d17d205f6071f08f2f\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 21:25:11 2019 +0200\n\n ASoC: sun4i-i2s: Move the channel configuration to a callback\n \n The two main generations of our I2S controller require a slightly different\n channel configuration, mostly because of a quite different register layout\n and some additional registers being needed on the newer generation.\n \n This used to be controlled through a bunch of booleans, however this proved\n to be quite impractical, especially since a bunch of SoCs forgot to set\n those parameters and therefore were broken from that point of view.\n \n Fixes: 21faaea1343f (\"ASoC: sun4i-i2s: Add support for A83T\")\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/6414463de69584e8227fa495b13aa5f4798e1f0e.1566242458.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit b2299e83815c59ab59c4ee4fb4842b3b28e5072f\nAuthor: Leon Romanovsky \u003cleonro@mellanox.com\u003e\nDate: Mon Aug 19 14:45:47 2019 +0300\n\n RDMA: Delete DEBUG code\n \n There is no need to keep DEBUG defines for out-of-the tree testing.\n \n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Link: https://lore.kernel.org/r/20190819114547.20704-1-leon@kernel.org\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit d5e120422db8808e1c8b1507900ca393a877c58f\nAuthor: Daniel Stuart \u003cdaniel.stuart14@gmail.com\u003e\nDate: Thu Aug 15 14:12:55 2019 -0300\n\n ASoC: intel: cht_bsw_max98090_ti: Add all Chromebooks that need pmc_plt_clk_0 quirk\n \n Every single baytrail chromebook sets PMC to 0, as can be seeing\n below by searching through coreboot source code:\n $ grep -rl \"PMC_PLT_CLK\\[0\\]\" .\n ./rambi/variants/glimmer/devicetree.cb\n ./rambi/variants/clapper/devicetree.cb\n ./rambi/variants/swanky/devicetree.cb\n ./rambi/variants/enguarde/devicetree.cb\n ./rambi/variants/winky/devicetree.cb\n ./rambi/variants/kip/devicetree.cb\n ./rambi/variants/squawks/devicetree.cb\n ./rambi/variants/orco/devicetree.cb\n ./rambi/variants/ninja/devicetree.cb\n ./rambi/variants/heli/devicetree.cb\n ./rambi/variants/sumo/devicetree.cb\n ./rambi/variants/banjo/devicetree.cb\n ./rambi/variants/candy/devicetree.cb\n ./rambi/variants/gnawty/devicetree.cb\n ./rambi/variants/rambi/devicetree.cb\n ./rambi/variants/quawks/devicetree.cb\n \n Plus, Cyan (only non-baytrail chromebook with max98090) also needs\n this patch for audio to work.\n \n Thus, this commit adds all the missing devices to bsw_max98090 quirk\n table, implemented by commit a182ecd3809c (\"ASoC: intel:\n cht_bsw_max98090_ti: Add quirk for boards using pmc_plt_clk_0\").\n \n Signed-off-by: Daniel Stuart \u003cdaniel.stuart14@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190815171300.30126-1-daniel.stuart14@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 8bcf62b73e5421df94deca95d7d7c152997fe5b4\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 21:25:13 2019 +0200\n\n ASoC: sun4i-i2s: Rework MCLK divider calculation\n \n The MCLK divider calculation is currently computing the ideal divider using\n the oversample rate, the sample rate and the parent rate.\n \n However, since we have access to the frequency is supposed to be running at\n already, and as it turns out we\u0027re using it to compute the oversample rate,\n we can just use the ratio between the parent rate and the MCLK rate to\n simplify a bit the formula.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/dcc5deb2eb650758d268bddd20f60ba58856d024.1566242458.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c7dd0828c088a71f30de8d249f63b2fa9f0d322d\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 21:25:10 2019 +0200\n\n ASoC: sun4i-i2s: Replace call to params_channels by local variable\n \n The sun4i_i2s_hw_params already has a variable holding the value returned\n by params_channels, so let\u0027s just use that variable instead of calling\n params_channels multiple times.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/c0faaac69ad40248f24eed3c3b2fa1ccc4a85b70.1566242458.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 7df8f9a20196072162d9dc8fe99943f2d35f23d5\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 21:25:14 2019 +0200\n\n ASoC: sun4i-i2s: Don\u0027t use the oversample to calculate BCLK\n \n The BCLK divider should be calculated using the parameters that actually\n make the BCLK rate: the number of channels, the sampling rate and the\n sample width.\n \n We\u0027ve been using the oversample_rate previously because in the former SoCs,\n the BCLK\u0027s parent is MCLK, which in turn is being used to generate the\n oversample rate, so we end up with something like this:\n \n oversample \u003d mclk_rate / sampling_rate\n bclk_div \u003d oversample / word_size / channels\n \n So, bclk_div \u003d mclk_rate / sampling_rate / word_size / channels.\n \n And this is actually better, since the oversampling ratio only plays a role\n because the MCLK is its parent, not because of what BCLK is supposed to be.\n \n Furthermore, that assumption of MCLK being the parent has been broken on\n newer SoCs, so let\u0027s use the proper formula, and have the parent rate as an\n argument.\n \n Fixes: 7d2993811a1e (\"ASoC: sun4i-i2s: Add support for H3\")\n Fixes: 21faaea1343f (\"ASoC: sun4i-i2s: Add support for A83T\")\n Fixes: 66ecce332538 (\"ASoC: sun4i-i2s: Add compatibility with A64 codec I2S\")\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/c3595e3a9788c2ef2dcc30aa3c8c4953bb5cc249.1566242458.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit a49d24e7d8d4fd4edf59e6373983e0bf4a2cca15\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 21:25:09 2019 +0200\n\n ASoC: sun4i-i2s: Switch to devm for PCM register\n \n Since the introduction of the driver, a new managed helper for the\n dmaengine PCM registration has been created. Let\u0027s use it to simplify a bit\n our probe and remove functions.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/606d271187091e858e8c15e20555af0b79798fe1.1566242458.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 413235fcedc7f61e925fe9818bc3f5eff8ad2494\nAuthor: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nDate: Thu Aug 15 16:44:01 2019 +0100\n\n arm64: Change the tagged_addr sysctl control semantics to only prevent the opt-in\n \n First rename the sysctl control to abi.tagged_addr_disabled and make it\n default off (zero). When abi.tagged_addr_disabled \u003d\u003d 1, only block the\n enabling of the TBI ABI via prctl(PR_SET_TAGGED_ADDR_CTRL, PR_TAGGED_ADDR_ENABLE).\n Getting the status of the ABI or disabling it is still allowed.\n \n Acked-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\n Signed-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 3e91ec89f527b9870fe42dcbdb74fd389d123a95\nAuthor: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nDate: Thu Aug 15 16:44:00 2019 +0100\n\n arm64: Tighten the PR_{SET, GET}_TAGGED_ADDR_CTRL prctl() unused arguments\n \n Require that arg{3,4,5} of the PR_{SET,GET}_TAGGED_ADDR_CTRL prctl and\n arg2 of the PR_GET_TAGGED_ADDR_CTRL prctl() are zero rather than ignored\n for future extensions.\n \n Acked-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\n Signed-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit bf283a05c09b58db83afbb1a8a3c6a684c56c1bb\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Aug 19 21:25:08 2019 +0200\n\n ASoC: sun4i-i2s: Register regmap and PCM before our component\n \n So far the regmap and the dmaengine PCM are registered after our component\n has been, which means that our driver isn\u0027t properly initialised by then.\n \n Let\u0027s fix that.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/67e303f37f141ef73ce9ed47d7f831b63c694424.1566242458.git-series.maxime.ripard@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 0a05f2e865ad6e461cada8f931b34551287dbba2\nAuthor: Michał Mirosław \u003cmirq-linux@rere.qmqm.pl\u003e\nDate: Tue Aug 20 13:02:37 2019 +0200\n\n ASoC: wm8904: implement input mode select as a mux\n \n Make \u0027* Capture Mode\u0027 a mux. This makes DAPM know that in single-ended\n mode only inverting mux paths need to be enabled.\n \n Signed-off-by: Michał Mirosław \u003cmirq-linux@rere.qmqm.pl\u003e\n Acked-by: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n Link: https://lore.kernel.org/r/125cd3c9f298da9b08a4d6002d4c00d70a898950.1566298834.git.mirq-linux@rere.qmqm.pl\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 22afe6242387eca1ac3905fbde7bef38deb0357d\nAuthor: Michał Mirosław \u003cmirq-linux@rere.qmqm.pl\u003e\nDate: Tue Aug 20 13:02:35 2019 +0200\n\n ASoC: wm8904: fix typo in DAPM kcontrol name\n \n Trivial fix for typo in \"Capture Inverting Mux\"es\u0027 name.\n \n Signed-off-by: Michał Mirosław \u003cmirq-linux@rere.qmqm.pl\u003e\n Acked-by: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n Link: https://lore.kernel.org/r/f95ae1085f9f3c137a122c4d95728711613c15f7.1566298834.git.mirq-linux@rere.qmqm.pl\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit a7325af725e83ae1a4e7a2290250dd1c952b672d\nAuthor: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nDate: Fri Aug 16 14:39:07 2019 +0300\n\n RDMA/hns: Fix some white space check_mtu_validate()\n \n This line was indented a bit too far.\n \n Signed-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Link: https://lore.kernel.org/r/20190816113907.GA30799@mwanda\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit d70898e4ea7680e4b9d49cf41ace141a464455eb\nAuthor: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\nDate: Tue Aug 20 09:55:48 2019 -0700\n\n drm/i915: Update DRIVER_DATE to 20190820\n \n Signed-off-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n\ncommit 4d374ba0bf30a2a372167ee4b7cdd527e7b47b3b\nAuthor: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\nDate: Tue Aug 20 10:31:54 2019 +0100\n\n tools: bpftool: implement \"bpftool btf show|list\"\n \n Add a \"btf list\" (alias: \"btf show\") subcommand to bpftool in order to\n dump all BTF objects loaded on a system.\n \n When running the command, hash tables are built in bpftool to retrieve\n all the associations between BTF objects and BPF maps and programs. This\n allows for printing all such associations when listing the BTF objects.\n \n The command is added at the top of the subcommands for \"bpftool btf\", so\n that typing only \"bpftool btf\" also comes down to listing the programs.\n We could not have this with the previous command (\"dump\"), which\n required a BTF object id, so it should not break any previous behaviour.\n This also makes the \"btf\" command behaviour consistent with \"prog\" or\n \"map\".\n \n Bash completion is updated to use \"bpftool btf\" instead of \"bpftool\n prog\" to list the BTF ids, as it looks more consistent.\n \n Example output (plain):\n \n # bpftool btf show\n 9: size 2989B prog_ids 21 map_ids 15\n 17: size 2847B prog_ids 36 map_ids 30,29,28\n 26: size 2847B\n \n Signed-off-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 09d7c2e32b6e06d58fe7a5aa38847f4719ab4cc7\nAuthor: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\nDate: Tue Aug 20 10:31:53 2019 +0100\n\n libbpf: add bpf_btf_get_next_id() to cycle through BTF objects\n \n Add an API function taking a BTF object id and providing the id of the\n next BTF object in the kernel. This can be used to list all BTF objects\n loaded on the system.\n \n v2:\n - Rebase on top of Andrii\u0027s changes regarding libbpf versioning.\n \n Signed-off-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit a6e130c4203bcc73450a00f647d99bd75ded95cb\nAuthor: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\nDate: Tue Aug 20 10:31:52 2019 +0100\n\n libbpf: refactor bpf_*_get_next_id() functions\n \n In preparation for the introduction of a similar function for retrieving\n the id of the next BTF object, consolidate the code from\n bpf_prog_get_next_id() and bpf_map_get_next_id() in libbpf.\n \n Signed-off-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit d2648e1ebbceb4da4f2edf5d471963f8831f3554\nAuthor: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\nDate: Tue Aug 20 10:31:51 2019 +0100\n\n tools: bpf: synchronise BPF UAPI header with tools\n \n Synchronise the bpf.h header under tools, to report the addition of the\n new BPF_BTF_GET_NEXT_ID syscall command for bpf().\n \n Signed-off-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 1b9ed84ecf268904d89edf2908426a8eb3b5a4ba\nAuthor: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\nDate: Tue Aug 20 10:31:50 2019 +0100\n\n bpf: add new BPF_BTF_GET_NEXT_ID syscall command\n \n Add a new command for the bpf() system call: BPF_BTF_GET_NEXT_ID is used\n to cycle through all BTF objects loaded on the system.\n \n The motivation is to be able to inspect (list) all BTF objects presents\n on the system.\n \n Signed-off-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit d91cc2f46ad5bd8020902093385fedf7e9e9d755\nAuthor: Raphael Gault \u003craphael.gault@arm.com\u003e\nDate: Tue Aug 20 16:57:45 2019 +0100\n\n arm64: perf_event: Add missing header needed for smp_processor_id()\n \n In perf_event.c we use smp_processor_id(), but we haven\u0027t included\n \u003clinux/smp.h\u003e where it is defined, and rely on this being pulled in\n via a transitive include. Let\u0027s make this more robust by including\n \u003clinux.smp.h\u003e explicitly.\n \n Acked-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Raphael Gault \u003craphael.gault@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 343bbb1a7f88df49a7166cd949d6cd3b4d3f0919\nAuthor: Nickey Yang \u003cnickey.yang@rock-chips.com\u003e\nDate: Mon Oct 1 14:38:42 2018 +0200\n\n dt-bindings: display: rockchip: update DSI controller\n \n This patch update describe panel/port links, including\n unit addresses in documentation of device tree bindings\n for the rockchip DSI controller based on the Synopsys\n DesignWare MIPI DSI host controller.\n \n Signed-off-by: Nickey Yang \u003cnickey.yang@rock-chips.com\u003e\n Reviewed-by: Brian Norris \u003cbriannorris@chromium.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n [this seems to have gotten lost when the original dsi-series was applied]\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20181001123845.11818-5-heiko@sntech.de\n\ncommit 238bdcdf5d0a087f8930c82b7c7cc142ca9399ce\nAuthor: Revanth Rajashekar \u003crevanth.rajashekar@intel.com\u003e\nDate: Tue Aug 20 09:30:51 2019 -0600\n\n block: sed-opal: Removed duplicate OPAL_METHOD_LENGTH definition\n \n The original commit adding the sed-opal library by mistake added two\n definitions of OPAL_METHOD_LENGTH, remove one of them.\n \n Signed-off-by: Revanth Rajashekar \u003crevanth.rajashekar@intel.com\u003e\n Reviewed-by: Scott Bauer \u003csbauer@plzdonthack.me\u003e\n Reviewed-by: Jon Derrick \u003cjonathan.derrick@intel.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 89c6cc2cab7e5090dc85ce0162ce92903b3aac5d\nAuthor: Revanth Rajashekar \u003crevanth.rajashekar@intel.com\u003e\nDate: Tue Aug 20 09:30:50 2019 -0600\n\n block: sed-opal: Remove always false conditional statement\n \n In the function \u0027response_parse\u0027, num_entries will never be 0 as\n slen is checked for 0. Hence, the condition \u0027if (num_entries \u003d\u003d 0)\u0027\n can never be true.\n \n Signed-off-by: Revanth Rajashekar \u003crevanth.rajashekar@intel.com\u003e\n Reviewed-by: Scott Bauer \u003csbauer@plzdonthack.me\u003e\n Reviewed-by: Jon Derrick \u003cjonathan.derrick@intel.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 5cc23ed75b629dfb0f8f7a7d0c80e0bab36b3960\nAuthor: Revanth Rajashekar \u003crevanth.rajashekar@intel.com\u003e\nDate: Tue Aug 20 09:30:49 2019 -0600\n\n block: sed-opal: Add/remove spaces\n \n Signed-off-by: Revanth Rajashekar \u003crevanth.rajashekar@intel.com\u003e\n Reviewed-by: Scott Bauer \u003csbauer@plzdonthack.me\u003e\n Reviewed-by: Jon Derrick \u003cjonathan.derrick@intel.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit c9cc9bec36d055077094db9fa57dfbf1870cbb73\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Tue Aug 20 16:40:47 2019 +0200\n\n soc: amlogic: meson-gx-socinfo: Add SM1 and S905X3 IDs\n \n Add the SoC IDs for the S905X3 Amlogic SM1 SoC.\n \n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Reviewed-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 9b543419652917f048310d0863c47c107c26fb0c\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Aug 19 15:41:00 2019 +0300\n\n Tools: hv: move to tools buildsystem\n \n There is a nice buildsystem dedicated for userspace tools in Linux kernel tree.\n Switch Hyper-V daemons to be built by it.\n \n Cc: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Tested-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\n Signed-off-by: Sasha Levin \u003csashal@kernel.org\u003e\n\ncommit b81d39c7a1efb83caa3f4419939a46e96191abb6\nAuthor: Jiri Olsa \u003cjolsa@redhat.com\u003e\nDate: Tue Aug 20 14:46:24 2019 +0200\n\n libperf: Fix arch include paths\n \n Guenter Roeck reported problem with compilation when the ARCH is\n specified:\n \n $ make ARCH\u003dx86_64\n In file included from tools/include/asm/atomic.h:6:0,\n from include/linux/atomic.h:5,\n from tools/include/linux/refcount.h:41,\n from cpumap.c:4: tools/include/asm/../../arch/x86/include/asm/atomic.h:11:10:\n fatal error: asm/cmpxchg.h: No such file or directory\n \n The problem is that we don\u0027t use SRCARCH (the sanitized ARCH version)\n and we don\u0027t get the proper include path.\n \n Reported-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Tested-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Fixes: 314350491810 (\"libperf: Make libperf.a part of the perf build\")\n Link: http://lkml.kernel.org/r/20190820124624.GG24105@krava\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 42fc2e9ef9603a7948aaa4ffd8dfb94b30294ad8\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Aug 20 11:45:17 2019 -0300\n\n tools headers: Fixup bitsperlong per arch includes\n \n We were getting the file by luck, from one of the paths in -I, fix it to\n get it from the proper place:\n \n $ cd tools/include/uapi/asm/\n [acme@quaco asm]$ grep include bitsperlong.h\n #include \"../../arch/x86/include/uapi/asm/bitsperlong.h\"\n #include \"../../arch/arm64/include/uapi/asm/bitsperlong.h\"\n #include \"../../arch/powerpc/include/uapi/asm/bitsperlong.h\"\n #include \"../../arch/s390/include/uapi/asm/bitsperlong.h\"\n #include \"../../arch/sparc/include/uapi/asm/bitsperlong.h\"\n #include \"../../arch/mips/include/uapi/asm/bitsperlong.h\"\n #include \"../../arch/ia64/include/uapi/asm/bitsperlong.h\"\n #include \"../../arch/riscv/include/uapi/asm/bitsperlong.h\"\n #include \"../../arch/alpha/include/uapi/asm/bitsperlong.h\"\n #include \u003casm-generic/bitsperlong.h\u003e\n $ ls -la ../../arch/x86/include/uapi/asm/bitsperlong.h\n ls: cannot access \u0027../../arch/x86/include/uapi/asm/bitsperlong.h\u0027: No such file or directory\n $ ls -la ../../../arch/*/include/uapi/asm/bitsperlong.h\n -rw-rw-r--. 1 237 ../../../arch/alpha/include/uapi/asm/bitsperlong.h\n -rw-rw-r--. 1 841 ../../../arch/arm64/include/uapi/asm/bitsperlong.h\n -rw-rw-r--. 1 966 ../../../arch/hexagon/include/uapi/asm/bitsperlong.h\n -rw-rw-r--. 1 234 ../../../arch/ia64/include/uapi/asm/bitsperlong.h\n -rw-rw-r--. 1 100 ../../../arch/microblaze/include/uapi/asm/bitsperlong.h\n -rw-rw-r--. 1 244 ../../../arch/mips/include/uapi/asm/bitsperlong.h\n -rw-rw-r--. 1 352 ../../../arch/parisc/include/uapi/asm/bitsperlong.h\n -rw-rw-r--. 1 312 ../../../arch/powerpc/include/uapi/asm/bitsperlong.h\n -rw-rw-r--. 1 353 ../../../arch/riscv/include/uapi/asm/bitsperlong.h\n -rw-rw-r--. 1 292 ../../../arch/s390/include/uapi/asm/bitsperlong.h\n -rw-rw-r--. 1 323 ../../../arch/sparc/include/uapi/asm/bitsperlong.h\n -rw-rw-r--. 1 320 ../../../arch/x86/include/uapi/asm/bitsperlong.h\n $\n \n Found while fixing some other problem, before it was escaping the\n tools/ chroot and using stuff in the kernel sources:\n \n CC /tmp/build/perf/util/find_bit.o\n In file included from /git/linux/tools/include/../../arch/x86/include/uapi/asm/bitsperlong.h:11,\n from /git/linux/tools/include/uapi/asm/bitsperlong.h:3,\n from /git/linux/tools/include/linux/bits.h:6,\n from /git/linux/tools/include/linux/bitops.h:13,\n from ../lib/find_bit.c:17:\n \n # cd /git/linux/tools/include/../../arch/x86/include/uapi/asm/\n # pwd\n /git/linux/arch/x86/include/uapi/asm\n #\n \n Now it is getting the one we want it to, i.e. the one inside tools/:\n \n CC /tmp/build/perf/util/find_bit.o\n In file included from /git/linux/tools/arch/x86/include/uapi/asm/bitsperlong.h:11,\n from /git/linux/tools/include/linux/bits.h:6,\n from /git/linux/tools/include/linux/bitops.h:13,\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-8f8cfqywmf6jk8a3ucr0ixhu@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 5c959b6d8f9369e42a3cb0423bcce9ac069cee55\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Aug 19 16:48:25 2019 -0300\n\n perf top: Show info message while collecting samples\n \n Give visual cue about what is happening while initially collecting the\n minimal set of samples to collect/sort/display.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-xcui60p1v6ozijfam2o89ya8@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 2284cf8074ff12b6304665618f81ced3aa0b41de\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Aug 19 16:43:58 2019 -0300\n\n perf ui browser: Allow specifying message to show when no samples are available to display\n \n The \u0027perf top\u0027 tool will use that to avoid having a initial blank screen\n while collecting the minimum number of samples to sort and display.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-89ciceg8cy4442he3t0jzo3f@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 9b01611934c045ac7ca71aebf5ee1906951d6bce\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Aug 19 16:38:24 2019 -0300\n\n perf ui: Introduce non-interactive ui__info_window() function\n \n Sometimes we want just to print a message on the center of the screen,\n like in \u0027perf top\u0027 while we wait for the minimum amount of samples to be\n collected before sorting and showing them.\n \n Also expose __ui__info_window() as an optimization for cases where such\n message is to be printed while holding the ui lock.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-uat0f89vfwl2w52kv9wzwd8a@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 9e79ff77e4195e40e7a47a2001132db66e25d541\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Aug 19 16:09:50 2019 -0300\n\n perf ui: Make \u0027exit_msg\u0027 optional in ui__question_window()\n \n We will not need it when refactoring this function to be\n non-interactive, so make it optional.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-pnx1dn17bsz7lqt9ty95nnjx@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit a4973d8f7bea98a0795ba853e7bd2cef11363824\nAuthor: Leo Yan \u003cleo.yan@linaro.org\u003e\nDate: Thu Aug 15 16:28:54 2019 +0800\n\n perf cs-etm: Support sample flags \u0027insn\u0027 and \u0027insnlen\u0027\n \n The synthetic branch and instruction samples are missed to set\n instruction related info, thus the perf tool fails to display samples\n with flags \u0027-F,+insn,+insnlen\u0027.\n \n The CoreSight trace decoder provides sufficient information to decide\n the instruction size based on the ISA type: A64/A32 instructions are\n 32-bit size, but one exception is the T32 instruction size, which might\n be 32-bit or 16-bit.\n \n This patch handles these cases and it reads the instruction values from\n DSO file; thus can support the flags \u0027-F,+insn,+insnlen\u0027.\n \n Before:\n \n # perf script -F,insn,insnlen,ip,sym\n 0 [unknown] ilen: 0\n ffff97174044 _start ilen: 0\n ffff97174938 _dl_start ilen: 0\n ffff97174938 _dl_start ilen: 0\n ffff97174938 _dl_start ilen: 0\n ffff97174938 _dl_start ilen: 0\n ffff97174938 _dl_start ilen: 0\n ffff97174938 _dl_start ilen: 0\n ffff97174938 _dl_start ilen: 0\n ffff97174938 _dl_start ilen: 0\n \n [...]\n \n After:\n \n # perf script -F,insn,insnlen,ip,sym\n 0 [unknown] ilen: 0\n ffff97174044 _start ilen: 4 insn: 2f 02 00 94\n ffff97174938 _dl_start ilen: 4 insn: c1 ff ff 54\n ffff97174938 _dl_start ilen: 4 insn: c1 ff ff 54\n ffff97174938 _dl_start ilen: 4 insn: c1 ff ff 54\n ffff97174938 _dl_start ilen: 4 insn: c1 ff ff 54\n ffff97174938 _dl_start ilen: 4 insn: c1 ff ff 54\n ffff97174938 _dl_start ilen: 4 insn: c1 ff ff 54\n ffff97174938 _dl_start ilen: 4 insn: c1 ff ff 54\n ffff97174938 _dl_start ilen: 4 insn: c1 ff ff 54\n \n [...]\n \n Signed-off-by: Leo Yan \u003cleo.yan@linaro.org\u003e\n Reviewed-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Tested-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Mike Leach \u003cmike.leach@linaro.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Robert Walker \u003crobert.walker@arm.com\u003e\n Cc: Suzuki Poulouse \u003csuzuki.poulose@arm.com\u003e\n Cc: coresight@lists.linaro.org\n Cc: linux-arm-kernel@lists.infradead.org\n Link: http://lkml.kernel.org/r/20190815082854.18191-1-leo.yan@linaro.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 10ccbc1cc0b8a05a5c8491630d36d1e2672036c1\nAuthor: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\nDate: Fri Aug 9 18:31:28 2019 +0300\n\n perf report: Prefer DWARF callstacks to LBR ones when captured both\n \n Display DWARF based callchains when the perf.data file contains raw thread\n stack data as LBR callstack data.\n \n Commiter testing:\n \n This changes the output from the branch stack based one, i.e. without\n this patch, for the same file as in the previous csets:\n \n # perf report --stdio\n # To display the perf.data header info, please use --header/--header-only options.\n #\n # Total Lost Samples: 0\n #\n # Samples: 13 of event \u0027cycles\u0027\n # Event count (approx.): 13\n #\n # Overhead Command Source Shared Object Source Symbol Target Symbol Basic Block Cycles\n # ........ ....... .................... ........................... ......................................... ..................\n #\n 7.69% ls libpthread-2.29.so [.] _init [.] __pthread_initialize_minimal_internal 6827\n 7.69% ls ld-2.29.so [k] _start [k] _dl_start -\n 7.69% ls ld-2.29.so [.] _dl_start_user [.] _dl_init -24790\n 7.69% ls ld-2.29.so [k] _dl_start [k] _dl_sysdep_start 278\n 7.69% ls ld-2.29.so [k] dl_main [k] _dl_map_object_deps 15581\n 7.69% ls ld-2.29.so [k] open_verify.constprop.0 [k] lseek64 4228\n 7.69% ls ld-2.29.so [k] _dl_map_object [k] open_verify.constprop.0 55\n 7.69% ls ld-2.29.so [k] openaux [k] _dl_map_object 67\n 7.69% ls ld-2.29.so [k] _dl_map_object_deps [k] 0x00007f441b57c090 112\n 7.69% ls ld-2.29.so [.] call_init.part.0 [.] _init 334\n 7.69% ls ld-2.29.so [.] _dl_init [.] call_init.part.0 383\n 7.69% ls ld-2.29.so [k] _dl_sysdep_start [k] dl_main 45\n 7.69% ls ld-2.29.so [k] _dl_catch_exception [k] openaux 116\n \n #\n # (Tip: For memory address profiling, try: perf mem record / perf mem report)\n #\n \n To the one that shows call chains:\n \n # perf report --stdio\n # To display the perf.data header info, please use --header/--header-only options.\n #\n #\n # Total Lost Samples: 0\n #\n # Samples: 10 of event \u0027cycles\u0027\n # Event count (approx.): 3204047\n #\n # Children Self Command Shared Object Symbol\n # ........ ........ ....... .................. .........................................\n #\n 55.01% 0.00% ls [kernel.vmlinux] [k] entry_SYSCALL_64_after_hwframe\n |\n ---entry_SYSCALL_64_after_hwframe\n do_syscall_64\n |\n --16.01%--__x64_sys_execve\n __do_execve_file.isra.0\n search_binary_handler\n load_elf_binary\n elf_map\n vm_mmap_pgoff\n do_mmap\n mmap_region\n perf_event_mmap\n perf_iterate_sb\n perf_iterate_ctx\n perf_event_mmap_output\n perf_output_copy\n memcpy_erms\n \n 55.01% 39.00% ls [kernel.vmlinux] [k] do_syscall_64\n |\n |--39.00%--0xffffffffffffffff\n | _dl_map_object\n | open_verify.constprop.0\n | __lseek64 (inlined)\n | entry_SYSCALL_64_after_hwframe\n | do_syscall_64\n |\n --16.01%--do_syscall_64\n __x64_sys_execve\n __do_execve_file.isra.0\n search_binary_handler\n load_elf_binary\n elf_map\n vm_mmap_pgoff\n do_mmap\n mmap_region\n perf_event_mmap\n perf_iterate_sb\n perf_iterate_ctx\n perf_event_mmap_output\n perf_output_copy\n memcpy_erms\n \n 42.95% 42.95% ls libpthread-2.29.so [.] __pthread_initialize_minimal_internal\n |\n ---_init\n __pthread_initialize_minimal_internal\n \n 42.95% 0.00% ls libpthread-2.29.so [.] _init\n |\n ---_init\n __pthread_initialize_minimal_internal\n \n \u003cSNIP\u003e\n \n #\n # (Tip: Profiling branch (mis)predictions with: perf record -b / perf report)\n #\n #\n \n The branch stack view be explicitely selected using:\n \n # perf report -h branch-stack\n \n Usage: perf report [\u003coptions\u003e]\n \n -b, --branch-stack use branch records for per branch histogram filling\n \n #\n \n I.e. after this patch:\n \n # perf report -b --stdio\n # To display the perf.data header info, please use --header/--header-only options.\n #\n #\n # Total Lost Samples: 0\n #\n # Samples: 13 of event \u0027cycles\u0027\n # Event count (approx.): 13\n #\n # Overhead Command Source Shared Object Source Symbol Target Symbol Basic Block Cycles\n # ........ ....... .................... ........................... ......................................... ..................\n #\n 7.69% ls libpthread-2.29.so [.] _init [.] __pthread_initialize_minimal_internal 6827\n 7.69% ls ld-2.29.so [k] _start [k] _dl_start -\n 7.69% ls ld-2.29.so [.] _dl_start_user [.] _dl_init -24790\n 7.69% ls ld-2.29.so [k] _dl_start [k] _dl_sysdep_start 278\n 7.69% ls ld-2.29.so [k] dl_main [k] _dl_map_object_deps 15581\n 7.69% ls ld-2.29.so [k] open_verify.constprop.0 [k] lseek64 4228\n 7.69% ls ld-2.29.so [k] _dl_map_object [k] open_verify.constprop.0 55\n 7.69% ls ld-2.29.so [k] openaux [k] _dl_map_object 67\n 7.69% ls ld-2.29.so [k] _dl_map_object_deps [k] 0x00007f441b57c090 112\n 7.69% ls ld-2.29.so [.] call_init.part.0 [.] _init 334\n 7.69% ls ld-2.29.so [.] _dl_init [.] call_init.part.0 383\n 7.69% ls ld-2.29.so [k] _dl_sysdep_start [k] dl_main 45\n 7.69% ls ld-2.29.so [k] _dl_catch_exception [k] openaux 116\n \n #\n # (Tip: Show current config key-value pairs: perf config --list)\n #\n #\n \n Signed-off-by: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Tested-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Jin Yao \u003cyao.jin@linux.intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Kan Liang \u003ckan.liang@linux.intel.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/ccbd9583-82f4-dec5-7e84-64bf56e351fb@linux.intel.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit d2720c3dad58def723f9617f7cf2a48c752ef50a\nAuthor: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\nDate: Fri Aug 9 18:26:30 2019 +0300\n\n perf report: Dump LBR callstack data by -D jointly with thread stack\n \n Make perf report -D command print captured LBR callstack chain when it is\n collected together with raw thread stack data:\n \n 2752673087247083 0x5d10 [0x548]: PERF_RECORD_SAMPLE(IP, 0x4002): 5841/5841: 0x40121f period: 1543862 addr: 0\n ... FP chain: nr:0\n ... branch callstack: nr:3\n ..... 0: 00000000004011d0\n ..... 1: 00007f393c388411\n ..... 2: 0000000000401098\n ... user regs: mask 0xff0fff ABI 64-bit\n .... AX 0x34e7\n .... BX 0x7fff5f6dd3c0\n .... CX 0xffffffff\n .... DX 0x34e6\n .... SI 0x7f393c5268d0\n .... DI 0x0\n .... BP 0x401260\n .... SP 0x7fff5f6dd3c0\n .... IP 0x40121f\n .... FLAGS 0x29f\n .... CS 0x33\n .... SS 0x2b\n .... R8 0x7f393c526800\n .... R9 0x7f393c525da0\n .... R10 0xfffffffffffff70a\n .... R11 0x246\n .... R12 0x401070\n .... R13 0x7fff5f6ddcb0\n .... R14 0x0\n .... R15 0x0\n ... ustack: size 1024, offset 0x130\n . data_src: 0x5080021\n ... thread: stack_test:5841\n ...... dso: /root/abudanko/stacks/stack_test\n \n Committer testing:\n \n # perf record -g --call-graph dwarf,1024 -j stack,u ls \u003e /dev/null\n [ perf record: Woken up 1 times to write data ]\n [ perf record: Captured and wrote 0.042 MB perf.data (10 samples) ]\n #\n \n Before:\n \n # perf report -D |\u0026 grep PERF_RECORD_SAMPLE -A28 | tail -29\n 67538909824483 0xa7a0 [0x560]: PERF_RECORD_SAMPLE(IP, 0x4002): 9721/9721: 0x7f441b2b1e20 period: 1376095 addr: 0\n ... FP chain: nr:0\n ... user regs: mask 0xff0fff ABI 64-bit\n .... AX 0x7f441b2b1000\n .... BX 0x7f441b55b970\n .... CX 0x7fff6e2db218\n .... DX 0x7fff6e2db218\n .... SI 0x7fff6e2db208\n .... DI 0x1\n .... BP 0x1\n .... SP 0x7fff6e2db178\n .... IP 0x7f441b2b1e20\n .... FLAGS 0x20a\n .... CS 0x33\n .... SS 0x2b\n .... R8 0x1\n .... R9 0x7f441b371c18\n .... R10 0x7f441b5a5f10\n .... R11 0x202\n .... R12 0x7fff6e2db208\n .... R13 0x7fff6e2db218\n .... R14 0x7f441b5a7150\n .... R15 0x0\n ... ustack: size 1024, offset 0x148\n . data_src: 0x5080021\n ... thread: ls:9721\n ...... dso: /usr/lib64/libpthread-2.29.so\n \n 0xad00 [0x60]: event: 10\n #\n \n After:\n \n # perf report -D |\u0026 grep PERF_RECORD_SAMPLE -A31 | tail -32\n 67538909824483 0xa7a0 [0x560]: PERF_RECORD_SAMPLE(IP, 0x4002): 9721/9721: 0x7f441b2b1e20 period: 1376095 addr: 0\n ... FP chain: nr:0\n ... branch callstack: nr:4\n ..... 0: 00007f441b2b1e20\n ..... 1: 00007f441b58af1a\n ..... 2: 00007f441b58b0e1\n ..... 3: 00007f441b57c145\n ... user regs: mask 0xff0fff ABI 64-bit\n .... AX 0x7f441b2b1000\n .... BX 0x7f441b55b970\n .... CX 0x7fff6e2db218\n .... DX 0x7fff6e2db218\n .... SI 0x7fff6e2db208\n .... DI 0x1\n .... BP 0x1\n .... SP 0x7fff6e2db178\n .... IP 0x7f441b2b1e20\n .... FLAGS 0x20a\n .... CS 0x33\n .... SS 0x2b\n .... R8 0x1\n .... R9 0x7f441b371c18\n .... R10 0x7f441b5a5f10\n .... R11 0x202\n .... R12 0x7fff6e2db208\n .... R13 0x7fff6e2db218\n .... R14 0x7f441b5a7150\n .... R15 0x0\n ... ustack: size 1024, offset 0x148\n . data_src: 0x5080021\n ... thread: ls:9721\n ...... dso: /usr/lib64/libpthread-2.29.so\n #\n \n Signed-off-by: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Tested-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Jin Yao \u003cyao.jin@linux.intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Kan Liang \u003ckan.liang@linux.intel.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/aa82e5dd-def2-0ca8-a064-db9e2e8ad076@linux.intel.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 2566349648b40aa3f5edf6af8f7f893ccd6e4eae\nAuthor: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\nDate: Fri Aug 9 18:23:58 2019 +0300\n\n perf record: Enable LBR callstack capture jointly with thread stack\n \n Enable \u0027-j stack\u0027 applicability together with \u0027--call-graph dwarf\u0027\n option so thread stack data and LBR call stack could be captured\n jointly:\n \n $ perf record -g --call-graph dwarf,1024 -j stack,u -- stack_test\n \n Collected LBR call stack can be used to augment DWARF call stack\n calculated from the raw thread stack data and to provide more\n comprehensive call stack information for cases when collected SIZE is\n not enough to cover complete thread stack.\n \n Such cases are typical for workloads that allocate large arrays of data\n on its threads stacks or the possible SIZE to collect can\u0027t be large\n enough due to workload nature or system configuration and this is where\n hardware captured LBR call stacks can provide missing stack frames.\n Possible DWARF plus LBR call stacks consolidation algorithm description\n follows.\n \n With this patch set perf report command UI currently ignores collected\n LBR call stack data and still provides DWARF based call stacks\n information.\n \n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n \n Overview:\n \n Legend:\n \n THS - thread stack\n CTX - thread register context\n SWS - software stack\n SSF - skipped stack frames\n PSS - Perf sample stack\n \n ip,sp,bp - HW registers values\n d - allocated stack regions\n kip - ip address in the kernel space\n K - captured thread stack size\n \n THS\n \n -----\n | |\u003c-stack bottom\n ...\n |---|\n |ip4|\n |---| PSS \u003d SWS(THS(K))\n | |\n --\u003e | |\n | |d3 | user/\n | |---| user PSS kernel PSS\n | |ip3| ------ ------\n | |---| |SSF | |SSF |\n | | | .... ....\n | | | ------ ------\n | |d2 | | -1 | | -1 |\n |---| user ------ ------\n K |ip2| CTX |ip3 | |ip3 |\n |---| |----| |----|\n | |d1 | ... |ip2 | , |ip2 |\n | |---| |---| |----| |----|\n | |ip1| |bp0| |ip1 | |ip1 |\n | |---| |---| |----| |----|\n | | | |ip0|-\u003e|ip0 | |ip0 |\u003c-user stack top\n | | | |---| ------ ------\n | | |\u003c-|sp0|\u003c-stack |kip0|\u003c-kernel stack bottom\n --\u003e ----- ----- top |----|\n |kip1|\n |----|\n |kip2|\n |----|\n ....\n | |\u003c-kernel stack top\n ------\n \n Algorithm details:\n \n Legend:\n \n HWS - hardware stack\n K-SWS - kernel software stack\n \n BRANCH\n TABLE\n \n HWS ip ip\n from to\n ------ -----------\n |ip7`| |ip7`| |\n |----| |----|----|\n |ip6`| |ip6`| |\n user PSS |----| |----|----|\n |ip5`| |ip5`| |\n ------ |----| |----|----|\n | -1 | |ip4`| |ip4`| |\n ------ |----| |----|----|\n |ip3 |~~~|ip3`| |ip3`| |\n |----| |----| |----|----|\n |ip2 |~~~|ip2`| |ip2`| |\n |----| |----| |----|----|\n |ip1 |~~~|ip1`| |ip1`|ip0`|\n |----| |----| -----------\n |ip0 |~~~|ip0`|\u003c---------\u0027\n ------ ------\n \n 1. if (sym(ipj) \u003d\u003d sym(ipj`)), j\u003d0-3 \u003d\u003d\u003d\u003e user PSS\n 2. ipj` , j\u003d4-7 \u003d\u003d\u003d\u003e user PSS\n \n Augmented PSS \u003d A_SWS(SWS(THS(K)), HWS):\n \n user/\n user PSS kernel PSS\n \n ------ ------\n |ip7`| |ip7`|\u003c-user PSS bottom\n |----| |----|\n |ip6`| |ip6`|\n |----| |----|\n HWS |ip5`| |ip5`|\n |----| |----|\n |ip4`| |ip4`|\n ------ ------\n |ip3 | |ip3 |\n |----| |----|\n SWS |ip2 | |ip2 |\n |----| |----|\n |ip1 | |ip1 |\n |----| |----|\n |ip0 | |ip0 |\u003c-user PSS top\n ------ ------\n |kip0|\u003c-kernel PSS bottom\n |----|\n |kip1|\n K-SWS |----|\n |kip2|\n |----|\n |kip3|\u003c-kernel PSS top\n ------\n \n APSS\n \n Committer testing:\n \n Before:\n \n # perf record -g --call-graph dwarf,1024 -j stack,u ls \u003e /dev/null\n unknown branch filter stack, check man page\n \n Usage: perf record [\u003coptions\u003e] [\u003ccommand\u003e]\n or: perf record [\u003coptions\u003e] -- \u003ccommand\u003e [\u003coptions\u003e]\n \n -j, --branch-filter \u003cbranch filter mask\u003e\n branch stack filter modes\n # perf record -g --call-graph dwarf,1024 -j u ls \u003e /dev/null\n [ perf record: Woken up 1 times to write data ]\n [ perf record: Captured and wrote 0.054 MB perf.data (12 samples) ]\n # perf evlist -v\n cycles: size: 112, { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|ADDR|CALLCHAIN|PERIOD|BRANCH_STACK|REGS_USER|STACK_USER|DATA_SRC, read_format: ID, disabled: 1, inherit: 1, mmap: 1, comm: 1, freq: 1, enable_on_exec: 1, task: 1, precise_ip: 3, mmap_data: 1, sample_id_all: 1, exclude_guest: 1, exclude_callchain_user: 1, mmap2: 1, comm_exec: 1, ksymbol: 1, bpf_event: 1, branch_sample_type: ANY, sample_regs_user: 0xff0fff, sample_stack_user: 1024\n #\n \n After:\n \n # perf record -g --call-graph dwarf,1024 -j stack,u ls \u003e /dev/null\n [ perf record: Woken up 1 times to write data ]\n [ perf record: Captured and wrote 0.044 MB perf.data (11 samples) ]\n [root@quaco ~]# perf evlist -v\n cycles: size: 112, { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|ADDR|CALLCHAIN|PERIOD|BRANCH_STACK|REGS_USER|STACK_USER|DATA_SRC, read_format: ID, disabled: 1, inherit: 1, mmap: 1, comm: 1, freq: 1, enable_on_exec: 1, task: 1, precise_ip: 3, mmap_data: 1, sample_id_all: 1, exclude_guest: 1, exclude_callchain_user: 1, mmap2: 1, comm_exec: 1, ksymbol: 1, bpf_event: 1, branch_sample_type: USER|CALL_STACK, sample_regs_user: 0xff0fff, sample_stack_user: 1024\n #\n \n Signed-off-by: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Tested-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Jin Yao \u003cyao.jin@linux.intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Kan Liang \u003ckan.liang@linux.intel.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/e9e00090-66fb-d2a4-c90f-1d12344f7788@linux.intel.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 82a2f88458d70704be843961e10b5cef9a6e95d3\nAuthor: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\nDate: Mon Aug 5 13:01:50 2019 -0400\n\n tools lib traceevent: Fix \"robust\" test of do_generate_dynamic_list_file\n \n The tools/lib/traceevent/Makefile had a test added to it to detect a failure\n of the \"nm\" when making the dynamic list file (whatever that is). The\n problem is that the test sorts the values \"U W w\" and some versions of sort\n will place \"w\" ahead of \"W\" (even though it has a higher ASCII value, and\n break the test.\n \n Add \u0027tr \"w\" \"W\"\u0027 to merge the two and not worry about the ordering.\n \n Reported-by: Tzvetomir Stoyanov \u003ctstoyanov@vmware.com\u003e\n Signed-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: David Carrillo-Cisneros \u003cdavidcc@google.com\u003e\n Cc: He Kuang \u003chekuang@huawei.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Michal rarek \u003cmmarek@suse.com\u003e\n Cc: Paul Turner \u003cpjt@google.com\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Stephane Eranian \u003ceranian@google.com\u003e\n Cc: Uwe Kleine-König \u003cu.kleine-koenig@pengutronix.de\u003e\n Cc: Wang Nan \u003cwangnan0@huawei.com\u003e\n Cc: stable@vger.kernel.org\n Fixes: 6467753d61399 (\"tools lib traceevent: Robustify do_generate_dynamic_list_file\")\n Link: http://lkml.kernel.org/r/20190805130150.25acfeb1@gandalf.local.home\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 3c84e65a533dbaa1a29bfd847deca73704b675eb\nAuthor: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\nDate: Mon Aug 12 12:09:35 2019 +0300\n\n perf evsel: Add comment for \u0027idx\u0027 member in \u0027struct perf_sample_id\n \n The \u0027idx\u0027 member was added as preparation for AUX area sampling. Add a\n comment to describe why.\n \n Signed-off-by: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/83ff264f-84c3-5372-8976-dd9293d20c6f@intel.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit aeaed4848234c97fb720b0e51a0c56dc8de0eeed\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:40:58 2019 -0400\n\n xprtrdma: Boost client\u0027s max slot table size to match Linux server\n \n I\u0027ve heard rumors of an NFS/RDMA server implementation that has a\n default credit limit of 1024. The client\u0027s default setting remains\n at 128.\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit 0ac10d87a571ae7d68c7f70f1c2229388f1dce9e\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Aug 19 10:53:20 2019 -0300\n\n tools arch x86: Sync asm/cpufeatures.h with the with the kernel\n \n To pick up the changes in:\n \n f36cf386e3fe (\"x86/speculation/swapgs: Exclude ATOMs from speculation through SWAPGS\")\n 18ec54fdd6d1 (\"x86/speculation: Prepare entry code for Spectre v1 swapgs mitigations\")\n \n That don\u0027t affect anything in tools/.\n \n This silences this perf build warning:\n \n Warning: Kernel ABI header at \u0027tools/arch/x86/include/asm/cpufeatures.h\u0027 differs from latest version at \u0027arch/x86/include/asm/cpufeatures.h\u0027\n diff -u tools/arch/x86/include/asm/cpufeatures.h arch/x86/include/asm/cpufeatures.h\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/n/tip-860dq1qie2cpnfghlpcnxrzr@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit bb13f35b96f4c83dc4015d71fa2b543c665fbdae\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Tue Aug 20 01:32:43 2019 +0000\n\n nfsd: remove duplicated include from filecache.c\n \n Remove duplicated include.\n \n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit f3c66a2f56683a20ced4e700a8167d6b357641da\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:40:11 2019 -0400\n\n xprtrdma: Boost maximum transport header size\n \n Although I haven\u0027t seen any performance results that justify it,\n I\u0027ve received several complaints that NFS/RDMA no longer supports\n a maximum rsize and wsize of 1MB. These days it is somewhat smaller.\n \n To simplify the logic that determines whether a chunk list is\n necessary, the implementation uses a fixed maximum size of the\n transport header. Currently that maximum size is 256 bytes, one\n quarter of the default inline threshold size for RPC/RDMA v1.\n \n Since commit a78868497c2e (\"xprtrdma: Reduce max_frwr_depth\"), the\n size of chunks is also smaller to take advantage of inline page\n lists in device internal MR data structures.\n \n The combination of these two design choices has reduced the maximum\n NFS rsize and wsize that can be used for most RNIC/HCAs. Increasing\n the maximum transport header size and the maximum number of RDMA\n segments it can contain increases the negotiated maximum rsize/wsize\n on common RNIC/HCAs.\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit b658911731d41a91b9de5458d6f38d6e1c90a453\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Aug 19 11:14:28 2019 -0300\n\n tools headers: Synchronize linux/bits.h with the kernel sources\n \n To pick up the changes in this cset:\n \n 95b980d62d52 (\"linux/bits.h: make BIT(), GENMASK(), and friends available in assembly\")\n \n To address this tools/perf build warning:\n \n Warning: Kernel ABI header at \u0027tools/include/linux/bits.h\u0027 differs from latest version at \u0027include/linux/bits.h\u0027\n diff -u tools/include/linux/bits.h include/linux/bits.h\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-1if3iga5r3di6oyddgxsr225@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit e7023898034ea3142661ecf02813177c00f3298c\nAuthor: Jungseung Lee \u003cjs07.lee@samsung.com\u003e\nDate: Sat Jul 13 22:56:20 2019 +0900\n\n mtd: spi-nor : Remove SPI_NOR_HAS_TB flag on s25fl512s\n \n Currently, the Top/Bottom protection function (SPI_NOR_HAS_TB) is\n implemented to fit some flashes with TB bit on SR.\n \n s25fl512s has TBPROT bit on CR1, so the TB protection is not working on it.\n Fix the wrong flag on s25fl512s.\n \n Signed-off-by: Jungseung Lee \u003cjs07.lee@samsung.com\u003e\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n\ncommit aaa6ef8aa85f33c6dd593e139b7f7d901bbde0e2\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Aug 19 11:00:54 2019 -0300\n\n tools headers: Grab copy of linux/const.h, needed by linux/bits.h\n \n So that can update the copy of linux/bits.h that now uses macros defined\n in const.h and that are not available in older systems.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-c2qfcbl58hxyfb5u5xivp7is@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 098454362a06029c1abc175b74409bf7d6e6604b\nAuthor: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\nDate: Sun Aug 18 21:34:20 2019 -0700\n\n test_bpf: Fix a new clang warning about xor-ing two numbers\n \n r369217 in clang added a new warning about potential misuse of the xor\n operator as an exponentiation operator:\n \n ../lib/test_bpf.c:870:13: warning: result of \u002710 ^ 300\u0027 is 294; did you\n mean \u00271e300\u0027? [-Wxor-used-as-pow]\n { { 4, 10 ^ 300 }, { 20, 10 ^ 300 } },\n ~~~^~~~~\n 1e300\n ../lib/test_bpf.c:870:13: note: replace expression with \u00270xA ^ 300\u0027 to\n silence this warning\n ../lib/test_bpf.c:870:31: warning: result of \u002710 ^ 300\u0027 is 294; did you\n mean \u00271e300\u0027? [-Wxor-used-as-pow]\n { { 4, 10 ^ 300 }, { 20, 10 ^ 300 } },\n ~~~^~~~~\n 1e300\n ../lib/test_bpf.c:870:31: note: replace expression with \u00270xA ^ 300\u0027 to\n silence this warning\n \n The commit link for this new warning has some good logic behind wanting\n to add it but this instance appears to be a false positive. Adopt its\n suggestion to silence the warning but not change the code. According to\n the differential review link in the clang commit, GCC may eventually\n adopt this warning as well.\n \n Link: https://github.com/ClangBuiltLinux/linux/issues/643\n Link: https://github.com/llvm/llvm-project/commit/920890e26812f808a74c60ebc14cc636dac661c1\n Signed-off-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Acked-by: Yonghong Song \u003cyhs@fb.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 146dc303630aff5fdf006614a67704539c519c33\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Aug 19 11:11:30 2019 -0300\n\n perf tools: tools/include should come before tools/uapi/include\n \n The next cset will grap const.h copies from the kernel to keep bits.h\n in sync as it started to use linux/const.h, that in turn includes\n uapi/linux/const.h.\n \n So now we have a file with the same name in tools/include and\n tools/uapi/include, and one includes the other, we need to have\n tools/include/uapi/ after tools/include/ for this to work, fix it.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-qzjqxa1wdrt51kwadyqawnuj@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 69ecfdaa534943efd95ee12b53fbb0f344e42e7e\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Tue Aug 20 01:10:35 2019 +0900\n\n bpf: add include guard to tnum.h\n \n Add a header include guard just in case.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 6e98bc349ea4219e21785d85809b10bd49e722df\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Aug 20 12:01:38 2019 -0300\n\n tools headers: Add limits.h to access __WORDSIZE\n \n We need to make sure limits.h is included before checking if we can use\n __WORDSIZE, do it.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-5yfoed4rnsck2n3cwhm9mvth@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 36bdd9056b6a83d573ffdde282a5a91ce734c536\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:39:25 2019 -0400\n\n xprtrdma: Fix calculation of ri_max_segs again\n \n Commit 302d3deb206 (\"xprtrdma: Prevent inline overflow\") added this\n calculation back in 2016, but got it wrong. I tested only the lower\n bound, which is why there is a max_t there. The upper bound should be\n rounded up too.\n \n Now, when using DIV_ROUND_UP, that takes care of the lower bound as\n well.\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit 670e90924bfe9cf80536eab66a31704e07b5fd00\nAuthor: Aaron Armstrong Skomra \u003cskomra@gmail.com\u003e\nDate: Fri Aug 16 12:02:50 2019 -0700\n\n HID: wacom: support named keys on older devices\n \n Some Wacom devices have keys with predefined meanings. However, when\n support was originally added for these devices, the codes for these\n keys were not available yet. These keys were thus reported with\n the numbered KEY_PROG* range.\n \n Some missing key codes were added with commit 4eb220cb35a9 (\"HID:\n wacom: generic: add 3 tablet touch keys\") and we are now able to\n report the proper key codes. We continue to report the original\n KEY_PROG codes so as not to break any unknown applications that\n may depend on them.\n \n Also, to support the touch key, track its state in the pad report.\n \n Signed-off-by: Aaron Armstrong Skomra \u003caaron.skomra@wacom.com\u003e\n Link: https://gitlab.freedesktop.org/libinput/libinput/merge_requests/155\n Link: https://github.com/linuxwacom/xf86-input-wacom/pull/46\n Reviewed-by: Ping Cheng \u003cping.cheng@wacom.com\u003e\n Reviewed-by: Jason Gerecke \u003cjason.gerecke@wacom.com\u003e\n Signed-off-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n\ncommit 073b50bccbbf99a3b79a1913604c656d0e1a56c9\nAuthor: Jason Gerecke \u003ckillertofu@gmail.com\u003e\nDate: Fri Aug 16 11:54:26 2019 -0700\n\n HID: wacom: Fix several minor compiler warnings\n \n Addresses a few issues that were noticed when compiling with non-default\n warnings enabled. The trimmed-down warnings in the order they are fixed\n below are:\n \n * declaration of \u0027size\u0027 shadows a parameter\n \n * \u0027%s\u0027 directive output may be truncated writing up to 5 bytes into a\n region of size between 1 and 64\n \n * pointer targets in initialization of \u0027char *\u0027 from \u0027unsigned char *\u0027\n differ in signedness\n \n * left shift of negative value\n \n Signed-off-by: Jason Gerecke \u003cjason.gerecke@wacom.com\u003e\n Reviewed-by: Aaron Armstrong Skomra \u003caaron.skomra@wacom.com\u003e\n Signed-off-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n\ncommit af08a7754a5da7ae704624cbe4ef83e46246c92d\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:38:38 2019 -0400\n\n xprtrdma: Update obsolete comment\n \n Comment was made obsolete by commit 8cec3dba76a4 (\"xprtrdma:\n rpcrdma_regbuf alignment\").\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit dc617f29dbe5ef0c8ced65ce62c464af1daaab3d\nAuthor: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nDate: Tue Aug 20 07:55:16 2019 -0700\n\n vfs: don\u0027t allow writes to swap files\n \n Don\u0027t let userspace write to an active swap file because the kernel\n effectively has a long term lease on the storage and things could get\n seriously corrupted if we let this happen.\n \n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit 1638045c36772b47a0765f7dca07cb90267e4942\nAuthor: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\nDate: Tue Aug 20 07:55:16 2019 -0700\n\n mm: set S_SWAPFILE on blockdev swap devices\n \n Set S_SWAPFILE on block device inodes so that they have the same\n protections as a swap flie.\n \n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit 2fb2a4d529fe3b1db4ced06d3b6568d46a991269\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:37:52 2019 -0400\n\n xprtrdma: Refresh the documenting comment in frwr_ops.c\n \n Things have changed since this comment was written. In particular,\n the reworking of connection closing, on-demand creation of MRs, and\n the removal of fr_state all mean that deferring MR recovery to\n frwr_map is no longer needed. The description is obsolete.\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit f13e18048bdfcea2c3e25ec691cb6b4d8ab3cf21\nAuthor: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\nDate: Mon Aug 12 17:03:54 2019 +0300\n\n mtd: spi-nor: intel-spi: Add support for Intel Tiger Lake SPI serial flash\n \n Intel Tiger Lake has the same SPI serial flash controller as Ice Lake.\n Add Tiger Lake PCI ID to the driver list of supported devices.\n \n Signed-off-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n\ncommit 6dc944db29ba9c1e9a50e5a16f28f8373fe0f5d5\nAuthor: Eugeniy Paltsev \u003cEugeniy.Paltsev@synopsys.com\u003e\nDate: Thu Jul 18 17:06:23 2019 +0300\n\n mtd: spi-nor: add support for sst26wf016b memory IC\n \n This commit adds support for the SST sst26wf016b flash memory IC.\n This IC was tested with \"snps,dw-apb-ssi\" SPI controller.\n We don\u0027t test dual/quad reads however sst26wf016b flash\u0027s datasheet\n advertises both dual and quad reads (and support of corresponding\n commands)\n \n Signed-off-by: Eugeniy Paltsev \u003cEugeniy.Paltsev@synopsys.com\u003e\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n\ncommit 6acbe9f630913a7ff433d7a5f30d3300fc4e0a94\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Aug 20 15:12:18 2019 +0100\n\n drm/i915/gtt: Relax pd_used assertion\n \n The current assertion tries to make sure that we do not over count the\n number of used PDE inside a page directory -- that is with an array of\n 512 pde, we do not expect more than 512 elements used! However, our\n assertion has to take into account that as we pin an element into the\n page directory, the caller first pins the page directory so the usage\n count is one higher. However, this should be one extra pin per thread,\n and the upper bound is that we may have one thread for each entry.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820141218.14714-1-chris@chris-wilson.co.uk\n\ncommit 95bd8304b34656d30bc0c908384de007b8fbcb55\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:37:05 2019 -0400\n\n SUNRPC: Inline xdr_commit_encode\n \n Micro-optimization: For xdr_commit_encode call sites in\n net/sunrpc/xdr.c, eliminate the extra calling sequence. On my\n client, this change saves about a microsecond for every 30 calls\n to xdr_reserve_space().\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit 1bcd8688bd9e122fdc47af97c5bed0c44ad60c98\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Mon Aug 19 19:01:46 2019 -0700\n\n drm/i915: Dynamically allocate s0ix struct for VLV\n \n This is only required for a single platform so no need to reserve the\n memory on all of them.\n \n This removes the last direct dependency of i915_drv.h on i915_reg.h\n (apart from the i915_reg_t definition).\n \n v2: drop unneeded diff, keep the vlv prefix, call functions\n unconditionally (Jani), fwd declaration of the struct (Chris)\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Imre Deak \u003cimre.deak@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820020147.5667-1-daniele.ceraolospurio@intel.com\n\ncommit 0aa5427a7fd3349d5d77ee8be8e47d15705d4f0e\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Sat Aug 17 02:38:48 2019 -0700\n\n drm/i915/tgl: Gen12 render context size\n \n Re-use Gen11 context size for now.\n \n [ Lucas: this is a temporary enabling patch that needs to be confirmed:\n we need to check BSpec 46255 and recompute ]\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Acked-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190817093902.2171-27-lucas.demarchi@intel.com\n\ncommit b41e63d88193babb6508a0d27bc4603ad9336ab1\nAuthor: Michel Thierry \u003cmichel.thierry@intel.com\u003e\nDate: Sat Aug 17 02:38:54 2019 -0700\n\n drm/i915/tgl: Updated Private PAT programming\n \n Gen12 removes the target-cache and age fields from the private PAT\n because MOCS now have the capability to set these itself. Only memory-type\n field should be programmed in the ppat, the reminded bits are reserved.\n \n Since now there are only 4 possible combinations, we could set only 4\n PPAT and leave the reminded 4 as UC, but I left them as WB as we used\n to have before.\n \n Also these registers have been relocated to the 0x4800-0x481c range.\n \n HSDES: 1406402661\n BSpec: 31654\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Michel Thierry \u003cmichel.thierry@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190817093902.2171-33-lucas.demarchi@intel.com\n\ncommit 13e53c5c533741a725cad04dfc2a5c467be60241\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Sat Aug 17 02:38:42 2019 -0700\n\n drm/i915/tgl: Introduce initial Tiger Lake workarounds\n \n Add empty workaround hooks for Tiger Lake. The workarounds will be added\n on separate patches. We were already applying\n WaRsForcewakeAddDelayForAck, which is indeed still valid, so also update\n the comment.\n \n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Radhakrishna Sripada \u003cradhakrishna.sripada@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190817093902.2171-21-lucas.demarchi@intel.com\n\ncommit f4785682c9af3ba27d3ebdb8332946962ee0c74e\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Tue Aug 20 11:22:01 2019 +0100\n\n drm/i915/tgl: Gen12 csb support\n \n The CSB format has been reworked for Gen12 to include information on\n both the context we\u0027re switching away from and the context we\u0027re\n switching to. After the change, some of the events don\u0027t have their\n own bit anymore and need to be inferred from other values in the csb.\n One of the context IDs (0x7FF) has also been reserved to indicate\n the invalid ctx, i.e. engine idle.\n \n Note that the full context ID includes the SW counter as well, but since\n we currently only care if the context is valid or not we can ignore that\n part.\n \n v2: fix mask size, fix and expand comments (Tvrtko),\n use if-ladder (Chris)\n \n Bspec: 45555, 46144\n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@linux.intel.com\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Acked-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820102201.29849-1-chris@chris-wilson.co.uk\n\ncommit 3123db1d2651853a963d38c8290b9443c23f2242\nAuthor: Avi Fishman \u003cavifishman70@gmail.com\u003e\nDate: Tue Jul 30 11:18:32 2019 +0300\n\n mtd: spi-nor: Add Winbond w25q256jvm\n \n Similar to w25q256 (besides not supporting QPI mode) but with different ID.\n The \"JVM\" suffix is in the datasheet.\n The datasheet indicates DUAL and QUAD are supported.\n https://www.winbond.com/resource-files/w25q256jv%20spi%20revi%2010232018%20plus.pdf\n \n Signed-off-by: Avi Fishman \u003cavifishman70@gmail.com\u003e\n Reviewed-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n\ncommit fccd20d9ed5bf9db59a5cd39914275a9d55588a0\nAuthor: Filipe Manana \u003cfdmanana@suse.com\u003e\nDate: Mon Aug 12 19:14:29 2019 +0100\n\n Btrfs: fix use-after-free when using the tree modification log\n \n At ctree.c:get_old_root(), we are accessing a root\u0027s header owner field\n after we have freed the respective extent buffer. This results in an\n use-after-free that can lead to crashes, and when CONFIG_DEBUG_PAGEALLOC\n is set, results in a stack trace like the following:\n \n [ 3876.799331] stack segment: 0000 [#1] SMP DEBUG_PAGEALLOC PTI\n [ 3876.799363] CPU: 0 PID: 15436 Comm: pool Not tainted 5.3.0-rc3-btrfs-next-54 #1\n [ 3876.799385] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.0-0-ga698c8995f-prebuilt.qemu.org 04/01/2014\n [ 3876.799433] RIP: 0010:btrfs_search_old_slot+0x652/0xd80 [btrfs]\n (...)\n [ 3876.799502] RSP: 0018:ffff9f08c1a2f9f0 EFLAGS: 00010286\n [ 3876.799518] RAX: ffff8dd300000000 RBX: ffff8dd85a7a9348 RCX: 000000038da26000\n [ 3876.799538] RDX: 0000000000000000 RSI: ffffe522ce368980 RDI: 0000000000000246\n [ 3876.799559] RBP: dae1922adadad000 R08: 0000000008020000 R09: ffffe522c0000000\n [ 3876.799579] R10: ffff8dd57fd788c8 R11: 000000007511b030 R12: ffff8dd781ddc000\n [ 3876.799599] R13: ffff8dd9e6240578 R14: ffff8dd6896f7a88 R15: ffff8dd688cf90b8\n [ 3876.799620] FS: 00007f23ddd97700(0000) GS:ffff8dda20200000(0000) knlGS:0000000000000000\n [ 3876.799643] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n [ 3876.799660] CR2: 00007f23d4024000 CR3: 0000000710bb0005 CR4: 00000000003606f0\n [ 3876.799682] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\n [ 3876.799703] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\n [ 3876.799723] Call Trace:\n [ 3876.799735] ? do_raw_spin_unlock+0x49/0xc0\n [ 3876.799749] ? _raw_spin_unlock+0x24/0x30\n [ 3876.799779] resolve_indirect_refs+0x1eb/0xc80 [btrfs]\n [ 3876.799810] find_parent_nodes+0x38d/0x1180 [btrfs]\n [ 3876.799841] btrfs_check_shared+0x11a/0x1d0 [btrfs]\n [ 3876.799870] ? extent_fiemap+0x598/0x6e0 [btrfs]\n [ 3876.799895] extent_fiemap+0x598/0x6e0 [btrfs]\n [ 3876.799913] do_vfs_ioctl+0x45a/0x700\n [ 3876.799926] ksys_ioctl+0x70/0x80\n [ 3876.799938] ? trace_hardirqs_off_thunk+0x1a/0x20\n [ 3876.799953] __x64_sys_ioctl+0x16/0x20\n [ 3876.799965] do_syscall_64+0x62/0x220\n [ 3876.799977] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n [ 3876.799993] RIP: 0033:0x7f23e0013dd7\n (...)\n [ 3876.800056] RSP: 002b:00007f23ddd96ca8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010\n [ 3876.800078] RAX: ffffffffffffffda RBX: 00007f23d80210f8 RCX: 00007f23e0013dd7\n [ 3876.800099] RDX: 00007f23d80210f8 RSI: 00000000c020660b RDI: 0000000000000003\n [ 3876.800626] RBP: 000055fa2a2a2440 R08: 0000000000000000 R09: 00007f23ddd96d7c\n [ 3876.801143] R10: 00007f23d8022000 R11: 0000000000000246 R12: 00007f23ddd96d80\n [ 3876.801662] R13: 00007f23ddd96d78 R14: 00007f23d80210f0 R15: 00007f23ddd96d80\n (...)\n [ 3876.805107] ---[ end trace e53161e179ef04f9 ]---\n \n Fix that by saving the root\u0027s header owner field into a local variable\n before freeing the root\u0027s extent buffer, and then use that local variable\n when needed.\n \n Fixes: 30b0463a9394d9 (\"Btrfs: fix accessing the root pointer in tree mod log functions\")\n CC: stable@vger.kernel.org # 3.10+\n Reviewed-by: Nikolay Borisov \u003cnborisov@suse.com\u003e\n Reviewed-by: Anand Jain \u003canand.jain@oracle.com\u003e\n Signed-off-by: Filipe Manana \u003cfdmanana@suse.com\u003e\n Reviewed-by: David Sterba \u003cdsterba@suse.com\u003e\n Signed-off-by: David Sterba \u003cdsterba@suse.com\u003e\n\ncommit 3481e64bbe9876e359aec74c9e93039b94f678b0\nAuthor: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\nDate: Tue Aug 20 14:53:46 2019 +0100\n\n bpf: add BTF ids in procfs for file descriptors to BTF objects\n \n Implement the show_fdinfo hook for BTF FDs file operations, and make it\n print the id of the BTF object. This allows for a quick retrieval of the\n BTF id from its FD; or it can help understanding what type of object\n (BTF) the file descriptor points to.\n \n v2:\n - Do not expose data_size, only btf_id, in FD info.\n \n Signed-off-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit ede6bc88d6bbe16a938de2b00f60f4e03768c988\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Tue Aug 20 01:36:52 2019 +0000\n\n bpf: Use PTR_ERR_OR_ZERO in xsk_map_inc()\n \n Use PTR_ERR_OR_ZERO rather than if(IS_ERR(...)) + PTR_ERR.\n \n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Acked-by: Björn Töpel \u003cbjorn.topel@intel.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 691b45ddbd182a4ce0bc91953c70c845cf0935f1\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Aug 19 18:36:19 2019 -0400\n\n SUNRPC: Remove rpc_wake_up_queued_task_on_wq()\n \n Clean up: commit c544577daddb (\"SUNRPC: Clean up transport write\n space handling\") appears to have removed the last caller of\n rpc_wake_up_queued_task_on_wq().\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit e2751463eaa6f9fec8fea80abbdc62dbc487b3c5\nAuthor: Jia-Ju Bai \u003cbaijiaju1990@gmail.com\u003e\nDate: Fri Jul 26 15:48:53 2019 +0800\n\n fs: nfs: Fix possible null-pointer dereferences in encode_attrs()\n \n In encode_attrs(), there is an if statement on line 1145 to check\n whether label is NULL:\n if (label \u0026\u0026 (attrmask[2] \u0026 FATTR4_WORD2_SECURITY_LABEL))\n \n When label is NULL, it is used on lines 1178-1181:\n *p++ \u003d cpu_to_be32(label-\u003elfs);\n *p++ \u003d cpu_to_be32(label-\u003epi);\n *p++ \u003d cpu_to_be32(label-\u003elen);\n p \u003d xdr_encode_opaque_fixed(p, label-\u003elabel, label-\u003elen);\n \n To fix these bugs, label is checked before being used.\n \n These bugs are found by a static analysis tool STCheck written by us.\n \n Signed-off-by: Jia-Ju Bai \u003cbaijiaju1990@gmail.com\u003e\n Signed-off-by: Anna Schumaker \u003cAnna.Schumaker@Netapp.com\u003e\n\ncommit 6d26d9bbf59a01ac99488252356348ae066a2c38\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Sat Aug 17 02:38:50 2019 -0700\n\n drm/i915/tgl: add GEN12_MAX_CONTEXT_HW_ID\n \n Like Gen11, Gen12 has 11 available bits for the ctx id field. However,\n the last value (0x7FF) is reserved to indicate engine idle, so we\n need to reduce the maximum number of contexts by 1 compared to Gen11.\n \n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Acked-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190817093902.2171-29-lucas.demarchi@intel.com\n\ncommit 487f471da34d2a29383e792ec512d0c4ab10da12\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Sat Aug 17 02:38:49 2019 -0700\n\n drm/i915/tgl: add Gen12 default indirect ctx offset\n \n Gen12 uses a new indirect ctx offset.\n \n Bspec: 11740\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Radhakrishna Sripada \u003cradhakrishna.sripada@intel.com\u003e\n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190817093902.2171-28-lucas.demarchi@intel.com\n\ncommit bd3b3004b1d303f2cbb17020bb338c4653b90f76\nAuthor: Michel Thierry \u003cmichel.thierry@intel.com\u003e\nDate: Tue Jul 30 17:49:02 2019 -0700\n\n drm/i915/tgl: Report valid VDBoxes with SFC capability\n \n In Gen11, only even numbered \"logical\" VDBoxes are hooked up to a SFC\n (Scaler \u0026 Format Converter) unit. This is not the case in Tigerlake,\n where each VDBox can access a SFC.\n \n We will use this information to decide when the SFC units need to be reset\n and also pass it to the GuC.\n \n Bspec: 48077\n Signed-off-by: Michel Thierry \u003cmichel.thierry@intel.com\u003e\n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Cc: Vinay Belgaumkar \u003cvinay.belgaumkar@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190731004902.34672-5-daniele.ceraolospurio@intel.com\n\ncommit 2833ddccbdb4a23d79070603eb271acb680818e2\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Aug 20 11:05:31 2019 +0100\n\n drm/i915: Be defensive when starting vma activity\n \n Before we acquire the vma for GPU activity, ensure that the underlying\n object is not already in the process of being freed.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820100531.8430-1-chris@chris-wilson.co.uk\n\ncommit 44c22f3f1a0a1a24b7d9b0afdd77bbb949e05ba1\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Aug 20 09:09:07 2019 +0100\n\n drm/i915: Serialize insertion into the file-\u003emm.request_list\n \n Currently, we remove the from per-file request list for throttling and\n retirement under a dedicated spinlock, but insertion is governed by\n struct_mutex. This needs to be the same lock so that the\n retirement/insertion of neighbouring requests (at the tail) doesn\u0027t\n break the list.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820080907.4665-1-chris@chris-wilson.co.uk\n\ncommit 149d6deb880c3a9002161bb6f978bc81c7c1ca8e\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Aug 16 12:55:23 2019 +0300\n\n drm/i915: Sanitize PHY state during display core uninit\n \n To work around a DMC/Punit issue on ICL where the driver\u0027s\n ICL_PORT_COMP_DW8/IREFGEN PHY setting is lost when entering/exiting DC6\n state, make sure to reinit the PHY whenever disabling DC states.\n Similarly the driver\u0027s PHY/DBUF/CDCLK settings should have been preserved\n across DC5/6 transitions, so check this on all platforms.\n \n This gets rid of the following WARN during suspend:\n Combo PHY A HW state changed unexpectedly\n \n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816095523.15800-1-imre.deak@intel.com\n\ncommit 5f912f7ced427db56729e6419b1fdff9226e0728\nAuthor: Peter Griffin \u003cpeter.griffin@linaro.org\u003e\nDate: Mon Aug 19 23:48:40 2019 +0000\n\n dt-bindings: reset: hisilicon: Add ao reset controller\n \n This is required to bring Mali450 gpu out of reset.\n \n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: devicetree@vger.kernel.org\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Peter Griffin \u003cpeter.griffin@linaro.org\u003e\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 33a1c553d026f18e5dd6bde5c50ad16b57553f55\nAuthor: Peter Griffin \u003cpeter.griffin@linaro.org\u003e\nDate: Mon Aug 19 23:48:39 2019 +0000\n\n dt-bindings: reset: hisilicon: Update compatible documentation\n \n The reset driver now supports the ao reset controller, so update the\n documentation to match.\n \n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: devicetree@vger.kernel.org\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Peter Griffin \u003cpeter.griffin@linaro.org\u003e\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit f84470749bb9d849678cc852f26b0968bc361b83\nAuthor: Peter Griffin \u003cpeter.griffin@linaro.org\u003e\nDate: Mon Aug 19 23:48:38 2019 +0000\n\n dt-bindings: gpu: mali-utgard: add hisilicon,hi6220-mali compatible\n \n The Hisilicon hi6220 uses a Mali-450MP4 with 4 PPs, so add\n a compatible for it.\n \n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: devicetree@vger.kernel.org\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Peter Griffin \u003cpeter.griffin@linaro.org\u003e\n Signed-off-by: John Stultz \u003cjohn.stultz@linaro.org\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit a63af99f0d0d1c0992736f1aaf9e3b42bee0e2b8\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Sun Aug 18 21:01:13 2019 +0300\n\n spi: spi-fsl-dspi: Move dspi_interrupt above dspi_transfer_one_message\n \n The two functions are loosely coupled through dspi-\u003ewaitq, but\n logically, dspi_transfer_one_message depends on dspi_interrupt in order\n to complete. Move its definition above it so the I/O functions are\n grouped closer together.\n \n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190818180115.31114-13-olteanv@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 404be07f4ed27697f5fa69162f67a94555738595\nAuthor: Daniel Stuart \u003cdaniel.stuart14@gmail.com\u003e\nDate: Thu Aug 15 14:12:55 2019 -0300\n\n ASoC: intel: cht_bsw_max98090_ti: Add all Chromebooks that need pmc_plt_clk_0 quirk\n \n Every single baytrail chromebook sets PMC to 0, as can be seeing\n below by searching through coreboot source code:\n $ grep -rl \"PMC_PLT_CLK\\[0\\]\" .\n ./rambi/variants/glimmer/devicetree.cb\n ./rambi/variants/clapper/devicetree.cb\n ./rambi/variants/swanky/devicetree.cb\n ./rambi/variants/enguarde/devicetree.cb\n ./rambi/variants/winky/devicetree.cb\n ./rambi/variants/kip/devicetree.cb\n ./rambi/variants/squawks/devicetree.cb\n ./rambi/variants/orco/devicetree.cb\n ./rambi/variants/ninja/devicetree.cb\n ./rambi/variants/heli/devicetree.cb\n ./rambi/variants/sumo/devicetree.cb\n ./rambi/variants/banjo/devicetree.cb\n ./rambi/variants/candy/devicetree.cb\n ./rambi/variants/gnawty/devicetree.cb\n ./rambi/variants/rambi/devicetree.cb\n ./rambi/variants/quawks/devicetree.cb\n \n Plus, Cyan (only non-baytrail chromebook with max98090) also needs\n this patch for audio to work.\n \n Thus, this commit adds all the missing devices to bsw_max98090 quirk\n table, implemented by commit a182ecd3809c (\"ASoC: intel:\n cht_bsw_max98090_ti: Add quirk for boards using pmc_plt_clk_0\").\n \n Signed-off-by: Daniel Stuart \u003cdaniel.stuart14@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190815171300.30126-1-daniel.stuart14@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 351b31002c1853af078ebfffd4b67bfc3d19e3dd\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Tue Aug 20 14:35:10 2019 +0200\n\n ASoC: meson: g12a-tohdmitx: require regmap mmio\n \n The tohdmitx glue uses regmap MMIO so it should require it.\n \n Fixes: c8609f3870f7 (\"ASoC: meson: add g12a tohdmitx control\")\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Link: https://lore.kernel.org/r/20190820123510.22491-1-jbrunet@baylibre.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 18dd62ae3bc31baa0473e4a09e46c02e0bdc57a0\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Tue Aug 20 14:34:13 2019 +0200\n\n ASoC: meson: axg-tdm-formatter: free reset on device removal\n \n Use the devm variant to get the formatter reset so it is properly freed\n on device removal\n \n Fixes: 751bd5db5260 (\"ASoC: meson: axg-tdm-formatter: add reset\")\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Link: https://lore.kernel.org/r/20190820123413.22249-1-jbrunet@baylibre.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 206175306da1de0c116e8bc12db3230edb230d5a\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Sun Aug 18 21:01:12 2019 +0300\n\n spi: spi-fsl-dspi: Fix typos\n \n mask of -\u003e mask off\n at and -\u003e and\n \n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190818180115.31114-12-olteanv@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit d6bdfa6c203d16e7f290ecc34137c396b1602782\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Sun Aug 18 21:01:11 2019 +0300\n\n spi: spi-fsl-dspi: Use reverse Christmas tree declaration order\n \n This patch puts variable declaration in the reverse order of their\n length for cosmetic purposes.\n \n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190818180115.31114-11-olteanv@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 3a11ea664be6ff16387759d57aed3a015f6339a1\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Sun Aug 18 21:01:10 2019 +0300\n\n spi: spi-fsl-dspi: Replace legacy spi_master names with spi_controller\n \n This adapts the spi-fsl-dspi driver to the API changes introduced in\n commit 8caab75fd2c2 (\"spi: Generalize SPI \"master\" to \"controller\"\").\n \n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190818180115.31114-10-olteanv@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit bee44359169ecf293a0bd2351ada186a0864a94a\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Sun Aug 18 21:01:09 2019 +0300\n\n spi: spi-fsl-dspi: Remove pointless assignment of master-\u003etransfer to NULL\n \n Introduced in commit 9298bc727385 (\"spi: spi-fsl-dspi: Remove\n spi-bitbang\") for less than obvious reasons, this assignment is\n confusing and serves no purpose.\n \n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190818180115.31114-9-olteanv@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9b536d3647ee1348860657c239d247364e4374ec\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Sun Aug 18 21:01:08 2019 +0300\n\n spi: spi-fsl-dspi: Remove unused initialization of \u0027ret\u0027 in dspi_probe\n \n There is no code path for reaching \u0027return ret;\u0027 without it first being\n assigned to an error code. Therefore the initialization with 0 is\n pointless.\n \n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190818180115.31114-8-olteanv@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit abbd0ef1f2542621e0f94a7273b7f7e458a38497\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Sun Aug 18 21:01:07 2019 +0300\n\n spi: spi-fsl-dspi: Reduce indentation in dspi_release_dma()\n \n There is no point in surrounding an entire function block in an if\n condition. Rather, exit early if the condition is false.\n \n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190818180115.31114-7-olteanv@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 06d5dd29976fb49f3236b1e66142680619ea259d\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Sun Aug 18 21:01:06 2019 +0300\n\n spi: spi-fsl-dspi: Change usage pattern of SPI_MCR_* and SPI_CTAR_* macros\n \n These are macros that accept 0 or 1 as argument (a boolean value). Their\n use encourages the abuse of complex ternary operations inside their\n argument list, which detracts from the code readability. Replace these\n with simple if-else statements.\n \n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190818180115.31114-6-olteanv@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9e6f784e66cb865de25a1f5fd1706a45d9e298a7\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Sun Aug 18 21:01:05 2019 +0300\n\n spi: spi-fsl-dspi: Demistify magic value in SPI_SR_CLEAR\n \n This patch adds the field definitions for the SPI_SR register. The SPI\n status register is write-1-to-clear and this value is written at init\n time.\n \n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190818180115.31114-5-olteanv@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit b2655196cf9ce783d0b6cfec1f47ac4ecf3dcf95\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Sun Aug 18 21:01:04 2019 +0300\n\n spi: spi-fsl-dspi: Use BIT() and GENMASK() macros\n \n Switch to using more idiomatic register field definitions, which makes\n it easier to look them up in the datasheet. Cosmetic patch.\n \n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190818180115.31114-4-olteanv@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 5782a17fe528ce596eb5de66777a73b16f4f1549\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Sun Aug 18 21:01:03 2019 +0300\n\n spi: spi-fsl-dspi: Remove unused defines and includes\n \n This is a cosmetic patch.\n \n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190818180115.31114-3-olteanv@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 50fcd84764fcbf40b64a8798f12a3d4be5dc8777\nAuthor: Vladimir Oltean \u003colteanv@gmail.com\u003e\nDate: Sun Aug 18 21:01:02 2019 +0300\n\n spi: spi-fsl-dspi: Fix code alignment\n \n This is a cosmetic patch that changes nothing except makes sure the code\n is aligned to the same column, which makes it easier to the eye.\n \n Signed-off-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190818180115.31114-2-olteanv@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 6ac3a0ebfcc2f0c75ca0ca6974389ce421aa5cbd\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Aug 20 13:21:18 2019 +0100\n\n dmabuf: Mark up onstack timer for selftests\n \n The dma-fence selftest uses an on-stack timer that requires explicit\n annotation for debugobjects.\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111442\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Acked-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190820122118.13698-1-chris@chris-wilson.co.uk\n\ncommit 2969a78aead38b49e80c821a5c683544ab16160d\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Thu Aug 8 19:25:47 2019 +0300\n\n drm/i915: Fix HW readout for crtc_clock in HDMI mode\n \n The conversion during HDMI HW readout from port_clock to crtc_clock was\n missed when HDMI 10bpc support was added, so fix that.\n \n v2:\n - Unscrew the non-HDMI case.\n \n Fixes: cd9e11a8bf25 (\"drm/i915/icl: Add 10-bit support for hdmi\")\n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d109593\n Cc: Radhakrishna Sripada \u003cradhakrishna.sripada@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808162547.7009-1-imre.deak@intel.com\n\ncommit 6869b7b206595ae0e326f59719090351eb8f4f5d\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Sun Aug 18 11:05:57 2019 +0200\n\n memremap: provide a not device managed memremap_pages\n \n The kvmppc ultravisor code wants a device private memory pool that is\n system wide and not attached to a device. Instead of faking up one\n provide a low-level memremap_pages for it.\n \n Link: https://lore.kernel.org/r/20190818090557.17853-5-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Ira Weiny \u003cira.weiny@intel.com\u003e\n Reviewed-by: Dan Williams \u003cdan.j.williams@intel.com\u003e\n Tested-by: Bharata B Rao \u003cbharata@linux.ibm.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 6f42193fd86eb5eb518984032c85e9b5582b7625\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Sun Aug 18 11:05:56 2019 +0200\n\n memremap: don\u0027t use a separate devm action for devmap_managed_enable_get\n \n Just clean up for early failures and then piggy back on\n devm_memremap_pages_release. This helps with a pending not device managed\n version of devm_memremap_pages.\n \n Link: https://lore.kernel.org/r/20190818090557.17853-4-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Ira Weiny \u003cira.weiny@intel.com\u003e\n Reviewed-by: Dan Williams \u003cdan.j.williams@intel.com\u003e\n Tested-by: Bharata B Rao \u003cbharata@linux.ibm.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit fdc029b19dfd190840d23e5c70bd231140b0e4a4\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Sun Aug 18 11:05:55 2019 +0200\n\n memremap: remove the dev field in struct dev_pagemap\n \n The dev field in struct dev_pagemap is only used to print dev_name in two\n places, which are at best nice to have. Just remove the field and thus\n the name in those two messages.\n \n Link: https://lore.kernel.org/r/20190818090557.17853-3-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Ira Weiny \u003cira.weiny@intel.com\u003e\n Reviewed-by: Dan Williams \u003cdan.j.williams@intel.com\u003e\n Tested-by: Bharata B Rao \u003cbharata@linux.ibm.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 0c385190392d8c7128fb7517b3c676e19c7b8808\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Sun Aug 18 11:05:54 2019 +0200\n\n resource: add a not device managed request_free_mem_region variant\n \n Factor out the guts of devm_request_free_mem_region so that we can\n implement both a device managed and a manually release version as tiny\n wrappers around it.\n \n Link: https://lore.kernel.org/r/20190818090557.17853-2-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Ira Weiny \u003cira.weiny@intel.com\u003e\n Reviewed-by: Dan Williams \u003cdan.j.williams@intel.com\u003e\n Tested-by: Bharata B Rao \u003cbharata@linux.ibm.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 8402ce61bec28a1b73ec467b4c146f1e3f587044\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Wed Aug 14 22:20:23 2019 +0200\n\n mm/mmu_notifiers: check if mmu notifier callbacks are allowed to fail\n \n Just a bit of paranoia, since if we start pushing this deep into\n callchains it\u0027s hard to spot all places where an mmu notifier\n implementation might fail when it\u0027s not allowed to.\n \n Inspired by some confusion we had discussing i915 mmu notifiers and\n whether we could use the newly-introduced return value to handle some\n corner cases. Until we realized that these are only for when a task has\n been killed by the oom reaper.\n \n An alternative approach would be to split the callback into two versions,\n one with the int return value, and the other with void return value like\n in older kernels. But that\u0027s a lot more churn for fairly little gain I\n think.\n \n Summary from the m-l discussion on why we want something at warning level:\n This allows automated tooling in CI to catch bugs without humans having to\n look at everything. If we just upgrade the existing pr_info to a pr_warn,\n then we\u0027ll have false positives. And as-is, no one will ever spot the\n problem since it\u0027s lost in the massive amounts of overall dmesg noise.\n \n Link: https://lore.kernel.org/r/20190814202027.18735-2-daniel.vetter@ffwll.ch\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 9b2ed9cb975c63f5534daaebeb225ab52b589372\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Wed Aug 14 09:59:28 2019 +0200\n\n mm: remove CONFIG_MIGRATE_VMA_HELPER\n \n CONFIG_MIGRATE_VMA_HELPER guards helpers that are required for proper\n devic private memory support. Remove the option and just check for\n CONFIG_DEVICE_PRIVATE instead.\n \n Link: https://lore.kernel.org/r/20190814075928.23766-11-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Tested-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 06d462beb470d361ffa8bd7b3d865509a8606987\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Wed Aug 14 09:59:27 2019 +0200\n\n mm: remove the unused MIGRATE_PFN_DEVICE flag\n \n No one ever checks this flag, and we could easily get that information\n from the page if needed.\n \n Link: https://lore.kernel.org/r/20190814075928.23766-10-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Tested-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 2a915acf88ac0c4c583cd654bdc87371e38016c1\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Wed Aug 14 09:59:26 2019 +0200\n\n mm: remove the unused MIGRATE_PFN_ERROR flag\n \n Now that we can rely errors in the normal control flow there is no\n need for this flag, remove it.\n \n Link: https://lore.kernel.org/r/20190814075928.23766-9-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Tested-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit f268307ec7c728fbc042608c7505926460117816\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Wed Aug 14 09:59:25 2019 +0200\n\n nouveau: simplify nouveau_dmem_migrate_vma\n \n Factor the main copy page to vram routine out into a helper that acts\n on a single page and which doesn\u0027t require the nouveau_dmem_migrate\n structure for argument passing. As an added benefit the new version\n only allocates the dma address array once and reuses it for each\n subsequent chunk of work.\n \n Link: https://lore.kernel.org/r/20190814075928.23766-8-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Tested-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit bfe69ef94aeeae76531b8f37799b8ac6fca56957\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Wed Aug 14 09:59:24 2019 +0200\n\n nouveau: simplify nouveau_dmem_migrate_to_ram\n \n Factor the main copy page to ram routine out into a helper that acts on\n a single page and which doesn\u0027t require the nouveau_dmem_fault\n structure for argument passing. Also remove the loop over multiple\n pages as we only handle one at the moment, although the structure of\n the main worker function makes it relatively easy to add multi page\n support back if needed in the future. But at least for now this avoid\n the needed to dynamically allocate memory for the dma addresses in\n what is essentially the page fault path.\n \n Link: https://lore.kernel.org/r/20190814075928.23766-7-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Tested-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 107ba59fc6d761ca996243dd0a1ee89addc295af\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Wed Aug 14 09:59:23 2019 +0200\n\n nouveau: remove a few function stubs\n \n nouveau_dmem_migrate_vma and nouveau_dmem_convert_pfn are only called\n when CONFIG_DRM_NOUVEAU_SVM is enabled, so there is no need to provide\n !CONFIG_DRM_NOUVEAU_SVM stubs for them.\n \n Link: https://lore.kernel.org/r/20190814075928.23766-6-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Tested-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 2ab2bda53c986a93e46228bff4362b6267202f81\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Wed Aug 14 09:59:22 2019 +0200\n\n nouveau: factor out dmem fence completion\n \n Factor out the end of fencing logic from the two migration routines.\n \n Link: https://lore.kernel.org/r/20190814075928.23766-5-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Tested-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 64de8b8d65f9bfd0fced9029bcec37257614ccaf\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Wed Aug 14 09:59:21 2019 +0200\n\n nouveau: factor out device memory address calculation\n \n Factor out the repeated device memory address calculation into\n a helper.\n \n Link: https://lore.kernel.org/r/20190814075928.23766-4-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Tested-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit dea027f28209aa7a23a4e56b2ae270f16d48055d\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Wed Aug 14 09:59:20 2019 +0200\n\n nouveau: reset dma_nr in nouveau_dmem_migrate_alloc_and_copy\n \n When we start a new batch of dma_map operations we need to reset dma_nr,\n as we start filling a newly allocated array.\n \n Link: https://lore.kernel.org/r/20190814075928.23766-3-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Tested-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit a7d1f22bb74f32cf3cd93f52776007e161f1a738\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Wed Aug 14 09:59:19 2019 +0200\n\n mm: turn migrate_vma upside down\n \n There isn\u0027t any good reason to pass callbacks to migrate_vma. Instead\n we can just export the three steps done by this function to drivers and\n let them sequence the operation without callbacks. This removes a lot\n of boilerplate code as-is, and will allow the drivers to drastically\n improve code flow and error handling further on.\n \n Link: https://lore.kernel.org/r/20190814075928.23766-2-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Tested-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 471f390205698ed5225af052a80b71d2bce6b93a\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Tue Aug 6 20:15:47 2019 -0300\n\n drm/amdkfd: use mmu_notifier_put\n \n The sequence of mmu_notifier_unregister_no_release(),\n mmu_notifier_call_srcu() is identical to mmu_notifier_put() with the\n free_notifier callback.\n \n As this is the last user of those APIs, converting it means we can drop\n them.\n \n Link: https://lore.kernel.org/r/20190806231548.25242-11-jgg@ziepe.ca\n Reviewed-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 0029cab3146a53b820e73c9cd70809b831c590d7\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Tue Aug 6 20:15:46 2019 -0300\n\n drm/amdkfd: fix a use after free race with mmu_notifer unregister\n \n When using mmu_notifer_unregister_no_release() the caller must ensure\n there is a SRCU synchronize before the mn memory is freed, otherwise use\n after free races are possible, for instance:\n \n CPU0 CPU1\n invalidate_range_start\n hlist_for_each_entry_rcu(..)\n mmu_notifier_unregister_no_release(\u0026p-\u003emn)\n kfree(mn)\n if (mn-\u003eops-\u003einvalidate_range_end)\n \n The error unwind in amdkfd misses the SRCU synchronization.\n \n amdkfd keeps the kfd_process around until the mm is released, so split the\n flow to fully initialize the kfd_process and register it for find_process,\n and with the notifier. Past this point the kfd_process does not need to be\n cleaned up as it is fully ready.\n \n The final failable step does a vm_mmap() and does not seem to impact the\n kfd_process global state. Since it also cannot be undone (and already has\n problems with undo if it internally fails), it has to be last.\n \n This way we don\u0027t have to try to unwind the mmu_notifier_register() and\n avoid the problem with the SRCU.\n \n Along the way this also fixes various other error unwind bugs in the flow.\n \n Fixes: 45102048f77e (\"amdkfd: Add process queue manager module\")\n Link: https://lore.kernel.org/r/20190806231548.25242-10-jgg@ziepe.ca\n Reviewed-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 534e5f84b7a9c1e139ee8944fee6313d3a24d5d3\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Tue Aug 6 20:15:45 2019 -0300\n\n drm/radeon: use mmu_notifier_get/put for struct radeon_mn\n \n radeon is using a device global hash table to track what mmu_notifiers\n have been registered on struct mm. This is better served with the new\n get/put scheme instead.\n \n radeon has a bug where it was not blocking notifier release() until all\n the BO\u0027s had been invalidated. This could result in a use after free of\n pages the BOs. This is tied into a second bug where radeon left the\n notifiers running endlessly even once the interval tree became\n empty. This could result in a use after free with module unload.\n \n Both are fixed by changing the lifetime model, the BOs exist in the\n interval tree with their natural lifetimes independent of the mm_struct\n lifetime using the get/put scheme. The release runs synchronously and just\n does invalidate_start across the entire interval tree to create the\n required DMA fence.\n \n Additions to the interval tree after release are already impossible as\n only current-\u003emm is used during the add.\n \n Link: https://lore.kernel.org/r/20190806231548.25242-9-jgg@ziepe.ca\n Acked-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit c7d8b7824ff9de866a356e1892dbe9f191aa5d06\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Tue Aug 6 20:15:42 2019 -0300\n\n hmm: use mmu_notifier_get/put for \u0027struct hmm\u0027\n \n This is a significant simplification, it eliminates all the remaining\n \u0027hmm\u0027 stuff in mm_struct, eliminates krefing along the critical notifier\n paths, and takes away all the ugly locking and abuse of page_table_lock.\n \n mmu_notifier_get() provides the single struct hmm per struct mm which\n eliminates mm-\u003ehmm.\n \n It also directly guarantees that no mmu_notifier op callback is callable\n while concurrent free is possible, this eliminates all the krefs inside\n the mmu_notifier callbacks.\n \n The remaining krefs in the range code were overly cautious, drivers are\n already not permitted to free the mirror while a range exists.\n \n Link: https://lore.kernel.org/r/20190806231548.25242-6-jgg@ziepe.ca\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Tested-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 7f01494fd85f3ac40635bc2a7bf005b607084c96\nAuthor: Manivannan Sadhasivam \u003cmanivannan.sadhasivam@linaro.org\u003e\nDate: Tue Aug 20 17:20:00 2019 +0530\n\n spi: Fix the number of CS lines documented as an example\n \n The number of CS lines is mentioned as 2 in the spi-controller binding\n but however in the example, 4 cs-gpios are used. Hence fix that to\n mention 4.\n \n Signed-off-by: Manivannan Sadhasivam \u003cmanivannan.sadhasivam@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190820115000.32041-1-manivannan.sadhasivam@linaro.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c7e6f0e99227b3dcdc5e62f789119e000887ff79\nAuthor: Guo Ren \u003cren_guo@c-sky.com\u003e\nDate: Tue Aug 20 20:15:44 2019 +0800\n\n csky: Support kernel non-aligned access\n \n We prohibit non-aligned access in kernel mode, but some special NIC\n driver needs to support kernel-state unaligned access. For example,\n when the bus does not support unaligned access, IP header parsing\n will cause non-aligned access and driver does not recopy the skb\n buffer to dma for performance reasons.\n \n Added kernel_enable \u0026 user_enable to control unaligned access and\n added kernel_count \u0026 user_count for statistical unaligned access.\n \n Signed-off-by: Guo Ren \u003cren_guo@c-sky.com\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 8eb2fd00f65a96143ed1535bdbf4ca4e129d30d1\nAuthor: Axel Lin \u003caxel.lin@ingics.com\u003e\nDate: Sun Aug 18 17:51:13 2019 +0800\n\n spi: zynq-qspi: Fix missing spi_unregister_controller when unload module\n \n Use devm_spi_register_controller to fix missing spi_unregister_controller\n when unload module.\n \n Signed-off-by: Axel Lin \u003caxel.lin@ingics.com\u003e\n Acked-by: Michal Simek \u003cmichal.simek@xilinx.com\u003e\n Link: https://lore.kernel.org/r/20190818095113.2397-1-axel.lin@ingics.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c372a35550c8d60f673b20210eea58a06d6d38cb\nAuthor: Kunihiko Hayashi \u003chayashi.kunihiko@socionext.com\u003e\nDate: Tue Aug 20 15:16:04 2019 +0900\n\n ASoC: uniphier: Fix double reset assersion when transitioning to suspend state\n \n When transitioning to supend state, uniphier_aio_dai_suspend() is called\n and asserts reset lines and disables clocks.\n \n However, if there are two or more DAIs, uniphier_aio_dai_suspend() are\n called multiple times, and double reset assersion will cause.\n \n This patch defines the counter that has the number of DAIs at first, and\n whenever uniphier_aio_dai_suspend() are called, it decrements the\n counter. And only if the counter is zero, it asserts reset lines and\n disables clocks.\n \n In the same way, uniphier_aio_dai_resume() are called, it increments the\n counter after deasserting reset lines and enabling clocks.\n \n Fixes: 139a34200233 (\"ASoC: uniphier: add support for UniPhier AIO CPU DAI driver\")\n Signed-off-by: Kunihiko Hayashi \u003chayashi.kunihiko@socionext.com\u003e\n Link: https://lore.kernel.org/r/1566281764-14059-1-git-send-email-hayashi.kunihiko@socionext.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit dc140045c0cace809af872e3799e8fbe1b7d7f86\nAuthor: Guo Ren \u003cren_guo@c-sky.com\u003e\nDate: Tue Aug 20 12:47:24 2019 +0800\n\n csky: Fixup defer cache flush for 610\n \n We use defer cache flush mechanism to improve the performance of\n 610, but the implementation is wrong. We fix it up now and update\n the mechanism:\n \n - Zero page needn\u0027t be flushed.\n - If page is file mapping \u0026 non-touched in user space, defer flush.\n - If page is anon mapping or dirty file mapping, flush immediately.\n - In update_mmu_cache finish the defer flush by flush_dcache_page().\n \n For 610 we need take care the dcache aliasing issue:\n - VIPT cache with 8K-bytes size per way in 4K page granularity.\n \n Signed-off-by: Guo Ren \u003cren_guo@c-sky.com\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit d720e64150c79d14f4faf931604faa1f0755134d\nAuthor: Robin Murphy \u003crobin.murphy@arm.com\u003e\nDate: Tue Aug 20 12:38:49 2019 +0100\n\n iommu/arm-smmu: Ensure 64-bit I/O accessors are available on 32-bit CPU\n \n As part of the grand SMMU driver refactoring effort, the I/O register\n accessors were moved into \u0027arm-smmu.h\u0027 in commit 6d7dff62afb0\n (\"iommu/arm-smmu: Move Secure access quirk to implementation\").\n \n On 32-bit architectures (such as ARM), the 64-bit accessors are defined\n in \u0027linux/io-64-nonatomic-hi-lo.h\u0027, so include this header to fix the\n build.\n \n Signed-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 962411b05a6d3342aa649e39cda1704c1fc042c6\nAuthor: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nDate: Fri Aug 16 01:56:08 2019 -0500\n\n dmaengine: ti: omap-dma: Add cleanup in omap_dma_probe()\n \n If devm_request_irq() fails to disable all interrupts, no cleanup is\n performed before retuning the error. To fix this issue, invoke\n omap_dma_free() to do the cleanup.\n \n Signed-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\n Acked-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Link: https://lore.kernel.org/r/1565938570-7528-1-git-send-email-wenwen@cs.uga.edu\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 2c231c0c1dec42192aca0f87f2dc68b8f0cbc7d2\nAuthor: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nDate: Fri Aug 16 01:48:55 2019 -0500\n\n dmaengine: ti: dma-crossbar: Fix a memory leak bug\n \n In ti_dra7_xbar_probe(), \u0027rsv_events\u0027 is allocated through kcalloc(). Then\n of_property_read_u32_array() is invoked to search for the property.\n However, if this process fails, \u0027rsv_events\u0027 is not deallocated, leading to\n a memory leak bug. To fix this issue, free \u0027rsv_events\u0027 before returning\n the error.\n \n Signed-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\n Acked-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Link: https://lore.kernel.org/r/1565938136-7249-1-git-send-email-wenwen@cs.uga.edu\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit df58525df395561551440392d92daa6416b4f8f3\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Mon Aug 19 17:08:29 2019 +0200\n\n can: mcp251x: remove custom DMA mapped buffer\n \n There is no need to duplicate what SPI core already does, i.e. mapping buffers\n for DMA capable transfers. This patch removes all related pices of code.\n \n Tested-by: Sean Nyekjaer \u003csean@geanix.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 6a07c2305ab200844f250f4dba47ed60d0640505\nAuthor: Phil Elwell \u003cphil@raspberrypi.org\u003e\nDate: Tue Nov 14 11:03:22 2017 +0000\n\n can: mcp251x: Use DT-supplied interrupt flags\n \n The MCP2515 datasheet clearly describes a level-triggered interrupt pin.\n Therefore the receiving interrupt controller must also be configured for\n level-triggered operation otherwise there is a danger of a missed\n interrupt condition blocking all subsequent interrupts. The ONESHOT\n flag ensures that the interrupt is masked until the threaded interrupt\n handler exits.\n \n Rather than change the flags globally (they must have worked for at\n least one user), keep the old behavior for for non DT devices. DT based\n devices specify the flags in their corresonding DT node.\n \n See: https://github.com/raspberrypi/linux/issues/2175\n https://github.com/raspberrypi/linux/issues/2263\n \n Signed-off-by: Phil Elwell \u003cphil@raspberrypi.org\u003e\n Tested-by: Sean Nyekjaer \u003csean@geanix.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 3964576307edf408435949633b1f600262d70bb6\nAuthor: Alexander Shiyan \u003cshc_work@mail.ru\u003e\nDate: Thu Jan 31 11:46:11 2019 +0300\n\n can: mcp251x: Use dev_name() during request_threaded_irq()\n \n Passing driver name as name during request_threaded_irq() results in all\n CAN IRQs have same name. This does not help much to easily identify which\n IRQ belongs to which CAN instance. Therefore pass dev_name() during\n request_threaded_irq() so that better identifiable name is listed for\n CAN devices in cat /proc/interrupts output.\n \n Output of cat /proc/interrupts\n Before this patch:\n 253: 2 gpio-mxc 13 Edge mcp251x\n 259: 2 gpio-mxc 19 Edge mcp251x\n After this patch:\n 253: 2 gpio-mxc 13 Edge spi1.1\n 259: 2 gpio-mxc 19 Edge spi1.2\n \n Signed-off-by: Alexander Shiyan \u003cshc_work@mail.ru\u003e\n Tested-by: Sean Nyekjaer \u003csean@geanix.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit d84ea2123f8d27144e3f4d58cd88c9c6ddc799de\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Tue Aug 13 16:01:02 2019 +0200\n\n can: mcp251x: mcp251x_hw_reset(): allow more time after a reset\n \n Some boards take longer than 5ms to power up after a reset, so allow\n some retries attempts before giving up.\n \n Fixes: ff06d611a31c (\"can: mcp251x: Improve mcp251x_hw_reset()\")\n Cc: linux-stable \u003cstable@vger.kernel.org\u003e\n Tested-by: Sean Nyekjaer \u003csean@geanix.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit af669cd26e0ac4ba7bd6743cdd5ef10e9332129e\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Tue Aug 13 14:26:45 2019 +0200\n\n can: mcp251x: use u8 instead of uint8_t\n \n This patch changes all the uint8_t in the arguments in several function\n to u8.\n \n Acked-by: Sean Nyekjaer \u003csean@geanix.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit d344c6d6c33f0a22f6166730eb02a8e9db589685\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Tue Aug 13 15:33:01 2019 +0200\n\n can: mcp251x: fix print formating strings\n \n This patch fixes the print format strings in the driver.\n \n Acked-by: Sean Nyekjaer \u003csean@geanix.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 4669597496dc7fce034cc595baeaff9c1f1b6412\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Wed Jul 24 14:34:42 2019 +0200\n\n can: mcp251x: avoid long lines\n \n This patch fixes long lines in the driver.\n \n Acked-by: Sean Nyekjaer \u003csean@geanix.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 77654a6da00f7a712288f2684229bf63b9c41760\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Wed Jul 24 14:28:21 2019 +0200\n\n can: mcp251x: remove unnecessary blank lines\n \n This patch removes unnecessary blank lines, so that checkpatch doesn\u0027t\n complain anymore.\n \n Acked-by: Sean Nyekjaer \u003csean@geanix.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 3b9bcede4d9fc00cdab064529cec62699a83bbed\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Wed Jul 24 14:16:29 2019 +0200\n\n can: mcp251x: convert block comments to network style comments\n \n This patch converts all block comments to network subsystem style block\n comments.\n \n Acked-by: Sean Nyekjaer \u003csean@geanix.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit b3402c405747c4b6a9fe2d0a225835d95abd8496\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Mon Aug 19 19:34:28 2019 +0200\n\n can: m_can_platform: m_can_plat_probe(): add missing error handling if mcan_class is NULL\n \n This patch adds the missing error handling in m_can_plat_probe() if\n mcan_class is NULL.\n \n Fixes: f524f829b75a (\"can: m_can: Create a m_can platform framework\")\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 28b0ffe98b53cf1f3e58023699886f813fb26144\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Mon Aug 19 19:17:13 2019 +0200\n\n can: m_can_platform: remove not needed casts to struct m_can_plat_priv *\n \n The struct m_can_classdev::device_data is a void pointer, so there\u0027s no\n need to cast it to struct m_can_plat_priv *, when assigning the struct\n m_can_plat_priv pointer.\n \n This patch removes the not needed casts from the m_can_platform driver.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 6093f744fec77e06f98b0267ced88b9caa64eae8\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Fri Aug 16 10:44:49 2019 +0200\n\n can: tcan4x5x: fix data length in regmap write path\n \n In regmap_spi_gather_write() the \"addr\" is prepared. The chip expects\n the number of 32 bit words to write in the lower 8 bits of addr. However\n the number of byte to write in shifted left by 3 (\u003d\u003d divided by 8).\n \n The function tcan4x5x_regmap_write() is called with a data buffer, which\n holds the register information in the first 32 bits, followed by the\n actual data. tcan4x5x_regmap_write() calls regmap_spi_gather_write()\n with the val pointer pointing to the actual data (i.e. the original\n pointer is incremented by 4 bytes), but without decrementing the count.\n \n If the regmap framework only calls tcan4x5x_regmap_write() to read\n single 32 bit registers these two bugs cancel each other.\n \n This patch fixes the code.\n \n Fixes: 5443c226ba91 (\"can: tcan4x5x: Add tcan4x5x driver to the kernel\")\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 7fbda1306542d89984547faf6bfce584e182a541\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Mon Aug 19 19:34:28 2019 +0200\n\n can: tcan4x5x: tcan4x5x_can_probe(): add missing error handling if mcan_class is NULL\n \n This patch adds the missing error handling in tcan4x5x_can_probe() if\n mcan_class is NULL.\n \n Fixes: 5443c226ba91 (\"can: tcan4x5x: Add tcan4x5x driver to the kernel\")\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit ad07819f223e26b30cc5ec25a31d906db0788600\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Mon Aug 19 19:17:13 2019 +0200\n\n can: tcan4x5x: remove not needed casts to struct tcan4x5x_priv *\n \n The struct m_can_classdev::device_data is a void pointer, so there\u0027s no\n need to cast it to struct tcan4x5x_priv *, when assigning the struct\n tcan4x5x_priv pointer.\n \n This patch removes the not needed casts from the tcan4x5x driver.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 65668b32695652e81c86671f075ed45f9efcf0d7\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Fri Aug 16 18:44:04 2019 +0200\n\n can: tcan4x5x: remove unused struct tcan4x5x_priv::tcan4x5x_lock\n \n The mutex struct tcan4x5x_priv::tcan4x5x_lock is unused in the driver,\n so this patch removes the variable from the driver.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 653ee35ce6d58261ef507cb270a5bd534813db8a\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Mon Aug 19 17:08:29 2019 +0200\n\n can: hi311x: remove custom DMA mapped buffer\n \n There is no need to duplicate what SPI core already does, i.e. mapping buffers\n for DMA capable transfers. This patch removes all related pices of code.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 1f0dee39e3cb272c32bb878d0a78d91300326d2b\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Aug 19 13:30:18 2019 +0530\n\n can: peak_pci: Make structure peak_pciec_i2c_bit_ops constant\n \n Static structure peak_pciec_i2c_bit_ops, of type i2c_algo_bit_data, is\n not used except to be copied into another variable. Hence make it const\n to protect it from modification.\n \n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 30cc0ed73e3390af7c5573b159826e8162f09fe7\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Aug 14 11:22:21 2019 +0200\n\n can: rcar_can: Remove unused platform data support\n \n All R-Car platforms use DT for describing CAN controllers. R-Car CAN\n platform data support was never used in any upstream kernel.\n \n Move the Clock Select Register settings enum into the driver, and remove\n platform data support and the corresponding header file.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit d071fd294f2474118629f4021a6a3dedef28e09f\nAuthor: Mao Wenan \u003cmaowenan@huawei.com\u003e\nDate: Wed Aug 14 15:21:05 2019 +0800\n\n dmaengine: change alignment of mux_configure32 and fsl_edma_chan_mux\n \n The alignment of mux_configure32() and fsl_edma_chan_mux() need\n to be adjusted, it must start precisely at the first column after\n the openning parenthesis of the first line.\n \n Fixes: 9d831528a656 (\"dmaengine: fsl-edma: extract common fsl-edma code (no changes in behavior intended)\")\n Signed-off-by: Mao Wenan \u003cmaowenan@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190814072105.144107-3-maowenan@huawei.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 4f48e29f7673726a8c1a18b10906ccd58a031b95\nAuthor: Mao Wenan \u003cmaowenan@huawei.com\u003e\nDate: Wed Aug 14 15:21:04 2019 +0800\n\n dmaengine: make mux_configure32 static\n \n There is one sparse warning in drivers/dma/fsl-edma-common.c:\n drivers/dma/fsl-edma-common.c:93:6: warning: symbol \u0027mux_configure32\u0027\n was not declared. Should it be static?\n \n Fix it by setting mux_configure32() as static.\n \n Fixes: 232a7f18cf8ec (\"dmaengine: fsl-edma: add i.mx7ulp edma2 version support\")\n Signed-off-by: Mao Wenan \u003cmaowenan@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190814072105.144107-2-maowenan@huawei.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 698f7a9be9b2389384c69750d87a0c68202c0621\nAuthor: Nathan Huckleberry \u003cnhuck@google.com\u003e\nDate: Tue Aug 13 10:34:48 2019 -0700\n\n dmaengine: mv_xor_v2: Fix -Wshift-negative-value\n \n Clang produces the following warning\n \n drivers/dma/mv_xor_v2.c:264:40: warning: shifting a negative signed value\n is undefined [-Wshift-negative-value]\n reg \u0026\u003d (~MV_XOR_V2_DMA_IMSG_THRD_MASK \u003c\u003c\n MV_XOR_V2_DMA_IMSG_THRD_SHIFT);\n ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ^\n drivers/dma/mv_xor_v2.c:271:46: warning: shifting a negative signed value\n is undefined [-Wshift-negative-value]\n reg \u0026\u003d (~MV_XOR_V2_DMA_IMSG_TIMER_THRD_MASK \u003c\u003c\n ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ^\n \n Upon further investigation MV_XOR_V2_DMA_IMSG_THRD_SHIFT and\n MV_XOR_V2_DMA_IMSG_TIMER_THRD_SHIFT are both 0. Since shifting by 0 does\n nothing, these variables can be removed.\n \n Cc: clang-built-linux@googlegroups.com\n Link: https://github.com/ClangBuiltLinux/linux/issues/521\n Signed-off-by: Nathan Huckleberry \u003cnhuck@google.com\u003e\n Reviewed-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Link: https://lore.kernel.org/r/20190813173448.109859-1-nhuck@google.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 3a1144f83c531dea36bad013b04888fd6e69aa0b\nAuthor: Denis Efremov \u003cefremov@linux.com\u003e\nDate: Tue Aug 13 09:00:04 2019 +0300\n\n MAINTAINERS: dmaengine: dw axi dmac: Fix typo in a path\n \n Fix typo (s/dwi-/dw-/) in the directory path.\n \n Fixes: 1fe20f1b8454 (\"dmaengine: Introduce DW AXI DMAC driver\")\n Signed-off-by: Denis Efremov \u003cefremov@linux.com\u003e\n Link: https://lore.kernel.org/r/20190813060004.10594-1-efremov@linux.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit ccfb5bd71d3d1228090a8633800ae7cdf42a94ac\nAuthor: Nathan Lynch \u003cnathanl@linux.ibm.com\u003e\nDate: Fri Aug 2 14:29:26 2019 -0500\n\n powerpc/pseries/mobility: use cond_resched when updating device tree\n \n After a partition migration, pseries_devicetree_update() processes\n changes to the device tree communicated from the platform to\n Linux. This is a relatively heavyweight operation, with multiple\n device tree searches, memory allocations, and conversations with\n partition firmware.\n \n There\u0027s a few levels of nested loops which are bounded only by\n decisions made by the platform, outside of Linux\u0027s control, and indeed\n we have seen RCU stalls on large systems while executing this call\n graph. Use cond_resched() in these loops so that the cpu is yielded\n when needed.\n \n Signed-off-by: Nathan Lynch \u003cnathanl@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802192926.19277-4-nathanl@linux.ibm.com\n\ncommit 10e4850d7c7f2af2e5c40520b8caf73bf9d7e2d1\nAuthor: Nathan Lynch \u003cnathanl@linux.ibm.com\u003e\nDate: Fri Aug 2 14:29:25 2019 -0500\n\n powerpc/rtas: allow rescheduling while changing cpu states\n \n rtas_cpu_state_change_mask() potentially operates on scores of cpus,\n so explicitly allow rescheduling in the loop body.\n \n Signed-off-by: Nathan Lynch \u003cnathanl@linux.ibm.com\u003e\n Reviewed-by: Gautham R. Shenoy \u003cego@linux.vnet.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802192926.19277-3-nathanl@linux.ibm.com\n\ncommit a6717c01ddc259f6f73364779df058e2c67309f8\nAuthor: Nathan Lynch \u003cnathanl@linux.ibm.com\u003e\nDate: Fri Aug 2 14:29:24 2019 -0500\n\n powerpc/rtas: use device model APIs and serialization during LPM\n \n The LPAR migration implementation and userspace-initiated cpu hotplug\n can interleave their executions like so:\n \n 1. Set cpu 7 offline via sysfs.\n \n 2. Begin a partition migration, whose implementation requires the OS\n to ensure all present cpus are online; cpu 7 is onlined:\n \n rtas_ibm_suspend_me -\u003e rtas_online_cpus_mask -\u003e cpu_up\n \n This sets cpu 7 online in all respects except for the cpu\u0027s\n corresponding struct device; dev-\u003eoffline remains true.\n \n 3. Set cpu 7 online via sysfs. _cpu_up() determines that cpu 7 is\n already online and returns success. The driver core (device_online)\n sets dev-\u003eoffline \u003d false.\n \n 4. The migration completes and restores cpu 7 to offline state:\n \n rtas_ibm_suspend_me -\u003e rtas_offline_cpus_mask -\u003e cpu_down\n \n This leaves cpu7 in a state where the driver core considers the cpu\n device online, but in all other respects it is offline and\n unused. Attempts to online the cpu via sysfs appear to succeed but the\n driver core actually does not pass the request to the lower-level\n cpuhp support code. This makes the cpu unusable until the cpu device\n is manually set offline and then online again via sysfs.\n \n Instead of directly calling cpu_up/cpu_down, the migration code should\n use the higher-level device core APIs to maintain consistent state and\n serialize operations.\n \n Fixes: 120496ac2d2d (\"powerpc: Bring all threads online prior to migration/hibernation\")\n Signed-off-by: Nathan Lynch \u003cnathanl@linux.ibm.com\u003e\n Reviewed-by: Gautham R. Shenoy \u003cego@linux.vnet.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802192926.19277-2-nathanl@linux.ibm.com\n\ncommit 415480dce2ef03bb8335deebd2f402f475443ce0\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Mon Aug 19 06:40:25 2019 +0000\n\n powerpc/603: Fix handling of the DIRTY flag\n \n If a page is already mapped RW without the DIRTY flag, the DIRTY\n flag is never set and a TLB store miss exception is taken forever.\n \n This is easily reproduced with the following app:\n \n void main(void)\n {\n volatile char *ptr \u003d mmap(0, 4096, PROT_READ | PROT_WRITE, MAP_SHARED | MAP_ANONYMOUS, -1, 0);\n \n *ptr \u003d *ptr;\n }\n \n When DIRTY flag is not set, bail out of TLB miss handler and take\n a minor page fault which will set the DIRTY flag.\n \n Fixes: f8b58c64eaef (\"powerpc/603: let\u0027s handle PAGE_DIRTY directly\")\n Cc: stable@vger.kernel.org # v5.1+\n Reported-by: Doug Crawford \u003cdoug.crawford@intelight-its.com\u003e\n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/80432f71194d7ee75b2f5043ecf1501cf1cca1f3.1566196646.git.christophe.leroy@c-s.fr\n\ncommit 6bb25170d7a44ef0ed9677814600f0785e7421d1\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Wed Jul 24 18:46:38 2019 +1000\n\n powerpc/64s/radix: Remove redundant pfn_pte bitop, add VM_BUG_ON\n \n pfn_pte is never given a pte above the addressable physical memory\n limit, so the masking is redundant. In case of a software bug, it\n is not obviously better to silently truncate the pfn than to corrupt\n the pte (either one will result in memory corruption or crashes),\n so there is no reason to add this to the fast path.\n \n Add VM_BUG_ON to catch cases where the pfn is invalid. These would\n catch the create_section_mapping bug fixed by a previous commit.\n \n [16885.256466] ------------[ cut here ]------------\n [16885.256492] kernel BUG at arch/powerpc/include/asm/book3s/64/pgtable.h:612!\n cpu 0x0: Vector: 700 (Program Check) at [c0000000ee0a36d0]\n pc: c000000000080738: __map_kernel_page+0x248/0x6f0\n lr: c000000000080ac0: __map_kernel_page+0x5d0/0x6f0\n sp: c0000000ee0a3960\n msr: 9000000000029033\n current \u003d 0xc0000000ec63b400\n paca \u003d 0xc0000000017f0000 irqmask: 0x03 irq_happened: 0x01\n pid \u003d 85, comm \u003d sh\n kernel BUG at arch/powerpc/include/asm/book3s/64/pgtable.h:612!\n Linux version 5.3.0-rc1-00001-g0fe93e5f3394\n enter ? for help\n [c0000000ee0a3a00] c000000000d37378 create_physical_mapping+0x260/0x360\n [c0000000ee0a3b10] c000000000d370bc create_section_mapping+0x1c/0x3c\n [c0000000ee0a3b30] c000000000071f54 arch_add_memory+0x74/0x130\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Reviewed-by: Aneesh Kumar K.V \u003caneesh.kumar@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190724084638.24982-5-npiggin@gmail.com\n\ncommit 4dd7554a6456d124c85e0a4ad156625b71390b5c\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Wed Jul 24 18:46:37 2019 +1000\n\n powerpc/64: Add VIRTUAL_BUG_ON checks for __va and __pa addresses\n \n Ensure __va is given a physical address below PAGE_OFFSET, and __pa is\n given a virtual address above PAGE_OFFSET.\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Reviewed-by: Aneesh Kumar K.V \u003caneesh.kumar@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190724084638.24982-4-npiggin@gmail.com\n\ncommit 10c4bd7cd28e77aeb8cfa65b23cb3c632ede2a49\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Wed Jul 24 18:46:36 2019 +1000\n\n powerpc/perf: fix imc allocation failure handling\n \n The alloc_pages_node return value should be tested for failure\n before being passed to page_address.\n \n Tested-by: Anju T Sudhakar \u003canju@linux.vnet.ibm.com\u003e\n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Reviewed-by: Aneesh Kumar K.V \u003caneesh.kumar@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190724084638.24982-3-npiggin@gmail.com\n\ncommit 31f210cf42d4b308eacef89b6cb0b1459338b8de\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Wed Jul 24 18:46:35 2019 +1000\n\n powerpc/64s/radix: Fix memory hot-unplug page table split\n \n create_physical_mapping expects physical addresses, but splitting\n these mapping on hot unplug is supplying virtual (effective)\n addresses.\n \n Fixes: 4dd5f8a99e791 (\"powerpc/mm/radix: Split linear mapping on hot-unplug\")\n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Reviewed-by: Aneesh Kumar K.V \u003caneesh.kumar@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190724084638.24982-2-npiggin@gmail.com\n\ncommit 8f51e3929470942e6a8744061254fdeef646cd36\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Wed Jul 24 18:46:34 2019 +1000\n\n powerpc/64s/radix: Fix memory hotplug section page table creation\n \n create_physical_mapping expects physical addresses, but creating and\n splitting these mappings after boot is supplying virtual (effective)\n addresses. This can be irritated by booting with mem\u003d to limit memory\n then probing an unused physical memory range:\n \n echo \u003caddr\u003e \u003e /sys/devices/system/memory/probe\n \n This mostly works by accident, firstly because __va(__va(x)) \u003d\u003d __va(x)\n so the virtual address does not get corrupted. Secondly because pfn_pte\n masks out the upper bits of the pfn beyond the physical address limit,\n so a pfn constructed with a 0xc000000000000000 virtual linear address\n will be masked back to the correct physical address in the pte.\n \n Fixes: 6cc27341b21a8 (\"powerpc/mm: add radix__create_section_mapping()\")\n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Reviewed-by: Aneesh Kumar K.V \u003caneesh.kumar@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190724084638.24982-1-npiggin@gmail.com\n\ncommit e354d7dc81d0e81bea33165f381aff1eda45f5d9\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Thu Jun 13 00:03:17 2019 +1000\n\n powerpc/64: allow compiler to cache \u0027current\u0027\n \n current may be cached by the compiler, so remove the volatile asm\n restriction. This results in better generated code, as well as being\n smaller and fewer dependent loads, it can avoid store-hit-load flushes\n like this one that shows up in irq_exit():\n \n preempt_count_sub(HARDIRQ_OFFSET);\n if (!in_interrupt() \u0026\u0026 ...)\n \n Which ends up as:\n \n ((struct thread_info *)current)-\u003epreempt_count -\u003d HARDIRQ_OFFSET;\n if (((struct thread_info *)current)-\u003epreempt_count ...\n \n Evaluating current twice presently means it has to be loaded twice, and\n here gcc happens to pick a different register each time, then\n preempt_count is accessed via that base register:\n \n 1058: ld r10,2392(r13) \u003c-- current\n 105c: lwz r9,0(r10) \u003c-- preempt_count\n 1060: addis r9,r9,-1\n 1064: stw r9,0(r10) \u003c-- preempt_count\n 1068: ld r9,2392(r13) \u003c-- current\n 106c: lwz r9,0(r9) \u003c-- preempt_count\n 1070: rlwinm. r9,r9,0,11,23\n 1074: bne 1090 \u003cirq_exit+0x60\u003e\n \n This can frustrate store-hit-load detection heuristics and cause\n flushes. Allowing the compiler to cache current in a reigster with this\n patch results in the same base register being used for all accesses,\n which is more likely to be detected as an alias:\n \n 1058: ld r31,2392(r13)\n ...\n 1070: lwz r9,0(r31)\n 1074: addis r9,r9,-1\n 1078: stw r9,0(r31)\n 107c: lwz r9,0(r31)\n 1080: rlwinm. r9,r9,0,11,23\n 1084: bne 10a0 \u003cirq_exit+0x60\u003e\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190612140317.24490-1-npiggin@gmail.com\n\ncommit 7ab0b7cb8951d4095d73e203759b74d41916e455\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Fri Aug 16 07:52:20 2019 +0000\n\n powerpc/32: Add warning on misaligned copy_page() or clear_page()\n \n copy_page() and clear_page() expect page aligned destination, and\n use dcbz instruction to clear entire cache lines based on the\n assumption that the destination is cache aligned.\n \n As shown during analysis of a bug in BTRFS filesystem, a misaligned\n copy_page() can create bugs that are difficult to locate (see Link).\n \n Add an explicit WARNING when copy_page() or clear_page() are called\n with misaligned destination.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://bugzilla.kernel.org/show_bug.cgi?id\u003d204371\n Link: https://lore.kernel.org/r/c6cea38f90480268d439ca44a645647e260fff09.1565941808.git.christophe.leroy@c-s.fr\n\ncommit f204338f8e29777302042cba578b0da44f69695f\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Fri Aug 16 05:41:44 2019 +0000\n\n powerpc/mm: ppc 603 doesn\u0027t need update_mmu_cache()\n \n On powerpc 603, there is no hash table so get out of\n update_mmu_cache() early.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/6133e0f115d955fac4061536dab0fa7480a1c433.1565933217.git.christophe.leroy@c-s.fr\n\ncommit f49f4e2b68b683491263e92c229ff344d44759a7\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Fri Aug 16 05:41:43 2019 +0000\n\n powerpc/mm: Simplify update_mmu_cache() on BOOK3S32\n \n On BOOK3S32, hash_preload() neither use is_exec nor trap,\n so drop those parameters and simplify update_mmu_cached().\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/35f143c6fe29f9fd25c7f3cd4448ae401029ce3c.1565933217.git.christophe.leroy@c-s.fr\n\ncommit e5a1edb9fe4cfa07e37a59475f8f7d0a8939c73e\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Fri Aug 16 05:41:42 2019 +0000\n\n powerpc/mm: move update_mmu_cache() into book3s hash utils.\n \n update_mmu_cache() is only for BOOK3S, and can be simplified for\n BOOK3S32.\n \n Move it out of mem.c into respective BOOK3S32 and BOOK3S64 files\n containing hash utils.\n \n BOOK3S64 version of hash_preload() is only used locally, declare it\n static.\n \n Remove the radix_enabled() stuff in BOOK3S32 version.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/107aaf43583a5f5d09e0d4e84c4c4390ecfcd512.1565933217.git.christophe.leroy@c-s.fr\n\ncommit 4c1616ef036ffaaea95a29d7b6abf9d3e8eb9d92\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Fri Aug 16 05:41:41 2019 +0000\n\n powerpc/mm: move FSL_BOOK3 version of update_mmu_cache()\n \n Move FSL_BOOK3E version of update_mmu_cache() at the same\n place as book3e_hugetlb_preload() as update_mmu_cache() is\n the only user of book3e_hugetlb_preload().\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/4d69fdc86df9c74adc71a60331a86f6afb8b5e9e.1565933217.git.christophe.leroy@c-s.fr\n\ncommit d9642117914c9d3f800b3bacc19d7e388b04edb4\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Fri Aug 16 05:41:40 2019 +0000\n\n powerpc/mm: define empty update_mmu_cache() as static inline\n \n Only BOOK3S and FSL_BOOK3E have a usefull update_mmu_cache().\n \n For the others, just define it static inline.\n \n In the meantime, simplify the FSL_BOOK3E related ifdef as\n book3e_hugetlb_preload() only exists when CONFIG_PPC_FSL_BOOK3E\n is selected.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/668aba4db6b9af6d8a151174e11a4289f1a6bbcd.1565933217.git.christophe.leroy@c-s.fr\n\ncommit ad628a34ec4e3558bf838195f60bbaa4c2b68f2a\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Wed Aug 14 14:36:10 2019 +0000\n\n powerpc/mm: don\u0027t display empty early ioremap area\n \n On the 8xx, the layout displayed at boot is:\n \n [ 0.000000] Memory: 121856K/131072K available (5728K kernel code, 592K rwdata, 1248K rodata, 560K init, 448K bss, 9216K reserved, 0K cma-reserved)\n [ 0.000000] Kernel virtual memory layout:\n [ 0.000000] * 0xffefc000..0xffffc000 : fixmap\n [ 0.000000] * 0xffefc000..0xffefc000 : early ioremap\n [ 0.000000] * 0xc9000000..0xffefc000 : vmalloc \u0026 ioremap\n [ 0.000000] SLUB: HWalign\u003d16, Order\u003d0-3, MinObjects\u003d0, CPUs\u003d1, Nodes\u003d1\n \n Remove display of an empty early ioremap.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/f6267226038cb25a839b567319e240576e3f8565.1565793287.git.christophe.leroy@c-s.fr\n\ncommit 9d6d712fbf7766f21c838940eebcd7b4d476c5e6\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Wed Aug 14 10:02:20 2019 +0000\n\n powerpc/32s: Fix boot failure with DEBUG_PAGEALLOC without KASAN.\n \n When KASAN is selected, the definitive hash table has to be\n set up later, but there is already an early temporary one.\n \n When KASAN is not selected, there is no early hash table,\n so the setup of the definitive hash table cannot be delayed.\n \n Fixes: 72f208c6a8f7 (\"powerpc/32s: move hash code patching out of MMU_init_hw()\")\n Cc: stable@vger.kernel.org # v5.2+\n Reported-by: Jonathan Neuschafer \u003cj.neuschaefer@gmx.net\u003e\n Tested-by: Jonathan Neuschafer \u003cj.neuschaefer@gmx.net\u003e\n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/b7860c5e1e784d6b96ba67edf47dd6cbc2e78ab6.1565776892.git.christophe.leroy@c-s.fr\n\ncommit 38a0d0cdb46d3f91534e5b9839ec2d67be14c59d\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Wed Aug 14 09:25:52 2019 +0000\n\n powerpc/futex: Fix warning: \u0027oldval\u0027 may be used uninitialized in this function\n \n We see warnings such as:\n kernel/futex.c: In function \u0027do_futex\u0027:\n kernel/futex.c:1676:17: warning: \u0027oldval\u0027 may be used uninitialized in this function [-Wmaybe-uninitialized]\n return oldval \u003d\u003d cmparg;\n ^\n kernel/futex.c:1651:6: note: \u0027oldval\u0027 was declared here\n int oldval, ret;\n ^\n \n This is because arch_futex_atomic_op_inuser() only sets *oval if ret\n is 0 and GCC doesn\u0027t see that it will only use it when ret is 0.\n \n Anyway, the non-zero ret path is an error path that won\u0027t suffer from\n setting *oval, and as *oval is a local var in futex_atomic_op_inuser()\n it will have no impact.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n [mpe: reword change log slightly]\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/86b72f0c134367b214910b27b9a6dd3321af93bb.1565774657.git.christophe.leroy@c-s.fr\n\ncommit 663c0c9496a69f80011205ba3194049bcafd681d\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Fri Aug 9 14:58:10 2019 +0000\n\n powerpc/kasan: Fix shadow area set up for modules.\n \n When loading modules, from time to time an Oops is encountered during\n the init of shadow area for globals. This is due to the last page not\n always being mapped depending on the exact distance between the start\n and the end of the shadow area and the alignment with the page\n addresses.\n \n Fix this by aligning the starting address with the page address.\n \n Fixes: 2edb16efc899 (\"powerpc/32: Add KASAN support\")\n Cc: stable@vger.kernel.org # v5.2+\n Reported-by: Erhard F. \u003cerhard_f@mailbox.org\u003e\n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/4f887e9b77d0d725cbb52035c7ece485c1c5fc14.1565361881.git.christophe.leroy@c-s.fr\n\ncommit 45ff3c55958542c3b76075d59741297b8cb31cbb\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Fri Aug 9 14:58:09 2019 +0000\n\n powerpc/kasan: Fix parallel loading of modules.\n \n Parallel loading of modules may lead to bad setup of shadow page table\n entries.\n \n First, lets align modules so that two modules never share the same\n shadow page.\n \n Second, ensure that two modules cannot allocate two page tables for\n the same PMD entry at the same time. This is done by using\n init_mm.page_table_lock in the same way as __pte_alloc_kernel()\n \n Fixes: 2edb16efc899 (\"powerpc/32: Add KASAN support\")\n Cc: stable@vger.kernel.org # v5.2+\n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/c97284f912128cbc3f2fe09d68e90e65fb3e6026.1565361876.git.christophe.leroy@c-s.fr\n\ncommit 658d029df0bce6472c94b724ca54d74bc6659c2e\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Fri Jun 28 15:55:52 2019 +0000\n\n powerpc/hw_breakpoint: move instruction stepping out of hw_breakpoint_handler()\n \n On 8xx, breakpoints stop after executing the instruction, so\n stepping/emulation is not needed. Move it into a sub-function and\n remove the #ifdefs.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Reviewed-by: Ravi Bangoria \u003cravi.bangoria@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/f8cdc3f1c66ad3c43ebc568abcc6c39ed4676284.1561737231.git.christophe.leroy@c-s.fr\n\ncommit 65e701b2d2a8593c44c8855aee2e087be7e11e75\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Wed Aug 14 12:36:13 2019 +0000\n\n powerpc/ptdump: drop non vital #ifdefs\n \n hashpagetable.c is only compiled when CONFIG_PPC_BOOK3S_64 is\n defined, so drop the test and its \u0027else\u0027 branch.\n \n Use IS_ENABLED(CONFIG_PPC_PSERIES) instead of #ifdef, this allows the\n code to be checked at any build. It is still optimised out by GCC.\n \n Use IS_ENABLED(CONFIG_PPC_64K_PAGES) instead of #ifdef.\n \n Use IS_ENABLED(CONFIG_SPARSEMEN_VMEMMAP) instead of #ifdef.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/c8998ed32e4e3954b56a8dacecfe43319a2a0483.1565786091.git.christophe.leroy@c-s.fr\n\ncommit f3a2ac05894b57e78cd6cc0adba1de642034d940\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Wed Aug 14 12:36:12 2019 +0000\n\n powerpc/ptdump: get out of note_prot_wx() when CONFIG_PPC_DEBUG_WX is not selected.\n \n When CONFIG_PPC_DEBUG_WX, note_prot_wx() is useless.\n \n Get out of it early and inconditionnally in that case,\n so that GCC can kick all the code out.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/ff6c8f631bd4ce3a10e0cc241eb569816187bc20.1565786091.git.christophe.leroy@c-s.fr\n\ncommit 82242352728125ecd86cea4631b89a5e6be0d0fd\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Wed Aug 14 12:36:11 2019 +0000\n\n powerpc/ptdump: drop dummy KERN_VIRT_START on PPC32\n \n PPC32 doesn\u0027t have KERN_VIRT_START. Make PAGE_OFFSET the\n default starting address for the dump, and drop the dummy\n definition of KERN_VIRT_START. Only use KERN_VIRT_START for\n non radix PPC64.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/334632b1df4775b0ccf3bdc8d6b201d14e3daedd.1565786091.git.christophe.leroy@c-s.fr\n\ncommit e033829d2aaad85cb7cf46986c3be0bcc72f791e\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Wed Aug 14 12:36:10 2019 +0000\n\n powerpc/ptdump: fix walk_pagetables() address mismatch\n \n walk_pagetables() always walk the entire pgdir from address 0\n but considers PAGE_OFFSET or KERN_VIRT_START as the starting\n address of the walk, resulting in a possible mismatch in the\n displayed addresses.\n \n Ex: on PPC32, when KERN_VIRT_START was locally defined as\n PAGE_OFFSET, ptdump displayed 0x80000000\n instead of 0xc0000000 for the first kernel page,\n because 0xc0000000 + 0xc0000000 \u003d 0x80000000\n \n Start the walk at st-\u003estart_address instead of starting at 0.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/5aa2ac513295f594cce8ddb1c649f61947bd063d.1565786091.git.christophe.leroy@c-s.fr\n\ncommit 7c7a532ba3fc51bf9527d191fb410786c1fdc73c\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Wed Aug 14 12:36:09 2019 +0000\n\n powerpc/ptdump: Fix addresses display on PPC32\n \n Commit 453d87f6a8ae (\"powerpc/mm: Warn if W+X pages found on boot\")\n wrongly changed KERN_VIRT_START from 0 to PAGE_OFFSET, leading to a\n shift in the displayed addresses.\n \n Lets revert that change to resync walk_pagetables()\u0027s addr val and\n pgd_t pointer for PPC32.\n \n Fixes: 453d87f6a8ae (\"powerpc/mm: Warn if W+X pages found on boot\")\n Cc: stable@vger.kernel.org # v5.2+\n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/eb4d626514e22f85814830012642329018ef6af9.1565786091.git.christophe.leroy@c-s.fr\n\ncommit b48b8bc45a8fe4bc1d1369f0e2eb98b15a4b6c56\nAuthor: Jarkko Nikula \u003cjarkko.nikula@linux.intel.com\u003e\nDate: Tue Aug 13 11:06:02 2019 +0300\n\n dmaengine: dw: Update Intel Elkhart Lake Service Engine acronym\n \n Intel Elkhart Lake Offload Service Engine (OSE) will be called as\n Intel(R) Programmable Services Engine (Intel(R) PSE) in documentation.\n \n Update the comment here accordingly.\n \n Signed-off-by: Jarkko Nikula \u003cjarkko.nikula@linux.intel.com\u003e\n Acked-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190813080602.15376-1-jarkko.nikula@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 14c2a7b934e43527e38adc40cb9466811baaf8a9\nAuthor: Simon Horman \u003chorms+renesas@verge.net.au\u003e\nDate: Mon Aug 19 15:52:44 2019 +0200\n\n dt-bindings: dmaengine: nbpfaxi: Rename bindings documentation file\n \n Rename the device tree bindings for renesas \"Type-AXI\" NBPFAXI* DMA\n controllers from nbpfaxi.txt to renesas,nbpfaxi.txt.\n \n This is part of an ongoing effort to name bindings documentation files for\n Renesas IP blocks consistently, in line with the compat strings they\n document.\n \n Signed-off-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Link: https://lore.kernel.org/r/20190819135244.18183-1-horms+renesas@verge.net.au\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit a8c16b7593bd1a4e613164a47c526ca9d1be764b\nAuthor: Mihail Atanassov \u003cMihail.Atanassov@arm.com\u003e\nDate: Mon Aug 5 09:56:25 2019 +0000\n\n drm/komeda: Add support for \u0027memory-region\u0027 DT node property\n \n The \u0027memory-region\u0027 property of the komeda display driver DT binding\n allows the use of a \u0027reserved-memory\u0027 node for buffer allocations. Add\n the requisite of_reserved_mem_device_{init,release} calls to actually\n make use of the memory if present.\n \n Changes since v1:\n - Move handling inside komeda_parse_dt\n \n Signed-off-by: Mihail Atanassov \u003cmihail.atanassov@arm.com\u003e\n Signed-off-by: Ayan Kumar Halder \u003cayan.halder@arm.com\u003e\n Reviewed-by: James Qian Wang (Arm Technology China) \u003cjames.qian.wang@arm.com\u003e\n Link:- https://patchwork.kernel.org/patch/11076413/\n\ncommit 4b67f1ddcf23d6dc4b5a3a486b758cb91f725e72\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Aug 20 10:58:03 2019 +0100\n\n iommu/arm-smmu: Make private implementation details static\n \n Many of the device-specific implementation details in \u0027arm-smmu-impl.c\u0027\n are exposed to other compilation units. Whilst we may require this in\n the future, let\u0027s make it all \u0027static\u0027 for now so that we can expose\n things on a case-by-case basic.\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 0688587a71200911721a0cb025fe805ddc0404b6\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Mon Aug 12 14:32:52 2019 +0200\n\n dt-bindings: clock: meson: add resets to the audio clock controller\n \n Add the documentation and bindings for the resets provided by the g12a\n audio clock controller\n \n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n\ncommit cfddf9f4c9f038c91c6c61d5cf3a161731b5c418\nAuthor: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nDate: Mon Aug 19 18:47:34 2019 -0500\n\n locks: fix a memory leak bug in __break_lease()\n \n In __break_lease(), the file lock \u0027new_fl\u0027 is allocated in lease_alloc().\n However, it is not deallocated in the following execution if\n smp_load_acquire() fails, leading to a memory leak bug. To fix this issue,\n free \u0027new_fl\u0027 before returning the error.\n \n Signed-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n\ncommit 6c9050a73469268c4c82129e2c840f33d4333bd5\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:23 2019 -0700\n\n irqchip: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Jason Cooper \u003cjason@lakedaemon.net\u003e\n Cc: Marc Zyngier \u003cmarc.zyngier@arm.com\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit a512584abd7ab860560ac21e6daac1aaebc1c14f\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Mon Aug 19 16:45:34 2019 +0900\n\n irqchip: Add include guard to irq-partition-percpu.h\n \n Add a header include guard just in case.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 7224cec4e76c8d8169e328923597e659b705760d\nAuthor: Lubomir Rintel \u003clkundrak@v3.sk\u003e\nDate: Fri Aug 16 20:18:49 2019 +0200\n\n irqchip/mmp: Do not call irq_set_default_host() on DT platforms\n \n Using a default domain on DT platforms is unnecessary, as the firmware\n tables describe the full topology, and nothing is implicit.\n \n Signed-off-by: Lubomir Rintel \u003clkundrak@v3.sk\u003e\n [maz: wrote an actual changelog]\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 342be1068d9b5b1fd364d270b4f731764e23de2b\nAuthor: Zenghui Yu \u003cyuzenghui@huawei.com\u003e\nDate: Sat Jul 27 06:14:22 2019 +0000\n\n irqchip/gic-v3-its: Remove the redundant set_bit for lpi_map\n \n We try to find a free LPI region in device\u0027s lpi_map and allocate them\n (set them to 1) when we want to allocate LPIs for this device. This is\n what bitmap_find_free_region() has done for us. The following set_bit\n is redundant and a bit confusing (since we only set_bit against the first\n allocated LPI idx). Remove it, and make the set_bit explicit by comment.\n \n Signed-off-by: Zenghui Yu \u003cyuzenghui@huawei.com\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 7f2481b39b4c776fb9c03081ffcfe81f4961601c\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Wed Jul 31 17:29:33 2019 +0100\n\n irqchip/gic-v3: Add quirks for HIP06/07 invalid GICD_TYPER erratum 161010803\n \n It looks like the HIP06/07 SoCs have extra bits in their GICD_TYPER\n registers, which confuse the GICv3.1 code (these systems appear to\n expose ESPIs while they actually don\u0027t).\n \n Detect these systems as early as possible and wipe the fields that\n should be RES0 in the register.\n \n Tested-by: John Garry \u003cjohn.garry@huawei.com\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 25edaed6c7feca8fded25348732c628283304364\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Wed Jul 31 17:28:49 2019 +0100\n\n irqchip/gic: Skip DT quirks when evaluating IIDR-based quirks\n \n When evaluating potential quirks matched by reads of the IIDR\n register, skip the quirk entries that use a \"compatible\"\n property attached to them, as these are DT based.\n \n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit ad5a78d3da81836c88d1f2d53310484462660997\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Thu Jul 25 15:30:51 2019 +0100\n\n irqchip/gic-v3: Warn about inconsistent implementations of extended ranges\n \n As is it usual for the GIC, it isn\u0027t disallowed to put together a system\n that is majorly inconsistent, with a distributor supporting the\n extended ranges while some of the CPUs don\u0027t.\n \n Kindly tell the user that things are sailing isn\u0027t going to be smooth.\n \n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 5f51f803826e4f4aedff415ddaf14efa707be5a7\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Thu Jul 18 13:19:25 2019 +0100\n\n irqchip/gic-v3: Add EPPI range support\n \n Expand the pre-existing PPI support to be able to deal with the\n Extended PPI range (EPPI). This includes obtaining the number of PPIs\n from each individual redistributor, and compute the minimum set\n (just in case someone builds something really clever...).\n \n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 4b049063e0bcbfd302f6bf867de9d55a965d622e\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Thu Jul 18 13:18:51 2019 +0100\n\n dt-bindings: interrupt-controller: arm,gic-v3: Describe EPPI range support\n \n Update the GICv3 binding to allow interrupts in the EPPI range.\n \n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 52085d3f2028d853f8d6ce7ead2f8a504f6077fa\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Thu Jul 18 13:05:17 2019 +0100\n\n irqchip/gic-v3: Dynamically allocate PPI partition descriptors\n \n Again, PPIs are becoming a variable set. Let\u0027s hack the PPI partition\n code to make the top-level array dynamically allocated.\n \n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 81a43273045b116901e569ca27ddf55550f92caf\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Thu Jul 18 12:53:05 2019 +0100\n\n irqchip/gic-v3: Dynamically allocate PPI NMI refcounts\n \n As we\u0027re about to have a variable number of PPIs, let\u0027s make the\n allocation of the NMI refcounts dynamic. Also apply some minor\n cleanups (moving things around).\n \n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 1a60e1e6439164c06636dce5d32660de505d23c3\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Thu Jul 18 11:15:14 2019 +0100\n\n irqchip/gic: Prepare for more than 16 PPIs\n \n GICv3.1 allows up to 80 PPIs (16 legaci PPIs and 64 Extended PPIs),\n meaning we can\u0027t just leave the old 16 hardcoded everywhere.\n \n We also need to add the infrastructure to discover the number of PPIs\n on a per redistributor basis, although we still pretend there is only\n 16 of them for now.\n \n No functional change.\n \n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 211bddd210a6746e4fdfa9b6cdfbdb15026530a7\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Tue Jul 16 15:17:31 2019 +0100\n\n irqchip/gic-v3: Add ESPI range support\n \n Add the required support for the ESPI range, which behave exactly like\n the SPIs of old, only with new funky INTIDs.\n \n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit ea4e537ae149b468963bf500cfb0a99f8df4b3e7\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 19 20:57:40 2019 +0100\n\n dma-buf: Use %zu for printing sizeof\n \n Use the %zu format specifier for a size_t returned by sizeof.\n \n Reported-by: kbuild-all@01.org\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190819195740.27608-1-chris@chris-wilson.co.uk\n\ncommit 2d122942484c20bde308bdf694459800b63211d1\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Aug 20 10:11:54 2019 +0100\n\n Revert \"init/Kconfig: Fix infinite Kconfig recursion on PPC\"\n \n This reverts commit 71c67a31f09fa8fdd1495dffd96a5f0d4cef2ede.\n \n Commit 117acf5c29dd (\"powerpc/Makefile: Always pass --synthetic to nm if\n supported\") removed the only conditional definition of $(NM), so we can\n revert our temporary bodge to avoid Kconfig recursion and go back to\n passing $(NM) through to the \u0027tools-support-relr.sh\u0027 when detecting\n support for RELR relocations.\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 866246534836c60f706076cdefcd45072ad9eab2\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Tue Jul 16 15:18:40 2019 +0100\n\n dt-bindings: interrupt-controller: arm,gic-v3: Describe ESPI range support\n \n GICv3.1 introduces support for new interrupt ranges, one of them being\n the Extended SPI range (ESPI). The DT binding is extended to deal with\n it as a new interrupt class.\n \n Reviewed-by: Lokesh Vutla \u003clokeshvutla@ti.com\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit e91b036e1c20d80419164ddfc32125052df3fb39\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Tue Jul 16 14:41:40 2019 +0100\n\n irqchip/gic-v3: Add INTID range and convertion primitives\n \n In the beginning, life was simple. The GIC driver mostly cared about\n PPIs, SPIs and LPIs, all with nicely layed out ranges.\n \n We\u0027re about to change all that, with new ranges such as EPPI and ESPI\n interleaved in the middle of the no-irq-land between the \"special IDs\"\n and the LPI range. Boo.\n \n In order to make our life less hellish, let\u0027s introduce a set of primitives\n that will allow ranges to be identified easily and offsets to be remapped.\n \n So far, there is no functionnal change.\n \n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 13d22e2e1f35f2d3cc7ddc002c23e733c2782dd4\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Tue Jul 16 14:35:17 2019 +0100\n\n irqchip/gic: Rework gic_configure_irq to take the full ICFGR base\n \n gic_configure_irq is currently passed the (re)distributor address,\n to which it applies an a fixed offset to get to the configuration\n registers. This offset is constant across all GICs, or rather it was\n until to v3.1...\n \n An easy way out is for the individual drivers to pass the base\n address of the configuration register for the considered interrupt.\n At the same time, move part of the error handling back to the\n individual drivers, as things are about to change on that front.\n \n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 117acf5c29dd89e4c86761c365b9724dba0d9763\nAuthor: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\nDate: Tue Aug 20 17:23:19 2019 +1000\n\n powerpc/Makefile: Always pass --synthetic to nm if supported\n \n Back in 2004 we added logic to arch/ppc64/Makefile to pass\n the --synthetic option to nm, if it was supported by nm.\n \n Then in 2005 when arch/ppc64 and arch/ppc were merged, the logic to\n add --synthetic was moved inside an #ifdef CONFIG_PPC64 block within\n arch/powerpc/Makefile, and has remained there since.\n \n That was fine, though crufty, until recently when a change to\n init/Kconfig added a config time check that uses $(NM). On powerpc\n that leads to an infinite loop because Kconfig uses $(NM) to calculate\n some values, then the powerpc Makefile changes $(NM), which Kconfig\n notices and restarts.\n \n The original commit that added --synthetic simply said:\n On new toolchains we need to use nm --synthetic or we miss code\n symbols.\n \n And the nm man page says that the --synthetic option causes nm to:\n Include synthetic symbols in the output. These are special symbols\n created by the linker for various purposes.\n \n So it seems safe to always pass --synthetic if nm supports it, ie. on\n 32-bit and 64-bit, it just means 32-bit kernels might have more\n symbols reported (and in practice I see no extra symbols). Making it\n unconditional avoids the #ifdef CONFIG_PPC64, which in turn avoids the\n infinite loop.\n \n Debugged-by: Peter Collingbourne \u003cpcc@google.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit a7e42142926f815c776f745d027f69a53415d99c\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Mon Aug 19 10:27:04 2019 +0200\n\n gpio: ftgpio: Move hardware initialization\n \n It is probably wise to initialize the hardware before registering\n the irq chip.\n \n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190819082704.14237-1-linus.walleij@linaro.org\n\ncommit eb1e8bd6e3900e3a3b9776cd1b09ecfc05573619\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Mon Aug 19 11:30:58 2019 +0200\n\n gpio: Use callback presence to determine need of valid_mask\n \n After we switched the two drivers that have .need_valid_mask\n set to use the callback for setting up the .valid_mask,\n we can just use the presence of the .init_valid_mask()\n callback (or the OF reserved ranges, nota bene) to determine\n whether to allocate the mask or not and we can drop the\n .need_valid_mask field altogether.\n \n Cc: Benjamin Gaignard \u003cbenjamin.gaignard@st.com\u003e\n Cc: Amelie Delaunay \u003camelie.delaunay@st.com\u003e\n Cc: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Cc: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190819093058.10863-1-linus.walleij@linaro.org\n\ncommit da9b142ab2c55ae8c46242a7624fdf62c4089e7f\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Mon Aug 19 11:11:40 2019 +0200\n\n pinctrl: stmfx: Use the callback to populate valid_mask\n \n This makes use of the existing callback to populate the\n valid mask instead of iteratively setting it up during\n probe.\n \n Cc: Benjamin Gaignard \u003cbenjamin.gaignard@st.com\u003e\n Cc: Amelie Delaunay \u003camelie.delaunay@st.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190819091140.622-1-linus.walleij@linaro.org\n\ncommit c9fc5aff217267a9ef3a76450760534488870c69\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Mon Aug 19 10:49:04 2019 +0200\n\n gpio: Pass mask and size with the init_valid_mask()\n \n It is more helpful for drivers to have the affected fields\n directly available when we use the callback to set up the\n valid mask. Change this and switch over the only user\n (MSM) to use the passed parameters. If we do this we can\n also move the mask out of publicly visible struct fields.\n \n Cc: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Cc: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190819084904.30027-1-linus.walleij@linaro.or\n Reviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit f52a0c7b5eb907a12bc08a33cc998ed02dad3507\nAuthor: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nDate: Fri Aug 16 21:38:12 2019 +1000\n\n gpio: stubs in headers should be inline\n \n Fixes: fdd61a013a24 (\"gpio: Add support for hierarchical IRQ domains\")\n Signed-off-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Link: https://lore.kernel.org/r/20190816213812.40a130db@canb.auug.org.au\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit f3260e3ea1f570cc3c13a65d3e46c86ca581266d\nAuthor: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\nDate: Thu Jul 25 15:10:02 2019 +0200\n\n gpio: mockup: don\u0027t depend twice on GPIOLIB\n \n config GPIO_MOCKUP is defined in a big if GPIOLIB ... endif block so it\n doesn\u0027t need to depend explicitly on GPIOLIB.\n \n Signed-off-by: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\n Link: https://lore.kernel.org/r/20190725131002.14597-1-uwe@kleine-koenig.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit be819aa6f11145de32dab8690ec6055348488c18\nAuthor: Guo Ren \u003cren_guo@c-sky.com\u003e\nDate: Tue Aug 20 15:31:29 2019 +0800\n\n csky: Fixup arch_get_unmapped_area() implementation\n \n Current arch_get_unmapped_area() of abiv1 doesn\u0027t use standard kernel\n api. After referring to the implementation of arch/arm, we implement\n it with vm_unmapped_area() from linux/mm.h.\n \n Signed-off-by: Guo Ren \u003cren_guo@c-sky.com\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit f25896ebfe0cf818ebd1adb5e6a05dc40b820e45\nAuthor: Krzysztof Wilczynski \u003ckw@linux.com\u003e\nDate: Tue Aug 20 08:51:21 2019 +0200\n\n x86/PCI: Remove superfluous returns from void functions\n \n Remove unnecessary empty return statements at the end of void functions\n in arch/x86/kernel/quirks.c.\n \n Signed-off-by: Krzysztof Wilczynski \u003ckw@linux.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Cc: Bjorn Helgaas \u003chelgaas@kernel.org\u003e\n Cc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: linux-pci@vger.kernel.org\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: x86-ml \u003cx86@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190820065121.16594-1-kw@linux.com\n\ncommit 01b45d3c8f873135287800c56c8c5aa041a6f69e\nAuthor: Dariusz Marcinkiewicz \u003cdarekm@google.com\u003e\nDate: Wed Aug 14 12:45:01 2019 +0200\n\n dw-hdmi-cec: use cec_notifier_cec_adap_(un)register\n \n Use the new cec_notifier_cec_adap_(un)register() functions to\n (un)register the notifier for the CEC adapter.\n \n Also adds CEC_CAP_CONNECTOR_INFO capability to the adapter.\n \n Changes since v3:\n - add CEC_CAP_CONNECTOR_INFO to cec_allocate_adapter,\n - replace CEC_CAP_LOG_ADDRS | CEC_CAP_TRANSMIT |\n CEC_CAP_RC | CEC_CAP_PASSTHROUGH with CEC_CAP_DEFAULTS.\n \n Signed-off-by: Dariusz Marcinkiewicz \u003cdarekm@google.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Tested-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190814104520.6001-4-darekm@google.com\n\ncommit 7070fe7386c1a49e0d48a498a562a3bd952070d2\nAuthor: Dariusz Marcinkiewicz \u003cdarekm@google.com\u003e\nDate: Wed Aug 14 12:45:06 2019 +0200\n\n drm: dw-hdmi: use cec_notifier_conn_(un)register\n \n Use the new cec_notifier_conn_(un)register() functions to\n (un)register the notifier for the HDMI connector, and fill in\n the cec_connector_info.\n \n Changes since v6:\n - move cec_notifier_conn_unregister to a bridge detach\n function,\n - add a mutex protecting a CEC notifier.\n Changes since v4:\n - typo fix\n Changes since v2:\n - removed unnecessary NULL check before a call to\n cec_notifier_conn_unregister,\n - use cec_notifier_phys_addr_invalidate to invalidate physical\n address.\n Changes since v1:\n Add memory barrier to make sure that the notifier\n becomes visible to the irq thread once it is fully\n constructed.\n \n Signed-off-by: Dariusz Marcinkiewicz \u003cdarekm@google.com\u003e\n Acked-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Tested-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190814104520.6001-9-darekm@google.com\n\ncommit 6311b6521bcc804e4d2fd45a5640562a7b8b5241\nAuthor: Jordan Crouse \u003cjcrouse@codeaurora.org\u003e\nDate: Mon Aug 5 14:33:46 2019 -0600\n\n drivers: qcom: Add BCM vote macro to header\n \n The macro to generate a Bus Controller Manager (BCM) TCS command is used\n by the interconnect driver but might also be interesting to other\n drivers that need to construct TCS commands for sub processors so move\n it out of the sdm845 specific file and into the header.\n \n Signed-off-by: Jordan Crouse \u003cjcrouse@codeaurora.org\u003e\n Acked-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Acked-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Signed-off-by: Georgi Djakov \u003cgeorgi.djakov@linaro.org\u003e\n\ncommit b4a81b87a4cfe2bb26a4a943b748d96a43ef20e8\nAuthor: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nDate: Tue Aug 20 00:33:54 2019 -0500\n\n ecryptfs: fix a memory leak bug in ecryptfs_init_messaging()\n \n In ecryptfs_init_messaging(), if the allocation for \u0027ecryptfs_msg_ctx_arr\u0027\n fails, the previously allocated \u0027ecryptfs_daemon_hash\u0027 is not deallocated,\n leading to a memory leak bug. To fix this issue, free\n \u0027ecryptfs_daemon_hash\u0027 before returning the error.\n \n Cc: stable@vger.kernel.org\n Fixes: 88b4a07e6610 (\"[PATCH] eCryptfs: Public key transport mechanism\")\n Signed-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\n Signed-off-by: Tyler Hicks \u003ctyhicks@canonical.com\u003e\n\ncommit fe2e082f5da5b4a0a92ae32978f81507ef37ec66\nAuthor: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nDate: Tue Aug 20 00:16:40 2019 -0500\n\n ecryptfs: fix a memory leak bug in parse_tag_1_packet()\n \n In parse_tag_1_packet(), if tag 1 packet contains a key larger than\n ECRYPTFS_MAX_ENCRYPTED_KEY_BYTES, no cleanup is executed, leading to a\n memory leak on the allocated \u0027auth_tok_list_item\u0027. To fix this issue, go to\n the label \u0027out_free\u0027 to perform the cleanup work.\n \n Cc: stable@vger.kernel.org\n Fixes: dddfa461fc89 (\"[PATCH] eCryptfs: Public key; packet management\")\n Signed-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\n Signed-off-by: Tyler Hicks \u003ctyhicks@canonical.com\u003e\n\ncommit 769a807d0b41df4201dbeb01c22eaeb3e5905532\nAuthor: Florian Westphal \u003cfw@strlen.de\u003e\nDate: Mon Aug 12 10:32:13 2019 +0200\n\n xfrm: policy: avoid warning splat when merging nodes\n \n syzbot reported a splat:\n xfrm_policy_inexact_list_reinsert+0x625/0x6e0 net/xfrm/xfrm_policy.c:877\n CPU: 1 PID: 6756 Comm: syz-executor.1 Not tainted 5.3.0-rc2+ #57\n Call Trace:\n xfrm_policy_inexact_node_reinsert net/xfrm/xfrm_policy.c:922 [inline]\n xfrm_policy_inexact_node_merge net/xfrm/xfrm_policy.c:958 [inline]\n xfrm_policy_inexact_insert_node+0x537/0xb50 net/xfrm/xfrm_policy.c:1023\n xfrm_policy_inexact_alloc_chain+0x62b/0xbd0 net/xfrm/xfrm_policy.c:1139\n xfrm_policy_inexact_insert+0xe8/0x1540 net/xfrm/xfrm_policy.c:1182\n xfrm_policy_insert+0xdf/0xce0 net/xfrm/xfrm_policy.c:1574\n xfrm_add_policy+0x4cf/0x9b0 net/xfrm/xfrm_user.c:1670\n xfrm_user_rcv_msg+0x46b/0x720 net/xfrm/xfrm_user.c:2676\n netlink_rcv_skb+0x1f0/0x460 net/netlink/af_netlink.c:2477\n xfrm_netlink_rcv+0x74/0x90 net/xfrm/xfrm_user.c:2684\n netlink_unicast_kernel net/netlink/af_netlink.c:1302 [inline]\n netlink_unicast+0x809/0x9a0 net/netlink/af_netlink.c:1328\n netlink_sendmsg+0xa70/0xd30 net/netlink/af_netlink.c:1917\n sock_sendmsg_nosec net/socket.c:637 [inline]\n sock_sendmsg net/socket.c:657 [inline]\n \n There is no reproducer, however, the warning can be reproduced\n by adding rules with ever smaller prefixes.\n \n The sanity check (\"does the policy match the node\") uses the prefix value\n of the node before its updated to the smaller value.\n \n To fix this, update the prefix earlier. The bug has no impact on tree\n correctness, this is only to prevent a false warning.\n \n Reported-by: syzbot+8cc27ace5f6972910b31@syzkaller.appspotmail.com\n Signed-off-by: Florian Westphal \u003cfw@strlen.de\u003e\n Signed-off-by: Steffen Klassert \u003csteffen.klassert@secunet.com\u003e\n\ncommit d4507d421355a268f0eeb1e2b2310c6a2837b319\nAuthor: Matthias Kaehlcke \u003cmka@chromium.org\u003e\nDate: Wed Aug 7 11:44:44 2019 -0700\n\n arm64: dts: sdm845: Add dynamic CPU power coefficients\n \n Add dynamic power coefficients for the Silver and Gold CPU cores of\n the Qualcomm SDM845.\n \n Signed-off-by: Matthias Kaehlcke \u003cmka@chromium.org\u003e\n Reviewed-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n Reviewed-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit b602614a81078bf29c82b2671bb96a63488f68d6\nAuthor: Matthew Garrett \u003cmatthewgarrett@google.com\u003e\nDate: Mon Aug 19 17:18:05 2019 -0700\n\n lockdown: Print current-\u003ecomm in restriction messages\n \n Print the content of current-\u003ecomm in messages generated by lockdown to\n indicate a restriction that was hit. This makes it a bit easier to find\n out what caused the message.\n \n The message now patterned something like:\n \n Lockdown: \u003ccomm\u003e: \u003cwhat\u003e is restricted; see man kernel_lockdown.7\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 1957a85b0032a81e6482ca4aab883643b8dae06e\nAuthor: Matthew Garrett \u003cmatthewgarrett@google.com\u003e\nDate: Mon Aug 19 17:18:04 2019 -0700\n\n efi: Restrict efivar_ssdt_load when the kernel is locked down\n \n efivar_ssdt_load allows the kernel to import arbitrary ACPI code from an\n EFI variable, which gives arbitrary code execution in ring 0. Prevent\n that when the kernel is locked down.\n \n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Acked-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Cc: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Cc: linux-efi@vger.kernel.org\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit ccbd54ff54e8b1880456b81c4aea352ebe208843\nAuthor: Matthew Garrett \u003cmatthewgarrett@google.com\u003e\nDate: Mon Aug 19 17:18:03 2019 -0700\n\n tracefs: Restrict tracefs when the kernel is locked down\n \n Tracefs may release more information about the kernel than desirable, so\n restrict it when the kernel is locked down in confidentiality mode by\n preventing open().\n \n (Fixed by Ben Hutchings to avoid a null dereference in\n default_file_open())\n \n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Reviewed-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n Cc: Ben Hutchings \u003cben@decadent.org.uk\u003e\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 5496197f9b084f086cb410dd566648b0896fcc74\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Aug 19 17:18:02 2019 -0700\n\n debugfs: Restrict debugfs when the kernel is locked down\n \n Disallow opening of debugfs files that might be used to muck around when\n the kernel is locked down as various drivers give raw access to hardware\n through debugfs. Given the effort of auditing all 2000 or so files and\n manually fixing each one as necessary, I\u0027ve chosen to apply a heuristic\n instead. The following changes are made:\n \n (1) chmod and chown are disallowed on debugfs objects (though the root dir\n can be modified by mount and remount, but I\u0027m not worried about that).\n \n (2) When the kernel is locked down, only files with the following criteria\n are permitted to be opened:\n \n - The file must have mode 00444\n - The file must not have ioctl methods\n - The file must not have mmap\n \n (3) When the kernel is locked down, files may only be opened for reading.\n \n Normal device interaction should be done through configfs, sysfs or a\n miscdev, not debugfs.\n \n Note that this makes it unnecessary to specifically lock down show_dsts(),\n show_devs() and show_call() in the asus-wmi driver.\n \n I would actually prefer to lock down all files by default and have the\n the files unlocked by the creator. This is tricky to manage correctly,\n though, as there are 19 creation functions and ~1600 call sites (some of\n them in loops scanning tables).\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n cc: Andy Shevchenko \u003candy.shevchenko@gmail.com\u003e\n cc: acpi4asus-user@lists.sourceforge.net\n cc: platform-driver-x86@vger.kernel.org\n cc: Matthew Garrett \u003cmjg59@srcf.ucam.org\u003e\n cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Greg KH \u003cgreg@kroah.com\u003e\n Cc: Rafael J. Wysocki \u003crafael@kernel.org\u003e\n Signed-off-by: Matthew Garrett \u003cmatthewgarrett@google.com\u003e\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 29d3c1c8dfe752c01b7115ecd5a3142b232a38e1\nAuthor: Matthew Garrett \u003cmatthewgarrett@google.com\u003e\nDate: Mon Aug 19 17:18:01 2019 -0700\n\n kexec: Allow kexec_file() with appropriate IMA policy when locked down\n \n Systems in lockdown mode should block the kexec of untrusted kernels.\n For x86 and ARM we can ensure that a kernel is trustworthy by validating\n a PE signature, but this isn\u0027t possible on other architectures. On those\n platforms we can use IMA digital signatures instead. Add a function to\n determine whether IMA has or will verify signatures for a given event type,\n and if so permit kexec_file() even if the kernel is otherwise locked down.\n This is restricted to cases where CONFIG_INTEGRITY_TRUSTED_KEYRING is set\n in order to prevent an attacker from loading additional keys at runtime.\n \n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Acked-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n Cc: Dmitry Kasatkin \u003cdmitry.kasatkin@gmail.com\u003e\n Cc: linux-integrity@vger.kernel.org\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit b0c8fdc7fdb77586c3d1937050925b960743306e\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Aug 19 17:18:00 2019 -0700\n\n lockdown: Lock down perf when in confidentiality mode\n \n Disallow the use of certain perf facilities that might allow userspace to\n access kernel data.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: Arnaldo Carvalho de Melo \u003cacme@kernel.org\u003e\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 9d1f8be5cf42b497a3bddf1d523f2bb142e9318c\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Aug 19 17:17:59 2019 -0700\n\n bpf: Restrict bpf when kernel lockdown is in confidentiality mode\n \n bpf_read() and bpf_read_str() could potentially be abused to (eg) allow\n private keys in kernel memory to be leaked. Disable them if the kernel\n has been locked down in confidentiality mode.\n \n Suggested-by: Alexei Starovoitov \u003calexei.starovoitov@gmail.com\u003e\n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n cc: netdev@vger.kernel.org\n cc: Chun-Yi Lee \u003cjlee@suse.com\u003e\n cc: Alexei Starovoitov \u003calexei.starovoitov@gmail.com\u003e\n Cc: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit a94549dd87f5ea4ca50fee493df08a2dc6256b53\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Aug 19 17:17:58 2019 -0700\n\n lockdown: Lock down tracing and perf kprobes when in confidentiality mode\n \n Disallow the creation of perf and ftrace kprobes when the kernel is\n locked down in confidentiality mode by preventing their registration.\n This prevents kprobes from being used to access kernel memory to steal\n crypto data, but continues to allow the use of kprobes from signed\n modules.\n \n Reported-by: Alexei Starovoitov \u003calexei.starovoitov@gmail.com\u003e\n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Acked-by: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Cc: Naveen N. Rao \u003cnaveen.n.rao@linux.ibm.com\u003e\n Cc: Anil S Keshavamurthy \u003canil.s.keshavamurthy@intel.com\u003e\n Cc: davem@davemloft.net\n Cc: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 02e935bf5b34edcc4cb0dc532dd0e1a1bfb33b51\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Aug 19 17:17:57 2019 -0700\n\n lockdown: Lock down /proc/kcore\n \n Disallow access to /proc/kcore when the kernel is locked down to prevent\n access to cryptographic data. This is limited to lockdown\n confidentiality mode and is still permitted in integrity mode.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 906357f77a077508d160e729f917c5f0a4304f25\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Aug 19 17:17:56 2019 -0700\n\n x86/mmiotrace: Lock down the testmmiotrace module\n \n The testmmiotrace module shouldn\u0027t be permitted when the kernel is locked\n down as it can be used to arbitrarily read and write MMIO space. This is\n a runtime check rather than buildtime in order to allow configurations\n where the same kernel may be run in both locked down or permissive modes\n depending on local policy.\n \n Suggested-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Signed-off-by: David Howells \u003cdhowells@redhat.com\n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Acked-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n cc: Steven Rostedt \u003crostedt@goodmis.org\u003e\n cc: Ingo Molnar \u003cmingo@kernel.org\u003e\n cc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\n cc: x86@kernel.org\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 20657f66ef52e5005369e4ef539d4cbf01eab10d\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Aug 19 17:17:55 2019 -0700\n\n lockdown: Lock down module params that specify hardware parameters (eg. ioport)\n \n Provided an annotation for module parameters that specify hardware\n parameters (such as io ports, iomem addresses, irqs, dma channels, fixed\n dma buffers and other types).\n \n Suggested-by: Alan Cox \u003cgnomes@lxorguk.ukuu.org.uk\u003e\n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Cc: Jessica Yu \u003cjeyu@kernel.org\u003e\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 794edf30ee6cd088d5f4079b1d4a4cfe5371203e\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Aug 19 17:17:54 2019 -0700\n\n lockdown: Lock down TIOCSSERIAL\n \n Lock down TIOCSSERIAL as that can be used to change the ioport and irq\n settings on a serial port. This only appears to be an issue for the serial\n drivers that use the core serial code. All other drivers seem to either\n ignore attempts to change port/irq or give an error.\n \n Reported-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-serial@vger.kernel.org\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 3f19cad3fa0d0fff18ee126f03a80420ae7bcbc9\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Aug 19 17:17:53 2019 -0700\n\n lockdown: Prohibit PCMCIA CIS storage when the kernel is locked down\n \n Prohibit replacement of the PCMCIA Card Information Structure when the\n kernel is locked down.\n \n Suggested-by: Dominik Brodowski \u003clinux@dominikbrodowski.net\u003e\n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 6ea0e815fc5e18597724169caa6e4d46dd8e693d\nAuthor: Linn Crosetto \u003clcrosetto@gmail.com\u003e\nDate: Mon Aug 19 17:17:52 2019 -0700\n\n acpi: Disable ACPI table override if the kernel is locked down\n \n \u003eFrom the kernel documentation (initrd_table_override.txt):\n \n If the ACPI_INITRD_TABLE_OVERRIDE compile option is true, it is possible\n to override nearly any ACPI table provided by the BIOS with an\n instrumented, modified one.\n \n When lockdown is enabled, the kernel should disallow any unauthenticated\n changes to kernel space. ACPI tables contain code invoked by the kernel,\n so do not allow ACPI tables to be overridden if the kernel is locked down.\n \n Signed-off-by: Linn Crosetto \u003clcrosetto@gmail.com\u003e\n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n cc: linux-acpi@vger.kernel.org\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 41fa1ee9c6d687afb05760dd349f361855f1d7f5\nAuthor: Josh Boyer \u003cjwboyer@redhat.com\u003e\nDate: Mon Aug 19 17:17:51 2019 -0700\n\n acpi: Ignore acpi_rsdp kernel param when the kernel has been locked down\n \n This option allows userspace to pass the RSDP address to the kernel, which\n makes it possible for a user to modify the workings of hardware. Reject\n the option when the kernel is locked down. This requires some reworking\n of the existing RSDP command line logic, since the early boot code also\n makes use of a command-line passed RSDP when locating the SRAT table\n before the lockdown code has been initialised. This is achieved by\n separating the command line RSDP path in the early boot code from the\n generic RSDP path, and then copying the command line RSDP into boot\n params in the kernel proper if lockdown is not enabled. If lockdown is\n enabled and an RSDP is provided on the command line, this will only be\n used when parsing SRAT (which shouldn\u0027t permit kernel code execution)\n and will be ignored in the rest of the kernel.\n \n (Modified by Matthew Garrett in order to handle the early boot RSDP\n environment)\n \n Signed-off-by: Josh Boyer \u003cjwboyer@redhat.com\u003e\n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n cc: Dave Young \u003cdyoung@redhat.com\u003e\n cc: linux-acpi@vger.kernel.org\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit f474e1486b78ac15322f8a1cda48a32a1deff9d3\nAuthor: Matthew Garrett \u003cmjg59@srcf.ucam.org\u003e\nDate: Mon Aug 19 17:17:50 2019 -0700\n\n ACPI: Limit access to custom_method when the kernel is locked down\n \n custom_method effectively allows arbitrary access to system memory, making\n it possible for an attacker to circumvent restrictions on module loading.\n Disable it if the kernel is locked down.\n \n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n cc: linux-acpi@vger.kernel.org\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 95f5e95f41dff31b2a4566c5a8975c08a49ae4e3\nAuthor: Matthew Garrett \u003cmjg59@srcf.ucam.org\u003e\nDate: Mon Aug 19 17:17:49 2019 -0700\n\n x86/msr: Restrict MSR access when the kernel is locked down\n \n Writing to MSRs should not be allowed if the kernel is locked down, since\n it could lead to execution of arbitrary code in kernel mode. Based on a\n patch by Kees Cook.\n \n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Acked-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Reviewed-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n cc: x86@kernel.org\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 96c4f67293e4cd8b3394adce5a8041a2784e68a3\nAuthor: Matthew Garrett \u003cmjg59@srcf.ucam.org\u003e\nDate: Mon Aug 19 17:17:48 2019 -0700\n\n x86: Lock down IO port access when the kernel is locked down\n \n IO port access would permit users to gain access to PCI configuration\n registers, which in turn (on a lot of hardware) give access to MMIO\n register space. This would potentially permit root to trigger arbitrary\n DMA, so lock it down by default.\n \n This also implicitly locks down the KDADDIO, KDDELIO, KDENABIO and\n KDDISABIO console ioctls.\n \n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n cc: x86@kernel.org\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit eb627e17727ebeede70697ae1798688b0d328b54\nAuthor: Matthew Garrett \u003cmjg59@srcf.ucam.org\u003e\nDate: Mon Aug 19 17:17:47 2019 -0700\n\n PCI: Lock down BAR access when the kernel is locked down\n \n Any hardware that can potentially generate DMA has to be locked down in\n order to avoid it being possible for an attacker to modify kernel code,\n allowing them to circumvent disabled module loading or module signing.\n Default to paranoid - in future we can potentially relax this for\n sufficiently IOMMU-isolated devices.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Acked-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n cc: linux-pci@vger.kernel.org\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 38bd94b8a1bd46e6d3d9718c7ff582e4c8ccb440\nAuthor: Josh Boyer \u003cjwboyer@fedoraproject.org\u003e\nDate: Mon Aug 19 17:17:46 2019 -0700\n\n hibernate: Disable when the kernel is locked down\n \n There is currently no way to verify the resume image when returning\n from hibernate. This might compromise the signed modules trust model,\n so until we can work with signed hibernate images we disable it when the\n kernel is locked down.\n \n Signed-off-by: Josh Boyer \u003cjwboyer@fedoraproject.org\u003e\n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Cc: rjw@rjwysocki.net\n Cc: pavel@ucw.cz\n cc: linux-pm@vger.kernel.org\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 155bdd30af17e90941589b5db4dab9a29b28c112\nAuthor: Jiri Bohac \u003cjbohac@suse.cz\u003e\nDate: Mon Aug 19 17:17:45 2019 -0700\n\n kexec_file: Restrict at runtime if the kernel is locked down\n \n When KEXEC_SIG is not enabled, kernel should not load images through\n kexec_file systemcall if the kernel is locked down.\n \n [Modified by David Howells to fit with modifications to the previous patch\n and to return -EPERM if the kernel is locked down for consistency with\n other lockdowns. Modified by Matthew Garrett to remove the IMA\n integration, which will be replaced by integrating with the IMA\n architecture policy patches.]\n \n Signed-off-by: Jiri Bohac \u003cjbohac@suse.cz\u003e\n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n cc: kexec@lists.infradead.org\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 99d5cadfde2b1acb7650021df5abaa5ec447dd10\nAuthor: Jiri Bohac \u003cjbohac@suse.cz\u003e\nDate: Mon Aug 19 17:17:44 2019 -0700\n\n kexec_file: split KEXEC_VERIFY_SIG into KEXEC_SIG and KEXEC_SIG_FORCE\n \n This is a preparatory patch for kexec_file_load() lockdown. A locked down\n kernel needs to prevent unsigned kernel images from being loaded with\n kexec_file_load(). Currently, the only way to force the signature\n verification is compiling with KEXEC_VERIFY_SIG. This prevents loading\n usigned images even when the kernel is not locked down at runtime.\n \n This patch splits KEXEC_VERIFY_SIG into KEXEC_SIG and KEXEC_SIG_FORCE.\n Analogous to the MODULE_SIG and MODULE_SIG_FORCE for modules, KEXEC_SIG\n turns on the signature verification but allows unsigned images to be\n loaded. KEXEC_SIG_FORCE disallows images without a valid signature.\n \n Signed-off-by: Jiri Bohac \u003cjbohac@suse.cz\u003e\n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n cc: kexec@lists.infradead.org\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit fef5dad9876034253d59acbf8c0c314f4d94cf87\nAuthor: Dave Young \u003cdyoung@redhat.com\u003e\nDate: Mon Aug 19 17:17:43 2019 -0700\n\n lockdown: Copy secure_boot flag in boot params across kexec reboot\n \n Kexec reboot in case secure boot being enabled does not keep the secure\n boot mode in new kernel, so later one can load unsigned kernel via legacy\n kexec_load. In this state, the system is missing the protections provided\n by secure boot.\n \n Adding a patch to fix this by retain the secure_boot flag in original\n kernel.\n \n secure_boot flag in boot_params is set in EFI stub, but kexec bypasses the\n stub. Fixing this issue by copying secure_boot flag across kexec reboot.\n \n Signed-off-by: Dave Young \u003cdyoung@redhat.com\u003e\n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n cc: kexec@lists.infradead.org\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 7d31f4602f8d366072471ca138e4ea7b8edf9be0\nAuthor: Matthew Garrett \u003cmjg59@srcf.ucam.org\u003e\nDate: Mon Aug 19 17:17:42 2019 -0700\n\n kexec_load: Disable at runtime if the kernel is locked down\n \n The kexec_load() syscall permits the loading and execution of arbitrary\n code in ring 0, which is something that lock-down is meant to prevent. It\n makes sense to disable kexec_load() in this situation.\n \n This does not affect kexec_file_load() syscall which can check for a\n signature on the image to be booted.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Acked-by: Dave Young \u003cdyoung@redhat.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n cc: kexec@lists.infradead.org\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 9b9d8dda1ed72e9bd560ab0ca93d322a9440510e\nAuthor: Matthew Garrett \u003cmjg59@srcf.ucam.org\u003e\nDate: Mon Aug 19 17:17:41 2019 -0700\n\n lockdown: Restrict /dev/{mem,kmem,port} when the kernel is locked down\n \n Allowing users to read and write to core kernel memory makes it possible\n for the kernel to be subverted, avoiding module loading restrictions, and\n also to steal cryptographic information.\n \n Disallow /dev/mem and /dev/kmem from being opened this when the kernel has\n been locked down to prevent this.\n \n Also disallow /dev/port from being opened to prevent raw ioport access and\n thus DMA from being used to accomplish the same thing.\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Cc: x86@kernel.org\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 49fcf732bdae0550721ef73af7c45109ce26b2a9\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Mon Aug 19 17:17:40 2019 -0700\n\n lockdown: Enforce module signatures if the kernel is locked down\n \n If the kernel is locked down, require that all modules have valid\n signatures that we can verify.\n \n I have adjusted the errors generated:\n \n (1) If there\u0027s no signature (ENODATA) or we can\u0027t check it (ENOPKG,\n ENOKEY), then:\n \n (a) If signatures are enforced then EKEYREJECTED is returned.\n \n (b) If there\u0027s no signature or we can\u0027t check it, but the kernel is\n locked down then EPERM is returned (this is then consistent with\n other lockdown cases).\n \n (2) If the signature is unparseable (EBADMSG, EINVAL), the signature fails\n the check (EKEYREJECTED) or a system error occurs (eg. ENOMEM), we\n return the error we got.\n \n Note that the X.509 code doesn\u0027t check for key expiry as the RTC might not\n be valid or might not have been transferred to the kernel\u0027s clock yet.\n \n [Modified by Matthew Garrett to remove the IMA integration. This will\n be replaced with integration with the IMA architecture policy\n patchset.]\n \n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: Matthew Garrett \u003cmatthewgarrett@google.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Cc: Jessica Yu \u003cjeyu@kernel.org\u003e\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 000d388ed3bbed745f366ce71b2bb7c2ee70f449\nAuthor: Matthew Garrett \u003cmatthewgarrett@google.com\u003e\nDate: Mon Aug 19 17:17:39 2019 -0700\n\n security: Add a static lockdown policy LSM\n \n While existing LSMs can be extended to handle lockdown policy,\n distributions generally want to be able to apply a straightforward\n static policy. This patch adds a simple LSM that can be configured to\n reject either integrity or all lockdown queries, and can be configured\n at runtime (through securityfs), boot time (via a kernel parameter) or\n build time (via a kconfig option). Based on initial code by David\n Howells.\n \n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Cc: David Howells \u003cdhowells@redhat.com\u003e\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 9e47d31d6a57b5babaca36d42b0d11b6db6019b7\nAuthor: Matthew Garrett \u003cmatthewgarrett@google.com\u003e\nDate: Mon Aug 19 17:17:38 2019 -0700\n\n security: Add a \"locked down\" LSM hook\n \n Add a mechanism to allow LSMs to make a policy decision around whether\n kernel functionality that would allow tampering with or examining the\n runtime state of the kernel should be permitted.\n \n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Acked-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Acked-by: Casey Schaufler \u003ccasey@schaufler-ca.com\u003e\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit e6b1db98cf4d54d9ea59cfcc195f70dc946fdd38\nAuthor: Matthew Garrett \u003cmatthewgarrett@google.com\u003e\nDate: Mon Aug 19 17:17:37 2019 -0700\n\n security: Support early LSMs\n \n The lockdown module is intended to allow for kernels to be locked down\n early in boot - sufficiently early that we don\u0027t have the ability to\n kmalloc() yet. Add support for early initialisation of some LSMs, and\n then add them to the list of names when we do full initialisation later.\n Early LSMs are initialised in link order and cannot be overridden via\n boot parameters, and cannot make use of kmalloc() (since the allocator\n isn\u0027t initialised yet).\n \n (Fixed by Stephen Rothwell to include a stub to fix builds when\n !CONFIG_SECURITY)\n \n Signed-off-by: Matthew Garrett \u003cmjg59@google.com\u003e\n Acked-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Acked-by: Casey Schaufler \u003ccasey@schaufler-ca.com\u003e\n Cc: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: James Morris \u003cjmorris@namei.org\u003e\n\ncommit 10541f037bd925c6752250e0b37b1ea43de27068\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:57:12 2019 -0700\n\n scsi: lpfc: Update lpfc version to 12.4.0.0\n \n Update lpfc version to 12.4.0.0\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit c00f62e6c5468ed0673c583f1ff284274e817410\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:57:11 2019 -0700\n\n scsi: lpfc: Merge per-protocol WQ/CQ pairs into single per-cpu pair\n \n Currently, each hardware queue, typically allocated per-cpu, consists of a\n WQ/CQ pair per protocol. Meaning if both SCSI and NVMe are supported 2\n WQ/CQ pairs will exist for the hardware queue. Separate queues are\n unnecessary. The current implementation wastes memory backing the 2nd set\n of queues, and the use of double the SLI-4 WQ/CQ\u0027s means less hardware\n queues can be supported which means there may not always be enough to have\n a pair per cpu. If there is only 1 pair per cpu, more cpu\u0027s may get their\n own WQ/CQ.\n \n Rework the implementation to use a single WQ/CQ pair by both protocols.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 0d8af096430ed810aacade5f564e5b98347137d4\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:57:10 2019 -0700\n\n scsi: lpfc: Add NVMe sequence level error recovery support\n \n FC-NVMe-2 added support for sequence level error recovery in the FC-NVME\n protocol. This allows for the detection of errors and lost frames and\n immediate retransmission of data to avoid exchange termination, which\n escalates into NVMeoFC connection and association failures. A significant\n RAS improvement.\n \n The driver is modified to indicate support for SLER in the NVMe PRLI is\n issues and to check for support in the PRLI response. When both sides\n support it, the driver will set a bit in the WQE to enable the recovery\n behavior on the exchange. The adapter will take care of all detection and\n retransmission.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit d79c9e9d4b3d9330ee38f392a7c98e0fc494f7f8\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:57:09 2019 -0700\n\n scsi: lpfc: Support dynamic unbounded SGL lists on G7 hardware.\n \n Typical SLI-4 hardware supports up to 2 4KB pages to be registered per XRI\n to contain the exchanges Scatter/Gather List. This caps the number of SGL\n elements that can be in the SGL. There are not extensions to extend the\n list out of the 2 pages.\n \n The G7 hardware adds a SGE type that allows the SGL to be vectored to a\n different scatter/gather list segment. And that segment can contain a SGE\n to go to another segment and so on. The initial segment must still be\n pre-registered for the XRI, but it can be a much smaller amount (256Bytes)\n as it can now be dynamically grown. This much smaller allocation can\n handle the SG list for most normal I/O, and the dynamic aspect allows it to\n support many MB\u0027s if needed.\n \n The implementation creates a pool which contains \"segments\" and which is\n initially sized to hold the initial small segment per xri. If an I/O\n requires additional segments, they are allocated from the pool. If the\n pool has no more segments, the pool is grown based on what is now\n needed. After the I/O completes, the additional segments are returned to\n the pool for use by other I/Os. Once allocated, the additional segments are\n not released under the assumption of \"if needed once, it will be needed\n again\". Pools are kept on a per-hardware queue basis, which is typically\n 1:1 per cpu, but may be shared by multiple cpus.\n \n The switch to the smaller initial allocation significantly reduces the\n memory footprint of the driver (which only grows if large ios are\n issued). Based on the several K of XRIs for the adapter, the 8KB-\u003e256B\n reduction can conserve 32MBs or more.\n \n It has been observed with per-cpu resource pools that allocating a resource\n on CPU A, may be put back on CPU B. While the get routines are distributed\n evenly, only a limited subset of CPUs may be handling the put routines.\n This can put a strain on the lpfc_put_cmd_rsp_buf_per_cpu routine because\n all the resources are being put on a limited subset of CPUs.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit e62245d923caebc02582b12ce861c3d780b4106f\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:57:08 2019 -0700\n\n scsi: lpfc: Add MDS driver loopback diagnostics support\n \n Added code to support driver loopback with MDS Diagnostics. This style of\n diagnostics passes frames from the fabric to the driver who then echo them\n back out the link. SEND_FRAME WQEs are used to transmit the frames. Added\n the SOF and EOF field location definitions for use by SEND_FRAME.\n \n Also ensure that enable_mds_diags is a RW parameter.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit ec76242f3be73e5cbb950a7ac99d4fba7e226e2e\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:57:07 2019 -0700\n\n scsi: lpfc: Add first and second level hardware revisions to sysfs reporting\n \n To aid better hardware detection when there are issues, report the first\n and second level hardware revisions from the READ_REV command. Add the\n elements to the existing hardware id string.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 32350664497279f4ddd96164caafc8a1b573ca2a\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:57:06 2019 -0700\n\n scsi: lpfc: Migrate to %px and %pf in kernel print calls\n \n In order to see real addresses, convert %p with %px for kernel addresses\n and replace %p with %pf for functions.\n \n While converting, standardize on \"x%px\" throughout (not %px or 0x%px).\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 1df09449902d1d4023a46ad02eab16483c54c49b\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:57:05 2019 -0700\n\n scsi: lpfc: Add simple unlikely optimizations to reduce NVME latency\n \n While performing code review, several relatively simple optimizations can\n be done in the fast path.\n \n Add these optimizations (unlikely designators).\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit d9f492a1a1ab2f3847e2c65b87a738f3831e6751\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:57:04 2019 -0700\n\n scsi: lpfc: Fix coverity warnings\n \n Running on Coverity produced the following errors:\n \n - coding style (indentation)\n \n - memset size mismatch errors\n note: comment cases where it is purposely a mismatch\n \n Fix the errors.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit db197bc4699b3f0347367d25345fe681fe73f1f6\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:57:03 2019 -0700\n\n scsi: lpfc: Fix nvme first burst module parameter description\n \n modinfo for lpfc_nvme_enable_fb is incorrect. FirstBurst on lpfc target is\n not fully supported.\n \n Update the attribute description\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 4945c0f95bbfb35e230148f226c46868f4fdd55f\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:57:02 2019 -0700\n\n scsi: lpfc: Fix BlockGuard enablement on FCoE adapters\n \n The driver is allowing the user to change lpfc_enable_bg while loading the\n driver against a FCoE adapter. This is not supported.\n \n No check is made for the adapter type when applying the blockguard\n enablement value.\n \n Fix by verifying the adapter type before setting the enablement flag.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 845d0327bf0885e7d2fb6cb18e800491be793250\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:57:01 2019 -0700\n\n scsi: lpfc: Fix reported physical link speed on a disabled trunked link\n \n GetTrunkInfo is displaying an incorrect link speed when the link is a trunk\n and the link has gone down. The driver is not clearing the logical speed\n as part of the link down transition.\n \n Fix by setting the logical speed to UNKNOWN SPEED when the link goes down.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit f98b2fd796d0f3dfa472e1ff977d38180301c982\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:57:00 2019 -0700\n\n scsi: lpfc: Fix Max Frame Size value shown in fdmishow output\n \n Max Frame Size value is shown as 34816 in fdmishow from Switch.\n \n The driver uses bbRcvSize in common service param which is obtained from\n the READ_SPARM mailbox command. The bbRcvSize field which is displayed is a\n three nibble field but the driver is printing a full four nibbles.\n \n Fix by masking off the upper nibble.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 6db51abb8dc3919b85c5d2afd35b3871399e8a81\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:59 2019 -0700\n\n scsi: lpfc: Fix upcall to bsg done in non-success cases\n \n The scsi transport fc bsg interface does not expect the bsg_job_done()\n callback to be done if the bsg request call returns failure. Several of the\n HST_VENDOR cases in the driver unconditionally call bsg_job_done()\n regardless of the returning value.\n \n Fix the code to only call bsg_job_done() if the call to lpfc_bsg_request()\n will return success.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 07b1b91412c801030752da3d32a543c5522a79d6\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:58 2019 -0700\n\n scsi: lpfc: Fix sli4 adapter initialization with MSI\n \n When forcing the use of MSI (vs MSI-X) the driver is crashing in\n pci_irq_get_affinity.\n \n The driver was not using the new pci_alloc_irq_vectors interface in the MSI\n path.\n \n Fix by using pci_alloc_irq_vectors() with PCI_RQ_MSI in the MSI path.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 6a224b47fd019d299037e14381ddff438f2d34c8\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:57 2019 -0700\n\n scsi: lpfc: Fix nvme sg_seg_cnt display if HBA does not support NVME\n \n The driver is currently reporting a non-zero nvme sg_seg_cnt value of 256\n when nvme is disabled. It should be zero.\n \n Fix by ensuring the value is cleared.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 005d8eb92847e6823d9fea6041f7c6f26b6cf2da\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:56 2019 -0700\n\n scsi: lpfc: Fix nvme target mode ABTSing a received ABTS\n \n If an unsolicited ABTS was received, the driver looks up the exchange it\n references. It it does various searches looking for the exchange\n context. When one is eventually matched and it is associated with an XRI\n context, the driver sends an ABORT WQE to terminate the exchange. Current\n code looks at whether the transport had taken action on the XRI yet or not\n (no action if set to LPFC_NVMET_STE_RCV; action if non-LPFC_NVMET_STE_RCV).\n Based on action or not one of two (sol vs unsol) issue abort routines are\n called. The unsol version cheats and transmits a sequence containing an\n ABTS with no interaction with the adapter. The sol version issues an Abort\n WQE and lets the adapter manage whether the ABTS is sent to not.\n \n The issue is the unsol version is sending ABTS unconditionally for the\n exchange that received the ABTS. It\u0027s unnecessary.\n \n Remove the conditional and just call the adapter command-based routine to\n let the adapter manage the ABTS.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 84f2ddf8cf317afcfd92e06ed5a25aea48dc200e\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:55 2019 -0700\n\n scsi: lpfc: Fix hang when downloading fw on port enabled for nvme\n \n As part of firmware download, the adapter is reset. On the adapter the\n reset causes the function to stop and all outstanding io is terminated\n (without responses). The reset path then starts teardown of the adapter,\n starting with deregistration of the remote ports with the nvme-fc\n transport. The local port is then deregistered and the driver waits for\n local port deregistration. This never finishes.\n \n The remote port deregistrations terminated the nvme controllers, causing\n them to send aborts for all the outstanding io. The aborts were serviced in\n the driver, but stalled due to its state. The nvme layer then stops to\n reclaim it\u0027s outstanding io before continuing. The io must be returned\n before the reset on the controller is deemed complete and the controller\n delete performed. The remote port deregistration won\u0027t complete until all\n the controllers are terminated. And the local port deregistration won\u0027t\n complete until all controllers and remote ports are terminated. Thus things\n hang.\n \n The issue is the reset which stopped the adapter also stopped all the\n responses that would drive i/o completions, and the aborts were also\n stopped that stopped i/o completions. The driver, when resetting the\n adapter like this, needs to be generating the completions as part of the\n adapter reset so that I/O complete (in error), and any aborts are not\n queued.\n \n Fix by adding flush routines whenever the adapter port has been reset or\n discovered in error. The flush routines will generate the completions for\n the scsi and nvme outstanding io. The abort ios, if waiting, will be caught\n and flushed as well.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 5e0e2318aa2a6fb8c2c693fb7ff995650e452054\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:54 2019 -0700\n\n scsi: lpfc: Fix too many sg segments spamming in kernel log\n \n This issue is specific to SLI-3 adapters, specifically when DIF is used.\n \n Once seen, this message floods the logs:\n 9064 BLKGRD: lpfc_scsi_prep_dma_buf_s3: Too many sg segments from\n dma_map_sg\n \n The driver, upon detecting an error such as too many elements in an sglist,\n misrepresents the error by treating it as a temporary resource issue by\n returning MLQUEUE_HOST_BUSY. In these cases, no retry will fix it and it\n should have been a hard error. The repeated retry was causing the spamming\n of the log.\n \n As for the initial reason of why an I/O encountered this issue at all is\n not clear as parameters set by the driver should have avoided this. The\n dm multipath maintainer has been notified of the issue.\n \n Fix by changing the return code for the dma mapping routines to indicate\n cases that are not retryable and return DID_ERROR on those cases.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 8c24a4f643edbcc7c8281b1f7527568f565dfbf8\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:53 2019 -0700\n\n scsi: lpfc: Fix crash due to port reset racing vs adapter error handling\n \n If the adapter encounters a condition which causes the adapter to fail\n (driver must detect the failure) simultaneously to a request to the driver\n to reset the adapter (such as a host_reset), the reset path will be racing\n with the asynchronously-detect adapter failure path. In the failing\n situation, one path has started to tear down the adapter data structures\n (io_wq\u0027s) while the other path has initiated a repeat of the teardown and\n is in the lpfc_sli_flush_xxx_rings path and attempting to access the\n just-freed data structures.\n \n Fix by the following:\n \n - In cases where an adapter failure is detected, rather than explicitly\n calling offline_eratt() to start the teardown, change the adapter state\n and let the later calls of posted work to the slowpath thread invoke the\n adapter recovery. In essence, this means all requests to reset are\n serialized on the slowpath thread.\n \n - Clean up the routine that restarts the adapter. If there is a failure\n from brdreset, don\u0027t immediately error and leave things in a partial\n state. Instead, ensure the adapter state is set and finish the teardown\n of structures before returning.\n \n - If in the scsi host reset handler and the board fails to reset and\n restart (which can be due to parallel reset/recovery paths), instead of\n hard failing and explicitly calling offline_eratt() (which gets into the\n redundant path), just fail out and let the asynchronous path resolve the\n adapter state.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 894bb17f0cb07a6503d096dbbd156e38a3640822\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:52 2019 -0700\n\n scsi: lpfc: Fix deadlock on host_lock during cable pulls\n \n During cable pull testing a deadlock was seen between lpfc_nlp_counters()\n vs lpfc_mbox_process_link_up() vs lpfc_work_list_done(). They are all\n waiting on the shost-\u003ehost_lock.\n \n Issue is all of these cases raise irq when taking out the lock but use\n spin_unlock_irq() when unlocking. The unlock path is will unconditionally\n re-enable interrupts in cases where irq state should be preserved. The\n re-enablement allowed the other paths to execute which then causes the\n deadlock.\n \n Fix by converting the lock/unlock to irqsave/irqrestore.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 6825b7bd32030486cc94bdd4bb512756fcd5b901\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:51 2019 -0700\n\n scsi: lpfc: Fix error in remote port address change\n \n In a test with high nvme remote port counts connected via a multi-hop FC\n switch config where switches were systematically reset (e.g. fabric\n partitioning and re-establishment), the nvme remote ports would switch\n addresses based on the switch reconfiguration events. The driver would get\n into a situation where the nvme port changed address, PLOGI and PRLI would\n succeed nvme transport registration occurred, but subsequent LS requests by\n the nvme subsystem failed due to a bad ndlp state and connectivity to the\n device failed.\n \n The driver hit a race condition on multiple devices that address swapped\n simultaneously. In cases where the driver notices the remote port structure\n came back as the same value as previously (meaning a nvme_rport structure\n was re-enabled and did not go through devloss_tmo/connect_tmo_failures on\n all controllers) the driver would unconditionally exit assuming the ndlp\n information was correct. But, if the ndlp\u0027s had been swapped, the ndlp had\n stale port state information, which when used by the LS request commands,\n would fail the commands.\n \n Fix by checking whether a node swap had occurred, and only exit if no ndlp\n swap had occurred.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit a6d10f24a04d658078582fc0a89b20f050c0ae68\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:50 2019 -0700\n\n scsi: lpfc: Fix driver nvme rescan logging\n \n In situations where zoning is not being used, thus NVMe initiators see\n other NVMe initiators as well as NVMe targets, a link bounce on an\n initiator will cause the NVMe initiators to spew \"6169\" State Error\n messages.\n \n The driver is not qualifying whether the remote port is a NVMe targer or\n not before calling the lpfc_nvme_rescan_port(), which validates the role\n and prints the message if its only an NVMe initiator.\n \n Fix by the following:\n \n - Before calling lpfc_nvme_rescan_port() ensure that the node is a NVMe\n storage target or a NVMe discovery controller.\n \n - Clean up implementation of lpfc_nvme_rescan_port. remoteport pointer\n will always be NULL if a NVMe initiator only. But, grabbing of\n remoteport pointer should be done under lock to coincide with the\n registering of the remote port with the fc transport.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit c26c265b168bb09013922846a51447603bcdfdc4\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:49 2019 -0700\n\n scsi: lpfc: Fix sg_seg_cnt for HBAs that don\u0027t support NVME\n \n On an SLI-3 adapter which does not support NVMe, but with the driver global\n attribute to enable nvme on any adapter if it does support NVMe\n (e.g. module parameter lpfc_enable_fc4_type\u003d3), the SGL and total SGE\n values are being munged by the protocol enablement when it shouldn\u0027t be.\n \n Correct by changing the location of where the NVME sgl information is being\n applied, which will avoid any SLI-3-based adapter.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit a643c6de1441e5cdab88452c46fe7c38b318009e\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:48 2019 -0700\n\n scsi: lpfc: Fix propagation of devloss_tmo setting to nvme transport\n \n If admin changes the devloss_tmo on an rport via the fc_remote_port rport\n dev_loss_tmo attribute, the value is on set on scsi stack. The change is\n not propagated to NVMe.\n \n The set routine in the lldd lacks the call to\n nvme_fc_set_remoteport_devloss() to set the value.\n \n Fix by adding the call to the lldd set routine.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit b95b21193c85db1b2797e38eea4f0fa72c7a087c\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:47 2019 -0700\n\n scsi: lpfc: Fix loss of remote port after devloss due to lack of RPIs\n \n In tests with remote ports contantly logging out/logging coupled with\n occassional local link bounce, if a remote port is disocnnected for longer\n than devloss_tmo and then subsequently reconnected, eventually the test\n will fail to login with the remote port and remote port connectivity is\n lost.\n \n When devloss_tmo expires, the driver does not free the node struct until\n the port or npiv instances is being deleted. The node is left allocated but\n the state set to UNUSED. If the node was in the process of logging in when\n the local link drop occurred, meaning the RPI was allocated for the node in\n order to send the ELS, but not yet registered which comes after successful\n login, the node is moved to the NPR state, and if devloss expires, to\n UNUSED state. If the remote port comes back, the node associated with it\n is restarted and this path happens to allocate a new RPI and overwrites the\n prior RPI value. In the cases where the port was logged in and loggs out,\n the path did release the RPI but did not set the node rpi value. In the\n cases where the remote port never finished logging in, the path never did\n the call to release the rpi. In this latter case, when the node is\n subsequently restore, the new rpi allocation overwrites the rpi that was\n not released, and the rpi is now leaked. Eventually the port will run out\n of RPI resources to log into new remote ports.\n \n Fix by following changes:\n \n - When an rpi is released, do so under locks and ensure the node rpi value\n is set to a non-allocated value (LPFC_RPI_ALLOC_ERROR). Note:\n refactored to a small service routine to avoid indentation issues.\n \n - When re-enabling a node, check the rpi value to determine if a new\n allocation is necessary. If already set, use the prior rpi.\n \n Enhanced logging to help in the future.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 96d156f95c92eb1d8a8bfae84ceb70396a036f43\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:46 2019 -0700\n\n scsi: lpfc: Fix devices that don\u0027t return after devloss followed by rediscovery\n \n If a remote port is removed and remains removed for devloss_tmo, if an RSCN\n is subsequently received indicating the presence of the remte port, the\n driver does not login to and rediscovery the remote port.\n \n Currently, in order to for a port to be rediscovered post an RSCN, the node\n state must be NPR to reflect not logged in. When devloss expires, the node\n state is marked UNUSED. When an RSCN occurs, the nodes referenced by the\n RSCN will have a NPR_2B_DISC flag set, but the re-login will only be\n attempted if the node is in NPR_NODE state. Thus the node is skipped over.\n \n Fix by recognizing the NPR_2B_DISC and UNUSED and transition the node back\n to NPR state to allow the re-login to take place.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 07f50997d66c3273121dd6b8a7d433cdfb5395c1\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:45 2019 -0700\n\n scsi: lpfc: Fix null ptr oops updating lpfc_devloss_tmo via sysfs attribute\n \n If an admin updates lpfc\u0027s devloss_tmo sysfs attribute, the kernel will\n oops.\n \n Coding of a loop allowed a new value (rport) to be set/checked for null\n followed by an older value (remoteport) checked for null to allow progress\n where the new value, even though null, will be referenced.\n \n Rework the logic to validate and prevent any reference to the null ptr.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 6ede2ddd8bdd28bdaf43d15e70dc0ef38c5448db\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:44 2019 -0700\n\n scsi: lpfc: Fix FLOGI handling across multiple link up/down conditions\n \n It\u0027s possible for the driver to initiate an FLOGI and before it completes,\n another link down/up transition occurs requiring a new FLOGI. Currently,\n nothing is done to abort/noop the older FLOGI request to the adapter, so if\n this transition occurs and the FLOGI completion is received after the link\n down/up transition, the driver may erroneously act on the older FLOGI. In\n most cases, the adapter properly terminates/fails the FLOGI, but there is a\n timing condition where the FLOGI may complete on the wire prior to the\n transition, but the response may not be seen/processed by the driver before\n the driver sees the link transition.\n \n Fix by having the link down handler in the driver run through any\n outstanding ELS\u0027s and change the completion handler of the ELS so that it\n will be no-op\u0027d and released.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 3ad348d94452d6ab3aa0316105fb9f34f8ed3140\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:43 2019 -0700\n\n scsi: lpfc: Fix oops when fewer hdwqs than cpus\n \n When tearing down the adapter for a reset, online/offline, or driver\n unload, the queue free routine would hit a GPF oops. This only occurs on\n conditions where the number of hardware queues created is fewer than the\n number of cpus in the system. In this condition cpus share a hardware\n queue. And of course, it\u0027s the 2nd cpu that shares a hardware that\n attempted to free it a second time and hit the oops.\n \n Fix by reworking the cpu to hardware queue mapping such that:\n Assignment of hardware queues to cpus occur in two passes:\n first pass: is first time assignment of a hardware queue to a cpu.\n This will set the LPFC_CPU_FIRST_IRQ flag for the cpu.\n second pass: for cpus that did not get a hardware queue they will\n be assigned one from a primary cpu (one set in first pass).\n \n Deletion of hardware queues is driven by cpu itteration, and queues\n will only be deleted if the LPFC_CPU_FIRST_IRQ flag is set.\n \n Also contains a few small cleanup fixes and a little better logging.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 4b0a42be26dd89cefaac89330070222ac8e14c01\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:42 2019 -0700\n\n scsi: lpfc: Fix irq raising in lpfc_sli_hba_down\n \n The adapter reset path (lpfc_sli_hba_down) is taking/releasing a lock with\n irq. But, the path is already under the hbalock which raised irq so it\u0027s\n unnecessary.\n \n Convert to simple lock/unlock.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 61184f1742c901160ceead85b3f1b3d03038f1d0\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:41 2019 -0700\n\n scsi: lpfc: Fix Oops in nvme_register with target logout/login\n \n lpfc_nvme_register_port hit a null prev_ndlp pointer in a test with lots of\n target ports swapping addresses. The oldport value was stale, thus it\u0027s\n ndlp (prev_ndlp set to it) was used.\n \n Fix by moving oldrport pointer checks, and if used prev_ndlp pointer\n assignment, to be done while the lock is held.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 08180db2542b94e93cbdffd2a42cac2db147114a\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:40 2019 -0700\n\n scsi: lpfc: Fix issuing init_vpi mbox on SLI-3 card\n \n The driver is inadvertently trying to issue an INIT_VPI mailbox command on\n an SLI-3 driver. The command is specific to SLI-4. When the call is made to\n send the command, if on an SLI-3 adapter, an array pointer is NULL and the\n driver will oops.\n \n Fix by restricting the command to SLI-4 adapters only.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 26d824ca45a80f7a95e7afe578e54680cef5236f\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:39 2019 -0700\n\n scsi: lpfc: Fix ADISC reception terminating login state if a NVME target\n \n If a target issues an ADISC to the port and the target is a NVME target,\n the driver is inadvertantly invalidating the login and marking the remote\n port as logged out. Communication with the target is lost.\n \n Revise the ADISC check so that FCP or NVME targets will be marked valid at\n the end of ADISC processing. Enhance logging to recognize condition\n better.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 7f20c1cb236afd75237f1302e60daef8b90b818b\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:38 2019 -0700\n\n scsi: lpfc: Fix discovery when target has no GID_FT information\n \n Some remote ports may be slow in registering their GID_FT protocol\n information with the fabric. If the remote port is an initiator, it may\n send PLOGI to the port before the GID_FT logic is complete. Meaning, after\n accepting the PLOGI, when the driver may see no response to the GID_FT that\n is issued after the login to determine the protocols supported so that\n proper PRLI\u0027s may be transmit. If the driver has no fc4 information, it\n currently stops and the remote port is not discovered.\n \n Fix by issuing a LOGO when there is no GID_FT information. The LOGO\n completion handling will attempt to re-login if the nport_id is still\n present.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 57178b9275508e59d6daf89b874c03d597342776\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:37 2019 -0700\n\n scsi: lpfc: Fix port relogin failure due to GID_FT interaction\n \n In cases of remote-port-side cable pull/replug, there happens to be a\n target that upon replug will send the port a PLOGI, a PRLI, and a LOGO.\n When this sequence is received by the driver, the PLOGI accepted and a\n GFT_ID is issued to find the protocol support for the remote port. While\n the GFT_ID is outstanding, a LOGO is received. The driver logs the remote\n port out and unregisters the RPI and schedules a new PLOGI transmission.\n However, the GFT_ID was not terminated. When it completed, the driver\n attempted to transition the remote port to PRLI transmission, which cancels\n the PLOGI scheduling. The PRLI transmit attempt is rejected by the adapter\n as the remote port is not logged in. No retry is attempted as it\u0027s expected\n the logout is noted and the supposedly scheduled PLOGI should address the\n state. As there is no PLOGI, the remote port does not get re-discovered.\n \n Fix by aborting the outstanding GFT_ID if the related remote port is logged\n out.\n \n Ensure a PRLI transmit attempt only occurs if the remote port is logging\n in. This avoids the incorrect attempt while logged out.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 296012285c908f6aacea83913ab9eaf273225d57\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:36 2019 -0700\n\n scsi: lpfc: Fix leak of ELS completions on adapter reset\n \n If the adapter is reset while there are outstanding ELS\u0027s, subsequent\n reinitialization of the adapter will fail as it has not recovered all of\n the io contexts relative to the ELS\u0027s.\n \n If an ELS timed out or otherwise failed and an the ELS was attempted to be\n aborted (which changes the ELS completion context), in causes where the\n driver generates completions for the outstanding IO as the adapter would\n not due to being reset, the driver released only the ELS context and failed\n to release the abort context. When the adapter went to reinit, as it had\n not received all of the contexts, it failed to reinit.\n \n Fix by having the ELS completion handler identify the driver-generated\n completion status and release the abort context.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 8d34a59caecda9a7ce0cad108ee64c37aa0c9812\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:35 2019 -0700\n\n scsi: lpfc: Fix failure to clear non-zero eq_delay after io rate reduction\n \n Unusually high IO latency can be observed with little IO in progress. The\n latency may remain high regardless of amount of IO and can only be cleared\n by forcing lpfc_fcp_imax values to non-zero and then back to zero.\n \n The driver\u0027s eq_delay mechanism that scales the interrupt coalescing based\n on io completion load failed to reduce or turn off coalescing when load\n decreased. Specifically, if no io completed on a cpu within an eq_delay\n polling window, the eq delay processing was skipped and no change was made\n to the coalescing values. This left the coalescing values set when they\n were no longer applicable.\n \n Fix by always clearing the percpu counters for each time period and always\n run the eq_delay calculations if an eq has a non-zero coalescing value.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 3cee98db2610fb633fcbdd0f85385f60dabbaf75\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:34 2019 -0700\n\n scsi: lpfc: Fix crash on driver unload in wq free\n \n If a timer routine uses workqueues, it could fire before the workqueue is\n allocated.\n \n Fix by allocating the workqueue before the timer routines are setup\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 1d755d6477cafea1d371ecc841e22adda0a43d27\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:33 2019 -0700\n\n scsi: lpfc: Fix ELS field alignments\n \n After seeing some interoperability issues with ADISC, it was determined the\n ELS definitions in lpfc were using types that allowed the compiler to add\n pad to the structure, causing the structure to no longer be per spec. The\n offending structures are ADISC, FAN, and RNID.\n \n This patch implements the simple fix of eliminating the pad by forcing the\n compiler to pack the structure. Care was taken to ensure field accesses\n won\u0027t be by operations that would hit a bad field alignment.\n \n The better solution would be to convert to the uapi fc header definitions,\n but the number of changes required to do is rather intrusive so this course\n of action was deferred.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 4f1a2fef2af6b8c746bc965e5f87cd3cdaf5f1f8\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:32 2019 -0700\n\n scsi: lpfc: Fix PLOGI failure with high remoteport count\n \n When connected to a high number of remote ports, the driver is encountering\n PLOGI errors. The errors are due to adapter detected failures indicating\n illegal field values.\n \n Turns out the driver was prematurely clearing an RPI bitmask before waiting\n for an UNREG_RPI mailbox completion. This allowed the RPI to be reused\n before it was actually available.\n \n Fix by clearing RPI bitmask only after UNREG_RPI mailbox completion.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 31f06d2e73726160645f8d9976a0b3f42e136da7\nAuthor: James Smart \u003cjsmart2021@gmail.com\u003e\nDate: Wed Aug 14 16:56:31 2019 -0700\n\n scsi: lpfc: Limit xri count for kdump environment\n \n scsi-mq operation inherently performs pre-allocation of resources for\n blk-mq request queues. Even though the kdump environment reduces the\n configuration to a single CPU, thus 1 hardware queue, which helps\n significantly, the resources are still rather large due to the per request\n allocations. blk-mq pre-allocations can be over 4KB per request. With\n adapter can_queue values in the 4k or 8k range, this can easily be 32MBs\n before any other driver memory is factored in. Driver SGL DMA buffer\n allocation can be up to 8KB per request as well adding an additional\n 64MB. Totals are well over 100MB for a single shost. Given kdump memory\n auto-sizing utilities don\u0027t accommodate this amount of memory well, it\u0027s\n very possible for kdump to fail due to lack of memory.\n \n Fix by having the driver recognize that it is booting within a kdump\n context and reduce the number of requests it will support to a more\n reasonable value.\n \n Signed-off-by: Dick Kennedy \u003cdick.kennedy@broadcom.com\u003e\n Signed-off-by: James Smart \u003cjsmart2021@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit a967783300bf2ba55a29255c0a591a554c651993\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Sun Jul 21 17:12:29 2019 +0530\n\n scsi: lpfc: remove NULL check before some freeing functions\n \n As dma_pool_destroy and mempool_destroy functions has NULL check. We may\n not need NULL check before calling them.\n \n Fix below warnings reported by coccicheck\n ./drivers/scsi/lpfc/lpfc_mem.c:252:2-18: WARNING: NULL check before some\n freeing functions is not needed.\n ./drivers/scsi/lpfc/lpfc_mem.c:255:2-18: WARNING: NULL check before some\n freeing functions is not needed.\n ./drivers/scsi/lpfc/lpfc_mem.c:258:2-18: WARNING: NULL check before some\n freeing functions is not needed.\n ./drivers/scsi/lpfc/lpfc_mem.c:261:2-18: WARNING: NULL check before some\n freeing functions is not needed.\n ./drivers/scsi/lpfc/lpfc_mem.c:265:2-18: WARNING: NULL check before some\n freeing functions is not needed.\n ./drivers/scsi/lpfc/lpfc_mem.c:269:2-17: WARNING: NULL check before some\n freeing functions is not needed.\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Reviewed-by: James Smart \u003cjames.smart@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit d508e31df10ff1f309fddb88d7e830315bdbb93a\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Aug 19 13:25:57 2019 +0530\n\n scsi: ufs-qcom: Make structure ufs_hba_qcom_vops constant\n \n Static structure ufs_hba_qcom_vops, of type ufs_hba_variant_ops, is used\n only once, when it is passed as the second argument to function\n ufshcd_pltfrm_init(). In the definition of ufshcd_pltfrm_init(), its second\n parameter (corresponding to ufs_hba_qcom_vops) is declared as\n constant. Hence declare ufs_hba_qcom_vops itself constant as well to\n protect it from unintended modification. Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Reviewed-by: Vivek Gautam \u003cvivek.gautam@codeaurora.org\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 78d4b1327d8bd60b58f422f6df02853c33558b11\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Mon Aug 12 16:31:34 2019 +0800\n\n scsi: lpfc: use spin_lock_irqsave in IRQ context\n \n As spin_unlock_irq will enable interrupts.\n Function lpfc_findnode_rpi is called from\n lpfc_sli_abts_err_handler (./drivers/scsi/lpfc/lpfc_sli.c)\n \u003c- lpfc_sli_async_event_handler\n \u003c- lpfc_sli_process_unsol_iocb\n \u003c- lpfc_sli_handle_fast_ring_event\n \u003c- lpfc_sli_fp_intr_handler\n \u003c- lpfc_sli_intr_handler\n and lpfc_sli_intr_handler is an interrupt handler.\n \n Interrupts are enabled in interrupt handler. Use\n spin_lock_irqsave/spin_unlock_irqrestore instead of spin_(un)lock_irq in\n IRQ context to avoid this.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Reviewed-by: James Smart \u003cjames.smart@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit ee9a256cd8d0f7d23d7dd310d7fdc23f28cada0d\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Thu Aug 8 09:35:25 2019 +0800\n\n scsi: lpfc: remove redundant code\n \n Remove the redundant initialization code.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Reviewed-by: James Smart \u003cjames.smart@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 9730ddfb123db3f37ace7571176661257712c99b\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Tue Aug 13 14:23:49 2019 +0100\n\n scsi: fnic: remove redundant assignment of variable rc\n \n Variable ret is initialized to a value that is never read and it is\n re-assigned later and immediately returns. Clean up the code by removing\n rc and just returning 0.\n \n [mkp: typo]\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Acked-by: Karan Tilak Kumar \u003ckartilak@cisco.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 3cf92f4bfccbb3e378cd86dc46e09c2bc18cda26\nAuthor: Martin Wilck \u003cmwilck@suse.com\u003e\nDate: Wed Aug 14 13:28:29 2019 +0000\n\n scsi: qla2xxx: cleanup trace buffer initialization\n \n Avoid code duplication between qla2x00_alloc_offload_mem() and\n qla2x00_alloc_fw_dump() by moving the FCE and EFT buffer allocation and\n initialization to separate functions. Cleanly track failure and success by\n making sure that the ha-\u003eeft, ha-\u003efce and respective eft_dma, fce_dma\n members are set if and only if the buffers are properly allocated and\n initialized. Avoid pointless buffer reallocation. Eliminate some goto\n statements. Make sure the fce_enabled flag is cleared when the FCE buffer\n is freed.\n \n Fixes: ad0a0b01f088 (\"scsi: qla2xxx: Fix Firmware dump size for Extended login and Exchange Offload\")\n Fixes: a28d9e4ef997 (\"scsi: qla2xxx: Add support for multiple fwdump templates/segments\")\n Cc: Joe Carnuccio \u003cjoe.carnuccio@cavium.com\u003e\n Cc: Quinn Tran \u003cqutran@marvell.com\u003e\n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Cc: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Martin Wilck \u003cmwilck@suse.com\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit edbd56472a636ab396f5ee6783e8438fa725a6ee\nAuthor: Martin Wilck \u003cmwilck@suse.com\u003e\nDate: Wed Aug 14 13:28:27 2019 +0000\n\n scsi: qla2xxx: qla2x00_alloc_fw_dump: set ha-\u003eeft\n \n In qla2x00_alloc_fw_dump(), an existing EFT buffer (e.g. from previous\n invocation of qla2x00_alloc_offload_mem()) is freed. The buffer is then\n re-allocated, but without setting the eft and eft_dma fields to the new\n values.\n \n Fixes: a28d9e4ef997 (\"scsi: qla2xxx: Add support for multiple fwdump templates/segments\")\n Cc: Joe Carnuccio \u003cjoe.carnuccio@cavium.com\u003e\n Cc: Quinn Tran \u003cqutran@marvell.com\u003e\n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Cc: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Martin Wilck \u003cmwilck@suse.com\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 2f7576347cf3dd993f1ffe5e6018b4b764319c52\nAuthor: Xin Long \u003clucien.xin@gmail.com\u003e\nDate: Mon Aug 19 22:02:50 2019 +0800\n\n sctp: remove net sctp.x_enable working as a global switch\n \n The netns sctp feature flags shouldn\u0027t work as a global switch,\n which is mostly like a firewall/netfilter\u0027s job. Also, it will\n break asoc as it discard or accept chunks incorrectly when net\n sctp.x_enable is changed after the asoc is created.\n \n Since each type of chunk\u0027s processing function will check the\n corresp asoc\u0027s feature flag, this \u0027global switch\u0027 should be\n removed, and net sctp.x_enable will only work as the default\n feature flags for the future sctp sockets/endpoints.\n \n Signed-off-by: Xin Long \u003clucien.xin@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 56dd525abd56f7acd7b44a52935726e3ada4916c\nAuthor: Xin Long \u003clucien.xin@gmail.com\u003e\nDate: Mon Aug 19 22:02:49 2019 +0800\n\n sctp: add SCTP_AUTH_SUPPORTED sockopt\n \n SCTP_AUTH_SUPPORTED sockopt is used to set enpoint\u0027s auth\n flag. With this feature, each endpoint will have its own\n flag for its future asoc\u0027s auth_capable, instead of netns\n auth flag.\n \n Note that when both ep\u0027s auth_enable is enabled, endpoint\n auth related data should be initialized. If asconf_enable\n is also set, SCTP_CID_ASCONF/SCTP_CID_ASCONF_ACK should\n be added into auth_chunk_list.\n \n Signed-off-by: Xin Long \u003clucien.xin@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 03f961270f4256fe9f47b94aea889bd26877216b\nAuthor: Xin Long \u003clucien.xin@gmail.com\u003e\nDate: Mon Aug 19 22:02:48 2019 +0800\n\n sctp: add sctp_auth_init and sctp_auth_free\n \n This patch is to factor out sctp_auth_init and sctp_auth_free\n functions, and sctp_auth_init will also be used in the next\n patch for SCTP_AUTH_SUPPORTED sockopt.\n \n Signed-off-by: Xin Long \u003clucien.xin@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 219f9ea4d3b797f0337dece61e4e8255840e47d0\nAuthor: Xin Long \u003clucien.xin@gmail.com\u003e\nDate: Mon Aug 19 22:02:47 2019 +0800\n\n sctp: use ep and asoc auth_enable properly\n \n sctp has per endpoint auth flag and per asoc auth flag, and\n the asoc one should be checked when coming to asoc and the\n endpoint one should be checked when coming to endpoint.\n \n Signed-off-by: Xin Long \u003clucien.xin@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit df2c71ffdfae58961981d7cbcccea93688fc4e96\nAuthor: Xin Long \u003clucien.xin@gmail.com\u003e\nDate: Mon Aug 19 22:02:46 2019 +0800\n\n sctp: add SCTP_ASCONF_SUPPORTED sockopt\n \n SCTP_ASCONF_SUPPORTED sockopt is used to set enpoint\u0027s asconf\n flag. With this feature, each endpoint will have its own flag\n for its future asoc\u0027s asconf_capable, instead of netns asconf\n flag.\n \n Note that when both ep\u0027s asconf_enable and auth_enable are\n enabled, SCTP_CID_ASCONF and SCTP_CID_ASCONF_ACK should be\n added into auth_chunk_list.\n \n Signed-off-by: Xin Long \u003clucien.xin@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a2eeacc830438d738d21230325662fe5c22c3bf0\nAuthor: Xin Long \u003clucien.xin@gmail.com\u003e\nDate: Mon Aug 19 22:02:45 2019 +0800\n\n sctp: check asoc peer.asconf_capable before processing asconf\n \n asconf chunks should be dropped when the asoc doesn\u0027t support\n asconf feature.\n \n Signed-off-by: Xin Long \u003clucien.xin@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit bb2ded26028be00204dbe2153f98fbd1902a2187\nAuthor: Xin Long \u003clucien.xin@gmail.com\u003e\nDate: Mon Aug 19 22:02:44 2019 +0800\n\n sctp: not set peer.asconf_capable in sctp_association_init\n \n asoc-\u003epeer.asconf_capable is to be set during handshake, and its\n value should be initialized to 0. net-\u003esctp.addip_noauth will be\n checked in sctp_process_init when processing INIT_ACK on client\n and COOKIE_ECHO on server.\n \n Signed-off-by: Xin Long \u003clucien.xin@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4e27428fb5626f966aa961b1aad8751f2ebeef72\nAuthor: Xin Long \u003clucien.xin@gmail.com\u003e\nDate: Mon Aug 19 22:02:43 2019 +0800\n\n sctp: add asconf_enable in struct sctp_endpoint\n \n This patch is to make addip/asconf flag per endpoint,\n and its value is initialized by the per netns flag,\n net-\u003esctp.addip_enable.\n \n It also replaces the checks of net-\u003esctp.addip_enable\n with ep-\u003easconf_enable in some places.\n \n Signed-off-by: Xin Long \u003clucien.xin@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit af809709e9df2a44137429ba3694c339a11b710d\nAuthor: Li RongQing \u003clirongqing@baidu.com\u003e\nDate: Mon Aug 19 20:05:15 2019 +0800\n\n net: remove empty inet_exit_net\n \n Pointer members of an object with static storage duration, if not\n explicitly initialized, will be initialized to a NULL pointer. The\n net namespace API checks if this pointer is not NULL before using it,\n it are safe to remove the function.\n \n Signed-off-by: Li RongQing \u003clirongqing@baidu.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 14b54ac4fbb92f1a52966cc33477436de0f72965\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Mon Aug 19 10:52:08 2019 +0300\n\n tc-testing: concurrency: wrap piped rule update commands\n \n Concurrent tests use several commands to update rules in parallel: \u0027find\u0027\n prints names of batch files in tmp directory and pipes result to \u0027xargs\u0027\n which runs instance of tc per batch file in parallel. This breaks when used\n with ns plugin that adds \u0027ip netns exec $NS\u0027 prefix to the command, which\n causes only first command in pipe to be executed in namespace:\n \n \u003d\u003d\u003d\u003d\u003d\u003e Test e41d: Add 1M flower filters with 10 parallel tc instances\n -----\u003e prepare stage\n ns/SubPlugin.adjust_command\n adjust_command: stage is setup; inserting netns stuff in command [/bin/mkdir tmp] list [[\u0027/bin/mkdir\u0027, \u0027tmp\u0027]]\n adjust_command: return command [ip netns exec tcut /bin/mkdir tmp]\n command \"ip netns exec tcut /bin/mkdir tmp\"\n ns/SubPlugin.adjust_command\n adjust_command: stage is setup; inserting netns stuff in command [/sbin/tc qdisc add dev ens1f0 ingress] list [[\u0027/sbin/tc\u0027, \u0027qdisc\u0027, \u0027add\u0027, \u0027dev\u0027, \u0027ens1f0\u0027, \u0027ingress\u0027]]\n adjust_command: return command [ip netns exec tcut /sbin/tc qdisc add dev ens1f0 ingress]\n command \"ip netns exec tcut /sbin/tc qdisc add dev ens1f0 ingress\"\n ns/SubPlugin.adjust_command\n adjust_command: stage is setup; inserting netns stuff in command [./tdc_multibatch.py ens1f0 tmp 100000 10 add] list [[\u0027./tdc_multibatch.py\u0027, \u0027ens1f0\u0027, \u0027tmp\u0027, \u0027100000\u0027, \u002710\u0027, \u0027add\u0027]]\n adjust_command: return command [ip netns exec tcut ./tdc_multibatch.py ens1f0 tmp 100000 10 add]\n command \"ip netns exec tcut ./tdc_multibatch.py ens1f0 tmp 100000 10 add\"\n -----\u003e execute stage\n ns/SubPlugin.adjust_command\n adjust_command: stage is execute; inserting netns stuff in command [find tmp/add* -print | xargs -n 1 -P 10 /sbin/tc -b] list [[\u0027find\u0027, \u0027tmp/add*\u0027, \u0027-print\u0027, \u0027|\u0027, \u0027xargs\u0027, \u0027-n\u0027, \u00271\u0027, \u0027-P\u0027, \u002710\u0027, \u0027/sbin/tc\u0027, \u0027-b\u0027]\n ]\n adjust_command: return command [ip netns exec tcut find tmp/add* -print | xargs -n 1 -P 10 /sbin/tc -b]\n command \"ip netns exec tcut find tmp/add* -print | xargs -n 1 -P 10 /sbin/tc -b\"\n exit: 123\n exit: 0\n Cannot find device \"ens1f0\"\n Cannot find device \"ens1f0\"\n Command failed tmp/add_0:1\n Command failed tmp/add_1:1\n Cannot find device \"ens1f0\"\n Command failed tmp/add_2:1\n Cannot find device \"ens1f0\"\n Command failed tmp/add_4:1\n Cannot find device \"ens1f0\"\n Command failed tmp/add_3:1\n Cannot find device \"ens1f0\"\n Command failed tmp/add_5:1\n Cannot find device \"ens1f0\"\n Command failed tmp/add_6:1\n Cannot find device \"ens1f0\"\n Command failed tmp/add_8:1\n Cannot find device \"ens1f0\"\n Command failed tmp/add_7:1\n Cannot find device \"ens1f0\"\n Command failed tmp/add_9:1\n \n Fix the issue by executing whole compound command in namespace by wrapping\n it in \u0027bash -c\u0027 invocation.\n \n Fixes: 489ce2f42514 (\"tc-testing: Restore original behaviour for namespaces in tdc\")\n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c11a99e7942383a11ef858aa1679a94d6eea5917\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Mon Aug 19 10:52:07 2019 +0300\n\n tc-testing: use dedicated DUMMY interface name for dummy dev\n \n A lot of tests reuse $DEV1 veth name for naming dummy device. This causes\n problem when tdc is invoked without specifying a test group and tries to\n execute all tests. In this case tdc instantiates ns plugin, which creates\n veth pair once before running tests. However, if any of the tests that\n reuse $DEV1 run before test that depend on ns plugin, it will delete $DEV1\n as a part of teardown section:\n \n \u003d\u003d\u003d\u003d\u003d\u003e Test 3b88: Delete ingress qdisc twice [3770/41080]\n -----\u003e prepare stage\n ns/SubPlugin.adjust_command\n adjust_command: stage is setup; inserting netns stuff in command [/sbin/ip link add dev v0p1 type dummy || /bin/true] list [[\u0027/sbin/ip\u0027, \u0027link\u0027, \u0027add\u0027, \u0027dev\u0027, \u0027v0p1\u0027, \u0027type\u0027, \u0027dummy\u0027, \u0027||\u0027, \u0027/bin/true\u0027]]\n adjust_command: return command [ip netns exec tcut /sbin/ip link add dev v0p1 type dummy || /bin/true]\n command \"ip netns exec tcut /sbin/ip link add dev v0p1 type dummy || /bin/true\"\n ns/SubPlugin.adjust_command\n adjust_command: stage is setup; inserting netns stuff in command [/sbin/tc qdisc add dev v0p1 ingress] list [[\u0027/sbin/tc\u0027, \u0027qdisc\u0027, \u0027add\u0027, \u0027dev\u0027, \u0027v0p1\u0027, \u0027ingress\u0027]]\n adjust_command: return command [ip netns exec tcut /sbin/tc qdisc add dev v0p1 ingress]\n command \"ip netns exec tcut /sbin/tc qdisc add dev v0p1 ingress\"\n ns/SubPlugin.adjust_command\n adjust_command: stage is setup; inserting netns stuff in command [/sbin/tc qdisc del dev v0p1 ingress] list [[\u0027/sbin/tc\u0027, \u0027qdisc\u0027, \u0027del\u0027, \u0027dev\u0027, \u0027v0p1\u0027, \u0027ingress\u0027]]\n adjust_command: return command [ip netns exec tcut /sbin/tc qdisc del dev v0p1 ingress]\n command \"ip netns exec tcut /sbin/tc qdisc del dev v0p1 ingress\"\n -----\u003e execute stage\n ns/SubPlugin.adjust_command\n adjust_command: stage is execute; inserting netns stuff in command [/sbin/tc qdisc del dev v0p1 ingress] list [[\u0027/sbin/tc\u0027, \u0027qdisc\u0027, \u0027del\u0027, \u0027dev\u0027, \u0027v0p1\u0027, \u0027ingress\u0027]]\n adjust_command: return command [ip netns exec tcut /sbin/tc qdisc del dev v0p1 ingress]\n command \"ip netns exec tcut /sbin/tc qdisc del dev v0p1 ingress\"\n -----\u003e verify stage\n ns/SubPlugin.adjust_command\n adjust_command: stage is verify; inserting netns stuff in command [/sbin/tc qdisc show dev v0p1] list [[\u0027/sbin/tc\u0027, \u0027qdisc\u0027, \u0027show\u0027, \u0027dev\u0027, \u0027v0p1\u0027]]\n adjust_command: return command [ip netns exec tcut /sbin/tc qdisc show dev v0p1]\n command \"ip netns exec tcut /sbin/tc qdisc show dev v0p1\"\n -----\u003e teardown stage\n ns/SubPlugin.adjust_command\n adjust_command: stage is teardown; inserting netns stuff in command [/sbin/ip link del dev v0p1 type dummy] list [[\u0027/sbin/ip\u0027, \u0027link\u0027, \u0027del\u0027, \u0027dev\u0027, \u0027v0p1\u0027, \u0027type\u0027, \u0027dummy\u0027]]\n adjust_command: return command [ip netns exec tcut /sbin/ip link del dev v0p1 type dummy]\n command \"ip netns exec tcut /sbin/ip link del dev v0p1 type dummy\"\n \n After this ns-dependent tests will fail because dev doesn\u0027t exist:\n \n \u003d\u003d\u003d\u003d\u003d\u003e Test 901f: Add fw filter with prio at 32-bit maxixum\n -----\u003e prepare stage\n ns/SubPlugin.adjust_command\n adjust_command: stage is setup; inserting netns stuff in command [/sbin/tc qdisc add dev v0p1 ingress] list [[\u0027/sbin/tc\u0027, \u0027qdisc\u0027, \u0027add\u0027, \u0027dev\u0027, \u0027v0p1\u0027, \u0027ingress\u0027]]\n adjust_command: return command [ip netns exec tcut /sbin/tc qdisc add dev v0p1 ingress]\n command \"ip netns exec tcut /sbin/tc qdisc add dev v0p1 ingress\"\n \n -----\u003e prepare stage *** Could not execute: \"$TC qdisc add dev $DEV1 ingress\"\n \n -----\u003e prepare stage *** Error message: \"Cannot find device \"v0p1\"\n \"\n returncode 1; expected [0]\n \n -----\u003e prepare stage *** Aborting test run.\n \n \u003c_io.BufferedReader name\u003d3\u003e *** stdout ***\n \n \u003c_io.BufferedReader name\u003d5\u003e *** stderr ***\n \"-----\u003e prepare stage\" did not complete successfully\n Exception \u003cclass \u0027__main__.PluginMgrTestFail\u0027\u003e (\u0027setup\u0027, None, \u0027\"-----\u003e prepare stage\" did not complete successfully\u0027) (caught in test_runner, running test 477 901f Add fw filter with prio at 32-bit maxixum stage\n setup)\n ---------------\n traceback\n File \"./tdc.py\", line 371, in test_runner\n res \u003d run_one_test(pm, args, index, tidx)\n File \"./tdc.py\", line 272, in run_one_test\n prepare_env(args, pm, \u0027setup\u0027, \"-----\u003e prepare stage\", tidx[\"setup\"])\n File \"./tdc.py\", line 247, in prepare_env\n \u0027\"{}\" did not complete successfully\u0027.format(prefix))\n ---------------\n \n Fix the issue by introducing standalone $DUMMY config variable and\n substitute all usage of $DEV1 in tests that don\u0027t depend on ns plugin.\n \n Fixes: 489ce2f42514 (\"tc-testing: Restore original behaviour for namespaces in tdc\")\n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6636fb310681d716f50c2e4064daf44ff358685e\nAuthor: Hayes Wang \u003chayeswang@realtek.com\u003e\nDate: Mon Aug 19 11:15:19 2019 +0800\n\n r8152: fix accessing skb after napi_gro_receive\n \n Fix accessing skb after napi_gro_receive which is caused by\n commit 47922fcde536 (\"r8152: support skb_add_rx_frag\").\n \n Fixes: 47922fcde536 (\"r8152: support skb_add_rx_frag\")\n Signed-off-by: Hayes Wang \u003chayeswang@realtek.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 64b95df91f44837f961321a6c555409885ca59c4\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 19 21:07:05 2019 +0100\n\n drm/i915: Assume exclusive access to objects inside resume\n \n Inside gtt_restore_mappings() we currently take the obj-\u003eresv-\u003elock, but\n in the future we need to avoid taking this fs-reclaim tainted lock as we\n need to extend the coverage of the vm-\u003emutex. Take advantage of the\n single-threaded nature of the early resume phase, and do a single\n wbinvd() to flush all the GTT objects en masse.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190819200705.3631-1-chris@chris-wilson.co.uk\n\ncommit 8725fcd99a3084a7a15a6e70882bfa3fe7925f52\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Mon Aug 19 21:36:39 2019 +0200\n\n x86/irq: Check for VECTOR_UNUSED directly\n \n It\u0027s simpler and more intuitive to directly check for VECTOR_UNUSED than\n checking whether the other error codes are not set.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/caeaca93-5ee1-cea1-8894-3aa0d5b19241@gmail.com\n\ncommit d6f83427ff422b3a65e4a6bd84c010f78d4a30a5\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Mon Aug 19 21:36:09 2019 +0200\n\n x86/irq: Move IS_ERR_OR_NULL() check into common do_IRQ() code\n \n Both the 64bit and the 32bit handle_irq() implementation check the irq\n descriptor pointer with IS_ERR_OR_NULL() and return failure. That can be\n done simpler in the common do_IRQ() code.\n \n This reduces the 64bit handle_irq() function to a wrapper around\n generic_handle_irq_desc(). Invoke it directly from do_IRQ() to spare the\n extra function call.\n \n [ tglx: Got rid of the #ifdef and massaged changelog ]\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/2ec758c7-9aaa-73ab-f083-cc44c86aa741@gmail.com\n\ncommit e30c44e2e59c98023997632815cbc5e273991c25\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Mon Aug 19 21:34:47 2019 +0200\n\n x86/irq: Improve definition of VECTOR_SHUTDOWN et al\n \n These values are used with IS_ERR(), so it\u0027s more intuitive to define\n them like a standard PTR_ERR() of a negative errno.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/146835e8-c086-4e85-7ece-bcba6795e6db@gmail.com\n\ncommit 8084499bd7d44b2c96658d788e329f3c232eb050\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Tue Aug 13 09:56:02 2019 +0800\n\n irqchip/irq-ingenic-tcu: Fix COMPILE_TEST building\n \n While do COMPILE_TEST building, if GENERIC_IRQ_CHIP is\n not selected, it fails:\n \n drivers/irqchip/irq-ingenic-tcu.o: In function `ingenic_tcu_intc_cascade\u0027:\n irq-ingenic-tcu.c:(.text+0x13f): undefined reference to `irq_get_domain_generic_chip\u0027\n drivers/irqchip/irq-ingenic-tcu.o: In function `ingenic_tcu_irq_init\u0027:\n irq-ingenic-tcu.c:(.init.text+0x97): undefined reference to `irq_generic_chip_ops\u0027\n irq-ingenic-tcu.c:(.init.text+0xdd): undefined reference to `__irq_alloc_domain_generic_chips\u0027\n irq-ingenic-tcu.c:(.init.text+0x10b): undefined reference to `irq_get_domain_generic_chip\u0027\n \n select GENERIC_IRQ_CHIP to fix this.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Fixes: 9536eba03ec7 (\"irqchip: Add irq-ingenic-tcu driver\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Acked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: \u003cjason@lakedaemon.net\u003e\n Cc: \u003cmaz@kernel.org\u003e\n Cc: \u003cpaul@crapouillou.net\u003e\n Cc: \u003cmalat@debian.org\u003e\n Cc: \u003clinux-kernel@vger.kernel.org\u003e\n Cc: \u003clinux-mips@vger.kernel.org\u003e\n Cc: \u003clinux-clk@vger.kernel.org\u003e\n\ncommit 3becd97e032a756f26e48d70a46bf776509ba6b2\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Sun Aug 18 17:51:24 2019 +0200\n\n MIPS: Octeon: Fix a typo in #define OCTOEN_SERIAL_LEN\n \n It should be OCTEON_SERIAL_LEN.\n Update the #define and use it accordingly\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: ralf@linux-mips.org\n Cc: jhogan@kernel.org\n Cc: linux-mips@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: kernel-janitors@vger.kernel.org\n\ncommit 0ac33e4e9b5e4ccdd43c9d4e77882a1af0e41dec\nAuthor: Ilya Leoshkevich \u003ciii@linux.ibm.com\u003e\nDate: Fri Aug 16 18:06:04 2019 +0200\n\n selftests: use \"$(MAKE)\" instead of \"make\"\n \n When doing \"make kselftest TARGETS\u003dbpf -j12\", bpf progs end up being\n compiled sequentially and thus slowly.\n \n The reason is that parent make (tools/testing/selftests/Makefile) does\n not share its jobserver with child make\n (tools/testing/selftests/bpf/Makefile), therefore the latter runs with\n -j1.\n \n Change all instances of \"make\" to \"$(MAKE)\", so that the whole make\n hierarchy runs using a single jobserver.\n \n Signed-off-by: Ilya Leoshkevich \u003ciii@linux.ibm.com\u003e\n Signed-off-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n\ncommit edde25e55d87f0046db64ed7ce634e5a586229e3\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Fri Aug 16 21:57:38 2019 +0200\n\n net: phy: realtek: support NBase-T MMD EEE registers on RTL8125\n \n Emulate the 802.3bz MMD EEE registers for 2.5Gbps EEE on RTL8125.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 99b60d56a35b18af267f275559a530db372bfad7\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Fri Aug 16 21:56:27 2019 +0200\n\n net: phy: add EEE-related constants\n \n Add EEE-related constants. This includes the new MMD EEE registers for\n NBase-T / 802.3bz.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 607f625b86f9ba4c5c04df190c73010de65f4e91\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Fri Aug 16 18:06:54 2019 +0300\n\n net: flow_offload: convert block_ing_cb_list to regular list type\n \n RCU list block_ing_cb_list is protected by rcu read lock in\n flow_block_ing_cmd() and with flow_indr_block_ing_cb_lock mutex in all\n functions that use it. However, flow_block_ing_cmd() needs to call blocking\n functions while iterating block_ing_cb_list which leads to following\n suspicious RCU usage warning:\n \n [ 401.510948] \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n [ 401.510952] WARNING: suspicious RCU usage\n [ 401.510993] 5.3.0-rc3+ #589 Not tainted\n [ 401.510996] -----------------------------\n [ 401.511001] include/linux/rcupdate.h:265 Illegal context switch in RCU read-side critical section!\n [ 401.511004]\n other info that might help us debug this:\n \n [ 401.511008]\n rcu_scheduler_active \u003d 2, debug_locks \u003d 1\n [ 401.511012] 7 locks held by test-ecmp-add-v/7576:\n [ 401.511015] #0: 00000000081d71a5 (sb_writers#4){.+.+}, at: vfs_write+0x166/0x1d0\n [ 401.511037] #1: 000000002bd338c3 (\u0026of-\u003emutex){+.+.}, at: kernfs_fop_write+0xef/0x1b0\n [ 401.511051] #2: 00000000c921c634 (kn-\u003ecount#317){.+.+}, at: kernfs_fop_write+0xf7/0x1b0\n [ 401.511062] #3: 00000000a19cdd56 (\u0026dev-\u003emutex){....}, at: sriov_numvfs_store+0x6b/0x130\n [ 401.511079] #4: 000000005425fa52 (pernet_ops_rwsem){++++}, at: unregister_netdevice_notifier+0x30/0x140\n [ 401.511092] #5: 00000000c5822793 (rtnl_mutex){+.+.}, at: unregister_netdevice_notifier+0x35/0x140\n [ 401.511101] #6: 00000000c2f3507e (rcu_read_lock){....}, at: flow_block_ing_cmd+0x5/0x130\n [ 401.511115]\n stack backtrace:\n [ 401.511121] CPU: 21 PID: 7576 Comm: test-ecmp-add-v Not tainted 5.3.0-rc3+ #589\n [ 401.511124] Hardware name: Supermicro SYS-2028TP-DECR/X10DRT-P, BIOS 2.0b 03/30/2017\n [ 401.511127] Call Trace:\n [ 401.511138] dump_stack+0x85/0xc0\n [ 401.511146] ___might_sleep+0x100/0x180\n [ 401.511154] __mutex_lock+0x5b/0x960\n [ 401.511162] ? find_held_lock+0x2b/0x80\n [ 401.511173] ? __tcf_get_next_chain+0x1d/0xb0\n [ 401.511179] ? mark_held_locks+0x49/0x70\n [ 401.511194] ? __tcf_get_next_chain+0x1d/0xb0\n [ 401.511198] __tcf_get_next_chain+0x1d/0xb0\n [ 401.511251] ? uplink_rep_async_event+0x70/0x70 [mlx5_core]\n [ 401.511261] tcf_block_playback_offloads+0x39/0x160\n [ 401.511276] tcf_block_setup+0x1b0/0x240\n [ 401.511312] ? mlx5e_rep_indr_setup_tc_cb+0xca/0x290 [mlx5_core]\n [ 401.511347] ? mlx5e_rep_indr_tc_block_unbind+0x50/0x50 [mlx5_core]\n [ 401.511359] tc_indr_block_get_and_ing_cmd+0x11b/0x1e0\n [ 401.511404] ? mlx5e_rep_indr_tc_block_unbind+0x50/0x50 [mlx5_core]\n [ 401.511414] flow_block_ing_cmd+0x7e/0x130\n [ 401.511453] ? mlx5e_rep_indr_tc_block_unbind+0x50/0x50 [mlx5_core]\n [ 401.511462] __flow_indr_block_cb_unregister+0x7f/0xf0\n [ 401.511502] mlx5e_nic_rep_netdevice_event+0x75/0xb0 [mlx5_core]\n [ 401.511513] unregister_netdevice_notifier+0xe9/0x140\n [ 401.511554] mlx5e_cleanup_rep_tx+0x6f/0xe0 [mlx5_core]\n [ 401.511597] mlx5e_detach_netdev+0x4b/0x60 [mlx5_core]\n [ 401.511637] mlx5e_vport_rep_unload+0x71/0xc0 [mlx5_core]\n [ 401.511679] esw_offloads_disable+0x5b/0x90 [mlx5_core]\n [ 401.511724] mlx5_eswitch_disable.cold+0xdf/0x176 [mlx5_core]\n [ 401.511759] mlx5_device_disable_sriov+0xab/0xb0 [mlx5_core]\n [ 401.511794] mlx5_core_sriov_configure+0xaf/0xd0 [mlx5_core]\n [ 401.511805] sriov_numvfs_store+0xf8/0x130\n [ 401.511817] kernfs_fop_write+0x122/0x1b0\n [ 401.511826] vfs_write+0xdb/0x1d0\n [ 401.511835] ksys_write+0x65/0xe0\n [ 401.511847] do_syscall_64+0x5c/0xb0\n [ 401.511857] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n [ 401.511862] RIP: 0033:0x7fad892d30f8\n [ 401.511868] Code: 89 02 48 c7 c0 ff ff ff ff eb bb 0f 1f 80 00 00 00 00 f3 0f 1e fa 48 8d 05 25 96 0d 00 8b 00 85 c0 75 17 b8 01 00 00 00 0f 05 \u003c48\u003e 3d 00 f0 ff ff 77 60 c3 0f 1f 80 00 00 00 00 48 83\n ec 28 48 89\n [ 401.511871] RSP: 002b:00007ffca2a9fad8 EFLAGS: 00000246 ORIG_RAX: 0000000000000001\n [ 401.511875] RAX: ffffffffffffffda RBX: 0000000000000002 RCX: 00007fad892d30f8\n [ 401.511878] RDX: 0000000000000002 RSI: 000055afeb072a90 RDI: 0000000000000001\n [ 401.511881] RBP: 000055afeb072a90 R08: 00000000ffffffff R09: 000000000000000a\n [ 401.511884] R10: 000055afeb058710 R11: 0000000000000246 R12: 0000000000000002\n [ 401.511887] R13: 00007fad893a8780 R14: 0000000000000002 R15: 00007fad893a3740\n \n To fix the described incorrect RCU usage, convert block_ing_cb_list from\n RCU list to regular list and protect it with flow_indr_block_ing_cb_lock\n mutex in flow_block_ing_cmd().\n \n Fixes: 1150ab0f1b33 (\"flow_offload: support get multi-subsystem block\")\n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Acked-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b09d6e47397409faf12f72edd140956e4cfe3011\nAuthor: Shameer Kolothum \u003cshameerali.kolothum.thodi@huawei.com\u003e\nDate: Tue Jul 23 17:06:37 2019 +0100\n\n vfio/type1: remove duplicate retrieval of reserved regions\n \n As we now already have the reserved regions list, just pass that into\n vfio_iommu_has_sw_msi() fn.\n \n Signed-off-by: Shameer Kolothum \u003cshameerali.kolothum.thodi@huawei.com\u003e\n Reviewed-by: Eric Auger \u003ceric.auger@redhat.com\u003e\n Signed-off-by: Alex Williamson \u003calex.williamson@redhat.com\u003e\n\ncommit a717072007e8aedd3f951726d8cf55454860b30d\nAuthor: Shameer Kolothum \u003cshameerali.kolothum.thodi@huawei.com\u003e\nDate: Tue Jul 23 17:06:36 2019 +0100\n\n vfio/type1: Add IOVA range capability support\n \n This allows the user-space to retrieve the supported IOVA\n range(s), excluding any non-relaxable reserved regions. The\n implementation is based on capability chains, added to\n VFIO_IOMMU_GET_INFO ioctl.\n \n Signed-off-by: Shameer Kolothum \u003cshameerali.kolothum.thodi@huawei.com\u003e\n Reviewed-by: Eric Auger \u003ceric.auger@redhat.com\u003e\n Signed-off-by: Alex Williamson \u003calex.williamson@redhat.com\u003e\n\ncommit 9b77e5c79840fc334a5b7f770c5ab0c09dc0e028\nAuthor: Shameer Kolothum \u003cshameerali.kolothum.thodi@huawei.com\u003e\nDate: Tue Jul 23 17:06:35 2019 +0100\n\n vfio/type1: check dma map request is within a valid iova range\n \n This checks and rejects any dma map request outside valid iova\n range.\n \n Signed-off-by: Shameer Kolothum \u003cshameerali.kolothum.thodi@huawei.com\u003e\n Reviewed-by: Eric Auger \u003ceric.auger@redhat.com\u003e\n Signed-off-by: Alex Williamson \u003calex.williamson@redhat.com\u003e\n\ncommit c84b82dd3e593db217f23c60f7edae02c76a3c4c\nAuthor: Cao jin \u003ccaoj.fnst@cn.fujitsu.com\u003e\nDate: Fri Aug 9 19:46:12 2019 +0800\n\n x86/fixmap: Cleanup outdated comments\n \n Remove stale comments and fix the not longer valid pagetable entry\n reference.\n \n Signed-off-by: Cao jin \u003ccaoj.fnst@cn.fujitsu.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190809114612.2569-1-caoj.fnst@cn.fujitsu.com\n\ncommit 78becab98b8f2f7d22e615675d1a701a67a885af\nAuthor: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\nDate: Mon Aug 19 11:51:17 2019 +1000\n\n vfio/spapr_tce: Fix incorrect tce_iommu_group memory free\n \n The @tcegrp variable is used in 1) a loop over attached groups\n 2) it stores a pointer to a newly allocated tce_iommu_group if 1) found\n nothing. However the error handler does not distinguish how we got there\n and incorrectly releases memory for a found+incompatible group.\n \n This fixes it by adding another error handling case.\n \n Fixes: 0bd971676e68 (\"powerpc/powernv/npu: Add compound IOMMU groups\")\n Signed-off-by: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\n Reviewed-by: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\n Signed-off-by: Alex Williamson \u003calex.williamson@redhat.com\u003e\n\ncommit eee413e620f408e1e88d94b16d05f45407939d1b\nAuthor: Parav Pandit \u003cparav@mellanox.com\u003e\nDate: Thu Aug 8 09:12:54 2019 -0500\n\n vfio-mdev/mtty: Simplify interrupt generation\n \n While generating interrupt, mdev_state is already available for which\n interrupt is generated.\n Instead of doing indirect way from state-\u003edevice-\u003euuid-\u003e to searching\n state linearly in linked list on every interrupt generation,\n directly use the available state.\n \n Hence, simplify the code to use mdev_state and remove unused helper\n function with that.\n \n Reviewed-by: Cornelia Huck \u003ccohuck@redhat.com\u003e\n Signed-off-by: Parav Pandit \u003cparav@mellanox.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Alex Williamson \u003calex.williamson@redhat.com\u003e\n\ncommit db2cb969e8aed2395620fe2cb0bffd194c02b4b1\nAuthor: Cornelia Huck \u003ccohuck@redhat.com\u003e\nDate: Tue Aug 6 11:30:00 2019 +0200\n\n vfio: re-arrange vfio region definitions\n \n It is easy to miss already defined region types. Let\u0027s re-arrange\n the definitions a bit and add more comments to make it hopefully\n a bit clearer.\n \n No functional change.\n \n Signed-off-by: Cornelia Huck \u003ccohuck@redhat.com\u003e\n Signed-off-by: Alex Williamson \u003calex.williamson@redhat.com\u003e\n\ncommit f45daadfe1add9f468f3ae1f7e2d9b235fe80748\nAuthor: Shameer Kolothum \u003cshameerali.kolothum.thodi@huawei.com\u003e\nDate: Tue Jul 23 17:06:34 2019 +0100\n\n vfio/type1: Update iova list on detach\n \n Get a copy of iova list on _group_detach and try to update the list.\n On success replace the current one with the copy. Leave the list as\n it is if update fails.\n \n Signed-off-by: Shameer Kolothum \u003cshameerali.kolothum.thodi@huawei.com\u003e\n Reviewed-by: Eric Auger \u003ceric.auger@redhat.com\u003e\n Signed-off-by: Alex Williamson \u003calex.williamson@redhat.com\u003e\n\ncommit af029169b8fdae31064624d60b5469a3da95ad32\nAuthor: Shameer Kolothum \u003cshameerali.kolothum.thodi@huawei.com\u003e\nDate: Tue Jul 23 17:06:33 2019 +0100\n\n vfio/type1: Check reserved region conflict and update iova list\n \n This retrieves the reserved regions associated with dev group and\n checks for conflicts with any existing dma mappings. Also update\n the iova list excluding the reserved regions.\n \n Reserved regions with type IOMMU_RESV_DIRECT_RELAXABLE are\n excluded from above checks as they are considered as directly\n mapped regions which are known to be relaxable.\n \n Signed-off-by: Shameer Kolothum \u003cshameerali.kolothum.thodi@huawei.com\u003e\n Reviewed-by: Eric Auger \u003ceric.auger@redhat.com\u003e\n Signed-off-by: Alex Williamson \u003calex.williamson@redhat.com\u003e\n\ncommit 1108696aecf048433bf77806570f57bdbb6ef724\nAuthor: Shameer Kolothum \u003cshameerali.kolothum.thodi@huawei.com\u003e\nDate: Tue Jul 23 17:06:32 2019 +0100\n\n vfio/type1: Introduce iova list and add iommu aperture validity check\n \n This introduces an iova list that is valid for dma mappings. Make\n sure the new iommu aperture window doesn\u0027t conflict with the current\n one or with any existing dma mappings during attach.\n \n Signed-off-by: Shameer Kolothum \u003cshameerali.kolothum.thodi@huawei.com\u003e\n Reviewed-by: Eric Auger \u003ceric.auger@redhat.com\u003e\n Signed-off-by: Alex Williamson \u003calex.williamson@redhat.com\u003e\n\ncommit a87fa5bcf86df2f232ee8b69e64a46825bcf6cc6\nAuthor: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\nDate: Wed Mar 6 09:51:20 2019 -0800\n\n arm64: dts: qcom: qcs404: Update memory map to v3\n \n Update the reserved-memory map to version 3, to adjust to changes in the\n remoteprocs.\n \n Reviewed-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit cc3375607d79c368ec1dbeb4c2922371fffa19ca\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 19 19:44:03 2019 +0100\n\n drm/i915: Use 0 for the unordered context\n \n Since commit 078dec3326e2 (\"dma-buf: add dma_fence_get_stub\") the 0\n fence context became an impossible match as it is used for an always\n signaled fence. We can simplify our timeline tracking by knowing that 0\n always means no match.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190819184404.24200-1-chris@chris-wilson.co.uk\n Link: https://patchwork.freedesktop.org/patch/msgid/20190819175109.5241-1-chris@chris-wilson.co.uk\n\ncommit 00452ba9fdb5bf6fb5fea1dae5227b4bbed44fc4\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Mon Aug 12 12:21:13 2019 +0200\n\n x86/platform/intel/iosf_mbi Rewrite locking\n \n There are 2 problems with the old iosf PMIC I2C bus arbritration code which\n need to be addressed:\n \n 1. The lockdep code complains about a possible deadlock in the\n iosf_mbi_[un]block_punit_i2c_access code:\n \n [ 6.712662] \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n [ 6.712673] WARNING: possible circular locking dependency detected\n [ 6.712685] 5.3.0-rc2+ #79 Not tainted\n [ 6.712692] ------------------------------------------------------\n [ 6.712702] kworker/0:1/7 is trying to acquire lock:\n [ 6.712712] 00000000df1c5681 (iosf_mbi_block_punit_i2c_access_count_mutex){+.+.}, at: iosf_mbi_unblock_punit_i2c_access+0x13/0x90\n [ 6.712739]\n but task is already holding lock:\n [ 6.712749] 0000000067cb23e7 (iosf_mbi_punit_mutex){+.+.}, at: iosf_mbi_block_punit_i2c_access+0x97/0x186\n [ 6.712768]\n which lock already depends on the new lock.\n \n [ 6.712780]\n the existing dependency chain (in reverse order) is:\n [ 6.712792]\n -\u003e #1 (iosf_mbi_punit_mutex){+.+.}:\n [ 6.712808] __mutex_lock+0xa8/0x9a0\n [ 6.712818] iosf_mbi_block_punit_i2c_access+0x97/0x186\n [ 6.712831] i2c_dw_acquire_lock+0x20/0x30\n [ 6.712841] i2c_dw_set_reg_access+0x15/0xb0\n [ 6.712851] i2c_dw_probe+0x57/0x473\n [ 6.712861] dw_i2c_plat_probe+0x33e/0x640\n [ 6.712874] platform_drv_probe+0x38/0x80\n [ 6.712884] really_probe+0xf3/0x380\n [ 6.712894] driver_probe_device+0x59/0xd0\n [ 6.712905] bus_for_each_drv+0x84/0xd0\n [ 6.712915] __device_attach+0xe4/0x170\n [ 6.712925] bus_probe_device+0x9f/0xb0\n [ 6.712935] deferred_probe_work_func+0x79/0xd0\n [ 6.712946] process_one_work+0x234/0x560\n [ 6.712957] worker_thread+0x50/0x3b0\n [ 6.712967] kthread+0x10a/0x140\n [ 6.712977] ret_from_fork+0x3a/0x50\n [ 6.712986]\n -\u003e #0 (iosf_mbi_block_punit_i2c_access_count_mutex){+.+.}:\n [ 6.713004] __lock_acquire+0xe07/0x1930\n [ 6.713015] lock_acquire+0x9d/0x1a0\n [ 6.713025] __mutex_lock+0xa8/0x9a0\n [ 6.713035] iosf_mbi_unblock_punit_i2c_access+0x13/0x90\n [ 6.713047] i2c_dw_set_reg_access+0x4d/0xb0\n [ 6.713058] i2c_dw_probe+0x57/0x473\n [ 6.713068] dw_i2c_plat_probe+0x33e/0x640\n [ 6.713079] platform_drv_probe+0x38/0x80\n [ 6.713089] really_probe+0xf3/0x380\n [ 6.713099] driver_probe_device+0x59/0xd0\n [ 6.713109] bus_for_each_drv+0x84/0xd0\n [ 6.713119] __device_attach+0xe4/0x170\n [ 6.713129] bus_probe_device+0x9f/0xb0\n [ 6.713140] deferred_probe_work_func+0x79/0xd0\n [ 6.713150] process_one_work+0x234/0x560\n [ 6.713160] worker_thread+0x50/0x3b0\n [ 6.713170] kthread+0x10a/0x140\n [ 6.713180] ret_from_fork+0x3a/0x50\n [ 6.713189]\n other info that might help us debug this:\n \n [ 6.713202] Possible unsafe locking scenario:\n \n [ 6.713212] CPU0 CPU1\n [ 6.713221] ---- ----\n [ 6.713229] lock(iosf_mbi_punit_mutex);\n [ 6.713239] lock(iosf_mbi_block_punit_i2c_access_count_mutex);\n [ 6.713253] lock(iosf_mbi_punit_mutex);\n [ 6.713265] lock(iosf_mbi_block_punit_i2c_access_count_mutex);\n [ 6.713276]\n *** DEADLOCK ***\n \n In practice can never happen because only the first caller which\n increments iosf_mbi_block_punit_i2c_access_count will also take\n iosf_mbi_punit_mutex, that is the whole purpose of the counter, which\n itself is protected by iosf_mbi_block_punit_i2c_access_count_mutex.\n \n But there is no way to tell the lockdep code about this and we really\n want to be able to run a kernel with lockdep enabled without these\n warnings being triggered.\n \n 2. The lockdep warning also points out another real problem, if 2 threads\n both are in a block of code protected by iosf_mbi_block_punit_i2c_access\n and the first thread to acquire the block exits before the second thread\n then the second thread will call mutex_unlock on iosf_mbi_punit_mutex,\n but it is not the thread which took the mutex and unlocking by another\n thread is not allowed.\n \n Fix this by getting rid of the notion of holding a mutex for the entire\n duration of the PMIC accesses, be it either from the PUnit side, or from an\n in kernel I2C driver. In general holding a mutex after exiting a function\n is a bad idea and the above problems show this case is no different.\n \n Instead 2 counters are now used, one for PMIC accesses from the PUnit\n and one for accesses from in kernel I2C code. When access is requested\n now the code will wait (using a waitqueue) for the counter of the other\n type of access to reach 0 and on release, if the counter reaches 0 the\n wakequeue is woken.\n \n Note that the counter approach is necessary to allow nested calls.\n The main reason for this is so that a series of i2c transfers can be done\n with the punit blocked from accessing the bus the whole time. This is\n necessary to be able to safely read/modify/write a PMIC register without\n racing with the PUNIT doing the same thing.\n \n Allowing nested iosf_mbi_block_punit_i2c_access() calls also is desirable\n from a performance pov since the whole dance necessary to block the PUnit\n from accessing the PMIC I2C bus is somewhat expensive.\n \n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Andy Shevchenko \u003candy.shevchenko@gmail.com\u003e\n Link: https://lkml.kernel.org/r/20190812102113.95794-1-hdegoede@redhat.com\n\ncommit 96158346b59bfb1d531b61fd7696ce370acebbf1\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 19 18:19:00 2019 +0100\n\n drm/i915: Select DMABUF_SELFTESTS for the default i915.ko debug build\n \n Include the DMABUF_SELFTESTS as part of the standard build for IGT, so\n that they can be run by igt/dmabuf\n \n Testcase: igt/dmabuf\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tomi Sarvela \u003ctomi.p.sarvela@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190819171900.4501-1-chris@chris-wilson.co.uk\n\ncommit 1ee170ea3f0dcf3a4b34f7e7c36559e84bb0d3d6\nAuthor: Kees Cook \u003ckeescook@chromium.org\u003e\nDate: Mon Aug 19 10:24:52 2019 -0700\n\n lkdtm: Split WARNING into separate tests\n \n There are three paths through the kernel code exception logging:\n \n - BUG (no configurable printk message)\n - WARN_ON (no configurable printk message)\n - WARN (configurable printk message)\n \n LKDTM was not testing WARN_ON(). This is needed to evaluate the placement\n of the \"cut here\" line, which needs special handling in each of the\n three exceptions (and between architectures that implement instruction\n exceptions to implement the code exceptions).\n \n Signed-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\n\ncommit d4e0f82ac840bf3d16b25d60f261b429603138a9\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Thu Aug 15 11:44:52 2019 -0300\n\n media: pixfmt-compressed.rst: improve H264/HEVC/MPEG1+2/VP8+9 documentation\n \n The existing documentation was incorrect and did not correspond\n to how actual codec drivers implemented this.\n \n Update the documentation to explicitly specify what is actually\n expected.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 051f8d2d86f848cd29632908f56d277ac5800a50\nAuthor: Tomasz Figa \u003ctfiga@chromium.org\u003e\nDate: Thu Aug 15 11:44:51 2019 -0300\n\n media: docs-rst: Document memory-to-memory video decoder interface\n \n Due to complexity of the video decoding process, the V4L2 drivers of\n stateful decoder hardware require specific sequences of V4L2 API calls\n to be followed. These include capability enumeration, initialization,\n decoding, seek, pause, dynamic resolution change, drain and end of\n stream.\n \n Specifics of the above have been discussed during Media Workshops at\n LinuxCon Europe 2012 in Barcelona and then later Embedded Linux\n Conference Europe 2014 in Düsseldorf. The de facto Codec API that\n originated at those events was later implemented by the drivers we already\n have merged in mainline, such as s5p-mfc or coda.\n \n The only thing missing was the real specification included as a part of\n Linux Media documentation. Fix it now and document the decoder part of\n the Codec API.\n \n Signed-off-by: Tomasz Figa \u003ctfiga@chromium.org\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit b867d9ce0915a1edf580add42db64e6c764928eb\nAuthor: Maxime Jourdan \u003cmjourdan@baylibre.com\u003e\nDate: Thu Aug 15 11:44:50 2019 -0300\n\n media: vicodec: set flags for vdec/stateful OUTPUT coded formats\n \n Tag all the coded formats where the vicodec stateful decoder supports\n dynamic resolution switching and bytestream parsing.\n \n Signed-off-by: Maxime Jourdan \u003cmjourdan@baylibre.com\u003e\n Reviewed-by: Paul Kocialkowski \u003cpaul.kocialkowski@bootlin.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit abf8dd2922b51710387e44307884615f072abc18\nAuthor: Maxime Jourdan \u003cmjourdan@baylibre.com\u003e\nDate: Thu Aug 15 11:44:49 2019 -0300\n\n media: mtk-vcodec: flag OUTPUT formats with V4L2_FMT_FLAG_DYN_RESOLUTION\n \n Tag all the coded formats where the mtk-vcodec decoder supports dynamic\n resolution switching.\n \n Signed-off-by: Maxime Jourdan \u003cmjourdan@baylibre.com\u003e\n Reviewed-by: Paul Kocialkowski \u003cpaul.kocialkowski@bootlin.com\u003e\n Reviewed-by: Alexandre Courbot \u003cacourbot@chromium.org\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 5aecb7d030370fc76cfd15649c218ba702c46b29\nAuthor: Maxime Jourdan \u003cmjourdan@baylibre.com\u003e\nDate: Thu Aug 15 11:44:47 2019 -0300\n\n media: venus: vdec: flag OUTPUT formats with V4L2_FMT_FLAG_DYN_RESOLUTION\n \n Tag all the coded formats where the venus vdec supports dynamic\n resolution switching.\n \n Signed-off-by: Maxime Jourdan \u003cmjourdan@baylibre.com\u003e\n Acked-by: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\n Reviewed-by: Paul Kocialkowski \u003cpaul.kocialkowski@bootlin.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 60a039eb27f921abaef4778999db510dd75d5e48\nAuthor: Maxime Jourdan \u003cmjourdan@baylibre.com\u003e\nDate: Thu Aug 15 11:44:46 2019 -0300\n\n media: videodev2.h: add V4L2_FMT_FLAG_DYN_RESOLUTION\n \n Add an enum_fmt format flag to specifically tag coded formats where\n dynamic resolution switching is supported by the device.\n \n This is useful for some codec drivers that can support dynamic\n resolution switching for one or more of their listed coded formats. It\n allows userspace to know whether it should extract the video parameters\n itself, or if it can rely on the device to send V4L2_EVENT_SOURCE_CHANGE\n when such changes are detected.\n \n Signed-off-by: Maxime Jourdan \u003cmjourdan@baylibre.com\u003e\n Reviewed-by: Paul Kocialkowski \u003cpaul.kocialkowski@bootlin.com\u003e\n Reviewed-by: Alexandre Courbot \u003cacourbot@chromium.org\u003e\n Acked-by: Tomasz Figa \u003ctfiga@chromium.org\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 2b770bee787daec66ed86baf4fc83275f949c8ac\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Thu Aug 15 11:44:45 2019 -0300\n\n media: videodev2.h: add V4L2_FMT_FLAG_CONTINUOUS_BYTESTREAM\n \n Add an enum_fmt format flag to specifically tag coded formats where\n full bytestream parsing is supported by the device.\n \n Some stateful decoders are capable of fully parsing a bytestream,\n but others require that userspace pre-parses the bytestream into\n frames or fields (see the corresponding pixelformat descriptions\n for details).\n \n If this flag is set, then this pre-parsing step is not required\n (but still possible, of course).\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Reviewed-by: Paul Kocialkowski \u003cpaul.kocialkowski@bootlin.com\u003e\n Reviewed-by: Alexandre Courbot \u003cacourbot@chromium.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 760327930e10f6108ce9e44854ec3b5ef30cdcbd\nAuthor: Hertz Wong \u003chertz.wong@rock-chips.com\u003e\nDate: Fri Aug 16 13:01:32 2019 -0300\n\n media: hantro: Enable H264 decoding on rk3288\n \n Now that the generic bits have been added, we can activate H264 decoding\n on rk3288.\n \n Signed-off-by: Hertz Wong \u003chertz.wong@rock-chips.com\u003e\n Signed-off-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit dea0a82f3d22ad8a4e113a60ca94aa173af36756\nAuthor: Hertz Wong \u003chertz.wong@rock-chips.com\u003e\nDate: Fri Aug 16 13:01:31 2019 -0300\n\n media: hantro: Add support for H264 decoding on G1\n \n Add the G1 specific bits to support H264 decoding.\n \n Signed-off-by: Hertz Wong \u003chertz.wong@rock-chips.com\u003e\n Signed-off-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Tested-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit a9471e25629b02f8a110232994dc4b63788e766f\nAuthor: Hertz Wong \u003chertz.wong@rock-chips.com\u003e\nDate: Fri Aug 16 13:01:30 2019 -0300\n\n media: hantro: Add core bits to support H264 decoding\n \n Add helpers and patch hantro_{drv,v4l2}.c to prepare addition of H264\n decoding support.\n \n Signed-off-by: Hertz Wong \u003chertz.wong@rock-chips.com\u003e\n Signed-off-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Tested-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 56d8d6413d0218a0b3cebc5decdcaedc9a0b7e4e\nAuthor: Jonathan Neuschäfer \u003cj.neuschaefer@gmx.net\u003e\nDate: Thu Aug 8 18:36:28 2019 +0200\n\n drm/drv: Use // for comments in example code\n \n This improves Sphinx output in two ways:\n \n - It avoids an unmatched single-quote (\u0027), about which Sphinx complained:\n \n Documentation/gpu/drm-internals.rst:298:\n WARNING: Could not lex literal_block as \"c\". Highlighting skipped.\n \n An alternative approach would be to replace \"can\u0027t\" with a word that\n doesn\u0027t have a single-quote.\n \n - It lets Sphinx format the comments in italics and grey, making the\n code slightly easier to read.\n \n Signed-off-by: Jonathan Neuschäfer \u003cj.neuschaefer@gmx.net\u003e\n Acked-by: Daniel Vetter \u003cdaniel@ffwll.ch\u003e [via irc]\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808163629.14280-1-j.neuschaefer@gmx.net\n\ncommit 2989f6451084aed3f8cc9992477f7a9bf57a3716\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 19 10:59:27 2019 +0100\n\n dma-buf: Add selftests for dma-fence\n \n Exercise the dma-fence API exported to drivers.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Acked-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190819095928.32091-2-chris@chris-wilson.co.uk\n\ncommit 3636e82135ce5ade87c238e61a315df23dbcd4d6\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Aug 16 23:21:51 2019 +0100\n\n soc: samsung: chipid: Fix memory leak in error path\n \n Currently when the call to product_id_to_soc_id fails there\n is a memory leak of soc_dev_attr-\u003erevision and soc_dev_attr\n on the error return path. Fix this by adding a common error\n return path that frees there obects and use this for two\n error return paths.\n \n Addresses-Coverity: (\"Resource leak\")\n Fixes: 3253b7b7cd44 (\"soc: samsung: Add exynos chipid driver support\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 9536b64ac0d6e3151963a11441dde7ade045fb29\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 19 10:59:26 2019 +0100\n\n dma-buf: Introduce selftesting framework\n \n In light of recent review slip ups, the absence of a suite of tests for\n dma-buf became apparent. Given the current plethora of testing\n frameworks, opt for one already in use by Intel\u0027s CI and so allow easy\n hook up into igt.\n \n We introduce a new module that when loaded will execute the list of\n selftests and their subtest. The names of the selftests are put into the\n modinfo as parameters so that igt can identify each, and run them\n independently, principally for ease of error reporting.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: Tomi Sarvela \u003ctomi.p.sarvela@intel.com\u003e\n Acked-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190819095928.32091-1-chris@chris-wilson.co.uk\n\ncommit aa4fffec310da5eb6d6ef09509b03e780bdf248a\nAuthor: Steven Price \u003csteven.price@arm.com\u003e\nDate: Fri Aug 16 10:31:07 2019 +0100\n\n drm/panfrost: Remove opp table when unloading\n \n The devfreq opp table needs to be removed when unloading the driver to\n free the memory associated with it.\n \n Signed-off-by: Steven Price \u003csteven.price@arm.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816093107.30518-3-steven.price@arm.com\n\ncommit e21dd290881b20c9a37e4a79568841e9ffebeb67\nAuthor: Steven Price \u003csteven.price@arm.com\u003e\nDate: Fri Aug 16 10:31:05 2019 +0100\n\n drm/panfrost: Enable devfreq to work without regulator\n \n If there is no regulator defined for the GPU then still control the\n frequency using the supplied clock.\n \n Some boards have clock control but no (direct) control of the regulator.\n For example the HiKey960 uses a mailbox protocol to a MCU to control\n frequencies and doesn\u0027t directly control the voltage. This patch allows\n frequency control of the GPU on this system.\n \n Signed-off-by: Steven Price \u003csteven.price@arm.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816093107.30518-1-steven.price@arm.com\n\ncommit 7282f7645d06bf0afe0a3c11ab92d9392528b819\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Tue Aug 13 09:01:15 2019 -0600\n\n drm/panfrost: Implement per FD address spaces\n \n Up until now, a single shared GPU address space was used. This is not\n ideal as there\u0027s no protection between processes and doesn\u0027t work for\n supporting the same GPU/CPU VA feature. Most importantly, this will\n hopefully mitigate Alyssa\u0027s fear of WebGL, whatever that is.\n \n Most of the changes here are moving struct drm_mm and struct\n panfrost_mmu objects from the per device struct to the per FD struct.\n The critical function is panfrost_mmu_as_get() which handles allocating\n and switching the h/w address spaces.\n \n There\u0027s 3 states an AS can be in: free, allocated, and in use. When a\n job runs, it requests an address space and then marks it not in use when\n job is complete(but stays assigned). The first time thru, we find a free\n AS in the alloc_mask and assign the AS to the FD. Then the next time\n thru, we most likely already have our AS and we just mark it in use with\n a ref count. We need a ref count because we have multiple job slots. If\n the job/FD doesn\u0027t have an AS assigned and there are no free ones, then\n we pick an allocated one not in use from our LRU list and switch the AS\n from the old FD to the new one.\n \n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Cc: Steven Price \u003csteven.price@arm.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813150115.30338-1-robh@kernel.org\n\ncommit 3efdf83ca0f9d3149f8c2201dad86a74fd952f91\nAuthor: Wei Yongjun \u003cweiyongjun1@huawei.com\u003e\nDate: Wed Aug 14 04:48:14 2019 +0000\n\n drm/panfrost: Fix missing unlock on error in panfrost_mmu_map_fault_addr()\n \n Add the missing unlock before return from function panfrost_mmu_map_fault_addr()\n in the error handling case.\n \n Fixes: 187d2929206e (\"drm/panfrost: Add support for GPU heap allocations\")\n Signed-off-by: Wei Yongjun \u003cweiyongjun1@huawei.com\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190814044814.102294-1-weiyongjun1@huawei.com\n\ncommit 6c2eb77bf5667c0755f5a9a408f02f949925256d\nAuthor: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\nDate: Fri Aug 16 13:01:29 2019 -0300\n\n media: hantro: Move copy_metadata() before doing a decode operation\n \n Some decoders use intra slice/frame references. The capture buffer\n pointed by these references might be new and thus have invalid\n timestamp which prevents the decoder logic from retrieving the\n vb2_buffer object based on the output buf timestamp.\n Copy all metadata (including the timestamp) before starting the decode\n operation.\n \n Suggested-by: Jonas Karlman \u003cjonas@kwiboo.se\u003e\n Signed-off-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Tested-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 341772b82a3b83e7a7a2c0605c8c728e81b38319\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Fri Aug 16 13:01:28 2019 -0300\n\n media: cedrus: Specify H264 startcode and decoding mode\n \n The cedrus VPU is slice-based and expects V4L2_PIX_FMT_H264_SLICE\n buffers to contain H264 slices with no start code.\n \n Expose this to userspace with the newly added menu control.\n \n These two controls are specified as mandatory for applications,\n but we mark them as non-required on the driver side for\n backwards compatibility.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 3f715c64be6e6e1e1bb140fb1179ab0a712f94c3\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Fri Aug 16 13:01:27 2019 -0300\n\n media: cedrus: Cleanup control initialization\n \n In order to introduce other controls, the control initialization\n needs to support an initial struct v4l2_ctrl_control.\n \n While here, let\u0027s cleanup the control initialization,\n removing unneeded fields.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit c3adb85745ca6cc19532b2ee197d7abece1ac732\nAuthor: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\nDate: Fri Aug 16 13:01:26 2019 -0300\n\n media: uapi: h264: Get rid of the p0/b0/b1 ref-lists\n \n Those lists can be extracted from the dpb, let\u0027s simplify userspace\n life and build that list kernel-side (generic helpers will be provided\n for drivers that need this list).\n \n Signed-off-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Reviewed-by: Nicolas Dufresne \u003cnicolas.dufresne@collabora.com\u003e\n Reviewed-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Reviewed-by: Paul Kocialkowski \u003cpaul.kocialkowski@bootlin.com\u003e\n Tested-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 8cae93e090113e46bd29a99c1727d8f13ea12fdf\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Fri Aug 16 13:01:25 2019 -0300\n\n media: uapi: h264: Add the concept of start code\n \n Stateless decoders have different expectations about the\n start code that is prepended on H264 slices. Add a\n menu control to express the supported start code types\n (including no start code).\n \n Drivers are allowed to support only one start code type,\n but they can support both too.\n \n Note that this is independent of the H264 decoding mode,\n which specifies the granularity of the decoding operations.\n Either in frame-based or slice-based mode, this new control\n will allow to define the start code expected on H264 slices.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Tested-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 5604be66a56867a784e162299a48c214921ffa1b\nAuthor: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\nDate: Fri Aug 16 13:01:24 2019 -0300\n\n media: uapi: h264: Add the concept of decoding mode\n \n Some stateless decoders don\u0027t support per-slice decoding granularity\n (or at least not in a way that would make them efficient or easy to use).\n \n Expose a menu to control the supported decoding modes. Drivers are\n allowed to support only one decoding but they can support both too.\n \n To fully specify the decoding operation, we need to introduce\n a start_byte_offset, to indicate where slices start.\n \n Signed-off-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Reviewed-by: Paul Kocialkowski \u003cpaul.kocialkowski@bootlin.com\u003e\n Tested-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 7bb3c32abd7bafd346f667cccb7dfe9686f14ddd\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Fri Aug 16 13:01:23 2019 -0300\n\n media: uapi: h264: Rename pixel format\n \n The V4L2_PIX_FMT_H264_SLICE_RAW name was originally suggested\n because the pixel format would represent H264 slices without any\n start code.\n \n However, as we will now introduce a start code menu control,\n give the pixel format a more meaningful name, while it\u0027s\n still early enough to do so.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Tested-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 4333fb96ca1086d1cec0f93f78c453aa2dee8a5c\nAuthor: Rasmus Villemoes \u003clinux@rasmusvillemoes.dk\u003e\nDate: Fri Aug 16 13:01:22 2019 -0300\n\n media: lib/sort.c: implement sort() variant taking context argument\n \n Our list_sort() utility has always supported a context argument that\n is passed through to the comparison routine. Now there\u0027s a use case\n for the similar thing for sort().\n \n This implements sort_r by simply extending the existing sort function\n in the obvious way. To avoid code duplication, we want to implement\n sort() in terms of sort_r(). The naive way to do that is\n \n static int cmp_wrapper(const void *a, const void *b, const void *ctx)\n {\n int (*real_cmp)(const void*, const void*) \u003d ctx;\n return real_cmp(a, b);\n }\n \n sort(..., cmp) { sort_r(..., cmp_wrapper, cmp) }\n \n but this would do two indirect calls for each comparison. Instead, do\n as is done for the default swap functions - that only adds a cost of a\n single easily predicted branch to each comparison call.\n \n Aside from introducing support for the context argument, this also\n serves as preparation for patches that will eliminate the indirect\n comparison calls in common cases.\n \n Requested-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n \n Signed-off-by: Rasmus Villemoes \u003clinux@rasmusvillemoes.dk\u003e\n Signed-off-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Acked-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n Tested-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit ba7e4a08bbf7441664b3d140671db8d08ea15f22\nAuthor: Robin Murphy \u003crobin.murphy@arm.com\u003e\nDate: Thu Aug 15 19:37:37 2019 +0100\n\n iommu/arm-smmu: Add context init implementation hook\n \n Allocating and initialising a context for a domain is another point\n where certain implementations are known to want special behaviour.\n Currently the other half of the Cavium workaround comes into play here,\n so let\u0027s finish the job to get the whole thing right out of the way.\n \n Signed-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 62b993a36e4c2d60669beb3d1afe038ed44a41ec\nAuthor: Robin Murphy \u003crobin.murphy@arm.com\u003e\nDate: Thu Aug 15 19:37:36 2019 +0100\n\n iommu/arm-smmu: Add reset implementation hook\n \n Reset is an activity rife with implementation-defined poking. Add a\n corresponding hook, and use it to encapsulate the existing MMU-500\n details.\n \n Signed-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 3995e18689fda2434863bdf2a8cd19acea4e61f0\nAuthor: Robin Murphy \u003crobin.murphy@arm.com\u003e\nDate: Thu Aug 15 19:37:35 2019 +0100\n\n iommu/arm-smmu: Add configuration implementation hook\n \n Probing the ID registers and setting up the SMMU configuration is an\n area where overrides and workarounds may well be needed. Indeed, the\n Cavium workaround detection lives there at the moment, so let\u0027s break\n that out.\n \n Signed-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 6d7dff62afb0c7a880860148a8984d0cddc6e589\nAuthor: Robin Murphy \u003crobin.murphy@arm.com\u003e\nDate: Thu Aug 15 19:37:34 2019 +0100\n\n iommu/arm-smmu: Move Secure access quirk to implementation\n \n Move detection of the Secure access quirk to its new home, trimming it\n down in the process - time has proven that boolean DT flags are neither\n ideal nor necessarily sufficient, so it\u0027s highly unlikely we\u0027ll ever add\n more, let alone enough to justify the frankly overengineered parsing\n machinery.\n \n Signed-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit fc058d37b3450db3e146d475f85e6afd51888997\nAuthor: Robin Murphy \u003crobin.murphy@arm.com\u003e\nDate: Thu Aug 15 19:37:33 2019 +0100\n\n iommu/arm-smmu: Add implementation infrastructure\n \n Add some nascent infrastructure for handling implementation-specific\n details outside the flow of the architectural code. This will allow us\n to keep mutually-incompatible vendor-specific hooks in their own files\n where the respective interested parties can maintain them with minimal\n chance of conflicts. As somewhat of a template, we\u0027ll start with a\n general place to collect the relatively trivial existing quirks.\n \n Signed-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit c5fc64881f073504581ac55132e30f02ee61320b\nAuthor: Robin Murphy \u003crobin.murphy@arm.com\u003e\nDate: Thu Aug 15 19:37:32 2019 +0100\n\n iommu/arm-smmu: Rename arm-smmu-regs.h\n \n We\u0027re about to start using it for more than just register definitions,\n so generalise the name.\n \n Signed-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 00320ce6505821b405992c241ebe4c79f178bf8f\nAuthor: Robin Murphy \u003crobin.murphy@arm.com\u003e\nDate: Thu Aug 15 19:37:31 2019 +0100\n\n iommu/arm-smmu: Abstract GR0 accesses\n \n Clean up the remaining accesses to GR0 registers, so that everything is\n now neatly abstracted. This folds up the Non-Secure alias quirk as the\n first step towards moving it out of the way entirely. Although GR0 does\n technically contain some 64-bit registers (sGFAR and the weird SMMUv2\n HYPC and MONC stuff), they\u0027re not ones we have any need to access.\n \n Signed-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 19713fd40df8e65759e836129671be5f6f21c626\nAuthor: Robin Murphy \u003crobin.murphy@arm.com\u003e\nDate: Thu Aug 15 19:37:30 2019 +0100\n\n iommu/arm-smmu: Abstract context bank accesses\n \n Context bank accesses are fiddly enough to deserve a number of extra\n helpers to keep the callsites looking sane, even though there are only\n one or two of each.\n \n Signed-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit aadbf2143aeb61368f4b9240b4f8960a0d08c473\nAuthor: Robin Murphy \u003crobin.murphy@arm.com\u003e\nDate: Thu Aug 15 19:37:29 2019 +0100\n\n iommu/arm-smmu: Abstract GR1 accesses\n \n Introduce some register access abstractions which we will later use to\n encapsulate various quirks. GR1 is the easiest page to start with.\n \n Signed-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 6100576284e9df139477817e1a207b5fae1a3df8\nAuthor: Robin Murphy \u003crobin.murphy@arm.com\u003e\nDate: Thu Aug 15 19:37:28 2019 +0100\n\n iommu/arm-smmu: Get rid of weird \"atomic\" write\n \n The smmu_write_atomic_lq oddity made some sense when the context\n format was effectively tied to CONFIG_64BIT, but these days it\u0027s\n simpler to just pick an explicit access size based on the format\n for the one-and-a-half times we actually care.\n \n Signed-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 71e8a8cdaff995a46e3e186a736636747cbd2f50\nAuthor: Robin Murphy \u003crobin.murphy@arm.com\u003e\nDate: Thu Aug 15 19:37:27 2019 +0100\n\n iommu/arm-smmu: Split arm_smmu_tlb_inv_range_nosync()\n \n Since we now use separate iommu_gather_ops for stage 1 and stage 2\n contexts, we may as well divide up the monolithic callback into its\n respective stage 1 and stage 2 parts.\n \n Signed-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 490325e0c123b2f1f393a679805c580ee69cd2f6\nAuthor: Robin Murphy \u003crobin.murphy@arm.com\u003e\nDate: Thu Aug 15 19:37:26 2019 +0100\n\n iommu/arm-smmu: Rework cb_base handling\n \n To keep register-access quirks manageable, we want to structure things\n to avoid needing too many individual overrides. It seems fairly clean to\n have a single interface which handles both global and context registers\n in terms of the architectural pages, so the first preparatory step is to\n rework cb_base into a page number rather than an absolute address.\n \n Signed-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 620565a76bb31657045a6d4cc37f9df0b4ab5f56\nAuthor: Robin Murphy \u003crobin.murphy@arm.com\u003e\nDate: Thu Aug 15 19:37:25 2019 +0100\n\n iommu/arm-smmu: Convert context bank registers to bitfields\n \n Finish the final part of the job, once again updating some names to\n match the current spec.\n \n Signed-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 5114e96cb27ed71559f2b674bb3c6ee14eafe9f9\nAuthor: Robin Murphy \u003crobin.murphy@arm.com\u003e\nDate: Thu Aug 15 19:37:24 2019 +0100\n\n iommu/arm-smmu: Convert GR1 registers to bitfields\n \n As for GR0, use the bitfield helpers to make GR1 usage a little cleaner,\n and use it as an opportunity to audit and tidy the definitions. This\n tweaks the handling of CBAR types to match what we did for S2CR a while\n back, and fixes a couple of names which didn\u0027t quite match the latest\n architecture spec (IHI0062D.c).\n \n Signed-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 0caf5f4e84fe0a870917984abf1f0eb29fa6e375\nAuthor: Robin Murphy \u003crobin.murphy@arm.com\u003e\nDate: Thu Aug 15 19:37:23 2019 +0100\n\n iommu/arm-smmu: Convert GR0 registers to bitfields\n \n FIELD_PREP remains a terrible name, but the overall simplification will\n make further work on this stuff that much more manageable. This also\n serves as an audit of the header, wherein we can impose a consistent\n grouping and ordering of the offset and field definitions\n \n Signed-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit a5b396ce5001a8b8c8594c7fa90b0df61d6b3272\nAuthor: Robin Murphy \u003crobin.murphy@arm.com\u003e\nDate: Thu Aug 15 19:37:22 2019 +0100\n\n iommu/qcom: Mask TLBI addresses correctly\n \n As with arm-smmu from whence this code was borrowed, the IOVAs passed in\n here happen to be at least page-aligned anyway, but still; oh dear.\n \n Signed-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 353b325047fdfd437dc2afaee89fc66de3657b7e\nAuthor: Robin Murphy \u003crobin.murphy@arm.com\u003e\nDate: Thu Aug 15 19:37:21 2019 +0100\n\n iommu/arm-smmu: Mask TLBI address correctly\n \n The less said about \"~12UL\" the better. Oh dear.\n \n We get away with it due to calling constraints that mean IOVAs are\n implicitly at least page-aligned to begin with, but still; oh dear.\n \n Signed-off-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit e1d7b66b815a542114169609c8f1b4e46856d7be\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 19 08:58:22 2019 +0100\n\n drm/i915: i915_active.retire() is optional\n \n Check that i915_active.retire() exists before calling.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190819075835.20065-6-chris@chris-wilson.co.uk\n\ncommit 74585fcb7b3ccb35135e2418dd66251022a916e5\nAuthor: Andrey Konovalov \u003candreyknvl@google.com\u003e\nDate: Mon Aug 19 15:14:42 2019 +0200\n\n selftests, arm64: fix uninitialized symbol in tags_test.c\n \n Fix tagged_ptr not being initialized when TBI is not enabled.\n \n Link: https://www.spinics.net/lists/linux-kselftest/msg09446.html\n Reported-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Signed-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 4843a543fad3bf8221cf14e5d5f32d15cee89e84\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Fri Aug 16 03:38:13 2019 -0300\n\n media: gspca: zero usb_buf on error\n \n If reg_r() fails, then gspca_dev-\u003eusb_buf was left uninitialized,\n and some drivers used the contents of that buffer in logic.\n \n This caused several syzbot errors:\n \n https://syzkaller.appspot.com/bug?extid\u003d397fd082ce5143e2f67d\n https://syzkaller.appspot.com/bug?extid\u003d1a35278dd0ebfb3a038a\n https://syzkaller.appspot.com/bug?extid\u003d06ddf1788cfd048c5e82\n \n I analyzed the gspca drivers and zeroed the buffer where needed.\n \n Reported-and-tested-by: syzbot+1a35278dd0ebfb3a038a@syzkaller.appspotmail.com\n Reported-and-tested-by: syzbot+397fd082ce5143e2f67d@syzkaller.appspotmail.com\n Reported-and-tested-by: syzbot+06ddf1788cfd048c5e82@syzkaller.appspotmail.com\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 2509d725e02adcf4d1375f80f2beef925b4a10aa\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Fri Aug 16 01:24:51 2019 -0300\n\n media: rockchip/rga: Update maintainership\n \n Update Jacob\u0027s address and add myself as co-maintainer.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit ed9927533a64ae55fa5adc91bf7383ee12f5710d\nAuthor: Trond Myklebust \u003ctrondmy@gmail.com\u003e\nDate: Sun Aug 18 14:18:59 2019 -0400\n\n nfsd: Fix the documentation for svcxdr_tmpalloc()\n \n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit b96811cd02466c6bb65c17639a96f2a766b7db9c\nAuthor: Trond Myklebust \u003ctrondmy@gmail.com\u003e\nDate: Sun Aug 18 14:18:58 2019 -0400\n\n nfsd: Fix up some unused variable warnings\n \n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit 7775ec57f4c71309109f9e98cf8854a2ca8c9df1\nAuthor: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\nDate: Sun Aug 18 14:18:57 2019 -0400\n\n nfsd: close cached files prior to a REMOVE or RENAME that would replace target\n \n It\u0027s not uncommon for some workloads to do a bunch of I/O to a file and\n delete it just afterward. If knfsd has a cached open file however, then\n the file may still be open when the dentry is unlinked. If the\n underlying filesystem is nfs, then that could trigger it to do a\n sillyrename.\n \n On a REMOVE or RENAME scan the nfsd_file cache for open files that\n correspond to the inode, and proactively unhash and put their\n references. This should prevent any delete-on-last-close activity from\n occurring, solely due to knfsd\u0027s open file cache.\n \n This must be done synchronously though so we use the variants that call\n flush_delayed_fput. There are deadlock possibilities if you call\n flush_delayed_fput while holding locks, however. In the case of\n nfsd_rename, we don\u0027t even do the lookups of the dentries to be renamed\n until we\u0027ve locked for rename.\n \n Once we\u0027ve figured out what the target dentry is for a rename, check to\n see whether there are cached open files associated with it. If there\n are, then unwind all of the locking, close them all, and then reattempt\n the rename.\n \n Signed-off-by: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit 501cb1849f865960501d19d54e6a5af306f9b6fd\nAuthor: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\nDate: Sun Aug 18 14:18:56 2019 -0400\n\n nfsd: rip out the raparms cache\n \n The raparms cache was set up in order to ensure that we carry readahead\n information forward from one RPC call to the next. In other words, it\n was set up because each RPC call was forced to open a struct file, then\n close it, causing the loss of readahead information that is normally\n cached in that struct file, and used to keep the page cache filled when\n a user calls read() multiple times on the same file descriptor.\n \n Now that we cache the struct file, and reuse it for all the I/O calls\n to a given file by a given user, we no longer have to keep a separate\n readahead cache.\n \n Signed-off-by: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit 6b556ca2872be223176c1a1e249606248ce0b201\nAuthor: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\nDate: Sun Aug 18 14:18:55 2019 -0400\n\n nfsd: have nfsd_test_lock use the nfsd_file cache\n \n Signed-off-by: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit 5c4583b2b78eef4eb33fca9a4598e72e08dd514b\nAuthor: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\nDate: Sun Aug 18 14:18:54 2019 -0400\n\n nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache\n \n Have nfs4_preprocess_stateid_op pass back a nfsd_file instead of a filp.\n Since we now presume that the struct file will be persistent in most\n cases, we can stop fiddling with the raparms in the read code. This\n also means that we don\u0027t really care about the rd_tmp_file field\n anymore.\n \n Signed-off-by: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit eb82dd393744107ebc365a53e7813c7c67cb203b\nAuthor: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\nDate: Sun Aug 18 14:18:53 2019 -0400\n\n nfsd: convert fi_deleg_file and ls_file fields to nfsd_file\n \n Have them keep an nfsd_file reference instead of a struct file.\n \n Signed-off-by: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit fd4f83fd7dfb1bce2f1af51fcbaf6575f4b9d189\nAuthor: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\nDate: Sun Aug 18 14:18:52 2019 -0400\n\n nfsd: convert nfs4_file-\u003efi_fds array to use nfsd_files\n \n Signed-off-by: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit 8b8900b729e4f31f12ac1127bde137c775c327e6\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Thu Aug 15 10:00:33 2019 -0300\n\n media: hdpvr: add terminating 0 at end of string\n \n dev-\u003eusbc_buf was passed as argument for %s, but it was not safeguarded\n by a terminating 0.\n \n This caused this syzbot issue:\n \n https://syzkaller.appspot.com/bug?extid\u003d79d18aac4bf1770dd050\n \n Reported-and-tested-by: syzbot+79d18aac4bf1770dd050@syzkaller.appspotmail.com\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 0d616f2a3fdbf1304db44d451d9f07008556923b\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Thu Aug 15 09:40:52 2019 -0300\n\n media: radio/si470x: kill urb on error\n \n In the probe() function radio-\u003eint_in_urb was not killed if an\n error occurred in the probe sequence. It was also missing in\n the disconnect.\n \n This caused this syzbot issue:\n \n https://syzkaller.appspot.com/bug?extid\u003d2d4fc2a0c45ad8da7e99\n \n Reported-and-tested-by: syzbot+2d4fc2a0c45ad8da7e99@syzkaller.appspotmail.com\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 3cec570b6d9c371bef1688a00ceb039aa4221fb7\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Aug 14 11:07:06 2019 -0300\n\n media: vidioc-queryctrl.rst: fix broken c:type references\n \n One typo in the function name, one missing : after :c:type.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit f4378baf07a200cbf7a3f0f4ba646aca6d3b11ac\nAuthor: Hugues Fruchet \u003chugues.fruchet@st.com\u003e\nDate: Wed Aug 14 10:48:53 2019 -0300\n\n media: stm32-dcmi: add support of several sub-devices\n \n Add support of several sub-devices within pipeline instead\n of a single one.\n This allows to support a CSI-2 camera sensor connected\n through a CSI-2 to parallel bridge.\n \n Signed-off-by: Hugues Fruchet \u003chugues.fruchet@st.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 34f8d704a32a32064ff5ae35802a6011107ea5c1\nAuthor: Hugues Fruchet \u003chugues.fruchet@st.com\u003e\nDate: Wed Aug 14 10:48:52 2019 -0300\n\n media: stm32-dcmi: add media controller support\n \n Add media controller support to dcmi in order\n to walk within remote subdevices pipeline.\n \n Signed-off-by: Hugues Fruchet \u003chugues.fruchet@st.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 585b18ede91de794c44660145f14a840702fc713\nAuthor: Hugues Fruchet \u003chugues.fruchet@st.com\u003e\nDate: Wed Aug 14 10:48:51 2019 -0300\n\n media: stm32-dcmi: trace the supported fourcc/mbus_code\n \n Add a trace of the set of supported fourcc/mbus_code which\n intersect between DCMI and source sub-device.\n \n Signed-off-by: Hugues Fruchet \u003chugues.fruchet@st.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 510b3188250961d9a4dc87f46ebcedf8a90c0c9e\nAuthor: Hugues Fruchet \u003chugues.fruchet@st.com\u003e\nDate: Wed Aug 14 10:48:50 2019 -0300\n\n media: stm32-dcmi: improve sensor subdev naming\n \n Rename \"subdev\" entity struct field to \"source\"\n to prepare for several subdev support.\n Move asd field on top of entity struct.\n \n Acked-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Hugues Fruchet \u003chugues.fruchet@st.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 6b338c720c139acd4a0688cea4cd5018743422be\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Aug 14 06:12:31 2019 -0300\n\n media: Fix various misspellings of disconnected\n \n Fix typos.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 5920afa3c85ff38642f652b6e3880e79392fcc89\nAuthor: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\nDate: Sun Aug 18 14:18:51 2019 -0400\n\n nfsd: hook nfsd_commit up to the nfsd_file cache\n \n Use cached filps if possible instead of opening a new one every time.\n \n Signed-off-by: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit 48cd7b51258c1a158293cefb97dda988080f5e13\nAuthor: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\nDate: Sun Aug 18 14:18:50 2019 -0400\n\n nfsd: hook up nfsd_read to the nfsd_file cache\n \n Signed-off-by: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit b493523926f9b466db3c440ac64beb93d8068cdf\nAuthor: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\nDate: Sun Aug 18 14:18:49 2019 -0400\n\n nfsd: hook up nfsd_write to the new nfsd_file cache\n \n Signed-off-by: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit 65294c1f2c5e72b15b76e16c8c8cfd9359fc9f6f\nAuthor: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\nDate: Sun Aug 18 14:18:48 2019 -0400\n\n nfsd: add a new struct file caching facility to nfsd\n \n Currently, NFSv2/3 reads and writes have to open a file, do the read or\n write and then close it again for each RPC. This is highly inefficient,\n especially when the underlying filesystem has a relatively slow open\n routine.\n \n This patch adds a new open file cache to knfsd. Rather than doing an\n open for each RPC, the read/write handlers can call into this cache to\n see if there is one already there for the correct filehandle and\n NFS_MAY_READ/WRITE flags.\n \n If there isn\u0027t an entry, then we create a new one and attempt to\n perform the open. If there is, then we wait until the entry is fully\n instantiated and return it if it is at the end of the wait. If it\u0027s\n not, then we attempt to take over construction.\n \n Since the main goal is to speed up NFSv2/3 I/O, we don\u0027t want to\n close these files on last put of these objects. We need to keep them\n around for a little while since we never know when the next READ/WRITE\n will come in.\n \n Cache entries have a hardcoded 1s timeout, and we have a recurring\n workqueue job that walks the cache and purges any entries that have\n expired.\n \n Signed-off-by: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\n Signed-off-by: Weston Andros Adamson \u003cdros@primarydata.com\u003e\n Signed-off-by: Richard Sharpe \u003crichard.sharpe@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit 7239a40ca8bfd88dc5d2f66a14882054fe8e3b92\nAuthor: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\nDate: Sun Aug 18 14:18:47 2019 -0400\n\n vfs: Export flush_delayed_fput for use by knfsd.\n \n Allow knfsd to flush the delayed fput list so that it can ensure the\n cached struct file is closed before it is unlinked.\n \n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit b72679ee89a0a0ecd26f7b6fcae96cdaababff94\nAuthor: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\nDate: Sun Aug 18 14:18:46 2019 -0400\n\n notify: export symbols for use by the knfsd file cache\n \n The knfsd file cache will need to detect when files are unlinked, so that\n it can close the associated cached files. Export a minimal set of notifier\n functions to allow it to do so.\n \n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit 18f6622ebbdea56a83f8e553c159ce2d62d3ad0c\nAuthor: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\nDate: Sun Aug 18 14:18:45 2019 -0400\n\n locks: create a new notifier chain for lease attempts\n \n With the new file caching infrastructure in nfsd, we can end up holding\n files open for an indefinite period of time, even when they are still\n idle. This may prevent the kernel from handing out leases on the file,\n which is something we don\u0027t want to block.\n \n Fix this by running a SRCU notifier call chain whenever on any\n lease attempt. nfsd can then purge the cache for that inode before\n returning.\n \n Since SRCU is only conditionally compiled in, we must only define the\n new chain if it\u0027s enabled, and users of the chain must ensure that\n SRCU is enabled.\n \n Signed-off-by: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit f69d6d8eef7807f8d937b81da24bebd2e926e4d2\nAuthor: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\nDate: Sun Aug 18 14:18:44 2019 -0400\n\n sunrpc: add a new cache_detail operation for when a cache is flushed\n \n When the exports table is changed, exportfs will usually write a new\n time to the \"flush\" file in the nfsd.export cache procfile. This tells\n the kernel to flush any entries that are older than that value.\n \n This gives us a mechanism to tell whether an unexport might have\n occurred. Add a new -\u003eflush cache_detail operation that is called after\n flushing the cache whenever someone writes to a \"flush\" file.\n \n Signed-off-by: Jeff Layton \u003cjeff.layton@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@primarydata.com\u003e\n Signed-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit 4fa4ef39e1c48b576a6a7f49c7410277e7ba09fa\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Sat Jul 27 10:31:29 2019 -0300\n\n media: Clarify how menus are hidden by SUBDRV_AUTOSELECT\n \n Some users have been having a hard time finding the hidden\n menus. A typically case are camera sensor drivers\n (e.g IMX219, OV5645, etc), which are common on embedded\n platforms and not really \"ancillary\" devices.\n \n The problem with MEDIA_SUBDRV_AUTOSELECT seems to be related\n to the fact that it uses the \"visible\" syntax to hide\n the menus.\n \n This is not obvious and it normally takes some time to\n figure out.\n \n To fix the problem, add a comment on each of hidden menus,\n which should clarify what option is causing menus to be hidden.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 4866073e6ddf03066c925d3237903d7f4ca68982\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Fri Aug 16 17:49:38 2019 -0400\n\n svcrdma: Use llist for managing cache of recv_ctxts\n \n Use a wait-free mechanism for managing the svc_rdma_recv_ctxts free\n list. Subsequently, sc_recv_lock can be eliminated.\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit d6dfe43ec6062beea5ba1172b957e74a13c95b86\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Fri Aug 16 17:48:36 2019 -0400\n\n svcrdma: Remove svc_rdma_wq\n \n Clean up: the system workqueue will work just as well.\n \n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit 988721db93b2f5e6477cb0ea0b64ba9bcfd67778\nAuthor: Junxiao Bi \u003cjunxiao.bi@oracle.com\u003e\nDate: Fri Aug 16 14:12:33 2019 -0700\n\n block: remove struct request_queue queue_head\n \n The dispatch list is not used any more, as the legacy block IO stack\n has been removed.\n \n Reviewed-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Reviewed-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Junxiao Bi \u003cjunxiao.bi@oracle.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 5a6b7ef69e3d9807aa9b38d9098772ae05916de5\nAuthor: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\nDate: Fri Aug 16 17:50:41 2019 -0700\n\n drm/i915/gen11: Allow usage of all GPIO pins\n \n Our pin mapping tables for ICP and MCC currently only list the standard\n GPIO pins used for various output ports. Even through ICP\u0027s standard\n pin usage only utilizes pins 1, 2, and 9-12, and MCC\u0027s standard pin\n usage only uses pins 1, 2, and 9, these platforms do still have GPIO\n registers to address pins in the range 1-3 and 9-14. OEM\u0027s may remap\n GPIO usage in non-standard ways (and provide the actual mapping via VBT\n settings), so we shouldn\u0027t exclude pins on these platforms just because\n they aren\u0027t part of the standard mappings.\n \n TGP\u0027s standard pin tables contains all the possible pins, so let\u0027s\n rename them to \"icp\" and use them for all PCH \u003e\u003d PCH_ICP. This will\n prevent intel_gmbus_is_valid_pin from rejecting non-standard pin usage\n that an OEM specifies via the VBT.\n \n Note that this will cause pin 9 to be labeled as \"tc1\" instead of \"dpc\"\n in debug messages on platforms with the MCC PCH, but that may actually\n help avoid confusion since the text strings will now be the same on all\n gen11+ platforms instead of being different on just EHL.\n \n v2: Drop now-unused MCC_DDC_BUS_DDI_* names.\n \n v3: We want to compare against INTEL_PCH_TYPE, not INTEL_PCH_ID.\n \n Bspec: 8417\n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Cc: Vivek Kasireddy \u003cvivek.kasireddy@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Signed-off-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190817005041.20651-1-matthew.d.roper@intel.com\n\ncommit 9301d28dccd05bbcea0e69c382886d6a8b8e6bc2\nAuthor: Kever Yang \u003ckever.yang@rock-chips.com\u003e\nDate: Thu Aug 15 16:12:52 2019 +0800\n\n arm64: dts: rockchip: fix vcc_host_5v regulator for usb3 host\n \n According to rock64 schemetic V2 and V3, the VCC_HOST_5V output is\n controlled by USB_20_HOST_DRV, which is the same as VCC_HOST1_5V.\n \n V1 hardware was never sold and only V2/V3 is with customers,\n so there is no need to keep a seaprate v1 version around.\n \n Signed-off-by: Kever Yang \u003ckever.yang@rock-chips.com\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit 70d6894d1456de95a3b8b3c80f6d0714fc04fcec\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 19 12:20:33 2019 +0100\n\n drm/i915: Serialize against vma moves\n \n Make sure that when submitting requests, we always serialize against\n potential vma moves and clflushes.\n \n Time for a i915_request_await_vma() interface!\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190819112033.30638-1-chris@chris-wilson.co.uk\n\ncommit 4d24376370fbfc87231d54434a683f2913abcce4\nAuthor: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\nDate: Wed Aug 14 10:53:30 2019 +0200\n\n gpu: ipu-v3: image-convert: only sample into the next tile if necessary\n \n The first pixel of the next tile is only sampled by the hardware if the\n fractional input position corresponding to the last written output pixel\n is not an integer position.\n \n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit fbefb84dd142c7d75232ae9d446ae36829c2dc97\nAuthor: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\nDate: Wed Aug 14 10:50:13 2019 +0200\n\n gpu: ipu-v3: image-convert: move tile burst alignment out of loop\n \n Burst aligned input and output width can be calculated once per column,\n instead of repeatedly for each tile in the column. The same goes for\n input and output height per row. Also don\u0027t round up the same values\n repeatedly.\n \n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit 5fb8b650cc1170f5e7a0f1a4ab3bb2042c007102\nAuthor: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\nDate: Tue Aug 13 14:39:50 2019 +0200\n\n gpu: ipu-v3: image-convert: bail on invalid tile sizes\n \n If we managed to create tiles sized 0x0 because of a bug in the seam\n calculation, return with an error message instead of letting the driver\n run into a division by zero later. Also check for tile sizes that are\n larger than supported by the hardware.\n \n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit de2564c70fcc1dfee89fff92eccba5a711a921cc\nAuthor: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\nDate: Tue Aug 13 13:42:58 2019 +0200\n\n gpu: ipu-v3: image-convert: fix image downsize coefficients and tiling calculation\n \n This patch effectively reverts commit 912bbf7e9ca4 (\"gpu: ipu-v3:\n image-convert: Fix image downsize coefficients\") and replaces it with a\n different solution based on the preceding patches.\n \n The previous fix tried to solve the problem of intermediate tile size\n between IC downsizing and main processing sections not being limited to\n 1024 pixels by downsizing the input image to a smaller intermediate size\n in the downsizing box filter. This causes unnecessary blurring,\n especially for scaling factors close to 1.\n \n Now that the seam position calculation makes sure that the 1024 pixel\n intermediate tile size limit is not exceeded, calculate the number of\n tiles from the maximum of intermediate size and output size and avoid\n unnecessary downsizing.\n \n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit 2e67a553e965b32a648c85fbe1c667fa440d6c92\nAuthor: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\nDate: Tue Aug 13 13:42:58 2019 +0200\n\n gpu: ipu-v3: image-convert: limit input seam position to hardware requirements\n \n Limit the input seam position to an interval that guarantees the tile\n size does not exceed 1024 pixels after the IC downsizing section and\n that space is left for the next tile.\n \n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit 82c3e948cc0fe7e01b39d437f7a58b197ebb422d\nAuthor: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\nDate: Tue Aug 13 14:30:25 2019 +0200\n\n gpu: ipu-v3: image-convert: fix output seam valid interval\n \n This fixes a failure to determine any seam if the output size is\n exactly 1024 multiplied by the number of tiles in a given direction.\n In that case an empty interval out_start \u003d\u003d out_end is being passed\n to find_best_seam, which looks for a seam out_start \u003c\u003d x \u003c out_end.\n \n Also reduce the interval for all but the left column / top row, to\n avoid returning position 0 as best fit.\n \n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit ca84b1b860407109b4b5f3802907f13bbef5d0fd\nAuthor: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\nDate: Tue Aug 13 18:49:26 2019 +0200\n\n gpu: ipu-v3: image-convert: move output seam valid interval calculation into find_best_seam\n \n This reduces code duplication and allows to apply the following\n modifications in a single place.\n \n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit cd42fa171d913627acb631ddf8b2a5d3a3d7e6ee\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Wed Aug 7 19:42:57 2019 +0300\n\n arm64: dts: imx8qxp: Add DSP DT node\n \n This includes DSP reserved memory, ADMA DSP device and DSP MU\n communication channels description.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 01c493144e52ad068098b5f497f897dc74c1610c\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Sun Aug 18 02:32:25 2019 -0400\n\n arm64: dts: imx8mn: Add cpu-freq support\n \n Add A53 OPP table, cpu regulator and speed grading node to\n support cpu-freq driver.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 3e44dd09736d98b4b14956e3cfe09260b4cceb50\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Sun Aug 18 02:32:20 2019 -0400\n\n arm64: dts: imx8mn-ddr4-evk: Add rohm,bd71847 PMIC support\n \n On i.MX8MN DDR4 EVK board, there is a rohm,bd71847 PMIC\n on i2c1 bus, enable it.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 089a6adeb0fbea7b8628986fdd7b8528554d7c71\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Sun Aug 18 02:32:19 2019 -0400\n\n arm64: dts: imx8mn-ddr4-evk: Add i2c1 support\n \n Enable i2c1 on i.MX8MN DDR4 EVK board.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 13cb15e0a6f5a51d14115525708f9deb30d975ad\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jun 19 10:21:44 2019 +0800\n\n arm64: dts: freescale: Add i.MX8MN DDR4 EVK board support\n \n This patch adds basic i.MM8MN DDR4 EVK board support.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit ee8696be6898fb7330e9a2b58f31303bba8f60ba\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Aug 14 22:57:30 2019 -0400\n\n arm64: dts: imx8mn: Add gpio-ranges property\n \n Add \"gpio-ranges\" property to establish connections between GPIOs\n and PINs on i.MX8MN pinctrl driver.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 6c3debcbae47a27a6fdf153e87d23e6410f6489b\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jun 19 10:21:43 2019 +0800\n\n arm64: dts: freescale: Add i.MX8MN dtsi support\n \n The i.MX8M Nano Media Applications Processor is a new SoC of the i.MX8M\n family, it is a 14nm FinFET product of the growing mScale family targeting\n the consumer market. It is built in Samsung 14LPP to achieve both high\n performance and low power consumption and relies on a powerful fully\n coherent core complex based on a quad core ARM Cortex-A53 cluster,\n Cortex-M7 low-power coprocessor and graphics accelerator.\n \n This patch adds the basic dtsi support for i.MX8MN.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 9b75651f41a05f1aee9006e2c6fa9f505e5287bc\nAuthor: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\nDate: Mon Jul 29 14:29:31 2019 +0200\n\n gpu: ipu-v3: image-convert: enable V4L2_PIX_FMT_BGRX32 and _RGBX32\n \n Enable image converter support for V4L2_PIX_FMT_BGRX32 and\n V4L2_PIX_FMT_RGBX32 pixel formats.\n \n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit a59957172b0c608a19aaadb77d65387f9a542c85\nAuthor: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\nDate: Mon Jul 29 14:22:10 2019 +0200\n\n gpu: ipu-v3: enable remaining 32-bit RGB V4L2 pixel formats\n \n Support is already implemented for the corresponding DRM formats,\n just hook up the remaining V4L2 pixel formats.\n \n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Reviewed-by: Marco Felsch \u003cm.felsch@pengutronix.de\u003e\n\ncommit a1406b72cbca4d9044cc1c26b00e780b36699d1f\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Fri Aug 16 06:13:03 2019 -0400\n\n arm64: dts: imx8mm: Enable cpu-idle driver\n \n Enable i.MX8MM cpu-idle using generic ARM cpu-idle driver, 2 states\n are supported, details as below:\n \n root@imx8mmevk:~# cat /sys/devices/system/cpu/cpu0/cpuidle/state0/name\n WFI\n root@imx8mmevk:~# cat /sys/devices/system/cpu/cpu0/cpuidle/state0/usage\n 3973\n root@imx8mmevk:~# cat /sys/devices/system/cpu/cpu0/cpuidle/state1/name\n cpu-pd-wait\n root@imx8mmevk:~# cat /sys/devices/system/cpu/cpu0/cpuidle/state1/usage\n 6647\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Acked-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 491d3a3fc113a6b2902a2920d9e0c4a43bdf4f0e\nAuthor: Ashish Kumar \u003cAshish.Kumar@nxp.com\u003e\nDate: Thu Aug 15 11:39:01 2019 +0800\n\n arm64: dts: ls1028a: Add esdhc node in dts\n \n This patch is to add esdhc node and enable SD UHS-I,\n eMMC HS200 for ls1028ardb/ls1028aqds board.\n \n Signed-off-by: Ashish Kumar \u003cAshish.Kumar@nxp.com\u003e\n Signed-off-by: Yangbo Lu \u003cyangbo.lu@nxp.com\u003e\n Signed-off-by: Yinbo Zhu \u003cyinbo.zhu@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 81f36887ea337b59c12d179eda8f7ea4398bac0e\nAuthor: Wen He \u003cwen.he_1@nxp.com\u003e\nDate: Mon Aug 12 18:02:24 2019 +0800\n\n arm64: dts: ls1028a: Add properties node for Display output pixel clock\n \n The LS1028A has a clock domain PXLCLK0 used for the Display output\n interface in the display core, independent of the system bus frequency,\n for flexible clock design. This display core has its own pixel clock.\n \n This patch enable the pixel clock provider on the LS1028A.\n \n Signed-off-by: Wen He \u003cwen.he_1@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 7cb220a75ff35fe5723428e9d83533bea0d3e3f6\nAuthor: Chuanhua Han \u003cchuanhua.han@nxp.com\u003e\nDate: Tue Aug 6 16:42:23 2019 +0800\n\n arm64: dts: lx2160a: Fix incorrect I2C clock divider\n \n Lx2160a platform, the i2c input clock is actually platform pll CLK / 16\n (this is the hardware connection), other clock divider can not get the\n correct i2c clock, resulting in the output of SCL pin clock is not\n accurate.\n \n Signed-off-by: Chuanhua Han \u003cchuanhua.han@nxp.com\u003e\n Acked-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit ced41bb1caefe4f20a03ce14fd252ba1875351da\nAuthor: Chuanhua Han \u003cchuanhua.han@nxp.com\u003e\nDate: Tue Aug 6 16:42:22 2019 +0800\n\n arm64: dts: ls1028a: Fix incorrect I2C clock divider\n \n Ls1028a platform, the i2c input clock is actually platform pll CLK / 4\n (this is the hardware connection), other clock divider can not get the\n correct i2c clock, resulting in the output of SCL pin clock is not\n accurate.\n \n Signed-off-by: Chuanhua Han \u003cchuanhua.han@nxp.com\u003e\n Acked-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 52d3406ec72d20f847f108d70b17b2ab844956f6\nAuthor: Chuanhua Han \u003cchuanhua.han@nxp.com\u003e\nDate: Tue Aug 6 16:42:21 2019 +0800\n\n arm64: dts: ls1012a: Fix incorrect I2C clock divider\n \n Ls1012a platform, the i2c input clock is actually platform pll CLK / 4\n (this is the hardware connection), other clock divider can not get the\n correct i2c clock, resulting in the output of SCL pin clock is not\n accurate.\n \n Signed-off-by: Chuanhua Han \u003cchuanhua.han@nxp.com\u003e\n Acked-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 86c457e3991a89e866792f05bc5ae43f476d80d2\nAuthor: Chuanhua Han \u003cchuanhua.han@nxp.com\u003e\nDate: Tue Aug 6 16:42:20 2019 +0800\n\n arm64: dts: ls1088a: Fix incorrect I2C clock divider\n \n Ls1088a platform, the i2c input clock is actually platform pll CLK / 8\n (this is the hardware connection), other clock divider can not get the\n correct i2c clock, resulting in the output of SCL pin clock is not\n accurate.\n \n Signed-off-by: Chuanhua Han \u003cchuanhua.han@nxp.com\u003e\n Acked-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit f64697bd0b9e3d050e8dcfe9605f2be5a936f98f\nAuthor: Song Hui \u003chui.song_1@nxp.com\u003e\nDate: Mon Aug 5 14:57:00 2019 +0800\n\n arm64: dts: ls1028a: fix gpio nodes\n \n Update the nodes to include little-endian\n property to be consistent with the hardware.\n \n Signed-off-by: Song Hui \u003chui.song_1@nxp.com\u003e\n Acked-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 571cebfe8e2b8678de0681633d3933c08082f7c1\nAuthor: Yuantian Tang \u003candy.tang@nxp.com\u003e\nDate: Tue Aug 6 13:35:07 2019 +0800\n\n arm64: dts: ls1028a: Add Thermal Monitor Unit node\n \n The Thermal Monitoring Unit (TMU) monitors and reports the\n temperature from 2 remote temperature measurement sites\n located on ls1028a chip.\n Add TMU dts node to enable this feature.\n \n Signed-off-by: Yuantian Tang \u003candy.tang@nxp.com\u003e\n Acked-by: Eduardo Valentin \u003cedubezval@gmail.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit df5eff6d2a44c8f01b0dcd61b0ed5715116e10b4\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Aug 19 13:22:36 2019 +0530\n\n udlfb: Make dlfb_ops constant\n \n Static structure dlfb_ops, of type fb_ops, is not used except to be\n copied into another variable. Hence make dlfb_ops constant to protect it\n from unintended modification.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Cc: Bernie Thompson \u003cbernie@plugable.com\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190819075236.1051-1-nishkadg.linux@gmail.com\n\ncommit 06b1f4b9f2a050e84f34f507bbd5f9983eeb0a06\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Wed Aug 7 11:13:12 2019 -0500\n\n video: fbdev/mmp/core: Use struct_size() in kzalloc()\n \n One of the more common cases of allocation size calculations is finding\n the size of a structure that has a zero-sized array at the end, along\n with memory for some number of elements for that array. For example:\n \n struct mmp_path {\n ...\n struct mmp_overlay overlays[0];\n };\n \n size \u003d sizeof(struct mmp_path) + count * sizeof(struct mmp_overlay);\n instance \u003d kzalloc(size, GFP_KERNEL)\n \n Instead of leaving these open-coded and prone to type mistakes, we can\n now use the new struct_size() helper:\n \n instance \u003d kzalloc(struct_size(instance, overlays, count), GFP_KERNEL)\n \n Notice that, in this case, variable size is not necessary, hence it\n is removed.\n \n This code was detected with the help of Coccinelle.\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807161312.GA26835@embeddedor\n\ncommit 2012f7762e8d800fbb3e33af409921bc8bade761\nAuthor: Souptick Joarder \u003cjrdr.linux@gmail.com\u003e\nDate: Wed Jul 31 01:03:20 2019 +0530\n\n video: fbdev: viafb: Remove dead code\n \n This is dead code since 3.15. If there is no plan to use\n it further, this can be removed forever.\n \n Signed-off-by: Souptick Joarder \u003cjrdr.linux@gmail.com\u003e\n Cc: Florian Tobias Schandinat \u003cFlorianSchandinat@gmx.de\u003e\n [b.zolnierkie: fix patch summary]\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/1564515200-5020-1-git-send-email-jrdr.linux@gmail.com\n\ncommit efbd44abaf742127a7bb2f9afcac7a526c52eccb\nAuthor: Souptick Joarder \u003cjrdr.linux@gmail.com\u003e\nDate: Wed Jul 31 00:44:13 2019 +0530\n\n video: fbdev: aty[128]fb: Remove dead code\n \n This is dead code since 3.15. If there is no plan to use it\n further, this can be removed forever.\n \n Signed-off-by: Souptick Joarder \u003cjrdr.linux@gmail.com\u003e\n Cc: Paul Mackerras \u003cpaulus@samba.org\u003e\n Cc: David S. Miller \u003cdavem@davemloft.net\u003e\n Cc: Mikulas Patocka \u003cmpatocka@redhat.com\u003e\n Cc: \"Ville Syrjälä\" \u003csyrjala@sci.fi\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n [b.zolnierkie: fix patch summary]\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/1564514053-4571-1-git-send-email-jrdr.linux@gmail.com\n\ncommit cc0c3e39e51b8d0c450e23b2dbfc655ac64ff683\nAuthor: Anders Roxell \u003canders.roxell@linaro.org\u003e\nDate: Tue Jul 30 17:25:30 2019 +0200\n\n video: fbdev: sh_mobile_lcdcfb: Mark expected switch fall-through\n \n Now that -Wimplicit-fallthrough is passed to GCC by default, the\n following warnings shows up:\n \n ../drivers/video/fbdev/sh_mobile_lcdcfb.c: In function ‘sh_mobile_lcdc_channel_fb_init’:\n ../drivers/video/fbdev/sh_mobile_lcdcfb.c:2086:22: warning: this statement may fall\n through [-Wimplicit-fallthrough\u003d]\n info-\u003efix.ypanstep \u003d 2;\n ~~~~~~~~~~~~~~~~~~~^~~\n ../drivers/video/fbdev/sh_mobile_lcdcfb.c:2087:2: note: here\n case V4L2_PIX_FMT_NV16:\n ^~~~\n ../drivers/video/fbdev/sh_mobile_lcdcfb.c: In function ‘sh_mobile_lcdc_overlay_fb_init’:\n ../drivers/video/fbdev/sh_mobile_lcdcfb.c:1596:22: warning: this statement may fall\n through [-Wimplicit-fallthrough\u003d]\n info-\u003efix.ypanstep \u003d 2;\n ~~~~~~~~~~~~~~~~~~~^~~\n ../drivers/video/fbdev/sh_mobile_lcdcfb.c:1597:2: note: here\n case V4L2_PIX_FMT_NV16:\n ^~~~\n \n Rework to address a warnings due to the enablement of\n -Wimplicit-fallthrough.\n \n Signed-off-by: Anders Roxell \u003canders.roxell@linaro.org\u003e\n Cc: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n [b.zolnierkie: fix patch summary]\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730152530.3055-1-anders.roxell@linaro.org\n\ncommit af70a48698cad800b2e7e71256fb373707b08959\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Wed Jul 24 21:19:00 2019 +0800\n\n video: fbdev: radeonfb: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Cc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n [b.zolnierkie: fix patch summary and intendation]\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190724131900.2039-1-hslester96@gmail.com\n\ncommit 70fc43c0998d52714c80a96d5e74f8b126e1f8b4\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Wed Jul 24 21:17:44 2019 +0800\n\n video: fbdev: sm712fb: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Cc: Sudip Mukherjee \u003csudipm.mukherjee@gmail.com\u003e\n Cc: Teddy Wang \u003cteddy.wang@siliconmotion.com\u003e\n [b.zolnierkie: fix patch summary]\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190724131744.1709-1-hslester96@gmail.com\n\ncommit dbb4a75b9c405dbd0a53698a6d21626145138193\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Jul 22 15:33:58 2019 -0500\n\n video: fbdev: pvr2fb: remove unnecessary comparison of unsigned integer with \u003c 0\n \n There is no need to compare *var-\u003exoffset* or *var-\u003eyoffset* with \u003c 0\n because such variables are of type unsigned, making it impossible to\n hold a negative value.\n \n Fix this by removing such comparisons.\n \n Addresses-Coverity-ID: 1451964 (\"Unsigned compared against 0\")\n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190722203358.GA29111@embeddedor\n\ncommit 51677dfcc17f88ed754143df670ff064eae67f84\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Sun Jul 21 15:19:18 2019 +0200\n\n efifb: BGRT: Improve efifb_bgrt_sanity_check\n \n For various reasons, at least with x86 EFI firmwares, the xoffset and\n yoffset in the BGRT info are not always reliable.\n \n Extensive testing has shown that when the info is correct, the\n BGRT image is always exactly centered horizontally (the yoffset variable\n is more variable and not always predictable).\n \n This commit simplifies / improves the bgrt_sanity_check to simply\n check that the BGRT image is exactly centered horizontally and skips\n (re)drawing it when it is not.\n \n This fixes the BGRT image sometimes being drawn in the wrong place.\n \n Cc: stable@vger.kernel.org\n Fixes: 88fe4ceb2447 (\"efifb: BGRT: Do not copy the boot graphics for non native resolutions\")\n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Cc: Peter Jones \u003cpjones@redhat.com\u003e,\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190721131918.10115-1-hdegoede@redhat.com\n\ncommit b6a32bbd8735def2d0d696ba59205d1874b7800f\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Fri Aug 16 18:09:23 2019 +0200\n\n genirq: Force interrupt threading on RT\n \n Switch force_irqthreads from a boot time modifiable variable to a compile\n time constant when CONFIG_PREEMPT_RT is enabled.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190816160923.12855-1-bigeasy@linutronix.de\n\ncommit 613cc5cd745c77da2b9090bcb29f82c05b853567\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Sun Aug 18 02:32:24 2019 -0400\n\n clk: imx8mn: Add necessary frequency support for ARM PLL table\n \n i.MX8MN supports CPU running at 1.5GHz/1.4GHz/1.2GHz, add missing\n frequency for ARM PLL table.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 313ccbad732dfcede65cbbb7444c00f20c08f766\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Sun Aug 18 02:32:23 2019 -0400\n\n clk: imx8mn: Add missing rate_count assignment for each PLL structure\n \n Add .rate_count assignment which is necessary for searching required\n PLL rate from the each PLL table.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 62a17029611af92ff447bfe2677b24438ebcbbad\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Jul 31 13:27:39 2019 +0200\n\n arm64: dts: renesas: r8a77980: Fix IPMMU-VC0 base address\n \n According to the R-Car Gen3 Hardware Manual Errata for Rev 1.50 of Feb\n 12, 2019, the base address of the IPMMU-VC0 block on R-Car V3H is\n 0xfe990000.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Ulrich Hecht \u003culi+renesas@fpond.eu\u003e\n\ncommit 8703ba77ec555b08c538beb728a4df1b72e0213e\nAuthor: Simon Horman \u003chorms+renesas@verge.net.au\u003e\nDate: Mon Jul 29 10:03:56 2019 +0200\n\n arm64: dts: renesas: ebisu, draak: Limit EtherAVB to 100Mbps\n \n * According to the R-Car Gen3 Hardware Manual Errata for Rev 1.00 of\n August 24, 2018, the TX clock internal delay mode isn\u0027t supported\n on R-Car E3 (r8a77990) and D3 (r8a77995).\n \n * TX clock internal delay mode is required for reliable 1Gbps communication\n using the KSZ9031RNX phy present on the Ebisu and Draak boards.\n \n Thus, the E3 based Ebisu and D3 based Draak boards can not reliably\n use 1Gbps and the speed should be limited to 100Mbps.\n \n Based on work by Kazuya Mizuguchi.\n \n Signed-off-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit e77ad88d0c6228af65d0a0d49b264c2fb249afcf\nAuthor: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\nDate: Mon Jul 15 10:32:49 2019 +0100\n\n arm64: dts: renesas: hihope-common: Add HDMI audio support\n \n This patch adds support for HDMI audio to the device tree\n common to the HiHope RZ/G2M and the HiHope RZ/G2N.\n \n Signed-off-by: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 24ce0bab260b59964629bb4d9570f0548b89e490\nAuthor: Suman Anna \u003cs-anna@ti.com\u003e\nDate: Fri Aug 16 17:58:37 2019 -0500\n\n iommu/omap: Use the correct type for SLAB_HWCACHE_ALIGN\n \n The macro SLAB_HWCACHE_ALIGN is of type slab_flags_t, but is currently\n assigned in the OMAP IOMMU driver using a unsigned long variable. This\n generates a sparse warning around the type check. Fix this by defining\n the variable flags using the correct type.\n \n Signed-off-by: Suman Anna \u003cs-anna@ti.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit e0e1df61524f6c26421f813889934f63a1f6c06e\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Fri Aug 9 16:14:32 2019 +0200\n\n soc: renesas: rcar-sysc: Eliminate local variable gov\n \n As of commit 980532a5dda319ee (\"soc: renesas: rcar-sysc: Use\n GENPD_FLAG_ALWAYS_ON\"), the local variable \"gov\" is assigned just once,\n so it can be eliminated.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\n\ncommit da51ceda8ab054bf3b4c5cf86a8deec7d7849a5c\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 15 11:43:54 2019 +0530\n\n soc: renesas: rcar-sysc: Add goto to of_node_put() before return\n \n The local variable np in function rcar_sysc_pd_init takes the return\n value of of_find_matching_node_and_match(), which gets a node but does\n not put it. If np is not put before the function returns, it may cause\n a memory leak.\n \n Hence, remove the return statement that does not immediately follow a\n putting of np. Replace it with a goto pointing to a pre-existing label\n that first puts np and then returns the required value.\n \n Issue found with Coccinelle.\n \n Fixes: afa6f53df6052968 (\"soc: renesas: rcar-sysc: Add support for fixing up power area tables\")\n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit a03fa77d85a736d3edef1a0b93100902b6db58c0\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Fri Aug 16 10:22:29 2019 +0200\n\n ARM: dts: r8a7779: Use SYSC \"always-on\" PM Domain for HSCIF\n \n Hook up HSCIF serial devices that are part of the CPG/MSTP Clock Domain\n to the SYSC \"always-on\" PM Domain, for a more consistent\n device-power-area description in DT.\n \n Cfr. commit 751e29bbb64ad091 (\"ARM: dts: r8a7779: Use SYSC \"always-on\"\n PM Domain\").\n \n Fixes: 055d15a88f66b096 (\"ARM: dts: r8a7779: Add HSCIF0/1 device nodes\")\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Ulrich Hecht \u003culi+renesas@fpond.eu\u003e\n\ncommit 876eb9bfa9d6945c2762af004246de8c04e6f8be\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Wed Aug 14 12:35:36 2019 -0700\n\n ARM: vf610-zii-cfu1: Add node for switch watchdog\n \n Add I2C child node for switch watchdog present on CFU1.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Signed-off-by: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Shawn Guo \u003cshawnguo@kernel.org\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: linux-kernel@vger.kernel.org\n Tested-by: Chris Healy \u003ccphealy@gmail.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 4a92f07816ba3020c690cc0f6e9627fec15b335e\nAuthor: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\nDate: Wed Aug 14 10:23:16 2019 +0200\n\n bus: imx-weim: use module_platform_driver()\n \n Switch from module_platform_driver_probe() to module_platform_driver().\n The former is not suitable for booting with device tree as the driver\n will be registered before the device and thus won\u0027t be probed again\n when the device is present.\n \n Signed-off-by: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit fc19a57dd4834a939b5eef574c61cd4f81af6cf8\nAuthor: Zhang Lixu \u003clixu.zhang@intel.com\u003e\nDate: Thu Aug 8 18:21:13 2019 +0800\n\n HID: intel-ish-hid: ipc: check the NO_D3 flag to distinguish resume paths\n \n The NO_D3 flag would be set if the ISH enter D0i3 in ish_suspend(),\n The resume paths can be distinguished by checking the NO_D3 flag.\n It\u0027s more reasonable than checking the FW status.\n \n Signed-off-by: Zhang Lixu \u003clixu.zhang@intel.com\u003e\n Acked-by: Srinivas Pandruvada \u003csrinivas.pandruvada@linux.intel.com\u003e\n Signed-off-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n\ncommit 2db8edaa88c1208f7346c65041120081e2434d2a\nAuthor: Zhang Lixu \u003clixu.zhang@intel.com\u003e\nDate: Thu Aug 8 18:21:12 2019 +0800\n\n HID: intel-ish-hid: ipc: make ish suspend paths clear\n \n For suspend-to-idle, send suspend message and set N0_D3 flag to put\n the ISH into D0i3 state.\n For suspend-to-mem, disable the DMA bit before ISH entering D3, and\n NO_D3 flag is cleared by default, then the ISH would enter D3.\n \n Signed-off-by: Zhang Lixu \u003clixu.zhang@intel.com\u003e\n Acked-by: Srinivas Pandruvada \u003csrinivas.pandruvada@linux.intel.com\u003e\n Signed-off-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n\ncommit c1ca58f6982bb815c27a4a75f0f430f87b624f66\nAuthor: Zhang Lixu \u003clixu.zhang@intel.com\u003e\nDate: Thu Aug 8 18:21:11 2019 +0800\n\n HID: intel-ish-hid: ipc: set NO_D3 flag only when needed\n \n Currently, the NO_D3 flag is set in ish_probe(), and cleared in\n ish_remove(). So even if the system goes into S3, ISH is still\n in D0i3 state. It makes more sense that put ISH into D3 as system\n goes into S3 and put ISH into D0i3 as system goes into suspend-to-idle.\n I remove the NO_D3 setting in ish_probe(), so that ISH can enter\n D3 state when system enters S3. Only set N0_D3 flag when system\n enters the suspend-to-idle or platform specified, and clear it\n when system resume.\n \n When the ISH enters D3, the FW will check the DMA bit status.\n If the DMA bit is set, the FW will reset automatically. So the\n DMA bit need be clear before putting ISH into D3 state.\n \n Signed-off-by: Zhang Lixu \u003clixu.zhang@intel.com\u003e\n Acked-by: Srinivas Pandruvada \u003csrinivas.pandruvada@linux.intel.com\u003e\n Signed-off-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n\ncommit 580ea2c2ae07d7c576fe289a6d22a6c3515f9f81\nAuthor: Marco Felsch \u003cm.felsch@pengutronix.de\u003e\nDate: Wed Aug 14 09:55:58 2019 +0200\n\n ARM: dts: imx6: drop gpmi-nand address and size cells\n \n Since commit fe2585e9c29a (\"doc: dt: mtd: support partitions in a\n special \u0027partitions\u0027 subnode\") and commit 5cfdedb7b9a0 (\"mtd: ofpart:\n move ofpart partitions to a dedicated dt node\") the partitioning should\n be within a partitions sub-node.\n \n Baseboard device trees following that scheme will get a dtc warning due\n to the predefined #address-cells and #size-cells properties:\n \n arch/arm/boot/dts/imx6qdl.dtsi:171.26-189.5: Warning (avoid_unnecessary_addr_size): /soc/gpmi-nand@112000: unnecessary #address-cells/#size-cells without \"ranges\" or child \"reg\" property\n \n I found no upstream imx6 baseboard using the old partitioning scheme, so\n we can drop the two properties to avoid such warnings.\n \n Signed-off-by: Marco Felsch \u003cm.felsch@pengutronix.de\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 291b983846d5fc455e128d285eb070d62b44d029\nAuthor: Marco Felsch \u003cm.felsch@pengutronix.de\u003e\nDate: Wed Aug 14 09:55:57 2019 +0200\n\n ARM: dts: imx6: replace simple-bus by simple-mfd for anatop\n \n Replace it to fix the following DTC warnings:\n \n arch/arm/boot/dts/imx6qdl.dtsi:702.19-715.7: Warning (simple_bus_reg): /soc/aips-bus@2000000/anatop@20c8000/regulator-1p1: missing or empty reg/ranges property\n arch/arm/boot/dts/imx6qdl.dtsi:717.19-730.7: Warning (simple_bus_reg): /soc/aips-bus@2000000/anatop@20c8000/regulator-3p0: missing or empty reg/ranges property\n arch/arm/boot/dts/imx6qdl.dtsi:732.19-745.7: Warning (simple_bus_reg): /soc/aips-bus@2000000/anatop@20c8000/regulator-2p5: missing or empty reg/ranges property\n arch/arm/boot/dts/imx6qdl.dtsi:747.32-762.7: Warning (simple_bus_reg): /soc/aips-bus@2000000/anatop@20c8000/regulator-vddcore: missing or empty reg/ranges property\n arch/arm/boot/dts/imx6qdl.dtsi:764.29-779.7: Warning (simple_bus_reg): /soc/aips-bus@2000000/anatop@20c8000/regulator-vddpu: missing or empty reg/ranges property\n arch/arm/boot/dts/imx6qdl.dtsi:781.31-796.7: Warning (simple_bus_reg): /soc/aips-bus@2000000/anatop@20c8000/regulator-vddsoc: missing or empty reg/ranges property\n \n Signed-off-by: Marco Felsch \u003cm.felsch@pengutronix.de\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit f8cade831018d80eaa3d9f64dac5a52d8715de55\nAuthor: Peng Fan \u003cpeng.fan@nxp.com\u003e\nDate: Wed Aug 14 09:53:12 2019 +0800\n\n clk: imx8mn: fix int pll clk gate\n \n To Frac pll, the gate shift is 13, however to Int PLL the gate shift\n is 11.\n \n Signed-off-by: Peng Fan \u003cpeng.fan@nxp.com\u003e\n Reviewed-by: Jacky Bai \u003cping.bai@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit be378b600791044cdc9820fe0ae13efa9e5499aa\nAuthor: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\nDate: Tue Aug 13 20:05:31 2019 +0300\n\n clk: imx8mn: Add GIC clock\n \n This is enabled by default but if it\u0027s not explicitly defined and marked\n as critical then its parent might get turned off.\n \n Signed-off-by: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit b3d08a4b1e46cd8146eaf6d6cb3779f756f0b3f3\nAuthor: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\nDate: Tue Aug 13 20:05:30 2019 +0300\n\n clk: imx8mn: Fix incorrect parents\n \n * Replace to audio_pll2_clk with audio_pll2_out\n * Replace sys3_pll2_out with sys_pll3_out\n * Replace sys1_pll_40m with sys_pll1_40m\n * qspi parent[2] is sys_pll2_333m not sys_pll1_800m\n \n Fixes: 96d6392b54db (\"clk: imx: Add support for i.MX8MN clock driver\")\n Signed-off-by: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 3125c9eb01e34ced138626d6d88bf94842a3fc2f\nAuthor: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\nDate: Tue Aug 13 20:05:29 2019 +0300\n\n clk: imx8mm: Fix incorrect parents\n \n * There is no video_pll2 on imx8mm, replace with dummy\n * Replace reference to sys_pll3_clk with sys_pll3_out\n * qspi parent[2] is sys_pll2_333m not sys_pll1_800m\n \n Fixes: ba5625c3e272 (\"clk: imx: Add clock driver support for imx8mm\")\n Signed-off-by: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 78f5666c18f33346f89633e517028f1caccfdec1\nAuthor: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\nDate: Tue Aug 13 20:05:28 2019 +0300\n\n clk: imx8mq: Fix sys3 pll references\n \n The \"sys3_pll2_out\" CLK was removed in refactoring so all references\n need to be updated to \"sys3_pll_out\"\n \n Fixes: e9dda4af685f (\"clk: imx: Refactor entire sccg pll clk\")\n Signed-off-by: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit a1e37b0259e77b1db2ff2636059cd513971ca7fa\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 19 08:58:21 2019 +0100\n\n drm/i915: Only emit the \u0027send bug report\u0027 once for a GPU hang\n \n Use a locked xchg to ensure that the global log message giving\n instructions on how to send a bug report is emitted precisely once.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190819075835.20065-5-chris@chris-wilson.co.uk\n\ncommit 49ff824a028ac602bb89defd6e43f91c7cf1c84a\nAuthor: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\nDate: Mon Jul 29 08:16:14 2019 +0200\n\n m68k: defconfig: Update defconfigs for v5.3-rc2\n \n Actual changes:\n -CONFIG_CHR_DEV_OSST\u003dm\n +CONFIG_CRYPTO_AES\u003dy\n +CONFIG_CRYPTO_ARC4\u003dm\n +CONFIG_CRYPTO_XXHASH\u003dm\n +CONFIG_NF_CONNTRACK_BRIDGE\u003dm\n +CONFIG_NF_TABLES_BRIDGE\u003dm\n -CONFIG_NF_TABLES_BRIDGE\u003dy\n +CONFIG_NFT_BRIDGE_META\u003dm\n +CONFIG_NFT_SYNPROXY\u003dm\n +CONFIG_REED_SOLOMON_TEST\u003dm\n +CONFIG_TEST_BLACKHOLE_DEV\u003dm\n +CONFIG_TEST_MEMINIT\u003dm\n -# CONFIG_VALIDATE_FS_PARSER is not set\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n\ncommit 053b514295694f3336e97f56d5f41c0d4972c109\nAuthor: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\nDate: Mon Aug 12 08:23:48 2019 +0200\n\n m68k: atari: Rename shifter to shifter_st to avoid conflict\n \n When test-compiling the BCM2835 pin control driver on m68k:\n \n In file included from arch/m68k/include/asm/io_mm.h:32:0,\n from arch/m68k/include/asm/io.h:8,\n from include/linux/io.h:13,\n from include/linux/irq.h:20,\n from include/linux/gpio/driver.h:7,\n from drivers/pinctrl/bcm/pinctrl-bcm2835.c:17:\n drivers/pinctrl/bcm/pinctrl-bcm2835.c: In function \u0027bcm2711_pull_config_set\u0027:\n arch/m68k/include/asm/atarihw.h:190:22: error: expected identifier or \u0027(\u0027 before \u0027volatile\u0027\n # define shifter ((*(volatile struct SHIFTER *)SHF_BAS))\n \n \"shifter\" is a too generic name for a global definition.\n \n As the corresponding definition for Atari TT is already called\n \"shifter_tt\", fix this by renaming the definition for Atari ST to\n \"shifter_st\".\n \n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Suggested-by: Michael Schmitz \u003cschmitzmic@gmail.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 94c04390225bcd8283103fd0c04be20cc30cc979\nAuthor: Finn Thain \u003cfthain@telegraphics.com.au\u003e\nDate: Fri Aug 2 10:10:25 2019 +1000\n\n m68k: Prevent some compiler warnings in Coldfire builds\n \n Since commit d3b41b6bb49e (\"m68k: Dispatch nvram_ops calls to Atari or\n Mac functions\"), Coldfire builds generate compiler warnings due to the\n unconditional inclusion of asm/atarihw.h and asm/macintosh.h.\n \n The inclusion of asm/atarihw.h causes warnings like this:\n \n In file included from ./arch/m68k/include/asm/atarihw.h:25:0,\n from arch/m68k/kernel/setup_mm.c:41,\n from arch/m68k/kernel/setup.c:3:\n ./arch/m68k/include/asm/raw_io.h:39:0: warning: \"__raw_readb\" redefined\n #define __raw_readb in_8\n \n In file included from ./arch/m68k/include/asm/io.h:6:0,\n from arch/m68k/kernel/setup_mm.c:36,\n from arch/m68k/kernel/setup.c:3:\n ./arch/m68k/include/asm/io_no.h:16:0: note: this is the location of the previous definition\n #define __raw_readb(addr) \\\n ...\n \n This issue is resolved by dropping the asm/raw_io.h include. It turns out\n that asm/io_mm.h already includes that header file.\n \n Moving the relevant macro definitions helps to clarify this dependency\n and make it safe to include asm/atarihw.h.\n \n The other warnings look like this:\n \n In file included from arch/m68k/kernel/setup_mm.c:48:0,\n from arch/m68k/kernel/setup.c:3:\n ./arch/m68k/include/asm/macintosh.h:19:35: warning: \u0027struct irq_data\u0027 declared inside parameter list will not be visible outside of this definition or declaration\n extern void mac_irq_enable(struct irq_data *data);\n ^~~~~~~~\n ...\n \n This issue is resolved by adding the missing linux/irq.h include.\n \n Signed-off-by: Finn Thain \u003cfthain@telegraphics.com.au\u003e\n Acked-by: Greg Ungerer \u003cgerg@linux-m68k.org\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n\ncommit aee6bff1c325d7a8b317a0e6d3441799ff8ac8ce\nAuthor: Finn Thain \u003cfthain@telegraphics.com.au\u003e\nDate: Mon Jul 8 19:31:22 2019 +1000\n\n m68k: mac: Revisit floppy disc controller base addresses\n \n Rename floppy_type macros to make them more consistent with the scsi_type\n macros, which are named after classes of models with similar memory maps.\n \n The MAC_FLOPPY_OLD symbol is introduced to change the relevant base\n address from 0x50F00000 to 0x50000000 (consistent with MAC_SCSI_OLD).\n \n The documentation for LC-class machines has the IO devices at offsets\n from $50F00000. Use these addresses for MAC_FLOPPY_LC (consistent with\n MAC_SCSI_LC) because they may not be aliased elsewhere in the memory map.\n \n Add comments with controller type information from \u0027Designing Cards and\n Drivers for the Macintosh Family\u0027, relevant Developer Notes and\n http://mess.redump.net/mess/driver_info/mac_technical_notes\n \n Adopt phys_addr_t to avoid type casts.\n \n Signed-off-by: Finn Thain \u003cfthain@telegraphics.com.au\u003e\n Tested-by: Stan Johnson \u003cuserm57@yahoo.com\u003e\n Acked-by: Laurent Vivier \u003clvivier@redhat.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n\ncommit b2ffaa0c7fca832f41f477f30fbe8e64304a2901\nAuthor: Philippe Schenker \u003cphilippe.schenker@toradex.com\u003e\nDate: Mon Aug 12 14:21:28 2019 +0000\n\n ARM: dts: imx6qdl-colibri: add phy to fec\n \n Add the phy-node and mdio bus to the fec-node, represented as is on\n hardware.\n This commit includes micrel,led-mode that is set to the default\n value, prepared for someone who wants to change this.\n \n Signed-off-by: Philippe Schenker \u003cphilippe.schenker@toradex.com\u003e\n Acked-by: Marcel Ziswiler \u003cmarcel.ziswiler@toradex.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 4e9ccbd6748259a755a59c031c3571e37ae61da6\nAuthor: Oleksandr Suvorov \u003coleksandr.suvorov@toradex.com\u003e\nDate: Mon Aug 12 14:21:21 2019 +0000\n\n ARM: dts: imx7-colibri: add recovery for I2C for iMX7\n \n - add recovery mode for applicable i2c buses for\n Colibri iMX7 module.\n \n Signed-off-by: Oleksandr Suvorov \u003coleksandr.suvorov@toradex.com\u003e\n Signed-off-by: Philippe Schenker \u003cphilippe.schenker@toradex.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 02200d29c32427d6dba651ca638a300978c11d95\nAuthor: Philippe Schenker \u003cphilippe.schenker@toradex.com\u003e\nDate: Mon Aug 12 14:21:19 2019 +0000\n\n ARM: dts: imx7-colibri: Add sleep pinctrl to ethernet\n \n Add sleep pinmux to the fec so it can properly sleep.\n \n Signed-off-by: Philippe Schenker \u003cphilippe.schenker@toradex.com\u003e\n Acked-by: Marcel Ziswiler \u003cmarcel.ziswiler@toradex.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 03d655ff251c8296cd887055e470796eb1e47e47\nAuthor: Philippe Schenker \u003cphilippe.schenker@toradex.com\u003e\nDate: Mon Aug 12 14:21:18 2019 +0000\n\n ARM: dts: imx7-colibri: prepare module device tree for FlexCAN\n \n Prepare FlexCAN use on SODIMM 55/63 178/188. Those SODIMM pins are\n compatible for CAN bus use with several modules from the Colibri\n family.\n Add Better drivestrength and also add flexcan2.\n \n Signed-off-by: Philippe Schenker \u003cphilippe.schenker@toradex.com\u003e\n Acked-by: Marcel Ziswiler \u003cmarcel.ziswiler@toradex.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit a95fbda08ee20cd063ce5826e0df95a2c22ea8c5\nAuthor: Stefan Agner \u003cstefan.agner@toradex.com\u003e\nDate: Mon Aug 12 14:21:17 2019 +0000\n\n ARM: dts: imx7-colibri: disable HS400\n \n Force HS200 by masking bit 63 of the SDHCI capability register.\n The i.MX ESDHC driver uses SDHCI_QUIRK2_CAPS_BIT63_FOR_HS400. With\n that the stack checks bit 63 to descide whether HS400 is available.\n Using sdhci-caps-mask allows to mask bit 63. The stack then selects\n HS200 as operating mode.\n \n This prevents rare communication errors with minimal effect on\n performance:\n sdhci-esdhc-imx 30b60000.usdhc: warning! HS400 strobe DLL\n status REF not lock!\n \n Signed-off-by: Stefan Agner \u003cstefan.agner@toradex.com\u003e\n Signed-off-by: Philippe Schenker \u003cphilippe.schenker@toradex.com\u003e\n Reviewed-by: Oleksandr Suvorov \u003coleksandr.suvorov@toradex.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 4ff96fb52c6964ad42e0a878be8f86a2e8052ddd\nAuthor: Miroslav Benes \u003cmbenes@suse.cz\u003e\nDate: Fri Jul 19 14:28:39 2019 +0200\n\n livepatch: Nullify obj-\u003emod in klp_module_coming()\u0027s error path\n \n klp_module_coming() is called for every module appearing in the system.\n It sets obj-\u003emod to a patched module for klp_object obj. Unfortunately\n it leaves it set even if an error happens later in the function and the\n patched module is not allowed to be loaded.\n \n klp_is_object_loaded() uses obj-\u003emod variable and could currently give a\n wrong return value. The bug is probably harmless as of now.\n \n Signed-off-by: Miroslav Benes \u003cmbenes@suse.cz\u003e\n Reviewed-by: Petr Mladek \u003cpmladek@suse.com\u003e\n Acked-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Signed-off-by: Petr Mladek \u003cpmladek@suse.com\u003e\n\ncommit 037ab580f07568425b991ad0e97674e79bbeea5c\nAuthor: Marcel Ziswiler \u003cmarcel.ziswiler@toradex.com\u003e\nDate: Mon Aug 12 14:21:15 2019 +0000\n\n ARM: dts: imx7-colibri: make sure module supplies are always on\n \n Prevent regulators from being switched off.\n \n Signed-off-by: Marcel Ziswiler \u003cmarcel.ziswiler@toradex.com\u003e\n Signed-off-by: Philippe Schenker \u003cphilippe.schenker@toradex.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit ee050dc83bc326ad5ef8ee93bca344819371e7a5\nAuthor: Miles Chen \u003cmiles.chen@mediatek.com\u003e\nDate: Thu Aug 15 19:32:46 2019 +0800\n\n lib/stackdepot: Fix outdated comments\n \n Replace \"depot_save_stack\" with \"stack_depot_save\" in code comments because\n depot_save_stack() was replaced in commit c0cfc337264c (\"lib/stackdepot:\n Provide functions which operate on plain storage arrays\") and removed in\n commit 56d8f079c51a (\"lib/stackdepot: Remove obsolete functions\")\n \n Signed-off-by: Miles Chen \u003cmiles.chen@mediatek.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190815113246.18478-1-miles.chen@mediatek.com\n\ncommit 6cb0880f08229360c6c57416de075aa96930be78\nAuthor: Chris Chiu \u003cchiu@endlessm.com\u003e\nDate: Fri Aug 16 17:38:38 2019 +0800\n\n pinctrl: intel: remap the pin number to gpio offset for irq enabled pin\n \n On Asus X571GT, GPIO 297 is configured as an interrupt and serves\n for the touchpad. The touchpad will report input events much less\n than expected after S3 suspend/resume, which results in extremely\n slow cursor movement. However, the number of interrupts observed\n from /proc/interrupts increases much more than expected even no\n touching touchpad.\n \n This is due to the value of PADCFG0 of PIN 225 for the interrupt\n has been changed from 0x80800102 to 0x80100102. The GPIROUTIOXAPIC\n is toggled on which results in the spurious interrupts. The PADCFG0\n of PIN 225 is expected to be saved during suspend, but the 297 is\n saved instead because the gpiochip_line_is_irq() expect the GPIO\n offset but what\u0027s really passed to it is PIN number. In this case,\n the /sys/kernel/debug/pinctrl/INT3450:00/gpio-ranges shows\n \n 288: INT3450:00 GPIOS [436 - 459] PINS [216 - 239]\n \n So gpiochip_line_is_irq() returns true for GPIO offset 297, the\n suspend routine spuriously saves the content for PIN 297 which\n we expect to save for PIN 225.\n \n This commit maps the PIN number to GPIO offset first in the\n intel_pinctrl_should_save() to make sure the values for the\n specific PINs can be correctly saved and then restored.\n \n Fixes: c538b9436751 (\"pinctrl: intel: Only restore pins that are used by the driver\")\n Signed-off-by: Chris Chiu \u003cchiu@endlessm.com\u003e\n Acked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit d67739268cf0ee928cdc5e8224090c59efacd653\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 19 08:58:19 2019 +0100\n\n drm/i915/gt: Mark up the nested engine-pm timeline lock as irqsafe\n \n We use a fake timeline-\u003emutex lock to reassure lockdep that the timeline\n is always locked when emitting requests. However, the use inside\n __engine_park() may be inside hardirq and so lockdep now complains about\n the mixed irq-state of the nested locked. Disable irqs around the\n lockdep tracking to keep it happy.\n \n Fixes: 6c69a45445af (\"drm/i915/gt: Mark context-\u003eactive_count as protected by timeline-\u003emutex\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190819075835.20065-3-chris@chris-wilson.co.uk\n\ncommit 608b038aec9f5ffa80683e0574d772dbc84c3547\nAuthor: André Draszik \u003cgit@andred.net\u003e\nDate: Fri Aug 9 12:29:46 2019 +0100\n\n ARM: dts: imx7d: cl-som-imx7: add compatible for phy\n \n While not strictly needed as \"ethernet-phy-ieee802.3-c22\"\n is assumed by default if not given explicitly, having\n the compatible string here makes it more clear what\n this is and which driver handles this - an Ethernet\n phy attached to mdio, handled by of_mdio.c\n \n Signed-off-by: André Draszik \u003cgit@andred.net\u003e\n CC: Ilya Ledvich \u003cilya@compulab.co.il\u003e\n CC: Igor Grinberg \u003cgrinberg@compulab.co.il\u003e\n CC: Rob Herring \u003crobh+dt@kernel.org\u003e\n CC: Mark Rutland \u003cmark.rutland@arm.com\u003e\n CC: Shawn Guo \u003cshawnguo@kernel.org\u003e\n CC: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\n CC: Pengutronix Kernel Team \u003ckernel@pengutronix.de\u003e\n CC: Fabio Estevam \u003cfestevam@gmail.com\u003e\n CC: NXP Linux Team \u003clinux-imx@nxp.com\u003e\n CC: devicetree@vger.kernel.org\n CC: linux-arm-kernel@lists.infradead.org\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 342061c53a049569fc7f56d237753c26b4b2166d\nAuthor: Borislav Petkov \u003cbp@suse.de\u003e\nDate: Mon Aug 19 09:01:40 2019 +0200\n\n x86/msr-index: Move AMD MSRs where they belong\n \n ... sort them in and fixup comment, while at it.\n \n No functional changes.\n \n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: http://lkml.kernel.org/r/20190819070140.23708-1-bp@alien8.de\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 6a736ebf314a0b0813bae248ad3daa9154463798\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 19 08:58:18 2019 +0100\n\n drm/i915: Always wrap the ring offset before resetting\n \n We were passing in an unwrapped offset into intel_ring_reset() on\n unpinning. Sooner or later that had to land on ring-\u003esize.\n \n \u003c3\u003e [314.872147] intel_ring_reset:1237 GEM_BUG_ON(!intel_ring_offset_valid(ring, tail))\n \u003c4\u003e [314.872272] ------------[ cut here ]------------\n \u003c2\u003e [314.872276] kernel BUG at drivers/gpu/drm/i915/gt/intel_ringbuffer.c:1237!\n \u003c4\u003e [314.872320] invalid opcode: 0000 [#1] PREEMPT SMP PTI\n \u003c4\u003e [314.872331] CPU: 1 PID: 3466 Comm: i915_selftest Tainted: G U 5.3.0-rc4-CI-Patchwork_14061+ #1\n \u003c4\u003e [314.872346] Hardware name: Hewlett-Packard HP Compaq 8000 Elite CMT PC/3647h, BIOS 786G7 v01.02 10/22/2009\n \u003c4\u003e [314.872477] RIP: 0010:intel_ring_reset+0x51/0x70 [i915]\n \u003c4\u003e [314.872487] Code: 9e db 51 e0 48 8b 35 b6 c7 22 00 49 c7 c0 f8 d9 d6 a0 b9 d5 04 00 00 48 c7 c2 70 5b d4 a0 48 c7 c7 6c fc c0 a0 e8 cf be 58 e0 \u003c0f\u003e 0b 89 77 20 89 77 1c 89 77 24 e9 4f ed ff ff 0f 1f 44 00 00 66\n \u003c4\u003e [314.872512] RSP: 0018:ffffc9000034fa98 EFLAGS: 00010282\n \u003c4\u003e [314.872523] RAX: 0000000000000010 RBX: ffff8881019412c8 RCX: 0000000000000000\n \u003c4\u003e [314.872534] RDX: 0000000000000001 RSI: 0000000000000008 RDI: 0000000000000f20\n \u003c4\u003e [314.872545] RBP: ffff888104e0f740 R08: 0000000000000000 R09: 0000000000000f20\n \u003c4\u003e [314.872557] R10: 0000000000000000 R11: ffff888117094518 R12: ffffffffa0d3d2c0\n \u003c4\u003e [314.872569] R13: ffffffffa0e2a250 R14: ffffffffa0e2a1e0 R15: ffffc9000034fe88\n \u003c4\u003e [314.872581] FS: 00007fe6d49f6e40(0000) GS:ffff888117a80000(0000) knlGS:0000000000000000\n \u003c4\u003e [314.872595] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n \u003c4\u003e [314.872605] CR2: 000055e3283e9cc8 CR3: 0000000108842000 CR4: 00000000000406e0\n \u003c4\u003e [314.872616] Call Trace:\n \u003c4\u003e [314.872701] intel_ring_unpin+0x1a/0x220 [i915]\n \u003c4\u003e [314.872787] ring_destroy+0x48/0xc0 [i915]\n \u003c4\u003e [314.872870] intel_engines_cleanup+0x24/0x40 [i915]\n \u003c4\u003e [314.872964] i915_gem_driver_release+0x1b/0xf0 [i915]\n \u003c4\u003e [314.872984] i915_driver_release+0x1c/0x80 [i915]\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190819075835.20065-2-chris@chris-wilson.co.uk\n\ncommit 9846a4524ac90b63496580b7ad50674b40d92a8f\nAuthor: André Draszik \u003cgit@andred.net\u003e\nDate: Fri Aug 9 04:12:27 2019 +0100\n\n ARM: dts: imx7d: cl-som-imx7: make ethernet work again\n \n Recent changes to the atheros at803x driver caused\n ethernet to stop working on this board.\n In particular commit 6d4cd041f0af\n (\"net: phy: at803x: disable delay only for RGMII mode\")\n and commit cd28d1d6e52e\n (\"net: phy: at803x: Disable phy delay for RGMII mode\")\n fix the AR8031 driver to configure the phy\u0027s (RX/TX)\n delays as per the \u0027phy-mode\u0027 in the device tree.\n \n This now prevents ethernet from working on this board.\n \n It used to work before those commits, because the\n AR8031 comes out of reset with RX delay enabled, and\n the at803x driver didn\u0027t touch the delay configuration\n at all when \"rgmii\" mode was selected, and because\n arch/arm/mach-imx/mach-imx7d.c:ar8031_phy_fixup()\n unconditionally enables TX delay.\n \n Since above commits ar8031_phy_fixup() also has no\n effect anymore, and the end-result is that all delays\n are disabled in the phy, no ethernet.\n \n Update the device tree to restore functionality.\n \n Signed-off-by: André Draszik \u003cgit@andred.net\u003e\n CC: Ilya Ledvich \u003cilya@compulab.co.il\u003e\n CC: Igor Grinberg \u003cgrinberg@compulab.co.il\u003e\n CC: Rob Herring \u003crobh+dt@kernel.org\u003e\n CC: Mark Rutland \u003cmark.rutland@arm.com\u003e\n CC: Shawn Guo \u003cshawnguo@kernel.org\u003e\n CC: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\n CC: Pengutronix Kernel Team \u003ckernel@pengutronix.de\u003e\n CC: Fabio Estevam \u003cfestevam@gmail.com\u003e\n CC: NXP Linux Team \u003clinux-imx@nxp.com\u003e\n CC: devicetree@vger.kernel.org\n CC: linux-arm-kernel@lists.infradead.org\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit dce84022376df581ea9975ea32011342ec9220a5\nAuthor: Sébastien Szymanski \u003csebastien.szymanski@armadeus.com\u003e\nDate: Wed Jul 31 18:32:57 2019 +0200\n\n ARM: dts: imx6ul: Add csi node\n \n Add csi node for i.MX6UL SoC.\n \n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Sébastien Szymanski \u003csebastien.szymanski@armadeus.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 0cc4e8f1a86cf4d57d6efd7ee82636534834d870\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Wed Jul 17 13:26:39 2019 -0600\n\n thermal: intel: int340x_thermal: Remove unnecessary acpi_has_method() uses\n \n acpi_evaluate_object() will already return in error if the method does not\n exist. Checking if the method is absent before the acpi_evaluate_object()\n call is not needed. Remove acpi_has_method() calls to avoid additional\n work.\n \n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Reviewed-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Signed-off-by: Zhang Rui \u003crui.zhang@intel.com\u003e\n\ncommit beb22d6d899387a14685c10af17afdf687d9e669\nAuthor: Srinivas Pandruvada \u003csrinivas.pandruvada@linux.intel.com\u003e\nDate: Thu Aug 1 15:39:37 2019 -0700\n\n thermal: int340x: processor_thermal: Add Ice Lake support\n \n Add new PCI id for Ice lake processor thermal device. Also enabled\n the RAPL mmio interface. The MMIO offsets match Skylake.\n \n Signed-off-by: Srinivas Pandruvada \u003csrinivas.pandruvada@linux.intel.com\u003e\n Signed-off-by: Zhang Rui \u003crui.zhang@intel.com\u003e\n\ncommit 062d79a1ce88243f5338b48f8c7eb8e63cd7c0b2\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Wed Jul 24 20:23:37 2019 +0800\n\n thermal: processor_thermal_device: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Zhang Rui \u003crui.zhang@intel.com\u003e\n\ncommit fdf4f2fb8e8990c131b2b1a5a9c03681bb16e87a\nAuthor: Srinivas Pandruvada \u003csrinivas.pandruvada@linux.intel.com\u003e\nDate: Mon Jul 22 18:03:02 2019 -0700\n\n drivers: thermal: processor_thermal_device: Export sysfs interface for TCC offset\n \n This change exports an interface to read tcc offset and allow writing if\n the platform is not locked.\n \n Refer to Intel SDM for details on the MSR: MSR_TEMPERATURE_TARGET.\n Here TCC Activation Offset (R/W) bits allow temperature offset in degrees\n in relation to TjMAX.\n \n This change will be useful for improving performance from user space for\n some platforms, if the current offset is not optimal.\n \n Signed-off-by: Srinivas Pandruvada \u003csrinivas.pandruvada@linux.intel.com\u003e\n Tested-by: Benjamin Berg \u003cbberg@redhat.com\u003e\n Signed-off-by: Zhang Rui \u003crui.zhang@intel.com\u003e\n\ncommit 5bc90eea02a87fd26c939fa7f41d351c983c55b2\nAuthor: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\nDate: Fri Jul 12 02:01:58 2019 +0530\n\n thermal: Add some error messages\n \n When registering a thermal zone device, we currently return -EINVAL in\n four cases. This makes it a little hard to debug the real cause of the\n failure.\n \n Print some error messages to make it easier for developer to figure out\n what happened.\n \n Signed-off-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Signed-off-by: Zhang Rui \u003crui.zhang@intel.com\u003e\n\ncommit 83aa81e64af8dc6ed58efeebb09a5e4e48b2ec34\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Wed Jul 10 13:14:52 2019 +0300\n\n thermal: Fix use-after-free when unregistering thermal zone device\n \n thermal_zone_device_unregister() cancels the delayed work that polls the\n thermal zone, but it does not wait for it to finish. This is racy with\n respect to the freeing of the thermal zone device, which can result in a\n use-after-free [1].\n \n Fix this by waiting for the delayed work to finish before freeing the\n thermal zone device. Note that thermal_zone_device_set_polling() is\n never invoked from an atomic context, so it is safe to call\n cancel_delayed_work_sync() that can block.\n \n [1]\n [ +0.002221] \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n [ +0.000064] BUG: KASAN: use-after-free in __mutex_lock+0x1076/0x11c0\n [ +0.000016] Read of size 8 at addr ffff8881e48e0450 by task kworker/1:0/17\n \n [ +0.000023] CPU: 1 PID: 17 Comm: kworker/1:0 Not tainted 5.2.0-rc6-custom-02495-g8e73ca3be4af #1701\n [ +0.000010] Hardware name: Mellanox Technologies Ltd. MSN2100-CB2FO/SA001017, BIOS 5.6.5 06/07/2016\n [ +0.000016] Workqueue: events_freezable_power_ thermal_zone_device_check\n [ +0.000012] Call Trace:\n [ +0.000021] dump_stack+0xa9/0x10e\n [ +0.000020] print_address_description.cold.2+0x9/0x25e\n [ +0.000018] __kasan_report.cold.3+0x78/0x9d\n [ +0.000016] kasan_report+0xe/0x20\n [ +0.000016] __mutex_lock+0x1076/0x11c0\n [ +0.000014] step_wise_throttle+0x72/0x150\n [ +0.000018] handle_thermal_trip+0x167/0x760\n [ +0.000019] thermal_zone_device_update+0x19e/0x5f0\n [ +0.000019] process_one_work+0x969/0x16f0\n [ +0.000017] worker_thread+0x91/0xc40\n [ +0.000014] kthread+0x33d/0x400\n [ +0.000015] ret_from_fork+0x3a/0x50\n \n [ +0.000020] Allocated by task 1:\n [ +0.000015] save_stack+0x19/0x80\n [ +0.000015] __kasan_kmalloc.constprop.4+0xc1/0xd0\n [ +0.000014] kmem_cache_alloc_trace+0x152/0x320\n [ +0.000015] thermal_zone_device_register+0x1b4/0x13a0\n [ +0.000015] mlxsw_thermal_init+0xc92/0x23d0\n [ +0.000014] __mlxsw_core_bus_device_register+0x659/0x11b0\n [ +0.000013] mlxsw_core_bus_device_register+0x3d/0x90\n [ +0.000013] mlxsw_pci_probe+0x355/0x4b0\n [ +0.000014] local_pci_probe+0xc3/0x150\n [ +0.000013] pci_device_probe+0x280/0x410\n [ +0.000013] really_probe+0x26a/0xbb0\n [ +0.000013] driver_probe_device+0x208/0x2e0\n [ +0.000013] device_driver_attach+0xfe/0x140\n [ +0.000013] __driver_attach+0x110/0x310\n [ +0.000013] bus_for_each_dev+0x14b/0x1d0\n [ +0.000013] driver_register+0x1c0/0x400\n [ +0.000015] mlxsw_sp_module_init+0x5d/0xd3\n [ +0.000014] do_one_initcall+0x239/0x4dd\n [ +0.000013] kernel_init_freeable+0x42b/0x4e8\n [ +0.000012] kernel_init+0x11/0x18b\n [ +0.000013] ret_from_fork+0x3a/0x50\n \n [ +0.000015] Freed by task 581:\n [ +0.000013] save_stack+0x19/0x80\n [ +0.000014] __kasan_slab_free+0x125/0x170\n [ +0.000013] kfree+0xf3/0x310\n [ +0.000013] thermal_release+0xc7/0xf0\n [ +0.000014] device_release+0x77/0x200\n [ +0.000014] kobject_put+0x1a8/0x4c0\n [ +0.000014] device_unregister+0x38/0xc0\n [ +0.000014] thermal_zone_device_unregister+0x54e/0x6a0\n [ +0.000014] mlxsw_thermal_fini+0x184/0x35a\n [ +0.000014] mlxsw_core_bus_device_unregister+0x10a/0x640\n [ +0.000013] mlxsw_devlink_core_bus_device_reload+0x92/0x210\n [ +0.000015] devlink_nl_cmd_reload+0x113/0x1f0\n [ +0.000014] genl_family_rcv_msg+0x700/0xee0\n [ +0.000013] genl_rcv_msg+0xca/0x170\n [ +0.000013] netlink_rcv_skb+0x137/0x3a0\n [ +0.000012] genl_rcv+0x29/0x40\n [ +0.000013] netlink_unicast+0x49b/0x660\n [ +0.000013] netlink_sendmsg+0x755/0xc90\n [ +0.000013] __sys_sendto+0x3de/0x430\n [ +0.000013] __x64_sys_sendto+0xe2/0x1b0\n [ +0.000013] do_syscall_64+0xa4/0x4d0\n [ +0.000013] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n \n [ +0.000017] The buggy address belongs to the object at ffff8881e48e0008\n which belongs to the cache kmalloc-2k of size 2048\n [ +0.000012] The buggy address is located 1096 bytes inside of\n 2048-byte region [ffff8881e48e0008, ffff8881e48e0808)\n [ +0.000007] The buggy address belongs to the page:\n [ +0.000012] page:ffffea0007923800 refcount:1 mapcount:0 mapping:ffff88823680d0c0 index:0x0 compound_mapcount: 0\n [ +0.000020] flags: 0x200000000010200(slab|head)\n [ +0.000019] raw: 0200000000010200 ffffea0007682008 ffffea00076ab808 ffff88823680d0c0\n [ +0.000016] raw: 0000000000000000 00000000000d000d 00000001ffffffff 0000000000000000\n [ +0.000007] page dumped because: kasan: bad access detected\n \n [ +0.000012] Memory state around the buggy address:\n [ +0.000012] ffff8881e48e0300: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb\n [ +0.000012] ffff8881e48e0380: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb\n [ +0.000012] \u003effff8881e48e0400: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb\n [ +0.000008] ^\n [ +0.000012] ffff8881e48e0480: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb\n [ +0.000012] ffff8881e48e0500: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb\n [ +0.000007] \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n \n Fixes: b1569e99c795 (\"ACPI: move thermal trip handling to generic thermal layer\")\n Reported-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Zhang Rui \u003crui.zhang@intel.com\u003e\n\ncommit dc62c2398e2abf552d7e29e9f38bf0197a2eb867\nAuthor: Yue Hu \u003chuyue2@yulong.com\u003e\nDate: Wed Aug 7 11:01:30 2019 +0800\n\n thermal/drivers/core: Use put_device() if device_register() fails\n \n Never directly free @dev after calling device_register(), even if it\n returned an error! Always use put_device() to give up the reference\n initialized. Clean up the rollback block also.\n \n Signed-off-by: Yue Hu \u003chuyue2@yulong.com\u003e\n Signed-off-by: Zhang Rui \u003crui.zhang@intel.com\u003e\n\ncommit 144e5d6ebeaecbed7220540df8201c9a6fefa360\nAuthor: Stefan Mavrodiev \u003cstefan@olimex.com\u003e\nDate: Fri Jul 26 16:32:36 2019 +0300\n\n thermal_hwmon: Sanitize thermal_zone type\n \n When calling thermal_add_hwmon_sysfs(), the device type is sanitized by\n replacing \u0027-\u0027 with \u0027_\u0027. However tz-\u003etype remains unsanitized. Thus\n calling thermal_hwmon_lookup_by_type() returns no device. And if there is\n no device, thermal_remove_hwmon_sysfs() fails with \"hwmon device lookup\n failed!\".\n \n The result is unregisted hwmon devices in the sysfs.\n \n Fixes: 409ef0bacacf (\"thermal_hwmon: Sanitize attribute name passed to hwmon\")\n \n Signed-off-by: Stefan Mavrodiev \u003cstefan@olimex.com\u003e\n Signed-off-by: Zhang Rui \u003crui.zhang@intel.com\u003e\n\ncommit af3e0fe9c66370a6a69007da33f826322c11e77e\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jun 28 09:29:54 2019 -0300\n\n docs: thermal: add it to the driver API\n \n The file contents mostly describes driver internals.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Zhang Rui \u003crui.zhang@intel.com\u003e\n\ncommit d777478599f781fc5162d1ae95dbee6e5ae05a41\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Aug 13 11:57:11 2019 +0530\n\n drm/xen-front: Make structure fb_funcs constant\n \n Static structure fb_funcs, of type drm_framebuffer_funcs, is used only\n when it is passed to drm_gem_fb_create_with_funcs() as its last\n argument. drm_gem_fb_create_with_funcs does not modify its lst argument\n (fb_funcs) and hence fb_funcs is never modified. Therefore make fb_funcs\n constant to protect it from further modification.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Oleksandr Andrushchenko \u003coleksandr_andrushchenko@epam.com\u003e\n Reviewed-by: Oleksandr Andrushchenko \u003coleksandr_andrushchenko@epam.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813062712.24993-1-nishkadg.linux@gmail.com\n\ncommit 39f14e79b15a40709ef177bc4c07f193b6d3bce3\nAuthor: Cédric Le Goater \u003cclg@kaod.org\u003e\nDate: Wed Aug 14 17:47:54 2019 +0200\n\n powerpc/xmon: Add a dump of all XIVE interrupts\n \n Modify the xmon \u0027dxi\u0027 command to query all interrupts if no IRQ number\n is specified.\n \n Signed-off-by: Cédric Le Goater \u003cclg@kaod.org\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190814154754.23682-4-clg@kaod.org\n\ncommit b4868ff55d082bc66b0c287a41e4888f6d3e5f87\nAuthor: Cédric Le Goater \u003cclg@kaod.org\u003e\nDate: Wed Aug 14 17:47:53 2019 +0200\n\n powerpc/xive: Fix dump of XIVE interrupt under pseries\n \n The xmon \u0027dxi\u0027 command calls OPAL to query the XIVE configuration of a\n interrupt. This can only be done on baremetal (PowerNV) and it will\n crash a pseries machine.\n \n Introduce a new XIVE get_irq_config() operation which implements a\n different query depending on the platform, PowerNV or pseries, and\n modify xmon to use a top level wrapper.\n \n Signed-off-by: Cédric Le Goater \u003cclg@kaod.org\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190814154754.23682-3-clg@kaod.org\n\ncommit c3e0dbd7f780a58c4695f1cd8fc8afde80376737\nAuthor: Cédric Le Goater \u003cclg@kaod.org\u003e\nDate: Wed Aug 14 17:47:52 2019 +0200\n\n powerpc/xmon: Check for HV mode when dumping XIVE info from OPAL\n \n Currently, the xmon \u0027dx\u0027 command calls OPAL to dump the XIVE state in\n the OPAL logs and also outputs some of the fields of the internal XIVE\n structures in Linux. The OPAL calls can only be done on baremetal\n (PowerNV) and they crash a pseries machine. Fix by checking the\n hypervisor feature of the CPU.\n \n Signed-off-by: Cédric Le Goater \u003cclg@kaod.org\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190814154754.23682-2-clg@kaod.org\n\ncommit 201ed7f327a17577debec52c33786d4b3259d0dc\nAuthor: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\nDate: Thu Jul 18 15:11:39 2019 +1000\n\n powerpc/powernv/ioda2: Create bigger default window with 64k IOMMU pages\n \n At the moment we create a small window only for 32bit devices, the window\n maps 0..2GB of the PCI space only. For other devices we either use\n a sketchy bypass or hardware bypass but the former can only work if\n the amount of RAM is no bigger than the device\u0027s DMA mask and the latter\n requires devices to support at least 59bit DMA.\n \n This extends the default DMA window to the maximum size possible to allow\n a wider DMA mask than just 32bit. The default window size is now limited\n by the the iommu_table::it_map allocation bitmap which is a contiguous\n array, 1 bit per an IOMMU page.\n \n This increases the default IOMMU page size from hard coded 4K to\n the system page size to allow wider DMA masks.\n \n This increases the level number to not exceed the max order allocation\n limit per TCE level. By the same time, this keeps minimal levels number\n as 2 in order to save memory.\n \n As the extended window now overlaps the 32bit MMIO region, this adds\n an area reservation to iommu_init_table().\n \n After this change the default window size is 0x80000000000\u003d\u003d1\u003c\u003c43 so\n devices limited to DMA mask smaller than the amount of system RAM can\n still use more than just 2GB of memory for DMA.\n \n This is an optimization and not a bug fix for DMA API usage.\n \n With the on-demand allocation of indirect TCE table levels enabled and\n 2 levels, the first TCE level size is just\n 1\u003c\u003cceil((log2(0x7ffffffffff+1)-16)/2)\u003d16384 TCEs or 2 system pages.\n \n Signed-off-by: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190718051139.74787-5-aik@ozlabs.ru\n\ncommit c37c792dec0929dbb6360a609fb00fa20bb16fc2\nAuthor: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\nDate: Thu Jul 18 15:11:38 2019 +1000\n\n powerpc/powernv/ioda2: Allocate TCE table levels on demand for default DMA window\n \n We allocate only the first level of multilevel TCE tables for KVM\n already (alloc_userspace_copy\u003d\u003dtrue), and the rest is allocated on demand.\n This is not enabled though for bare metal.\n \n This removes the KVM limitation (implicit, via the alloc_userspace_copy\n parameter) and always allocates just the first level. The on-demand\n allocation of missing levels is already implemented.\n \n As from now on DMA map might happen with disabled interrupts, this\n allocates TCEs with GFP_ATOMIC; otherwise lockdep reports errors 1].\n In practice just a single page is allocated there so chances for failure\n are quite low.\n \n To save time when creating a new clean table, this skips non-allocated\n indirect TCE entries in pnv_tce_free just like we already do in\n the VFIO IOMMU TCE driver.\n \n This changes the default level number from 1 to 2 to reduce the amount\n of memory required for the default 32bit DMA window at the boot time.\n The default window size is up to 2GB which requires 4MB of TCEs which is\n unlikely to be used entirely or at all as most devices these days are\n 64bit capable so by switching to 2 levels by default we save 4032KB of\n RAM per a device.\n \n While at this, add __GFP_NOWARN to alloc_pages_node() as the userspace\n can trigger this path via VFIO, see the failure and try creating a table\n again with different parameters which might succeed.\n \n [1]:\n \u003d\u003d\u003d\n BUG: sleeping function called from invalid context at mm/page_alloc.c:4596\n in_atomic(): 1, irqs_disabled(): 1, pid: 1038, name: scsi_eh_1\n 2 locks held by scsi_eh_1/1038:\n #0: 000000005efd659a (\u0026host-\u003eeh_mutex){+.+.}, at: ata_eh_acquire+0x34/0x80\n #1: 0000000006cf56a6 (\u0026(\u0026host-\u003elock)-\u003erlock){....}, at: ata_exec_internal_sg+0xb0/0x5c0\n irq event stamp: 500\n hardirqs last enabled at (499): [\u003cc000000000cb8a74\u003e] _raw_spin_unlock_irqrestore+0x94/0xd0\n hardirqs last disabled at (500): [\u003cc000000000cb85c4\u003e] _raw_spin_lock_irqsave+0x44/0x120\n softirqs last enabled at (0): [\u003cc000000000101120\u003e] copy_process.isra.4.part.5+0x640/0x1a80\n softirqs last disabled at (0): [\u003c0000000000000000\u003e] 0x0\n CPU: 73 PID: 1038 Comm: scsi_eh_1 Not tainted 5.2.0-rc6-le_nv2_aikATfstn1-p1 #634\n Call Trace:\n [c000003d064cef50] [c000000000c8e6c4] dump_stack+0xe8/0x164 (unreliable)\n [c000003d064cefa0] [c00000000014ed78] ___might_sleep+0x2f8/0x310\n [c000003d064cf020] [c0000000003ca084] __alloc_pages_nodemask+0x2a4/0x1560\n [c000003d064cf220] [c0000000000c2530] pnv_alloc_tce_level.isra.0+0x90/0x130\n [c000003d064cf290] [c0000000000c2888] pnv_tce+0x128/0x3b0\n [c000003d064cf360] [c0000000000c2c00] pnv_tce_build+0xb0/0xf0\n [c000003d064cf3c0] [c0000000000bbd9c] pnv_ioda2_tce_build+0x3c/0xb0\n [c000003d064cf400] [c00000000004cfe0] ppc_iommu_map_sg+0x210/0x550\n [c000003d064cf510] [c00000000004b7a4] dma_iommu_map_sg+0x74/0xb0\n [c000003d064cf530] [c000000000863944] ata_qc_issue+0x134/0x470\n [c000003d064cf5b0] [c000000000863ec4] ata_exec_internal_sg+0x244/0x5c0\n [c000003d064cf700] [c0000000008642d0] ata_exec_internal+0x90/0xe0\n [c000003d064cf780] [c0000000008650ac] ata_dev_read_id+0x2ec/0x640\n [c000003d064cf8d0] [c000000000878e28] ata_eh_recover+0x948/0x16d0\n [c000003d064cfa10] [c00000000087d760] sata_pmp_error_handler+0x480/0xbf0\n [c000003d064cfbc0] [c000000000884624] ahci_error_handler+0x74/0xe0\n [c000003d064cfbf0] [c000000000879fa8] ata_scsi_port_error_handler+0x2d8/0x7c0\n [c000003d064cfca0] [c00000000087a544] ata_scsi_error+0xb4/0x100\n [c000003d064cfd00] [c000000000802450] scsi_error_handler+0x120/0x510\n [c000003d064cfdb0] [c000000000140c48] kthread+0x1b8/0x1c0\n [c000003d064cfe20] [c00000000000bd8c] ret_from_kernel_thread+0x5c/0x70\n ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300)\n irq event stamp: 2305\n \n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n hardirqs last enabled at (2305): [\u003cc00000000000e4c8\u003e] fast_exc_return_irq+0x28/0x34\n hardirqs last disabled at (2303): [\u003cc000000000cb9fd0\u003e] __do_softirq+0x4a0/0x654\n WARNING: possible irq lock inversion dependency detected\n 5.2.0-rc6-le_nv2_aikATfstn1-p1 #634 Tainted: G W\n softirqs last enabled at (2304): [\u003cc000000000cba054\u003e] __do_softirq+0x524/0x654\n softirqs last disabled at (2297): [\u003cc00000000010f278\u003e] irq_exit+0x128/0x180\n --------------------------------------------------------\n swapper/0/0 just changed the state of lock:\n 0000000006cf56a6 (\u0026(\u0026host-\u003elock)-\u003erlock){-...}, at: ahci_single_level_irq_intr+0xac/0x120\n but this lock took another, HARDIRQ-unsafe lock in the past:\n (fs_reclaim){+.+.}\n \n and interrupts could create inverse lock ordering between them.\n \n other info that might help us debug this:\n Possible interrupt unsafe locking scenario:\n \n CPU0 CPU1\n ---- ----\n lock(fs_reclaim);\n local_irq_disable();\n lock(\u0026(\u0026host-\u003elock)-\u003erlock);\n lock(fs_reclaim);\n \u003cInterrupt\u003e\n lock(\u0026(\u0026host-\u003elock)-\u003erlock);\n \n *** DEADLOCK ***\n \n no locks held by swapper/0/0.\n \n the shortest dependencies between 2nd lock and 1st lock:\n -\u003e (fs_reclaim){+.+.} ops: 167579 {\n HARDIRQ-ON-W at:\n lock_acquire+0xf8/0x2a0\n fs_reclaim_acquire.part.23+0x44/0x60\n kmem_cache_alloc_node_trace+0x80/0x590\n alloc_desc+0x64/0x270\n __irq_alloc_descs+0x2e4/0x3a0\n irq_domain_alloc_descs+0xb0/0x150\n irq_create_mapping+0x168/0x2c0\n xics_smp_probe+0x2c/0x98\n pnv_smp_probe+0x40/0x9c\n smp_prepare_cpus+0x524/0x6c4\n kernel_init_freeable+0x1b4/0x650\n kernel_init+0x2c/0x148\n ret_from_kernel_thread+0x5c/0x70\n SOFTIRQ-ON-W at:\n lock_acquire+0xf8/0x2a0\n fs_reclaim_acquire.part.23+0x44/0x60\n kmem_cache_alloc_node_trace+0x80/0x590\n alloc_desc+0x64/0x270\n __irq_alloc_descs+0x2e4/0x3a0\n irq_domain_alloc_descs+0xb0/0x150\n irq_create_mapping+0x168/0x2c0\n xics_smp_probe+0x2c/0x98\n pnv_smp_probe+0x40/0x9c\n smp_prepare_cpus+0x524/0x6c4\n kernel_init_freeable+0x1b4/0x650\n kernel_init+0x2c/0x148\n ret_from_kernel_thread+0x5c/0x70\n INITIAL USE at:\n lock_acquire+0xf8/0x2a0\n fs_reclaim_acquire.part.23+0x44/0x60\n kmem_cache_alloc_node_trace+0x80/0x590\n alloc_desc+0x64/0x270\n __irq_alloc_descs+0x2e4/0x3a0\n irq_domain_alloc_descs+0xb0/0x150\n irq_create_mapping+0x168/0x2c0\n xics_smp_probe+0x2c/0x98\n pnv_smp_probe+0x40/0x9c\n smp_prepare_cpus+0x524/0x6c4\n kernel_init_freeable+0x1b4/0x650\n kernel_init+0x2c/0x148\n ret_from_kernel_thread+0x5c/0x70\n }\n \u003d\u003d\u003d\n \n Signed-off-by: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\n Reviewed-by: Alistair Popple \u003calistair@popple.id.au\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190718051139.74787-4-aik@ozlabs.ru\n\ncommit 4f7e0babbc7c46fc0db4f5c14fe96bf6f4d69502\nAuthor: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\nDate: Thu Jul 18 15:11:37 2019 +1000\n\n powerpc/iommu: Allow bypass-only for DMA\n \n POWER8 and newer support a bypass mode which maps all host memory to\n PCI buses so an IOMMU table is not always required. However if we fail to\n create such a table, the DMA setup fails and the kernel does not boot.\n \n This skips the 32bit DMA setup check if the bypass is selected.\n \n Signed-off-by: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\n Reviewed-by: David Gibson \u003cdavid@gibson.dropbear.id.au\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190718051139.74787-3-aik@ozlabs.ru\n\ncommit 56090a3902c80c296e822d11acdb6a101b322c52\nAuthor: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\nDate: Thu Jul 18 15:11:36 2019 +1000\n\n powerpc/powernv/ioda: Fix race in TCE level allocation\n \n pnv_tce() returns a pointer to a TCE entry and originally a TCE table\n would be pre-allocated. For the default case of 2GB window the table\n needs only a single level and that is fine. However if more levels are\n requested, it is possible to get a race when 2 threads want a pointer\n to a TCE entry from the same page of TCEs.\n \n This adds cmpxchg to handle the race. Note that once TCE is non-zero,\n it cannot become zero again.\n \n Fixes: a68bd1267b72 (\"powerpc/powernv/ioda: Allocate indirect TCE levels on demand\")\n CC: stable@vger.kernel.org # v4.19+\n Signed-off-by: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190718051139.74787-2-aik@ozlabs.ru\n\ncommit c784be435d5dae28d3b03db31753dd7a18733f0c\nAuthor: Gautham R. Shenoy \u003cego@linux.vnet.ibm.com\u003e\nDate: Wed May 15 13:15:52 2019 +0530\n\n powerpc/pseries: Fix cpu_hotplug_lock acquisition in resize_hpt()\n \n The calls to arch_add_memory()/arch_remove_memory() are always made\n with the read-side cpu_hotplug_lock acquired via memory_hotplug_begin().\n On pSeries, arch_add_memory()/arch_remove_memory() eventually call\n resize_hpt() which in turn calls stop_machine() which acquires the\n read-side cpu_hotplug_lock again, thereby resulting in the recursive\n acquisition of this lock.\n \n In the absence of CONFIG_PROVE_LOCKING, we hadn\u0027t observed a system\n lockup during a memory hotplug operation because cpus_read_lock() is a\n per-cpu rwsem read, which, in the fast-path (in the absence of the\n writer, which in our case is a CPU-hotplug operation) simply\n increments the read_count on the semaphore. Thus a recursive read in\n the fast-path doesn\u0027t cause any problems.\n \n However, we can hit this problem in practice if there is a concurrent\n CPU-Hotplug operation in progress which is waiting to acquire the\n write-side of the lock. This will cause the second recursive read to\n block until the writer finishes. While the writer is blocked since the\n first read holds the lock. Thus both the reader as well as the writers\n fail to make any progress thereby blocking both CPU-Hotplug as well as\n Memory Hotplug operations.\n \n Memory-Hotplug CPU-Hotplug\n CPU 0 CPU 1\n ------ ------\n \n 1. down_read(cpu_hotplug_lock.rw_sem)\n [memory_hotplug_begin]\n 2. down_write(cpu_hotplug_lock.rw_sem)\n [cpu_up/cpu_down]\n 3. down_read(cpu_hotplug_lock.rw_sem)\n [stop_machine()]\n \n Lockdep complains as follows in these code-paths.\n \n swapper/0/1 is trying to acquire lock:\n (____ptrval____) (cpu_hotplug_lock.rw_sem){++++}, at: stop_machine+0x2c/0x60\n \n but task is already holding lock:\n (____ptrval____) (cpu_hotplug_lock.rw_sem){++++}, at: mem_hotplug_begin+0x20/0x50\n \n other info that might help us debug this:\n Possible unsafe locking scenario:\n \n CPU0\n ----\n lock(cpu_hotplug_lock.rw_sem);\n lock(cpu_hotplug_lock.rw_sem);\n \n *** DEADLOCK ***\n \n May be due to missing lock nesting notation\n \n 3 locks held by swapper/0/1:\n #0: (____ptrval____) (\u0026dev-\u003emutex){....}, at: __driver_attach+0x12c/0x1b0\n #1: (____ptrval____) (cpu_hotplug_lock.rw_sem){++++}, at: mem_hotplug_begin+0x20/0x50\n #2: (____ptrval____) (mem_hotplug_lock.rw_sem){++++}, at: percpu_down_write+0x54/0x1a0\n \n stack backtrace:\n CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.0.0-rc5-58373-gbc99402235f3-dirty #166\n Call Trace:\n dump_stack+0xe8/0x164 (unreliable)\n __lock_acquire+0x1110/0x1c70\n lock_acquire+0x240/0x290\n cpus_read_lock+0x64/0xf0\n stop_machine+0x2c/0x60\n pseries_lpar_resize_hpt+0x19c/0x2c0\n resize_hpt_for_hotplug+0x70/0xd0\n arch_add_memory+0x58/0xfc\n devm_memremap_pages+0x5e8/0x8f0\n pmem_attach_disk+0x764/0x830\n nvdimm_bus_probe+0x118/0x240\n really_probe+0x230/0x4b0\n driver_probe_device+0x16c/0x1e0\n __driver_attach+0x148/0x1b0\n bus_for_each_dev+0x90/0x130\n driver_attach+0x34/0x50\n bus_add_driver+0x1a8/0x360\n driver_register+0x108/0x170\n __nd_driver_register+0xd0/0xf0\n nd_pmem_driver_init+0x34/0x48\n do_one_initcall+0x1e0/0x45c\n kernel_init_freeable+0x540/0x64c\n kernel_init+0x2c/0x160\n ret_from_kernel_thread+0x5c/0x68\n \n Fix this issue by\n 1) Requiring all the calls to pseries_lpar_resize_hpt() be made\n with cpu_hotplug_lock held.\n \n 2) In pseries_lpar_resize_hpt() invoke stop_machine_cpuslocked()\n as a consequence of 1)\n \n 3) To satisfy 1), in hpt_order_set(), call mmu_hash_ops.resize_hpt()\n with cpu_hotplug_lock held.\n \n Fixes: dbcf929c0062 (\"powerpc/pseries: Add support for hash table resizing\")\n Cc: stable@vger.kernel.org # v4.11+\n Reported-by: Aneesh Kumar K.V \u003caneesh.kumar@linux.ibm.com\u003e\n Signed-off-by: Gautham R. Shenoy \u003cego@linux.vnet.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/1557906352-29048-1-git-send-email-ego@linux.vnet.ibm.com\n\ncommit 20e79a0a2cfd15b6cfb18119f2e108396be56716\nAuthor: Yonglong Liu \u003cliuyonglong@huawei.com\u003e\nDate: Sat Aug 17 10:21:17 2019 +0800\n\n net: hns: add phy_attached_info() to the hns driver\n \n This patch adds the call to phy_attached_info() to the hns driver\n to identify which exact PHY drivers is in use.\n \n Suggested-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: Yonglong Liu \u003cliuyonglong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 296c9120752bab93151bd7220896dd302683a91c\nAuthor: Stefan Roese \u003csr@denx.de\u003e\nDate: Fri Aug 16 15:23:25 2019 +0200\n\n net: ethernet: mediatek: Add MT7628/88 SoC support\n \n This patch adds support for the MediaTek MT7628/88 SoCs to the common\n MediaTek ethernet driver. Some minor changes are needed for this and\n a bigger change, as the MT7628 does not support QDMA (only PDMA).\n \n Signed-off-by: Stefan Roese \u003csr@denx.de\u003e\n Cc: René van Dorst \u003copensource@vdorst.com\u003e\n Cc: Daniel Golle \u003cdaniel@makrotopia.org\u003e\n Cc: Sean Wang \u003csean.wang@mediatek.com\u003e\n Cc: John Crispin \u003cjohn@phrozen.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 08df5fa63accfc1c847cc2d91be4ca698946394d\nAuthor: Stefan Roese \u003csr@denx.de\u003e\nDate: Fri Aug 16 15:23:24 2019 +0200\n\n net: ethernet: mediatek: Rename NEXT_RX_DESP_IDX to NEXT_DESP_IDX\n \n Rename the NEXT_RX_DESP_IDX macro to NEXT_DESP_IDX, so that it better\n can be used for TX ops as well. This will be used in the upcoming\n MT7628/88 support (same functionality for RX and TX in this macro).\n \n Signed-off-by: Stefan Roese \u003csr@denx.de\u003e\n Cc: René van Dorst \u003copensource@vdorst.com\u003e\n Cc: Daniel Golle \u003cdaniel@makrotopia.org\u003e\n Cc: Sean Wang \u003csean.wang@mediatek.com\u003e\n Cc: John Crispin \u003cjohn@phrozen.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 45487403e1045c1019b65930a6719fec43f986db\nAuthor: Stefan Roese \u003csr@denx.de\u003e\nDate: Fri Aug 16 15:23:23 2019 +0200\n\n net: ethernet: mediatek: Rename MTK_QMTK_INT_STATUS to MTK_QDMA_INT_STATUS\n \n Currently all QDMA registers are named \"MTK_QDMA_foo\" in this driver\n with one exception: MTK_QMTK_INT_STATUS. This patch renames\n MTK_QMTK_INT_STATUS to MTK_QDMA_INT_STATUS so that all macros follow\n this rule.\n \n Signed-off-by: Stefan Roese \u003csr@denx.de\u003e\n Cc: René van Dorst \u003copensource@vdorst.com\u003e\n Cc: Daniel Golle \u003cdaniel@makrotopia.org\u003e\n Cc: Sean Wang \u003csean.wang@mediatek.com\u003e\n Cc: John Crispin \u003cjohn@phrozen.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 5be4480d4656020c17dbac2831d62615af4f1631\nAuthor: Stefan Roese \u003csr@denx.de\u003e\nDate: Fri Aug 16 15:23:22 2019 +0200\n\n dt-bindings: net: mediatek: Add support for MediaTek MT7628/88 SoC\n \n Add compatible for the ethernet IP core on MT7628/88 SoCs. Its\n compatible with the older Ralink Rt5350F SoC. And OpenWrt already\n uses this compatible string for the MT76x8.\n \n Signed-off-by: Stefan Roese \u003csr@denx.de\u003e\n Cc: René van Dorst \u003copensource@vdorst.com\u003e\n Cc: Daniel Golle \u003cdaniel@makrotopia.org\u003e\n Cc: Sean Wang \u003csean.wang@mediatek.com\u003e\n Cc: John Crispin \u003cjohn@phrozen.org\u003e\n Cc: devicetree@vger.kernel.org\n Cc: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9041f047df3afe545089173e5ae71592382de169\nAuthor: Ivan Vecera \u003civecera@redhat.com\u003e\nDate: Fri Aug 16 09:15:35 2019 +0200\n\n be2net: eliminate enable field from be_aic_obj\n \n Adaptive coalescing is managed per adapter not per event queue so it\n does not needed to store \u0027enable\u0027 flag for each event queue.\n \n Signed-off-by: Ivan Vecera \u003civecera@redhat.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit e654f9f53b45fde3fcc8051830b212c7a8f36148\nAuthor: Jon Maloy \u003cjon.maloy@ericsson.com\u003e\nDate: Thu Aug 15 16:42:50 2019 +0200\n\n tipc: clean up skb list lock handling on send path\n \n The policy for handling the skb list locks on the send and receive paths\n is simple.\n \n - On the send path we never need to grab the lock on the \u0027xmitq\u0027 list\n when the destination is an exernal node.\n \n - On the receive path we always need to grab the lock on the \u0027inputq\u0027\n list, irrespective of source node.\n \n However, when transmitting node local messages those will eventually\n end up on the receive path of a local socket, meaning that the argument\n \u0027xmitq\u0027 in tipc_node_xmit() will become the \u0027ínputq\u0027 argument in the\n function tipc_sk_rcv(). This has been handled by always initializing\n the spinlock of the \u0027xmitq\u0027 list at message creation, just in case it\n may end up on the receive path later, and despite knowing that the lock\n in most cases never will be used.\n \n This approach is inaccurate and confusing, and has also concealed the\n fact that the stated \u0027no lock grabbing\u0027 policy for the send path is\n violated in some cases.\n \n We now clean up this by never initializing the lock at message creation,\n instead doing this at the moment we find that the message actually will\n enter the receive path. At the same time we fix the four locations\n where we incorrectly access the spinlock on the send/error path.\n \n This patch also reverts commit d12cffe9329f (\"tipc: ensure head-\u003elock\n is initialised\") which has now become redundant.\n \n CC: Eric Dumazet \u003cedumazet@google.com\u003e\n Reported-by: Chris Packham \u003cchris.packham@alliedtelesis.co.nz\u003e\n Acked-by: Ying Xue \u003cying.xue@windriver.com\u003e\n Signed-off-by: Jon Maloy \u003cjon.maloy@ericsson.com\u003e\n Reviewed-by: Xin Long \u003clucien.xin@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1bef61fc7eedf6ac341fd73574c85ae199ca9aae\nAuthor: Yonglong Liu \u003cliuyonglong@huawei.com\u003e\nDate: Fri Aug 16 16:09:42 2019 +0800\n\n net: hns3: add phy_attached_info() to the hns3 driver\n \n This patch adds the call to phy_attached_info() to the hns3 driver\n to identify which exact PHY drivers and models is in use.\n \n Signed-off-by: Yonglong Liu \u003cliuyonglong@huawei.com\u003e\n Reviewed-by: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 20981a1e6b708536493e10ba90016ae28a9f3749\nAuthor: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\nDate: Fri Aug 16 16:09:41 2019 +0800\n\n net: hns3: prevent unnecessary MAC TNL interrupt\n \n MAC TNL interrupt is used to collect statistic info about\n link status changing suddenly when netdev is running.\n \n But when stopping netdev, the enabled MAC TNL interrupt is\n unnecessary, and may add some noises to the statistic info.\n So this patch disables it before stopping MAC.\n \n Fixes: a63457878b12 (\"net: hns3: Add handling of MAC tunnel interruption\")\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Reviewed-by: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ac887be5b0fea53104fd6e2e5c904fb85f171f78\nAuthor: Xiaofei Tan \u003ctanxiaofei@huawei.com\u003e\nDate: Fri Aug 16 16:09:40 2019 +0800\n\n net: hns3: change print level of RAS error log from warning to error\n \n This patch changes print level of RAS error log from warning to error.\n Because RAS error and its recovery process could cause application\n failure. Also uses %u instead of %d when the parameter is unsigned.\n \n Signed-off-by: Xiaofei Tan \u003ctanxiaofei@huawei.com\u003e\n Signed-off-by: Weihang Li \u003cliweihang@hisilicon.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 37417c6625cae97762d7095863b0cf5a127f1c4e\nAuthor: Guojia Liao \u003cliaoguojia@huawei.com\u003e\nDate: Fri Aug 16 16:09:39 2019 +0800\n\n net: hns3: fix error and incorrect format\n \n The pointer type parameter should be declare as const for preventing\n from its pointed value being unexpected modified.\n \n The uninitialized variable can not be return directly. The default\n return value is 0 if no abnormal result.\n \n This patch fixes the preceding two errors, deletes redundant\n declaration of a function and align one parameter.\n \n Signed-off-by: Guojia Liao \u003cliaoguojia@huawei.com\u003e\n Signed-off-by: Weihang Li \u003cliweihang@hisilicon.com\u003e\n Signed-off-by: Jian Shen \u003cshenjian15@huawei.com\u003e\n Signed-off-by: Guangbin Huang \u003chuangguangbin2@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit cdd332accd4a93ac17d86024d52971d387b9a042\nAuthor: Guojia Liao \u003cliaoguojia@huawei.com\u003e\nDate: Fri Aug 16 16:09:38 2019 +0800\n\n net: hns3: modify redundant initialization of variable\n \n Some temporary variables do not need to be initialized that\n they will be set before used, so this patch deletes the\n initialization value of these temporary variables.\n \n Signed-off-by: Guojia Liao \u003cliaoguojia@huawei.com\u003e\n Signed-off-by: Guangbin Huang \u003chuangguangbin2@huawei.com\u003e\n Signed-off-by: Huzhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 46ee73508cbbfcb03216823043a2e24764f18664\nAuthor: Guojia Liao \u003cliaoguojia@huawei.com\u003e\nDate: Fri Aug 16 16:09:37 2019 +0800\n\n net: hns3: add or modify comments\n \n To explain some code, this patch adds some comments, and modifies or\n merges some comments to make them more neat.\n \n Signed-off-by: Guojia Liao \u003cliaoguojia@huawei.com\u003e\n Signed-off-by: Zhongzhu Liu \u003cliuzhongzhu@huawei.com\u003e\n Signed-off-by: Weihang Li \u003cliweihang@hisilicon.com\u003e\n Signed-off-by: Guangbin Huang \u003chuangguangbin2@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1bd231538c21d1cd691e71cbeeb4100fabc58068\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Aug 12 19:14:01 2019 +0300\n\n pinctrl: intel: Allow to request locked pads\n \n Some firmwares would like to protect pads from being modified by OS\n and at the same time provide them to OS as a resource. So, the driver\n in such circumstances may request pad and may not change its state.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Acked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n\ncommit 07ab0f8d9a129add914aff3e988da4033471dfc0\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Fri Aug 2 11:37:09 2019 +0100\n\n KVM: Call kvm_arch_vcpu_blocking early into the blocking sequence\n \n When a vpcu is about to block by calling kvm_vcpu_block, we call\n back into the arch code to allow any form of synchronization that\n may be required at this point (SVN stops the AVIC, ARM synchronises\n the VMCR and enables GICv4 doorbells). But this synchronization\n comes in quite late, as we\u0027ve potentially waited for halt_poll_ns\n to expire.\n \n Instead, let\u0027s move kvm_arch_vcpu_blocking() to the beginning of\n kvm_vcpu_block(), which on ARM has several benefits:\n \n - VMCR gets synchronised early, meaning that any interrupt delivered\n during the polling window will be evaluated with the correct guest\n PMR\n - GICv4 doorbells are enabled, which means that any guest interrupt\n directly injected during that window will be immediately recognised\n \n Tang Nianyao ran some tests on a GICv4 machine to evaluate such\n change, and reported up to a 10% improvement for netperf:\n \n \u003cquote\u003e\n netperf result:\n D06 as server, intel 8180 server as client\n with change:\n package 512 bytes - 5500 Mbits/s\n package 64 bytes - 760 Mbits/s\n without change:\n package 512 bytes - 5000 Mbits/s\n package 64 bytes - 710 Mbits/s\n \u003c/quote\u003e\n \n Acked-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 0ed5f5d63963673ccdd00d101637bf56e6d3d6dc\nAuthor: Alexandru Elisei \u003calexandru.elisei@arm.com\u003e\nDate: Thu Aug 15 10:56:22 2019 +0100\n\n KVM: arm/arm64: vgic: Make function comments match function declarations\n \n Since commit 503a62862e8f (\"KVM: arm/arm64: vgic: Rely on the GIC driver to\n parse the firmware tables\"), the vgic_v{2,3}_probe functions stopped using\n a DT node. Commit 909777324588 (\"KVM: arm/arm64: vgic-new: vgic_init:\n implement kvm_vgic_hyp_init\") changed the functions again, and now they\n require exactly one argument, a struct gic_kvm_info populated by the GIC\n driver. Unfortunately the comments regressed and state that a DT node is\n used instead. Change the function comments to reflect the current\n prototypes.\n \n Signed-off-by: Alexandru Elisei \u003calexandru.elisei@arm.com\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 363de99b770500b2102e9375249726254fdeae8a\nAuthor: Mark Rutland \u003cmark.rutland@arm.com\u003e\nDate: Tue Aug 6 16:57:37 2019 +0100\n\n arm64/kvm: Remove VMID rollover I-cache maintenance\n \n For VPIPT I-caches, we need I-cache maintenance on VMID rollover to\n avoid an ABA problem. Consider a single vCPU VM, with a pinned stage-2,\n running with an idmap VA-\u003eIPA and idmap IPA-\u003ePA. If we don\u0027t do\n maintenance on rollover:\n \n // VMID A\n Writes insn X to PA 0xF\n Invalidates PA 0xF (for VMID A)\n \n I$ contains [{A,F}-\u003eX]\n \n [VMID ROLLOVER]\n \n // VMID B\n Writes insn Y to PA 0xF\n Invalidates PA 0xF (for VMID B)\n \n I$ contains [{A,F}-\u003eX, {B,F}-\u003eY]\n \n [VMID ROLLOVER]\n \n // VMID A\n I$ contains [{A,F}-\u003eX, {B,F}-\u003eY]\n \n Unexpectedly hits stale I$ line {A,F}-\u003eX.\n \n However, for PIPT and VIPT I-caches, the VMID doesn\u0027t affect lookup or\n constrain maintenance. Given the VMID doesn\u0027t affect PIPT and VIPT\n I-caches, and given VMID rollover is independent of changes to stage-2\n mappings, I-cache maintenance cannot be necessary on VMID rollover for\n PIPT or VIPT I-caches.\n \n This patch removes the maintenance on rollover for VIPT and PIPT\n I-caches. At the same time, the unnecessary colons are removed from the\n asm statement to make it more legible.\n \n Signed-off-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Christoffer Dall \u003cchristoffer.dall@arm.com\u003e\n Reviewed-by: James Morse \u003cjames.morse@arm.com\u003e\n Cc: Julien Thierry \u003cjulien.thierry.kdev@gmail.com\u003e\n Cc: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Cc: kvmarm@lists.cs.columbia.edu\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 41108170d98069b34837c4512f93ef2ce651b812\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Mon Mar 18 10:34:48 2019 +0000\n\n KVM: arm/arm64: vgic-irqfd: Implement kvm_arch_set_irq_inatomic\n \n Now that we have a cache of MSI-\u003eLPI translations, it is pretty\n easy to implement kvm_arch_set_irq_inatomic (this cache can be\n parsed without sleeping).\n \n Hopefully, this will improve some LPI-heavy workloads.\n \n Tested-by: Andre Przywara \u003candre.przywara@arm.com\u003e\n Reviewed-by: Eric Auger \u003ceric.auger@redhat.com\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 86a7dae884f38e11f3af8137ddf1fb969cd2699c\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Mon Mar 18 10:29:30 2019 +0000\n\n KVM: arm/arm64: vgic-its: Check the LPI translation cache on MSI injection\n \n When performing an MSI injection, let\u0027s first check if the translation\n is already in the cache. If so, let\u0027s inject it quickly without\n going through the whole translation process.\n \n Tested-by: Andre Przywara \u003candre.przywara@arm.com\u003e\n Reviewed-by: Eric Auger \u003ceric.auger@redhat.com\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 89489ee9ced8924f64f99c0470eae38e9e4e204b\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Mon Mar 18 10:17:39 2019 +0000\n\n KVM: arm/arm64: vgic-its: Cache successful MSI-\u003eLPI translation\n \n On a successful translation, preserve the parameters in the LPI\n translation cache. Each translation is reusing the last slot\n in the list, naturally evicting the least recently used entry.\n \n Tested-by: Andre Przywara \u003candre.przywara@arm.com\u003e\n Reviewed-by: Eric Auger \u003ceric.auger@redhat.com\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit cbfda481d87e92ce635e426099946cd413b251be\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Mon Jun 10 10:20:02 2019 +0100\n\n KVM: arm/arm64: vgic-its: Invalidate MSI-LPI translation cache on vgic teardown\n \n In order to avoid leaking vgic_irq structures on teardown, we need to\n drop all references to LPIs before deallocating the cache itself.\n \n This is done by invalidating the cache on vgic teardown.\n \n Tested-by: Andre Przywara \u003candre.przywara@arm.com\u003e\n Reviewed-by: Eric Auger \u003ceric.auger@redhat.com\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 363518f37a86acc515defae6d1ba91a6c7617de9\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Wed May 22 18:16:49 2019 +0100\n\n KVM: arm/arm64: vgic-its: Invalidate MSI-LPI translation cache on ITS disable\n \n If an ITS gets disabled, we need to make sure that further interrupts\n won\u0027t hit in the cache. For that, we invalidate the translation cache\n when the ITS is disabled.\n \n Tested-by: Andre Przywara \u003candre.przywara@arm.com\u003e\n Reviewed-by: Eric Auger \u003ceric.auger@redhat.com\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit b4931afcde1ffccd4a406009aef33c14bc6c6cb8\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Wed May 22 18:16:49 2019 +0100\n\n KVM: arm/arm64: vgic-its: Invalidate MSI-LPI translation cache on disabling LPIs\n \n If a vcpu disables LPIs at its redistributor level, we need to make sure\n we won\u0027t pend more interrupts. For this, we need to invalidate the LPI\n translation cache.\n \n Tested-by: Andre Przywara \u003candre.przywara@arm.com\u003e\n Reviewed-by: Eric Auger \u003ceric.auger@redhat.com\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 0c14484866194382c0631d6e9bb14cc2fe83e814\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Mon Mar 18 10:23:16 2019 +0000\n\n KVM: arm/arm64: vgic-its: Invalidate MSI-LPI translation cache on specific commands\n \n The LPI translation cache needs to be discarded when an ITS command\n may affect the translation of an LPI (DISCARD, MAPC and MAPD with V\u003d0)\n or the routing of an LPI to a redistributor with disabled LPIs (MOVI,\n MOVALL).\n \n We decide to perform a full invalidation of the cache, irrespective\n of the LPI that is affected. Commands are supposed to be rare enough\n that it doesn\u0027t matter.\n \n Tested-by: Andre Przywara \u003candre.przywara@arm.com\u003e\n Reviewed-by: Eric Auger \u003ceric.auger@redhat.com\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 7d825fd6eaa7467c9b737fd5cf46e45f298d1c20\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Mon Jun 10 10:26:37 2019 +0100\n\n KVM: arm/arm64: vgic-its: Add MSI-LPI translation cache invalidation\n \n There\u0027s a number of cases where we need to invalidate the caching\n of translations, so let\u0027s add basic support for that.\n \n Tested-by: Andre Przywara \u003candre.przywara@arm.com\u003e\n Reviewed-by: Eric Auger \u003ceric.auger@redhat.com\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 1bb3691d8330f09a0699c124fcb71991b5fd105b\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Mon Mar 18 12:45:22 2019 +0000\n\n KVM: arm/arm64: vgic: Add __vgic_put_lpi_locked primitive\n \n Our LPI translation cache needs to be able to drop the refcount\n on an LPI whilst already holding the lpi_list_lock.\n \n Let\u0027s add a new primitive for this.\n \n Tested-by: Andre Przywara \u003candre.przywara@arm.com\u003e\n Reviewed-by: Eric Auger \u003ceric.auger@redhat.com\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 24cab82c34aa6f3ede3de1d8621624cb5ef33feb\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Mon Mar 18 10:13:01 2019 +0000\n\n KVM: arm/arm64: vgic: Add LPI translation cache definition\n \n Add the basic data structure that expresses an MSI to LPI\n translation as well as the allocation/release hooks.\n \n The size of the cache is arbitrarily defined as 16*nr_vcpus.\n \n Tested-by: Andre Przywara \u003candre.przywara@arm.com\u003e\n Reviewed-by: Eric Auger \u003ceric.auger@redhat.com\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit ef468849751251864a50576a3b0df43448a960fe\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Sun Aug 18 00:25:11 2019 +0100\n\n drm/i915: Propagate fence errors\n \n Errors spread like wildfire, and must eventually be returned to the\n user. They need to be captured and passed along the flow of fences,\n infecting each in turn with the existing error, until finally they fall\n out of a user visible result.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190817232511.11391-1-chris@chris-wilson.co.uk\n\ncommit 1e296b5be40d309a1585c14bc55da6ff6a29ecf0\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Sat Aug 17 20:43:40 2019 +0200\n\n usb: typec: fusb302: Call fusb302_debugfs_init earlier\n \n tcpm_register_port() will call some of the fusb302 code\u0027s callbacks\n wich in turn will call fusb302_log(). So we need to call\n fusb302_debugfs_init() before we call tcpm_register_port().\n \n This fixes the following warning, which was caused by the logbuffer_lock\n not yet being initialized (which is done by fusb302_debugfs_init):\n \n DEBUG_LOCKS_WARN_ON(lock-\u003emagic !\u003d lock)\n WARNING: CPU: 0 PID: 1306 at kernel/locking/mutex.c:912 __mutex_lock+0x978/0x9a0\n Modules linked in: fusb302(+) tcpm pi3usb30532 typec bq24190_charger snd_soc_sst_cht_bsw_rt5645 mei_hdcp dwc3 intel_rapl_msr udc_core ulpi gpio_keys intel_powerclamp coretemp kvm_intel brcmfmac kvm brcmutil joydev cfg80211 wdat_wdt irqbypass pcspkr intel_cstate extcon_intel_cht_wc i2c_cht_wc(E) snd_intel_sst_acpi snd_intel_sst_core snd_soc_rt5645 snd_soc_sst_atom_hifi2_platform snd_soc_acpi_intel_match snd_soc_rl6231 snd_soc_acpi intel_xhci_usb_role_switch roles hci_uart snd_soc_core btqca mei_txe btrtl processor_thermal_device mei snd_hdmi_lpe_audio lpc_ich snd_compress btbcm intel_rapl_common ac97_bus dwc3_pci snd_pcm_dmaengine intel_soc_dts_iosf btintel snd_seq bluetooth snd_seq_device snd_pcm intel_cht_int33fe_musb snd_timer intel_cht_int33fe_typec intel_hid intel_cht_int33fe_common sparse_keymap snd ecdh_generic goodix rfkill soundcore ecc spi_pxa2xx_platform max17042_battery dw_dmac int3406_thermal dptf_power acpi_pad soc_button_array int3400_thermal int3403_thermal\n gpd_pocket_fan intel_int0002_vgpio int340x_thermal_zone acpi_thermal_rel dm_crypt mmc_block i915 crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel i2c_algo_bit drm_kms_helper drm video sdhci_acpi sdhci mmc_core pwm_lpss_platform pwm_lpss i2c_dev\n CPU: 0 PID: 1306 Comm: systemd-udevd Tainted: G E 5.3.0-rc4+ #83\n Hardware name: Default string Default string/Default string, BIOS 5.11 06/28/2017\n RIP: 0010:__mutex_lock+0x978/0x9a0\n Code: c0 0f 84 26 f7 ff ff 44 8b 05 24 25 c8 00 45 85 c0 0f 85 16 f7 ff ff 48 c7 c6 da 55 2f ae 48 c7 c7 98 8c 2d ae e8 a0 f9 5c ff \u003c0f\u003e 0b e9 fc f6 ff ff 4c 89 f0 4d 89 fe 49 89 c7 e9 cf fa ff ff e8\n RSP: 0018:ffffb7a8c0523800 EFLAGS: 00010286\n RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000\n RDX: 0000000000000002 RSI: 0000000000000001 RDI: 0000000000000246\n RBP: ffffb7a8c05238c0 R08: 0000000000000000 R09: 0000000000000000\n R10: ffffb7a8c0523648 R11: 0000000000000030 R12: 0000000000000000\n R13: ffffb7a8c0523990 R14: ffff9bf22f70c028 R15: ffff9bf22f70c360\n FS: 00007f39ca234940(0000) GS:ffff9bf237400000(0000) knlGS:0000000000000000\n CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n CR2: 00007f1f108481a0 CR3: 0000000271f28000 CR4: 00000000001006f0\n Call Trace:\n ? find_held_lock+0x39/0x90\n ? _fusb302_log+0x81/0x1d0 [fusb302]\n ? vsnprintf+0x3aa/0x4f0\n ? _fusb302_log+0x81/0x1d0 [fusb302]\n _fusb302_log+0x81/0x1d0 [fusb302]\n ...\n \n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Reviewed-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Link: https://lore.kernel.org/r/20190817184340.64086-3-hdegoede@redhat.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b7cd6f64ff577d88ff532d9bdbdc7bbc53f27616\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Sat Aug 17 20:43:39 2019 +0200\n\n usb: typec: fusb302: Use usb_debug_root as root for our debugfs entry\n \n Use usb_debug_root as root for our debugfs entry instead of creating our\n own subdirectory under the debugfs root.\n \n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Reviewed-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Link: https://lore.kernel.org/r/20190817184340.64086-2-hdegoede@redhat.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 9f4bc2ed74086e51cfc75754adfcbff86f28597a\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Sat Aug 17 20:43:38 2019 +0200\n\n usb: typec: tcpm: Use usb_debug_root as root for our debugfs entry\n \n Use usb_debug_root as root for our debugfs entry instead of creating our\n own subdirectory under the debugfs root.\n \n Another patch in this series will make the same change to the fusb302\n driver, which also uses dev_name() (on the same device) for the debugfs\n entry name. So we also prefix dev_name() with \"tcpm-\" here to avoid a\n name conflict.\n \n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Reviewed-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Link: https://lore.kernel.org/r/20190817184340.64086-1-hdegoede@redhat.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a8dc0f6d187bcccc7c1a41f0062badbf3d74e44c\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Sun Aug 18 09:52:04 2019 +0000\n\n drm/i915/uc: Never fail on HuC firmware errors\n \n There is no need to mark whole GPU as wedged just because\n of the custom HuC fw failure as users can always verify\n actual HuC firmware status using existing HUC_STATUS ioctl.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190818095204.31568-4-michal.wajdeczko@intel.com\n\ncommit ee402140274e87030bbdcf92b145974f69ae01d3\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Sun Aug 18 09:52:03 2019 +0000\n\n drm/i915/uc: Don\u0027t always fail on unavailable GuC firmware\n \n If we failed to fetch default GuC firmware and we didn\u0027t plan\n to use it for the submission and we never have used GuC before\n then we may continue normal driver load, no need to declare\n GPU wedged (we can use execlist for submission) and it is safe\n to run without the HuC (users will check HuC status anyway).\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190818095204.31568-3-michal.wajdeczko@intel.com\n\ncommit 1ce5ba970cb713b8336ef4074c7a42a21c4e82e1\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Sun Aug 18 09:52:02 2019 +0000\n\n drm/i915/guc: Don\u0027t open log relay if GuC is not running\n \n As we plan to continue driver load after GuC initialization\n failure, we can\u0027t assume that GuC log data will be available\n just because GuC was initially enabled. We must check that\n GuC is still running instead.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190818095204.31568-2-michal.wajdeczko@intel.com\n\ncommit a3c59cf2971088c4bf889d8c093990ebf59fb789\nAuthor: Fugang Duan \u003cfugang.duan@nxp.com\u003e\nDate: Sun Aug 18 10:33:45 2019 +0100\n\n nvmem: imx: add i.MX8QM platform support\n \n i.MX8QM efuse table has some difference with i.MX8QXP platform,\n so add i.MX8QM platform support.\n \n Signed-off-by: Fugang Duan \u003cfugang.duan@nxp.com\u003e\n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190818093345.29647-8-srinivas.kandagatla@linaro.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit e3fd9d360c7de0563a65c61989a3c6b243da93e5\nAuthor: Fugang Duan \u003cfugang.duan@nxp.com\u003e\nDate: Sun Aug 18 10:33:44 2019 +0100\n\n dt-bindings: fsl: scu: add new compatible string for ocotp\n \n Add new compatible string \"fsl,imx8qm-scu-ocotp\" into binding\n doc for i.MX8 SCU OCOTP driver.\n \n Signed-off-by: Fugang Duan \u003cfugang.duan@nxp.com\u003e\n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190818093345.29647-7-srinivas.kandagatla@linaro.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 5fb812293ad67574146ff1fe9725252aa1dca724\nAuthor: Stefan Wahren \u003cwahrenst@gmx.net\u003e\nDate: Sun Aug 18 10:33:43 2019 +0100\n\n nvmem: mxs-ocotp: update MODULE_AUTHOR() email address\n \n The email address listed in MODULE_AUTHOR() will be disabled in the\n near future. Replace it with my private one.\n \n Signed-off-by: Stefan Wahren \u003cwahrenst@gmx.net\u003e\n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190818093345.29647-6-srinivas.kandagatla@linaro.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit cb6b0a393c4a6d43e29ef476d19313c5fef8b6c7\nAuthor: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\nDate: Sun Aug 18 10:33:42 2019 +0100\n\n nvmem: meson-mx-efuse: allow reading data smaller than word_size\n \n Some Amlogic boards store the Ethernet MAC address inside the eFuse. The\n Ethernet MAC address uses 6 bytes. The existing logic in\n meson_mx_efuse_read() would write beyond the end of the data buffer when\n trying to read data with a size that is not aligned to word_size (4\n bytes on Meson8, Meson8b and Meson8m2).\n \n Calculate the remaining data to copy inside meson_mx_efuse_read() so\n reading 6 bytes doesn\u0027t write beyond the end of the data buffer.\n \n Signed-off-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190818093345.29647-5-srinivas.kandagatla@linaro.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 2ac00e34acfc93af19376f83bb31db1aea1f006a\nAuthor: Stefan Mavrodiev \u003cstefan@olimex.com\u003e\nDate: Sun Aug 18 10:33:41 2019 +0100\n\n nvmem: sunxi_sid: fix A64 SID controller support\n \n Like in H3, A64 SID controller doesn\u0027t return correct data\n when using direct access. It appears that on A64, SID needs\n 8 bytes of word_size.\n \n Workaround is to enable read by registers.\n \n Signed-off-by: Stefan Mavrodiev \u003cstefan@olimex.com\u003e\n Acked-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\n Tested-by: Vasily Khoruzhick \u003canarsoul@gmail.com\u003e\n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190818093345.29647-4-srinivas.kandagatla@linaro.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d93b5d4a2145acf2c67a24e26e17a7908c95ebbc\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Sun Aug 18 10:33:40 2019 +0100\n\n nvmem: imx-ocotp: Add i.MX8MN support\n \n i.MX8MN is a new SoC of i.MX8M series, it is similar to i.MX8MM\n in terms of addressing and clock setup, add support for its fuse\n read/write.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190818093345.29647-3-srinivas.kandagatla@linaro.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 443c396f08a4bb383da19bd9c8b1b413dab0a2ab\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Sun Aug 18 10:33:39 2019 +0100\n\n dt-bindings: imx-ocotp: Add i.MX8MN compatible\n \n Add compatible for i.MX8MN and add i.MX8MM/i.MX8MN to the description.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190818093345.29647-2-srinivas.kandagatla@linaro.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit fcaf3d933926e36a264939e6d3d3cb62b42695b8\nAuthor: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\nDate: Sun Aug 18 10:39:02 2019 +0100\n\n slimbus: fix slim_tid_txn()\n \n fix below issue reported by coccicheck\n ./drivers/slimbus/slimbus.h:440:3-46: duplicated argument to \u0026\u0026 or ||\n \n Looks like this was a typo, SLIM_MSG_MC_REQUEST_CHANGE_VALUE is command\n which requires transaction ID, so fix it, this also fix the warning.\n \n Reported-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190818093902.29993-3-srinivas.kandagatla@linaro.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 04eb94d526423ff082efce61f4f26b0369d0bfdd\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Sun Aug 18 10:39:01 2019 +0100\n\n slimbus: qcom-ngd-ctrl: Add of_node_put() before return\n \n Each iteration of for_each_available_child_of_node puts the previous\n node, but in the case of a return from the middle of the loop, there is\n no put, thus causing a memory leak. Hence add an of_node_put before the\n return in two places.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Reviewed-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190818093902.29993-2-srinivas.kandagatla@linaro.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 596cb85218d7aff659b99bbaa50059f48d00d971\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Aug 13 11:55:48 2019 +0530\n\n drm/vboxvideo: Make structure vbox_fb_helper_funcs constant\n \n The static structure vbox_fb_helper_funcs, of type drm_fb_helper_funcs,\n is used only when it is passed as the third argument to\n drm_fb_helper_fbdev_setup(), which does not modify it. Hence make it\n constant to protect it from unintended modifications.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Reviewed-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813062548.24770-1-nishkadg.linux@gmail.com\n\ncommit 8555b95dbd8b781b8c87076fd50d5b54b76fd092\nAuthor: Fugang Duan \u003cfugang.duan@nxp.com\u003e\nDate: Wed Aug 7 12:03:19 2019 +0800\n\n nvmem: imx: add i.MX8QM platform support\n \n i.MX8QM efuse table has some difference with i.MX8QXP platform,\n so add i.MX8QM platform support.\n \n Signed-off-by: Fugang Duan \u003cfugang.duan@nxp.com\u003e\n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n\ncommit b738a55405a0cab2e197cb0e411f306ec2db5173\nAuthor: Fugang Duan \u003cfugang.duan@nxp.com\u003e\nDate: Wed Aug 7 12:03:20 2019 +0800\n\n dt-bindings: fsl: scu: add new compatible string for ocotp\n \n Add new compatible string \"fsl,imx8qm-scu-ocotp\" into binding\n doc for i.MX8 SCU OCOTP driver.\n \n Signed-off-by: Fugang Duan \u003cfugang.duan@nxp.com\u003e\n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n\ncommit 8a617c0d762c7787d92eefda2025c73d697e6c27\nAuthor: Stefan Wahren \u003cwahrenst@gmx.net\u003e\nDate: Tue Aug 13 20:17:27 2019 +0200\n\n nvmem: mxs-ocotp: update MODULE_AUTHOR() email address\n \n The email address listed in MODULE_AUTHOR() will be disabled in the\n near future. Replace it with my private one.\n \n Signed-off-by: Stefan Wahren \u003cwahrenst@gmx.net\u003e\n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n\ncommit 87b42aa3cb5512036ca98028c6f6b914a96743ba\nAuthor: Sumera Priyadarsini \u003csylphrenadin@gmail.com\u003e\nDate: Sat Aug 17 03:07:02 2019 +0530\n\n staging: gasket: Remove unnecessary line-breaks in function signatures\n \n This patch fixes the function signatures for gasket_read_page_table_size,\n gasket_read_simple_page_table_size, gasket_partition_page_table,\n gasket_config_coherent_allocator to avoid the checkpatch.pl warning:\n \n CHECK: Lines should not end with a \u0027(\u0027\n \n Signed-off-by: Sumera Priyadarsini \u003csylphrenadin@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190816213702.32116-1-sylphrenadin@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 1d819c5406d8a01867df55e90af97ee64c03b037\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Fri Aug 16 15:11:42 2019 +0800\n\n staging: erofs: use common file type conversion\n \n Deduplicate the EROFS file type conversion implementation and\n remove EROFS_FT_* definitions since it\u0027s the same as defined\n by POSIX, let\u0027s follow ext2 as Linus pointed out [1]\n commit e10892189428 (\"ext2: use common file type conversion\").\n \n [1] https://lore.kernel.org/r/CAHk-\u003dwiUs+b\u003diVKM3mVooXgVk7cmmC67KTmnAuL0cd_cMMVAKw@mail.gmail.com/\n \n Reported-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190816071142.8633-1-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 5ee5265674ce69db0973e7de6b30016efa1d78fd\nAuthor: Christina Quast \u003ccontact@christina-quast.de\u003e\nDate: Fri Aug 16 08:59:36 2019 +0200\n\n staging: rtl8192e: rtllib_crypt_ccmp.c: Use crypto API ccm(aes)\n \n Use ccm(aes) aead transform instead of invoking the AES block cipher\n block by block.\n \n Signed-off-by: Christina Quast \u003ccontact@christina-quast.de\u003e\n Reviewed-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190816065936.12214-3-contact@christina-quast.de\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit eb0e7bf3ca94d71247d5bb2b3925183698741d7d\nAuthor: Christina Quast \u003ccontact@christina-quast.de\u003e\nDate: Fri Aug 16 08:59:35 2019 +0200\n\n staging: rtl8192u: ieee80211: ieee80211_crypt_ccmp.c: Use crypto API ccm(aes)\n \n Use ccm(aes) aead transform instead of invoking the AES block cipher\n block by block.\n \n Signed-off-by: Christina Quast \u003ccontact@christina-quast.de\u003e\n Reviewed-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190816065936.12214-2-contact@christina-quast.de\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit ef13e55c27e128b6d84aa939dcc565b1d2ee7724\nAuthor: Rishi Gupta \u003cgupt21@gmail.com\u003e\nDate: Sat Aug 17 12:05:59 2019 +0530\n\n clk: Remove extraneous \u0027for\u0027 word in comments\n \n An extra \u0027for\u0027 word is grammatically incorrect in the comment\n \u0027verifying ops for multi-parent clks\u0027. This commit removes\n this extra for word.\n \n Signed-off-by: Rishi Gupta \u003cgupt21@gmail.com\u003e\n Link: https://lkml.kernel.org/r/1566023759-7880-1-git-send-email-gupt21@gmail.com\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit 36cc34358caf631115e1566485fffed3cf4196b5\nAuthor: Björn Töpel \u003cbjorn.topel@intel.com\u003e\nDate: Thu Aug 15 11:30:14 2019 +0200\n\n xsk: support BPF_EXIST and BPF_NOEXIST flags in XSKMAP\n \n The XSKMAP did not honor the BPF_EXIST/BPF_NOEXIST flags when updating\n an entry. This patch addresses that.\n \n Signed-off-by: Björn Töpel \u003cbjorn.topel@intel.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 0402acd683c678874df6bdbc23530ca07ea19353\nAuthor: Björn Töpel \u003cbjorn.topel@intel.com\u003e\nDate: Thu Aug 15 11:30:13 2019 +0200\n\n xsk: remove AF_XDP socket from map when the socket is released\n \n When an AF_XDP socket is released/closed the XSKMAP still holds a\n reference to the socket in a \"released\" state. The socket will still\n use the netdev queue resource, and block newly created sockets from\n attaching to that queue, but no user application can access the\n fill/complete/rx/tx queues. This results in that all applications need\n to explicitly clear the map entry from the old \"zombie state\"\n socket. This should be done automatically.\n \n In this patch, the sockets tracks, and have a reference to, which maps\n it resides in. When the socket is released, it will remove itself from\n all maps.\n \n Suggested-by: Bruce Richardson \u003cbruce.richardson@intel.com\u003e\n Signed-off-by: Björn Töpel \u003cbjorn.topel@intel.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit c3bbf176fbad5d7470f8a4f311f7c11126ad36c2\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Wed Aug 14 10:37:51 2019 -0700\n\n selftests/bpf: add sockopt clone/inheritance test\n \n Add a test that calls setsockopt on the listener socket which triggers\n BPF program. This BPF program writes to the sk storage and sets\n clone flag. Make sure that sk storage is cloned for a newly\n accepted connection.\n \n We have two cloned maps in the tests to make sure we hit both cases\n in bpf_sk_storage_clone: first element (sk_storage_alloc) and\n non-first element(s) (selem_link_map).\n \n Cc: Martin KaFai Lau \u003ckafai@fb.com\u003e\n Cc: Yonghong Song \u003cyhs@fb.com\u003e\n Acked-by: Martin KaFai Lau \u003ckafai@fb.com\u003e\n Acked-by: Yonghong Song \u003cyhs@fb.com\u003e\n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 9e819ffcfe35c3779166d0fc0088215cf4924c33\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Wed Aug 14 10:37:50 2019 -0700\n\n bpf: sync bpf.h to tools/\n \n Sync new sk storage clone flag.\n \n Cc: Martin KaFai Lau \u003ckafai@fb.com\u003e\n Cc: Yonghong Song \u003cyhs@fb.com\u003e\n Acked-by: Martin KaFai Lau \u003ckafai@fb.com\u003e\n Acked-by: Yonghong Song \u003cyhs@fb.com\u003e\n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 8f51dfc73bf181f2304e1498f55d5f452e060cbe\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Wed Aug 14 10:37:49 2019 -0700\n\n bpf: support cloning sk storage on accept()\n \n Add new helper bpf_sk_storage_clone which optionally clones sk storage\n and call it from sk_clone_lock.\n \n Cc: Martin KaFai Lau \u003ckafai@fb.com\u003e\n Cc: Yonghong Song \u003cyhs@fb.com\u003e\n Acked-by: Martin KaFai Lau \u003ckafai@fb.com\u003e\n Acked-by: Yonghong Song \u003cyhs@fb.com\u003e\n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit b0e4701ce15d0381cdea0643c7f0a35dc529cec2\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Wed Aug 14 10:37:48 2019 -0700\n\n bpf: export bpf_map_inc_not_zero\n \n Rename existing bpf_map_inc_not_zero to __bpf_map_inc_not_zero to\n indicate that it\u0027s caller\u0027s responsibility to do proper locking.\n Create and export bpf_map_inc_not_zero wrapper that properly\n locks map_idr_lock. Will be used in the next commit to\n hold a map while cloning a socket.\n \n Cc: Martin KaFai Lau \u003ckafai@fb.com\u003e\n Cc: Yonghong Song \u003cyhs@fb.com\u003e\n Acked-by: Martin KaFai Lau \u003ckafai@fb.com\u003e\n Acked-by: Yonghong Song \u003cyhs@fb.com\u003e\n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit fae55527ac1164b66bee983a4d82ade2bfedb332\nAuthor: Petar Penkov \u003cppenkov@google.com\u003e\nDate: Fri Aug 16 10:08:25 2019 -0700\n\n selftests/bpf: fix race in test_tcp_rtt test\n \n There is a race in this test between receiving the ACK for the\n single-byte packet sent in the test, and reading the values from the\n map.\n \n This patch fixes this by having the client wait until there are no more\n unacknowledged packets.\n \n Before:\n for i in {1..1000}; do ../net/in_netns.sh ./test_tcp_rtt; \\\n done | grep -c PASSED\n \u003c trimmed error messages \u003e\n 993\n \n After:\n for i in {1..10000}; do ../net/in_netns.sh ./test_tcp_rtt; \\\n done | grep -c PASSED\n 10000\n \n Fixes: b55873984dab (\"selftests/bpf: test BPF_SOCK_OPS_RTT_CB\")\n Signed-off-by: Petar Penkov \u003cppenkov@google.com\u003e\n Reviewed-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 929ffa6e9df0832fdf541ff2d9532e209153c0ec\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Thu Aug 15 22:45:43 2019 -0700\n\n libbpf: relicense bpf_helpers.h and bpf_endian.h\n \n bpf_helpers.h and bpf_endian.h contain useful macros and BPF helper\n definitions essential to almost every BPF program. Which makes them\n useful not just for selftests. To be able to expose them as part of\n libbpf, though, we need them to be dual-licensed as LGPL-2.1 OR\n BSD-2-Clause. This patch updates licensing of those two files.\n \n Acked-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n Acked-by: Hechao Li \u003chechaol@fb.com\u003e\n Acked-by: Martin KaFai Lau \u003ckafai@fb.com\u003e\n Acked-by: Andrey Ignatov \u003crdna@fb.com\u003e\n Acked-by: Yonghong Song \u003cyhs@fb.com\u003e\n Acked-by: Lawrence Brakmo \u003cbrakmo@fb.com\u003e\n Acked-by: Adam Barth \u003carb@fb.com\u003e\n Acked-by: Roman Gushchin \u003cguro@fb.com\u003e\n Acked-by: Josef Bacik \u003cjbacik@fb.com\u003e\n Acked-by: Joe Stringer \u003cjoe@wand.net.nz\u003e\n Acked-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n Acked-by: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\n Acked-by: David Ahern \u003cdsahern@gmail.com\u003e\n Acked-by: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\n Acked-by: Ilya Leoshkevich \u003ciii@linux.ibm.com\u003e\n Acked-by: Lorenz Bauer \u003clmb@cloudflare.com\u003e\n Acked-by: Adrian Ratiu \u003cadrian.ratiu@collabora.com\u003e\n Acked-by: Nikita V. Shirokov \u003ctehnerd@tehnerd.com\u003e\n Acked-by: Willem de Bruijn \u003cwillemb@google.com\u003e\n Acked-by: Petar Penkov \u003cppenkov@google.com\u003e\n Acked-by: Teng Qin \u003cpalmtenor@gmail.com\u003e\n Cc: Michael Holzheu \u003cholzheu@linux.vnet.ibm.com\u003e\n Cc: Naveen N. Rao \u003cnaveen.n.rao@linux.vnet.ibm.com\u003e\n Cc: David S. Miller \u003cdavem@davemloft.net\u003e\n Cc: Michal Rostecki \u003cmrostecki@opensuse.org\u003e\n Cc: John Fastabend \u003cjohn.fastabend@gmail.com\u003e\n Cc: Sargun Dhillon \u003csargun@sargun.me\u003e\n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit c14a9f633d9eb5c0fdd0cb4135b6be978fd538f3\nAuthor: Maxim Mikityanskiy \u003cmaximmi@mellanox.com\u003e\nDate: Wed Aug 14 14:34:06 2019 +0000\n\n net: Don\u0027t call XDP_SETUP_PROG when nothing is changed\n \n Don\u0027t uninstall an XDP program when none is installed, and don\u0027t install\n an XDP program that has the same ID as the one already installed.\n \n dev_change_xdp_fd doesn\u0027t perform any checks in case it uninstalls an\n XDP program. It means that the driver\u0027s ndo_bpf can be called with\n XDP_SETUP_PROG asking to set it to NULL even if it\u0027s already NULL. This\n case happens if the user runs `ip link set eth0 xdp off` when there is\n no XDP program attached.\n \n The symmetrical case is possible when the user tries to set the program\n that is already set.\n \n The drivers typically perform some heavy operations on XDP_SETUP_PROG,\n so they all have to handle these cases internally to return early if\n they happen. This patch puts this check into the kernel code, so that\n all drivers will benefit from it.\n \n Signed-off-by: Maxim Mikityanskiy \u003cmaximmi@mellanox.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit a7bd4018d6424f652f3aabdc5d0d9c64303fa36a\nAuthor: Maxim Mikityanskiy \u003cmaximmi@mellanox.com\u003e\nDate: Wed Aug 14 09:27:23 2019 +0200\n\n net/mlx5e: Add AF_XDP need_wakeup support\n \n This commit adds support for the new need_wakeup feature of AF_XDP. The\n applications can opt-in by using the XDP_USE_NEED_WAKEUP bind() flag.\n When this feature is enabled, some behavior changes:\n \n RX side: If the Fill Ring is empty, instead of busy-polling, set the\n flag to tell the application to kick the driver when it refills the Fill\n Ring.\n \n TX side: If there are pending completions or packets queued for\n transmission, set the flag to tell the application that it can skip the\n sendto() syscall and save time.\n \n The performance testing was performed on a machine with the following\n configuration:\n \n - 24 cores of Intel Xeon E5-2620 v3 @ 2.40 GHz\n - Mellanox ConnectX-5 Ex with 100 Gbit/s link\n \n The results with retpoline disabled:\n \n | without need_wakeup | with need_wakeup |\n |----------------------|----------------------|\n | one core | two cores | one core | two cores |\n -------|----------|-----------|----------|-----------|\n txonly | 20.1 | 33.5 | 29.0 | 34.2 |\n rxdrop | 0.065 | 14.1 | 12.0 | 14.1 |\n l2fwd | 0.032 | 7.3 | 6.6 | 7.2 |\n \n \"One core\" means the application and NAPI run on the same core. \"Two\n cores\" means they are pinned to different cores.\n \n Signed-off-by: Maxim Mikityanskiy \u003cmaximmi@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Reviewed-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 871aa189a69f7bbe6254459d17b78e1cce65c9ae\nAuthor: Maxim Mikityanskiy \u003cmaximmi@mellanox.com\u003e\nDate: Wed Aug 14 09:27:22 2019 +0200\n\n net/mlx5e: Move the SW XSK code from NAPI poll to a separate function\n \n Two XSK tasks are performed during NAPI polling, that are not bound to\n hardware interrupts: TXing packets and polling for frames in the Fill\n Ring. They are special in a way that the hardware doesn\u0027t know about\n these tasks, so it doesn\u0027t trigger interrupts if there is still some\n work to be done, it\u0027s our driver\u0027s responsibility to ensure NAPI will be\n rescheduled if needed.\n \n Create a new function to handle these tasks and move the corresponding\n code from mlx5e_napi_poll to the new function to improve modularity and\n prepare for the changes in the following patch.\n \n Signed-off-by: Maxim Mikityanskiy \u003cmaximmi@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Reviewed-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 46738f73ea4f94906491cb8399bdeafa5b9cc5ec\nAuthor: Magnus Karlsson \u003cmagnus.karlsson@intel.com\u003e\nDate: Wed Aug 14 09:27:21 2019 +0200\n\n samples/bpf: add use of need_wakeup flag in xdpsock\n \n This commit adds using the need_wakeup flag to the xdpsock sample\n application. It is turned on by default as we think it is a feature\n that seems to always produce a performance benefit, if the application\n has been written taking advantage of it. It can be turned off in the\n sample app by using the \u0027-m\u0027 command line option.\n \n The txpush and l2fwd sub applications have also been updated to\n support poll() with multiple sockets.\n \n Signed-off-by: Magnus Karlsson \u003cmagnus.karlsson@intel.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit a4500432c2587cb2ae7554537886a4516ff2e7aa\nAuthor: Magnus Karlsson \u003cmagnus.karlsson@intel.com\u003e\nDate: Wed Aug 14 09:27:20 2019 +0200\n\n libbpf: add support for need_wakeup flag in AF_XDP part\n \n This commit adds support for the new need_wakeup flag in AF_XDP. The\n xsk_socket__create function is updated to handle this and a new\n function is introduced called xsk_ring_prod__needs_wakeup(). This\n function can be used by the application to check if Rx and/or Tx\n processing needs to be explicitly woken up.\n \n Signed-off-by: Magnus Karlsson \u003cmagnus.karlsson@intel.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 5c129241e2de79f09cb4e50bbca09e1c14ad787d\nAuthor: Magnus Karlsson \u003cmagnus.karlsson@intel.com\u003e\nDate: Wed Aug 14 09:27:19 2019 +0200\n\n ixgbe: add support for AF_XDP need_wakeup feature\n \n This patch adds support for the need_wakeup feature of AF_XDP. If the\n application has told the kernel that it might sleep using the new bind\n flag XDP_USE_NEED_WAKEUP, the driver will then set this flag if it has\n no more buffers on the NIC Rx ring and yield to the application. For\n Tx, it will set the flag if it has no outstanding Tx completion\n interrupts and return to the application.\n \n Signed-off-by: Magnus Karlsson \u003cmagnus.karlsson@intel.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 3d0c5f1cd268fdd7d2e75eaf3e0d2fe460ac748b\nAuthor: Magnus Karlsson \u003cmagnus.karlsson@intel.com\u003e\nDate: Wed Aug 14 09:27:18 2019 +0200\n\n i40e: add support for AF_XDP need_wakeup feature\n \n This patch adds support for the need_wakeup feature of AF_XDP. If the\n application has told the kernel that it might sleep using the new bind\n flag XDP_USE_NEED_WAKEUP, the driver will then set this flag if it has\n no more buffers on the NIC Rx ring and yield to the application. For\n Tx, it will set the flag if it has no outstanding Tx completion\n interrupts and return to the application.\n \n Signed-off-by: Magnus Karlsson \u003cmagnus.karlsson@intel.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 77cd0d7b3f257fd0e3096b4fdcff1a7d38e99e10\nAuthor: Magnus Karlsson \u003cmagnus.karlsson@intel.com\u003e\nDate: Wed Aug 14 09:27:17 2019 +0200\n\n xsk: add support for need_wakeup flag in AF_XDP rings\n \n This commit adds support for a new flag called need_wakeup in the\n AF_XDP Tx and fill rings. When this flag is set, it means that the\n application has to explicitly wake up the kernel Rx (for the bit in\n the fill ring) or kernel Tx (for bit in the Tx ring) processing by\n issuing a syscall. Poll() can wake up both depending on the flags\n submitted and sendto() will wake up tx processing only.\n \n The main reason for introducing this new flag is to be able to\n efficiently support the case when application and driver is executing\n on the same core. Previously, the driver was just busy-spinning on the\n fill ring if it ran out of buffers in the HW and there were none on\n the fill ring. This approach works when the application is running on\n another core as it can replenish the fill ring while the driver is\n busy-spinning. Though, this is a lousy approach if both of them are\n running on the same core as the probability of the fill ring getting\n more entries when the driver is busy-spinning is zero. With this new\n feature the driver now sets the need_wakeup flag and returns to the\n application. The application can then replenish the fill queue and\n then explicitly wake up the Rx processing in the kernel using the\n syscall poll(). For Tx, the flag is only set to one if the driver has\n no outstanding Tx completion interrupts. If it has some, the flag is\n zero as it will be woken up by a completion interrupt anyway.\n \n As a nice side effect, this new flag also improves the performance of\n the case where application and driver are running on two different\n cores as it reduces the number of syscalls to the kernel. The kernel\n tells user space if it needs to be woken up by a syscall, and this\n eliminates many of the syscalls.\n \n This flag needs some simple driver support. If the driver does not\n support this, the Rx flag is always zero and the Tx flag is always\n one. This makes any application relying on this feature default to the\n old behaviour of not requiring any syscalls in the Rx path and always\n having to call sendto() in the Tx path.\n \n For backwards compatibility reasons, this feature has to be explicitly\n turned on using a new bind flag (XDP_USE_NEED_WAKEUP). I recommend\n that you always turn it on as it so far always have had a positive\n performance impact.\n \n The name and inspiration of the flag has been taken from io_uring by\n Jens Axboe. Details about this feature in io_uring can be found in\n http://kernel.dk/io_uring.pdf, section 8.3.\n \n Signed-off-by: Magnus Karlsson \u003cmagnus.karlsson@intel.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 9116e5e2b1fff71dce501d971e86a3695acc3dba\nAuthor: Magnus Karlsson \u003cmagnus.karlsson@intel.com\u003e\nDate: Wed Aug 14 09:27:16 2019 +0200\n\n xsk: replace ndo_xsk_async_xmit with ndo_xsk_wakeup\n \n This commit replaces ndo_xsk_async_xmit with ndo_xsk_wakeup. This new\n ndo provides the same functionality as before but with the addition of\n a new flags field that is used to specifiy if Rx, Tx or both should be\n woken up. The previous ndo only woke up Tx, as implied by the\n name. The i40e and ixgbe drivers (which are all the supported ones)\n are updated with this new interface.\n \n This new ndo will be used by the new need_wakeup functionality of XDP\n sockets that need to be able to wake up both Rx and Tx driver\n processing.\n \n Signed-off-by: Magnus Karlsson \u003cmagnus.karlsson@intel.com\u003e\n Acked-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 94e18382003c87fa60004ad577dd6f532d4dae99\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Sat Aug 17 20:54:51 2019 +0200\n\n net: stmmac: selftests: Add selftest for VLAN TX Offload\n \n Add 2 new selftests for VLAN Insertion offloading. Tests are for inner\n and outer VLAN offloading.\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 30d932279dc226d4471c823618dba5c1ea1e681e\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Sat Aug 17 20:54:50 2019 +0200\n\n net: stmmac: Add support for VLAN Insertion Offload\n \n Adds the logic to insert a given VLAN ID in a packet. This is offloaded\n to HW and its descriptor based. For now, only XGMAC implements the\n necessary callbacks.\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 81b945aea0ead6fd1ed9a57c01b0b8699efea426\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Sat Aug 17 20:54:49 2019 +0200\n\n net: stmmac: xgmac: Add EEE support\n \n Add support for EEE in XGMAC cores by implementing the necessary\n callbacks.\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 8180d5797a1d4a93e3a1a653da9742b65ef98d77\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Sat Aug 17 20:54:48 2019 +0200\n\n net: stmmac: selftests: Add tests for SA Insertion/Replacement\n \n Add 4 new tests:\n - SA Insertion (register based)\n - SA Insertion (descriptor based)\n - SA Replacament (register based)\n - SA Replacement (descriptor based)\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 8000ddc0eceb8e504cb9f95a1d72af0a8d2fc76e\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Sat Aug 17 20:54:47 2019 +0200\n\n net: stmmac: Add support for SA Insertion/Replacement in XGMAC cores\n \n Add the support for Source Address Insertion and Replacement in XGMAC\n cores. Two methods are supported: Descriptor based and register based.\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit bfc56530697d939d07608d221451cd875aefa295\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Sat Aug 17 20:54:46 2019 +0200\n\n net: stmmac: Add ethtool register dump for XGMAC cores\n \n Add the ethtool interface to dump the register map in XGMAC cores.\n \n Changes from v2:\n - Remove uneeded memset (Jakub)\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 95eaf3cd0a9044b2cbfe1081692e568879363cc7\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Sat Aug 17 20:54:45 2019 +0200\n\n net: stmmac: dwxgmac: Add Flexible PPS support\n \n Add the support for Flexible PPS in XGMAC cores.\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b5418e130e5f2529b6db82725658bfc6d5485a6d\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Sat Aug 17 20:54:44 2019 +0200\n\n net: stmmac: Add a counter for Split Header packets\n \n Add a counter that increments each time a packet with split header is\n received.\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 67afd6d1cfdf0d0461fe3c4c922447f3e9b1c6ee\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Sat Aug 17 20:54:43 2019 +0200\n\n net: stmmac: Add Split Header support and enable it in XGMAC cores\n \n Add the support for Split Header feature in the RX path and enable it in\n XGMAC cores.\n \n This does not impact neither beneficts bandwidth but it does reduces CPU\n usage because without the feature all the entire packet is memcpy\u0027ed,\n while that with the feature only the header is.\n \n With Split Header disabled \u0027perf stat -d\u0027 gives:\n 86870.624945 task-clock (msec) # 0.429 CPUs utilized\n 1073352 context-switches # 0.012 M/sec\n 1 cpu-migrations # 0.000 K/sec\n 213 page-faults # 0.002 K/sec\n 327113872376 cycles # 3.766 GHz (62.53%)\n 56618161216 instructions # 0.17 insn per cycle (75.06%)\n 10742205071 branches # 123.658 M/sec (75.36%)\n 584309242 branch-misses # 5.44% of all branches (75.19%)\n 17594787965 L1-dcache-loads # 202.540 M/sec (74.88%)\n 4003773131 L1-dcache-load-misses # 22.76% of all L1-dcache hits (74.89%)\n 1313301468 LLC-loads # 15.118 M/sec (49.75%)\n 355906510 LLC-load-misses # 27.10% of all LL-cache hits (49.92%)\n \n With Split Header enabled \u0027perf stat -d\u0027 gives:\n 49324.456539 task-clock (msec) # 0.245 CPUs utilized\n 2542387 context-switches # 0.052 M/sec\n 1 cpu-migrations # 0.000 K/sec\n 213 page-faults # 0.004 K/sec\n 177092791469 cycles # 3.590 GHz (62.30%)\n 68555756017 instructions # 0.39 insn per cycle (75.16%)\n 12697019382 branches # 257.418 M/sec (74.81%)\n 442081897 branch-misses # 3.48% of all branches (74.79%)\n 20337958358 L1-dcache-loads # 412.330 M/sec (75.46%)\n 3820210140 L1-dcache-load-misses # 18.78% of all L1-dcache hits (75.35%)\n 1257719198 LLC-loads # 25.499 M/sec (49.73%)\n 685543923 LLC-load-misses # 54.51% of all LL-cache hits (49.86%)\n \n Changes from v2:\n - Reword commit message (Jakub)\n Changes from v1:\n - Add performance info (David)\n - Add misssing dma_sync_single_for_device()\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c887e02a938d7cae9837322448de0af2dd75ee1d\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Sat Aug 17 20:54:42 2019 +0200\n\n net: stmmac: xgmac: Correctly return that RX descriptor is not last one\n \n Return the correct value when RX descriptor is not the last one.\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ec222003bd948de8f3ffbca522d328af1a8452ad\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Sat Aug 17 20:54:41 2019 +0200\n\n net: stmmac: Prepare to add Split Header support\n \n In order to add Split Header support, stmmac_rx() needs to take into\n account that packet may be split accross multiple descriptors.\n \n Refactor the logic of this function in order to support this scenario.\n \n Changes from v2:\n - Fixup if condition detection (Jakub)\n - Don\u0027t stop NAPI with unfinished packet (Jakub)\n - Use napi_alloc_skb() (Jakub)\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 25e80cd05ff81e51750106f1ae12364449ec38e7\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Sat Aug 17 20:54:40 2019 +0200\n\n net: stmmac: Get correct timestamp values from XGMAC\n \n TX Timestamp in XGMAC comes from MAC instead of descriptors. Implement\n this in a new callback.\n \n Also, RX Timestamp in XGMAC must be cheked against corruption and we need\n a barrier to make sure that descriptor fields are read correctly.\n \n Changes from v2:\n - Rework return code check (Jakub)\n Changes from v1:\n - Rework the get timestamp function (David)\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 95766451bfb82f972bf3fea93fc6e91a904cf624\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sat Aug 17 16:28:25 2019 +0300\n\n Documentation: Add a section for devlink-trap testing\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b3cb7df9ecb5d73a7a116a50a92ad992f20798cf\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sat Aug 17 16:28:24 2019 +0300\n\n selftests: devlink_trap: Add test cases for devlink-trap\n \n Add test cases for devlink-trap on top of the netdevsim implementation.\n \n The tests focus on the devlink-trap core infrastructure and user space\n API. They test both good and bad flows and also dismantle of the netdev\n and devlink device used to report trapped packets.\n \n This allows device drivers to focus their tests on device-specific\n functionality.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a054c8d90bac321d804b789a82f9238f5a80a668\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sat Aug 17 16:28:23 2019 +0300\n\n selftests: forwarding: devlink_lib: Add devlink-trap helpers\n \n Add helpers to interact with devlink-trap, such as setting the action of\n a trap and retrieving statistics.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit bc030d9c91c3764a143fc3ac4516bbfd330b41ed\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sat Aug 17 16:28:22 2019 +0300\n\n selftests: forwarding: devlink_lib: Allow tests to define devlink device\n \n For tests that create their network interfaces dynamically or do not use\n interfaces at all (as with netdevsim) it is useful to define their own\n devlink device instead of deriving it from the first network interface.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9e087457048884c57c51a32b52d8aa1d9b0d7385\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sat Aug 17 16:28:21 2019 +0300\n\n Documentation: Add description of netdevsim traps\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit da58f90f11f597520f226caff1d3cfc115abedc9\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sat Aug 17 16:28:20 2019 +0300\n\n netdevsim: Add devlink-trap support\n \n Have netdevsim register its trap groups and traps with devlink during\n initialization and periodically report trapped packets to devlink core.\n \n Since netdevsim is not a real device, the trapped packets are emulated\n using a workqueue that periodically reports a UDP packet with a random\n 5-tuple from each active packet trap and from each running netdev.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f3047ca01f12bf997ef199710b51490b0f750b34\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sat Aug 17 16:28:19 2019 +0300\n\n Documentation: Add devlink-trap documentation\n \n Add initial documentation of the devlink-trap mechanism, explaining the\n background, motivation and the semantics of the interface.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 391203ab11df9b23cd0b867122bccbe33fe16f02\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sat Aug 17 16:28:18 2019 +0300\n\n devlink: Add generic packet traps and groups\n \n Add generic packet traps and groups that can report dropped packets as\n well as exceptions such as TTL error.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 0f420b6c52e9799f664429e739a421fb4c5527a3\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sat Aug 17 16:28:17 2019 +0300\n\n devlink: Add packet trap infrastructure\n \n Add the basic packet trap infrastructure that allows device drivers to\n register their supported packet traps and trap groups with devlink.\n \n Each driver is expected to provide basic information about each\n supported trap, such as name and ID, but also the supported metadata\n types that will accompany each packet trapped via the trap. The\n currently supported metadata type is just the input port, but more will\n be added in the future. For example, output port and traffic class.\n \n Trap groups allow users to set the action of all member traps. In\n addition, users can retrieve per-group statistics in case per-trap\n statistics are too narrow. In the future, the trap group object can be\n extended with more attributes, such as policer settings which will limit\n the amount of traffic generated by member traps towards the CPU.\n \n Beside registering their packet traps with devlink, drivers are also\n expected to report trapped packets to devlink along with relevant\n metadata. devlink will maintain packets and bytes statistics for each\n packet trap and will potentially report the trapped packet with its\n metadata to user space via drop monitor netlink channel.\n \n The interface towards the drivers is simple and allows devlink to set\n the action of the trap. Currently, only two actions are supported:\n \u0027trap\u0027 and \u0027drop\u0027. When set to \u0027trap\u0027, the device is expected to provide\n the sole copy of the packet to the driver which will pass it to devlink.\n When set to \u0027drop\u0027, the device is expected to drop the packet and not\n send a copy to the driver. In the future, more actions can be added,\n such as \u0027mirror\u0027.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 8e94c3bc922e70225bd35891c8e6002bddd984eb\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sat Aug 17 16:28:16 2019 +0300\n\n drop_monitor: Allow user to start monitoring hardware drops\n \n Drop monitor has start and stop commands, but so far these were only\n used to start and stop monitoring of software drops.\n \n Now that drop monitor can also monitor hardware drops, we should allow\n the user to control these as well.\n \n Do that by adding SW and HW flags to these commands. If no flag is\n specified, then only start / stop monitoring software drops. This is\n done in order to maintain backward-compatibility with existing user\n space applications.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d40e1deb930f4bd51a5214983e50aafc26db686e\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sat Aug 17 16:28:15 2019 +0300\n\n drop_monitor: Add support for summary alert mode for hardware drops\n \n In summary alert mode a notification is sent with a list of recent drop\n reasons and a count of how many packets were dropped due to this reason.\n \n To avoid expensive operations in the context in which packets are\n dropped, each CPU holds an array whose number of entries is the maximum\n number of drop reasons that can be encoded in the netlink notification.\n Each entry stores the drop reason and a count. When a packet is dropped\n the array is traversed and a new entry is created or the count of an\n existing entry is incremented.\n \n Later, in process context, the array is replaced with a newly allocated\n copy and the old array is encoded in a netlink notification. To avoid\n breaking user space, the notification includes the ancillary header,\n which is \u0027struct net_dm_alert_msg\u0027 with number of entries set to \u00270\u0027.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 5e58109b1ea454b93e455e0e8fc0bc4c226b8c0a\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sat Aug 17 16:28:14 2019 +0300\n\n drop_monitor: Add support for packet alert mode for hardware drops\n \n In a similar fashion to software drops, extend drop monitor to send\n netlink events when packets are dropped by the underlying hardware.\n \n The main difference is that instead of encoding the program counter (PC)\n from which kfree_skb() was called in the netlink message, we encode the\n hardware trap name. The two are mostly equivalent since they should both\n help the user understand why the packet was dropped.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 80cebed85c88e71fb435b9eb5948548f10f15cba\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sat Aug 17 16:28:13 2019 +0300\n\n drop_monitor: Consider all monitoring states before performing configuration\n \n The drop monitor configuration (e.g., alert mode) is global, but user\n will be able to enable monitoring of only software or hardware drops.\n \n Therefore, ensure that monitoring of both software and hardware drops are\n disabled before allowing drop monitor configuration to take place.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit edd3d0074c256848bbf5805350b39d40b1e2bf2b\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sat Aug 17 16:28:12 2019 +0300\n\n drop_monitor: Add basic infrastructure for hardware drops\n \n Export a function that can be invoked in order to report packets that\n were dropped by the underlying hardware along with metadata.\n \n Subsequent patches will add support for the different alert modes.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit cac1174fa17f2cf5d504cbe00c0053f0b9e3b58d\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sat Aug 17 16:28:11 2019 +0300\n\n drop_monitor: Initialize hardware per-CPU data\n \n Like software drops, hardware drops also need the same type of per-CPU\n data. Therefore, initialize it during module initialization and\n de-initialize it during module exit.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9b63f57d4ab981e18e35def3349040527ead93c3\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sat Aug 17 16:28:10 2019 +0300\n\n drop_monitor: Move per-CPU data init/fini to separate functions\n \n Currently drop monitor only reports software drops to user space, but\n subsequent patches are going to add support for hardware drops.\n \n Like software drops, the per-CPU data of hardware drops needs to be\n initialized and de-initialized upon module initialization and exit. To\n avoid code duplication, break this code into separate functions, so that\n these could be re-used for hardware drops.\n \n No functional changes intended.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1bc844ee0faa1b92e3ede00bdd948021c78d7088\nAuthor: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\nDate: Sat Aug 17 14:22:13 2019 +0300\n\n net: bridge: mdb: allow add/delete for host-joined groups\n \n Currently this is needed only for user-space compatibility, so similar\n object adds/deletes as the dumped ones would succeed. Later it can be\n used for L2 mcast MAC add/delete.\n \n v3: fix compiler warning (DaveM)\n v2: don\u0027t send a notification when used from user-space, arm the group\n timer if no ports are left after host entry del\n \n Signed-off-by: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit e77b0c84e33c766728991fb637ce0ffe41be2fb1\nAuthor: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\nDate: Sat Aug 17 14:22:12 2019 +0300\n\n net: bridge: mdb: dump host-joined entries as well\n \n Currently we dump only the port mdb entries but we can have host-joined\n entries on the bridge itself and they should be treated as normal temp\n mdbs, they\u0027re already notified:\n $ bridge monitor all\n [MDB]dev br0 port br0 grp ff02::8 temp\n \n The group will not be shown in the bridge mdb output, but it takes 1 slot\n and it\u0027s timing out. If it\u0027s only host-joined then the mdb show output\n can even be empty.\n \n After this patch we show the host-joined groups:\n $ bridge mdb show\n dev br0 port br0 grp ff02::8 temp\n \n Signed-off-by: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6545916ed9f4b805515a7546908a6b2ff2d060b5\nAuthor: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\nDate: Sat Aug 17 14:22:11 2019 +0300\n\n net: bridge: mdb: factor out mdb filling\n \n We have to factor out the mdb fill portion in order to re-use it later for\n the bridge mdb entries. No functional changes intended.\n \n Signed-off-by: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f59783f5bb1ac6c59d4ba35430df1b89caeadab1\nAuthor: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\nDate: Sat Aug 17 14:22:10 2019 +0300\n\n net: bridge: mdb: move vlan comments\n \n Trivial patch to move the vlan comments in their proper places above the\n vid 0 checks.\n \n Signed-off-by: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4b9cb2a5ceedd7f715d92570b773fad024ca9950\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Sat Aug 17 12:30:39 2019 +0200\n\n net: phy: remove genphy_config_init\n \n Now that all users have been removed we can remove genphy_config_init.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 00843d99ce15348775db3a9d043d4f7b714be73b\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Sat Aug 17 12:29:54 2019 +0200\n\n net: dsa: remove calls to genphy_config_init\n \n Supported PHY features are either auto-detected or explicitly set.\n In both cases calling genphy_config_init isn\u0027t needed.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c227ce4423855bddb41a05ad25a93f13b96c89bd\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Sat Aug 17 12:29:25 2019 +0200\n\n net: phy: remove calls to genphy_config_init\n \n Supported PHY features are either auto-detected or explicitly set.\n In both cases calling genphy_config_init isn\u0027t needed. All that\n genphy_config_init does is removing features that are set as\n supported but can\u0027t be auto-detected. Basically it duplicates the\n code in genphy_read_abilities. Therefore remove such calls from\n all PHY drivers.\n \n v2:\n - remove call also from new adin PHY driver\n v3:\n - pass NULL as config_init function pointer for dp83848\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f2cb60e9a3881e679465f84140754bc9d29956ea\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Sat Aug 17 16:30:22 2019 +0100\n\n dma-fence: Store the timestamp in the same union as the cb_list\n \n The timestamp and the cb_list are mutually exclusive, the cb_list can\n only be added to prior to being signaled (and once signaled we drain),\n while the timestamp is only valid upon being signaled. Both the\n timestamp and the cb_list are only valid while the fence is alive, and\n as soon as no references are held can be replaced by the rcu_head.\n \n By reusing the union for the timestamp, we squeeze the base dma_fence\n struct to 64 bytes on x86-64.\n \n v2: Sort the union chronologically\n \n Suggested-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Christian König \u003cchristian.koenig@amd.com\u003e\n Acked-by: Christian König \u003cchristian.koenig@amd.com\u003e.\n Link: https://patchwork.freedesktop.org/patch/msgid/20190817153022.5749-1-chris@chris-wilson.co.uk\n\ncommit 0fc89b6802ba1fcc561b0c906e0cefd384e3b2e5\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Sat Aug 17 16:23:00 2019 +0100\n\n dma-fence: Simply wrap dma_fence_signal_locked with dma_fence_signal\n \n Currently dma_fence_signal() tries to avoid the spinlock and only takes\n it if absolutely required to walk the callback list. However, to allow\n for some users to surreptitiously insert lazy signal callbacks that\n do not depend on enabling the signaling mechanism around every fence,\n we always need to notify the callbacks on signaling. As such, we will\n always need to take the spinlock and dma_fence_signal() effectively\n becomes a clone of dma_fence_signal_locked().\n \n v2: Update the test_and_set_bit() before entering the spinlock.\n v3: Drop the test_[and_set]_bit() before the spinlock, it\u0027s a caller\n error so expected to be very unlikely.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Christian König \u003cchristian.koenig@amd.com\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190817152300.5370-1-chris@chris-wilson.co.uk\n\ncommit f86a802bf3a7c28e4c4ea032fa2e61398dbccd5d\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Sat Aug 17 15:47:34 2019 +0100\n\n dma-fence: Avoid list_del during fence-\u003ecb_list iteration\n \n Before we notify the fence signal callback, we remove the cb from the\n list. However, since we are processing the entire list from underneath\n the spinlock, we do not need to individual delete each element, but can\n simply reset the link and the entire list.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190817144736.7826-4-chris@chris-wilson.co.uk\n\ncommit 4fe3997a68f3300c73adc196ff33a952febe6974\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Sat Aug 17 15:47:33 2019 +0100\n\n dma-fence: Shrink size of struct dma_fence\n \n Rearrange the couple of 32-bit atomics hidden amongst the field of\n pointers that unnecessarily caused the compiler to insert some padding,\n shrinks the size of the base struct dma_fence from 80 to 72 bytes on\n x86-64.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190817144736.7826-3-chris@chris-wilson.co.uk\n\ncommit b0baf85bb27ab028e726619f9162498f193cacbe\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Sat Aug 17 13:21:24 2019 +0100\n\n drm/panel: tpo-td043mtea1: remove redundant assignment\n \n Variable val is initialized to a value in a for-loop that is\n never read and hence it is redundant. Remove it.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190817122124.29650-1-colin.king@canonical.com\n\ncommit 0075a20a6dd150e12d9fcb8f22eb7bc4ba241baf\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Sat Aug 17 13:11:44 2019 +0000\n\n drm/i915/uc: Never fail on uC preparation step\n \n Let\u0027s wait with decision about importance of uC failure to\n hardware initialization step.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190817131144.26884-4-michal.wajdeczko@intel.com\n\ncommit 4a600cb707e1249472bd9c1c7d63bb971fc6675d\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Sat Aug 17 13:11:43 2019 +0000\n\n drm/i915/uc: Cleanup fw fetch on every GuC/HuC init failure\n \n Be consistent and always perform fw fetch cleanup in GuC/HuC specific\n init functions on every failure. Also while converting firmware\n status to error, stop treating SELECTED as non-error, as long term\n we should not see it.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190817131144.26884-3-michal.wajdeczko@intel.com\n\ncommit 5bde5a4572e122a68406ddb1d0798988cf9c2b0e\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Sat Aug 17 13:11:42 2019 +0000\n\n drm/i915/uc: Cleanup fw fetch only if it was successful\n \n We can rely on firmware status AVAILABLE to determine if any\n firmware cleanup is required. Also don\u0027t unconditionally reset\n fw status to SELECTED as we will loose MISSING/ERROR codes.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190817131144.26884-2-michal.wajdeczko@intel.com\n\ncommit 2a813f1392205654aea28098f3bcc3e6e2478fa5\nAuthor: Simon Wunderlich \u003csw@simonwunderlich.de\u003e\nDate: Fri Aug 16 10:26:26 2019 +0200\n\n batman-adv: Add Sven to MAINTAINERS file\n \n Sven is taking care of tracking our patches and merging most of them in\n our tree. Let\u0027s add him to the MAINTAINERS file so he will get all\n patch e-mails.\n \n Signed-off-by: Simon Wunderlich \u003csw@simonwunderlich.de\u003e\n Acked-by: Antonio Quartulli \u003ca@unstable.cc\u003e\n\ncommit bba10c5cab4ddd8725a7998e064fc72c9770c667\nAuthor: Rahul Tanwar \u003crahul.tanwar@linux.intel.com\u003e\nDate: Fri Aug 16 16:18:57 2019 +0800\n\n x86/cpu: Use constant definitions for CPU models\n \n Replace model numbers with their respective macro definitions when\n comparing CPU models.\n \n Suggested-by: Andy Shevchenko \u003candriy.shevchenko@intel.com\u003e\n Signed-off-by: Rahul Tanwar \u003crahul.tanwar@linux.intel.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Cc: alan@linux.intel.com\n Cc: cheol.yong.kim@intel.com\n Cc: Hans de Goede \u003chdegoede@redhat.com\u003e\n Cc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: qi-ming.wu@intel.com\n Cc: \"Rafael J. Wysocki\" \u003crafael.j.wysocki@intel.com\u003e\n Cc: Ricardo Neri \u003cricardo.neri-calderon@linux.intel.com\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Tony Luck \u003ctony.luck@intel.com\u003e\n Cc: x86-ml \u003cx86@kernel.org\u003e\n Link: https://lkml.kernel.org/r/f7a0e142faa953a53d5f81f78055e1b3c793b134.1565940653.git.rahul.tanwar@linux.intel.com\n\ncommit 9559c8751385ba2d9ff7fc1150694b4130f8a6d9\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Sat Aug 17 08:37:11 2019 +0100\n\n drm/i915/selftests: Check the context size\n \n Add a redzone to our context image and check the HW does not write into\n after a context save, to verify that we have the correct context size.\n (This does vary with feature bits, so test with a live setup that should\n match how we run userspace.)\n \n v2: Check the redzone on every context unpin\n v3: Use a kernel context to prevent loading garbage for ringbuffer\n submission\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190817073711.5897-1-chris@chris-wilson.co.uk\n\ncommit eb7c022ddc0c26c2969d04d2dd4dcc799dbe159c\nAuthor: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\nDate: Fri Aug 16 12:47:54 2019 +0300\n\n drm/i915/gtt: Fold gen8 insertions into one\n \n As we give page directory pointer (lvl 3) structure\n for pte insertion, we can fold both versions into\n one function by teaching it to get pdp regardless\n of top level.\n \n v2: naming and asserts (Chris)\n \n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816094754.26492-1-mika.kuoppala@linux.intel.com\n\ncommit 1c1daa47a0ae07a9f28aa2b07c6a8a9b1ea33aeb\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Jul 29 09:23:23 2019 -0700\n\n xfs: refactor the ioend merging code\n \n Introduce two nicely abstracted helper, which can be moved to the\n iomap code later. Also use list_pop_entry and list_first_entry_or_null\n to simplify the code a bit.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit 3d2d71b4fdff4167e39f011a4bf92ab1606c659b\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Jul 29 09:23:22 2019 -0700\n\n xfs: initialize iomap-\u003eflags in xfs_bmbt_to_iomap\n \n Currently we don\u0027t overwrite the flags field in the iomap in\n xfs_bmbt_to_iomap. This works fine with 0-initialized iomaps on stack,\n but is harmful once we want to be able to reuse an iomap in the\n writeback code. Replace the shared paramter with a set of initial\n flags an thus ensures the flags field is always reinitialized.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit 791ee52eb10b16066edf0b957597d5ce287a0079\nAuthor: Andreas Gruenbacher \u003cagruenba@redhat.com\u003e\nDate: Fri Aug 16 18:52:06 2019 -0700\n\n iomap: Fix trivial typo\n \n Signed-off-by: Andreas Gruenbacher \u003cagruenba@redhat.com\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit 54d895bea43c94f31304d59f82d755b7f4b59e7c\nAuthor: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\nDate: Tue Aug 13 20:09:42 2019 -0700\n\n arm64: dts: qcom: qcs404-evb: Mark WCSS clocks protected\n \n \u00277d0c76bdf227 (\"clk: qcom: Add WCSS gcc clock control for QCS404\")\u0027\n introduces two new clocks to gcc. These are not used before\n clk_disable_unused() and as such the clock framework tries to disable\n them.\n \n But on the EVB these registers are only accessible through TrustZone, so\n these clocks must be marked as \"protected\" to prevent the clock code\n from touching them.\n \n Numerical values are used as the constants are not yet available in a\n common tree.\n \n Reviewed-by: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n Reported-by: Mark Brown \u003cbroonie@kernel.org\u003e\n Reported-by: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit a4d2113e46c1d2ded1bfed9a19fe17b5ab2d584c\nAuthor: Bill Sommerfeld \u003cwsommerfeld@google.com\u003e\nDate: Wed Aug 14 17:10:43 2019 -0700\n\n ipvlan: set hw_enc_features like macvlan\n \n Allow encapsulated packets sent to tunnels layered over ipvlan to use\n offloads rather than forcing SW fallbacks.\n \n Since commit f21e5077010acda73a60 (\"macvlan: add offload features for\n encapsulation\"), macvlan has set dev-\u003ehw_enc_features to include\n everything in dev-\u003efeatures; do likewise in ipvlan.\n \n Signed-off-by: Bill Sommerfeld \u003cwsommerfeld@google.com\u003e\n Acked-by: Mahesh Bandewar \u003cmaheshb@google.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 356c484822e6ac97dad6947c083fc45d07332c7c\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Fri Aug 16 20:56:58 2019 +0000\n\n drm/i915/uc: Add explicit DISABLED state for firmware\n \n We really need to have separate NOT_SUPPORTED state (for\n lack of hardware support) and DISABLED state (to indicate\n user decision) as we will have to take special steps even\n if GuC firmware is now disabled but hardware exists and\n could have been previously used.\n \n v2: fix logic (Chris/CI)\n v3: use proper check to avoid probe failure (CI)\n v4: explain status transitions (Chris)\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816205658.15020-1-michal.wajdeczko@intel.com\n\ncommit 87caaaf2d19d4656924241aadb612782c296928f\nAuthor: David Ahern \u003cdsahern@gmail.com\u003e\nDate: Wed Aug 14 10:11:51 2019 -0700\n\n selftests: Fix get_ifidx and callers in nettest.c\n \n Dan reported:\n \n The patch acda655fefae: \"selftests: Add nettest\" from Aug 1, 2019,\n leads to the following static checker warning:\n \n ./tools/testing/selftests/net/nettest.c:1690 main()\n warn: unsigned \u0027tmp\u0027 is never less than zero.\n \n ./tools/testing/selftests/net/nettest.c\n 1680 case \u00271\u0027:\n 1681 args.has_expected_raddr \u003d 1;\n 1682 if (convert_addr(\u0026args, optarg,\n 1683 ADDR_TYPE_EXPECTED_REMOTE))\n 1684 return 1;\n 1685\n 1686 break;\n 1687 case \u00272\u0027:\n 1688 if (str_to_uint(optarg, 0, 0x7ffffff, \u0026tmp) !\u003d 0) {\n 1689 tmp \u003d get_ifidx(optarg);\n 1690 if (tmp \u003c 0) {\n \n \"tmp\" is unsigned so it can\u0027t be negative. Also all the callers assume\n that get_ifidx() returns negatives on error but it looks like it really\n returns zero on error so it\u0027s a bit unclear to me.\n \n Update get_ifidx to return -1 on errors and cleanup callers of it.\n \n Fixes: acda655fefae (\"selftests: Add nettest\")\n Reported-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Signed-off-by: David Ahern \u003cdsahern@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 523c620254b1dd406924c6270790be0e05b2b997\nAuthor: Bin Meng \u003cbmeng.cn@gmail.com\u003e\nDate: Sun Jul 28 02:30:19 2019 -0700\n\n dt-bindings: pci: pci-msi: Correct the unit-address of the pci node name\n \n The unit-address must match the first address specified in the\n reg property of the node.\n \n Signed-off-by: Bin Meng \u003cbmeng.cn@gmail.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 0d3d343560bad8e1b7879fe94251cfe731a2dd13\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:25:14 2019 +0200\n\n genirq: remove the is_affinity_mask_valid hook\n \n This override was only used by the ia64 SGI SN2 platform, which is\n gone now.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-29-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 6c13bb1d58d7a6002eb44e1595897250f0b0eec6\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:25:13 2019 +0200\n\n ia64: remove CONFIG_SWIOTLB ifdefs\n \n CONFIG_SWIOTLB is now unconditionally selected on ia64, so remove the\n ifdefs.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-28-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit df41017eafd267c08acbfff99d34e4f96bbfbc92\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:25:12 2019 +0200\n\n ia64: remove support for machvecs\n \n The only thing remaining of the machvecs is a few checks if we are\n running on an SGI UV system. Replace those with the existing\n is_uv_system() check that has been rewritten to simply check the\n OEM ID directly.\n \n That leaves us with a generic kernel that is as fast as the previous\n DIG/ZX1/UV kernels, but can support all hardware. Support for UV\n and the HP SBA IOMMU is now optional based on new config options.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-27-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 4cb3b44d6b7185611a4f0fd613c05a9e57d7ad94\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Thu Aug 15 18:23:43 2019 -0700\n\n drm/i915: Wrappers for display register waits\n \n To reduce the number of explicit dev_priv-\u003euncore calls in the display\n code ahead of the introduction of dev_priv-\u003ede_uncore, this patch\n introduces a wrapper for one of the main usages of it, the register\n waits. When we transition to the new uncore, we can just update the\n wrapper to point to the appropriate structure.\n \n Since the vast majority of waits are on a set or clear of a bit or mask,\n add set \u0026 clear flavours of the wrapper to simplify the code.\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816012343.36433-7-daniele.ceraolospurio@intel.com\n\ncommit 2033f6858970b98c18bed4d5ae68f43d17400abc\nAuthor: Corey Minyard \u003ccminyard@mvista.com\u003e\nDate: Fri Aug 16 16:13:42 2019 -0500\n\n ipmi: Free receive messages when in an oops\n \n If the driver handles a response in an oops, it was just ignoring\n the message. However, the IPMI watchdog timer was counting on the\n free happening to know when panic-time messages were complete. So\n free it in all cases.\n \n Signed-off-by: Corey Minyard \u003ccminyard@mvista.com\u003e\n\ncommit 75a037f3604ceb781ae23167e0cdfbe5d71533d7\nAuthor: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\nDate: Wed Jul 31 13:27:05 2019 -0700\n\n f2fs: fix livelock in swapfile writes\n \n This patch fixes livelock in the below call path when writing swap pages.\n \n [46374.617256] c2 701 __switch_to+0xe4/0x100\n [46374.617265] c2 701 __schedule+0x80c/0xbc4\n [46374.617273] c2 701 schedule+0x74/0x98\n [46374.617281] c2 701 rwsem_down_read_failed+0x190/0x234\n [46374.617291] c2 701 down_read+0x58/0x5c\n [46374.617300] c2 701 f2fs_map_blocks+0x138/0x9a8\n [46374.617310] c2 701 get_data_block_dio_write+0x74/0x104\n [46374.617320] c2 701 __blockdev_direct_IO+0x1350/0x3930\n [46374.617331] c2 701 f2fs_direct_IO+0x55c/0x8bc\n [46374.617341] c2 701 __swap_writepage+0x1d0/0x3e8\n [46374.617351] c2 701 swap_writepage+0x44/0x54\n [46374.617360] c2 701 shrink_page_list+0x140/0xe80\n [46374.617371] c2 701 shrink_inactive_list+0x510/0x918\n [46374.617381] c2 701 shrink_node_memcg+0x2d4/0x804\n [46374.617391] c2 701 shrink_node+0x10c/0x2f8\n [46374.617400] c2 701 do_try_to_free_pages+0x178/0x38c\n [46374.617410] c2 701 try_to_free_pages+0x348/0x4b8\n [46374.617419] c2 701 __alloc_pages_nodemask+0x7f8/0x1014\n [46374.617429] c2 701 pagecache_get_page+0x184/0x2cc\n [46374.617438] c2 701 f2fs_new_node_page+0x60/0x41c\n [46374.617449] c2 701 f2fs_new_inode_page+0x50/0x7c\n [46374.617460] c2 701 f2fs_init_inode_metadata+0x128/0x530\n [46374.617472] c2 701 f2fs_add_inline_entry+0x138/0xd64\n [46374.617480] c2 701 f2fs_do_add_link+0xf4/0x178\n [46374.617488] c2 701 f2fs_create+0x1e4/0x3ac\n [46374.617497] c2 701 path_openat+0xdc0/0x1308\n [46374.617507] c2 701 do_filp_open+0x78/0x124\n [46374.617516] c2 701 do_sys_open+0x134/0x248\n [46374.617525] c2 701 SyS_openat+0x14/0x20\n \n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n\ncommit 4e3f12d866092444758953c197344d30ff99f904\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Thu Aug 15 18:23:40 2019 -0700\n\n drm/i915: Move gmbus definitions out of i915_reg.h\n \n They\u0027re not related to registers, so move them to the more appropriate\n intel_gmbus.h\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Reviewed-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816012343.36433-4-daniele.ceraolospurio@intel.com\n\ncommit 3d7b3039741de382fba0f0b9aca7869202749f08\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Thu Aug 15 18:23:39 2019 -0700\n\n drm/i915: Move engine IDs out of i915_reg.h\n \n To remove the dependency between the GT headers and i915_reg.h, move the\n definition of the engine IDs/classes to intel_engine_types.h\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@linux.intel.com\u003e\n Reviewed-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816012343.36433-3-daniele.ceraolospurio@intel.com\n\ncommit 3e5d0641e8969b9ced7cfcc791b4620a86d900e0\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Thu Aug 15 18:23:38 2019 -0700\n\n drm/i915: Move i915_power_well_id out of i915_reg.h\n \n It has nothing to do with registers, so move it to the more appropriate\n intel_display_power.h\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816012343.36433-2-daniele.ceraolospurio@intel.com\n\ncommit 927441adea560d59cbbd68b6c26342f44e612271\nAuthor: Marek Behún \u003cmarek.behun@nic.cz\u003e\nDate: Wed Aug 14 16:40:24 2019 +0200\n\n net: dsa: mv88e6xxx: check for mode change in port_setup_mac\n \n The mv88e6xxx_port_setup_mac checks if the requested MAC settings are\n different from the current ones, and if not, does nothing (since chaning\n them requires putting the link down).\n \n In this check it only looks if the triplet [link, speed, duplex] is\n being changed.\n \n This patch adds support to also check if the mode parameter (of type\n phy_interface_t) is requested to be changed. The current mode is\n computed by the -\u003eport_link_state() method, and if it is different from\n PHY_INTERFACE_MODE_NA, we check for equality with the requested mode.\n \n In the implementations of the mv88e6250_port_link_state() method we set\n the current mode to PHY_INTERFACE_MODE_NA - so the code does not check\n for mode change on 6250.\n \n In the mv88e6352_port_link_state() method, we use the cached cmode of\n the port to determine the mode as phy_interface_t (and if it is not\n enough, eg. for RGMII, we also look at the port control register for\n RX/TX timings).\n \n Signed-off-by: Marek Behún \u003cmarek.behun@nic.cz\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit df403069029dc61e0fc09cbeb0b5900705edec5b\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 16 18:16:08 2019 +0100\n\n drm/i915/execlists: Lift process_csb() out of the irq-off spinlock\n \n If we only call process_csb() from the tasklet, though we lose the\n ability to bypass ksoftirqd interrupt processing on direct submission\n paths, we can push it out of the irq-off spinlock.\n \n The penalty is that we then allow schedule_out to be called concurrently\n with schedule_in requiring us to handle the usage count (baked into the\n pointer itself) atomically.\n \n As we do kick the tasklets (via local_bh_enable()) after our submission,\n there is a possibility there to see if we can pull the local softirq\n processing back from the ksoftirqd.\n \n v2: Store the \u0027switch_priority_hint\u0027 on submission, so that we can\n safely check during process_csb().\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816171608.11760-1-chris@chris-wilson.co.uk\n\ncommit eadf0b17b43db4e73a6bdde1ad745d3b582a71c5\nAuthor: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\nDate: Wed Jul 3 17:56:38 2019 +0300\n\n bus: fsl-mc: remove explicit device_link_del\n \n Starting with commit 72175d4ea4c4 (\"driver core: Make driver core own\n stateful device links\") stateful device links are owned by the driver\n core and should not be explicitly removed on device unbind. Delete all\n device_link_del appearances from the fsl-mc bus.\n \n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Acked-By: Laurentiu Tudor \u003claurentiu.tudor@nxp.com\u003e\n Signed-off-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n\ncommit 45f16c82db7e8020fafa1a470556d3460c951a17\nAuthor: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nDate: Fri Aug 16 15:22:28 2019 +0300\n\n drm/omap: displays: Remove unused panel drivers\n \n drm_panel-based drivers for the ACX565AKM, LB035Q02, LS037V7DW01,\n NL8048HL11, TD028TTEC1 and TD043MTEA1 are available, remove the\n omapdrm-specific drivers.\n \n Signed-off-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816122228.9475-3-laurent.pinchart@ideasonboard.com\n\ncommit 1e938755fa51673210cd2998230610086fcb5cd6\nAuthor: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nDate: Fri Aug 16 15:22:27 2019 +0300\n\n video: omapfb2: Make standard and custom panel drivers mutually exclusive\n \n Standard DRM panel drivers for several panels used by omapfb2 are now\n available. Their module name clashes with the modules from\n drivers/video/fbdev/omap2/omapfb/displays, part of the deprecated omapfb2\n fbdev driver. As omapfb2 can only be compiled when the omapdrm driver is\n disabled, and the DRM panel drivers are useless in that case, make the\n omapfb2 panels depend on the standard DRM panels being disabled to fix\n the name clash.\n \n Signed-off-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Fixes: dc2e1e5b279966 (\"drm/panel: Add driver for the Toppoly TD043MTEA1 panel\")\n Fixes: 415b8dd0871188 (\"drm/panel: Add driver for the Toppoly TD028TTEC1 panel\")\n Fixes: 1c8fc3f0c5d280 (\"drm/panel: Add driver for the Sony ACX565AKM panel\")\n Fixes: c9cf4c2a3bd35f (\"drm/panel: Add driver for the Sharp LS037V7DW01 panel\")\n Fixes: df439abe650177 (\"drm/panel: Add driver for the NEC NL8048HL11 panel\")\n Cc: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e [added tags]\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816122228.9475-2-laurent.pinchart@ideasonboard.com\n\ncommit 767078132ff9625396f1387f487ce9389f405b2f\nAuthor: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\nDate: Fri Aug 16 16:10:11 2019 +0300\n\n dt-bindings: net: add bindings for ADIN PHY driver\n \n This change adds bindings for the Analog Devices ADIN PHY driver, detailing\n all the properties implemented by the driver.\n \n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9fe0b8d6ba9fea0018dc3ac93f4677c8d44bb9a0\nAuthor: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\nDate: Fri Aug 16 16:10:10 2019 +0300\n\n net: phy: adin: add ethtool get_stats support\n \n This change implements retrieving all the error counters from the PHY.\n \n The counters require that the RxErrCnt register (0x0014) be read first,\n after which copies of the counters are latched into the registers. This\n ensures that all registers read after RxErrCnt are synchronized at the\n moment that they are read.\n \n The counter values need to be accumulated by the driver, as each time that\n RxErrCnt is read, the values that are latched are the ones that have\n incremented from the last read.\n \n Signed-off-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 2d99b58461e111f1fd662f275f9711027bab4913\nAuthor: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\nDate: Fri Aug 16 16:10:09 2019 +0300\n\n net: phy: adin: implement downshift configuration via phy-tunable\n \n Down-speed auto-negotiation may not always be enabled, in which case the\n PHY won\u0027t down-shift to 100 or 10 during auto-negotiation.\n \n This change enables downshift and configures the number of retries to\n default 4 (which is also in the datasheet\n \n The downshift control mechanism can also be controlled via the phy-tunable\n interface (ETHTOOL_PHY_DOWNSHIFT control).\n \n The change has been adapted from the Aquantia PHY driver.\n \n Reviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit fa5bd9c5f1cdb0fa90530113056f45d699009ede\nAuthor: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\nDate: Fri Aug 16 16:10:08 2019 +0300\n\n net: phy: adin: implement PHY subsystem software reset\n \n The ADIN PHYs supports 4 types of reset:\n 1. The standard PHY reset via BMCR_RESET bit in MII_BMCR reg\n 2. Reset via GPIO\n 3. Reset via reg GeSftRst (0xff0c) \u0026 reload previous pin configs\n 4. Reset via reg GeSftRst (0xff0c) \u0026 request new pin configs\n \n Resets 2, 3 \u0026 4 are almost identical, with the exception that the crystal\n oscillator is available during reset for 2.\n \n This change implements subsystem software reset via the GeSftRst and\n reloading the previous pin configuration (so reset number 3).\n This will also reset the PHY core regs (similar to reset 1).\n \n Since writing bit 1 to reg GeSftRst is self-clearing, the only thing that\n can be done, is to write to that register, wait a specific amount of time\n (10 milliseconds should be enough) and try to read back and check if there\n are no errors on read. A busy-wait-read won\u0027t work well, and may sometimes\n work or not work.\n \n In case phylib is configured to also do a reset via GPIO, the ADIN PHY may\n be reset twice when the PHY device registers, but that isn\u0027t a problem,\n since it\u0027s being done on boot (or PHY device register).\n \n Signed-off-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c6aa697c41fd3bbd0d8697a7462109ffe605a867\nAuthor: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\nDate: Fri Aug 16 16:10:07 2019 +0300\n\n net: phy: adin: add EEE translation layer from Clause 45 to Clause 22\n \n The ADIN1200 \u0026 ADIN1300 PHYs support EEE by using standard Clause 45 access\n to access MMD registers for EEE.\n \n The EEE register addresses (when using Clause 22) are available at\n different addresses (than Clause 45), and since accessing these regs (via\n Clause 22) needs a special mechanism, a translation table is required to\n convert these addresses.\n \n For Clause 45, this is not needed since the driver will likely never use\n this access mode.\n \n Reviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b422d1b6f761bf336e0219fee6901481d33453a0\nAuthor: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\nDate: Fri Aug 16 16:10:06 2019 +0300\n\n net: phy: adin: add support MDI/MDIX/Auto-MDI selection\n \n The ADIN PHYs support automatic MDI/MDIX negotiation. By default this is\n disabled, so this is enabled at `config_init`.\n \n This is controlled via the PHY Control 1 register.\n The supported modes are:\n 1. Manual MDI\n 2. Manual MDIX\n 3. Auto MDIX - prefer MDIX\n 4. Auto MDIX - prefer MDI\n \n The phydev mdix \u0026 mdix_ctrl fields include modes 3 \u0026 4 into a single\n auto-mode. So, the default mode this driver enables is 4 when Auto-MDI mode\n is used.\n \n When detecting MDI/MDIX mode, a combination of the PHY Control 1 register\n and PHY Status 1 register is used to determine the correct MDI/MDIX mode.\n \n If Auto-MDI mode is not set, then the manual MDI/MDIX mode is returned.\n If Auto-MDI mode is set, then MDIX mode is returned differs from the\n preferred MDI/MDIX mode.\n This covers all cases where:\n 1. MDI preferred \u0026 Pair01Swapped \u003d\u003d MDIX\n 2. MDIX preferred \u0026 Pair01Swapped \u003d\u003d MDI\n 3. MDI preferred \u0026 ! Pair01Swapped \u003d\u003d MDIX\n 4. MDIX preferred \u0026 ! Pair01Swapped \u003d\u003d MDI\n \n The preferred MDI/MDIX mode is not configured via SW, but can be configured\n via HW pins. Note that the `Pair01Swapped` is the Green-Yellow physical\n pairs.\n \n Reviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Signed-off-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f1012fb476082c8983251a1991183964502ea855\nAuthor: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\nDate: Fri Aug 16 16:10:05 2019 +0300\n\n net: phy: adin: make RMII fifo depth configurable\n \n The FIFO depth can be configured for the RMII mode. This change adds\n support for doing this via device-tree (or ACPI).\n \n Reviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c83e6163d852148e352181efab7e82f4cf0aa483\nAuthor: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\nDate: Fri Aug 16 16:10:04 2019 +0300\n\n net: phy: adin: make RGMII internal delays configurable\n \n The internal delays for the RGMII are configurable for both RX \u0026 TX. This\n change adds support for configuring them via device-tree (or ACPI).\n \n Reviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d6200c8fd5b385af8640e5213532092014d53152\nAuthor: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\nDate: Fri Aug 16 16:10:03 2019 +0300\n\n net: phy: adin: configure RGMII/RMII/MII modes on config\n \n The ADIN1300 chip supports RGMII, RMII \u0026 MII modes. Default (if\n unconfigured) is RGMII.\n This change adds support for configuring these modes via the device\n registers.\n \n For RGMII with internal delays (modes RGMII_ID,RGMII_TXID, RGMII_RXID),\n the default delay is 2 ns. This can be configurable and will be done in\n a subsequent change.\n \n Reviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 3e32d020d8641bfd75ef6dae04f0569119879486\nAuthor: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\nDate: Fri Aug 16 16:10:02 2019 +0300\n\n net: phy: adin: add {write,read}_mmd hooks\n \n Both ADIN1200 \u0026 ADIN1300 support Clause 45 access for some registers.\n The Extended Management Interface (EMI) registers are accessible via both\n Clause 45 (at register MDIO_MMD_VEND1) and using Clause 22.\n \n The Clause 22 access for MMD regs differs from the standard one defined by\n 802.3. The ADIN PHYs use registers ExtRegPtr (0x0010) and ExtRegData\n (0x0011) to access Clause 45 \u0026 EMI registers.\n \n The indirect access is done via the following mechanism (for both R/W):\n 1. Write the address of the register in the ExtRegPtr\n 2. Read/write the value of the register via reg ExtRegData\n \n This mechanism is needed to manage configuration of chip settings and to\n access EEE registers via Clause 22.\n \n Since Clause 45 access will likely never be used, it is not implemented via\n this hook.\n \n Reviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit fb44b8d62cc3f133bb30e8d47d151c306af9fe8f\nAuthor: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\nDate: Fri Aug 16 16:10:01 2019 +0300\n\n net: phy: adin: add support for interrupts\n \n This change hooks link-status-change interrupts to phylib.\n \n Reviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 49cc4c7db453af90010758b564c554562b1dc90a\nAuthor: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\nDate: Fri Aug 16 16:10:00 2019 +0300\n\n net: phy: adin: hook genphy_{suspend, resume} into the driver\n \n The chip supports standard suspend/resume via BMCR reg.\n Hook these functions into the `adin` driver.\n \n Reviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9c1029818c713eccdce945e45cd8fb927d6a3d6e\nAuthor: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\nDate: Fri Aug 16 16:09:59 2019 +0300\n\n net: phy: adin: add support for Analog Devices PHYs\n \n This change adds support for Analog Devices Industrial Ethernet PHYs.\n Particularly the PHYs this driver adds support for:\n * ADIN1200 - Robust, Industrial, Low Power 10/100 Ethernet PHY\n * ADIN1300 - Robust, Industrial, Low Latency 10/100/1000 Gigabit\n Ethernet PHY\n \n The 2 chips are register compatible with one another. The main difference\n being that ADIN1200 doesn\u0027t operate in gigabit mode.\n \n The chips can be operated by the Generic PHY driver as well via the\n standard IEEE PHY registers (0x0000 - 0x000F) which are supported by the\n kernel as well. This assumes that configuration of the PHY has been done\n completely in HW, according to spec.\n \n Configuration can also be done via registers, which will be supported by\n this driver.\n \n Datasheets:\n https://www.analog.com/media/en/technical-documentation/data-sheets/ADIN1300.pdf\n https://www.analog.com/media/en/technical-documentation/data-sheets/ADIN1200.pdf\n \n Reviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a8384e6ce5231d4f697f2dca1de4ff6bed7f3890\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:25:11 2019 +0200\n\n ia64: move the screen_info setup to common code\n \n There is nothing really platform specific about setting about the\n screen_info from the ia64_boot_param structure, so move it from the\n dig machvec to common code.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-26-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit fa809d709e7acb99aa1c74987829ae13d908af38\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:25:10 2019 +0200\n\n ia64: move the ROOT_DEV setup to common code\n \n I\u0027m not sure how useful a platform default ROOT_DEV is these days,\n but it pretty sure isn\u0027t machvec dependent.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-25-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 974f83ec1e5afeeb4c9028feb74ffa4ee70e71b7\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:25:09 2019 +0200\n\n ia64: rework iommu probing\n \n ia64 currently organizes the iommu probing along machves, which isn\u0027t\n very helpful. Instead just try to probe for Intel IOMMUs in mem_init\n as they are properly described in ACPI and if none was found initialize\n the swiotlb buffer. The HP SBA handling is then only done delayed when\n the actual hardware is probed. Only in the case that we actually found\n usable IOMMUs we then set up the DMA ops and free the not needed\n swiotlb buffer. This scheme gets rid of the need for the dma_init\n machvec operation, and the dig_vtd machvec.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-24-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 16567ca85fce11e2a20f22404ed0ceb133c4408e\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:25:08 2019 +0200\n\n ia64: remove the unused sn_coherency_id symbol\n \n The sn_coherency_id symbol isn\u0027t used anywhere, remove it.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-23-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 1164e757541b75f0bd87bb541f7ce49614471f5d\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:25:07 2019 +0200\n\n ia64: remove the SGI UV simulator support\n \n The simulator support was marked as temporary since the initial commit,\n so drop it more than 10 years later.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-22-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit df43acac8e32e2ee51a1e9376993c5ac18598d92\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:25:06 2019 +0200\n\n ia64: remove the zx1 swiotlb machvec\n \n The aim of this machvec is to support devices with \u003c 32-bit dma\n masks. But given that ia64 only has a ZONE_DMA32 and not a ZONE_DMA\n that isn\u0027t supported by swiotlb either.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-21-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 2e0f2b1659ddd7bb005ca0bf9f92915904974676\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:25:05 2019 +0200\n\n ia64: remove CONFIG_ACPI ifdefs\n \n Now that hpsim support is gone, CONFIG_ACPI is forced on for ia64, and\n we can remove a few ifdefs for it.\n \n Acked-by: Tom Vaden \u003ctom.vaden@hpe.com\u003e\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-20-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 768557ca5b062e14ccdc55d255f7ff50bb90d40a\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:25:04 2019 +0200\n\n ia64: remove CONFIG_PCI ifdefs\n \n Now that hpsim support is gone, CONFIG_PCI is forced on for ia64, and\n we can remove a few ifdefs for it.\n \n Acked-by: Tom Vaden \u003ctom.vaden@hpe.com\u003e\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-19-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit fc5bad03709f9c79ea299f554b6f99fc6f4fe31c\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:25:03 2019 +0200\n\n ia64: remove the hpsim platform\n \n The hpsim platform supports the HP IA64 simulator which was useful as a\n bring up platform. But it is fairly non-standard vs real IA64 system\n in that it for example doesn\u0027t support ACPI. It also comes with a\n whole bunch of simulator specific drivers. Remove it to simplify the\n IA64 port.\n \n Note that through a weird twist only them hpsim boot loader built the\n vmlinux.gz file, so the makefile targets for that are moved to the\n main ia64 Makefile now.\n \n Acked-by: Tom Vaden \u003ctom.vaden@hpe.com\u003e\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-18-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 05933aac7b11911955de307a329dc2a7a14b7bd0\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:25:02 2019 +0200\n\n ia64: remove now unused machvec indirections\n \n With the SGI SN2 machvec removal most of the indirections are unused\n now, so remove them. This includes the entire removal of the mmio\n read*/write* macros as the generic ones are identical to the\n asm-generic/io.h version.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-17-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit cf07cb1ff4ea008abf06c95878c700cf1dd65c3e\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:25:01 2019 +0200\n\n ia64: remove support for the SGI SN2 platform\n \n The SGI SN2 (early Altix) is a very non-standard IA64 platform that was\n at the very high end of even IA64 hardware, and has been discontinued\n a long time ago. Remove it because there no upstream users left, and it\n has magic hooks all over the kernel.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-16-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit f7bc6e42bf12487182fc442a08eca25d968dc543\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:25:00 2019 +0200\n\n drivers: remove the SGI SN2 IOC4 base support\n \n The IOC4 is a multi-function chip seen on SGI SN2 and some SGI MIPS\n systems. This removes the base driver, which while not having an SN2\n Kconfig dependency was only for sub-drivers that had one.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-15-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit c9fa9c327b5228c516f4a8c54b91b711526e3e96\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:24:59 2019 +0200\n\n drivers: remove the SGI SN2 IOC3 base support\n \n The IOC3 is a multi-function chip seen on SGI SN2 and some SGI\n MIPS systems. This removes the last bit of SN2 specific support,\n while the bits used by the mips ports are still around (and being\n substantially rewritten at the moment).\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-14-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 42ab658fccb6e13748ff4cb8bf33d1a2fe3f934a\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:24:58 2019 +0200\n\n qla2xxx: remove SGI SN2 support\n \n The SGI SN2 support is about to be removed, so drop the bits specific to\n it from this driver.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-13-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit f877a786b1dd314ec081355507b8605508e4dfab\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:24:57 2019 +0200\n\n qla1280: remove SGI SN2 support\n \n The SGI SN2 support is about to be removed, so drop the bits specific to\n it from this driver.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-12-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 9726bfcdb977d6f78074fcc5dd23003b450c1f35\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:24:56 2019 +0200\n\n misc/sgi-xp: remove SGI SN2 support\n \n Note this also marks xp broken on ia64 now, as the UV support, which\n was disable in generic kernels before actually never compiled due to\n undefined uv_gpa_to_soc_phys_ram and uv_gpa_in_mmr_space symbols since\n at least commit c2c9f1157414 (\"x86: uv: update XPC to handle updated\n BIOS interface\").\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-11-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 0fef2532d2a17100e57fad0569555536cac47eda\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:24:55 2019 +0200\n\n char/mspec: remove SGI SN2 support\n \n The SGI SN2 support is about to be removed, so drop the bits specific to\n it from this driver.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-10-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 9c860e4cf70870703ce57797660c19c192576c5c\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:24:54 2019 +0200\n\n tty/serial: remove the ioc3_serial driver\n \n The SGI SN2 support is about to be removed. Remove this driver that\n depends on the SN2 support.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-9-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit a017ef17cfd810b9e81f2b9f3202d02bc952d5f6\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:24:53 2019 +0200\n\n tty/serial: remove the ioc4_serial driver\n \n The SGI SN2 support is about to be removed. Remove this driver that\n depends on the SN2 support.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-8-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 08f9530a0f883b11fc4062a20703e8920c5980c7\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:24:52 2019 +0200\n\n tty/serial: remove the sn_console driver\n \n The SGI SN2 support is about to be removed. Remove this driver that\n depends on the SN2 support.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-7-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit c7532b601e774c472b8bf5c1fca8f39e82e4394a\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:24:51 2019 +0200\n\n PCI/hotplug: remove the sgi_hotplug driver\n \n The SGI SN2 support is about to be removed. Remove this driver that\n depends on the SN2 support.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-6-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 0b43ba0dbe81fdc71ff250b45a76bfa794d53987\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:24:50 2019 +0200\n\n ide: remove the sgiioc4 driver\n \n The SGI SN2 support is about to be removed. Remove this driver that\n depends on the SN2 support.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-5-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit defdeacb1ba891fa5084df5a809ceef488b0a9cf\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:24:49 2019 +0200\n\n char/agp: remove the sgi-agp driver\n \n The SGI SN2 support is about to be removed. Remove this driver that\n depends on the SN2 support.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-4-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 8334d1d415737198f64b0d31edb5daa0f14bb1e5\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:24:48 2019 +0200\n\n char: remove the SGI tiocx/mbcs driver\n \n The SGI SN2 support is about to be removed. Remove this driver that\n depends on the SN2 support.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-3-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit c116954b2c9707f540eca852b3b86a087dffa4b4\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 09:24:47 2019 +0200\n\n char: remove the SGI snsc driver\n \n The SGI SN2 support is about to be removed. Remove this driver that\n depends on the SN2 support.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190813072514.23299-2-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 62c3801619e16b68a37ea899b76572145dfe41c9\nAuthor: Fei Shao \u003cfshao@chromium.org\u003e\nDate: Wed Aug 14 15:06:00 2019 -0700\n\n Input: cros_ec_keyb - add back missing mask for event_type\n \n In the previous patch we didn\u0027t mask out event_type in case statement,\n so switches are always picked instead of buttons, which results in\n ChromeOS devices misbehaving when power button is pressed.\n This patch adds back the missing mask.\n \n Fixes: d096aa3eb604 (\"Input: cros_ec_keyb: mask out extra flags in event_type\")\n Signed-off-by: Fei Shao \u003cfshao@chromium.org\u003e\n Reviewed-by: Nicolas Boichat \u003cdrinkcat@chromium.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 10fa8acf0fa6ed79ddb662488addb3ba71e9db60\nAuthor: J. Bruce Fields \u003cbfields@redhat.com\u003e\nDate: Tue Jul 30 19:06:38 2019 -0400\n\n nfsd: Remove unnecessary NULL checks\n \n \"cb\" is never actually NULL in these functions.\n \n On a quick skim of the history, they seem to have been there from the\n beginning. I\u0027m not sure if they originally served a purpose.\n \n Reported-by: Jia-Ju Bai \u003cbaijiaju1990@gmail.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit b0740d71cb56149aae04adad1746c3565ac5b278\nAuthor: Stephen Boyd \u003csboyd@kernel.org\u003e\nDate: Wed Aug 14 21:25:00 2019 -0700\n\n clk: composite: Drop unused clk.h include\n \n This include isn\u0027t used. Drop it.\n \n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190815042500.9519-1-sboyd@kernel.org\n\ncommit 0214f33c4e0e1f5a9a9c3f4881310c61713294e6\nAuthor: Stephen Boyd \u003csboyd@kernel.org\u003e\nDate: Wed Jul 31 12:35:17 2019 -0700\n\n clk: Overwrite clk_hw::init with NULL during clk_register()\n \n We don\u0027t want clk provider drivers to use the init structure after clk\n registration time, but we leave a dangling reference to it by means of\n clk_hw::init. Let\u0027s overwrite the member with NULL during clk_register()\n so that this can\u0027t be used anymore after registration time.\n \n Cc: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Cc: Doug Anderson \u003cdianders@chromium.org\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190731193517.237136-10-sboyd@kernel.org\n Reviewed-by: Sylwester Nawrocki \u003cs.nawrocki@samsung.com\u003e\n\ncommit a7b85ad25a97cf897b4819a121655c483d86156f\nAuthor: Stephen Boyd \u003csboyd@kernel.org\u003e\nDate: Wed Aug 14 21:10:37 2019 -0700\n\n clk: sunxi: Don\u0027t call clk_hw_get_name() on a hw that isn\u0027t registered\n \n The implementation of clk_hw_get_name() relies on the clk_core\n associated with the clk_hw pointer existing. If of_clk_hw_register()\n fails, there isn\u0027t a clk_core created yet, so calling clk_hw_get_name()\n here fails. Extract the name first so we can print it later.\n \n Fixes: 1d80c14248d6 (\"clk: sunxi-ng: Add common infrastructure\")\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Chen-Yu Tsai \u003cwens@csie.org\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit e0e04fc866929f58600aad66b969d5e1572f9e54\nAuthor: Stephen Boyd \u003csboyd@kernel.org\u003e\nDate: Thu Aug 15 15:12:49 2019 -0700\n\n clk: ti: Don\u0027t reference clk_init_data after registration\n \n A future patch is going to change semantics of clk_register() so that\n clk_hw::init is guaranteed to be NULL after a clk is registered. Avoid\n referencing this member here so that we don\u0027t run into NULL pointer\n exceptions.\n \n Cc: Tero Kristo \u003ct-kristo@ti.com\u003e\n Cc: Tony Lindgren \u003ctony@atomide.com\u003e\n Reported-by: \"kernelci.org bot\" \u003cbot@kernelci.org\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190815221249.53235-1-sboyd@kernel.org\n\ncommit c42144139a383011327990a8c59a45f81b3493dd\nAuthor: Stephen Boyd \u003csboyd@kernel.org\u003e\nDate: Thu Aug 15 09:00:20 2019 -0700\n\n clk: qcom: Remove error prints from DFS registration\n \n These aren\u0027t useful and they reference the init structure name. Let\u0027s\n just drop them.\n \n Cc: Taniya Das \u003ctdas@codeaurora.org\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190815160020.183334-5-sboyd@kernel.org\n Acked-by: Taniya Das \u003ctdas@codeaurora.org\u003e\n\ncommit 21ef77de22c0d2b6c002150c25bab77199defd3f\nAuthor: Stephen Boyd \u003csboyd@kernel.org\u003e\nDate: Thu Aug 15 09:00:19 2019 -0700\n\n rtc: sun6i: Don\u0027t reference clk_init_data after registration\n \n A future patch is going to change semantics of clk_register() so that\n clk_hw::init is guaranteed to be NULL after a clk is registered. Avoid\n referencing this member here so that we don\u0027t run into NULL pointer\n exceptions.\n \n Cc: Alessandro Zummo \u003ca.zummo@towertech.it\u003e\n Cc: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Chen-Yu Tsai \u003cwens@csie.org\u003e\n Reported-by: \"kernelci.org bot\" \u003cbot@kernelci.org\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190815160020.183334-4-sboyd@kernel.org\n Acked-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\n Acked-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 1a4549c150e27dbc3aea762e879a88209df6d1a5\nAuthor: Stephen Boyd \u003csboyd@kernel.org\u003e\nDate: Thu Aug 15 09:00:18 2019 -0700\n\n clk: zx296718: Don\u0027t reference clk_init_data after registration\n \n A future patch is going to change semantics of clk_register() so that\n clk_hw::init is guaranteed to be NULL after a clk is registered. Avoid\n referencing this member here so that we don\u0027t run into NULL pointer\n exceptions.\n \n Cc: Jun Nie \u003cjun.nie@linaro.org\u003e\n Cc: Shawn Guo \u003cshawnguo@kernel.org\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190815160020.183334-3-sboyd@kernel.org\n\ncommit c8cec4f4af99288acefa2caea77b07e06a32641c\nAuthor: Stephen Boyd \u003csboyd@kernel.org\u003e\nDate: Thu Aug 15 09:00:17 2019 -0700\n\n clk: milbeaut: Don\u0027t reference clk_init_data after registration\n \n A future patch is going to change semantics of clk_register() so that\n clk_hw::init is guaranteed to be NULL after a clk is registered. Avoid\n referencing this member here so that we don\u0027t run into NULL pointer\n exceptions.\n \n Cc: Sugaya Taichi \u003csugaya.taichi@socionext.com\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190815160020.183334-2-sboyd@kernel.org\n\ncommit 1bc5557a3bffbac6c37c52c0f65ef78bfa147b85\nAuthor: Stephen Boyd \u003csboyd@kernel.org\u003e\nDate: Tue Aug 13 17:24:02 2019 -0700\n\n clk: socfpga: deindent code to proper indentation\n \n This code is indented oddly, causing checkpatch to complain. Indent it\n properly.\n \n Cc: Dinh Nguyen \u003cdinguyen@kernel.org\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190814002402.18154-1-sboyd@kernel.org\n Acked-by: Dinh Nguyen \u003cdinguyen@kernel.org\u003e\n\ncommit bd0e79f1da3cc5da586c92144fbab9fd24700735\nAuthor: Stephen Boyd \u003csboyd@kernel.org\u003e\nDate: Wed Jul 31 12:35:16 2019 -0700\n\n phy: ti: am654-serdes: Don\u0027t reference clk_init_data after registration\n \n A future patch is going to change semantics of clk_register() so that\n clk_hw::init is guaranteed to be NULL after a clk is registered. Avoid\n referencing this member here so that we don\u0027t run into NULL pointer\n exceptions.\n \n Cc: Roger Quadros \u003crogerq@ti.com\u003e\n Cc: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190731193517.237136-9-sboyd@kernel.org\n Acked-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit f6c90df8e7e33c3dc33d4d7471bc42c232b0510e\nAuthor: Stephen Boyd \u003csboyd@kernel.org\u003e\nDate: Wed Jul 31 12:35:15 2019 -0700\n\n clk: sprd: Don\u0027t reference clk_init_data after registration\n \n A future patch is going to change semantics of clk_register() so that\n clk_hw::init is guaranteed to be NULL after a clk is registered. Avoid\n referencing this member here so that we don\u0027t run into NULL pointer\n exceptions.\n \n Cc: Chunyan Zhang \u003czhang.chunyan@linaro.org\u003e\n Cc: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190731193517.237136-8-sboyd@kernel.org\n Acked-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Acked-by: Chunyan Zhang \u003czhang.chunyan@linaro.org\u003e\n\ncommit 09d4922d3c94832962c572808bac20dbee5004c5\nAuthor: Stephen Boyd \u003csboyd@kernel.org\u003e\nDate: Wed Jul 31 12:35:14 2019 -0700\n\n clk: socfpga: Don\u0027t reference clk_init_data after registration\n \n A future patch is going to change semantics of clk_register() so that\n clk_hw::init is guaranteed to be NULL after a clk is registered. Avoid\n referencing this member here so that we don\u0027t run into NULL pointer\n exceptions.\n \n Cc: Dinh Nguyen \u003cdinguyen@kernel.org\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190731193517.237136-7-sboyd@kernel.org\n Acked-by: Dinh Nguyen \u003cdinguyen@kernel.org\u003e\n\ncommit af55dadfbce35b4f4c6247244ce3e44b2e242b84\nAuthor: Stephen Boyd \u003csboyd@kernel.org\u003e\nDate: Wed Jul 31 12:35:13 2019 -0700\n\n clk: sirf: Don\u0027t reference clk_init_data after registration\n \n A future patch is going to change semantics of clk_register() so that\n clk_hw::init is guaranteed to be NULL after a clk is registered. Avoid\n referencing this member here so that we don\u0027t run into NULL pointer\n exceptions.\n \n Cc: Guo Zeng \u003cGuo.Zeng@csr.com\u003e\n Cc: Barry Song \u003cBaohua.Song@csr.com\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190731193517.237136-6-sboyd@kernel.org\n\ncommit af884a5dfdef06644bdbd7d54ca865616f24b256\nAuthor: Stephen Boyd \u003csboyd@kernel.org\u003e\nDate: Wed Jul 31 12:35:12 2019 -0700\n\n clk: qcom: Don\u0027t reference clk_init_data after registration\n \n A future patch is going to change semantics of clk_register() so that\n clk_hw::init is guaranteed to be NULL after a clk is registered. Avoid\n referencing this member here so that we don\u0027t run into NULL pointer\n exceptions.\n \n Cc: Taniya Das \u003ctdas@codeaurora.org\u003e\n Cc: Andy Gross \u003cagross@kernel.org\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190731193517.237136-5-sboyd@kernel.org\n Acked-by: Taniya Das \u003ctdas@codeaurora.org\u003e\n\ncommit 1610dd79d0f6202c5c1a91122255fa598679c13a\nAuthor: Stephen Boyd \u003csboyd@kernel.org\u003e\nDate: Wed Jul 31 12:35:11 2019 -0700\n\n clk: meson: axg-audio: Don\u0027t reference clk_init_data after registration\n \n A future patch is going to change semantics of clk_register() so that\n clk_hw::init is guaranteed to be NULL after a clk is registered. Avoid\n referencing this member here so that we don\u0027t run into NULL pointer\n exceptions.\n \n Cc: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Cc: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190731193517.237136-4-sboyd@kernel.org\n Acked-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n\ncommit d6d251f9bb74435a3e4ade641306f90f75645c5c\nAuthor: Stephen Boyd \u003csboyd@kernel.org\u003e\nDate: Wed Jul 31 12:35:10 2019 -0700\n\n clk: lochnagar: Don\u0027t reference clk_init_data after registration\n \n A future patch is going to change semantics of clk_register() so that\n clk_hw::init is guaranteed to be NULL after a clk is registered. Avoid\n referencing this member here so that we don\u0027t run into NULL pointer\n exceptions.\n \n Cc: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n Cc: Richard Fitzgerald \u003crf@opensource.cirrus.com\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190731193517.237136-3-sboyd@kernel.org\n Acked-by: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n\ncommit cf9ec1fc6d7cceb73e7f1efd079d2eae173fdf57\nAuthor: Stephen Boyd \u003csboyd@kernel.org\u003e\nDate: Wed Jul 31 12:35:09 2019 -0700\n\n clk: actions: Don\u0027t reference clk_init_data after registration\n \n A future patch is going to change semantics of clk_register() so that\n clk_hw::init is guaranteed to be NULL after a clk is registered. Avoid\n referencing this member here so that we don\u0027t run into NULL pointer\n exceptions.\n \n Cc: Manivannan Sadhasivam \u003cmanivannan.sadhasivam@linaro.org\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190731193517.237136-2-sboyd@kernel.org\n [sboyd@kernel.org: Move name to after checking for error or NULL hw]\n Acked-by: Manivannan Sadhasivam \u003cmanivannan.sadhasivam@linaro.org\u003e\n\ncommit 25ffd4b11d069300f018f7b04c3c6b8814a128d6\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 16 13:10:00 2019 +0100\n\n drm/i915: Markup expected timeline locks for i915_active\n \n As every i915_active_request should be serialised by a dedicated lock,\n i915_active consists of a tree of locks; one for each node. Markup up\n the i915_active_request with what lock is supposed to be guarding it so\n that we can verify that the serialised updated are indeed serialised.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816121000.8507-2-chris@chris-wilson.co.uk\n\ncommit 6c69a45445af924378371ce3ae3fbe20c4657b94\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 16 13:09:59 2019 +0100\n\n drm/i915/gt: Mark context-\u003eactive_count as protected by timeline-\u003emutex\n \n We use timeline-\u003emutex to protect modifications to\n context-\u003eactive_count, and the associated enable/disable callbacks.\n Due to complications with engine-pm barrier there is a path where we used\n a \"superlock\" to provide serialised protect and so could not\n unconditionally assert with lockdep that it was always held. However,\n we can mark the mutex as taken (noting that we may be nested underneath\n ourselves) which means we can be reassured the right timeline-\u003emutex is\n always treated as held and let lockdep roam free.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816121000.8507-1-chris@chris-wilson.co.uk\n\ncommit df2474a22c42ce419b67067c52d71da06c385501\nAuthor: Jeff Layton \u003cjlayton@kernel.org\u003e\nDate: Thu Aug 15 15:21:17 2019 -0400\n\n locks: print a warning when mount fails due to lack of \"mand\" support\n \n Since 9e8925b67a (\"locks: Allow disabling mandatory locking at compile\n time\"), attempts to mount filesystems with \"-o mand\" will fail.\n Unfortunately, there is no other indiciation of the reason for the\n failure.\n \n Change how the function is defined for better readability. When\n CONFIG_MANDATORY_FILE_LOCKING is disabled, printk a warning when\n someone attempts to mount with -o mand.\n \n Also, add a blurb to the mandatory-locking.txt file to explain about\n the \"mand\" option, and the behavior one should expect when it is\n disabled.\n \n Reported-by: Jan Kara \u003cjack@suse.cz\u003e\n Reviewed-by: Jan Kara \u003cjack@suse.cz\u003e\n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n\ncommit dd7c66ac777a70e224e66593e95ce6d22147c919\nAuthor: Justin Swartz \u003cjustin.swartz@risingedge.co.za\u003e\nDate: Sun Aug 11 23:00:13 2019 +0000\n\n ARM: dts: add device tree for Mecer Xtreme Mini S6\n \n The Mecer Xtreme Mini S6 features a Rockchip RK3229 SoC,\n 1GB DDR3 RAM, 8GB eMMC, MicroSD port, 10/100Mbps Ethernet,\n Realtek 8723BS WLAN module, 2 x USB 2.0 ports, HDMI output,\n and S/PDIF output.\n \n Signed-off-by: Justin Swartz \u003cjustin.swartz@risingedge.co.za\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit f789fbb1eb16de8db1e2eb7b6648f85464326195\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Fri Aug 16 10:55:01 2019 +0000\n\n drm/i915/wopcm: Fix SPDX tag location\n \n Move SPDX tag to first line, and update year to 2019.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816105501.31020-6-michal.wajdeczko@intel.com\n\ncommit 0d12ed982bbaf00966dd31b9963a8cbe4775f691\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Fri Aug 16 10:55:00 2019 +0000\n\n drm/i915/wopcm: Update error messages\n \n All WOPCM error messages are device specific, so use\n device specific error functions.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816105501.31020-5-michal.wajdeczko@intel.com\n\ncommit 311482651d5e7bd6e8ad4e4b42e078ed3a866b4a\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Fri Aug 16 10:54:59 2019 +0000\n\n drm/i915/wopcm: Try to use already locked WOPCM layout\n \n If WOPCM layout is already locked in HW we shouldn\u0027t continue\n with our own partitioning as it could be likely different and\n we will be unable to enforce it and fail. Instead we should try\n to reuse what is already programmed, maybe there will be a fit.\n \n This should enable us to reload driver with slightly different\n HuC firmware (or even without HuC) without need to reboot.\n \n v2: reordered/rebased\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Michal Winiarski \u003cmichal.winiarski@intel.com\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816105501.31020-4-michal.wajdeczko@intel.com\n\ncommit 851dfec946b6cb4265c207541d84c84334acfa9d\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Fri Aug 16 10:54:58 2019 +0000\n\n drm/i915/wopcm: Check WOPCM layout separately from calculations\n \n We can do WOPCM partitioning using rough estimates and limits\n and perform detailed check as separate step.\n \n v2: oops! s/max/min\n v3: consolidate overflow checks (Daniele)\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816105501.31020-3-michal.wajdeczko@intel.com\n\ncommit 1249dc1717381b37266201086d2806559d5715c1\nAuthor: Michał Winiarski \u003cmichal.winiarski@intel.com\u003e\nDate: Fri Aug 16 10:54:57 2019 +0000\n\n drm/i915/uc: Move FW size sanity check back to fetch\n \n While we need to know WOPCM size to do this sanity check, it has more to\n do with FW than with WOPCM. Let\u0027s move the check to fetch phase, it\u0027s\n not like WOPCM is going to grow in the meantime.\n \n v2: rebased\n v3: use __intel_uc_fw_get_upload_size (Daniele)\n \n Signed-off-by: Michał Winiarski \u003cmichal.winiarski@intel.com\u003e\n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Jackie Li \u003cyaodong.li@intel.com\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816105501.31020-2-michal.wajdeczko@intel.com\n\ncommit e2736219e6ca3117e10651e215b96d66775220da\nAuthor: John Keeping \u003cjohn@metanate.com\u003e\nDate: Thu Aug 15 11:01:46 2019 +0100\n\n perf unwind: Remove unnecessary test\n \n If dwarf_callchain_users is false, then unwind__prepare_access() will\n not set unwind_libunwind_ops so the remaining test here is sufficient.\n \n Signed-off-by: John Keeping \u003cjohn@metanate.com\u003e\n Reviewed-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Konstantin Khlebnikov \u003ckhlebnikov@yandex-team.ru\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: john keeping \u003cjohn@metanate.com\u003e\n Link: http://lkml.kernel.org/r/20190815100146.28842-3-john@metanate.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 3ba09632ce8bd42d1e4b57a6a423dcab937e211b\nAuthor: Matthew Auld \u003cmatthew.auld@intel.com\u003e\nDate: Fri Aug 16 11:53:57 2019 +0100\n\n drm/i915/buddy: use kmemleak_update_trace\n \n Since nodes are cached in a free-list, and potentially marked as free\n without actually being destroyed, thus allowing them to be\n opportunistically re-allocated, we should apply kmemleak_update_trace\n every time a node is given a new owner and marked as allocated, to aid\n in debugging.\n \n Suggested-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816105357.14340-2-matthew.auld@intel.com\n\ncommit 665c1c2166d3cc394f0cf2dd2da1e68b645552a7\nAuthor: Matthew Auld \u003cmatthew.auld@intel.com\u003e\nDate: Fri Aug 16 11:53:56 2019 +0100\n\n drm/i915/buddy: tidy up i915_buddy_fini\n \n If we are leaking nodes don\u0027t hide it. Also stop trying to be\n \"defensive\" and instead embrace Kasan et al.\n \n Signed-off-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816105357.14340-1-matthew.auld@intel.com\n\ncommit e8ba2906f6b9054102ad035ac9cafad9d4168589\nAuthor: John Keeping \u003cjohn@metanate.com\u003e\nDate: Thu Aug 15 11:01:45 2019 +0100\n\n perf unwind: Fix libunwind when tid !\u003d pid\n \n Commit e5adfc3e7e77 (\"perf map: Synthesize maps only for thread group\n leader\") changed the recording side so that we no longer get mmap events\n for threads other than the thread group leader (when synthesising these\n events for threads which exist before perf is started).\n \n When a file recorded after this change is loaded, the lack of mmap\n records mean that unwinding is not set up for any other threads.\n \n This can be seen in a simple record/report scenario:\n \n perf record --call-graph\u003ddwarf -t $TID\n perf report\n \n If $TID is a process ID then the report will show call graphs, but if\n $TID is a secondary thread the output is as if --call-graph\u003dnone was\n specified.\n \n Following the rationale in that commit, move the libunwind fields into\n struct map_groups and update the libunwind functions to take this\n instead of the struct thread. This is only required for\n unwind__finish_access which must now be called from map_groups__delete\n and the others are changed for symmetry.\n \n Note that unwind__get_entries keeps the thread argument since it is\n required for symbol lookup and the libdw unwind provider uses the thread\n ID.\n \n Signed-off-by: John Keeping \u003cjohn@metanate.com\u003e\n Reviewed-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Konstantin Khlebnikov \u003ckhlebnikov@yandex-team.ru\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Fixes: e5adfc3e7e77 (\"perf map: Synthesize maps only for thread group leader\")\n Link: http://lkml.kernel.org/r/20190815100146.28842-2-john@metanate.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit ab6cd0e5276e24403751e0b3b8ed807738a8571f\nAuthor: John Keeping \u003cjohn@metanate.com\u003e\nDate: Thu Aug 15 11:01:44 2019 +0100\n\n perf map: Use zalloc for map_groups\n \n In the next commit we will add new fields to map_groups and we need\n these to be null if no value is assigned. The simplest way to achieve\n this is to request zeroed memory from the allocator.\n \n Signed-off-by: John Keeping \u003cjohn@metanate.com\u003e\n Reviewed-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Konstantin Khlebnikov \u003ckhlebnikov@yandex-team.ru\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: john keeping \u003cjohn@metanate.com\u003e\n Link: http://lkml.kernel.org/r/20190815100146.28842-1-john@metanate.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit ef4b1a539f4b8776701752c5a09ee741a4232ae6\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 15 18:18:58 2019 -0300\n\n perf report: Add --switch-on/--switch-off events\n \n Since \u0027perf top\u0027 shares the histogram browser with \u0027perf report\u0027, then\n the same explanation in the previous cset applies.\n \n An additional example uses a pair of SDT events available for systemtap:\n \n # perf probe --exec\u003d/usr/bin/stap \u0027%*:*\u0027\n Added new events:\n sdt_stap:benchmark__thread__start (on %* in /usr/bin/stap)\n sdt_stap:benchmark (on %* in /usr/bin/stap)\n sdt_stap:benchmark__thread__end (on %* in /usr/bin/stap)\n sdt_stap:pass6__start (on %* in /usr/bin/stap)\n sdt_stap:pass6__end (on %* in /usr/bin/stap)\n sdt_stap:pass5__start (on %* in /usr/bin/stap)\n sdt_stap:pass5__end (on %* in /usr/bin/stap)\n sdt_stap:pass0__start (on %* in /usr/bin/stap)\n sdt_stap:pass0__end (on %* in /usr/bin/stap)\n sdt_stap:pass1a__start (on %* in /usr/bin/stap)\n sdt_stap:pass1b__start (on %* in /usr/bin/stap)\n sdt_stap:pass1__end (on %* in /usr/bin/stap)\n sdt_stap:pass2__start (on %* in /usr/bin/stap)\n sdt_stap:pass2__end (on %* in /usr/bin/stap)\n sdt_stap:pass3__start (on %* in /usr/bin/stap)\n sdt_stap:pass3__end (on %* in /usr/bin/stap)\n sdt_stap:pass4__start (on %* in /usr/bin/stap)\n sdt_stap:pass4__end (on %* in /usr/bin/stap)\n sdt_stap:benchmark__start (on %* in /usr/bin/stap)\n sdt_stap:benchmark__end (on %* in /usr/bin/stap)\n sdt_stap:cache__get (on %* in /usr/bin/stap)\n sdt_stap:cache__clean (on %* in /usr/bin/stap)\n sdt_stap:cache__add__module (on %* in /usr/bin/stap)\n sdt_stap:cache__add__source (on %* in /usr/bin/stap)\n sdt_stap:stap_system__complete (on %* in /usr/bin/stap)\n sdt_stap:stap_system__start (on %* in /usr/bin/stap)\n sdt_stap:stap_system__spawn (on %* in /usr/bin/stap)\n sdt_stap:stap_system__fork (on %* in /usr/bin/stap)\n sdt_stap:intern_string (on %* in /usr/bin/stap)\n sdt_stap:client__start (on %* in /usr/bin/stap)\n sdt_stap:client__end (on %* in /usr/bin/stap)\n \n You can now use it in all perf tools, such as:\n \n perf record -e sdt_stap:client__end -aR sleep 1\n \n #\n \n From these we\u0027re use the two below to run systemtap\u0027s test suite:\n \n # perf record -e sdt_stap:pass2__*,cycles:P make installcheck \u003e /dev/null\n ^C[ perf record: Woken up 8 times to write data ]\n [ perf record: Captured and wrote 2.691 MB perf.data (39638 samples) ]\n Terminated\n # perf script | grep sdt_stap\n stap 28979 [000] 19424.302660: sdt_stap:pass2__start: (561b9a537de3) arg1\u003d140730364262544\n stap 28979 [000] 19424.333083: sdt_stap:pass2__end: (561b9a53a9e1) arg1\u003d140730364262544\n stap 29045 [006] 19424.933460: sdt_stap:pass2__start: (563edddcede3) arg1\u003d140722674883152\n stap 29045 [006] 19424.963794: sdt_stap:pass2__end: (563edddd19e1) arg1\u003d140722674883152\n # perf script | grep cycles | wc -l\n 39634\n #\n \n Looking at the whole perf.data file:\n \n [root@quaco testsuite]# perf report | grep cycles:P -A25\n # Samples: 39K of event \u0027cycles:P\u0027\n # Event count (approx.): 34044267368\n #\n # Overhead Command Shared Object Symbol\n # ........ ....... .................... ................................\n #\n 3.50% cc1 cc1 [.] ht_lookup_with_hash\n 3.04% cc1 cc1 [.] _cpp_lex_token\n 2.11% cc1 cc1 [.] ggc_internal_alloc\n 1.83% cc1 cc1 [.] cpp_get_token_with_location\n 1.68% cc1 libc-2.29.so [.] _int_malloc\n 1.41% cc1 cc1 [.] linemap_position_for_column\n 1.25% cc1 cc1 [.] ggc_internal_cleared_alloc\n 1.20% cc1 cc1 [.] c_lex_with_flags\n 1.18% cc1 cc1 [.] get_combined_adhoc_loc\n 1.05% cc1 libc-2.29.so [.] malloc\n 1.01% cc1 libc-2.29.so [.] _int_free\n 0.96% stap stap [.] std::_Hashtable\u003cstd::__cxx11::basic_string\u003cchar, std::char_traits\u003cchar\u003e, std::allocator\u003cchar\u003e \u003e, std::__cxx11::basic_string\u003cchar, std::char_traits\u003cchar\u003e, std::allocator\u003cchar\u003e \u003e, std::allocator\u003cstd::__cxx11::basic_string\u003cchar, std::char_traits\u003cchar\u003e, std::allocator\u003cchar\u003e \u003e \u003e, std::__detail::_Identity, std::equal_to\u003cstd::__cxx11::basic_string\u003cchar, std::char_traits\u003cchar\u003e, std::allocator\u003cchar\u003e \u003e \u003e, stringtable_hash, std::__detail::_Mod_range_hashing, std::__detail::_Default_ranged_hash, std::__detail::_Prime_rehash_policy, std::__detail::_Hashtable_traits\u003ctrue, true, true\u003e \u003e::_M_insert\u003cstd::__cxx11::basic_string\u003cchar, std::char_traits\u003cchar\u003e, std::allocator\u003cchar\u003e \u003e const\u0026, std::__detail::_AllocNode\u003cstd::allocator\u003cstd::__detail::_Hash_node\u003cstd::__cxx11::basic_string\u003cchar, std::char_traits\u003cchar\u003e, std::allocator\u003cchar\u003e \u003e, true\u003e \u003e \u003e \u003e\n 0.78% stap stap [.] lexer::scan\n 0.74% cc1 cc1 [.] _cpp_lex_direct\n 0.70% cc1 cc1 [.] pop_scope\n 0.70% cc1 cc1 [.] c_parser_declspecs\n 0.69% stap libc-2.29.so [.] _int_malloc\n 0.68% cc1 cc1 [.] htab_find_slot\n 0.68% cc1 [kernel.vmlinux] [k] prepare_exit_to_usermode\n 0.64% cc1 [kernel.vmlinux] [k] clear_page_erms\n [root@quaco testsuite]#\n \n And now only what happens in slices demarcated by those start/end SDT\n events:\n \n [root@quaco testsuite]# perf report --switch-on\u003dsdt_stap:pass2__start --switch-off\u003dsdt_stap:pass2__end | grep cycles:P -A100\n # Samples: 240 of event \u0027cycles:P\u0027\n # Event count (approx.): 206491934\n #\n # Overhead Command Shared Object Symbol\n # ........ ....... ................... ................................................\n #\n 38.99% stap stap [.] systemtap_session::register_library_aliases\n 19.47% stap stap [.] match_key::operator\u003c\n 15.01% stap libc-2.29.so [.] __memcmp_avx2_movbe\n 5.19% stap libc-2.29.so [.] _int_malloc\n 2.50% stap libstdc++.so.6.0.26 [.] std::_Rb_tree_insert_and_rebalance\n 2.30% stap stap [.] match_node::build_no_more\n 2.07% stap libc-2.29.so [.] malloc\n 1.66% stap stap [.] std::_Rb_tree\u003cmatch_key, std::pair\u003cmatch_key const, match_node*\u003e, std::_Select1st\u003cstd::pair\u003cmatch_key const, match_node*\u003e \u003e, std::less\u003cmatch_key\u003e, std::allocator\u003cstd::pair\u003cmatch_key const, match_node*\u003e \u003e \u003e::find\n 1.66% stap stap [.] match_node::bind\n 1.58% stap [kernel.vmlinux] [k] prepare_exit_to_usermode\n 1.17% stap [kernel.vmlinux] [k] native_irq_return_iret\n 0.87% stap stap [.] 0x0000000000032ec4\n 0.77% stap libstdc++.so.6.0.26 [.] std::_Rb_tree_increment\n 0.47% stap stap [.] std::vector\u003cderived_probe_builder*, std::allocator\u003cderived_probe_builder*\u003e \u003e::_M_realloc_insert\u003cderived_probe_builder* const\u0026\u003e\n 0.47% stap [kernel.vmlinux] [k] get_page_from_freelist\n 0.47% stap [kernel.vmlinux] [k] swapgs_restore_regs_and_return_to_usermode\n 0.47% stap [kernel.vmlinux] [k] do_user_addr_fault\n 0.46% stap [kernel.vmlinux] [k] __pagevec_lru_add_fn\n 0.46% stap stap [.] std::_Rb_tree\u003cmatch_key, std::pair\u003cmatch_key const, match_node*\u003e, std::_Select1st\u003cstd::pair\u003cmatch_key const, match_node*\u003e \u003e, std::less\u003cmatch_key\u003e, std::allocator\u003cstd::pair\u003cmatch_key const, match_node*\u003e \u003e \u003e::_M_emplace_unique\u003cstd::pair\u003cmatch_key, match_node*\u003e \u003e\n 0.42% stap libstdc++.so.6.0.26 [.] 0x00000000000c18fa\n 0.40% stap [kernel.vmlinux] [k] interrupt_entry\n 0.40% stap [kernel.vmlinux] [k] update_load_avg\n 0.40% stap [kernel.vmlinux] [k] __intel_pmu_disable_all\n 0.40% stap [kernel.vmlinux] [k] clear_page_erms\n 0.39% stap [kernel.vmlinux] [k] __mod_node_page_state\n 0.39% stap [kernel.vmlinux] [k] error_entry\n 0.39% stap [kernel.vmlinux] [k] sync_regs\n 0.38% stap [kernel.vmlinux] [k] __handle_mm_fault\n 0.38% stap stap [.] derive_probes\n \n #\n # (Tip: System-wide collection from all CPUs: perf record -a)\n #\n [root@quaco testsuite]#\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Florian Weimer \u003cfweimer@redhat.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: William Cohen \u003cwcohen@redhat.com\u003e\n Link: https://lkml.kernel.org/n/tip-408hvumcnyn93a0auihnawew@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit e4c057d02c9ef0b39bb88544b0c3c0a0a4d55ed6\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Tue Aug 6 20:15:41 2019 -0300\n\n misc/sgi-gru: use mmu_notifier_get/put for struct gru_mm_struct\n \n GRU is already using almost the same algorithm as get/put, it even\n helpfully has a 10 year old comment to make this algorithm common, which\n is finally happening.\n \n There are a few differences and fixes from this conversion:\n - GRU used rcu not srcu to read the hlist\n - Unclear how the locking worked to prevent gru_register_mmu_notifier()\n from running concurrently with gru_drop_mmu_notifier() - this version is\n safe\n - GRU had a release function which only set a variable without any locking\n that skiped the synchronize_srcu during unregister which looks racey,\n but this makes it reliable via the integrated call_srcu().\n - It is unclear if the mmap_sem is actually held when\n __mmu_notifier_register() was called, lockdep will now warn if this is\n wrong\n \n Link: https://lore.kernel.org/r/20190806231548.25242-5-jgg@ziepe.ca\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Dimitri Sivanich \u003csivanich@hpe.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 2c7933f53f6bff7656e3324ca1a04e478bdc57c1\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Tue Aug 6 20:15:40 2019 -0300\n\n mm/mmu_notifiers: add a get/put scheme for the registration\n \n Many places in the kernel have a flow where userspace will create some\n object and that object will need to connect to the subsystem\u0027s\n mmu_notifier subscription for the duration of its lifetime.\n \n In this case the subsystem is usually tracking multiple mm_structs and it\n is difficult to keep track of what struct mmu_notifier\u0027s have been\n allocated for what mm\u0027s.\n \n Since this has been open coded in a variety of exciting ways, provide core\n functionality to do this safely.\n \n This approach uses the struct mmu_notifier_ops * as a key to determine if\n the subsystem has a notifier registered on the mm or not. If there is a\n registration then the existing notifier struct is returned, otherwise the\n ops-\u003ealloc_notifiers() is used to create a new per-subsystem notifier for\n the mm.\n \n The destroy side incorporates an async call_srcu based destruction which\n will avoid bugs in the callers such as commit 6d7c3cde93c1 (\"mm/hmm: fix\n use after free with struct hmm in the mmu notifiers\").\n \n Since we are inside the mmu notifier core locking is fairly simple, the\n allocation uses the same approach as for mmu_notifier_mm, the write side\n of the mmap_sem makes everything deterministic and we only need to do\n hlist_add_head_rcu() under the mm_take_all_locks(). The new users count\n and the discoverability in the hlist is fully serialized by the\n mmu_notifier_mm-\u003elock.\n \n Link: https://lore.kernel.org/r/20190806231548.25242-4-jgg@ziepe.ca\n Co-developed-by: Christoph Hellwig \u003chch@infradead.org\u003e\n Signed-off-by: Christoph Hellwig \u003chch@infradead.org\u003e\n Reviewed-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Tested-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 70df291bf81ffda47ff84e6e2da4fbe21f95a861\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Tue Aug 6 20:15:39 2019 -0300\n\n mm/mmu_notifiers: do not speculatively allocate a mmu_notifier_mm\n \n A prior commit e0f3c3f78da2 (\"mm/mmu_notifier: init notifier if necessary\")\n made an attempt at doing this, but had to be reverted as calling\n the GFP_KERNEL allocator under the i_mmap_mutex causes deadlock, see\n commit 35cfa2b0b491 (\"mm/mmu_notifier: allocate mmu_notifier in advance\").\n \n However, we can avoid that problem by doing the allocation only under\n the mmap_sem, which is already happening.\n \n Since all writers to mm-\u003emmu_notifier_mm hold the write side of the\n mmap_sem reading it under that sem is deterministic and we can use that to\n decide if the allocation path is required, without speculation.\n \n The actual update to mmu_notifier_mm must still be done under the\n mm_take_all_locks() to ensure read-side coherency.\n \n Link: https://lore.kernel.org/r/20190806231548.25242-3-jgg@ziepe.ca\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Tested-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 56c57103db17db9ecdad0507a3f0e3eea747fabe\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Tue Aug 6 20:15:38 2019 -0300\n\n mm/mmu_notifiers: hoist do_mmu_notifier_register down_write to the caller\n \n This simplifies the code to not have so many one line functions and extra\n logic. __mmu_notifier_register() simply becomes the entry point to\n register the notifier, and the other one calls it under lock.\n \n Also add a lockdep_assert to check that the callers are holding the lock\n as expected.\n \n Link: https://lore.kernel.org/r/20190806231548.25242-2-jgg@ziepe.ca\n Suggested-by: Christoph Hellwig \u003chch@infradead.org\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Tested-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 27235b15e4197b3a1014cf0ca0b08dbbf61a692b\nAuthor: Logan Gunthorpe \u003clogang@deltatee.com\u003e\nDate: Mon Aug 12 11:30:48 2019 -0600\n\n PCI/P2PDMA: Update pci_p2pdma_distance_many() documentation\n \n The comment describing pci_p2pdma_distance_many() still referred to\n the devices being behind the same root port. This no longer applies\n so reword the documentation.\n \n Link: https://lore.kernel.org/r/20190730163545.4915-15-logang@deltatee.com\n Link: https://lore.kernel.org/r/20190812173048.9186-15-logang@deltatee.com\n Signed-off-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit 5daf40b039249ef98d49610d14293dca3464b91f\nAuthor: Logan Gunthorpe \u003clogang@deltatee.com\u003e\nDate: Mon Aug 12 11:30:47 2019 -0600\n\n PCI/P2PDMA: Allow IOMMU for host bridge whitelist\n \n Now that we map the requests correctly we can remove the iommu_present()\n restriction.\n \n Link: https://lore.kernel.org/r/20190730163545.4915-14-logang@deltatee.com\n Link: https://lore.kernel.org/r/20190812173048.9186-14-logang@deltatee.com\n Signed-off-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit 5d52e1abcd47957f5f4cae26659768655e3ac9f5\nAuthor: Logan Gunthorpe \u003clogang@deltatee.com\u003e\nDate: Mon Aug 12 11:30:46 2019 -0600\n\n PCI/P2PDMA: dma_map() requests that traverse the host bridge\n \n Any requests that traverse the host bridge will need to be mapped into the\n IOMMU, so call dma_map_sg() inside pci_p2pdma_map_sg() when appropriate.\n \n Similarly, call dma_unmap_sg() inside pci_p2pdma_unmap_sg().\n \n Link: https://lore.kernel.org/r/20190730163545.4915-13-logang@deltatee.com\n Link: https://lore.kernel.org/r/20190812173048.9186-13-logang@deltatee.com\n Signed-off-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit 110203bee05f33637ccb44f046f14edaea94bb4c\nAuthor: Logan Gunthorpe \u003clogang@deltatee.com\u003e\nDate: Mon Aug 12 11:30:45 2019 -0600\n\n PCI/P2PDMA: Store mapping method in an xarray\n \n When upstream_bridge_distance() is called, store the method required to map\n the DMA transfers in an xarray so it can be looked up efficiently on the\n hot path in pci_p2pdma_map_sg().\n \n Link: https://lore.kernel.org/r/20190730163545.4915-12-logang@deltatee.com\n Link: https://lore.kernel.org/r/20190812173048.9186-12-logang@deltatee.com\n Signed-off-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit 142c242e6f12196f8064beb09198838611e029db\nAuthor: Logan Gunthorpe \u003clogang@deltatee.com\u003e\nDate: Mon Aug 12 11:30:44 2019 -0600\n\n PCI/P2PDMA: Factor out __pci_p2pdma_map_sg()\n \n Factor out the bus-only mapping into its own static function. No\n functional changes. The original pci_p2pdma_map_sg_attrs() will be used to\n decide whether this is an appropriate way to map.\n \n Link: https://lore.kernel.org/r/20190730163545.4915-11-logang@deltatee.com\n Link: https://lore.kernel.org/r/20190812173048.9186-11-logang@deltatee.com\n Signed-off-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit 7f73eac3a7137eabfb0c005c7ba55eb7994b9673\nAuthor: Logan Gunthorpe \u003clogang@deltatee.com\u003e\nDate: Mon Aug 12 11:30:43 2019 -0600\n\n PCI/P2PDMA: Introduce pci_p2pdma_unmap_sg()\n \n Add pci_p2pdma_unmap_sg() to the two places that call pci_p2pdma_map_sg().\n \n This is a prep patch to introduce correct mappings for p2pdma transactions\n that go through the root complex.\n \n Link: https://lore.kernel.org/r/20190730163545.4915-10-logang@deltatee.com\n Link: https://lore.kernel.org/r/20190812173048.9186-10-logang@deltatee.com\n Signed-off-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit 2b9f4bb2a4fb77da4862f9ddf5209de2bcdaa0c0\nAuthor: Logan Gunthorpe \u003clogang@deltatee.com\u003e\nDate: Mon Aug 12 11:30:42 2019 -0600\n\n PCI/P2PDMA: Add attrs argument to pci_p2pdma_map_sg()\n \n This is to match the dma_map_sg() API which this function will have to call\n in an future patch.\n \n Add a pci_p2pdma_map_sg_attrs() function and helper to call it with no\n attributes just like the dma_map_sg() function.\n \n Link: https://lore.kernel.org/r/20190730163545.4915-9-logang@deltatee.com\n Link: https://lore.kernel.org/r/20190812173048.9186-9-logang@deltatee.com\n Signed-off-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit 494d63b0d5d0a481f6ac23bc61e94647ab9c4390\nAuthor: Logan Gunthorpe \u003clogang@deltatee.com\u003e\nDate: Mon Aug 12 11:30:41 2019 -0600\n\n PCI/P2PDMA: Whitelist some Intel host bridges\n \n Intel devices do not have good support for P2P requests that span different\n host bridges as the transactions will cross the QPI/UPI bus and this does\n not perform well.\n \n Therefore, enable support for these devices only if the host bridges match.\n \n Add Intel devices that have been tested and are known to work. There are\n likely many others out there that will need to be tested and added.\n \n Link: https://lore.kernel.org/r/20190730163545.4915-8-logang@deltatee.com\n Link: https://lore.kernel.org/r/20190812173048.9186-8-logang@deltatee.com\n Signed-off-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit 2c84d818aee976390c055a417045a85c23d39662\nAuthor: Logan Gunthorpe \u003clogang@deltatee.com\u003e\nDate: Mon Aug 12 11:30:40 2019 -0600\n\n PCI/P2PDMA: Factor out host_bridge_whitelist()\n \n Push both PCI devices into the whitelist checking function seeing some\n hardware will require us ensuring they are on the same host bridge.\n \n At the same time we rename root_complex_whitelist() to\n host_bridge_whitelist() to match the terminology used in the code.\n \n Link: https://lore.kernel.org/r/20190730163545.4915-7-logang@deltatee.com\n Link: https://lore.kernel.org/r/20190812173048.9186-7-logang@deltatee.com\n Signed-off-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit e2cbfbf78968db3e3a71cb989af17f4f6448bf49\nAuthor: Logan Gunthorpe \u003clogang@deltatee.com\u003e\nDate: Mon Aug 12 11:30:39 2019 -0600\n\n PCI/P2PDMA: Apply host bridge whitelist for ACS\n \n When a P2PDMA transfer is rejected due to ACS being set, we can also check\n the whitelist and allow the transactions.\n \n Do this by pushing the whitelist check into the upstream_bridge_distance()\n function.\n \n Link: https://lore.kernel.org/r/20190730163545.4915-6-logang@deltatee.com\n Link: https://lore.kernel.org/r/20190812173048.9186-6-logang@deltatee.com\n Signed-off-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit c6bfaeb573a6caae4b96885b8bad63ee57495c77\nAuthor: Logan Gunthorpe \u003clogang@deltatee.com\u003e\nDate: Mon Aug 12 11:30:38 2019 -0600\n\n PCI/P2PDMA: Factor out __upstream_bridge_distance()\n \n This is a prep patch to create a second level helper. There are no\n functional changes.\n \n The root complex whitelist code will be moved into this function in a\n subsequent patch.\n \n Link: https://lore.kernel.org/r/20190730163545.4915-5-logang@deltatee.com\n Link: https://lore.kernel.org/r/20190812173048.9186-5-logang@deltatee.com\n Signed-off-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit 72583084e3fe333c27bc37527efe455d27eaf0b1\nAuthor: Logan Gunthorpe \u003clogang@deltatee.com\u003e\nDate: Mon Aug 12 11:30:37 2019 -0600\n\n PCI/P2PDMA: Add constants for map type results to upstream_bridge_distance()\n \n Add constant flags to indicate how two devices will be mapped or if they\n are unsupported. upstream_bridge_distance() will now return the\n mapping type and the distance in a passed-by-reference argument.\n \n This helps annotate the code better, but the main reason is so we can use\n the information to store the required mapping method in an xarray.\n \n Link: https://lore.kernel.org/r/20190730163545.4915-4-logang@deltatee.com\n Link: https://lore.kernel.org/r/20190812173048.9186-4-logang@deltatee.com\n Signed-off-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit 0afea3814358c97e2964710ba16fbb0b54ceda0e\nAuthor: Logan Gunthorpe \u003clogang@deltatee.com\u003e\nDate: Mon Aug 12 11:30:36 2019 -0600\n\n PCI/P2PDMA: Add provider\u0027s pci_dev to pci_p2pdma_pagemap struct\n \n The provider will be needed to figure out how to map a device.\n \n Link: https://lore.kernel.org/r/20190730163545.4915-3-logang@deltatee.com\n Link: https://lore.kernel.org/r/20190812173048.9186-3-logang@deltatee.com\n Signed-off-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit a6e6fe6549f609f5c00c91e988da9d25f8ae8604\nAuthor: Logan Gunthorpe \u003clogang@deltatee.com\u003e\nDate: Mon Aug 12 11:30:35 2019 -0600\n\n PCI/P2PDMA: Introduce private pagemap structure\n \n Move the PCI bus offset from the generic dev_pagemap structure to a\n specific pci_p2pdma_pagemap structure.\n \n This structure will grow in subsequent patches.\n \n Link: https://lore.kernel.org/r/20190730163545.4915-2-logang@deltatee.com\n Link: https://lore.kernel.org/r/20190812173048.9186-2-logang@deltatee.com\n Signed-off-by: Logan Gunthorpe \u003clogang@deltatee.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit 012bb028351419fb9cc481651ed363505207e9e0\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Aug 7 10:46:33 2019 +0200\n\n clocksource/drivers/renesas-ostm: Use DIV_ROUND_CLOSEST() helper\n \n Use the DIV_ROUND_CLOSEST() helper instead of open-coding the same\n operation.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 0b936e6122738f4cf474d1f3ff636cba0edc8b94\nAuthor: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\nDate: Fri Aug 2 08:34:05 2019 +0100\n\n drm: rcar-du: lvds: Fix bridge_to_rcar_lvds\n \n Using name \"bridge\" for macro bridge_to_rcar_lvds argument doesn\u0027t\n work when the pointer name used by the caller is not \"bridge\".\n Rename the argument to \"b\" to allow for any pointer name.\n \n While at it, fix the connector_to_rcar_lvds macro similarly.\n \n Fixes: c6a27fa41fab (\"drm: rcar-du: Convert LVDS encoder code to bridge driver\")\n Signed-off-by: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart+renesas@ideasonboard.com\u003e\n [Fix connector_to_rcar_lvds]\n Signed-off-by: Laurent Pinchart \u003claurent.pinchart+renesas@ideasonboard.com\u003e\n\ncommit dbbfaf5f2641af92cd06e94c91fff2898d26f422\nAuthor: Laurent Pinchart \u003claurent.pinchart+renesas@ideasonboard.com\u003e\nDate: Wed Jan 17 16:16:48 2018 +0200\n\n drm: Remove bridge support from legacy helpers\n \n DRM bridges are only used by atomic drivers, and none of them use the\n legacy helpers. Drop bridge support from those helpers to prepare for\n making the bridge operations atomic-aware.\n \n Signed-off-by: Laurent Pinchart \u003claurent.pinchart+renesas@ideasonboard.com\u003e\n Reviewed-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Acked-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n\ncommit 71f6bd791e0d6bec19306f5145d3c08617baf136\nAuthor: Laurent Pinchart \u003claurent.pinchart+renesas@ideasonboard.com\u003e\nDate: Wed Jan 17 15:54:15 2018 +0200\n\n drm: Don\u0027t include drm/drm_encoder_slave.h when not needed\n \n The dw-hdmi, kirin and imx drivers include the drm/drm_encoder_slave.h\n header but don\u0027t use the encoder slave API. Remove it or replace it with\n drm/drm_encoder.h as needed.\n \n Signed-off-by: Laurent Pinchart \u003claurent.pinchart+renesas@ideasonboard.com\u003e\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n\ncommit fc702ed49a8668a17343811ee28214d845bfc5e6\nAuthor: Andy Yan \u003candyshrk@gmail.com\u003e\nDate: Mon Aug 5 20:40:37 2019 +0800\n\n arm64: dts: rockchip: Add dts for Leez RK3399 P710 SBC\n \n P710 is a RK3399 based SBC, designed by Leez [0].\n \n Specification\n - Rockchip RK3399\n - 4/2GB LPDDR4\n - TF sd scard slot\n - eMMC\n - M.2 B-Key for 4G LTE\n - AP6256 for WiFi + BT\n - Gigabit ethernet\n - HDMI out\n - 40 pin header\n - USB 2.0 x 2\n - USB 3.0 x 1\n - USB 3.0 Type-C x 1\n - TYPE-C Power supply\n \n [0]https://leez.lenovo.com\n \n Signed-off-by: Andy Yan \u003candyshrk@gmail.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit d30bdfc0ecf885ca0495bdee522cb9d7ee005cf8\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Fri Aug 9 15:10:23 2019 +0800\n\n PM: sleep: Replace strncmp() with str_has_prefix()\n \n Use str_has_prefix() instead of strncmp() which is less\n straightforward in decode_state().\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n [ rjw: Subject \u0026 changelog ]\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit d5a1baddf1585885868cbab55989401fb97118c6\nAuthor: Rikard Falkeborn \u003crikard.falkeborn@gmail.com\u003e\nDate: Sun Aug 11 20:49:36 2019 +0200\n\n x86/boot: Use common BUILD_BUG_ON\n \n Defining BUILD_BUG_ON causes redefinition warnings when adding includes of\n include/linux/build_bug.h in files unrelated to x86/boot. For example,\n adding an include of build_bug.h to include/linux/bits.h shows the\n following warnings:\n \n CC arch/x86/boot/cpucheck.o\n In file included from ./include/linux/bits.h:22,\n from ./arch/x86/include/asm/msr-index.h:5,\n from arch/x86/boot/cpucheck.c:28:\n ./include/linux/build_bug.h:49: warning: \"BUILD_BUG_ON\" redefined\n 49 | #define BUILD_BUG_ON(condition) \\\n |\n In file included from arch/x86/boot/cpucheck.c:22:\n arch/x86/boot/boot.h:31: note: this is the location of the previous definition\n 31 | #define BUILD_BUG_ON(condition) ((void)sizeof(char[1 - 2*!!(condition)]))\n |\n \n The macro was added to boot.h in commit 62bd0337d0c4 (\"Top header file for\n new x86 setup code\"). At that time, BUILD_BUG_ON was defined in\n kernel.h. Presumably BUILD_BUG_ON was redefined to avoid pulling in\n kernel.h. Since then, BUILD_BUG_ON and similar macros have been split to a\n separate header file.\n \n Signed-off-by: Rikard Falkeborn \u003crikard.falkeborn@gmail.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Link: https://lkml.kernel.org/r/20190811184938.1796-2-rikard.falkeborn@gmail.com\n\ncommit 6e3360cd07ec2d4a53919dc62947f9769d8fdac4\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Thu Aug 15 14:20:18 2019 -0500\n\n ASoC: SOF: pcm: add ALH support\n \n Even if ALH has no specific configuration, we still need to handle the\n common parameters for all DAIs\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190815192018.30570-5-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 4d6bbf1a015c84ba05af9d8c59dd9c55cad4e668\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Thu Aug 15 14:20:17 2019 -0500\n\n ASoC: SOF: topology: initial support for Intel ALH DAI type\n \n The Audio Link Hub DAI does not require any static configuration from\n topology for now. We still need to pass the frame rate and format to\n firmware.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190815192018.30570-4-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 8207a1c4911226d0dd0592ffed5fce4a1f95e4b2\nAuthor: Bard liao \u003cyung-chuan.liao@linux.intel.com\u003e\nDate: Thu Aug 15 14:20:16 2019 -0500\n\n ASoC: SOF: rename SOUNDWIRE to ALH\n \n Rename SOUNDWIRE to ALH.\n \n Signed-off-by: Bard liao \u003cyung-chuan.liao@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190815192018.30570-3-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit f59b16ef4ccea1b52f1c4e4c60ce507dc0bcc0ad\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Thu Aug 15 14:20:15 2019 -0500\n\n ASoC: SOF: topology: Add dummy support for i.MX8 DAIs\n \n Add dummy support for SAI/ESAI digital audio interface\n IPs found on i.MX8 boards.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190815192018.30570-2-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 8661ab5b23d6d30d8687fc05bc1dba8f9a64b444\nAuthor: Shengjiu Wang \u003cshengjiu.wang@nxp.com\u003e\nDate: Fri Aug 16 01:03:14 2019 -0400\n\n ASoC: imx-audmux: Add driver suspend and resume to support MEGA Fast\n \n For i.MX6 SoloX, there is a mode of the SoC to shutdown all power\n source of modules during system suspend and resume procedure.\n Thus, AUDMUX needs to save all the values of registers before the\n system suspend and restore them after the system resume.\n \n Signed-off-by: Shengjiu Wang \u003cshengjiu.wang@nxp.com\u003e\n Link: https://lore.kernel.org/r/1565931794-7218-1-git-send-email-shengjiu.wang@nxp.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 22c65d80afb70538eccb96c0ebfd55fbae29478b\nAuthor: Andrius Štikonas \u003candrius@stikonas.eu\u003e\nDate: Sun Aug 11 22:11:08 2019 +0100\n\n arm64: dts: rockchip: enable internal SPI flash for RockPro64.\n \n The rockpro64 contains a nor-flash chip connected to spi1.\n \n Signed-off-by: Andrius Štikonas \u003candrius@stikonas.eu\u003e\n [a number of cleanups]\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit b016cd6ed4b772759804e0d6082bd1f5ca63b8ee\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Aug 14 19:24:01 2019 +0100\n\n dma-buf: Restore seqlock around dma_resv updates\n \n This reverts\n 67c97fb79a7f (\"dma-buf: add reservation_object_fences helper\")\n dd7a7d1ff2f1 (\"drm/i915: use new reservation_object_fences helper\")\n 0e1d8083bddb (\"dma-buf: further relax reservation_object_add_shared_fence\")\n 5d344f58da76 (\"dma-buf: nuke reservation_object seq number\")\n \n The scenario that defeats simply grabbing a set of shared/exclusive\n fences and using them blissfully under RCU is that any of those fences\n may be reallocated by a SLAB_TYPESAFE_BY_RCU fence slab cache. In this\n scenario, while keeping the rcu_read_lock we need to establish that no\n fence was changed in the dma_resv after a read (or full) memory barrier.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Acked-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190814182401.25009-1-chris@chris-wilson.co.uk\n\ncommit 3c29ba8e7c78dee1a8a59cadf235d6b93489a3b4\nAuthor: Douglas Anderson \u003cdianders@chromium.org\u003e\nDate: Thu Jun 20 11:20:56 2019 -0700\n\n Revert \"ARM: dts: rockchip: add startup delay to rk3288-veyron panel-regulators\"\n \n This reverts commit 1f45e8c6d0161f044d679f242fe7514e2625af4a.\n \n This 100 ms mystery delay is not on downstream kernels and no longer\n seems needed on upstream kernels either [1]. Presumably something in the\n meantime has made things better. A few possibilities for patches that\n have landed in the meantime that could have made this better are\n commit 3157694d8c7f (\"pwm-backlight: Add support for PWM delays\n proprieties.\"), commit 5fb5caee92ba (\"pwm-backlight: Enable/disable\n the PWM before/after LCD enable toggle.\"), and commit 6d5922dd0d60\n (\"ARM: dts: rockchip: set PWM delay backlight settings for Veyron\")\n \n Let\u0027s revert and get our 100 ms back.\n \n [1] https://lkml.kernel.org/r/2226970.BAPq4liE1j@diego\n \n Signed-off-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n [rebased on top of the recent veyron display cleanup]\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit 109e8adfbc665a3e463043cefd5d17e552dc020c\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Fri Aug 16 14:17:34 2019 +0300\n\n platform/x86: asus-wmi: Remove unnecessary blank lines\n \n Remove blank lines where they are not needed.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 127e1dfc66cd0ea9ebd0fbf95c817b1b8d1d408b\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Fri Aug 16 14:07:17 2019 +0300\n\n platform/x86: asus-wmi: Drop indentation level by inverting conditionals\n \n We have extra indentation level where it can be avoided by changing conditional\n to the inverted one.\n \n Do it here for three such locations in the driver.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 2275752004ab216c667cfdba378c5199f9068e96\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Fri Aug 16 14:10:03 2019 +0300\n\n platform/x86: asus-wmi: Use clamp_val() instead of open coded variant\n \n There is no need to open code clamp_val() macro implementation.\n Replace the corresponding lines with direct call to clamp_val().\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit a5556fa1107d5b9cda75632c6ece1b4612dd1607\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Fri Aug 16 14:15:55 2019 +0300\n\n platform/x86: asus-wmi: Replace sscanf() with kstrtoint()\n \n The use of sscanf() is an overkill here. Moreover, there is no need to check\n for count to be 0, since it\u0027s guaranteed by sysfs not to be.\n \n Taking above into account, replace sscanf() with kstrtoint() calls.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 6ac689d2e3162f2673136b59ea29e18c4cb3fbf1\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 16 09:31:43 2019 +0100\n\n drm/i915: Use the associated uncore for the vm\n \n We store the gt\u0026uncore to use in the i915_address_space, so use it!\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816083143.23558-1-chris@chris-wilson.co.uk\n\ncommit 6d03c140db2e9296407bef470473e0e8be31b2c5\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Thu Aug 15 14:59:03 2019 +0200\n\n USB: phy: fsl-usb: convert platform driver to use dev_groups\n \n Platform drivers now have the option to have the platform core create\n and remove any needed sysfs attribute files. So take advantage of that\n and do not register \"by hand\" any sysfs files.\n \n Acked-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190815125903.GA17065@kroah.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7c0a69394c265f2bb674c3f5daadfdd5c15ea0d1\nAuthor: Hui Wang \u003chui.wang@canonical.com\u003e\nDate: Fri Aug 16 14:27:40 2019 +0800\n\n ALSA: hda - Define a fallback_pin_fixup_tbl for alc269 family\n \n We have another Dell laptop which needs the DELL4_MIC_NO_PRESENCE,\n and this laptop has different pincfg definitions from existing\n ones in the pintbl, rather adding a new entry, let us define\n a tbl in the fallback_pin_fixup_tbl and this tbl will match\n all dell machines with alc289 codec and the pins of 0x19 and 0x1b\n are undef by default.\n \n Signed-off-by: Hui Wang \u003chui.wang@canonical.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 0fc1e447e9e474c2460df8d0378f899b8813e1d2\nAuthor: Hui Wang \u003chui.wang@canonical.com\u003e\nDate: Fri Aug 16 14:27:39 2019 +0800\n\n ALSA: hda - Expand pin_match function to match upcoming new tbls\n \n With the existing pintbl, we already have many entries in it. it is\n better to figure out a new way to reduce the size of the pintbl.\n \n We plan to define a new tbl which will match more machines with a\n single tbl, To do that, this function doesn\u0027t need to match all valid\n pins between machine and tbl, it just needs to match all pins defined\n in the tbl with the machine.\n \n And the plan is to move some tbls from pin_fixup_tbl to\n fallback_pin_fixup_tbl gradually.\n \n Signed-off-by: Hui Wang \u003chui.wang@canonical.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit b7f81cc13e0b050829619690eb740fdaf60a0ef1\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Thu Aug 15 20:38:44 2019 -0400\n\n arm64: dts: imx8mq: Add system counter node\n \n Add i.MX8MQ system counter node to enable timer-imx-sysctr\n broadcast timer driver.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 5820af9f6c9960ca2d09a3c639006f5cba2cebf5\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Thu Aug 15 20:38:43 2019 -0400\n\n arm64: dts: imx8mm: Add system counter node\n \n Add i.MX8MM system counter node to enable timer-imx-sysctr\n broadcast timer driver.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 8b75a1af78ae4598781c7a7fd6610ce914581538\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Thu Aug 15 20:38:42 2019 -0400\n\n clocksource/drivers/imx-sysctr: Add internal clock divider handle\n \n The system counter block guide states that the base clock is\n internally divided by 3 before use, that means the clock input of\n system counter defined in DT should be base clock which is normally\n from OSC, and then internally divided by 3 before use.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit c22f40c8e5e72354caee845c01ea88aff053e283\nAuthor: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\nDate: Tue Aug 13 15:30:50 2019 +0200\n\n clocksource/drivers/tcb_clksrc: Register delay timer\n \n Implement and register delay timer to allow get_cycles() to work properly.\n \n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit dcfc2f1cc312bd96d3efc42f28e46874e5b533f4\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jul 22 10:12:22 2019 +0200\n\n dt-bindings: timer: Convert Allwinner A13 HSTimer to a schema\n \n The newer Allwinner SoCs have a High Speed Timer supported in Linux, with a\n matching Device Tree binding.\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for that controller over to a YAML schemas.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit d700d8b6a54bb19628b188ed203ad0b014bb4da6\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jul 22 10:12:21 2019 +0200\n\n clocksource: sun4i: Add missing compatibles\n \n Newer Allwinner SoCs have different number of interrupts, let\u0027s add\n different compatibles for all of them to deal with this properly.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Acked-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 625fa209a22da60069b5923e6239771def5b532f\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jul 22 10:12:20 2019 +0200\n\n dt-bindings: timer: Add missing compatibles\n \n Newer Allwinner SoCs have different number of interrupts, let\u0027s add\n different compatibles for all of them to deal with this properly.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 2912a039d7260073dfb47472482646abfc7a51d9\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jul 22 10:12:19 2019 +0200\n\n dt-bindings: timer: Convert Allwinner A10 Timer to a schema\n \n The older Allwinner SoCs have a Timer supported in Linux, with a matching\n Device Tree binding.\n \n While the original binding only mentions one interrupt, the timer actually\n has 6 of them.\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for that controller over to a YAML schemas.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 128fc761306cb259c543737d532bda76deb03fb8\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:04 2019 -0700\n\n clocksource: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n\ncommit 84d8e80b0a3612aad746837e95d8efe324fdc5c6\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Fri Aug 16 12:46:25 2019 +0300\n\n platform/x86: asus-wmi: Refactor charge_threshold_store()\n \n There are few issues with the current code:\n - the error code from kstrtouint() is shadowed\n - the error code from asus_wmi_set_devstate() is ignored\n - the extra check against 0 for count (this is guaranteed by sysfs)\n \n Fix these issues by doing a slight refactoring of charge_threshold_store().\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit d507a54f5865d8dcbdd16c66a1a2da15640878ca\nAuthor: Kristian Klausen \u003ckristian@klausen.dk\u003e\nDate: Mon Aug 5 21:23:05 2019 +0200\n\n platform/x86: asus-wmi: Add support for charge threshold\n \n Most newer ASUS laptops supports limiting the battery charge level, which\n help prolonging the battery life.\n \n Tested on a Zenbook UX430UNR.\n \n Signed-off-by: Kristian Klausen \u003ckristian@klausen.dk\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 02a5e9bc86dee8c7698ac2cfd9e29650eb318442\nAuthor: Kristian Klausen \u003ckristian@klausen.dk\u003e\nDate: Sun Aug 4 17:21:43 2019 +0200\n\n platform/x86: asus-nb-wmi: Support ALS on the Zenbook UX430UNR\n \n This patch adds support for ALS on the Zenbook UX430UNR to the asus_nb_wmi\n driver.\n \n Signed-off-by: Kristian Klausen \u003ckristian@klausen.dk\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit f078d053c8696d4b0884ee81d132dd22fca6a5c9\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Sun Aug 4 15:40:23 2019 +0200\n\n platform/x86: touchscreen_dmi: Add info for the Irbis TW90 tablet\n \n Add touchscreen info for the Irbis TW90 tablet.\n \n Reported-by: russianneuromancer@ya.ru\n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit a0c809792183228a1d269d4112a2b64f2c7f8f9f\nAuthor: Giang Le \u003cohaibuzzle@gmail.com\u003e\nDate: Sat Aug 3 21:12:22 2019 +0700\n\n platform/x86: touchscreen_dmi: Add info for the Chuwi Surbook Mini tablet\n \n Add touchscreen platform data for the Chuwi Surbook Mini tablet.\n \n Signed-off-by: Giang Le \u003cohaibuzzle@gmail.com\u003e\n Reviewed-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 8e7cb1799b4f8bde3e7d9c80bf689e5408add271\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 16 08:46:35 2019 +0100\n\n drm/i915: Extract intel_frontbuffer active tracking\n \n Move the active tracking for the frontbuffer operations out of the\n i915_gem_object and into its own first class (refcounted) object. In the\n process of detangling, we switch from low level request tracking to the\n easier i915_active -- with the plan that this avoids any potential\n atomic callbacks as the frontbuffer tracking wishes to sleep as it\n flushes.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190816074635.26062-1-chris@chris-wilson.co.uk\n\ncommit e03250061b54041d3502696db421c44a4a8039f4\nAuthor: Wei Yongjun \u003cweiyongjun1@huawei.com\u003e\nDate: Fri Aug 16 02:40:44 2019 +0000\n\n btf: fix return value check in btf_vmlinux_init()\n \n In case of error, the function kobject_create_and_add() returns NULL\n pointer not ERR_PTR(). The IS_ERR() test in the return value check\n should be replaced with NULL test.\n \n Fixes: 341dfcf8d78e (\"btf: expose BTF info through sysfs\")\n Signed-off-by: Wei Yongjun \u003cweiyongjun1@huawei.com\u003e\n Acked-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 8918dc42dc85ba6981028f65a989c478eb80bc02\nAuthor: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\nDate: Thu Aug 15 15:32:20 2019 +0100\n\n tools: bpftool: move \"__printf()\" attributes to header file\n \n Some functions in bpftool have a \"__printf()\" format attributes to tell\n the compiler they should expect printf()-like arguments. But because\n these attributes are not used for the function prototypes in the header\n files, the compiler does not run the checks everywhere the functions are\n used, and some mistakes on format string and corresponding arguments\n slipped in over time.\n \n Let\u0027s move the __printf() attributes to the correct places.\n \n Note: We add guards around the definition of GCC_VERSION in\n tools/include/linux/compiler-gcc.h to prevent a conflict in jit_disasm.c\n on GCC_VERSION from headers pulled via libbfd.\n \n Fixes: c101189bc968 (\"tools: bpftool: fix -Wmissing declaration warnings\")\n Reported-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit b0ead6d75a5b335287337e602e6b815e1115481c\nAuthor: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\nDate: Thu Aug 15 15:32:19 2019 +0100\n\n tools: bpftool: fix format string for p_err() in detect_common_prefix()\n \n There is one call to the p_err() function in detect_common_prefix()\n where the message to print is passed directly as the first argument,\n without using a format string. This is harmless, but may trigger\n warnings if the \"__printf()\" attribute is used correctly for the p_err()\n function. Let\u0027s fix it by using a \"%s\" format string.\n \n Fixes: ba95c7452439 (\"tools: bpftool: add \"prog run\" subcommand to test-run programs\")\n Signed-off-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 8a15d5ced8c626c0331974c7281c1d651f7b0d83\nAuthor: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\nDate: Thu Aug 15 15:32:18 2019 +0100\n\n tools: bpftool: fix format string for p_err() in query_flow_dissector()\n \n The format string passed to one call to the p_err() function in\n query_flow_dissector() does not match the value that should be printed,\n resulting in some garbage integer being printed instead of\n strerror(errno) if /proc/self/ns/net cannot be open. Let\u0027s fix the\n format string.\n \n Fixes: 7f0c57fec80f (\"bpftool: show flow_dissector attachment status\")\n Signed-off-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit ed4a3983cd3eb93aaf80de8d8a36efed808acff2\nAuthor: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\nDate: Thu Aug 15 15:32:17 2019 +0100\n\n tools: bpftool: fix argument for p_err() in BTF do_dump()\n \n The last argument passed to one call to the p_err() function is not\n correct, it should be \"*argv\" instead of \"**argv\". This may lead to a\n segmentation fault error if BTF id cannot be parsed correctly. Let\u0027s fix\n this.\n \n Fixes: c93cc69004dt (\"bpftool: add ability to dump BTF types\")\n Signed-off-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 22c349e8db89df86804d3ba23cef037ccd44a8bf\nAuthor: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\nDate: Thu Aug 15 15:32:16 2019 +0100\n\n tools: bpftool: fix format strings and arguments for jsonw_printf()\n \n There are some mismatches between format strings and arguments passed to\n jsonw_printf() in the BTF dumper for bpftool, which seems harmless but\n may result in warnings if the \"__printf()\" attribute is used correctly\n for jsonw_printf(). Let\u0027s fix relevant format strings and type cast.\n \n Fixes: b12d6ec09730 (\"bpf: btf: add btf print functionality\")\n Signed-off-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 9def249dc8409ffc1f5a1d7195f1c462f2b49c07\nAuthor: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\nDate: Thu Aug 15 15:32:15 2019 +0100\n\n tools: bpftool: fix arguments for p_err() in do_event_pipe()\n \n The last argument passed to some calls to the p_err() functions is not\n correct, it should be \"*argv\" instead of \"**argv\". This may lead to a\n segmentation fault error if CPU IDs or indices from the command line\n cannot be parsed correctly. Let\u0027s fix this.\n \n Fixes: f412eed9dfde (\"tools: bpftool: add simple perf event output reader\")\n Signed-off-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit da15c03b047dca891d37b9f4ef9ca14d84a6484f\nAuthor: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\nDate: Tue Aug 13 20:06:48 2019 +1000\n\n powerpc/xive: Implement get_irqchip_state method for XIVE to fix shutdown race\n \n Testing has revealed the existence of a race condition where a XIVE\n interrupt being shut down can be in one of the XIVE interrupt queues\n (of which there are up to 8 per CPU, one for each priority) at the\n point where free_irq() is called. If this happens, can return an\n interrupt number which has been shut down. This can lead to various\n symptoms:\n \n - irq_to_desc(irq) can be NULL. In this case, no end-of-interrupt\n function gets called, resulting in the CPU\u0027s elevated interrupt\n priority (numerically lowered CPPR) never gets reset. That then\n means that the CPU stops processing interrupts, causing device\n timeouts and other errors in various device drivers.\n \n - The irq descriptor or related data structures can be in the process\n of being freed as the interrupt code is using them. This typically\n leads to crashes due to bad pointer dereferences.\n \n This race is basically what commit 62e0468650c3 (\"genirq: Add optional\n hardware synchronization for shutdown\", 2019-06-28) is intended to\n fix, given a get_irqchip_state() method for the interrupt controller\n being used. It works by polling the interrupt controller when an\n interrupt is being freed until the controller says it is not pending.\n \n With XIVE, the PQ bits of the interrupt source indicate the state of\n the interrupt source, and in particular the P bit goes from 0 to 1 at\n the point where the hardware writes an entry into the interrupt queue\n that this interrupt is directed towards. Normally, the code will then\n process the interrupt and do an end-of-interrupt (EOI) operation which\n will reset PQ to 00 (assuming another interrupt hasn\u0027t been generated\n in the meantime). However, there are situations where the code resets\n P even though a queue entry exists (for example, by setting PQ to 01,\n which disables the interrupt source), and also situations where the\n code leaves P at 1 after removing the queue entry (for example, this\n is done for escalation interrupts so they cannot fire again until\n they are explicitly re-enabled).\n \n The code already has a \u0027saved_p\u0027 flag for the interrupt source which\n indicates that a queue entry exists, although it isn\u0027t maintained\n consistently. This patch adds a \u0027stale_p\u0027 flag to indicate that\n P has been left at 1 after processing a queue entry, and adds code\n to set and clear saved_p and stale_p as necessary to maintain a\n consistent indication of whether a queue entry may or may not exist.\n \n With this, we can implement xive_get_irqchip_state() by looking at\n stale_p, saved_p and the ESB PQ bits for the interrupt.\n \n There is some additional code to handle escalation interrupts\n properly; because they are enabled and disabled in KVM assembly code,\n which does not have access to the xive_irq_data struct for the\n escalation interrupt. Hence, stale_p may be incorrect when the\n escalation interrupt is freed in kvmppc_xive_{,native_}cleanup_vcpu().\n Fortunately, we can fix it up by looking at vcpu-\u003earch.xive_esc_on,\n with some careful attention to barriers in order to ensure the correct\n result if xive_esc_irq() races with kvmppc_xive_cleanup_vcpu().\n \n Finally, this adds code to make noise on the console (pr_crit and\n WARN_ON(1)) if we find an interrupt queue entry for an interrupt\n which does not have a descriptor. While this won\u0027t catch the race\n reliably, if it does get triggered it will be an indication that\n the race is occurring and needs to be debugged.\n \n Fixes: 243e25112d06 (\"powerpc/xive: Native exploitation of the XIVE interrupt controller\")\n Cc: stable@vger.kernel.org # v4.12+\n Signed-off-by: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190813100648.GE9567@blackberry\n\ncommit 8d4ba9c931bc384bcc6889a43915aaaf19d3e499\nAuthor: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\nDate: Tue Aug 13 20:01:00 2019 +1000\n\n KVM: PPC: Book3S HV: Don\u0027t push XIVE context when not using XIVE device\n \n At present, when running a guest on POWER9 using HV KVM but not using\n an in-kernel interrupt controller (XICS or XIVE), for example if QEMU\n is run with the kernel_irqchip\u003doff option, the guest entry code goes\n ahead and tries to load the guest context into the XIVE hardware, even\n though no context has been set up.\n \n To fix this, we check that the \"CAM word\" is non-zero before pushing\n it to the hardware. The CAM word is initialized to a non-zero value\n in kvmppc_xive_connect_vcpu() and kvmppc_xive_native_connect_vcpu(),\n and is now cleared in kvmppc_xive_{,native_}cleanup_vcpu.\n \n Fixes: 5af50993850a (\"KVM: PPC: Book3S HV: Native usage of the XIVE interrupt controller\")\n Cc: stable@vger.kernel.org # v4.12+\n Reported-by: Cédric Le Goater \u003cclg@kaod.org\u003e\n Signed-off-by: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\n Reviewed-by: Cédric Le Goater \u003cclg@kaod.org\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190813100100.GC9567@blackberry\n\ncommit 959c5d5134786b4988b6fdd08e444aa67d1667ed\nAuthor: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\nDate: Tue Aug 13 20:03:49 2019 +1000\n\n KVM: PPC: Book3S HV: Fix race in re-enabling XIVE escalation interrupts\n \n Escalation interrupts are interrupts sent to the host by the XIVE\n hardware when it has an interrupt to deliver to a guest VCPU but that\n VCPU is not running anywhere in the system. Hence we disable the\n escalation interrupt for the VCPU being run when we enter the guest\n and re-enable it when the guest does an H_CEDE hypercall indicating\n it is idle.\n \n It is possible that an escalation interrupt gets generated just as we\n are entering the guest. In that case the escalation interrupt may be\n using a queue entry in one of the interrupt queues, and that queue\n entry may not have been processed when the guest exits with an H_CEDE.\n The existing entry code detects this situation and does not clear the\n vcpu-\u003earch.xive_esc_on flag as an indication that there is a pending\n queue entry (if the queue entry gets processed, xive_esc_irq() will\n clear the flag). There is a comment in the code saying that if the\n flag is still set on H_CEDE, we have to abort the cede rather than\n re-enabling the escalation interrupt, lest we end up with two\n occurrences of the escalation interrupt in the interrupt queue.\n \n However, the exit code doesn\u0027t do that; it aborts the cede in the sense\n that vcpu-\u003earch.ceded gets cleared, but it still enables the escalation\n interrupt by setting the source\u0027s PQ bits to 00. Instead we need to\n set the PQ bits to 10, indicating that an interrupt has been triggered.\n We also need to avoid setting vcpu-\u003earch.xive_esc_on in this case\n (i.e. vcpu-\u003earch.xive_esc_on seen to be set on H_CEDE) because\n xive_esc_irq() will run at some point and clear it, and if we race with\n that we may end up with an incorrect result (i.e. xive_esc_on set when\n the escalation interrupt has just been handled).\n \n It is extremely unlikely that having two queue entries would cause\n observable problems; theoretically it could cause queue overflow, but\n the CPU would have to have thousands of interrupts targetted to it for\n that to be possible. However, this fix will also make it possible to\n determine accurately whether there is an unhandled escalation\n interrupt in the queue, which will be needed by the following patch.\n \n Fixes: 9b9b13a6d153 (\"KVM: PPC: Book3S HV: Keep XIVE escalation interrupt masked unless ceded\")\n Cc: stable@vger.kernel.org # v4.16+\n Signed-off-by: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190813100349.GD9567@blackberry\n\ncommit 237aed48c642328ff0ab19b63423634340224a06\nAuthor: Cédric Le Goater \u003cclg@kaod.org\u003e\nDate: Tue Aug 6 19:25:38 2019 +0200\n\n KVM: PPC: Book3S HV: XIVE: Free escalation interrupts before disabling the VP\n \n When a vCPU is brought done, the XIVE VP (Virtual Processor) is first\n disabled and then the event notification queues are freed. When freeing\n the queues, we check for possible escalation interrupts and free them\n also.\n \n But when a XIVE VP is disabled, the underlying XIVE ENDs also are\n disabled in OPAL. When an END (Event Notification Descriptor) is\n disabled, its ESB pages (ESn and ESe) are disabled and loads return all\n 1s. Which means that any access on the ESB page of the escalation\n interrupt will return invalid values.\n \n When an interrupt is freed, the shutdown handler computes a \u0027saved_p\u0027\n field from the value returned by a load in xive_do_source_set_mask().\n This value is incorrect for escalation interrupts for the reason\n described above.\n \n This has no impact on Linux/KVM today because we don\u0027t make use of it\n but we will introduce in future changes a xive_get_irqchip_state()\n handler. This handler will use the \u0027saved_p\u0027 field to return the state\n of an interrupt and \u0027saved_p\u0027 being incorrect, softlockup will occur.\n \n Fix the vCPU cleanup sequence by first freeing the escalation interrupts\n if any, then disable the XIVE VP and last free the queues.\n \n Fixes: 90c73795afa2 (\"KVM: PPC: Book3S HV: Add a new KVM device for the XIVE native exploitation mode\")\n Fixes: 5af50993850a (\"KVM: PPC: Book3S HV: Native usage of the XIVE interrupt controller\")\n Cc: stable@vger.kernel.org # v4.12+\n Signed-off-by: Cédric Le Goater \u003cclg@kaod.org\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190806172538.5087-1-clg@kaod.org\n\ncommit dadb81d0afe732a7670f7c1bd287dada163a9f2f\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Wed Aug 14 13:05:48 2019 -0700\n\n libbpf: make libbpf.map source of truth for libbpf version\n \n Currently libbpf version is specified in 2 places: libbpf.map and\n Makefile. They easily get out of sync and it\u0027s very easy to update one,\n but forget to update another one. In addition, Github projection of\n libbpf has to maintain its own version which has to be remembered to be\n kept in sync manually, which is very error-prone approach.\n \n This patch makes libbpf.map a source of truth for libbpf version and\n uses shell invocation to parse out correct full and major libbpf version\n to use during build. Now we need to make sure that once new release\n cycle starts, we need to add (initially) empty section to libbpf.map\n with correct latest version.\n \n This also will make it possible to keep Github projection consistent\n with kernel sources version of libbpf by adopting similar parsing of\n version from libbpf.map.\n \n v2-\u003ev3:\n - grep -o + sort -rV (Andrey);\n \n v1-\u003ev2:\n - eager version vars evaluation (Jakub);\n - simplified version regex (Andrey);\n \n Cc: Andrey Ignatov \u003crdna@fb.com\u003e\n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Andrey Ignatov \u003crdna@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit cb9d9968661662a9743e50a32006d5e9d732d346\nAuthor: Daniel T. Lee \u003cdanieltimlee@gmail.com\u003e\nDate: Tue Aug 13 11:46:21 2019 +0900\n\n tools: bpftool: add documentation for net attach/detach\n \n Since, new sub-command \u0027net attach/detach\u0027 has been added for\n attaching XDP program on interface,\n this commit documents usage and sample output of `net attach/detach`.\n \n Signed-off-by: Daniel T. Lee \u003cdanieltimlee@gmail.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 10a708c24a31ae1be1ea23d1c38da2691d1fd65c\nAuthor: Daniel T. Lee \u003cdanieltimlee@gmail.com\u003e\nDate: Tue Aug 13 11:46:20 2019 +0900\n\n tools: bpftool: add bash-completion for net attach/detach\n \n This commit adds bash-completion for new \"net attach/detach\"\n subcommand for attaching XDP program on interface.\n \n Signed-off-by: Daniel T. Lee \u003cdanieltimlee@gmail.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 37c7f863ba92f0a7aa8bdfde9dfb6c392393fb83\nAuthor: Daniel T. Lee \u003cdanieltimlee@gmail.com\u003e\nDate: Tue Aug 13 11:46:19 2019 +0900\n\n tools: bpftool: add net detach command to detach XDP on interface\n \n By this commit, using `bpftool net detach`, the attached XDP prog can\n be detached. Detaching the BPF prog will be done through libbpf\n \u0027bpf_set_link_xdp_fd\u0027 with the progfd set to -1.\n \n Acked-by: Yonghong Song \u003cyhs@fb.com\u003e\n Signed-off-by: Daniel T. Lee \u003cdanieltimlee@gmail.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 04949ccc273e14d28aa004ac6954af0898ca3d12\nAuthor: Daniel T. Lee \u003cdanieltimlee@gmail.com\u003e\nDate: Tue Aug 13 11:46:18 2019 +0900\n\n tools: bpftool: add net attach command to attach XDP on interface\n \n By this commit, using `bpftool net attach`, user can attach XDP prog on\n interface. New type of enum \u0027net_attach_type\u0027 has been made, as stat ted at\n cover-letter, the meaning of \u0027attach\u0027 is, prog will be attached on interface.\n \n With \u0027overwrite\u0027 option at argument, attached XDP program could be replaced.\n Added new helper \u0027net_parse_dev\u0027 to parse the network device at argument.\n \n BPF prog will be attached through libbpf \u0027bpf_set_link_xdp_fd\u0027.\n \n Acked-by: Yonghong Song \u003cyhs@fb.com\u003e\n Signed-off-by: Daniel T. Lee \u003cdanieltimlee@gmail.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 4e3b0468e6d7f3c38ce52320803769b49ba87930\nAuthor: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\nDate: Mon Aug 12 16:45:37 2019 +0200\n\n net: mscc: PTP Hardware Clock (PHC) support\n \n This patch adds support for PTP Hardware Clock (PHC) to the Ocelot\n switch for both PTP 1-step and 2-step modes.\n \n Signed-off-by: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\n Acked-by: Richard Cochran \u003crichardcochran@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1f0239de58d60e0c4b44a35953004d180dd9ce6c\nAuthor: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\nDate: Mon Aug 12 16:45:36 2019 +0200\n\n net: mscc: remove the frame_info cpuq member\n \n In struct frame_info, the cpuq member is never used. This cosmetic patch\n removes it from the structure, and from the parsing of the frame header\n as it\u0027s only set but never used.\n \n Signed-off-by: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d8c964dccae30dc326e2bd7170d6919858bc5cff\nAuthor: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\nDate: Mon Aug 12 16:45:35 2019 +0200\n\n net: mscc: improve the frame header parsing readability\n \n This cosmetic patch improves the frame header parsing readability by\n introducing a new macro to access and mask its fields.\n \n Signed-off-by: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 45bce1719cbd1ad3fb48f840accb1d120b6b0209\nAuthor: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\nDate: Mon Aug 12 16:45:34 2019 +0200\n\n net: mscc: describe the PTP register range\n \n This patch adds support for using the PTP register range, and adds a\n description of its registers. This bank is used when configuring PTP.\n \n Signed-off-by: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 744350b98c3aa0295df84ce75603813a56423943\nAuthor: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\nDate: Mon Aug 12 16:45:33 2019 +0200\n\n Documentation/bindings: net: ocelot: document the PTP ready IRQ\n \n One additional interrupt needs to be described within the Ocelot device\n tree node: the PTP ready one. This patch documents the binding needed to\n do so.\n \n Signed-off-by: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit e36ea63b62ed7ec6dc644ca96830b72b577ebf18\nAuthor: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\nDate: Mon Aug 12 16:45:32 2019 +0200\n\n Documentation/bindings: net: ocelot: document the PTP bank\n \n One additional register range needs to be described within the Ocelot\n device tree node: the PTP. This patch documents the binding needed to do\n so.\n \n Signed-off-by: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c49a3a9cf26709d00211abbf79c2fd694ea4e86d\nAuthor: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\nDate: Wed Jul 3 17:56:39 2019 +0300\n\n soc: fsl: dpio: remove explicit device_link_remove\n \n Starting with commit 72175d4ea4c4 (\"driver core: Make driver core\n own stateful device links\") stateful device links are owned by the\n driver core and should not be explicitly removed on device unbind.\n Delete all device_link_remove appearances from the dpio driver.\n \n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Signed-off-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n\ncommit e5dadff4b09376e8ed92ecc0c12f1b9b3b1fbd19\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 15 21:57:09 2019 +0100\n\n drm/i915: Protect request retirement with timeline-\u003emutex\n \n Forgo the struct_mutex requirement for request retirement as we have\n been transitioning over to only using the timeline-\u003emutex for\n controlling the lifetime of a request on that timeline.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190815205709.24285-4-chris@chris-wilson.co.uk\n\ncommit ccb23d2dcc300d3fa176de654610ea82f635915d\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 15 21:57:08 2019 +0100\n\n drm/i915/gt: Guard timeline pinning without relying on struct_mutex\n \n In preparation for removing struct_mutex from around context retirement,\n we need to make timeline pinning and unpinning safe. Since multiple\n engines/contexts can share a single timeline, we cannot rely on\n borrowing the context mutex (otherwise we could state that the timeline\n is only pinned/unpinned inside the context pin/unpin and so guarded by\n it). However, we only perform a sequence of atomic operations inside the\n timeline pin/unpin and the sequence of those operations is safe for a\n concurrent unpin / pin, so we can relax the struct_mutex requirement.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190815205709.24285-3-chris@chris-wilson.co.uk\n\ncommit 338aade97c061da0403dbb00128d938782f30dbd\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 15 21:57:07 2019 +0100\n\n drm/i915/gt: Convert timeline tracking to spinlock\n \n Convert the active_list manipulation of timelines to use spinlocks so\n that we can perform the updates from underneath a quick interrupt\n callback, if need be.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190815205709.24285-2-chris@chris-wilson.co.uk\n\ncommit 531958f6f357cf7f812974152954484ff77d3964\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 15 21:57:06 2019 +0100\n\n drm/i915/gt: Track timeline activeness in enter/exit\n \n Lift moving the timeline to/from the active_list on enter/exit in order\n to shorten the active tracking span in comparison to the existing\n pin/unpin.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190815205709.24285-1-chris@chris-wilson.co.uk\n\ncommit 80f97619f5034888746ed11b01d275a90d75f7ee\nAuthor: Yinbo Zhu \u003cyinbo.zhu@nxp.com\u003e\nDate: Wed Aug 14 15:26:47 2019 +0800\n\n soc: fsl: guts: Add definition for LS1028A\n \n Adding compatible string \"ls1028a-dcfg\" to initialize guts driver\n for ls1028 and SoC die attribute definition for LS1028A\n \n Signed-off-by: Yinbo Zhu \u003cyinbo.zhu@nxp.com\u003e\n Signed-off-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n\ncommit 6ea4c0fe457094a62572a0b7ed781456a45b9a0d\nAuthor: Roy Pledge \u003croy.pledge@nxp.com\u003e\nDate: Thu Aug 1 20:17:05 2019 +0000\n\n soc/fsl/qbman: Update device tree with reserved memory\n \n When using the reserved memory node in the device tree there are\n two options - dynamic or static. If a dynamic allocation was\n selected (where the kernel selects the address for the allocation)\n convert it to a static allocation by inserting the reg property.\n This will ensure the same memory is reused after a kexec()\n \n Signed-off-by: Roy Pledge \u003croy.pledge@nxp.com\u003e\n Signed-off-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n\ncommit e844168a8f4a7de471059de3c4d60e4e770331e1\nAuthor: Roy Pledge \u003croy.pledge@nxp.com\u003e\nDate: Thu Aug 1 20:17:03 2019 +0000\n\n soc/fsl/qbman: Fixup qman_shutdown_fq()\n \n When shutting down a FQ on a dedicated channel only the\n SW portal associated with that channel can dequeue from it.\n Make sure the correct portal is use.\n \n Signed-off-by: Roy Pledge \u003croy.pledge@nxp.com\u003e\n Signed-off-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n\ncommit ea2b8488baa705be44e7c7ac8a42d33fd31315b7\nAuthor: Roy Pledge \u003croy.pledge@nxp.com\u003e\nDate: Thu Aug 1 20:17:01 2019 +0000\n\n soc/fsl/qbman: Disable interrupts during portal recovery\n \n Disable the QBMan interrupts during recovery.\n \n Signed-off-by: Roy Pledge \u003croy.pledge@nxp.com\u003e\n Signed-off-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n\ncommit 627da8bad530ec18c2111347fe559c5c21679220\nAuthor: Roy Pledge \u003croy.pledge@nxp.com\u003e\nDate: Thu Aug 1 20:16:59 2019 +0000\n\n soc/fsl/qbman: Fix drain_mr_fqni()\n \n The drain_mr_fqni() function may be called fron uninterruptable\n context so convert the msleep() to an mdelay(). Also ensure that\n the valid bit is updated while polling.\n \n Signed-off-by: Roy Pledge \u003croy.pledge@nxp.com\u003e\n Signed-off-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n\ncommit c5501aa92a881e599e1dc80f387d27763d07f67c\nAuthor: Roy Pledge \u003croy.pledge@nxp.com\u003e\nDate: Thu Aug 1 20:16:57 2019 +0000\n\n soc/fsl/qbman: Cleanup QMan queues if device was already initialized\n \n If the QMan device was previously initialized make sure all the\n frame queues are out of service once all the portals are probed.\n This handles the case where the kernel is restarted without the\n SoC being reset (kexec for example)\n \n Signed-off-by: Roy Pledge \u003croy.pledge@nxp.com\u003e\n Signed-off-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n\ncommit 0505d00c8dbaf06cb71e7a16818313ce1959e6d7\nAuthor: Roy Pledge \u003croy.pledge@nxp.com\u003e\nDate: Thu Aug 1 20:16:55 2019 +0000\n\n soc/fsl/qbman: Cleanup buffer pools if BMan was initialized prior to bootup\n \n Clean the BMan buffer pools if the device had been initialized\n previously. This will ensure a consistent state if the kernel\n was soft restarted (kexec for example)\n \n Signed-off-by: Roy Pledge \u003croy.pledge@nxp.com\u003e\n Signed-off-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n\ncommit 97777078d695d4294c999ad29363ff11f6384442\nAuthor: Roy Pledge \u003croy.pledge@nxp.com\u003e\nDate: Thu Aug 1 20:16:53 2019 +0000\n\n soc/fsl/qbman: Rework QBMan private memory setup\n \n Rework QBMan private memory setup so that the areas are not\n zeroed if the device was previously initialized\n \n If the QMan private memory was already initialized skip the PFDR\n initialization.\n \n Signed-off-by: Roy Pledge \u003croy.pledge@nxp.com\u003e\n Signed-off-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n\ncommit 031bca078e7554f1aa571a000272a52a07b1735f\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 14:36:06 2019 -0700\n\n soc: fsl: qbman: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n Cc: Li Yang \u003cleoyang.li@nxp.com\u003e\n Cc: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n\ncommit bfc4c359b2822bd2c457ccab271baeb33181c7c9\nAuthor: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\nDate: Mon Aug 12 15:27:37 2019 -0700\n\n drm/i915/cml: Add Missing PCI IDs\n \n The BSpec has added three new IDS for CML.\n Update the IDs in accordance to the Spec.\n \n Cc: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Reviewed-by: Anshuman Gupta \u003canshuman.gupta@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812222737.29356-1-anusha.srivatsa@intel.com\n\ncommit 0dabbe1bb3a4cf47d0cf52828355293f18acdae9\nAuthor: Sudarsana Reddy Kalluru \u003cskalluru@marvell.com\u003e\nDate: Wed Aug 14 01:11:53 2019 -0700\n\n qed: Add driver API for flashing the config attributes.\n \n The patch adds driver interface for reading the config attributes from user\n provided buffer, and updates these values on nvm config flash partition.\n \n This is basically an expansion of our existing ethtool -f implementation.\n The management FW has exposed an additional method of configuring some of\n the nvram options, and this makes use of that. This implementation will\n come into use when newer FW files which contain configuration directives\n employing this API will be provided to ethtool -f.\n \n Signed-off-by: Sudarsana Reddy Kalluru \u003cskalluru@marvell.com\u003e\n Signed-off-by: Ariel Elior \u003caelior@marvell.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 38eabdf0248e4b4f280dbda5e4e089aa5aaf806d\nAuthor: Sudarsana Reddy Kalluru \u003cskalluru@marvell.com\u003e\nDate: Wed Aug 14 01:11:52 2019 -0700\n\n qed: Add API for configuring NVM attributes.\n \n The patch adds API for configuring the NVM config attributes using\n Management FW (MFW) interfaces.\n \n Signed-off-by: Sudarsana Reddy Kalluru \u003cskalluru@marvell.com\u003e\n Signed-off-by: Ariel Elior \u003caelior@marvell.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 246bb4aaa4f4314816ae6835f3c6cc6dc7a10f4f\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Thu Aug 15 21:42:51 2019 +0200\n\n Revert \"ALSA: hda: Add codec on bus address table lately\"\n \n This reverts commit ee5f85d9290f (\"ALSA: hda: Add codec on bus address\n table lately\"). The commit caused several regression since I\u0027ve\n overlooked that the function doesn\u0027t manage only the caddr_tbl but\n also the codec linked list that is referred indirectly in the other\n drivers.\n \n Revert for now to make everything back to work.\n \n Fixes: ee5f85d9290f (\"ALSA: hda: Add codec on bus address table lately\")\n Reported-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 50e5bf56ba2fbbce3da74fb93aebb8e13f8eb8fc\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Thu Aug 15 21:38:01 2019 +0200\n\n arm-soc: document merges\n\ncommit 75bf1bd7d2f993787d0faa7d6e6534f7e88a2975\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 16:40:39 2019 +0200\n\n ARM: lpc32xx: allow multiplatform build\n \n All preparation work is done, so the platform can finally\n be moved into ARCH_MULTIPLATFORM. This requires a small\n change to the defconfig file to enable the platform.\n \n Link: https://lore.kernel.org/r/20190809144043.476786-14-arnd@arndb.de\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit d3532910038bb1e95e9c5952e98dd1d18b636e8b\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 16:40:38 2019 +0200\n\n ARM: lpc32xx: clean up header files\n \n All device drivers have stopped relying on mach/*.h headers,\n so move the remaining headers into arch/arm/mach-lpc32xx/lpc32xx.h\n to prepare for multiplatform builds.\n \n The mach/entry-macro.S file has been unused for a long time now\n and can simply get removed.\n \n Link: https://lore.kernel.org/r/20190809144043.476786-13-arnd@arndb.de\n Acked-by: Sylvain Lemieux \u003cslemieux.tyco@gmail.com\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit ffba29c9ebd0977dbf77bf6064776716a51b8ae5\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 16:40:37 2019 +0200\n\n serial: lpc32xx: allow compile testing\n \n The lpc32xx_loopback_set() function in hte lpc32xx_hs driver is the\n one thing that relies on platform header files. Move that into the\n core platform code so we only need a variable declaration for it,\n and enable COMPILE_TEST building.\n \n Link: https://lore.kernel.org/r/20190809144043.476786-12-arnd@arndb.de\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Acked-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 35974a7cc23c5deb5597c0a42183172498c4a0a8\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 16:40:36 2019 +0200\n\n net: lpc-enet: allow compile testing\n \n The lpc-enet driver can now be built on all platforms, so\n allow compile testing as well.\n \n Add one missing header inclusion that is required in some\n configurations.\n \n Link: https://lore.kernel.org/r/20190809144043.476786-11-arnd@arndb.de\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit de6f97b2bace0e2eb6c3a86e124d1e652a587b56\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 16:40:35 2019 +0200\n\n net: lpc-enet: fix printk format strings\n \n compile-testing this driver on other architectures showed\n multiple warnings:\n \n drivers/net/ethernet/nxp/lpc_eth.c: In function \u0027lpc_eth_drv_probe\u0027:\n drivers/net/ethernet/nxp/lpc_eth.c:1337:19: warning: format \u0027%d\u0027 expects argument of type \u0027int\u0027, but argument 4 has type \u0027resource_size_t {aka long long unsigned int}\u0027 [-Wformat\u003d]\n \n drivers/net/ethernet/nxp/lpc_eth.c:1342:19: warning: format \u0027%x\u0027 expects argument of type \u0027unsigned int\u0027, but argument 4 has type \u0027dma_addr_t {aka long long unsigned int}\u0027 [-Wformat\u003d]\n \n Use format strings that work on all architectures.\n \n Link: https://lore.kernel.org/r/20190809144043.476786-10-arnd@arndb.de\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 772775c1dfe00c99a9fe449ed35f74f1067f0128\nAuthor: kbuild test robot \u003clkp@intel.com\u003e\nDate: Fri Aug 9 16:40:34 2019 +0200\n\n net: lpc-enet: fix badzero.cocci warnings\n \n drivers/net/ethernet/nxp/lpc_eth.c:1316:31-32: WARNING comparing pointer to 0\n \n Compare pointer-typed values to NULL rather than 0\n \n Semantic patch information:\n This makes an effort to choose between !x and x \u003d\u003d NULL. !x is used\n if it has previously been used with the function used to initialize x.\n This relies on type information. More type information can be obtained\n using the option -all_includes and the option -I to specify an\n include path.\n \n Generated by: scripts/coccinelle/null/badzero.cocci\n \n Link: https://lore.kernel.org/r/20190809144043.476786-9-arnd@arndb.de\n Fixes: e42016eb3844 (\"net: lpc-enet: allow compile testing\")\n Signed-off-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit ecca1a6277aac10e40e4baba28adb893899b24b3\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 16:40:33 2019 +0200\n\n net: lpc-enet: move phy setup into platform code\n \n Setting the phy mode requires touching a platform specific\n register, which prevents us from building the driver without\n its header files.\n \n Move it into a separate function in arch/arm/mach/lpc32xx\n to hide the core registers from the network driver.\n \n Link: https://lore.kernel.org/r/20190809144043.476786-8-arnd@arndb.de\n Acked-by: Sylvain Lemieux \u003cslemieux.tyco@gmail.com\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 9dc03ffd996d4103cc2a11286d61e517bce27440\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 16:40:32 2019 +0200\n\n net: lpc-enet: factor out iram access\n \n The lpc_eth driver uses a platform specific method to find\n the internal sram. This prevents building it on other machines.\n \n Rework to only use one function call and keep the other platform\n internals where they belong. Ideally this would look up the\n sram location from DT, but as this is a rarely used driver,\n I want to keep the modifications to a minimum.\n \n Link: https://lore.kernel.org/r/20190809144043.476786-7-arnd@arndb.de\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 591f935db44867b38f3c91ec8e90104fb4e5dfb5\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Aug 14 11:29:23 2019 +0200\n\n serial: mxs-auart: Don\u0027t check for mctrl_gpio_to_gpiod() returning error\n \n Since commit 1d267ea6539f2663 (\"serial: mctrl-gpio: simplify init\n routine\"), mctrl_gpio_init() returns failure if the assignment to any\n member of the gpio array results in an error pointer.\n Since commit c359522194593815 (\"serial: mctrl_gpio: Avoid probe failures\n in case of missing gpiolib\"), mctrl_gpio_to_gpiod() returns NULL in the\n !CONFIG_GPIOLIB case.\n Hence there is no longer a need to check for mctrl_gpio_to_gpiod()\n returning an error value. A simple NULL check is sufficient.\n \n This follows the spirit of commit 445df7ff3fd1a0a9 (\"serial: mctrl-gpio:\n drop usages of IS_ERR_OR_NULL\") in the mctrl-gpio core.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Link: https://lore.kernel.org/r/20190814092924.13857-3-geert+renesas@glider.be\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d4d48e15040633c9382bb29b9b8a9eb6e262bf4b\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Aug 14 11:29:24 2019 +0200\n\n serial: sh-sci: Don\u0027t check for mctrl_gpio_to_gpiod() returning error\n \n Since commit 1d267ea6539f2663 (\"serial: mctrl-gpio: simplify init\n routine\"), mctrl_gpio_init() returns failure if the assignment to any\n member of the gpio array results in an error pointer.\n Since commit c359522194593815 (\"serial: mctrl_gpio: Avoid probe failures\n in case of missing gpiolib\"), mctrl_gpio_to_gpiod() returns NULL in the\n !CONFIG_GPIOLIB case.\n Hence there is no longer a need to check for mctrl_gpio_to_gpiod()\n returning an error value. A simple NULL check is sufficient.\n \n This follows the spirit of commit 445df7ff3fd1a0a9 (\"serial: mctrl-gpio:\n drop usages of IS_ERR_OR_NULL\") in the mctrl-gpio core.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Link: https://lore.kernel.org/r/20190814092924.13857-4-geert+renesas@glider.be\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7ccb0118e4295b3fb4b19f23012bdcdab2009399\nAuthor: Ahung Cheng \u003cahcheng@nvidia.com\u003e\nDate: Mon Aug 12 16:58:13 2019 +0530\n\n serial: tegra: protect IER against LCR.DLAB\n \n The IER and DLH registers occupy the same address space, selected by\n the LCR.DLAB bit. Hence, add port lock to protect IER when LCR.DLAB bit\n is set.\n \n Signed-off-by: Ahung Cheng \u003cahcheng@nvidia.com\u003e\n Signed-off-by: Krishna Yarlagadda \u003ckyarlagadda@nvidia.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Link: https://lore.kernel.org/r/1565609303-27000-5-git-send-email-kyarlagadda@nvidia.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f964a1e94fed3efdd207aa523b81c5c3b84ee936\nAuthor: Andreas Abel \u003caabel@nvidia.com\u003e\nDate: Mon Aug 12 16:58:10 2019 +0530\n\n serial: tegra: add internal loopback functionality\n \n Add the internal loopback functionality that can be enabled with\n TIOCM_LOOP.\n \n Signed-off-by: Andreas Abel \u003caabel@nvidia.com\u003e\n Signed-off-by: Krishna Yarlagadda \u003ckyarlagadda@nvidia.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Link: https://lore.kernel.org/r/1565609303-27000-2-git-send-email-kyarlagadda@nvidia.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 1bc102260d278de0af89c58536f4bbabd2ef28be\nAuthor: Razvan Stefanescu \u003crazvan.stefanescu@microchip.com\u003e\nDate: Tue Aug 13 10:40:25 2019 +0300\n\n tty/serial: atmel: reschedule TX after RX was started\n \n When half-duplex RS485 communication is used, after RX is started, TX\n tasklet still needs to be scheduled tasklet. This avoids console freezing\n when more data is to be transmitted, if the serial communication is not\n closed.\n \n Fixes: 69646d7a3689 (\"tty/serial: atmel: RS485 HD w/DMA: enable RX after TX is stopped\")\n Signed-off-by: Razvan Stefanescu \u003crazvan.stefanescu@microchip.com\u003e\n Cc: stable \u003cstable@vger.kernel.org\u003e\n Link: https://lore.kernel.org/r/20190813074025.16218-1-razvan.stefanescu@microchip.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b953815b819b0f327b5538feba3639d52db70172\nAuthor: Stefan-gabriel Mirea \u003cstefan-gabriel.mirea@nxp.com\u003e\nDate: Fri Aug 9 11:29:16 2019 +0000\n\n tty: serial: Add linflexuart driver for S32V234\n \n Introduce support for LINFlex driver, based on:\n - the version of Freescale LPUART driver after commit b3e3bf2ef2c7 (\"Merge\n 4.0-rc7 into tty-next\");\n - commit abf1e0a98083 (\"tty: serial: fsl_lpuart: lock port on console\n write\").\n In this basic version, the driver can be tested using initramfs and relies\n on the clocks and pin muxing set up by U-Boot.\n \n Remarks concerning the earlycon support:\n \n - LinFlexD does not allow character transmissions in the INIT mode (see\n section 47.4.2.1 in the reference manual[1]). Therefore, a mutual\n exclusion between the first linflex_setup_watermark/linflex_set_termios\n executions and linflex_earlycon_putchar was employed and the characters\n normally sent to earlycon during initialization are kept in a buffer and\n sent afterwards.\n \n - Empirically, character transmission is also forbidden within the last 1-2\n ms before entering the INIT mode, so we use an explicit timeout\n (PREINIT_DELAY) between linflex_earlycon_putchar and the first call to\n linflex_setup_watermark.\n \n - U-Boot currently uses the UART FIFO mode, while this driver makes the\n transition to the buffer mode. Therefore, the earlycon putchar function\n matches the U-Boot behavior before initializations and the Linux behavior\n after.\n \n [1] https://www.nxp.com/webapp/Download?colCode\u003dS32V234RM\n \n Signed-off-by: Stoica Cosmin-Stefan \u003ccosmin.stoica@nxp.com\u003e\n Signed-off-by: Adrian.Nitu \u003cadrian.nitu@freescale.com\u003e\n Signed-off-by: Larisa Grigore \u003cLarisa.Grigore@nxp.com\u003e\n Signed-off-by: Ana Nedelcu \u003cB56683@freescale.com\u003e\n Signed-off-by: Mihaela Martinas \u003cMihaela.Martinas@freescale.com\u003e\n Signed-off-by: Matthew Nunez \u003cmatthew.nunez@nxp.com\u003e\n [stefan-gabriel.mirea@nxp.com: Reduced for upstreaming and implemented\n earlycon support]\n Signed-off-by: Stefan-Gabriel Mirea \u003cstefan-gabriel.mirea@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190809112853.15846-6-stefan-gabriel.mirea@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d88ce24a0f3b30aba39e2a8966381fdfa9191106\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 16:40:31 2019 +0200\n\n gpio: lpc32xx: allow building on non-lpc32xx targets\n \n The driver uses hardwire MMIO addresses instead of the data\n that is passed in device tree. Change it over to only\n hardcode the register offset values and allow compile-testing.\n \n Link: https://lore.kernel.org/r/20190809144043.476786-6-arnd@arndb.de\n Acked-by: Sylvain Lemieux \u003cslemieux.tyco@gmail.com\u003e\n Tested-by: Sylvain Lemieux \u003cslemieux.tyco@gmail.com\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 5711e41b682ed4fc9b0803dc27722b1d8723a78a\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 16:40:30 2019 +0200\n\n serial: lpc32xx_hs: allow compile-testing\n \n The only thing that prevents building this driver on other\n platforms is the mach/hardware.h include, which is not actually\n used here at all, so remove the line and allow CONFIG_COMPILE_TEST.\n \n Link: https://lore.kernel.org/r/20190809144043.476786-5-arnd@arndb.de\n Acked-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Acked-by: Sylvain Lemieux \u003cslemieux.tyco@gmail.com\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 0a453d527ba4ce393d8d8bc0b3f21d5df3d640bf\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 16:40:29 2019 +0200\n\n watchdog: pnx4008_wdt: allow compile-testing\n \n The only thing that prevents building this driver on other\n platforms is the mach/hardware.h include, which is not actually\n used here at all, so remove the line and allow CONFIG_COMPILE_TEST.\n \n Link: https://lore.kernel.org/r/20190809144043.476786-4-arnd@arndb.de\n Acked-by: Sylvain Lemieux \u003cslemieux.tyco@gmail.com\u003e\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 50ad15282e7c75269ab754f9460dd36ae784e181\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 16:40:28 2019 +0200\n\n usb: udc: lpc32xx: allow compile-testing\n \n The only thing that prevents building this driver on other\n platforms is the mach/hardware.h include, which is not actually\n used here at all, so remove the line and allow CONFIG_COMPILE_TEST.\n \n Link: https://lore.kernel.org/r/20190809144043.476786-3-arnd@arndb.de\n Acked-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Acked-by: Sylvain Lemieux \u003cslemieux.tyco@gmail.com\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 23565baecee4fb5ac6f1d43e9b0685cca9438204\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 16:40:27 2019 +0200\n\n usb: ohci-nxp: enable compile-testing\n \n The driver hardcodes a hardware I/O address the way one should\n generally not do, and this prevents both compile-testing, and\n moving the platform to CONFIG_ARCH_MULTIPLATFORM.\n \n Change the code to be independent of the machine headers\n to allow those two. Removing the hardcoded address would\n be hard and is not necessary, so leave that in place for now.\n \n Link: https://lore.kernel.org/r/20190809144043.476786-2-arnd@arndb.de\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 6444f47eb8678a43d5260c67b89c18b1ea09e79e\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Fri Aug 2 12:08:13 2019 -0700\n\n writeback, cgroup: inode_switch_wbs() shouldn\u0027t give up on wb_switch_rwsem trylock fail\n \n As inode wb switching may make sync(2) miss some inodes, they\u0027re\n synchronized using wb_switch_rwsem so that no wb switching happens\n while sync(2) is in progress. In addition to synchronizing the actual\n switching, the rwsem is also used to prevent queueing new switch\n attempts while sync(2) is in progress. This is to avoid queueing too\n many instances while the rwsem is held by sync(2). Unfortunately,\n this is too agressive and can block wb switching for a long time if\n sync(2) is frequent.\n \n The goal is avoiding expolding the number of scheduled switches, not\n avoiding scheduling anything. Let\u0027s use wb_switch_rwsem only for\n synchronizing the actual switching and sync(2) and use\n isw_nr_in_flight instead for limiting the maximum number of scheduled\n switches. The limit is set to 1024 which should be more than enough\n while still avoiding extreme situations.\n \n Reviewed-by: Jan Kara \u003cjack@suse.cz\u003e\n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 55a694dffb7fd126b1e047aa46c437731d2700bb\nAuthor: Tejun Heo \u003ctj@kernel.org\u003e\nDate: Thu Aug 15 12:25:28 2019 -0700\n\n writeback, cgroup: Adjust WB_FRN_TIME_CUT_DIV to accelerate foreign inode switching\n \n WB_FRN_TIME_CUT_DIV is used to tell the foreign inode detection logic\n to ignore short writeback rounds to prevent getting confused by a\n burst of short writebacks. The parameter is currently 2 meaning that\n anything smaller than half of the running average writback duration\n will be ignored.\n \n This is unnecessarily aggressive. The detection logic uses 16 history\n slots and is already reasonably protected against some short bursts\n confusing it and the current parameter can lead to tens of seconds of\n missed detection depending on the writeback pattern.\n \n Let\u0027s change the parameter to 8, so that it only ignores writeback\n with are smaller than 12.5% of the current running average.\n \n v2: Add comment explaining what\u0027s going on with the foreign detection\n parameters.\n \n Reviewed-by: Jan Kara \u003cjack@suse.cz\u003e\n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit b98052e9b5c4878f0e13b11cd5b7a04242c2d848\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Aug 9 18:33:14 2019 +0100\n\n USB: storage: isd200: remove redundant assignment to variable sendToTransport\n \n The variable sendToTransport is being initialized with a value that is\n never read and is being re-assigned a little later on. The assignment\n is redundant and hence can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Acked-by: Alan Stern \u003cstern@rowland.harvard.edu\u003e\n Link: https://lore.kernel.org/r/20190809173314.4623-1-colin.king@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 8f6244055bd33695ce28f6b9b454fc1ae05bcd9c\nAuthor: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\nDate: Wed Aug 14 16:24:19 2019 +0300\n\n usb: typec: fusb302: Always provide fwnode for the port\n \n By registering a software fwnode for the port when the\n firmware does not supply one, we can always provide tcpm the\n connector capabilities by using the common USB connector\n device properties instead of using tcpc_config platform data.\n \n Signed-off-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\n Reviewed-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Link: https://lore.kernel.org/r/20190814132419.39759-4-heikki.krogerus@linux.intel.com\n Tested-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit c7316315a22164ee09059b8e148b52920a2521de\nAuthor: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\nDate: Wed Aug 14 16:24:18 2019 +0300\n\n dt-bindings: usb: fusb302: Remove deprecated properties\n \n There are no platforms using them anymore.\n \n Signed-off-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\n Reviewed-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Link: https://lore.kernel.org/r/20190814132419.39759-3-heikki.krogerus@linux.intel.com\n Tested-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit eceddc4071e39807f080913656252e4526227647\nAuthor: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\nDate: Wed Aug 14 16:24:17 2019 +0300\n\n usb: typec: fusb302: Remove unused properties\n \n Removing the deprecated fusb302 specific properties. There\n are no more platforms using them.\n \n Signed-off-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\n Reviewed-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Link: https://lore.kernel.org/r/20190814132419.39759-2-heikki.krogerus@linux.intel.com\n Tested-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 8d73f8f23e6b869b726cb01dd4747f56dc88660a\nAuthor: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\nDate: Tue Aug 13 10:45:09 2019 -0700\n\n page_pool: fix logic in __page_pool_get_cached\n \n __page_pool_get_cached() will return NULL when the ring is\n empty, even if there are pages present in the lookaside cache.\n \n It is also possible to refill the cache, and then return a\n NULL page.\n \n Restructure the logic so eliminate both cases.\n \n Signed-off-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Acked-by: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\n Acked-by: Ilias Apalodimas \u003cilias.apalodimas@linaro.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 11740ef4482914fcd8c9814ef7ceb7085715e554\nAuthor: Andy Grover \u003candy.grover@oracle.com\u003e\nDate: Thu Jan 13 11:40:31 2011 -0800\n\n rds: check for excessive looping in rds_send_xmit\n \n Original commit from 2011 updated to include a change by\n Yuval Shaia \u003cyuval.shaia@oracle.com\u003e\n that adds a new statistic counter \"send_stuck_rm\"\n to capture the messages looping exessively\n in the send path.\n \n Signed-off-by: Gerd Rausch \u003cgerd.rausch@oracle.com\u003e\n Acked-by: Santosh Shilimkar \u003csantosh.shilimkar@oracle.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 55c70ca00c982fbc0df4c4d3e31747fb73f4ddb5\nAuthor: Gerd Rausch \u003cgerd.rausch@oracle.com\u003e\nDate: Thu Jul 11 12:15:50 2019 -0700\n\n net/rds: Add a few missing rds_stat_names entries\n \n In a previous commit, fields were added to \"struct rds_statistics\"\n but array \"rds_stat_names\" was not updated accordingly.\n \n Please note the inconsistent naming of the string representations\n that is done in the name of compatibility\n with the Oracle internal code-base.\n \n s_recv_bytes_added_to_socket -\u003e \"recv_bytes_added_to_sock\"\n s_recv_bytes_removed_from_socket -\u003e \"recv_bytes_freed_fromsock\"\n \n Fixes: 192a798f5299 (\"RDS: add stat for socket recv memory usage\")\n Signed-off-by: Gerd Rausch \u003cgerd.rausch@oracle.com\u003e\n Acked-by: Santosh Shilimkar \u003csantosh.shilimkar@oracle.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 356feaad4f7d153ae429f2c7f7900b589dca6801\nAuthor: Chris Mason \u003cchris.mason@oracle.com\u003e\nDate: Fri Feb 3 11:08:51 2012 -0500\n\n RDS: don\u0027t use GFP_ATOMIC for sk_alloc in rds_create\n \n Signed-off-by: Chris Mason \u003cchris.mason@oracle.com\u003e\n Signed-off-by: Bang Nguyen \u003cbang.nguyen@oracle.com\u003e\n Signed-off-by: Gerd Rausch \u003cgerd.rausch@oracle.com\u003e\n Signed-off-by: Somasundaram Krishnasamy \u003csomasundaram.krishnasamy@oracle.com\u003e\n Acked-by: Santosh Shilimkar \u003csantosh.shilimkar@oracle.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 65dedd7fe1f224e9989181d161e23f0a544d304f\nAuthor: Chris Mason \u003cchris.mason@oracle.com\u003e\nDate: Fri Feb 3 11:07:54 2012 -0500\n\n RDS: limit the number of times we loop in rds_send_xmit\n \n This will kick the RDS worker thread if we have been looping\n too long.\n \n Original commit from 2012 updated to include a change by\n Venkat Venkatsubra \u003cvenkat.x.venkatsubra@oracle.com\u003e\n that triggers \"must_wake\" if \"rds_ib_recv_refill_one\" fails.\n \n Signed-off-by: Gerd Rausch \u003cgerd.rausch@oracle.com\u003e\n Acked-by: Santosh Shilimkar \u003csantosh.shilimkar@oracle.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 2f53ae347f597842683c4bde5b9ce76f5efae247\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 15 16:03:26 2019 -0300\n\n perf top: Add --switch-on/--switch-off events\n \n Just like \u0027perf trace\u0027 and \u0027perf script\u0027, should be useful for instance\n to only consider samples after the initialization phase of some\n workload.\n \n The man page has some examples and considerations about its current\n interface, that still doesn\u0027t handle the on/off events in a special way,\n behaving just like when multiple events are specified, i.e.:\n \n - In non-group mode (when the event list is not enclosed in {}) show a\n a menu to allow choosing which event the user wants to see in the\n histograms browser\n \n - In group mode, be it using {} or asking for --group, show one column\n per event.\n \n Try for instance:\n \n # perf top -e \u0027{cycles,instructions,probe:icmp_rcv}\u0027 --switch-on\u003dprobe:icmp_rcv\n \n Replace probe:icmp_rcv, that I put in place using:\n \n # perf probe icmp_rcv:59\n \n To hit when broadcast packets arrive, with a probe installed after an\n initialization phase is over or after some other point of interest, some\n garbage collection, etc, and also use --switch-off, for instance, on a\n probe installed after said garbage collection is over.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Florian Weimer \u003cfweimer@redhat.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: William Cohen \u003cwcohen@redhat.com\u003e\n Link: https://lkml.kernel.org/n/tip-c7q7qjeqtyvc9mkeipxza6ne@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 5156d7ef6d5e29f98f7302a5ea469fd9fa2af4ad\nAuthor: Jiri Pirko \u003cjiri@mellanox.com\u003e\nDate: Thu Aug 15 15:46:34 2019 +0200\n\n selftests: netdevsim: add devlink regions tests\n \n Test netdevsim devlink region implementation.\n \n Signed-off-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Tested-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Acked-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4418f862d675011a242a4e687d74ea107e6a40ed\nAuthor: Jiri Pirko \u003cjiri@mellanox.com\u003e\nDate: Thu Aug 15 15:46:33 2019 +0200\n\n netdevsim: implement support for devlink region and snapshots\n \n Implement dummy region of size 32K and allow user to create snapshots\n or random data using debugfs file trigger.\n \n Signed-off-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit dc8a670a87e6240d20c27f015b8c30f4ef811041\nAuthor: Jiri Pirko \u003cjiri@mellanox.com\u003e\nDate: Thu Aug 15 15:42:29 2019 +0200\n\n selftests: netdevsim: add devlink params tests\n \n Test recently added netdevsim devlink param implementation.\n \n Signed-off-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Tested-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Acked-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d87e5edb4cd7c954b71cfcb9fbfce65eac76e141\nAuthor: Jiri Pirko \u003cjiri@mellanox.com\u003e\nDate: Thu Aug 15 15:42:28 2019 +0200\n\n selftests: net: push jq workaround into separate helper\n \n Push the jq return value workaround code into a separate helper so it\n could be used by the rest of the code.\n \n Signed-off-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 873343e7d4964fa70cac22e2f7653f510bfaaa11\nAuthor: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\nDate: Thu Aug 15 20:41:00 2019 +0800\n\n page_pool: remove unnecessary variable init\n \n Remove variable initializations in functions that\n are followed by assignments before use\n \n Signed-off-by: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\n Acked-by: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b6cef26fb90964a298226caa524697413b8a1579\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Thu Aug 15 14:21:30 2019 +0200\n\n r8169: sync EEE handling for RTL8168h with vendor driver\n \n Sync EEE init for RTL8168h with vendor driver and add two writes to\n vendor-specific registers.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 2e779ddb5617928ee09842758c4734682264279d\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Thu Aug 15 14:14:18 2019 +0200\n\n r8169: use the generic EEE management functions\n \n Now that the Realtek PHY driver maps the vendor-specific EEE registers\n to the standard MMD registers, we can remove all special handling and\n use the generic functions phy_ethtool_get/set_eee.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Reviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 5b3f13950c91721ccabd001015ef09999c2306bb\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Thu Aug 15 14:12:55 2019 +0200\n\n net: phy: realtek: add support for EEE registers on integrated PHY\u0027s\n \n EEE-related registers on newer integrated PHY\u0027s have the standard\n layout, but are accessible not via MMD but via vendor-specific\n registers. Emulating the standard MMD registers allows to use the\n generic functions for EEE control.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Reviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 2441ba4806a184f82609aae07921d19c9377d040\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Thu Aug 15 13:19:22 2019 +0200\n\n net: phy: swphy: emulate register MII_ESTATUS\n \n When the genphy driver binds to a swphy it will call\n genphy_read_abilites that will try to read MII_ESTATUS if BMSR_ESTATEN\n is set in MII_BMSR. So far this would read the default value 0xffff\n and 1000FD and 1000HD are reported as supported just by chance.\n Better add explicit support for emulating MII_ESTATUS.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Reviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 916e571ebad37b820a2a4465dc2ea7643f2007fc\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Thu Aug 15 13:15:19 2019 +0200\n\n net: phy: read MII_CTRL1000 in genphy_read_status only if needed\n \n Value of MII_CTRL1000 is needed only if LPA_1000MSFAIL is set.\n Therefore move reading this register.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Reviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a46b5b6c27018102a4fe690d6516f5d5556646b7\nAuthor: Ka-Cheong Poon \u003cka-cheong.poon@oracle.com\u003e\nDate: Thu Aug 15 02:36:43 2019 -0700\n\n net/rds: Add RDS6_INFO_SOCKETS and RDS6_INFO_RECV_MESSAGES options\n \n Add support of the socket options RDS6_INFO_SOCKETS and\n RDS6_INFO_RECV_MESSAGES which update the RDS_INFO_SOCKETS and\n RDS_INFO_RECV_MESSAGES options respectively. The old options work\n for IPv4 sockets only.\n \n Signed-off-by: Ka-Cheong Poon \u003cka-cheong.poon@oracle.com\u003e\n Acked-by: Santosh Shilimkar \u003csantosh.shilimkar@oracle.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1fa70c7f49132513fb0da4afa7643395eedc7d35\nAuthor: Pankaj Dubey \u003cpankaj.dubey@samsung.com\u003e\nDate: Tue Aug 13 17:08:22 2019 +0200\n\n ARM: exynos: Enable exynos-chipid driver\n \n As now we have Chipid driver to initialize and expose SoC related\n information let\u0027s include it in build by default.\n \n Signed-off-by: Pankaj Dubey \u003cpankaj.dubey@samsung.com\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Signed-off-by: Sylwester Nawrocki \u003cs.nawrocki@samsung.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit a1da1f531dc08dc20da99e3c9837d08df80e3090\nAuthor: Pankaj Dubey \u003cpankaj.dubey@samsung.com\u003e\nDate: Tue Aug 13 17:08:23 2019 +0200\n\n arm64: exynos: Enable exynos-chipid driver\n \n Enable Exynos Chipid driver for accessing SoC related information.\n \n Signed-off-by: Pankaj Dubey \u003cpankaj.dubey@samsung.com\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Signed-off-by: Sylwester Nawrocki \u003cs.nawrocki@samsung.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 40d8aff614f71ab3cab20785b4f213e3802d4e87\nAuthor: Sylwester Nawrocki \u003cs.nawrocki@samsung.com\u003e\nDate: Tue Aug 13 17:08:20 2019 +0200\n\n soc: samsung: chipid: Convert exynos-chipid driver to use the regmap API\n \n Convert the driver to use regmap API in order to allow other\n drivers, like ASV, to access the CHIPID registers.\n \n Add definition of selected CHIPID register offsets and register bit\n fields for Exynos5422 SoC.\n \n Signed-off-by: Sylwester Nawrocki \u003cs.nawrocki@samsung.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 3253b7b7cd44c4dd029a4ce280ef9f409a256e5f\nAuthor: Pankaj Dubey \u003cpankaj.dubey@samsung.com\u003e\nDate: Tue Aug 13 17:08:19 2019 +0200\n\n soc: samsung: Add exynos chipid driver support\n \n Exynos SoCs have Chipid, for identification of product IDs and SoC\n revisions. This patch intends to provide initialization code for all\n these functionalities, at the same time it provides some sysfs entries\n for accessing these information to user-space.\n \n This driver uses existing binding for exynos-chipid.\n \n Changes by Bartlomiej:\n - fixed return values on errors\n - removed bogus kfree_const()\n - added missing Exynos4210 EVT0 id\n - converted code to use EXYNOS_MASK define\n - fixed np use after of_node_put()\n - fixed too early use of dev_info()\n - made driver fail for unknown SoC-s\n - added SPDX tag\n - updated Copyrights\n \n Signed-off-by: Pankaj Dubey \u003cpankaj.dubey@samsung.com\u003e\n [m.szyprowski: for suggestion and code snippet of product_id_to_soc_id]\n Signed-off-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n [s.nawrocki: updated copyright date, removed uneeded headers inclusion]\n Signed-off-by: Sylwester Nawrocki \u003cs.nawrocki@samsung.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit b552addc455ae71df831e7c402cd8d7e8d4434ca\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Wed Aug 14 09:52:44 2019 +0200\n\n ARM: samsung: Include GPIO driver header\n \n This file is using struct gpio_chip and needs to include\n \u003clinux/gpio/driver.h\u003e to get that.\n \n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 60c833d5664e1b3f71c4471233469790adf505ca\nAuthor: Stefan Wahren \u003cwahrenst@gmx.net\u003e\nDate: Thu Jul 25 21:03:42 2019 +0200\n\n ARM: dts: bcm283x: Enable HDMI at board level\n \n There might be headless setups of the Compute Module without HDMI,\n so better enable HDMI at board level. Btw this allows moving HDMI\n base definition into upcoming bcm2835-common.dtsi.\n \n Signed-off-by: Stefan Wahren \u003cwahrenst@gmx.net\u003e\n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n\ncommit 6173269038e98408c27cfb438ff07d463fbd0b33\nAuthor: Pan Xiuli \u003cxiuli.pan@linux.intel.com\u003e\nDate: Thu Aug 15 10:57:49 2019 -0500\n\n ASoC: SOF: Intel: initial support for Elkhart Lake\n \n Add Kconfig, PCI ID and chip info for EHL platform.\n \n Note that the core mask is different from previous platforms, only\n Core0 can be controlled by the host. Additional patches will be\n required for multi-core functionality.\n \n Signed-off-by: Pan Xiuli \u003cxiuli.pan@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190815155749.29304-5-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit f4ead53241c7dcd2f3130e059cb3dfdec0ec9871\nAuthor: Pan Xiuli \u003cxiuli.pan@linux.intel.com\u003e\nDate: Thu Aug 15 10:57:48 2019 -0500\n\n ASoC: Intel: common: add ACPI matching tables for EHL\n \n There are no upstream machine drivers just yet so just add dummy table\n for compilation in nocodec-mode.\n \n Signed-off-by: Pan Xiuli \u003cxiuli.pan@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190815155749.29304-4-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 1205c81edd5ef220ca32318572e499b4b180b1f3\nAuthor: Pan Xiuli \u003cxiuli.pan@linux.intel.com\u003e\nDate: Thu Aug 15 10:57:47 2019 -0500\n\n ASoC: SOF: Intel: initial support for Tiger Lake.\n \n Add Kconfig, PCI ID and chip info for Tiger Lake platform.\n \n Note that the core mask is different from previous platforms, only\n Core0 can be controlled by the host. Additional patches will be\n required for multi-core functionality.\n \n Signed-off-by: Pan Xiuli \u003cxiuli.pan@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190815155749.29304-3-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 5f7af9ecebce7fab78b0237b0cd998f2ed3de03c\nAuthor: Pan Xiuli \u003cxiuli.pan@linux.intel.com\u003e\nDate: Thu Aug 15 10:57:46 2019 -0500\n\n ASoC: Intel: common: add ACPI matching tables for Tiger Lake\n \n Initial support for TGL w/ RT1308\n \n Signed-off-by: Pan Xiuli \u003cxiuli.pan@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190815155749.29304-2-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 955741a5120bb1ed19d3b12da90bb947d3a0cb99\nAuthor: Raag Jadav \u003craagjadav@gmail.com\u003e\nDate: Wed Aug 14 00:02:55 2019 +0530\n\n regulator: act8865: operating mode and suspend state support\n \n Implement -\u003eset_mode(), -\u003eget_mode() and -\u003eset_suspend_xx() hooks\n for act8865 with unlocked expert registers.\n \n Based on work done by Borris Brezillon on [1].\n [1] https://www.spinics.net/lists/kernel/msg2942960.html\n \n Signed-off-by: Raag Jadav \u003craagjadav@gmail.com\u003e\n Link: https://lore.kernel.org/r/1565721176-8955-2-git-send-email-raagjadav@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 81eeb0a35c2e40bcaf122c6aae3be4f7d9abe201\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 15 11:07:04 2019 +0530\n\n regulator: core: Add label to collate of_node_put() statements\n \n In function of_get_child_regulator(), the loop for_each_child_of_node()\n contains two mid-loop return statements, each preceded by a statement\n putting child. In order to reduce this repetition, create a new label,\n err_node_put, that puts child and then returns the required value;\n edit the mid-loop return blocks to instead go to this new label.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190815053704.32156-1-nishkadg.linux@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit ff461ebfd4b7051d9b79ca023741e6c2d960a912\nAuthor: Raag Jadav \u003craagjadav@gmail.com\u003e\nDate: Wed Aug 14 00:02:56 2019 +0530\n\n regulator: act8865 regulator modes and suspend states\n \n Add documentation for act8865 regulator modes and suspend states.\n Add active-semi,8865-regulator.h file for device tree binding constants\n for act8865 regulators.\n \n Signed-off-by: Raag Jadav \u003craagjadav@gmail.com\u003e\n Link: https://lore.kernel.org/r/1565721176-8955-3-git-send-email-raagjadav@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 3a9477a06c6acb4234407b59999835a6f12f889d\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Thu Aug 15 10:50:30 2019 -0500\n\n ASoC: SOF: ipc: add ALH parameters\n \n The only configuration parameter is the ALH stream ID. No range\n checking is done by the driver, the firmware should check that the\n stream is valid for a specific hardware.\n \n Bump the ABI Minor number to keep the alignment with SOF firmware\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190815155032.29181-3-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c643c189f0fec116e75f9661e734545af0e3225e\nAuthor: Mac Chiang \u003cmac.chiang@intel.com\u003e\nDate: Thu Aug 15 14:51:49 2019 +0800\n\n ASoC: Intel: boards: Add Cometlake machine driver support\n \n reuse and add Cometlake support with:\n SSP0 for DA7219 headphone codec\n SSP1 for MAX98357a speaker amp codec\n \n Signed-off-by: Mac Chiang \u003cmac.chiang@intel.com\u003e\n Acked-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/1565851909-13825-1-git-send-email-mac.chiang@intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 672a749b4df68909e4a9dbb4ff7de3c795affd35\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Mon Aug 5 20:56:25 2019 +0800\n\n sgi-xpc: Use GFP_ATOMIC for kmalloc in atomic context.\n \n xpc_send_activate_IRQ_uv is called from\n \u003c-xpc_send_activate_IRQ_part_uv\n \u003c-xpc_indicate_partition_disengaged_uv\n (xpc_arch_ops.indicate_partition_disengaged)\n \u003c-xpc_die_deactivate\n \u003c-xpc_system_die\n \n xpc_system_die is registered by atomic_notifier_chain_register,\n which indicates xpc_system_die may be called in atomic context.\n So the kmalloc in xpc_send_activate_IRQ_uv may be in atomic context.\n \n Use GFP_ATOMIC instead of GFP_KERNEL in kmalloc.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190805125625.24963-1-huangfq.daxian@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit cd010d9b986f05b17adc02563eef4ebd38e6037f\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 8 13:34:22 2019 +0530\n\n sgi-xp: xpc_uv: Make structure xpc_arch_ops_uv constant\n \n The static xpc_arch_operations structure xpc_arch_ops_uv is only copied\n into the structure xpc_arch_ops, after which it is never modified; nor\n is it used in any other way. Hence it can be declared as a constant to\n prevent unintended modifications of its fields.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Acked-by: Robin Holt \u003crobinmholt@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190808080422.16503-1-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit c43b849f890af2b638b3bea79a4203875bcc54d7\nAuthor: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\nDate: Fri Jul 26 14:07:42 2019 -0400\n\n drm/amdgpu: Use new mode2 reset interface for RV.\n \n Integrate the mode2 reset into rest sequence.\n \n v2:\n Check ppfuncs pointer for NULL\n \n Signed-off-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e97204ead61c79c292761c5c934a02dfe9967b71\nAuthor: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\nDate: Wed Aug 14 16:26:35 2019 -0400\n\n drm/amd/poweplay: Add amd_pm_funcs callback for mode 2\n \n Add callback to call the new mode2 reset interface.\n \n Signed-off-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 10811ce4391b6edd496fbda103ff44eb0d6bc66f\nAuthor: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\nDate: Wed Aug 14 16:21:44 2019 -0400\n\n drm/amd/powerpay: Implement mode2 reset callback for SMU10\n \n Add implmenetion.\n \n Signed-off-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d6c23e6f1374180dc1197a1d25323ec68bd1e04c\nAuthor: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\nDate: Wed Aug 14 16:16:52 2019 -0400\n\n drm/amd/powerplay: add mode2 reset callback for pp_smu_mgr\n \n Also define reset modes (0, 1 and 2)\n \n Signed-off-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e84fb7bca63761c42201da7783f8a3a3abd8287b\nAuthor: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\nDate: Fri Jul 26 14:03:56 2019 -0400\n\n drm/amd/powerplay: Fix meaning of 0x1E PPSMC_MSG\n \n By comparing to windows it means ASIC reset.\n \n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b1f5b4538e7558deed89efdeccf7727f3f361720\nAuthor: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\nDate: Wed Aug 14 15:44:14 2019 -0400\n\n dmr/amdgpu: Fix compile error with CONFIG_DRM_AMDGPU_GART_DEBUGFS\n \n Double defintion of \u0027i\u0027\n \n Signed-off-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f659bb6dae58c113805f92822e4c16ddd3156b79\nAuthor: Ahzo \u003cAhzo@tutanota.com\u003e\nDate: Mon Aug 5 21:14:18 2019 +0200\n\n drm/amd/powerplay/smu7: enforce minimal VBITimeout (v2)\n \n This fixes screen corruption/flickering on 75 Hz displays.\n \n v2: make print statement debug only (Alex)\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d102646\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Ahzo \u003cAhzo@tutanota.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9bfe54e3b70e1d594a7fdc1e50043ab32e29eea7\nAuthor: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\nDate: Tue Aug 6 19:50:50 2019 -0700\n\n xilinx_sdfec: Convert to IDA\n \n This driver does not use the lookup abilities of the IDR, so convert it\n to the more space-efficient IDA.\n \n Signed-off-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n Link: https://lore.kernel.org/r/20190807025050.28367-1-willy@infradead.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 108b4d928c032b6fc3b9b10197154ee35d59961d\nAuthor: Gang Ba \u003cgaba@amd.com\u003e\nDate: Wed Aug 14 11:00:19 2019 -0400\n\n drm/amd/amdgpu: Update VM function pointer\n \n When VM state changed and system in large bar mode,\n make sure to use CPU update function, otherwise use\n SDMA function.\n \n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Gang Ba \u003cgaba@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f40c6912d2be1d9721d83b00abe312b2240407e6\nAuthor: Yong Zhao \u003cYong.Zhao@amd.com\u003e\nDate: Wed Aug 7 17:45:37 2019 -0400\n\n drm/amdkfd: Fill amdgpu_task_info for KFD VMs\n \n The amdgpu_task_info will be used when printing VM page fault for KFD\n processes.\n \n Signed-off-by: Yong Zhao \u003cYong.Zhao@amd.com\u003e\n Acked-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Harish Kasiviswanathan \u003charish.kasiviswanatha@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8b7d6157f2188ce94040ccb367490dbca929104a\nAuthor: Yong Zhao \u003cYong.Zhao@amd.com\u003e\nDate: Tue Aug 13 14:38:03 2019 -0400\n\n drm/amdgpu: Set VM_L2_CNTL.PDE_FAULT_CLASSIFICATION to 0 for GFX10\n \n We have done this for pre-GFX10 asics, but GFX10 did not pick up the\n new change. The below is the commit message for that change.\n \n This is recommended by HW designers. Previously when it was set to 1,\n the PDE walk error in VM fault will be treated as\n PERMISSION_OR_INVALID_PAGE_FAULT rather than usually expected OTHER_FAULT.\n As a result, the retry control in VM_CONTEXT*_CNTL will change accordingly.\n \n The above behavior is kind of abnormal. Furthermore, the\n PDE_FAULT_CLASSIFICATION \u003d\u003d 1 feature was targeted for very old ASICs\n and it never made it way to production. Therefore, we should set it to 0.\n \n Signed-off-by: Yong Zhao \u003cYong.Zhao@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5d36d4c97635f16ade47c86f4b3dd5843c79a7b2\nAuthor: Yong Zhao \u003cYong.Zhao@amd.com\u003e\nDate: Wed Aug 7 16:21:09 2019 -0400\n\n drm/amdgpu: Add more page fault info printing for GFX10\n \n The printing we did for GFX9 was not propogated to GFX10 somehow, so fix\n it now.\n \n Signed-off-by: Yong Zhao \u003cYong.Zhao@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4e0ae5e21436d8379404b356b8c807b42fbf24be\nAuthor: Yong Zhao \u003cYong.Zhao@amd.com\u003e\nDate: Tue Aug 13 14:04:26 2019 -0400\n\n drm/amdgpu: Add printing for RW extracted from VM_L2_PROTECTION_FAULT_STATUS\n \n RW is also useful in most cases.\n \n Signed-off-by: Yong Zhao \u003cYong.Zhao@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5413fce4b2147f8ebd0350ee5fdf82f9c30fd7dc\nAuthor: Oak Zeng \u003cOak.Zeng@amd.com\u003e\nDate: Thu Aug 1 14:24:42 2019 -0500\n\n drm/amdkfd/gfx10: Calling amdgpu functions to invalidate TLB\n \n Calling amdgpu function to invalidate TLB, instead of using a\n kfd implementation. Delete the kfd local TLB invalidation\n implementation.\n \n Signed-off-by: Oak Zeng \u003cOak.Zeng@amd.com\u003e\n Reviewed-by: Christian Konig \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 3ff985485b29693376bb470a40b7aba4394a189b\nAuthor: Oak Zeng \u003cOak.Zeng@amd.com\u003e\nDate: Thu Aug 1 14:55:45 2019 -0500\n\n drm/amdgpu: Export function to flush TLB of specific vm hub\n \n This is for kfd to reuse amdgpu TLB invalidation function.\n On gfx10, kfd only needs to flush TLB on gfx hub but not\n on mm hub. So export a function for KFD flush TLB only on\n specific hub.\n \n Signed-off-by: Oak Zeng \u003cOak.Zeng@amd.com\u003e\n Reviewed-by: Christian Konig \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 675a9e38b39c4114dbcf1ed2e7f7a0e5a6a5e4b4\nAuthor: Leo Li \u003csunpeng.li@amd.com\u003e\nDate: Tue Aug 6 13:50:28 2019 -0400\n\n drm/amd/display: Load NV12 SOC BB from firmware\n \n [Why]\n \n Previous SOC bounding box firmware loading logic was for NV10, when we\n still had it in firmware. Now that it\u0027s brought into driver code, and\n NV12 BB is in firmware, this logic needs to be repurposed for NV12.\n \n [How]\n \n Set SOC_BOUNDING_BOX_VALID to false, and add the dcn_2_0_nv12_soc BB\n struct. In init_soc_bounding_box, load firmware to nv12 BB instead.\n \n In addition, conditionally update and patch the BB, depending on ASIC\n REV.\n \n Signed-off-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Reviewed-by: Hersen Wu \u003chersenxs.wu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6d83a32d0bdd39cdcc0d70d7b719cd2ffb419268\nAuthor: Michael Strauss \u003cmichael.strauss@amd.com\u003e\nDate: Fri Jul 26 12:04:12 2019 -0400\n\n drm/amd/display: Enable MPO with pre-blend color processing (RGB)\n \n [Why]\n DCN10 performs color processing before MPC combination, causes color\n shift in RGB colorspaces when positive brightness offset is applied\n However, YCbCr is still unfixed and remains disabled\n \n [How]\n Add layerIndex to dc_plane_state and dc_plane_info structs\n Re-enable MPO when brightness is adjusted and colorspace is not YCbCr\n Set rear plane\u0027s brightness offset to 0 when front plane visible\n \n Signed-off-by: Michael Strauss \u003cmichael.strauss@amd.com\u003e\n Reviewed-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Acked-by: Nicholas Kazlauskas \u003cNicholas.Kazlauskas@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 30b7200c12fd2c951bcbc5bf0eefc20434917f28\nAuthor: yanyan kang \u003cYanyan.Kang@amd.com\u003e\nDate: Mon Jul 29 17:46:44 2019 +0800\n\n drm/amd/display: audio cannot switch to internal when display turns off\n \n [why]\n disable_az_endpoint has been skipped because\n dc-\u003edebug.az_endpoint_mute_only \u003d true.\n \n [how]\n set dc-\u003edebug.az_endpoint_mute_only false when PPLIB’s PME notification function\n pointer is not NULL at the dcn10_resource construct function,because right now\n SMU/PPLIB and DAL all have the AZ D3 force PME notification implemented. AZ D3 should work.\n \n Signed-off-by: yanyan kang \u003cYanyan.Kang@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e7f2c80cbaabf62966d02ba752d19c9a63f422ab\nAuthor: Wenjing Liu \u003cWenjing.Liu@amd.com\u003e\nDate: Fri Jul 26 14:53:20 2019 -0400\n\n drm/amd/display: check hpd before retry verify link cap\n \n [why]\n During detection link training if a display is disconnected,\n the current code will retry 3 times of link training\n on disconnected link before giving up.\n \n [how]\n Before each retry check for HPD status, only retry\n verify link cap when HPD is still high.\n Also put a 10ms delay between each retry to improve\n the chance of success.\n \n Signed-off-by: Wenjing Liu \u003cWenjing.Liu@amd.com\u003e\n Reviewed-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Acked-by: Abdoulaye Berthe \u003cAbdoulaye.Berthe@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 79e005204f75b54db9e42a39b3f66fac2ba190b9\nAuthor: Charlene Liu \u003ccharlene.liu@amd.com\u003e\nDate: Mon Jul 29 11:59:33 2019 -0400\n\n drm/amd/display: enable dcn_mem_pwr as golden setting updates\n \n Enable dcn_mem_pwr as golden setting updates\n \n Signed-off-by: Charlene Liu \u003ccharlene.liu@amd.com\u003e\n Reviewed-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8d966bdd48385d3ebdecfc25f195db034061e94f\nAuthor: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\nDate: Mon Jul 29 12:17:30 2019 -0400\n\n drm/amd/display: 3.2.46\n \n Signed-off-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 41f03a6d74c6f9c064b7edece13cd8716727f80a\nAuthor: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\nDate: Fri Jul 26 17:32:02 2019 -0400\n\n drm/amd/display: fix dcn20 global sync dml param extraction\n \n Currently the paremeters are extracted as if dml is calculating\n using pipes as we pass them in. in reality, dml internally merges\n pipes into planes if pipe split is detected.\n \n This change adds reverse logic to dcn20_calculate_dlg_params so\n that the global sync parameters can be correctly extracted for\n all the pipes when pipe split is enabled.\n \n Signed-off-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2c95e35fef21db2aec40c21f6d5267ae6e06a2e6\nAuthor: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\nDate: Sat Jul 27 15:10:54 2019 -0400\n\n drm/amd/display: 3.2.45\n \n Signed-off-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9b420a19bd077f81caadcb34738d7298550a7ed7\nAuthor: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\nDate: Thu Jul 25 18:46:54 2019 -0400\n\n drm/amd/display: Add and refine DSC logs in enable sequence\n \n [why]\n Some logs messages were not precise and some new log messages\n were needed after \"get packed PPS\" function was introduced\n \n Signed-off-by: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\n Reviewed-by: Wenjing Liu \u003cWenjing.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8221b6eacfc586b07aec101f9cad98e82867668d\nAuthor: Wyatt Wood \u003cwyatt.wood@amd.com\u003e\nDate: Wed Jul 24 13:37:08 2019 -0400\n\n drm/amd/display: Add Logging for Gamma Related information (2/2)\n \n [Why]\n A recent bug showed that logging would be useful in debugging\n various gamma issues.\n \n [How]\n Add logging in dc.\n \n Signed-off-by: Wyatt Wood \u003cwyatt.wood@amd.com\u003e\n Reviewed-by: Krunoslav Kovac \u003cKrunoslav.Kovac@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 40f08515ccf596f3f3f3adb4f8f632dfedae54c2\nAuthor: Julian Parkin \u003cjulian.parkin@amd.com\u003e\nDate: Thu Jul 25 16:53:27 2019 -0400\n\n drm/amd/display: Remove redundant definition of dwb_source enums\n \n There are repeated (but guarded) definitions of dwb_src enums. There are\n also unused entires. Clean them up.\n \n Signed-off-by: Julian Parkin \u003cjulian.parkin@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1ba0a5802fba04f2b0680ae12f478d9be93517d6\nAuthor: Qingqing Zhuo \u003cqingqing.zhuo@amd.com\u003e\nDate: Tue Jul 23 12:24:24 2019 -0400\n\n drm/amd/display: Add enum for H-timing divider mode\n \n Add h_timing_div_mode enum to better reflect possible register\n values. Replace previously programmed values with enum\n \n Signed-off-by: Qingqing Zhuo \u003cqingqing.zhuo@amd.com\u003e\n Reviewed-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e40837afb9b011757e17e9f71d97853ca574bcff\nAuthor: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\nDate: Thu Jul 25 11:53:16 2019 -0400\n\n drm/amd/display: Register VUPDATE_NO_LOCK interrupts for DCN2\n \n [Why]\n These are needed to send back DRM vblank events in the case where VRR\n is on. Without the interrupt enabled we\u0027re deferring the events into the\n vblank queue and userspace is left waiting forever to get back the\n events they need.\n \n Found using igt@kms_vrr - the test fails immediately due to vblank\n timeout.\n \n [How]\n Register them the same way we\u0027re handling it for DCN1.\n \n This fixes igt@kms_vrr for DCN2.\n \n Signed-off-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Reviewed-by: David Francis \u003cDavid.Francis@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 544618596fd58848f5f881e265a09320e91f2e97\nAuthor: Charlene Liu \u003ccharlene.liu@amd.com\u003e\nDate: Wed Jul 24 18:14:46 2019 -0400\n\n drm/amd/display: wake up ogam mem pwr before programming ocsc\n \n [Description]\n OGAM_MEM_PWR could stay in light up when driver woke up to update gamma.\n either disable MEM_LOW power feature or set to OGAM_bypass could make artificial color distortion goes away.\n Easy reproduce after LOW_MEM Power feature enables and resume from S3.\n \n Signed-off-by: Charlene Liu \u003ccharlene.liu@amd.com\u003e\n Reviewed-by: Julian Parkin \u003cjparkin@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1bb32e5a872daa8f918b73d4476862c5651d45bf\nAuthor: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\nDate: Thu Jul 25 17:27:03 2019 -0400\n\n drm/amd/display: Correct DSC PPS log\n \n [why]\n A misleading message \"Programming PPS\" appears before both programming\n and \"query PPS\" functions\n \n [how]\n Move the message from \"log PPS\" function to \"program PPS\" function\n \n Signed-off-by: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\n Reviewed-by: Wenjing Liu \u003cWenjing.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 90830e840e39df36e86dae17b41b501bc93dba49\nAuthor: Vitaly Prosyak \u003cvitaly.prosyak@amd.com\u003e\nDate: Fri Jul 19 16:07:33 2019 -0500\n\n drm/amd/display: Check if set_blank_data_double_buffer exists before call\n \n Not all ASIC types have this function implemented - check before\n calling.\n \n Signed-off-by: Vitaly Prosyak \u003cvitaly.prosyak@amd.com\u003e\n Reviewed-by: Julian Parkin \u003cjparkin@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 92a3701c438b069ac79b55e1ddfcc7907002ad39\nAuthor: Ilya Bakoulin \u003cIlya.Bakoulin@amd.com\u003e\nDate: Wed Jul 24 12:36:55 2019 -0400\n\n drm/amd/display: Fix type of ODMCombineType field\n \n The type was changed previously to better reflect possible register\n values.\n \n Signed-off-by: Ilya Bakoulin \u003cIlya.Bakoulin@amd.com\u003e\n Reviewed-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit bbdd620d3f2fd78ea09946c101ed9a5e6decb8cd\nAuthor: Wyatt Wood \u003cwyatt.wood@amd.com\u003e\nDate: Wed Jul 24 13:29:38 2019 -0400\n\n drm/amd/display: Add Logging for Gamma Related information (1/2)\n \n [Why]\n A recent bug showed that logging would be useful in\n debugging various gamma issues.\n \n [How]\n Add new log types and logging code to the color module.\n \n Signed-off-by: Wyatt Wood \u003cwyatt.wood@amd.com\u003e\n Reviewed-by: Aric Cyr \u003cAric.Cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit cf7180493cd3d47e4ee6f7e7a2d9da96ff53d73b\nAuthor: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\nDate: Tue Jul 23 16:38:02 2019 -0400\n\n drm/amd/display: Remove 4:2:2 DSC support\n \n [why]\n If DSC is available, a higher picture quality is achieved by using\n DSC with 4:4:4 format. Using 4:2:2 instead does not offer any benefit\n and would only introduce loss of quality. Removing it reduces\n maintenance and testing effort.\n \n Signed-off-by: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\n Reviewed-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e5382701c3520b3ed66169a6e4aa6ce5df8c56e0\nAuthor: Lewis Huang \u003cLewis.Huang@amd.com\u003e\nDate: Mon Jul 22 15:23:32 2019 -0400\n\n drm/amd/display: reprogram VM config when system resume\n \n [Why]\n The vm config will be clear to 0 when system enter S4. It will\n cause hubbub didn\u0027t know how to fetch data when system resume.\n The flip always pending because earliest_inuse_address and\n request_address are different.\n \n [How]\n Reprogram VM config when system resume\n \n Signed-off-by: Lewis Huang \u003cLewis.Huang@amd.com\u003e\n Reviewed-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Acked-by: Eric Yang \u003ceric.yang2@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b5b1f45549041c853568fe23a2da8bdc14d14190\nAuthor: Eric Yang \u003cEric.Yang2@amd.com\u003e\nDate: Wed Jul 10 22:41:51 2019 -0400\n\n drm/amd/display: Enable type C hotplug\n \n [Why and How]\n We want to change where timing is done for alt mode.\n Some of the commented out #ifs are needed for DCN20\n so we enable them for that case.\n \n Signed-off-by: Eric Yang \u003cEric.Yang2@amd.com\u003e\n Reviewed-by: Eric Yang \u003ceric.yang2@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2c58cc6dc0e15c62157430ee58f8efdae2aa7934\nAuthor: Ilya Bakoulin \u003cIlya.Bakoulin@amd.com\u003e\nDate: Mon Jul 22 14:12:25 2019 -0400\n\n drm/amd/display: HUBP/HUBBUB register programming fixes\n \n [Why]\n - Need to change interface function signature / add an enum\n to reflect the available register field values\n \n [How]\n - Add a new enum and modify existing functions to use it instead\n of bool\n \n Signed-off-by: Ilya Bakoulin \u003cIlya.Bakoulin@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2b162fd3024997b97c95290b0a8ed08d914f115d\nAuthor: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\nDate: Fri Jul 19 11:43:39 2019 -0400\n\n drm/amd/display: update optc odm interface for more than 2 opps\n \n Current optc odm interface only accepts 2 opps, we need to\n expand this to allow 4 to 1 odm combine.\n \n Signed-off-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1071a0ecf4a6a1fb7c98bf8351ebf414265cb780\nAuthor: Ilya Bakoulin \u003cIlya.Bakoulin@amd.com\u003e\nDate: Wed Jul 17 18:33:48 2019 -0400\n\n drm/amd/display: Update DML parameters\n \n [Why]\n Need to add DML struct members that were omitted in previous\n DML implemenations.\n \n [How]\n - Add missing enum values\n - Add missing struct members\n - Set new input values in the fetch functions\n \n Signed-off-by: Ilya Bakoulin \u003cIlya.Bakoulin@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5fc43055b03fabe7bc86ed890636f449e6353c6a\nAuthor: Julian Parkin \u003cjulian.parkin@amd.com\u003e\nDate: Mon Jul 15 12:16:01 2019 -0400\n\n drm/amd/display: Remove duplicate interface for programming FB\n \n [Why]\n There are currently two interfaces for exactly the same thing:\n hupb_update_dchub in hupb and update_dchub in hubbub. The hubbub\n version is currently unused past dcn10, largely because the call\n from the dcn10 hardware sequencer does not call through the\n interface, so the hupb interface was used instead. This is\n confusing because of the duplicate code, the unused functions,\n and the fact that more that one block currently owns this set\n of registers.\n \n [How]\n Remove the hubp interface entirely, as well as the register\n declarations that are not longer needed because of this. Change\n the call site to always call the hubbub version through the\n interface. Fix the update_dchub function in dcn20_hubbub.c to\n program the correct registers for dcn20.\n \n Signed-off-by: Julian Parkin \u003cjulian.parkin@amd.com\u003e\n Reviewed-by: Aric Cyr \u003cAric.Cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c681491a09213e6675c773a04f1e8801ffc201ff\nAuthor: Jun Lei \u003cJun.Lei@amd.com\u003e\nDate: Mon Jul 22 09:45:20 2019 -0400\n\n drm/amd/display: fix pipe selection logic in validate\n \n [why]\n Resource mapping done in dcn20_validate_bandwidth has a flaw: When a full\n update is performed, the HWSS will only update the MPCC tree for the stream\n that is updated as opposed to all streams. This means that when mapping pipes\n in validation, care must be taken to not change any existing mapping, otherwise it\n leads to partial hw programming\n \n [how]\n it\u0027s not strictly necessary to track which stream/mpcc tree is being updated, but\n rather it\u0027s sufficient to compare current and new state and just keep pipes that were\n previously already mapped unchanged.\n \n Signed-off-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Reviewed-by: Eric Yang \u003ceric.yang2@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 04cf85d2d9a2b177e8d3ee0a8a801ecf8fcf5666\nAuthor: Aric Cyr \u003caric.cyr@amd.com\u003e\nDate: Fri Jul 19 17:41:56 2019 -0400\n\n drm/amd/display: 3.2.44\n \n Signed-off-by: Aric Cyr \u003caric.cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1a9e3d4569fc9a9a1cbfb14cacf3268114646ab6\nAuthor: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\nDate: Wed Jul 17 19:02:14 2019 -0400\n\n drm/amd/display: Set DSC before DIG front-end is connected to its back-end\n \n [why]\n At the time DIG FE is connected to its BE, the clocks in OTG are enabled and\n PHY will also be set up. When DSC has to be used to fit the stream into the\n available bandwidth, without DSC being set DIG could get exposed to the\n higer bandwidth it (or link) could handle. This causes the HW to \"reject\"\n video enable setup (the register shows that video enable was attempted, but\n the status bit shows it as disabled).\n \n [how]\n - Separate DSC setup into DSC register config and DSC PPS SDP setup\n \n - Move most of the DSC setup (register config) to before\n dcn10_link_encoder_connect_dig_be_to_fe() is called\n \n - Set up DSC PPS SDP after DIG FE is connected to its BE. This is because\n setting DSC PPS SDP before that has no effect.\n \n Signed-off-by: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\n Reviewed-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7470673340b118b744e629383e8374ced4100b84\nAuthor: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\nDate: Thu Jul 18 19:13:26 2019 -0400\n\n drm/amd/display: Change DSC policy from slices per column to minimum slice height\n \n [why] Minimum slice height is recommended by VESA DSC Spreadsheet user guide\n \n Signed-off-by: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\n Reviewed-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 33ca0475b8c42fb5650b3b764a2b1217d9a1bb06\nAuthor: Vitaly Prosyak \u003cvitaly.prosyak@amd.com\u003e\nDate: Thu Jul 18 13:07:03 2019 -0500\n\n drm/amd/display: Add HLG support in color module\n \n [Why \u0026 How]\n Support hlg OETF and EOTF based on BT.2100-2\n Follow up is required.\n \n Signed-off-by: Vitaly Prosyak \u003cvitaly.prosyak@amd.com\u003e\n Reviewed-by: Krunoslav Kovac \u003cKrunoslav.Kovac@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Acked-by: Vitaly Prosyak \u003cVitaly.Prosyak@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e00feed03f8dcd3fd29b8da1cb21d0bad7eae10e\nAuthor: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\nDate: Sat Jul 27 09:33:58 2019 +0100\n\n MAINTAINERS: add maintainer for SD-FEC\n \n support\n \n Add maintainer entry for Xilinx SD-FEC driver support.\n \n Signed-off-by: Derek Kiernan \u003cderek.kiernan@xilinx.com\u003e\n Signed-off-by: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\n Link: https://lore.kernel.org/r/1564216438-322406-9-git-send-email-dragan.cvetic@xilinx.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit e6f4c346d32efafa18c99cf80182edf45f41a055\nAuthor: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\nDate: Sat Jul 27 09:33:57 2019 +0100\n\n Docs: misc: xilinx_sdfec: Add documentation\n \n Add SD-FEC driver documentation.\n \n Signed-off-by: Derek Kiernan \u003cderek.kiernan@xilinx.com\u003e\n Signed-off-by: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\n Link: https://lore.kernel.org/r/1564216438-322406-8-git-send-email-dragan.cvetic@xilinx.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 6bd6a690c2e7e710aa7ccefa4edc83f14099907e\nAuthor: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\nDate: Sat Jul 27 09:33:56 2019 +0100\n\n misc: xilinx_sdfec: Add stats \u0026 status ioctls\n \n SD-FEC statistic data are:\n - count of data interface errors (isr_err_count)\n - count of Correctable ECC errors (cecc_count)\n - count of Uncorrectable ECC errors (uecc_count)\n \n Add support:\n 1. clear stats ioctl callback which clears collected\n statistic data,\n 2. get stats ioctl callback which reads a collected\n statistic data,\n 3. set default configuration ioctl callback,\n 4. start ioctl callback enables SD-FEC HW,\n 5. stop ioctl callback disables SD-FEC HW.\n \n In a failed state driver enables the following ioctls:\n - get status\n - get statistics\n - clear stats\n - set default SD-FEC device configuration\n \n Tested-by: Santhosh Dyavanapally \u003cSDYAVANA@xilinx.com\u003e\n Tested by: Punnaiah Choudary Kalluri \u003cpunnaia@xilinx.com\u003e\n Tested-by: Derek Kiernan \u003cderek.kiernan@xilinx.com\u003e\n Tested-by: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\n Signed-off-by: Derek Kiernan \u003cderek.kiernan@xilinx.com\u003e\n Signed-off-by: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\n Link: https://lore.kernel.org/r/1564216438-322406-7-git-send-email-dragan.cvetic@xilinx.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit cc538f609dee49b73545569c49e3abd891fdd8b3\nAuthor: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\nDate: Sat Jul 27 09:33:55 2019 +0100\n\n misc: xilinx_sdfec: Support poll file operation\n \n Support monitoring and detecting the SD-FEC error events\n through IRQ and poll file operation.\n \n The SD-FEC device can detect one-error or multi-error events.\n An error triggers an interrupt which creates and run the ONE_SHOT\n IRQ thread.\n The ONE_SHOT IRQ thread detects type of error and pass that\n information to the poll function.\n The file_operation callback poll(), collects the events and\n updates the statistics accordingly.\n The function poll blocks() on waiting queue which can be\n unblocked by ONE_SHOT IRQ handling thread.\n \n Support SD-FEC interrupt set ioctl callback.\n The SD-FEC can detect two type of errors: coding errors (ECC) and\n a data interface errors (TLAST).\n The errors are events which can trigger an IRQ if enabled.\n The driver can monitor and detect these errors through IRQ.\n Also the driver updates the statistical data.\n \n Tested-by: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\n Signed-off-by: Derek Kiernan \u003cderek.kiernan@xilinx.com\u003e\n Signed-off-by: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\n Link: https://lore.kernel.org/r/1564216438-322406-6-git-send-email-dragan.cvetic@xilinx.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 77dd39d924e650cd20696d790f861dfe26e0cb64\nAuthor: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\nDate: Sat Jul 27 09:33:54 2019 +0100\n\n misc: xilinx_sdfec: Add ability to get/set config\n \n - Add capability to get SD-FEC config data using ioctl\n XSDFEC_GET_CONFIG.\n \n - Add capability to set SD-FEC data order using ioctl\n SDFEC_SET_ORDER.\n \n - Add capability to set SD-FEC bypass option using ioctl\n XSDFEC_SET_BYPASS.\n \n - Add capability to set SD-FEC active state using ioctl\n XSDFEC_IS_ACTIVE.\n \n Tested-by: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\n Signed-off-by: Derek Kiernan \u003cderek.kiernan@xilinx.com\u003e\n Signed-off-by: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\n Link: https://lore.kernel.org/r/1564216438-322406-5-git-send-email-dragan.cvetic@xilinx.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 20ec628e8007ec75c2f884e00004f39eab6289b5\nAuthor: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\nDate: Sat Jul 27 09:33:53 2019 +0100\n\n misc: xilinx_sdfec: Add ability to configure LDPC\n \n Add the capability to configure LDPC mode via the ioctl\n XSDFEC_ADD_LDPC_CODE_PARAMS.\n \n Tested-by: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\n Signed-off-by: Derek Kiernan \u003cderek.kiernan@xilinx.com\u003e\n Signed-off-by: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\n Link: https://lore.kernel.org/r/1564216438-322406-4-git-send-email-dragan.cvetic@xilinx.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 6f86ed820178ba35f87712548e6cd43b91608a6c\nAuthor: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\nDate: Sat Jul 27 09:33:52 2019 +0100\n\n misc: xilinx_sdfec: Add ability to configure turbo\n \n Add the capability to configure and retrieve turbo mode\n via the ioctls XSDFEC_SET_TURBO and XSDFEC_GET_TURBO.\n Add char device interface per DT node present and support\n file operations:\n - open(),\n - close(),\n - unlocked_ioctl(),\n - compat_ioctl().\n \n Tested-by: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\n Signed-off-by: Derek Kiernan \u003cderek.kiernan@xilinx.com\u003e\n Signed-off-by: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\n Link: https://lore.kernel.org/r/1564216438-322406-3-git-send-email-dragan.cvetic@xilinx.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 6d54e455689edcf1f9ef30761dd4fdfdc1cba33a\nAuthor: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\nDate: Sat Jul 27 09:33:51 2019 +0100\n\n misc: xilinx_sdfec: Store driver config and state\n \n Stores configuration based on parameters from the DT\n node and values from the SD-FEC core plus reads\n the default state from the SD-FEC core. To obtain\n values from the core register read, write capabilities\n have been added plus related register map details.\n \n Tested-by: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\n Signed-off-by: Derek Kiernan \u003cderek.kiernan@xilinx.com\u003e\n Signed-off-by: Dragan Cvetic \u003cdragan.cvetic@xilinx.com\u003e\n Link: https://lore.kernel.org/r/1564216438-322406-2-git-send-email-dragan.cvetic@xilinx.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b9e8d95a7bc2ebaea6f8752b19ea3e49c27e2271\nAuthor: Jun Lei \u003cJun.Lei@amd.com\u003e\nDate: Thu Jul 18 10:02:40 2019 -0400\n\n drm/amd/display: clean up DML for DCN2x\n \n [why]\n Previous \"less risky\" implemenation of 3 tiered fallback is no longer necessary since\n DMLv2 has gone through proper validation. v2 can now be used as the default and 1\n level of fallback can be removed\n \n [how]\n remove previous workaround implemenation\n \n Signed-off-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Reviewed-by: Nicholas Kazlauskas \u003cNicholas.Kazlauskas@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5ec43eda85506ddc2f91c3a4e28b38da3f14cf1e\nAuthor: Martin Leung \u003cmartin.leung@amd.com\u003e\nDate: Wed Jul 17 16:08:19 2019 -0400\n\n drm/amd/display: enabling seamless boot sequence for dcn2\n \n [Why]\n Seamless boot (building SW state inheriting BIOS-initialized timing) was\n enabled on DCN2, including fixes\n \n [How]\n Includes fixes for MPC, DPPCLK, and DIG FE mapping/OTG source select/\n Pixel clock.\n \n This is part 2 of 2 for seamless boot NV10\n \n Signed-off-by: Martin Leung \u003cmartin.leung@amd.com\u003e\n Reviewed-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 39bdac36cc139dfaf4ff324250319b79c6c224b8\nAuthor: Martin Leung \u003cmartin.leung@amd.com\u003e\nDate: Sun Jul 28 11:09:11 2019 -0400\n\n drm/amd/display: fix dcn-specific clk_mgr init_clocks\n \n [Why]\n underflow seen on certain monitor setups caused by making dcnxx_init_hw\n generic\n \n [How]\n by moving dcn20_init_hw into dcn10, we added a dcn-specific clk_mgr\n init (dc-\u003eclk_mgr-\u003efuncs-\u003einit_clocks()). Thus, put old clk_mgr\n memset in an else statement so both memsets don\u0027t get set\n \n Signed-off-by: Martin Leung \u003cmartin.leung@amd.com\u003e\n Reviewed-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8a31820b12187c7e31c52e8a00075359d9efac52\nAuthor: Martin Leung \u003cmartin.leung@amd.com\u003e\nDate: Tue Jul 9 15:15:17 2019 -0400\n\n drm/amd/display: Make init_hw and init_pipes generic for seamless boot\n \n [Why]\n For seamless boot the init_hw sequence must be split into\n actual hardware vs pipes, in order to defer pipe initialization to set mode\n and skip of pipe-destructive sequences\n \n [How]\n made dcn10_init_hw and dcn10_init_pipes generic for future dcns to inherit\n deleted dcn20 specific versions. This is part 1 of a 2 partimplementation\n of seamless boot\n \n Signed-off-by: Martin Leung \u003cmartin.leung@amd.com\u003e\n Reviewed-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 61e29b21cb2c6e87122015c8a0c4c3e26a7b4be4\nAuthor: Aric Cyr \u003caric.cyr@amd.com\u003e\nDate: Mon Jul 15 10:34:30 2019 -0400\n\n drm/amd/display: 3.2.43\n \n Signed-off-by: Aric Cyr \u003caric.cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 986936d1a9b29bd40be2acdf91d9e99ecda3bf33\nAuthor: Jun Lei \u003cJun.Lei@amd.com\u003e\nDate: Tue Jul 16 13:02:41 2019 -0400\n\n drm/amd/display: wait for pending complete when enabling a plane\n \n [why]\n When planes are enabled, they must be enabled using VSYNC update (not immediate).\n However, before the VUPDATE occurs, DM may call with an \"immediate\" flip which is address\n only. This operation would normally be okay, but if the locking for immediate flip happens\n to occur before the VUPDATE associated with the initial plane enablement, it will cause HW\n to hang.\n \n [how]\n HWSS should enforce plane enable in HW to be synchronous with the call that enables the plane.\n \n Signed-off-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Reviewed-by: Eric Yang \u003ceric.yang2@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 91db9311945f01901ddb9813ce11364de214a156\nAuthor: Su Sung Chung \u003cSu.Chung@amd.com\u003e\nDate: Mon Jul 8 11:31:39 2019 -0400\n\n drm/amd/display: refactor gpio to allocate hw_container in constructor\n \n [why]\n if dynamic allocation fails during gpio_open, it will cause crash due to\n page fault.\n \n [how]\n handle allocation when gpio object gets created and prevent from calling\n gpio_open if allocation failed\n \n Signed-off-by: Su Sung Chung \u003cSu.Chung@amd.com\u003e\n Reviewed-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 37495fbdf12d1bce30dbc228c901ef45eba35739\nAuthor: Jaehyun Chung \u003cjaehyun.chung@amd.com\u003e\nDate: Mon Jul 15 17:35:56 2019 -0400\n\n drm/amd/display: Add work-around option to skip DCN20 clock updates\n \n [Why] Auto Overclock Memory fails for some systems that don\u0027t support\n p-state.\n \n [How] Implement the workaround, and it\u0027s corresponding enable flag.\n \n Signed-off-by: Jaehyun Chung \u003cjaehyun.chung@amd.com\u003e\n Reviewed-by: Alvin Lee \u003cAlvin.Lee2@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f7f38ffef56b0138f902efd261a6d90680fec2d3\nAuthor: Jun Lei \u003cJun.Lei@amd.com\u003e\nDate: Mon Jul 15 10:41:47 2019 -0400\n\n drm/amd/display: fixup DPP programming sequence\n \n [why]\n DC does not correct account for the fact that DPP DTO is double buffered while DPP ref is not.\n This means that when DPP ref clock is lowered when it\u0027s \"safe to lower\", the DPP blocks that need\n an increased divider will temporarily have actual DPP clock drop below minimum while DTO\n double buffering takes effect. This results in temporary underflow.\n \n [how]\n To fix this, DPP clock cannot be programmed atomically, but rather be broken up into the DTO and the\n ref. Each has a separate \"safe to lower\" logic. When doing \"prepare\" the ref and dividers may only increase.\n When doing \"optimize\", both may decrease. It is guaranteed that we won\u0027t exceed max DPP clock because\n we do not use dividers larger than 1.\n \n Signed-off-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Reviewed-by: Eric Yang \u003ceric.yang2@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9adc8050bf3ca3e49c65e13259a4c310640542f1\nAuthor: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\nDate: Fri Jul 12 15:06:06 2019 -0400\n\n drm/amd/display: make firmware info only load once during dc_bios create\n \n Currently every time DC wants to access firmware info we make a call\n into VBIOS. This makes no sense as there is nothing that can change\n runtime inside fw info and can cause issues when calling unstable\n bios during bringup.\n \n This change eliminate this behavior by only calling bios once for fw\n info and keeping it stored as part of dc_bios.\n \n Signed-off-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Reviewed-by: Chris Park \u003cChris.Park@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 0b226322434c7786381d7a594efca9cc46f85211\nAuthor: David Galiffi \u003cdavid.galiffi@amd.com\u003e\nDate: Mon Jun 24 10:34:13 2019 -0400\n\n drm/amd/display: Synchronous DisplayPort Link Training\n \n [WHY]\n We require a method to perform synchronous link training.\n \n [HOW]\n Sync LT is broken into 3 basic steps.\n \"Begin\" starts the state machine, and resets \"preferred\" link settings.\n \"Attempt\" will attempt to train the link with a given set of training\n parameters.\n \"End\" stops the state machine, and will optionally disable the link phy.\n Between \"Begin\" and \"End\" DPCD:600h must not be set to \"2\"\n (D3:Powered Down).\n Between \"Begin\" and \"End\", there may be multiple \"Attempts\" with different\n training parameters.\n \n Signed-off-by: David Galiffi \u003cdavid.galiffi@amd.com\u003e\n Reviewed-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 36756dcbcf474466f7243fcec21af90ee40e72be\nAuthor: Julian Parkin \u003cjulian.parkin@amd.com\u003e\nDate: Fri Jul 12 17:30:06 2019 -0400\n\n drm/amd/display: Improve sharing of HUBBUB register lists\n \n Add DCN20 common register list that contains registers shared\n between DCN20 generations.\n \n Signed-off-by: Julian Parkin \u003cjulian.parkin@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Aric Cyr \u003cAric.Cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1ee299b9ee14579e6464229854ce30edd076c0ec\nAuthor: Vitaly Prosyak \u003cvitaly.prosyak@amd.com\u003e\nDate: Wed Jul 10 14:20:08 2019 -0500\n\n drm/amd/display: Add 22, 24, and 26 degamma\n \n [Why \u0026 How]\n Support degamma ROM and RAM based on hardware capabilities.\n Some refactoring into color module\n \n Signed-off-by: Vitaly Prosyak \u003cvitaly.prosyak@amd.com\u003e\n Reviewed-by: Gary Kattan \u003cGary.Kattan@amd.com\u003e\n Reviewed-by: Nevenko Stupar \u003cNevenko.Stupar@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Acked-by: Vitaly Prosyak \u003cVitaly.Prosyak@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a463b263032f7c98c5912207db43be1aa34a6438\nAuthor: Bayan Zabihiyan \u003cbayan.zabihiyan@amd.com\u003e\nDate: Wed Jul 10 16:00:53 2019 -0400\n\n drm/amd/display: Fix frames_to_insert math\n \n [Why]\n The math on deciding on how many\n \"frames to insert\" sometimes sent us over the max refresh rate.\n Also integer overflow can occur if we have high refresh rates.\n \n [How]\n Instead of clipping the frame duration such that it doesn’t go below the min,\n just remove a frame from the number of frames to insert. +\n Use unsigned long long for intermediate calculations to prevent\n integer overflow.\n \n Signed-off-by: Bayan Zabihiyan \u003cbayan.zabihiyan@amd.com\u003e\n Reviewed-by: Aric Cyr \u003cAric.Cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1cbcfc975164f397b449efb17f59d81a703090db\nAuthor: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\nDate: Fri Jul 12 10:52:54 2019 -0400\n\n drm/amd/display: fix issue where 252-255 values are clipped\n \n [Why]\n When endpoint is at the boundary of a region, such as at 2^0\u003d1\n we find that the last segment has a sharp slope and some points\n are clipped at the top.\n \n [How]\n If end point is 1, which is exactly at the 2^0 region boundary, we\n need to program an additional region beyond this point.\n \n Signed-off-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Reviewed-by: Aric Cyr \u003cAric.Cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ac42fd639550b386c41d8f3f2dc657d96fde7235\nAuthor: Wenjing Liu \u003cWenjing.Liu@amd.com\u003e\nDate: Wed Jul 10 18:35:18 2019 -0400\n\n drm/amd/display: reset hdmi tmds rate and data scramble on pipe reset\n \n [why]\n hdmi data scramble and tmds rate is not reset during pipe reset.\n \n [how]\n reset hdmi tmds rate and data scramble on pipe reset\n \n Signed-off-by: Wenjing Liu \u003cWenjing.Liu@amd.com\u003e\n Reviewed-by: Chris Park \u003cChris.Park@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 38df070135297746804bf8537ac2e8c66bb37bd5\nAuthor: Wenjing Liu \u003cWenjing.Liu@amd.com\u003e\nDate: Wed Jul 10 18:31:38 2019 -0400\n\n drm/amd/display: reset drr programming on pipe reset\n \n [why]\n drr is still enabled after driver is unloaded causing black screen\n \n [how]\n disable drr during pipe reset.\n \n Signed-off-by: Wenjing Liu \u003cWenjing.Liu@amd.com\u003e\n Reviewed-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9d19c73f37473bfc4c9216e5e48b4684cc2050d2\nAuthor: Yongqiang Sun \u003cyongqiang.sun@amd.com\u003e\nDate: Tue Jun 25 19:08:50 2019 -0400\n\n drm/amd/display: Add DFS reference clock field\n \n Add to clk_mgr_internal struct, for future use.\n \n Signed-off-by: Yongqiang Sun \u003cyongqiang.sun@amd.com\u003e\n Reviewed-by: Yongqiang Sun \u003cyongqiang.sun@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 99a423418aeabe9a572b3e5194f2c06180a066fb\nAuthor: Yongqiang Sun \u003cyongqiang.sun@amd.com\u003e\nDate: Sat Jun 22 18:52:41 2019 -0400\n\n drm/amd/display: Add PIXEL_RATE control regs for more instances\n \n For use by future ASICs\n \n Signed-off-by: Sung Lee \u003csung.lee@amd.com\u003e\n Signed-off-by: Yongqiang Sun \u003cyongqiang.sun@amd.com\u003e\n Reviewed-by: Yongqiang Sun \u003cyongqiang.sun@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 85cb9d506744742ac06114a794fe58a964a24a62\nAuthor: Christian König \u003cchristian.koenig@amd.com\u003e\nDate: Fri Aug 9 17:27:21 2019 +0200\n\n drm/scheduler: use job count instead of peek\n \n The spsc_queue_peek function is accessing queue-\u003ehead which belongs to\n the consumer thread and shouldn\u0027t be accessed by the producer\n \n This is fixing a rare race condition when destroying entities.\n \n Signed-off-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Acked-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Reviewed-by: Monk.liu@amd.com\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2568cedc130866ddca7b7d85824b5d237d2d3812\nAuthor: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nDate: Tue Aug 13 18:10:38 2019 +1000\n\n drm/amdgpu: MODULE_FIRMWARE requires linux/module.h\n \n Fixes: 6a7a0bdbfa0c (\"drm/amdgpu: add psp_v12_0 for renoir (v2)\")\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit dec4f1377c2c8556ce0512cb3ad8c6963313c0da\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Tue Aug 13 11:33:40 2019 +0100\n\n drm/amdgpu/powerplay: fix spelling mistake \"unsuported\" -\u003e \"unsupported\"\n \n There is a spelling mistake in a pr_err error message. Fix it. Also\n add a space after a comma to clean up a checkpatch warning.\n \n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d6e0cbb152b35833a26772c86b338d8297ce609d\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Tue Aug 13 15:46:03 2019 +0800\n\n drm/amdgpu: implement querying ras error count for mmhub\n \n get mmhub ea ras error count by accessing EDC_CNT register\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Guchun Chen \u003cguchun.chen@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f0f50dcfd427f9f4a179048f4b56bff176bacc94\nAuthor: Kevin Wang \u003ckevin1.wang@amd.com\u003e\nDate: Tue Aug 13 16:52:04 2019 +0800\n\n drm/amdgpu: use exiting amdgpu_ctx_total_num_entities function\n \n simplify driver code.\n \n Signed-off-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b81e57fbf92e76ce8b7329650440021ebb4e00a0\nAuthor: Kevin Wang \u003ckevin1.wang@amd.com\u003e\nDate: Tue Aug 13 16:48:28 2019 +0800\n\n drm/amdgpu: fix typo error amdgput -\u003e amdgpu\n \n fix typo error:\n change function name from \"amdgput_ctx_total_num_entities\" to\n \"amdgpu_ctx_total_num_entities\".\n \n Signed-off-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 60adad6ffdaf69b96bb7eedd4c8b6de79143dd0a\nAuthor: Kevin Wang \u003ckevin1.wang@amd.com\u003e\nDate: Tue Aug 13 10:25:25 2019 +0800\n\n drm/amd/powerplay: fix message of SetHardMinByFreq failed when feature is disabled\n \n the direct send message to smc to set hard clokc will failed\n when smc clock dpm feature is disabled.\n so use function of smu_set_hard_freq_range to replace it.\n the function will check feature enablement.\n \n eg: when uclk (mclk) dpm feature is disabled on navi10\n [  300.675901] amdgpu: [powerplay] failed send message: SetHardMinByFreq(28)\n    param: 0x00020064 response 0xfffffffb\n \n Signed-off-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 3f513baef6c437ae1ae7927c74b693b1d268f130\nAuthor: Chengming Gui \u003cJack.Gui@amd.com\u003e\nDate: Mon Aug 12 10:23:04 2019 +0800\n\n drm/amd/powerplay: add arcturus_is_dpm_running function for arcturus\n \n add arcturus_is_dpm_running function\n \n Signed-off-by: Chengming Gui \u003cJack.Gui@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d86c887f5edd398dfa62bcb14d5149a014b3c272\nAuthor: Chengming Gui \u003cJack.Gui@amd.com\u003e\nDate: Mon Aug 12 10:06:34 2019 +0800\n\n drm/amd/powerplay: remove redundancy debug log about smu unsupported features\n \n remove redundancy debug log about smu unsupported features\n \n Signed-off-by: Chengming Gui \u003cJack.Gui@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 244511f386ccb90f3bea889e4d86a289015898f9\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Thu Aug 15 09:27:03 2019 +0200\n\n drm/amdgpu: simplify and cleanup setting the dma mask\n \n Use dma_set_mask_and_coherent to set both masks in one go, and remove\n the no longer required fallback, as the kernel now always accepts\n larger than required DMA masks. Fail the driver probe if we can\u0027t\n set the DMA mask, as that means the system can only support a larger\n mask.\n \n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 03127c58b93d7da5edb53118bdafabdcae3581ef\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Thu Aug 15 09:27:02 2019 +0200\n\n drm/radeon: simplify and cleanup setting the dma mask\n \n Use dma_set_mask_and_coherent to set both masks in one go, and remove\n the no longer required fallback, as the kernel now always accepts\n larger than required DMA masks. Fail the driver probe if we can\u0027t\n set the DMA mask, as that means the system can only support a larger\n mask.\n \n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 90489ce18c3a504c781c8cfcec013258c3459328\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Thu Aug 15 09:27:01 2019 +0200\n\n drm/amdgpu: handle PCIe root ports with addressing limitations\n \n amdgpu uses a need_dma32 flag to indicate to the drm core that some\n allocations need to be done using GFP_DMA32, but it only checks the\n device addressing capabilities to make that decision. Unfortunately\n PCIe root ports that have limited addressing exist as well. Use the\n dma_addressing_limited instead to also take those into account.\n \n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 33b3ad3788aba846fc8b9a065fe2685a0b64f713\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Thu Aug 15 09:27:00 2019 +0200\n\n drm/radeon: handle PCIe root ports with addressing limitations\n \n radeon uses a need_dma32 flag to indicate to the drm core that some\n allocations need to be done using GFP_DMA32, but it only checks the\n device addressing capabilities to make that decision. Unfortunately\n PCIe root ports that have limited addressing exist as well. Use the\n dma_addressing_limited instead to also take those into account.\n \n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reported-by: Atish Patra \u003cAtish.Patra@wdc.com\u003e\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2a77eec0d3ca20481a340d4543930a704b3230f6\nAuthor: Akinobu Mita \u003cakinobu.mita@gmail.com\u003e\nDate: Wed Jul 31 12:00:07 2019 +0200\n\n devcoredump: fix typo in comment\n \n s/dev_coredumpmsg/dev_coredumpsg/ in the kernel-doc\n \n Reviewed-by: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\n Signed-off-by: Akinobu Mita \u003cakinobu.mita@gmail.com\u003e\n Link: https://lore.kernel.org/r/1564243146-5681-3-git-send-email-akinobu.mita@gmail.com\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n Link: https://lore.kernel.org/r/20190731100007.32684-2-johannes@sipsolutions.net\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit ce684d957c5672f3bb55e6b0872932b5b4c39c56\nAuthor: Akinobu Mita \u003cakinobu.mita@gmail.com\u003e\nDate: Wed Jul 31 12:00:06 2019 +0200\n\n devcoredump: use memory_read_from_buffer\n \n Use memory_read_from_buffer() to simplify devcd_readv().\n \n Reviewed-by: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\n Reviewed-by: Johannes Berg \u003cjohannes@sipsolutions.net\u003e\n Signed-off-by: Akinobu Mita \u003cakinobu.mita@gmail.com\u003e\n Link: https://lore.kernel.org/r/1564243146-5681-2-git-send-email-akinobu.mita@gmail.com\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n Link: https://lore.kernel.org/r/20190731100007.32684-1-johannes@sipsolutions.net\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 9c1c5e0bc5ec032e78bf15c302f3b20152f865ad\nAuthor: Saravana Kannan \u003csaravanak@google.com\u003e\nDate: Tue Aug 6 12:26:54 2019 -0700\n\n of/platform: Disable generic device linking code for PowerPC\n \n PowerPC platforms don\u0027t use the generic of/platform code to populate the\n devices from DT. Therefore the generic device linking code is never used\n in PowerPC. Compile it out to avoid warning about unused functions.\n \n If a specific PowerPC platform wants to use this code in the future,\n bringing this back for PowerPC would be trivial. We\u0027ll just need to export\n of_link_to_suppliers() and then let the machine specific files do the\n linking as they populate the devices from DT.\n \n Signed-off-by: Saravana Kannan \u003csaravanak@google.com\u003e\n Link: https://lore.kernel.org/r/20190806192654.138605-2-saravanak@google.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 22ac4318ad95847797de99dccaf059c76cd74efe\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 15 12:15:39 2019 -0300\n\n perf trace: Add --switch-on/--switch-off events\n \n Just like with \u0027perf script\u0027:\n \n # perf trace -e sched:*,syscalls:*sleep* sleep 1\n 0.000 :28345/28345 sched:sched_waking:comm\u003dperf pid\u003d28346 prio\u003d120 target_cpu\u003d005\n 0.005 :28345/28345 sched:sched_wakeup:perf:28346 [120] success\u003d1 CPU:005\n 0.383 sleep/28346 sched:sched_process_exec:filename\u003d/usr/bin/sleep pid\u003d28346 old_pid\u003d28346\n 0.613 sleep/28346 sched:sched_stat_runtime:comm\u003dsleep pid\u003d28346 runtime\u003d607375 [ns] vruntime\u003d23289041218 [ns]\n 0.689 sleep/28346 syscalls:sys_enter_nanosleep:rqtp: 0x7ffc491789b0\n 0.693 sleep/28346 sched:sched_stat_runtime:comm\u003dsleep pid\u003d28346 runtime\u003d72021 [ns] vruntime\u003d23289113239 [ns]\n 0.694 sleep/28346 sched:sched_switch:sleep:28346 [120] S \u003d\u003d\u003e swapper/5:0 [120]\n 1000.787 :0/0 sched:sched_waking:comm\u003dsleep pid\u003d28346 prio\u003d120 target_cpu\u003d005\n 1000.824 :0/0 sched:sched_wakeup:sleep:28346 [120] success\u003d1 CPU:005\n 1000.908 sleep/28346 syscalls:sys_exit_nanosleep:0x0\n 1001.218 sleep/28346 sched:sched_process_exit:comm\u003dsleep pid\u003d28346 prio\u003d120\n # perf trace -e sched:*,syscalls:*sleep* --switch-on\u003dsyscalls:sys_enter_nanosleep sleep 1\n 0.000 sleep/28349 sched:sched_stat_runtime:comm\u003dsleep pid\u003d28349 runtime\u003d603036 [ns] vruntime\u003d23873537697 [ns]\n 0.001 sleep/28349 sched:sched_switch:sleep:28349 [120] S \u003d\u003d\u003e swapper/4:0 [120]\n 1000.392 :0/0 sched:sched_waking:comm\u003dsleep pid\u003d28349 prio\u003d120 target_cpu\u003d004\n 1000.443 :0/0 sched:sched_wakeup:sleep:28349 [120] success\u003d1 CPU:004\n 1000.540 sleep/28349 syscalls:sys_exit_nanosleep:0x0\n 1000.852 sleep/28349 sched:sched_process_exit:comm\u003dsleep pid\u003d28349 prio\u003d120\n # perf trace -e sched:*,syscalls:*sleep* --switch-on\u003dsyscalls:sys_enter_nanosleep --switch-off\u003dsyscalls:sys_exit_nanosleep sleep 1\n 0.000 sleep/28352 sched:sched_stat_runtime:comm\u003dsleep pid\u003d28352 runtime\u003d610543 [ns] vruntime\u003d24811686681 [ns]\n 0.001 sleep/28352 sched:sched_switch:sleep:28352 [120] S \u003d\u003d\u003e swapper/0:0 [120]\n 1000.397 :0/0 sched:sched_waking:comm\u003dsleep pid\u003d28352 prio\u003d120 target_cpu\u003d000\n 1000.440 :0/0 sched:sched_wakeup:sleep:28352 [120] success\u003d1 CPU:000\n #\n # perf trace -e sched:*,syscalls:*sleep* --switch-on\u003dsyscalls:sys_enter_nanosleep --switch-off\u003dsyscalls:sys_exit_nanosleep --show-on-off sleep 1\n 0.000 sleep/28367 syscalls:sys_enter_nanosleep:rqtp: 0x7fffd1a25fc0\n 0.004 sleep/28367 sched:sched_stat_runtime:comm\u003dsleep pid\u003d28367 runtime\u003d628760 [ns] vruntime\u003d22170052672 [ns]\n 0.005 sleep/28367 sched:sched_switch:sleep:28367 [120] S \u003d\u003d\u003e swapper/2:0 [120]\n 1000.367 :0/0 sched:sched_waking:comm\u003dsleep pid\u003d28367 prio\u003d120 target_cpu\u003d002\n 1000.412 :0/0 sched:sched_wakeup:sleep:28367 [120] success\u003d1 CPU:002\n 1000.512 sleep/28367 syscalls:sys_exit_nanosleep:0x0\n #\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Florian Weimer \u003cfweimer@redhat.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: William Cohen \u003cwcohen@redhat.com\u003e\n Link: https://lkml.kernel.org/n/tip-t3ngpt1brcc1fm9gep9gxm4q@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 8b3c9ea7bf8f50ead6787c084cfc6d3a0b1e38aa\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 15 12:02:13 2019 -0300\n\n perf evswitch: Add hint when not finding specified on/off events\n \n If the user specifies a on or off switch event and it isn\u0027t in the\n perf.data file, provide a hint about how to see the events in the\n perf.data evlist:\n \n # perf script --switch-on\u003dsyscall:sys_enter_nanosleep --switch-off\u003dsyscalls:sys_exit_nanosleep\n ERROR: event_on event not found (syscall:sys_enter_nanosleep)\n HINT: use \u0027perf evlist\u0027 to see the available event names\n #\n # perf evlist\n sched:sched_kthread_stop\n sched:sched_kthread_stop_ret\n sched:sched_waking\n sched:sched_wakeup\n sched:sched_wakeup_new\n sched:sched_switch\n sched:sched_migrate_task\n sched:sched_process_free\n sched:sched_process_exit\n sched:sched_wait_task\n sched:sched_process_wait\n sched:sched_process_fork\n sched:sched_process_exec\n sched:sched_stat_wait\n sched:sched_stat_sleep\n sched:sched_stat_iowait\n sched:sched_stat_blocked\n sched:sched_stat_runtime\n sched:sched_pi_setprio\n sched:sched_move_numa\n sched:sched_stick_numa\n sched:sched_swap_numa\n sched:sched_wake_idle_without_ipi\n syscalls:sys_enter_clock_nanosleep\n syscalls:sys_exit_clock_nanosleep\n syscalls:sys_enter_nanosleep\n syscalls:sys_exit_nanosleep\n # Tip: use \u0027perf evlist --trace-fields\u0027 to show fields for tracepoint events\n #\n # perf script --switch-on\u003dsyscalls:sys_enter_nanosleep --switch-off\u003dsyscalls:sys_exit_nanosleep\n sleep 20919 [001] 109866.144411: sched:sched_stat_runtime: comm\u003dsleep pid\u003d20919 runtime\u003d521249 [ns] vruntime\u003d202919398131 [ns]\n sleep 20919 [001] 109866.144412: sched:sched_switch: sleep:20919 [120] S \u003d\u003d\u003e swapper/1:0 [120]\n swapper 0 [001] 109867.144568: sched:sched_waking: comm\u003dsleep pid\u003d20919 prio\u003d120 target_cpu\u003d001\n swapper 0 [001] 109867.144586: sched:sched_wakeup: sleep:20919 [120] success\u003d1 CPU:001\n #\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Florian Weimer \u003cfweimer@redhat.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: William Cohen \u003cwcohen@redhat.com\u003e\n Link: https://lkml.kernel.org/n/tip-iijjvdlyad973oskdq8gmi5w@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit c9a4269930dada68971a4a97f3abf079af8cde4e\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 15 11:35:56 2019 -0300\n\n perf evswitch: Move enoent error message printing to separate function\n \n Allows adding hints there, will be done in followup patch.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Florian Weimer \u003cfweimer@redhat.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: William Cohen \u003cwcohen@redhat.com\u003e\n Link: https://lkml.kernel.org/n/tip-1kvrdi7weuz3hxycwvarcu6v@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 124e02be72fdff05ab5d7f004a3c0d4061569380\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 15 11:31:29 2019 -0300\n\n perf evswitch: Introduce init() method to set the on/off evsels from the command line\n \n Another step in having all the boilerplate in just one place to then use\n in the other tools.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Florian Weimer \u003cfweimer@redhat.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: William Cohen \u003cwcohen@redhat.com\u003e\n Link: https://lkml.kernel.org/n/tip-snreb1wmwyjei3eefwotxp1l@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit add3a719c95f0443d563889b4af255b78ba54521\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 15 11:21:21 2019 -0300\n\n perf evswitch: Introduce OPTS_EVSWITCH() for cmd line processing\n \n All tools will want those, so provide a convenient way to get them.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Florian Weimer \u003cfweimer@redhat.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: William Cohen \u003cwcohen@redhat.com\u003e\n Link: https://lkml.kernel.org/n/tip-v16pe3sbf3wjmn152u18f649@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 0b495b121585a1b6ca120fe13f950e2f86ca8197\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 15 11:11:14 2019 -0300\n\n perf evswitch: Add the names of on/off events\n \n So that we can have macros for the OPT_ entries and also for finding\n those in an evlist, this way other tools will use this very easily.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Florian Weimer \u003cfweimer@redhat.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: William Cohen \u003cwcohen@redhat.com\u003e\n Link: https://lkml.kernel.org/n/tip-q0og1xoqqi0w38ve5u0a43k2@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 8829e56fa050998164e496d380cd69186ae9b8d0\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 15 11:00:11 2019 -0300\n\n perf evswitch: Move switch logic to use in other tools\n \n Now other tools that want switching can use an evswitch for that, just\n set it up and add it to the PERF_RECORD_SAMPLE processing function.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Florian Weimer \u003cfweimer@redhat.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: William Cohen \u003cwcohen@redhat.com\u003e\n Link: https://lkml.kernel.org/n/tip-b1trj1q97qwfv251l66q3noj@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit d2360442725fd29b3189887476c57059854a398c\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Aug 15 10:37:24 2019 -0300\n\n perf evswitch: Move struct to a separate header to use in other tools\n \n Now that we see that the simple userspace-based \"slicing\" of events\n using delimiter events (\"markers\") works, lets move it to a separate\n header to make it available to other tools, next step will be having\n the switch on/off check done at the PERF_RECORD_SAMPLE processing\n function moved too.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Florian Weimer \u003cfweimer@redhat.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: William Cohen \u003cwcohen@redhat.com\u003e\n Link: https://lkml.kernel.org/n/tip-z0cyi9ifzlr37cedr9xztc1k@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit dd41f660c03a6d8f2c2f3b2cccf50d8c4e06dd42\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 14 16:51:28 2019 -0300\n\n perf script: Allow specifying event to switch off processing of other events\n \n Counterpart of --switch-on:\n \n # perf record -e sched:*,syscalls:sys_*_nanosleep sleep 1\n [ perf record: Woken up 36 times to write data ]\n [ perf record: Captured and wrote 0.032 MB perf.data (10 samples) ]\n #\n # perf script\n :20918 20918 [002] 109866.143696: sched:sched_waking: comm\u003dperf pid\u003d20919 prio\u003d120 target_cpu\u003d001\n :20918 20918 [002] 109866.143702: sched:sched_wakeup: perf:20919 [120] success\u003d1 CPU:001\n sleep 20919 [001] 109866.144081: sched:sched_process_exec: filename\u003d/usr/bin/sleep pid\u003d20919 old_pid\u003d20919\n sleep 20919 [001] 109866.144408: syscalls:sys_enter_nanosleep: rqtp: 0x7ffc2384fef0, rmtp: 0x00000000\n sleep 20919 [001] 109866.144411: sched:sched_stat_runtime: comm\u003dsleep pid\u003d20919 runtime\u003d521249 [ns] vruntime\u003d202919398131 [n\u003e\n sleep 20919 [001] 109866.144412: sched:sched_switch: sleep:20919 [120] S \u003d\u003d\u003e swapper/1:0 [120]\n swapper 0 [001] 109867.144568: sched:sched_waking: comm\u003dsleep pid\u003d20919 prio\u003d120 target_cpu\u003d001\n swapper 0 [001] 109867.144586: sched:sched_wakeup: sleep:20919 [120] success\u003d1 CPU:001\n sleep 20919 [001] 109867.144614: syscalls:sys_exit_nanosleep: 0x0\n sleep 20919 [001] 109867.144753: sched:sched_process_exit: comm\u003dsleep pid\u003d20919 prio\u003d120\n #\n # perf script --switch-off syscalls:sys_exit_nanosleep\n :20918 20918 [002] 109866.143696: sched:sched_waking: comm\u003dperf pid\u003d20919 prio\u003d120 target_cpu\u003d001\n :20918 20918 [002] 109866.143702: sched:sched_wakeup: perf:20919 [120] success\u003d1 CPU:001\n sleep 20919 [001] 109866.144081: sched:sched_process_exec: filename\u003d/usr/bin/sleep pid\u003d20919 old_pid\u003d20919\n sleep 20919 [001] 109866.144408: syscalls:sys_enter_nanosleep: rqtp: 0x7ffc2384fef0, rmtp: 0x00000000\n sleep 20919 [001] 109866.144411: sched:sched_stat_runtime: comm\u003dsleep pid\u003d20919 runtime\u003d521249 [ns] vruntime\u003d202919398131 [n\u003e\n sleep 20919 [001] 109866.144412: sched:sched_switch: sleep:20919 [120] S \u003d\u003d\u003e swapper/1:0 [120]\n swapper 0 [001] 109867.144568: sched:sched_waking: comm\u003dsleep pid\u003d20919 prio\u003d120 target_cpu\u003d001\n swapper 0 [001] 109867.144586: sched:sched_wakeup: sleep:20919 [120] success\u003d1 CPU:001\n sleep 20919 [001] 109867.144753: sched:sched_process_exit: comm\u003dsleep pid\u003d20919 prio\u003d120\n #\n # perf script --switch-on syscalls:sys_enter_nanosleep --switch-off syscalls:sys_exit_nanosleep\n sleep 20919 [001] 109866.144411: sched:sched_stat_runtime: comm\u003dsleep pid\u003d20919 runtime\u003d521249 [ns] vruntime\u003d202919398131 [n\u003e\n sleep 20919 [001] 109866.144412: sched:sched_switch: sleep:20919 [120] S \u003d\u003d\u003e swapper/1:0 [120]\n swapper 0 [001] 109867.144568: sched:sched_waking: comm\u003dsleep pid\u003d20919 prio\u003d120 target_cpu\u003d001\n swapper 0 [001] 109867.144586: sched:sched_wakeup: sleep:20919 [120] success\u003d1 CPU:001\n #\n # perf script --switch-on syscalls:sys_enter_nanosleep --switch-off syscalls:sys_exit_nanosleep --show-on-off\n sleep 20919 [001] 109866.144408: syscalls:sys_enter_nanosleep: rqtp: 0x7ffc2384fef0, rmtp: 0x00000000\n sleep 20919 [001] 109866.144411: sched:sched_stat_runtime: comm\u003dsleep pid\u003d20919 runtime\u003d521249 [ns] vruntime\u003d202919398131 [n\u003e\n sleep 20919 [001] 109866.144412: sched:sched_switch: sleep:20919 [120] S \u003d\u003d\u003e swapper/1:0 [120]\n swapper 0 [001] 109867.144568: sched:sched_waking: comm\u003dsleep pid\u003d20919 prio\u003d120 target_cpu\u003d001\n swapper 0 [001] 109867.144586: sched:sched_wakeup: sleep:20919 [120] success\u003d1 CPU:001\n sleep 20919 [001] 109867.144614: syscalls:sys_exit_nanosleep: 0x0\n #\n \n Now think about using this together with \u0027perf probe\u0027 to create custom on/off\n events in your app :-)\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Florian Weimer \u003cfweimer@redhat.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: William Cohen \u003cwcohen@redhat.com\u003e\n Link: https://lkml.kernel.org/n/tip-li3j01c4tmj9kw6ydsl8swej@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 6469eb6dffeb44edfa3d4ca496b044b4a9c643b9\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 14 16:40:58 2019 -0300\n\n perf script: Allow showing the --switch-on event\n \n One may want to see the --switch-on event as well, allow for that, using\n the previous cset example:\n \n # perf script --switch-on syscalls:sys_enter_nanosleep --show-on-off\n sleep 13638 [001] 108237.582286: syscalls:sys_enter_nanosleep: rqtp: 0x7fff1948ac40, rmtp: 0x00000000\n sleep 13638 [001] 108237.582289: sched:sched_stat_runtime: comm\u003dsleep pid\u003d13638 runtime\u003d578104 [ns] vruntime\u003d202889459556 [ns]\n sleep 13638 [001] 108237.582291: sched:sched_switch: sleep:13638 [120] S \u003d\u003d\u003e swapper/1:0 [120]\n swapper 0 [001] 108238.582428: sched:sched_waking: comm\u003dsleep pid\u003d13638 prio\u003d120 target_cpu\u003d001\n swapper 0 [001] 108238.582458: sched:sched_wakeup: sleep:13638 [120] success\u003d1 CPU:001\n sleep 13638 [001] 108238.582698: sched:sched_stat_runtime: comm\u003dsleep pid\u003d13638 runtime\u003d173915 [ns] vruntime\u003d202889633471 [ns]\n sleep 13638 [001] 108238.582782: sched:sched_process_exit: comm\u003dsleep pid\u003d13638 prio\u003d120\n #\n # perf script --switch-on syscalls:sys_enter_nanosleep\n sleep 13638 [001] 108237.582289: sched:sched_stat_runtime: comm\u003dsleep pid\u003d13638 runtime\u003d578104 [ns] vruntime\u003d202889459556 [ns]\n sleep 13638 [001] 108237.582291: sched:sched_switch: sleep:13638 [120] S \u003d\u003d\u003e swapper/1:0 [120]\n swapper 0 [001] 108238.582428: sched:sched_waking: comm\u003dsleep pid\u003d13638 prio\u003d120 target_cpu\u003d001\n swapper 0 [001] 108238.582458: sched:sched_wakeup: sleep:13638 [120] success\u003d1 CPU:001\n sleep 13638 [001] 108238.582698: sched:sched_stat_runtime: comm\u003dsleep pid\u003d13638 runtime\u003d173915 [ns] vruntime\u003d202889633471 [ns]\n sleep 13638 [001] 108238.582782: sched:sched_process_exit: comm\u003dsleep pid\u003d13638 prio\u003d120\n #\n \n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: William Cohen \u003cwcohen@redhat.com\u003e\n Cc: Florian Weimer \u003cfweimer@redhat.com\u003e\n Link: https://lkml.kernel.org/n/tip-0omwwoywj1v63gu8cz0tr0cy@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit f90a24171a8179a29e5e1532fd5bb94e59b5380e\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 14 16:20:13 2019 -0300\n\n perf script: Allow specifying event to switch on processing of other events\n \n Sometime we want to only consider events after something happens, so\n allow discarding events till such events is found, e.g.:\n \n Record all scheduler tracepoints and the sys_enter_nanosleep syscall\n event for the \u0027sleep 1\u0027 workload:\n \n # perf record -e sched:*,syscalls:sys_enter_nanosleep sleep 1\n [ perf record: Woken up 31 times to write data ]\n [ perf record: Captured and wrote 0.032 MB perf.data (10 samples) ]\n #\n \n So we have these events in the generated perf data file:\n \n # perf evlist\n sched:sched_kthread_stop\n sched:sched_kthread_stop_ret\n sched:sched_waking\n sched:sched_wakeup\n sched:sched_wakeup_new\n sched:sched_switch\n sched:sched_migrate_task\n sched:sched_process_free\n sched:sched_process_exit\n sched:sched_wait_task\n sched:sched_process_wait\n sched:sched_process_fork\n sched:sched_process_exec\n sched:sched_stat_wait\n sched:sched_stat_sleep\n sched:sched_stat_iowait\n sched:sched_stat_blocked\n sched:sched_stat_runtime\n sched:sched_pi_setprio\n sched:sched_move_numa\n sched:sched_stick_numa\n sched:sched_swap_numa\n sched:sched_wake_idle_without_ipi\n syscalls:sys_enter_nanosleep\n # Tip: use \u0027perf evlist --trace-fields\u0027 to show fields for tracepoint events\n #\n \n Then show all of the events that actually took place in this \u0027perf record\u0027 session:\n \n # perf script\n :13637 13637 [002] 108237.581529: sched:sched_waking: comm\u003dperf pid\u003d13638 prio\u003d120 target_cpu\u003d001\n :13637 13637 [002] 108237.581537: sched:sched_wakeup: perf:13638 [120] success\u003d1 CPU:001\n sleep 13638 [001] 108237.581992: sched:sched_process_exec: filename\u003d/usr/bin/sleep pid\u003d13638 old_pid\u003d13638\n sleep 13638 [001] 108237.582286: syscalls:sys_enter_nanosleep: rqtp: 0x7fff1948ac40, rmtp: 0x00000000\n sleep 13638 [001] 108237.582289: sched:sched_stat_runtime: comm\u003dsleep pid\u003d13638 runtime\u003d578104 [ns] vruntime\u003d202889459556 [ns]\n sleep 13638 [001] 108237.582291: sched:sched_switch: sleep:13638 [120] S \u003d\u003d\u003e swapper/1:0 [120]\n swapper 0 [001] 108238.582428: sched:sched_waking: comm\u003dsleep pid\u003d13638 prio\u003d120 target_cpu\u003d001\n swapper 0 [001] 108238.582458: sched:sched_wakeup: sleep:13638 [120] success\u003d1 CPU:001\n sleep 13638 [001] 108238.582698: sched:sched_stat_runtime: comm\u003dsleep pid\u003d13638 runtime\u003d173915 [ns] vruntime\u003d202889633471 [ns]\n sleep 13638 [001] 108238.582782: sched:sched_process_exit: comm\u003dsleep pid\u003d13638 prio\u003d120\n #\n \n Now lets see only the ones that took place after a certain \"marker\":\n \n # perf script --switch-on syscalls:sys_enter_nanosleep\n sleep 13638 [001] 108237.582289: sched:sched_stat_runtime: comm\u003dsleep pid\u003d13638 runtime\u003d578104 [ns] vruntime\u003d202889459556 [ns]\n sleep 13638 [001] 108237.582291: sched:sched_switch: sleep:13638 [120] S \u003d\u003d\u003e swapper/1:0 [120]\n swapper 0 [001] 108238.582428: sched:sched_waking: comm\u003dsleep pid\u003d13638 prio\u003d120 target_cpu\u003d001\n swapper 0 [001] 108238.582458: sched:sched_wakeup: sleep:13638 [120] success\u003d1 CPU:001\n sleep 13638 [001] 108238.582698: sched:sched_stat_runtime: comm\u003dsleep pid\u003d13638 runtime\u003d173915 [ns] vruntime\u003d202889633471 [ns]\n sleep 13638 [001] 108238.582782: sched:sched_process_exit: comm\u003dsleep pid\u003d13638 prio\u003d120\n #\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Florian Weimer \u003cfweimer@redhat.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: William Cohen \u003cwcohen@redhat.com\u003e\n Link: https://lkml.kernel.org/n/tip-f1oo0ufdhrkx6nhy2lj1ierm@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 11e54d35e6d5c3533b706753224ef38ea235684b\nAuthor: Haiyan Song \u003chaiyanx.song@intel.com\u003e\nDate: Thu Aug 15 11:59:42 2019 +0800\n\n perf vendor events intel: Add Tremontx event file v1.02\n \n Add a Intel event file for perf.\n \n Signed-off-by: Haiyan Song \u003chaiyanx.song@intel.com\u003e\n Reviewed-by: Kan Liang \u003ckan.liang@linux.intel.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Jin Yao \u003cyao.jin@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190815035942.30602-1-haiyanx.song@intel.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 9e127fab67e3ec4451696da0c7872fd291b9372b\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Tue Aug 13 14:31:14 2019 +0100\n\n regulator: max8660: remove redundant assignment of variable ret\n \n Variable ret is initialized to a value that is never read before\n a return statement and hence can be removed. Remove it.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190813133114.14931-1-colin.king@canonical.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 3829100a63724f6dbf264b2a7f06e7f638ed952d\nAuthor: Andreas Kemnade \u003candreas@kemnade.info\u003e\nDate: Wed Aug 14 23:43:19 2019 +0200\n\n regulator: twl: voltage lists for vdd1/2 on twl4030\n \n _opp_supported_by_regulators() wrongly ignored errors from\n regulator_is_supported_voltage(), so it considered errors as\n success. Since\n commit 498209445124 (\"regulator: core: simplify return value on suported_voltage\")\n regulator_is_supported_voltage() returns a real boolean, so\n errors make _opp_supported_by_regulators() return false.\n \n That reveals a problem with the declaration of the VDD1/2\n regulators on twl4030.\n The VDD1/VDD2 regulators on twl4030 are neither defined with\n voltage lists nor with the continuous flag set, so\n regulator_is_supported_voltage() returns false and an error\n before above mentioned commit (which was considered success)\n The result is that after the above mentioned commit cpufreq\n does not work properly e.g. dm3730.\n \n [ 2.490997] core: _opp_supported_by_regulators: OPP minuV: 1012500 maxuV: 1012500, not supported by regulator\n [ 2.501617] cpu cpu0: _opp_add: OPP not supported by regulators (300000000)\n [ 2.509246] core: _opp_supported_by_regulators: OPP minuV: 1200000 maxuV: 1200000, not supported by regulator\n [ 2.519775] cpu cpu0: _opp_add: OPP not supported by regulators (600000000)\n [ 2.527313] core: _opp_supported_by_regulators: OPP minuV: 1325000 maxuV: 1325000, not supported by regulator\n [ 2.537750] cpu cpu0: _opp_add: OPP not supported by regulators (800000000)\n \n The patch fixes declaration of VDD1/2 regulators by\n adding proper voltage lists.\n \n Fixes: 498209445124 (\"regulator: core: simplify return value on suported_voltage\")\n Cc: stable@vger.kernel.org\n Signed-off-by: Andreas Kemnade \u003candreas@kemnade.info\u003e\n Tested-by: Adam Ford \u003caford173@gmail.com\u003e #logicpd-torpedo-37xx-devkit\n Link: https://lore.kernel.org/r/20190814214319.24087-1-andreas@kemnade.info\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 72b745e3ad65deac94ea4eb83262c52ba3ffdb5b\nAuthor: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\nDate: Tue Aug 13 13:45:32 2019 +0300\n\n ASoC: core: Move pcm_mutex up to card level from snd_soc_pcm_runtime\n \n The pcm_mutex is used to prevent concurrent execution of snd_pcm_ops\n callbacks. This works fine most of the cases but it can not handle setups\n when the same DAI is used by different rtd, for example:\n pcm3168a have two DAIs: one for Playback and one for Capture.\n If the codec is connected to a single CPU DAI we need to have two dai_link\n to support both playback and capture.\n \n In this case the snd_pcm_ops callbacks can be executed in parallel causing\n unexpected races in DAI drivers.\n \n By moving the pcm_mutex up to card level this can be solved\n while - hopefully - not breaking other setups.\n \n Signed-off-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Tested-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190813104532.16669-1-peter.ujfalusi@ti.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 12f0bfadf69bb154052722e7e4e5cd1639044c76\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 15 17:25:47 2019 +0800\n\n ASoC: 88pm860x: remove unused variables \u0027pcm_switch_controls\u0027 and \u0027aif1_mux\u0027\n \n sound/soc/codecs/88pm860x-codec.c:533:38: warning:\n pcm_switch_controls defined but not used [-Wunused-const-variable\u003d]\n sound/soc/codecs/88pm860x-codec.c:560:38: warning:\n aif1_mux defined but not used [-Wunused-const-variable\u003d]\n \n They are never used, so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190815092547.29564-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit e33d565795930d0341bb946aec457a814ccd53e6\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 15 17:24:36 2019 +0800\n\n ASoC: cs42l56: remove unused variable \u0027adc_swap_enum\u0027\n \n sound/soc/codecs/cs42l56.c:206:30: warning:\n adc_swap_enum defined but not used [-Wunused-const-variable\u003d]\n \n It is never used, so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190815092436.34632-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit d63887bc4f50fede7013bda7c733d58ecc43efc1\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 15 17:23:00 2019 +0800\n\n ASoC: es8328: Fix copy-paste error in es8328_right_line_controls\n \n It seems \u0027es8328_rline_enum\u0027 should be used\n in es8328_right_line_controls\n \n Fixes: 567e4f98922c (\"ASoC: add es8328 codec driver\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190815092300.68712-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 97d39be9ce5befc2e36f15d0df33832e0f633565\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 15 17:20:56 2019 +0800\n\n ASoC: es8328: remove unused variable \u0027pga_tlv\u0027\n \n sound/soc/codecs/es8328.c:102:35: warning:\n pga_tlv defined but not used [-Wunused-const-variable\u003d]\n \n They are never used, so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190815092056.28724-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 6d6376b143d59ab1b8635807c78d224d03580418\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 15 17:17:38 2019 +0800\n\n ASoC: tlv320aic31xx: remove unused variable \u0027cm_m_enum\u0027\n \n sound/soc/codecs/tlv320aic31xx.c:261:29: warning:\n cm_m_enum defined but not used [-Wunused-const-variable\u003d]\n \n It is never used, so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190815091738.21680-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit ab0ac2707784a966927c229752849c343bd1dbbf\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 15 17:15:34 2019 +0800\n\n ASoC: tlv320aic23: remove unused variable \u0027tlv320aic23_rec_src\u0027\n \n sound/soc/codecs/tlv320aic23.c:70:29: warning:\n tlv320aic23_rec_src defined but not used [-Wunused-const-variable\u003d]\n \n It is never used, so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190815091534.57780-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 5b366753c1c12feead0ae53b45482f569ed5399c\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 15 17:06:02 2019 +0800\n\n ASoC: rt1011: remove unused variable \u0027dac_vol_tlv\u0027 and \u0027adc_vol_tlv\u0027\n \n sound/soc/codecs/rt1011.c:981:35: warning:\n dac_vol_tlv defined but not used [-Wunused-const-variable\u003d]\n sound/soc/codecs/rt1011.c:982:35: warning:\n adc_vol_tlv defined but not used [-Wunused-const-variable\u003d]\n \n They are never used, so can be removed.\n \n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190815090602.9000-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9d22142c9b1ec8612b880121dd0bc27311cbb2b5\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 15 17:04:04 2019 +0800\n\n ASoC: max98371: remove unused variable \u0027max98371_noload_gain_tlv\u0027\n \n sound/soc/codecs/max98371.c:157:35: warning:\n max98371_noload_gain_tlv defined but not used [-Wunused-const-variable\u003d]\n \n It is never used, so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190815090404.72752-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit bed7f1469f08fd123cdec7a351ef0d875feadcf6\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 15 17:01:57 2019 +0800\n\n ASoC: cs4349: Use PM ops \u0027cs4349_runtime_pm\u0027\n \n sound/soc/codecs/cs4349.c:358:32: warning:\n cs4349_runtime_pm defined but not used [-Wunused-const-variable\u003d]\n \n cs4349_runtime_pm ops already defined, it seems\n we should enable it.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Fixes: e40da86 (\"ASoC: cs4349: Add support for Cirrus Logic CS4349\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190815090157.70036-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c25b456dc579298ac0ed7304f7d06a66288e96df\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 15 16:54:54 2019 +0800\n\n ASoC: cs42l73: remove unused variables \u0027vsp_output_mux\u0027 and \u0027xsp_output_mux\u0027\n \n sound/soc/codecs/cs42l73.c:276:38: warning:\n vsp_output_mux defined but not used [-Wunused-const-variable\u003d]\n sound/soc/codecs/cs42l73.c:279:38: warning:\n xsp_output_mux defined but not used [-Wunused-const-variable\u003d]\n \n They are never used, so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190815085454.30384-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit ff30779bd50eb823e2e75cfc7cc8a3373bfa3fd9\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Fri Jun 21 14:31:16 2019 +0300\n\n ASoC: Intel: Skylake: Print constant literals from format specifier\n \n Instead of using two additional \"%s\" specifiers, put the constant string\n literals directly to the format specifier.\n \n Cc: Liam Girdwood \u003clgirdwood@gmail.com\u003e\n Cc: Mark Brown \u003cbroonie@kernel.org\u003e\n Cc: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190621113116.47525-1-andriy.shevchenko@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit ccb1fa21ef58a2ac15519bb878470762e967e8b3\nAuthor: Jiaxin Yu \u003cjiaxin.yu@mediatek.com\u003e\nDate: Mon Aug 5 12:56:37 2019 +0800\n\n ASoC: mediatek: mt6358: add delay after dmic clock on\n \n Most dmics produce a high level when they receive clock. The difference\n between power-on and memory record time is about 10ms, but the dmic\n needs 50ms to output normal data.\n \n This commit add 100ms delay after SoC output clock so that we can cut\n off the pop noise at the beginning.\n \n Signed-off-by: Jiaxin Yu \u003cjiaxin.yu@mediatek.com\u003e\n Link: https://lore.kernel.org/r/1564980997-11359-1-git-send-email-jiaxin.yu@mediatek.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit d59170b42610c7cbc6e96431ca8357a8bdbf592b\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Tue Aug 13 22:41:22 2019 +0800\n\n ASoC: mediatek: mt8183-mt6358-ts3a227-max98357: remove unused variables\n \n sound/soc/mediatek/mt8183/mt8183-mt6358-ts3a227-max98357.c:50:1: warning:\n mt8183_mt6358_ts3a227_max98357_dapm_widgets defined but not used [-Wunused-const-variable\u003d]\n sound/soc/mediatek/mt8183/mt8183-mt6358-ts3a227-max98357.c:55:1: warning:\n mt8183_mt6358_ts3a227_max98357_dapm_routes defined but not used [-Wunused-const-variable\u003d]\n \n They are never used, so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190813144122.67676-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 57c3ed42f52cdc51f416c93b19708ef6ceb4a00b\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Tue Aug 13 22:39:52 2019 +0800\n\n ASoC: mediatek: mt8183-da7219-max98357: remove unused variable\n \n sound/soc/mediatek/mt8183/mt8183-da7219-max98357.c:120:1: warning:\n mt8183_da7219_max98357_dapm_widgets defined but not used [-Wunused-const-variable\u003d]\n sound/soc/mediatek/mt8183/mt8183-da7219-max98357.c:124:40: warning:\n mt8183_da7219_max98357_dapm_routes defined but not used [-Wunused-const-variable\u003d]\n \n They are never used, so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190813143952.29232-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit a9e792d006edbd33724f2eb858887d3b591d82c5\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Tue Aug 13 22:38:11 2019 +0800\n\n ASoC: mt2701: remove unused variables\n \n sound/soc/mediatek/mt2701/mt2701-afe-pcm.c:799:38: warning:\n mt2701_afe_o23_mix defined but not used [-Wunused-const-variable\u003d]\n sound/soc/mediatek/mt2701/mt2701-afe-pcm.c:803:38: warning:\n mt2701_afe_o24_mix defined but not used [-Wunused-const-variable\u003d]\n sound/soc/mediatek/mt2701/mt2701-afe-pcm.c:835:38: warning:\n mt2701_afe_multi_ch_out_i2s4 defined but not used [-Wunused-const-variable\u003d]\n \n They are never used, so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190813143811.31456-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 0faf1237c60a3791d7ff32035d3097d3e022e68f\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Tue Aug 13 22:25:01 2019 +0800\n\n ASoC: soc-core: Fix -Wunused-const-variable warning\n \n If CONFIG_DMI is not set, gcc warns:\n \n sound/soc/soc-core.c:81:27: warning:\n dmi_blacklist defined but not used [-Wunused-const-variable\u003d]\n \n Add #ifdef guard around it.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190813142501.13080-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 554b75bde64bcad9662530726d1483f7ef012069\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 15 17:19:20 2019 +0800\n\n ASoC: wm8737: Fix copy-paste error in wm8737_snd_controls\n \n sound/soc/codecs/wm8737.c:112:29: warning:\n high_3d defined but not used [-Wunused-const-variable\u003d]\n \n \u0027high_3d\u0027 should be used for 3D High Cut-off.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Fixes: 2a9ae13a2641 (\"ASoC: Add initial WM8737 driver\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Acked-by: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n Link: https://lore.kernel.org/r/20190815091920.64480-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9b4275c415acca6264a3d7f1182589959c93d530\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 15 17:01:57 2019 +0800\n\n ASoC: cs4349: Use PM ops \u0027cs4349_runtime_pm\u0027\n \n sound/soc/codecs/cs4349.c:358:32: warning:\n cs4349_runtime_pm defined but not used [-Wunused-const-variable\u003d]\n \n cs4349_runtime_pm ops already defined, it seems\n we should enable it.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Fixes: e40da86 (\"ASoC: cs4349: Add support for Cirrus Logic CS4349\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190815090157.70036-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 630742c296341a8cfe00dfd941392025ba8dd4e8\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 15 17:23:00 2019 +0800\n\n ASoC: es8328: Fix copy-paste error in es8328_right_line_controls\n \n It seems \u0027es8328_rline_enum\u0027 should be used\n in es8328_right_line_controls\n \n Fixes: 567e4f98922c (\"ASoC: add es8328 codec driver\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190815092300.68712-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 2b87a2553aa04105724d6c844e223415985246ed\nAuthor: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\nDate: Thu May 16 12:04:37 2019 +1000\n\n powerpc/64s: Make boot look nice(r)\n \n Radix boot looks like this:\n \n -----------------------------------------------------\n phys_mem_size \u003d 0x200000000\n dcache_bsize \u003d 0x80\n icache_bsize \u003d 0x80\n cpu_features \u003d 0x0000c06f8f5fb1a7\n possible \u003d 0x0000fbffcf5fb1a7\n always \u003d 0x00000003800081a1\n cpu_user_features \u003d 0xdc0065c2 0xaee00000\n mmu_features \u003d 0xbc006041\n firmware_features \u003d 0x0000000010000000\n hash-mmu: ppc64_pft_size \u003d 0x0\n hash-mmu: kernel vmalloc start \u003d 0xc008000000000000\n hash-mmu: kernel IO start \u003d 0xc00a000000000000\n hash-mmu: kernel vmemmap start \u003d 0xc00c000000000000\n -----------------------------------------------------\n \n Fix:\n \n -----------------------------------------------------\n phys_mem_size \u003d 0x200000000\n dcache_bsize \u003d 0x80\n icache_bsize \u003d 0x80\n cpu_features \u003d 0x0000c06f8f5fb1a7\n possible \u003d 0x0000fbffcf5fb1a7\n always \u003d 0x00000003800081a1\n cpu_user_features \u003d 0xdc0065c2 0xaee00000\n mmu_features \u003d 0xbc006041\n firmware_features \u003d 0x0000000010000000\n vmalloc start \u003d 0xc008000000000000\n IO start \u003d 0xc00a000000000000\n vmemmap start \u003d 0xc00c000000000000\n -----------------------------------------------------\n \n Signed-off-by: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190516020437.11783-1-npiggin@gmail.com\n\ncommit fd3806562f450a6189c31e0d2cb9cd4b208dcf2d\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Thu Aug 1 10:32:42 2019 +0200\n\n powerpc/xive: Add a check for memory allocation failure\n \n The result of this kzalloc is not checked. Add a check and corresponding\n error handling code.\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Reviewed-by: Greg Kurz \u003cgroug@kaod.org\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/cc53462734dfeaf15b6bad0e626b483de18656b4.1564647619.git.christophe.jaillet@wanadoo.fr\n\ncommit b214a8f2ea964b5ff2cc4a1e8bd8c74f64741f20\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Thu Aug 1 10:32:31 2019 +0200\n\n powerpc/xive: Use GFP_KERNEL instead of GFP_ATOMIC in \u0027xive_irq_bitmap_add()\u0027\n \n There is no need to use GFP_ATOMIC here. GFP_KERNEL should be enough.\n GFP_KERNEL is also already used for another allocation just a few lines\n below.\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Reviewed-by: Cédric Le Goater \u003cclg@kaod.org\u003e\n Reviewed-by: Greg Kurz \u003cgroug@kaod.org\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/85d5d247ce753befd6aa63c473f7823de6520ccd.1564647619.git.christophe.jaillet@wanadoo.fr\n\ncommit 81e0da703fdba1ee126868bf8350592c79cdba13\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Thu Aug 15 14:30:09 2019 +0200\n\n staging: sm750fb: fix odd license text\n \n There is some \"confusing\" license text in some of the sm750fb driver\n files. After discussing it with the company, it turns out to have been\n a mistake and these lines can be safely removed. The files are all to\n be licensed under the GPLv2 license.\n \n Cc: Sudip Mukherjee \u003csudipm.mukherjee@gmail.com\u003e\n Cc: Teddy Wang \u003cteddy.wang@siliconmotion.com\u003e\n Cc: linux-fbdev@vger.kernel.org\n Cc: devel@driverdev.osuosl.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Link: https://lore.kernel.org/r/20190815123009.16499-1-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 62520e3361d2d9e2e41d3958dcc34faee79a3c1a\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 15 05:20:30 2019 +0100\n\n drm/i915: Move tasklet kicking to __i915_request_queue caller\n \n Since __i915_request_queue() may be called from hardirq (timer) context,\n we cannot use local_bh_disable/enable at the lower level. As we do want\n to kick the tasklet to speed up initial submission or preemption for\n normal client submission, lift it to the normal process context\n callpath.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190815042031.27750-1-chris@chris-wilson.co.uk\n\ncommit ff784a78af11712c7fa97c292be5390e45b50639\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Aug 14 18:37:05 2019 +0800\n\n staging: erofs: correct all misused ENOTSUPP\n \n As Chao pointed out [1], ENOTSUPP is used for NFS\n protocol only, we should use EOPNOTSUPP instead...\n \n [1] https://lore.kernel.org/lkml/108ee2f9-75dd-b8ab-8da7-b81c17bafbf6@huawei.com/\n \n Reported-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190814103705.60698-3-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 382329a9d8553a98418a6f6e3425f0c288837897\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Aug 14 18:37:04 2019 +0800\n\n staging: erofs: differentiate unsupported on-disk format\n \n For some specific fields, use EOPNOTSUPP instead of EIO\n for values which look sane but aren\u0027t supported right now.\n \n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190814103705.60698-2-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a6b9b1d5eae61a68085030e50d56265dec5baa94\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Aug 14 18:37:03 2019 +0800\n\n staging: erofs: introduce EFSCORRUPTED and more logs\n \n Previously, EROFS uses EIO to indicate that filesystem\n is corrupted as well. However, as Pavel said [1], other\n filesystems tend to use EUCLEAN(EFSCORRUPTED) instead,\n let\u0027s follow what others do right now.\n \n Also, add some more prints to the syslog.\n \n [1] https://lore.kernel.org/lkml/20190813114821.GB11559@amd/\n \n Suggested-by: Pavel Machek \u003cpavel@denx.de\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190814103705.60698-1-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 34b5560db40d2941cfbe82eca1641353d5aed1a9\nAuthor: Mark Rutland \u003cmark.rutland@arm.com\u003e\nDate: Wed Aug 14 15:31:57 2019 +0100\n\n kasan/arm64: fix CONFIG_KASAN_SW_TAGS \u0026\u0026 KASAN_INLINE\n \n The generic Makefile.kasan propagates CONFIG_KASAN_SHADOW_OFFSET into\n KASAN_SHADOW_OFFSET, but only does so for CONFIG_KASAN_GENERIC.\n \n Since commit:\n \n 6bd1d0be0e97936d (\"arm64: kasan: Switch to using KASAN_SHADOW_OFFSET\")\n \n ... arm64 defines CONFIG_KASAN_SHADOW_OFFSET in Kconfig rather than\n defining KASAN_SHADOW_OFFSET in a Makefile. Thus, if\n CONFIG_KASAN_SW_TAGS \u0026\u0026 KASAN_INLINE are selected, we get build time\n splats due to KASAN_SHADOW_OFFSET not being set:\n \n | [mark@lakrids:~/src/linux]% usellvm 8.0.1 usekorg 8.1.0 make ARCH\u003darm64 CROSS_COMPILE\u003daarch64-linux- CC\u003dclang\n | scripts/kconfig/conf --syncconfig Kconfig\n | CC scripts/mod/empty.o\n | clang (LLVM option parsing): for the -hwasan-mapping-offset option: \u0027\u0027 value invalid for uint argument!\n | scripts/Makefile.build:273: recipe for target \u0027scripts/mod/empty.o\u0027 failed\n | make[1]: *** [scripts/mod/empty.o] Error 1\n | Makefile:1123: recipe for target \u0027prepare0\u0027 failed\n | make: *** [prepare0] Error 2\n \n Let\u0027s fix this by always propagating CONFIG_KASAN_SHADOW_OFFSET into\n KASAN_SHADOW_OFFSET if CONFIG_KASAN is selected, moving the existing\n common definition of +CFLAGS_KASAN_NOSANITIZE to the top of\n Makefile.kasan.\n \n Cc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Acked-by: Andrey Ryabinin \u003caryabinin@virtuozzo.com\u003e\n Tested-by Steve Capper \u003csteve.capper@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit d247ceda6ae65968bda1bf7a26b54a257a414c7a\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Thu Aug 15 11:53:14 2019 +0100\n\n staging: comedi: usbduxsigma: remove redundant assignment to variable fx2delay\n \n Variable fx2delay is being initialized with a value that is never read\n and fx2delay is being re-assigned a little later on. The assignment is\n redundant and hence can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190815105314.5756-1-colin.king@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7d5e9e350e60526779e293f32d8c32c4d5090b8b\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Thu Aug 15 09:40:34 2019 +0100\n\n staging: rtl8192u: fix up indentation of a statement\n \n There is a statement that is indented one level too deeply, remove\n the extraneous tab.\n \n Addresses-Coverity: (\"Identation does not match nesting level\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190815084034.13885-1-colin.king@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 1548547e1d82532db70a75b9635427a727cd5590\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 15 11:19:24 2019 +0530\n\n staging: gasket: apex: Make structure apex_desc constant\n \n Static structure apex_desc, of type gasket_driver_desc, is used only as\n an argument to the functions gasket_register_device() and\n gasket_unregister_device(). In the definitions of both these functions,\n their parameter is declared as const. Hence make apex_desc itself\n constant to protect it from modification.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190815054924.643-1-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a72553505f11d507ef8d86ea9cb94da68011f417\nAuthor: Pratik Shinde \u003cpratikshinde320@gmail.com\u003e\nDate: Wed Aug 14 02:08:40 2019 +0530\n\n staging: erofs: removing an extra call to iloc() in fill_inode()\n \n in fill_inode() we call iloc() twice.Avoiding the extra call by\n storing the result.\n \n Signed-off-by: Pratik Shinde \u003cpratikshinde320@gmail.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Cc: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190813203840.13782-1-pratikshinde320@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 845f7f7ecb1f67c3197f0a7c86207042736abf70\nAuthor: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\nDate: Thu Aug 15 12:49:29 2019 +0300\n\n drm/i915/icl: Add gen11 specific render breadcrumbs\n \n Flush according to what gen11 expects when writing\n breadcrumbs. As only the seqnowrite + flush differs\n between engine and gens, enclose the footer to\n helper.\n \n v2: avoid problem of sane local naming by not using them\n \n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190815094929.358-1-mika.kuoppala@linux.intel.com\n\ncommit 8a8b540a6dd2953fdd6746c18da89c8e0dcf1767\nAuthor: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\nDate: Thu Aug 15 11:30:54 2019 +0300\n\n drm/i915/icl: Add command cache invalidate\n \n On the set of invalidations, we need to add command\n cache invalidate as a new domain.\n \n Signed-off-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190815083055.14132-2-mika.kuoppala@linux.intel.com\n\ncommit cfba6bd8b07401de46dede31be33b98a15a4dd12\nAuthor: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\nDate: Thu Aug 15 11:30:53 2019 +0300\n\n drm/i915/icl: Implement gen11 flush including tile cache\n \n Add tile cache flushing for gen11. To relive us from the\n burden of previous obsolete workarounds, make a dedicated\n flush/invalidate callback for gen11.\n \n To fortify an independent single flush, do post\n sync op as there are indications that without it\n we don\u0027t flush everything. This should also make this\n callback more readily usable in tgl (see l3 fabric flush).\n \n v2: whitespacing\n \n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190815083055.14132-1-mika.kuoppala@linux.intel.com\n\ncommit 6f6333ba501e2bf4d040c30c618e9b313e114761\nAuthor: Matthew Auld \u003cmatthew.auld@intel.com\u003e\nDate: Thu Aug 15 11:32:10 2019 +0100\n\n drm/i915/selftest/buddy: fixup igt_buddy_alloc_range\n \n Dan reported the following static checker warning:\n \n drivers/gpu/drm/i915/selftests/i915_buddy.c:670 igt_buddy_alloc_range()\n error: we previously assumed \u0027block\u0027 could be null (see line 665)\n \n Reported-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Signed-off-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Cc: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190815103210.11802-1-matthew.auld@intel.com\n\ncommit 88f8065ca73ff8d374fe640e651bc3d37ddc1954\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 15 10:36:04 2019 +0100\n\n drm/i915: Convert a few more bland dmesg info to be device specific\n \n Looking around the GT initialisation, we have a few log messages we\n think are interesting enough present to the user (such as the amount of L4\n cache) and a few to inform them of the result of actions or conflicting\n HW restrictions (i.e. quirks). These are device specific messages, so\n use the dev family of printk.\n \n v2: shave off a few bytes of .rodata!\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190815093604.3618-1-chris@chris-wilson.co.uk\n\ncommit d225bb8d8a897d35c7beedcaba6caf57b3a4d292\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 13 11:01:41 2019 +0200\n\n arm64: unexport set_memory_x and set_memory_nx\n \n No module currently messed with clearing or setting the execute\n permission of kernel memory, and none really should.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 198429631a85622da1d08d360ef02cfb84c95919\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Mon Aug 12 01:59:12 2019 +0300\n\n crypto: arm64/aegis128 - implement plain NEON version\n \n Provide a version of the core AES transform to the aegis128 SIMD\n code that does not rely on the special AES instructions, but uses\n plain NEON instructions instead. This allows the SIMD version of\n the aegis128 driver to be used on arm64 systems that do not\n implement those instructions (which are not mandatory in the\n architecture), such as the Raspberry Pi 3.\n \n Since GCC makes a mess of this when using the tbl/tbx intrinsics\n to perform the sbox substitution, preload the Sbox into v16..v31\n in this case and use inline asm to emit the tbl/tbx instructions.\n Clang does not support this approach, nor does it require it, since\n it does a much better job at code generation, so there we use the\n intrinsics as usual.\n \n Cc: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Acked-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit a4397635afea5d127548d64e0055ed471ef2d5be\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Mon Aug 12 01:59:11 2019 +0300\n\n crypto: aegis128 - provide a SIMD implementation based on NEON intrinsics\n \n Provide an accelerated implementation of aegis128 by wiring up the\n SIMD hooks in the generic driver to an implementation based on NEON\n intrinsics, which can be compiled to both ARM and arm64 code.\n \n This results in a performance of 2.2 cycles per byte on Cortex-A53,\n which is a performance increase of ~11x compared to the generic\n code.\n \n Reviewed-by: Ondrej Mosnacek \u003comosnace@redhat.com\u003e\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit cf3d41adcc3595e7ccfbc9359a5daf39ee07aa8b\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Mon Aug 12 01:59:10 2019 +0300\n\n crypto: aegis128 - add support for SIMD acceleration\n \n Add some plumbing to allow the AEGIS128 code to be built with SIMD\n routines for acceleration.\n \n Reviewed-by: Ondrej Mosnacek \u003comosnace@redhat.com\u003e\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 8083b1bf8163e7ae7d8c90f221106d96450b8aa8\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Fri Aug 9 20:14:57 2019 +0300\n\n crypto: xts - add support for ciphertext stealing\n \n Add support for the missing ciphertext stealing part of the XTS-AES\n specification, which permits inputs of any size \u003e\u003d the block size.\n \n Cc: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Cc: Ondrej Mosnacek \u003comosnace@redhat.com\u003e\n Tested-by: Milan Broz \u003cgmazyland@gmail.com\u003e\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit a62084d299d950f2ad0649caf9a9b9a431346996\nAuthor: Pascal van Leeuwen \u003cpascalvanl@gmail.com\u003e\nDate: Fri Aug 9 17:51:07 2019 +0200\n\n crypto: aead - Do not allow authsize\u003d0 if auth. alg has digestsize\u003e0\n \n Return -EINVAL on an attempt to set the authsize to 0 with an auth.\n algorithm with a non-zero digestsize (i.e. anything but digest_null)\n as authenticating the data and then throwing away the result does not\n make any sense at all.\n \n The digestsize zero exception is for use with digest_null for testing\n purposes only.\n \n Signed-off-by: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 440dc9aab5286aec46b6bdf061c6eec4fd53887f\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 9 16:49:05 2019 +0800\n\n crypto: streebog - remove two unused variables\n \n crypto/streebog_generic.c:162:17: warning:\n Pi defined but not used [-Wunused-const-variable\u003d]\n crypto/streebog_generic.c:151:17: warning:\n Tau defined but not used [-Wunused-const-variable\u003d]\n \n They are never used, so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Reviewed-by: Vitaly Chikunov \u003cvt@altlinux.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit c2ccfa9ed7fbe0924782363c56b462f23e9c91ab\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 9 16:29:19 2019 +0800\n\n crypto: aes-generic - remove unused variable \u0027rco_tab\u0027\n \n crypto/aes_generic.c:64:18: warning:\n rco_tab defined but not used [-Wunused-const-variable\u003d]\n \n It is never used, so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Acked-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 5d2ded2823c30df6410a3d8934cd1df6fe0aa284\nAuthor: Jonathan Neuschäfer \u003cj.neuschaefer@gmx.net\u003e\nDate: Thu Aug 8 18:30:11 2019 +0200\n\n Documentation: crypto: crypto_engine: Fix Sphinx warning\n \n This fixes the following Sphinx warning:\n \n Documentation/crypto/crypto_engine.rst:2:\n WARNING: Explicit markup ends without a blank line; unexpected unindent.\n \n Signed-off-by: Jonathan Neuschäfer \u003cj.neuschaefer@gmx.net\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 5f05cdca2727dd54a3ec84cb291663af6fb68672\nAuthor: Phani Kiran Hemadri \u003cphemadri@marvell.com\u003e\nDate: Thu Aug 8 12:17:39 2019 +0000\n\n crypto: cavium/nitrox - Configure asymmetric queue manager Hardware unit\n \n This patch configures and initializes CNN55XX device AQM hardware unit.\n \n Signed-off-by: Phani Kiran Hemadri \u003cphemadri@marvell.com\u003e\n Reviewed-by: Srikanth Jampala \u003cjsrikanth@marvell.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit b31c17c852736846c917bbd7f1263ea7258fbdce\nAuthor: Phani Kiran Hemadri \u003cphemadri@marvell.com\u003e\nDate: Thu Aug 8 12:17:37 2019 +0000\n\n crypto: cavium/nitrox - Allocate asymmetric crypto command queues\n \n This patch adds support to allocate CNN55XX device AQMQ command queues\n required for submitting asymmetric crypto requests.\n \n Signed-off-by: Phani Kiran Hemadri \u003cphemadri@marvell.com\u003e\n Reviewed-by: Srikanth Jampala \u003cjsrikanth@marvell.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 43b970fa82131ab4f42789a06bfd68dbc6f2a829\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Thu Aug 8 16:00:22 2019 +0800\n\n crypto: cryptd - Use refcount_t for refcount\n \n Reference counters are preferred to use refcount_t instead of\n atomic_t.\n This is because the implementation of refcount_t can prevent\n overflows and detect possible use-after-free.\n So convert atomic_t ref counters to refcount_t.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 03a3bb7ae63150230c5de645dc95e673ebf17e1a\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Mon Aug 5 16:32:41 2019 -0700\n\n hwrng: core - Freeze khwrng thread during suspend\n \n The hwrng_fill() function can run while devices are suspending and\n resuming. If the hwrng is behind a bus such as i2c or SPI and that bus\n is suspended, the hwrng may hang the bus while attempting to add some\n randomness. It\u0027s been observed on ChromeOS devices with suspend-to-idle\n (s2idle) and an i2c based hwrng that this kthread may run and ask the\n hwrng device for randomness before the i2c bus has been resumed.\n \n Let\u0027s make this kthread freezable so that we don\u0027t try to touch the\n hwrng during suspend/resume. This ensures that we can\u0027t cause the hwrng\n backing driver to get into a bad state because the device is guaranteed\n to be resumed before the hwrng kthread is thawed.\n \n Cc: Andrey Pronin \u003capronin@chromium.org\u003e\n Cc: Duncan Laurie \u003cdlaurie@chromium.org\u003e\n Cc: Jason Gunthorpe \u003cjgg@ziepe.ca\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Guenter Roeck \u003cgroeck@chromium.org\u003e\n Cc: Alexander Steffen \u003cAlexander.Steffen@infineon.com\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit c9fbcf6815ce24c8da828c2a70cd6a16ce242a1b\nAuthor: Horia Geantă \u003choria.geanta@nxp.com\u003e\nDate: Mon Aug 5 15:49:55 2019 +0300\n\n crypto: caam/qi - execute library only on DPAA 1.x\n \n In the process of turning caam/qi into a library, the check of\n MCFGR[QI] bit has been inadvertently dropped.\n Fix the condition for DPAA 1.x QI detection, which should be:\n MCFGR[QI] \u0026\u0026 !MCFGR[DPAA2]\n \n A check in the library exit point is currently not needed,\n since the list of registered algorithms is empty.\n \n While here, silence the library initialization abort - since jr.c\n calls it unconditionally.\n \n Fixes: 1b46c90c8e00 (\"crypto: caam - convert top level drivers to libraries\")\n Signed-off-by: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 1ed2002f891dc1082eb110eaafccfde96d327aa3\nAuthor: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\nDate: Wed Aug 14 17:28:39 2019 +0800\n\n crypto: hisilicon - fix error handle in hisi_zip_create_req_q\n \n Directly return error in the first loop in hisi_zip_create_req_q.\n \n Fixes: 62c455ca853e (\"crypto: hisilicon - add HiSilicon ZIP accelerator support\")\n Signed-off-by: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 902f0babf5457578f6bfb9b6ac952c55c4482cb7\nAuthor: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\nDate: Wed Aug 14 17:28:38 2019 +0800\n\n crypto: hisilicon - add missing single_release\n \n Fix to add missing single_release in qm_regs_fops.\n \n Fixes: 263c9959c937 (\"crypto: hisilicon - add queue management driver for HiSilicon QM module\")\n Signed-off-by: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 5c0861989cc197c29433915774f805299a497977\nAuthor: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\nDate: Wed Aug 14 17:28:37 2019 +0800\n\n crypto: hisilicon - init curr_sgl_dma to fix compile warning\n \n Just init curr_sgl_dma \u003d 0 to avoid compile warning.\n \n Fixes: dfed0098ab91 (\"crypto: hisilicon - add hardware SGL support\")\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit db01e4818bbdab626c78747dd8f494d586ac968d\nAuthor: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\nDate: Wed Aug 14 17:28:36 2019 +0800\n\n crypto: hisilicon - add dependency for CRYPTO_DEV_HISI_ZIP\n \n Add ARM64/PCI/PCI_MSI dependency for CRYPTO_DEV_HISI_ZIP.\n \n Fixes: 62c455ca853e (\"crypto: hisilicon - add HiSilicon ZIP accelerator support\")\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 00ae05db583a7b5d5575bb601158139e25af3b02\nAuthor: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\nDate: Wed Aug 14 17:28:35 2019 +0800\n\n crypto: hisilicon - fix kbuild warnings\n \n Fix to use proper type of argument for dma_addr_t and size_t.\n \n Fixes: 263c9959c937 (\"crypto: hisilicon - add queue management driver for HiSilicon QM module\")\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 5336c17928cc464845ff765ce45b368c22f848e0\nAuthor: Guo Ren \u003cren_guo@c-sky.com\u003e\nDate: Thu Aug 15 16:24:56 2019 +0800\n\n csky: Fixup ioremap function losing\n \n Implement the following apis to meet usage in different scenarios.\n \n - ioremap (NonCache + StrongOrder)\n - ioremap_nocache (NonCache + StrongOrder)\n - ioremap_wc (NonCache + WeakOrder )\n - ioremap_cache ( Cache + WeakOrder )\n \n Also change flag VM_ALLOC to VM_IOREMAP in get_vm_area_caller.\n \n Signed-off-by: Guo Ren \u003cren_guo@c-sky.com\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: Christoph Hellwig \u003chch@infradead.org\u003e\n\ncommit 31d5d15dfc3418a57cfab419a353d8dc5f5698b5\nAuthor: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\nDate: Wed Aug 14 06:59:16 2019 -0300\n\n media: MAINTAINERS: Add entry for the ov5670 driver\n \n The ov5670 driver didn\u0027t get a MAINTAINERS entry when it was merged. Add\n one now.\n \n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 0486a18ce82bd00d69ddc0fab8faa4b80df2117b\nAuthor: Sébastien Szymanski \u003csebastien.szymanski@armadeus.com\u003e\nDate: Wed Jul 31 13:33:30 2019 -0300\n\n media: imx7-media-csi: add i.MX6UL support\n \n i.MX7 and i.MX6UL/L have the same CSI controller. So add i.MX6UL/L support\n to imx7-media-csi driver.\n \n Signed-off-by: Sébastien Szymanski \u003csebastien.szymanski@armadeus.com\u003e\n Reviewed-by: Rui Miguel Silva \u003crmfrfs@gmail.com\u003e\n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 0d90467348d82bce6dacc233ed4562e4288c06c2\nAuthor: Sébastien Szymanski \u003csebastien.szymanski@armadeus.com\u003e\nDate: Wed Jul 31 13:33:58 2019 -0300\n\n media: dt-bindings: imx7-csi: add i.MX6UL/L support\n \n Document \"fsl,imx6ul-csi\" entry.\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Sébastien Szymanski \u003csebastien.szymanski@armadeus.com\u003e\n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 0d5078c7172c46db6c58718d817b9fcf769554b4\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Thu Jun 27 19:29:12 2019 -0300\n\n media: imx: mipi csi-2: Don\u0027t fail if initial state times-out\n \n Not all sensors will be able to guarantee a proper initial state.\n This may be either because the driver is not properly written,\n or (probably unlikely) because the hardware won\u0027t support it.\n \n While the right solution in the former case is to fix the sensor\n driver, the real world not always allows right solutions, due to lack\n of available documentation and support on these sensors.\n \n Let\u0027s relax this requirement, and allow the driver to support stream start,\n even if the sensor initial sequence wasn\u0027t the expected.\n \n Also improve the warning message to better explain the problem and provide\n a hint that the sensor driver needs to be fixed.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Reviewed-by: Steve Longerbeam \u003cslongerbeam@gmail.com\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 9bf656e1a6b0709122ec3d71131171c92cf4cdae\nAuthor: Steve Longerbeam \u003cslongerbeam@gmail.com\u003e\nDate: Tue Aug 13 19:36:44 2019 -0300\n\n media: docs-rst: Clarify duration of LP-11 mode\n \n Add a sentence that makes it more clear when the CSI-2 transmitter\n must, if possible, exit LP-11 mode. That is, maintain LP-11 mode\n until stream on, at which point the transmitter activates the clock\n lane and transition to HS mode.\n \n Signed-off-by: Steve Longerbeam \u003cslongerbeam@gmail.com\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 24a90c926219505efd9b9ac773bb015e32d5519c\nAuthor: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\nDate: Tue Aug 13 08:31:28 2019 -0300\n\n media: v4l: Documentation: Raw Bayer formats are not RGB formats\n \n The raw Bayer formats have been listed under the label of RGB formats but\n in fact they\u0027re quite different. The latter are readily usable as such\n whereas the former require quite bit of image processing before useful.\n \n Split them into RGB and raw Bayer formats.\n \n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Acked-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 50435d8dbc0560c1b7ee2381edddb7e84b96d464\nAuthor: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\nDate: Tue Aug 13 08:26:43 2019 -0300\n\n media: v4l: Documentation: Serial busses use parallel mbus codes\n \n When the support for serial busses was introduced in V4L2, it was decided\n to use the existing parallel bus media bus pixel codes to describe them.\n While this was a practical choice at the time, it necessitates choosing\n which one of the many parallel mbus pixel codes to use, for on the serial\n busses these formats are effectively all equivalent.\n \n The practice has always been to use the pixel code that describes a bus\n that transfers a single sample per clock. Document this.\n \n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Acked-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 092e8eb90a7dc7dd210cd4e2ea36075d0a7f96af\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Thu Aug 8 08:05:40 2019 -0300\n\n media: i2c: ov5645: Fix power sequence\n \n This is mostly a port of Jacopo\u0027s fix:\n \n commit aa4bb8b8838ffcc776a79f49a4d7476b82405349\n Author: Jacopo Mondi \u003cjacopo@jmondi.org\u003e\n Date: Fri Jul 6 05:51:52 2018 -0400\n \n media: ov5640: Re-work MIPI startup sequence\n \n In the OV5645 case, the changes are:\n \n - At set_power(1) time power up MIPI Tx/Rx and set data and clock lanes in\n LP11 during \u0027sleep\u0027 and \u0027idle\u0027 with MIPI clock in non-continuous mode.\n - At set_power(0) time power down MIPI Tx/Rx (in addition to the current\n power down of regulators and clock gating).\n - At s_stream time enable/disable the MIPI interface output.\n \n With this commit the sensor is able to enter LP-11 mode during power up,\n as expected by some CSI-2 controllers.\n \n Many thanks to Fabio Estevam for his help debugging this issue.\n \n Tested-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Reviewed-by: Jacopo Mondi \u003cjacopo@jmondi.org\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 7ef57be07ac146e70535747797ef4aee0f06e9f9\nAuthor: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\nDate: Wed Aug 7 11:21:27 2019 -0300\n\n media: omap3isp: Don\u0027t set streaming state on random subdevs\n \n The streaming state should be set to the first upstream sub-device only,\n not everywhere, for a sub-device driver itself knows how to best control\n the streaming state of its own upstream sub-devices.\n \n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit e9eb103f027725053a4b02f93d7f2858b56747ce\nAuthor: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\nDate: Wed Aug 7 11:19:00 2019 -0300\n\n media: omap3isp: Set device on omap3isp subdevs\n \n The omap3isp driver registered subdevs without the dev field being set. Do\n that now.\n \n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 2f1fc5efe4a59117001544830c183da1765af86f\nAuthor: Luca Weiss \u003cluca@z3ntu.xyz\u003e\nDate: Fri Jun 21 14:20:25 2019 -0300\n\n media: ov5640: Add support for flash and lens devices\n \n Parse async sub-devices related to the sensor by switching the async\n sub-device registration function.\n \n Signed-off-by: Luca Weiss \u003cluca@z3ntu.xyz\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit fa4bb7d3acbfb7926d1b46ce6568922bfc99ffeb\nAuthor: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\nDate: Wed Aug 7 05:34:43 2019 -0300\n\n media: ov8856: Check reading clock frequency succeeded\n \n Instead of blindly trusting getting the clock frequency succeeded end then\n testing it against a pre-defined value, verify reading the value\n succeeded.\n \n Fixes: 879347f0c258 (\"media: ov8856: Add support for OV8856 sensor\")\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit bf27502b1f3bf8095bf81736e506d354a2ce9ec4\nAuthor: Shawn Tu \u003cshawnx.tu@intel.com\u003e\nDate: Wed Aug 7 07:07:56 2019 -0300\n\n media: ov5675: Add support for OV5675 sensor\n \n Add a V4L2 sub-device driver for Omnivision ov5675 image sensor.\n This is a camera sensor using the I2C bus for control and the\n CSI-2 bus for data.\n \n This driver supports following features:\n - manual exposure and analog/digital gain control support\n - vblank/hblank control support\n - test pattern support\n - media controller support\n - runtime PM support\n - support following resolutions:\n + 2592x1944 at 30FPS\n + 1296x972 at 30FPS\n \n [Sakari Ailus: Wrapped a few long lines.]\n [mchehab+samsung@kernel.org: fix a checkpatch warning]\n \n Signed-off-by: Shawn Tu \u003cshawnx.tu@intel.com\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 075af61c19cdc32b94fd697c610c9a07fa21866e\nAuthor: Stefan Agner \u003cstefan@agner.ch\u003e\nDate: Fri Jul 26 16:40:07 2019 +0200\n\n PCI: imx6: Limit DBI register length\n \n Define the length of the DBI registers and limit config space to its\n length. This makes sure that the kernel does not access registers\n beyond that point, avoiding the following abort on a i.MX 6Quad:\n \n # cat /sys/devices/soc0/soc/1ffc000.pcie/pci0000\\:00/0000\\:00\\:00.0/config\n [ 100.021433] Unhandled fault: imprecise external abort (0x1406) at 0xb6ea7000\n ...\n [ 100.056423] PC is at dw_pcie_read+0x50/0x84\n [ 100.060790] LR is at dw_pcie_rd_own_conf+0x44/0x48\n ...\n \n Signed-off-by: Stefan Agner \u003cstefan@agner.ch\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Reviewed-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n\ncommit d55c5f28afafb6b1f0a6978916b23338b383faab\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Wed Jun 12 13:51:37 2019 +0100\n\n arm64: smp: disable hotplug on trusted OS resident CPU\n \n The trusted OS may reject CPU_OFF calls to its resident CPU, so we must\n avoid issuing those. We never migrate a Trusted OS and we already take\n care to prevent CPU_OFF PSCI call. However, this is not reflected\n explicitly to the userspace. Any user can attempt to hotplug trusted OS\n resident CPU. The entire motion of going through the various state\n transitions in the CPU hotplug state machine gets executed and the\n PSCI layer finally refuses to make CPU_OFF call.\n \n This results is unnecessary unwinding of CPU hotplug state machine in\n the kernel. Instead we can mark the trusted OS resident CPU as not\n available for hotplug, so that the user attempt or request to do the\n same will get immediately rejected.\n \n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Cc: Will Deacon \u003cwill.deacon@arm.com\u003e\n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 088880ddc0b20086b71bb87b805fb63ff07c35f2\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Fri Aug 2 14:27:33 2019 +0200\n\n drm/etnaviv: implement softpin\n \n With softpin we allow the userspace to take control over the GPU virtual\n address space. The new capability is relected by a bump of the minor DRM\n version. There are a few restrictions for userspace to take into\n account:\n \n 1. The kernel reserves a bit of the address space to implement zero page\n faulting and mapping of the kernel internal ring buffer. Userspace can\n query the kernel for the first usable GPU VM address via\n ETNAVIV_PARAM_SOFTPIN_START_ADDR.\n \n 2. We only allow softpin on GPUs, which implement proper process\n separation via PPAS. If softpin is not available the softpin start\n address will be set to ~0.\n \n 3. Softpin is all or nothing. A submit using softpin must not use any\n address fixups via relocs.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Reviewed-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n\ncommit 17eae23b08207a8954f77ca74caeaeb72a04da67\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Fri Aug 2 14:19:38 2019 +0200\n\n drm/etnaviv: allow to request specific virtual address for gem mapping\n \n Allow the mapping code to request a specific virtual address for the gem\n mapping. If the virtual address is zero we fall back to the old mode of\n allocating a virtual address for the mapping.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Reviewed-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n\ncommit edb5ff07ef67962b81ea7abb716a74ef0a491ca1\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Fri Aug 2 14:24:06 2019 +0200\n\n drm/etnaviv: skip command stream validation on PPAS capable GPUs\n \n With per-process address spaces in place, a rogue process submitting\n bogus command streams can only hurt itself. There is no need to\n validate the command stream before execution anymore.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Reviewed-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n\ncommit f58ba5e3f6863ea4486952698898848a6db726c2\nAuthor: Randy Dunlap \u003crdunlap@infradead.org\u003e\nDate: Fri Jul 12 08:53:19 2019 -0700\n\n PCI: pci-hyperv: Fix build errors on non-SYSFS config\n \n Fix build errors when building almost-allmodconfig but with SYSFS\n not set (not enabled). Fixes these build errors:\n \n ERROR: \"pci_destroy_slot\" [drivers/pci/controller/pci-hyperv.ko] undefined!\n ERROR: \"pci_create_slot\" [drivers/pci/controller/pci-hyperv.ko] undefined!\n \n drivers/pci/slot.o is only built when SYSFS is enabled, so\n pci-hyperv.o has an implicit dependency on SYSFS.\n Make that explicit.\n \n Also, depending on X86 \u0026\u0026 X86_64 is not needed, so just change that\n to depend on X86_64.\n \n Fixes: a15f2c08c708 (\"PCI: hv: support reporting serial number as slot information\")\n Signed-off-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Reviewed-by: Haiyang Zhang \u003chaiyangz@microsoft.com\u003e\n Cc: Matthew Wilcox \u003cwilly@infradead.org\u003e\n Cc: Jake Oshins \u003cjakeo@microsoft.com\u003e\n Cc: \"K. Y. Srinivasan\" \u003ckys@microsoft.com\u003e\n Cc: Haiyang Zhang \u003chaiyangz@microsoft.com\u003e\n Cc: Stephen Hemminger \u003csthemmin@microsoft.com\u003e\n Cc: Stephen Hemminger \u003cstephen@networkplumber.org\u003e\n Cc: Sasha Levin \u003csashal@kernel.org\u003e\n Cc: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Cc: linux-pci@vger.kernel.org\n Cc: linux-hyperv@vger.kernel.org\n Cc: Dexuan Cui \u003cdecui@microsoft.com\u003e\n\ncommit 17e4660ae3d7e14120f0b355d3d1995cd10a3e43\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Fri Jul 5 19:17:27 2019 +0200\n\n drm/etnaviv: implement per-process address spaces on MMUv2\n \n This builds on top of the MMU contexts introduced earlier. Instead of having\n one context per GPU core, each GPU client receives its own context.\n \n On MMUv1 this still means a single shared pagetable set is used by all\n clients, but on MMUv2 there is now a distinct set of pagetables for each\n client. As the command fetch is also translated via the MMU on MMUv2 the\n kernel command ringbuffer is mapped into each of the client pagetables.\n \n As the MMU context switch is a bit of a heavy operation, due to the needed\n cache and TLB flushing, this patch implements a lazy way of switching the\n MMU context. The kernel does not have its own MMU context, but reuses the\n last client context for all of its operations. This has some visible impact,\n as the GPU can now only be started once a client has submitted some work and\n we got the client MMU context assigned. Also the MMU context has a different\n lifetime than the general client context, as the GPU might still execute the\n kernel command buffer in the context of a client even after the client has\n completed all GPU work and has been terminated. Only when the GPU is runtime\n suspended or switches to another clients MMU context is the old context\n freed up.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Reviewed-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n\ncommit e6364d70cfa3205233c131ae65f1aade8c8a050f\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Fri Jul 5 19:17:26 2019 +0200\n\n drm/etnaviv: provide MMU context to etnaviv_gem_mapping_get\n \n In preparation to having a context per process, etnaviv_gem_mapping_get\n should not use the current GPU context, but needs to be told which\n context to use.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Reviewed-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n\ncommit d80d842a47fbd029b18f8cff12535f2d1ece3f1e\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Fri Jul 5 19:17:25 2019 +0200\n\n drm/etnaviv: split out starting of FE idle loop\n \n Move buffer setup and starting of the FE loop in the kernel ringbuffer\n into a separate function. This is a preparation to start the FE later\n in the submit process.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Reviewed-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n\ncommit 27b67278e007b5475bc14918794bf73cf297a026\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Fri Jul 5 19:17:24 2019 +0200\n\n drm/etnaviv: rework MMU handling\n \n This reworks the MMU handling to make it possible to have multiple MMU contexts.\n A context is basically one instance of GPU page tables. Currently we have one\n set of page tables per GPU, which isn\u0027t all that clever, as it has the\n following two consequences:\n \n 1. All GPU clients (aka processes) are sharing the same pagetables, which means\n there is no isolation between clients, but only between GPU assigned memory\n spaces and the rest of the system. Better than nothing, but also not great.\n \n 2. Clients operating on the same set of buffers with different etnaviv GPU\n cores, e.g. a workload using both the 2D and 3D GPU, need to map the used\n buffers into the pagetable sets of each used GPU.\n \n This patch reworks all the MMU handling to introduce the abstraction of the\n MMU context. A context can be shared across different GPU cores, as long as\n they have compatible MMU implementations, which is the case for all systems\n with Vivante GPUs seen in the wild.\n \n As MMUv1 is not able to change pagetables on the fly, without a\n \"stop the world\" operation, which stops GPU, changes pagetables via CPU\n interaction, restarts GPU, the implementation introduces a shared context on\n MMUv1, which is returned whenever there is a request for a new context.\n \n This patch assigns a MMU context to each GPU, so on MMUv2 systems there is\n still one set of pagetables per GPU, but due to the shared context MMUv1\n systems see a change in behavior as now a single pagetable set is used\n across all GPU cores.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Reviewed-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n\ncommit 4900dda90af2cb13bc1d4c12ce94b98acc8fe64e\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Fri Jul 5 19:17:23 2019 +0200\n\n drm/etnaviv: replace MMU flush marker with flush sequence\n \n If a MMU is shared between multiple GPUs, all of them need to flush their\n TLBs, so a single marker that gets reset on the first flush won\u0027t do.\n Replace the flush marker with a sequence number, so that it\u0027s possible to\n check if the TLB is in sync with the current page table state for each GPU.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Reviewed-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n\ncommit bffe5db81ae63ec75f5e09d1c97de42c37cb181e\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Fri Jul 5 19:17:22 2019 +0200\n\n drm/etnaviv: share a single cmdbuf suballoc region across all GPUs\n \n There is no need for each GPU to have it\u0027s own cmdbuf suballocation\n region. Only allocate a single one for the the etnaviv virtual device\n and share it across all GPUs.\n \n As the suballoc space is now potentially shared by more hardware jobs\n running in parallel, double its size to 512KB to avoid contention.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Reviewed-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n\ncommit db82a0435b8be32d544bbed91c43c2f21b5f4ea7\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Fri Jul 5 19:17:21 2019 +0200\n\n drm/etnaviv: split out cmdbuf mapping into address space\n \n This allows to decouple the cmdbuf suballocator create and mapping\n the region into the GPU address space. Allowing multiple AS to share\n a single cmdbuf suballoc.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Reviewed-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n\ncommit db41fe7d2f75c946a0b138aa789ca7cf6cd8afc1\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Fri Jul 5 19:17:20 2019 +0200\n\n drm/etnaviv: simplify unbind checks\n \n Remember if the GPU has been sucessfully initialized. Only in that case\n do we need to clean up various structures in the unbind path. If the\n GPU hasn\u0027t been sucessfully initialized all the cleanups should happen\n in the failure paths of the init function.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Reviewed-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n\ncommit 3001eeb7f216f404d5c459876131d3d48f3c12fc\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Fri Aug 9 13:59:25 2019 +0200\n\n drm/etnaviv: pass mmu pointer to etnaviv_core_dump_mmu\n \n This function does only need the mmu part part of the gpu struct.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit 9a1fdae58760169bfe283489f1cb9a54a2b66d2d\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Fri Aug 9 13:58:02 2019 +0200\n\n drm/etnaviv: dump only failing submit\n \n Due to the tracking provided by the scheduler we know exactly which\n submit is failing. Only dump this single submit and the required\n auxiliary information. This cuts down the size of the devcoredumps\n by only including relevant information.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit dcc31687b8b9bf92aa1da1afbd7a91b1910f5512\nAuthor: Mao Wenan \u003cmaowenan@huawei.com\u003e\nDate: Thu Aug 15 15:48:48 2019 +0800\n\n interconnect: qcom: remove COMPILE_TEST from CONFIG_INTERCONNECT_QCOM_QCS404\n \n There is one compilation error when CONFIG_INTERCONNECT_QCOM_QCS404\u003dy and\n CONFIG_INTERCONNECT_QCOM_SMD_RPM\u003dy, as well as CONFIG_COMPILE_TEST\u003dy,\n but CONFIG_QCOM_SMD_RPM is not set, logs as below:\n \n drivers/interconnect/qcom/smd-rpm.o: In function `qcom_icc_rpm_smd_send\u0027:\n smd-rpm.c:(.text+0xe4): undefined reference to `qcom_rpm_smd_write\u0027\n Makefile:1071: recipe for target \u0027vmlinux\u0027 failed\n make: *** [vmlinux] Error 1\n \n This is because\n INTERCONNECT_QCOM_QCS404 depends on QCOM_SMD_RPM || COMPILE_TEST.\n Here CONFIG_COMPILE_TEST\u003dy, so CONFIG_INTERCONNECT_QCOM_SMD_RPM\n is selected. If CONFIG_QCOM_SMD_RPM is not set, then\n qcom_rpm_smd_write() is not defined, and compilation error happen.\n Fix this by removing COMPILE_TEST from CONFIG_INTERCONNECT_QCOM_QCS404.\n \n Fixes: 5e4e6c4d3ae0 (\"interconnect: qcom: Add QCS404 interconnect provider driver\")\n Signed-off-by: Mao Wenan \u003cmaowenan@huawei.com\u003e\n Signed-off-by: Georgi Djakov \u003cgeorgi.djakov@linaro.org\u003e\n\ncommit 8512ee31926296fd7d65c1b3c97bb20d67e7eb2a\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Fri Aug 9 14:55:15 2019 +0200\n\n gpio: aspeed: Pass irqchip when adding gpiochip\n \n We need to convert all old gpio irqchips to pass the irqchip\n setup along when adding the gpio_chip. For more info see\n drivers/gpio/TODO.\n \n For chained irqchips this is a pretty straight-forward\n conversion.\n \n Cc: Joel Stanley \u003cjoel@jms.id.au\u003e\n Cc: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Cc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n Link: https://lore.kernel.org/r/20190809125515.19094-1-linus.walleij@linaro.org\n Reviewed-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Tested-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Acked-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit a2ac3eb3656dd248f4e2b30a44a8ae622063d832\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Fri Aug 9 16:00:05 2019 +0200\n\n gpio: hlwd: Pass irqchip when adding gpiochip\n \n We need to convert all old gpio irqchips to pass the irqchip\n setup along when adding the gpio_chip. For more info see\n drivers/gpio/TODO.\n \n For chained irqchips this is a pretty straight-forward\n conversion.\n \n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n Link: https://lore.kernel.org/r/20190809140005.11654-1-linus.walleij@linaro.org\n Reviewed-by: Jonathan Neuschäfer \u003cj.neuschaefer@gmx.net\u003e\n Tested-by: Jonathan Neuschäfer \u003cj.neuschaefer@gmx.net\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit c7e66e48c05ac2de729999148daa56b563c30b2f\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Fri Aug 9 15:51:19 2019 +0200\n\n gpio: xlp: Pass irqchip when adding gpiochip\n \n We need to convert all old gpio irqchips to pass the irqchip\n setup along when adding the gpio_chip. For more info see\n drivers/gpio/TODO.\n \n For chained irqchips this is a pretty straight-forward\n conversion.\n \n Cc: Jayachandran C \u003cjnair@caviumnetworks.com\u003e\n Cc: Kamlakant Patel \u003ckamlakant.patel@broadcom.com\u003e\n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190809135119.6946-1-linus.walleij@linaro.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit fe8cf20d29309f552c8d9afb3e350e4df2e75150\nAuthor: Lokesh Vutla \u003clokeshvutla@ti.com\u003e\nDate: Fri Aug 9 13:59:42 2019 +0530\n\n dt-bindings: gpio: davinci: Add new compatible for J721E SoCs\n \n J721e SoCs have same gpio IP as K2G davinci gpio. Add a new compatible to\n handle J721E SoCs.\n \n Signed-off-by: Lokesh Vutla \u003clokeshvutla@ti.com\u003e\n Link: https://lore.kernel.org/r/20190809082947.30590-2-lokeshvutla@ti.com\n Reviewed-by: Keerthy \u003cj-keerthy@ti.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 607a0dcddbdc5b147c595f6544e97bd8d9491c14\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Fri Aug 9 15:18:04 2019 +0200\n\n gpio: cadence: Pass irqchip when adding gpiochip\n \n We need to convert all old gpio irqchips to pass the irqchip\n setup along when adding the gpio_chip. For more info see\n drivers/gpio/TODO.\n \n For chained irqchips this is a pretty straight-forward\n conversion.\n \n Cc: Jan Kotas \u003cjank@cadence.com\u003e\n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Tested-by: Jan Kotas \u003cjank@cadence.com\u003e\n Link: https://lore.kernel.org/r/20190809131804.20352-1-linus.walleij@linaro.org\n\ncommit a7fc89f9d5fcc10a5474cfe555f5a9e5df8b0f1f\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Thu Aug 8 14:32:40 2019 +0200\n\n gpio: thunderx: Switch to GPIOLIB_IRQCHIP\n \n Use the new infrastructure for hierarchical irqchips in\n gpiolib.\n \n The major part of the rewrite was dues to the fact that\n the driver was passing around a per-irq pointer to\n struct thunderx_line * data container, and the central\n handlers will assume struct gpio_chip * to be passed\n to we need to use the hwirq as index to look up the\n struct thunderx_line * for each IRQ.\n \n The pushing and pop:ing of the irqdomain was confusing\n because I\u0027ve never seen this before, but I tried to\n replicate it as best I could.\n \n I have no chance to test or debug this so I need\n help.\n \n Cc: David Daney \u003cdavid.daney@cavium.com\u003e\n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n Cc: Brian Masney \u003cmasneyb@onstation.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190808123242.5359-4-linus.walleij@linaro.org\n\ncommit 821c76c4c374adf0c7a7608ee4661aa801f3c1c5\nAuthor: Brian Masney \u003cmasneyb@onstation.org\u003e\nDate: Thu Aug 8 14:32:39 2019 +0200\n\n qcom: spmi-gpio: convert to hierarchical IRQ helpers in gpio core\n \n Now that the GPIO core has support for hierarchical IRQ chips, convert\n Qualcomm\u0027s spmi-gpio over to use these new helpers to reduce duplicated\n code across drivers.\n \n This change was tested on a LG Nexus 5 (hammerhead) phone.\n \n Signed-off-by: Brian Masney \u003cmasneyb@onstation.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190808123242.5359-3-linus.walleij@linaro.org\n\ncommit aa7d618ac65fb10697972549efc38d606595afc7\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Thu Aug 8 14:32:38 2019 +0200\n\n gpio: ixp4xx: Convert to hierarchical GPIOLIB_IRQCHIP\n \n This modifies the IXP4xx driver to use the new helpers\n to handle the remapping of parent to child hardware irqs\n in the gpiolib core.\n \n This pulls the majority of the code out of the driver\n and use the generic code in gpiolib.\n \n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Marc Zyngier \u003cmarc.zyngier@arm.com\u003e\n Cc: Lina Iyer \u003cilina@codeaurora.org\u003e\n Cc: Jon Hunter \u003cjonathanh@nvidia.com\u003e\n Cc: Sowjanya Komatineni \u003cskomatineni@nvidia.com\u003e\n Cc: Bitan Biswas \u003cbbiswas@nvidia.com\u003e\n Cc: linux-tegra@vger.kernel.org\n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n Cc: Brian Masney \u003cmasneyb@onstation.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190808123242.5359-2-linus.walleij@linaro.org\n\ncommit fdd61a013a24f2699aec1a446f0168682b6f9ec4\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Thu Aug 8 14:32:37 2019 +0200\n\n gpio: Add support for hierarchical IRQ domains\n \n Hierarchical IRQ domains can be used to stack different IRQ\n controllers on top of each other.\n \n Bring hierarchical IRQ domains into the GPIOLIB core with the\n following basic idea:\n \n Drivers that need their interrupts handled hierarchically\n specify a callback to translate the child hardware IRQ and\n IRQ type for each GPIO offset to a parent hardware IRQ and\n parent hardware IRQ type.\n \n Users have to pass the callback, fwnode, and parent irqdomain\n before calling gpiochip_irqchip_add().\n \n We use the new method of just filling in the struct\n gpio_irq_chip before adding the gpiochip for all hierarchical\n irqchips of this type.\n \n The code path for device tree is pretty straight-forward,\n while the code path for old boardfiles or anything else will\n be more convoluted requireing upfront allocation of the\n interrupts when adding the chip.\n \n One specific use-case where this can be useful is if a power\n management controller has top-level controls for wakeup\n interrupts. In such cases, the power management controller can\n be a parent to other interrupt controllers and program\n additional registers when an IRQ has its wake capability\n enabled or disabled.\n \n The hierarchical irqchip helper code will only be available\n when IRQ_DOMAIN_HIERARCHY is selected to GPIO chips using\n this should select or depend on that symbol. When using\n hierarchical IRQs, the parent interrupt controller must\n also be hierarchical all the way up to the top interrupt\n controller wireing directly into the CPU, so on systems\n that do not have this we can get rid of all the extra\n code for supporting hierarchical irqs.\n \n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Marc Zyngier \u003cmarc.zyngier@arm.com\u003e\n Cc: Lina Iyer \u003cilina@codeaurora.org\u003e\n Cc: Jon Hunter \u003cjonathanh@nvidia.com\u003e\n Cc: Sowjanya Komatineni \u003cskomatineni@nvidia.com\u003e\n Cc: Bitan Biswas \u003cbbiswas@nvidia.com\u003e\n Cc: linux-tegra@vger.kernel.org\n Cc: David Daney \u003cdavid.daney@cavium.com\u003e\n Cc: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Cc: Brian Masney \u003cmasneyb@onstation.org\u003e\n Signed-off-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: Brian Masney \u003cmasneyb@onstation.org\u003e\n Co-developed-by: Brian Masney \u003cmasneyb@onstation.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190808123242.5359-1-linus.walleij@linaro.org\n\ncommit a0a63940b0c9ebc6d3156bbfe0c2cf7560b580cd\nAuthor: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nDate: Mon Mar 4 19:37:40 2019 +0100\n\n drm/vmwgfx: Assign eviction priorities to resources\n \n TTM provides a means to assign eviction priorities to buffer object. This\n means that all buffer objects with a lower priority will be evicted first\n on memory pressure.\n Use this to make sure surfaces and in particular non-dirty surfaces are\n evicted first. Evicting in particular shaders, cotables and contexts imply\n a significant performance hit on vmwgfx, so make sure these resources are\n evicted last.\n Some buffer objects are sub-allocated in user-space which means we can have\n many resources attached to a single buffer object or resource. In that case\n the buffer object is given the highest priority of the attached resources.\n \n Signed-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Reviewed-by: Deepak Rawat \u003cdrawat@vmware.com\u003e\n\ncommit 6ae8748bf70630c1598bfdeb7e874624a57fd898\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 23 12:23:34 2019 +0200\n\n drm/vmwgfx: drop reminaing users of drmP.h\n \n Drop use of the deprecated drmP.h file from the\n remaining files.\n In several cases the drmP.h include could be removed without\n furter fixes. Other files required a few header files to be added.\n \n In all files divided includes files in blocks and sort them.\n \n v2:\n - fix warning in i386 build wiht HIGHMEM disabled\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e [warning in i386 build]\n Cc: VMware Graphics \u003clinux-graphics-maintainer@vmware.com\u003e\n Cc: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Reviewed-by: Deepak Rawat \u003cdrawat@vmware.com\u003e\n Signed-off-by: Deepak Rawat \u003cdrawat@vmware.com\u003e\n Signed-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n\ncommit d5c1f0110dd6cb182d698b2df7e0445282a1392d\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 23 12:23:33 2019 +0200\n\n drm/vmwgfx: drop use of drmP.h in header files\n \n To facilitate removal of drmP.h in the .c\n files remove the use from header files first.\n Fix fallout in the other files.\n \n Sorted include files in blocks and sorted files\n within each block in alphabetical order.\n \n This revealed a dependency from an uapi header to a header\n located below drivers/gpu/drm/vmwgfx/.\n Added FIXME to remind someone to fix this.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: VMware Graphics \u003clinux-graphics-maintainer@vmware.com\u003e\n Cc: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Reviewed-by: Deepak Rawat \u003cdrawat@vmware.com\u003e\n Signed-off-by: Deepak Rawat \u003cdrawat@vmware.com\u003e\n Signed-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n\ncommit 9c84aeba67cce6514374f134c9ce2c5f35ac3831\nAuthor: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\nDate: Tue May 28 08:08:55 2019 +0200\n\n drm/vmwgfx: Kill unneeded legacy security features\n \n At one point, the GPU command verifier and user-space handle manager\n couldn\u0027t properly protect GPU clients from accessing each other\u0027s data.\n Instead there was an elaborate mechanism to make sure only the active\n master\u0027s primary clients could render. The other clients were either\n put to sleep or even killed (if the master had exited). VRAM was\n evicted on master switch. With the advent of render-node functionality,\n we relaxed the VRAM eviction, but the other mechanisms stayed in place.\n \n Now that the GPU command verifier and ttm object manager properly\n isolates primary clients from different master realms we can remove the\n master switch related code and drop those legacy features.\n \n Signed-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Reviewed-by: Deepak Rawat \u003cdrawat@vmware.com\u003e\n\ncommit ecb9f80db23a7ab09b46b298b404e41dd7aff6e6\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Tue Aug 13 10:00:25 2019 +0200\n\n net/mvpp2: Replace tasklet with softirq hrtimer\n \n The tx_done_tasklet tasklet is used in invoke the hrtimer\n (mvpp2_hr_timer_cb) in softirq context. This can be also achieved without\n the tasklet but with HRTIMER_MODE_SOFT as hrtimer mode.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 777b4736933f774193065b3ad3708803d0c871d8\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Tue Aug 13 14:23:49 2019 +0100\n\n scsi: fnic: remove redundant assignment of variable rc\n \n Variable ret is initialized to a value that is never read and it is\n re-assigned later and immediately returns. Clean up the code by removing\n rc and just returning 0.\n \n [mkp: typo]\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Acked-by: Karan Tilak Kumar \u003ckartilak@cisco.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit cff1191553d99d46369dafec8f43e780278e4071\nAuthor: Martin Wilck \u003cmwilck@suse.com\u003e\nDate: Wed Aug 14 13:28:29 2019 +0000\n\n scsi: qla2xxx: cleanup trace buffer initialization\n \n Avoid code duplication between qla2x00_alloc_offload_mem() and\n qla2x00_alloc_fw_dump() by moving the FCE and EFT buffer allocation and\n initialization to separate functions. Cleanly track failure and success by\n making sure that the ha-\u003eeft, ha-\u003efce and respective eft_dma, fce_dma\n members are set if and only if the buffers are properly allocated and\n initialized. Avoid pointless buffer reallocation. Eliminate some goto\n statements. Make sure the fce_enabled flag is cleared when the FCE buffer\n is freed.\n \n Fixes: ad0a0b01f088 (\"scsi: qla2xxx: Fix Firmware dump size for Extended\n login and Exchange Offload\")\n Fixes: (a28d9e4ef997 \"scsi: qla2xxx: Add support for multiple fwdump\n templates/segments\")\n Cc: Joe Carnuccio \u003cjoe.carnuccio@cavium.com\u003e\n Cc: Quinn Tran \u003cqutran@marvell.com\u003e\n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Cc: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Martin Wilck \u003cmwilck@suse.com\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 530106fa95c998ff16fb80a6ad10829396083931\nAuthor: Martin Wilck \u003cmwilck@suse.com\u003e\nDate: Wed Aug 14 13:28:27 2019 +0000\n\n scsi: qla2xxx: qla2x00_alloc_fw_dump: set ha-\u003eeft\n \n In qla2x00_alloc_fw_dump(), an existing EFT buffer (e.g. from previous\n invocation of qla2x00_alloc_offload_mem()) is freed. The buffer is then\n re-allocated, but without setting the eft and eft_dma fields to the new\n values.\n \n Fixes: a28d9e4ef997 (\"scsi: qla2xxx: Add support for multiple fwdump templates/segments\")\n Cc: Joe Carnuccio \u003cjoe.carnuccio@cavium.com\u003e\n Cc: Quinn Tran \u003cqutran@marvell.com\u003e\n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Cc: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Martin Wilck \u003cmwilck@suse.com\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 5518ba4ebd7d1e15fc558af05fa600198535f074\nAuthor: Dan Williams \u003cdan.j.williams@intel.com\u003e\nDate: Wed Aug 7 14:31:11 2019 -0700\n\n tools/testing/nvdimm: Fix fallthrough warning\n \n Use the expected \u0027fall through\u0027 designation to fix:\n \n tools/testing/nvdimm/test/nfit.c: In function ‘nd_intel_test_finish_query’:\n tools/testing/nvdimm/test/nfit.c:433:13: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n fw-\u003estate \u003d FW_STATE_UPDATED;\n ~~~~~~~~~~^~~~~~~~~~~~~~~~~~\n tools/testing/nvdimm/test/nfit.c:435:2: note: here\n case FW_STATE_UPDATED:\n ^~~~\n \n Reviewed-by: Vishal Verma \u003cvishal.l.verma@intel.com\u003e\n Link: https://lore.kernel.org/r/156521347159.1442374.1381360879102718899.stgit@dwillia2-desk3.amr.corp.intel.com\n Signed-off-by: Dan Williams \u003cdan.j.williams@intel.com\u003e\n\ncommit 83c156d3ecc0121d27dc2b7f34e829b265c70c4f\nAuthor: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\nDate: Wed Aug 14 09:58:09 2019 -0700\n\n netfilter: nft_bitwise: Adjust parentheses to fix memcmp size argument\n \n clang warns:\n \n net/netfilter/nft_bitwise.c:138:50: error: size argument in \u0027memcmp\u0027\n call is a comparison [-Werror,-Wmemsize-comparison]\n if (memcmp(\u0026priv-\u003exor, \u0026zero, sizeof(priv-\u003exor) ||\n ~~~~~~~~~~~~~~~~~~^~\n net/netfilter/nft_bitwise.c:138:6: note: did you mean to compare the\n result of \u0027memcmp\u0027 instead?\n if (memcmp(\u0026priv-\u003exor, \u0026zero, sizeof(priv-\u003exor) ||\n ^\n )\n net/netfilter/nft_bitwise.c:138:32: note: explicitly cast the argument\n to size_t to silence this warning\n if (memcmp(\u0026priv-\u003exor, \u0026zero, sizeof(priv-\u003exor) ||\n ^\n (size_t)(\n 1 error generated.\n \n Adjust the parentheses so that the result of the sizeof is used for the\n size argument in memcmp, rather than the result of the comparison (which\n would always be true because sizeof is a non-zero number).\n \n Fixes: bd8699e9e292 (\"netfilter: nft_bitwise: add offload support\")\n Link: https://github.com/ClangBuiltLinux/linux/issues/638\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Reviewed-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit 707816c8b0508ba7afcee9a13b02bb4261e4bd8c\nAuthor: Jeremy Sowden \u003cjeremy@azazel.net\u003e\nDate: Wed Aug 14 09:01:28 2019 +0100\n\n netfilter: remove deprecation warnings from uapi headers.\n \n There are two netfilter userspace headers which contain deprecation\n warnings. While these headers are not used within the kernel, they are\n compiled stand-alone for header-testing.\n \n Pablo informs me that userspace iptables still refer to these headers,\n and the intention was to use xt_LOG.h instead and remove these, but\n userspace was never updated.\n \n Remove the warnings.\n \n Fixes: 2a475c409fe8 (\"kbuild: remove all netfilter headers from header-test blacklist.\")\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Jeremy Sowden \u003cjeremy@azazel.net\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit a9436dca115d121d98e0b30f078f3294ce13fa18\nAuthor: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\nDate: Wed Aug 14 12:37:24 2019 +0100\n\n tools: bpftool: compile with $(EXTRA_WARNINGS)\n \n Compile bpftool with $(EXTRA_WARNINGS), as defined in\n scripts/Makefile.include, and fix the new warnings produced.\n \n Simply leave -Wswitch-enum out of the warning list, as we have several\n switch-case structures where it is not desirable to process all values\n of an enum.\n \n Remove -Wshadow from the warnings we manually add to CFLAGS, as it is\n handled in $(EXTRA_WARNINGS).\n \n Signed-off-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit a7a03a6956487cf6462581449ab1de8661c882f2\nAuthor: Yabin Cui \u003cyabinc@google.com\u003e\nDate: Mon Aug 12 15:11:54 2019 -0700\n\n coresight: tmc-etr: Fix updating buffer in not-snapshot mode.\n \n TMC etr always copies all available data to perf aux buffer, which\n may exceed the available space in perf aux buffer. It isn\u0027t suitable\n for not-snapshot mode, because:\n 1) It may overwrite previously written data.\n 2) It may make the perf_event_mmap_page-\u003eaux_head report having more\n or less data than the reality.\n \n So change to only copy the latest data fitting the available space in\n perf aux buffer.\n \n Signed-off-by: Yabin Cui \u003cyabinc@google.com\u003e\n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n\ncommit dc2e1e5b279966affbd11ff7cfef52eb634ca2c9\nAuthor: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nDate: Tue Aug 13 23:11:01 2019 +0300\n\n drm/panel: Add driver for the Toppoly TD043MTEA1 panel\n \n This panel is used on the OMAP3 Pandora.\n \n The code is based on the omapdrm-specific panel-tpo-td043mtea1 driver.\n \n v2:\n - fix checkpatch warnings\n o (lcd \u003d\u003d NULL) \u003d\u003e (!lcd) (sam)\n o alignment to open \u0027(\u0027 (sam)\n \n Signed-off-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813201101.30980-10-laurent.pinchart@ideasonboard.com\n\ncommit 415b8dd087118829fd37548b0d75592e297c1fa9\nAuthor: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nDate: Tue Aug 13 23:11:00 2019 +0300\n\n drm/panel: Add driver for the Toppoly TD028TTEC1 panel\n \n This panel is used on the OpenMoko Neo FreeRunner and Neo 1973.\n \n The code is based on the omapdrm-specific panel-tpo-td028ttec1 driver.\n \n v2:\n - fix checkpatch warnings:\n o (lcd \u003d\u003d NULL) \u003d\u003e (!lcd) (sam)\n o (1 \u003c\u003c X) \u003d\u003e BIT(X) (sam)\n \n Signed-off-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813201101.30980-9-laurent.pinchart@ideasonboard.com\n\ncommit 1c8fc3f0c5d280d691c330388666c3726fa525c7\nAuthor: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nDate: Tue Aug 13 23:10:59 2019 +0300\n\n drm/panel: Add driver for the Sony ACX565AKM panel\n \n This panel is used on the Nokia N900.\n \n The code is based on the omapdrm-specific panel-sony-acx565akm driver.\n The hardware-related logic has been changed as little as possible to\n avoid regressions as hardware availability is lacking to test the\n changes. Follow-up patches should address the items listed in the TODO\n list.\n \n v2:\n - fix checkpatch warning (lcd \u003d\u003d NULL) \u003d\u003e (!lcd) (sam)\n \n Signed-off-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813201101.30980-8-laurent.pinchart@ideasonboard.com\n\ncommit c9cf4c2a3bd35f4e90e840979ba7872932f9f120\nAuthor: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nDate: Tue Aug 13 23:10:58 2019 +0300\n\n drm/panel: Add driver for the Sharp LS037V7DW01 panel\n \n This panel is used on the TI SDP3430 board.\n \n The code is based on the omapdrm-specific panel-sharp-ls037v7dw01\n driver.\n \n v2:\n - fix checkpatch warning (lcd \u003d\u003d NULL) \u003d\u003e (!lcd) (sam)\n - drop __exit_p() from remove. It caused a build warning.\n And no other panel drivers needs this (sam)\n \n Signed-off-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813201101.30980-7-laurent.pinchart@ideasonboard.com\n\ncommit df439abe650177aeb6f0f1baf085a7367cac1bc2\nAuthor: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nDate: Tue Aug 13 23:10:57 2019 +0300\n\n drm/panel: Add driver for the NEC NL8048HL11 panel\n \n This panel is used on the Zoom2/3/3630 SDP boards.\n \n The code is based on the omapdrm-specific panel-nec-nl8048hl11 driver\n \n v2:\n - fix checkpatch warning (lcd \u003d\u003d NULL) \u003d\u003e (!lcd) (sam)\n \n Signed-off-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813201101.30980-6-laurent.pinchart@ideasonboard.com\n\ncommit f5b0c6542476e9068ec2731b644ce668491cb8fa\nAuthor: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nDate: Tue Aug 13 23:10:56 2019 +0300\n\n drm/panel: Add driver for the LG Philips LB035Q02 panel\n \n This panel is used on the Gumstix Overo Palo35.\n \n The code is based on the omapdrm-specific panel-lgphilips-lb035q02\n driver.\n \n Signed-off-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813201101.30980-5-laurent.pinchart@ideasonboard.com\n\ncommit 90075dde44a26446b3850e847f997bc08786b71e\nAuthor: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nDate: Tue Aug 13 23:10:55 2019 +0300\n\n dt-bindings: display: panel: Add bindings for NEC NL8048HL11 panel\n \n The NEC NL8048HL11 is a 10.4cm WVGA (800x480) panel with a 24-bit RGB\n parallel data interface and an SPI control interface.\n \n Signed-off-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813201101.30980-4-laurent.pinchart@ideasonboard.com\n\ncommit 09e6890c3f1aea8150bc60e33392587321b840ce\nAuthor: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nDate: Tue Aug 13 23:10:54 2019 +0300\n\n dt-bindings: Add legacy \u0027toppoly\u0027 vendor prefix\n \n The \u0027toppoly\u0027 vendor prefix is in use and refers to TPO, whose DT vendor\n prefix is already defined as \u0027tpo\u0027. Add \u0027toppoly\u0027 as an alternative and\n document it as legacy.\n \n Signed-off-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813201101.30980-3-laurent.pinchart@ideasonboard.com\n\ncommit 684a4e6739d38943fb6357f421dafe3d5416afed\nAuthor: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\nDate: Tue Aug 13 23:10:53 2019 +0300\n\n dt-bindings: Add vendor prefix for LG Display\n \n LG Display is an LCD display manufacturer. Originally formed as a joint\n venture by LG Electronics and Philips Electronics, it was formerly known\n as LG.Philips LCD, hence the DT vendor prefix lgphilips (which is\n already in active use in the kernel).\n \n More information is available at\n https://en.wikipedia.org/wiki/LG_Display.\n \n Signed-off-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813201101.30980-2-laurent.pinchart@ideasonboard.com\n\ncommit 590ef0128b4277eaa9cb74d7ba878d48b261950c\nAuthor: Eric W. Biederman \u003cebiederm@xmission.com\u003e\nDate: Tue Jul 23 07:44:46 2019 -0500\n\n waitid: Add support for waiting for the current process group\n \n It was recently discovered that the linux version of waitid is not a\n superset of the other wait functions because it does not include support\n for waiting for the current process group. This has two downsides:\n 1. An extra system call is needed to get the current process group.\n 2. After the current process group is received and before it is passed\n to waitid a signal could arrive causing the current process group to change.\n Inherent race-conditions as these make it impossible for userspace to\n emulate this functionaly and thus violate async-signal safety\n requirements for waitpid.\n \n Arguments can be made for using a different choice of idtype and id\n for this case but the BSDs already use this P_PGID and 0 to indicate\n waiting for the current process\u0027s process group. So be nice to user\n space programmers and don\u0027t introduce an unnecessary incompatibility.\n \n Some people have noted that the posix description is that\n waitpid will wait for the current process group, and that in\n the presence of pthreads that process group can change. To get\n clarity on this issue I looked at XNU, FreeBSD, and Luminos. All of\n those flavors of unix waited for the current process group at the\n time of call and as written could not adapt to the process group\n changing after the call.\n \n At one point Linux did adapt to the current process group changing but\n that stopped in 161550d74c07 (\"pid: sys_wait... fixes\"). It has been\n over 11 years since Linux has that behavior, no programs that fail\n with the change in behavior have been reported, and I could not\n find any other unix that does this. So I think it is safe to clarify\n the definition of current process group, to current process group\n at the time of the wait function.\n \n Signed-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\n Signed-off-by: Christian Brauner \u003cchristian.brauner@ubuntu.com\u003e\n Reviewed-by: Oleg Nesterov \u003coleg@redhat.com\u003e\n Cc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: Palmer Dabbelt \u003cpalmer@sifive.com\u003e\n Cc: Rich Felker \u003cdalias@libc.org\u003e\n Cc: Alistair Francis \u003calistair23@gmail.com\u003e\n Cc: Zong Li \u003czongbox@gmail.com\u003e\n Cc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n Cc: Oleg Nesterov \u003coleg@redhat.com\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n Cc: Florian Weimer \u003cfweimer@redhat.com\u003e\n Cc: Adhemerval Zanella \u003cadhemerval.zanella@linaro.org\u003e\n Cc: GNU C Library \u003clibc-alpha@sourceware.org\u003e\n Link: https://lore.kernel.org/r/20190814154400.6371-2-christian.brauner@ubuntu.com\n\ncommit b8450d5218dde8112d990a1f17026709e8e67752\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Wed Aug 14 21:46:57 2019 +0200\n\n arm-soc: document merges\n \n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 70c55c1ad1a76e804ee5330e134674f5d2741cb7\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Sun Jul 28 14:11:24 2019 -0300\n\n media: vsp1: fix memory leak of dl on error return path\n \n Currently when the call vsp1_dl_body_get fails and returns null the\n error return path leaks the allocation of dl. Fix this by kfree\u0027ing\n dl before returning.\n \n Addresses-Coverity: (\"Resource leak\")\n \n Fixes: 5d7936b8e27d (\"media: vsp1: Convert display lists to use new body pool\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Reviewed-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Signed-off-by: Laurent Pinchart \u003claurent.pinchart+renesas@ideasonboard.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit ddf7cb83b0f45feb94ad89a987f600c766c463ca\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Wed Aug 14 19:59:44 2019 +0200\n\n ALSA: hda: Unexport a few more stuff\n \n Drop EXPORT_SYMBOL*() from a few more stuff in HD-audio core that\n aren\u0027t used outside. Particular the unsol event handler can be\n staticized now because the recent change removed all external\n callers.\n \n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 53eff75e5f4dd4b9bc489955fdc60fde48d85e93\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Wed Aug 14 18:27:08 2019 +0200\n\n ALSA: hda: Drop export of snd_hdac_bus_add/remove_device()\n \n snd_hdac_bus_add_device() and snd_hdac_remove_device() are called only\n internally in hda-core. Let\u0027s drop the exports of them and move the\n declarations into local.h.\n \n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 0bec8b7e5ca1a629f26173691526432f9d7cf8c1\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Wed Aug 14 10:46:38 2019 -0700\n\n Input: remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 95cbf02b8a5d93d8404b5568ebf6862e87314ead\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Aug 13 12:03:55 2019 +0530\n\n drm/aspeed: gfc_crtc: Make structure aspeed_gfx_funcs constant\n \n The static structure aspeed_gfx_funcs, of type\n drm_simple_display_pipe_funcs, is used only as an argument to\n drm_simple_display_pipe_init(), which does not modify it. Hence make it\n constant to protect it from unintended modification.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813063355.25549-1-nishkadg.linux@gmail.com\n\ncommit 5181b473d64ee278f24035ce335b89ddc4520fc0\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Tue Aug 13 08:09:32 2019 +0200\n\n net: phy: realtek: add NBase-T PHY auto-detection\n \n Realtek provided information on how the new NIC-integrated PHY\u0027s\n expose whether they support 2.5G/5G/10G. This allows to automatically\n differentiate 1Gbps and 2.5Gbps PHY\u0027s, and therefore allows to\n remove the fake PHY ID mechanism for RTL8125.\n So far RTL8125 supports 2.5Gbps only, but register layout for faster\n modes has been defined already, so let\u0027s use this information to be\n future-proof.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d082402e2174314e87131771d048925b2245b74c\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sun Aug 11 00:53:01 2019 +0900\n\n kbuild: remove meaningless \u0027targets\u0027 in ./Kbuild\n \n timeconst.h is generated by $(call filechk,...), missing-syscalls and\n old-atomics are invoked by $(call cmd,...)\n \n None of them needs to be added to \u0027targets\u0027.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit d4945049ad668f7472df50aaea98c7fea262e6d4\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sun Aug 11 00:53:00 2019 +0900\n\n kbuild: remove \u0027make /\u0027 support\n \n \u0027make /\u0027 is just an alias for \u0027make ./\u0027; this builds all objects of an\n external module, but skips the modpost stage.\n \n I am not a big fan of \u0027make /\u0027 since it looks as if it were touching\n the root directory of the system. I like \u0027make ./\u0027 better.\n \n I do not know how many people are using it, but let\u0027s show a hint if\n it is used. Also, move it close to the external module rules since\n this only makes sense for external modules.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit c2290f3286b6d1f2058b905cc954243c6027c37a\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sun Aug 11 00:52:59 2019 +0900\n\n kbuild: fix modkern_aflags implementation\n \n For the single target building %.symtypes from %.S, $(a_flags) is\n expanded into the _KERNEL flags even if the object is a part of a\n module.\n \n $(real-obj-m:.o\u003d.symtypes): modkern_aflags :\u003d $(KBUILD_AFLAGS_MODULE) $(AFLAGS_MODULE)\n \n ... would fix the issue, but it is not nice to duplicate similar code\n for every suffix.\n \n Implement modkern_aflags in the same way as modkern_cflags.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 986662b90352d79c4842dd7d4e678f50824ed729\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sun Aug 11 00:52:58 2019 +0900\n\n kbuild: refactor part-of-module more\n \n Make it even shorter.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 16b59cd78595df5d77c535a08537f2d9dae61496\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sun Aug 11 00:52:57 2019 +0900\n\n kbuild: move the Module.symvers check for external module build\n \n $(objtree)/Module.symvers is not required for descending into\n sub-directories. It is needed for the modpost stage.\n \n Move the Module.symvers check to the right place.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 33e84f2e79659e410c379b530dac57779f201d15\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Tue Aug 6 15:39:19 2019 +0900\n\n kbuild: treat an object as multi-used when $(foo-) is set\n \n Currently, Kbuild treats an object as multi-used when any of\n $(foo-objs), $(foo-y), $(foo-m) is set. It makes more sense to\n check $(foo-) as well.\n \n In the context of foo-$(CONFIG_FOO_FEATURE1), CONFIG_FOO_FEATURE1\n could be unset.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 3584be9ec3bfe2c12bcb40da13fa185d237bff7d\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Wed Jul 31 21:56:55 2019 +0200\n\n ARM: orion/mvebu: unify debug-ll virtual addresses\n \n In a multiplatform configuration, enabling DEBUG_LL breaks booting\n on all platforms with incompatible settings. In case of the Marvell\n platforms of the Orion/MVEBU family, the physical addresses are\n all the same, we just map them at different virtual addresses,\n which makes it impossible to run a kernel with DEBUG_LL enabled on\n a combination of the merged mvebu and the legacy boardfile based\n platforms.\n \n This is easily solved by using the same virtual address everywhere.\n I picked the address that is already used by mach-mvebu for UART0:\n 0xfec12000. All these platforms have a 1MB region with their internal\n registers, almost always at physical address 0xf1000000, so I\u0027m\n updating the iotable for that entry.\n \n In case of mach-dove, this is slightly trickier, as the existing\n mapping is 8MB and a second 8MB mapping is already at the 0xfec00000\n address. I have verified from the datasheet that the last 7MB of the\n physical mapping are \"reserved\" and nothing in Linux tries to use\n it either. I\u0027m putting this 1MB mapping at the same address as the\n others, and the second 8MB register area immediately before that.\n \n Link: https://lore.kernel.org/r/20190731195713.3150463-14-arnd@arndb.de\n Link: https://lore.kernel.org/linux-arm-kernel/87si3eb1z8.fsf@free-electrons.com/\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit ce78179ea6e042e2d3af0a1c71c105431ead483a\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Wed Jul 31 21:56:54 2019 +0200\n\n ARM: dove: clean up mach/*.h headers\n \n This is a simple move of all header files that are no longer\n included by anything else from the include/mach directory\n to the platform directory itself as preparation for\n multiplatform support.\n \n The mach/uncompress.h headers are left in place for now,\n and are mildly modified to be independent of the other\n headers. They will be removed entirely when ARCH_MULTIPLATFORM\n gets enabled and they become obsolete.\n \n Rather than updating the path names inside of the comments\n of each header, I delete those comments to avoid having to\n update them again, should they get moved or copied another\n time.\n \n Link: https://lore.kernel.org/r/20190731195713.3150463-13-arnd@arndb.de\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Acked-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 4773f9bdb476501df565b610adc0edf1932f2c6e\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Mon Aug 12 20:47:40 2019 +0200\n\n r8169: fix sporadic transmit timeout issue\n \n Holger reported sporadic transmit timeouts and it turned out that one\n path misses ringing the doorbell. Fix was suggested by Eric.\n \n Fixes: ef14358546b1 (\"r8169: make use of xmit_more\")\n Suggested-by: Eric Dumazet \u003cedumazet@google.com\u003e\n Tested-by: Holger Hoffstätte \u003cholger@applied-asynchrony.com\u003e\n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 535d1b947ba3bb6d37791277bf71f1c848522e6c\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Aug 4 11:41:32 2019 +0200\n\n drm/arm: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n \n While touching the list of include files divide them\n into blocks and sort within each block.\n Fix fallout.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Cc: Liviu Dudau \u003cliviu.dudau@arm.com\u003e\n Cc: Brian Starkey \u003cbrian.starkey@arm.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: malidp@foss.arm.com\n Link: https://patchwork.freedesktop.org/patch/msgid/20190804094132.29463-5-sam@ravnborg.org\n\ncommit 25e28ef2805f3196c7de1eaf47bbad537adefa41\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Aug 4 11:41:31 2019 +0200\n\n drm/armada: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n While touching the list of include files group them and sort them.\n Fix fallout from the header file removal.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Cc: Russell King \u003clinux@armlinux.org.uk\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190804094132.29463-4-sam@ravnborg.org\n\ncommit eb1df694cd7271632763bb99f7fb3891357461d8\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Aug 4 11:41:30 2019 +0200\n\n drm/tegra: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n \n For all touched files divide include files into blocks,\n and sort them within the blocks.\n Fix fallout.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: Jonathan Hunter \u003cjonathanh@nvidia.com\u003e\n Cc: linux-tegra@vger.kernel.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190804094132.29463-3-sam@ravnborg.org\n\ncommit 5c8a5f1d78d84faff0e5578abb4b53adab8aaa59\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Aug 4 11:41:29 2019 +0200\n\n drm/i2c/tda998x: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n Fix fallout.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Cc: Russell King \u003clinux@armlinux.org.uk\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190804094132.29463-2-sam@ravnborg.org\n\ncommit 718d58514ebc5740f80460ae7b73c57a01eac4c2\nAuthor: Robert Richter \u003crrichter@marvell.com\u003e\nDate: Mon Jun 24 15:09:13 2019 +0000\n\n EDAC/mc: Cleanup _edac_mc_free() code\n \n Remove needless and boilerplate variable declarations. No functional\n changes.\n \n [ bp: Add newlines for better readability. ]\n \n Signed-off-by: Robert Richter \u003crrichter@marvell.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Cc: linux-edac \u003clinux-edac@vger.kernel.org\u003e\n Cc: James Morse \u003cjames.morse@arm.com\u003e\n Cc: Mauro Carvalho Chehab \u003cmchehab@kernel.org\u003e\n Cc: Tony Luck \u003ctony.luck@intel.com\u003e\n Link: https://lkml.kernel.org/r/20190624150758.6695-10-rrichter@marvell.com\n\ncommit ee5f85d9290fe25d460bd320b7fe073075d72d33\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Tue Aug 13 17:25:39 2019 +0200\n\n ALSA: hda: Add codec on bus address table lately\n \n The call of snd_hdac_bus_add_device() is needed only for registering\n the codec onto the bus caddr_tbl[] that is referred essentially only\n in the unsol event handler. That is, the reason of this call and the\n release by the counter-part function snd_hdac_bus_remove_device() is\n just to assure that the unsol event gets notified to the codec.\n \n But the current implementation of the unsol notification wouldn\u0027t work\n properly when the codec is still in a premature init state. So this\n patch tries to work around it by delaying the caddr_tbl[] registration\n at the point of snd_hdac_device_register().\n \n Also, the order of snd_hdac_bus_remove_device() and device_del() calls\n are shuffled to make sure that the unsol event is masked before\n deleting the device.\n \n BugLink: https://bugzilla.kernel.org/show_bug.cgi?id\u003d204565\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit f2dbe87c5ac1f88e6007ba1f1374f4bd8a197fb6\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Tue Aug 13 17:11:28 2019 +0200\n\n ALSA: hda - Drop unsol event handler for Intel HDMI codecs\n \n We don\u0027t need to deal with the unsol events for Intel chips that are\n tied with the graphics via audio component notifier. Although the\n presence of the audio component is checked at the beginning of\n hdmi_unsol_event(), better to short cut by dropping unsol_event ops.\n \n BugLink: https://bugzilla.kernel.org/show_bug.cgi?id\u003d204565\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 90776dd1c427cbb4d381aa4b13338f1fb1d20f5e\nAuthor: Kees Cook \u003ckeescook@chromium.org\u003e\nDate: Tue Aug 13 16:04:50 2019 -0700\n\n arm64/efi: Move variable assignments after SECTIONS\n \n It seems that LLVM\u0027s linker does not correctly handle variable assignments\n involving section positions that are updated during the SECTIONS\n parsing. Commit aa69fb62bea1 (\"arm64/efi: Mark __efistub_stext_offset as\n an absolute symbol explicitly\") ran into this too, but found a different\n workaround.\n \n However, this was not enough, as other variables were also miscalculated\n which manifested as boot failures under UEFI where __efistub__end was\n not taking the correct _end value (they should be the same):\n \n $ ld.lld -EL -maarch64elf --no-undefined -X -shared \\\n -Bsymbolic -z notext -z norelro --no-apply-dynamic-relocs \\\n -o vmlinux.lld -T poc.lds --whole-archive vmlinux.o \u0026\u0026 \\\n readelf -Ws vmlinux.lld | egrep \u0027\\b(__efistub_|)_end\\b\u0027\n 368272: ffff000002218000 0 NOTYPE LOCAL HIDDEN 38 __efistub__end\n 368322: ffff000012318000 0 NOTYPE GLOBAL DEFAULT 38 _end\n \n $ aarch64-linux-gnu-ld.bfd -EL -maarch64elf --no-undefined -X -shared \\\n -Bsymbolic -z notext -z norelro --no-apply-dynamic-relocs \\\n -o vmlinux.bfd -T poc.lds --whole-archive vmlinux.o \u0026\u0026 \\\n readelf -Ws vmlinux.bfd | egrep \u0027\\b(__efistub_|)_end\\b\u0027\n 338124: ffff000012318000 0 NOTYPE LOCAL DEFAULT ABS __efistub__end\n 383812: ffff000012318000 0 NOTYPE GLOBAL DEFAULT 15325 _end\n \n To work around this, all of the __efistub_-prefixed variable assignments\n need to be moved after the linker script\u0027s SECTIONS entry. As it turns\n out, this also solves the problem fixed in commit aa69fb62bea1, so those\n changes are reverted here.\n \n Link: https://github.com/ClangBuiltLinux/linux/issues/634\n Link: https://bugs.llvm.org/show_bug.cgi?id\u003d42990\n Acked-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 38d16667604e31f30a715baefec2fe3aa88024f0\nAuthor: Mark Brown \u003cbroonie@kernel.org\u003e\nDate: Thu Aug 8 15:05:54 2019 +0100\n\n arm64: Clarify when cpu_enable() is called\n \n Strengthen the wording in the documentation for cpu_enable() to make it\n more obvious to readers not already familiar with the code when the core\n will call this callback and that this is intentional.\n \n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n Reviewed-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n [will: minor tweak to emphasis in the comment]\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 77ad4ce69321abbe26ec92b2a2691a66531eb688\nAuthor: Mark Rutland \u003cmark.rutland@arm.com\u003e\nDate: Wed Aug 14 14:28:48 2019 +0100\n\n arm64: memory: rename VA_START to PAGE_END\n \n Prior to commit:\n \n 14c127c957c1c607 (\"arm64: mm: Flip kernel VA space\")\n \n ... VA_START described the start of the TTBR1 address space for a given\n VA size described by VA_BITS, where all kernel mappings began.\n \n Since that commit, VA_START described a portion midway through the\n address space, where the linear map ends and other kernel mappings\n begin.\n \n To avoid confusion, let\u0027s rename VA_START to PAGE_END, making it clear\n that it\u0027s not the start of the TTBR1 address space and implying that\n it\u0027s related to PAGE_OFFSET. Comments and other mnemonics are updated\n accordingly, along with a typo fix in the decription of VMEMMAP_SIZE.\n \n There should be no functional change as a result of this patch.\n \n Signed-off-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Tested-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Reviewed-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 233947ef16a18952d22786770dab1ddafa1ac377\nAuthor: Mark Rutland \u003cmark.rutland@arm.com\u003e\nDate: Wed Aug 14 14:28:47 2019 +0100\n\n arm64: memory: fix flipped VA space fallout\n \n VA_START used to be the start of the TTBR1 address space, but now it\u0027s a\n point midway though. In a couple of places we still use VA_START to get\n the start of the TTBR1 address space, so let\u0027s fix these up to use\n PAGE_OFFSET instead.\n \n Fixes: 14c127c957c1c607 (\"arm64: mm: Flip kernel VA space\")\n Signed-off-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Tested-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Reviewed-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit d7da429339f503854b777ed941456841a72b48b6\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Tue Aug 13 15:01:51 2019 +0100\n\n ALSA: sb: remove redundant assignment to variable result\n \n Variable result is initialized to a value that is never read and it is\n re-assigned later. The initialization is redundant and can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 4ed3350539aa931f58c939fcd803c7510584e143\nAuthor: Alan Stern \u003cstern@rowland.harvard.edu\u003e\nDate: Tue Aug 13 16:15:38 2019 -0400\n\n USB: usbfs: Add a capability flag for runtime suspend\n \n The recent commit 7794f486ed0b (\"usbfs: Add ioctls for runtime power\n management\") neglected to add a corresponding capability flag. This\n patch rectifies the omission.\n \n Signed-off-by: Alan Stern \u003cstern@rowland.harvard.edu\u003e\n CC: Mayuresh Kulkarni \u003cmkulkarni@opensource.cirrus.com\u003e\n \n Link: https://lore.kernel.org/r/Pine.LNX.4.44L0.1908131613490.1941-100000@iolanthe.rowland.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b8e24a9300b0836a9d39f6b20746766b3b81f1bd\nAuthor: Johannes Weiner \u003channes@cmpxchg.org\u003e\nDate: Thu Aug 8 15:03:00 2019 -0400\n\n block: annotate refault stalls from IO submission\n \n psi tracks the time tasks wait for refaulting pages to become\n uptodate, but it does not track the time spent submitting the IO. The\n submission part can be significant if backing storage is contended or\n when cgroup throttling (io.latency) is in effect - a lot of time is\n spent in submit_bio(). In that case, we underreport memory pressure.\n \n Annotate submit_bio() to account submission time as memory stall when\n the bio is reading userspace workingset pages.\n \n Tested-by: Suren Baghdasaryan \u003csurenb@google.com\u003e\n Signed-off-by: Johannes Weiner \u003channes@cmpxchg.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit c391818a10def7f5aa989842892a4306e75c917d\nAuthor: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\nDate: Tue Aug 13 15:43:07 2019 +0300\n\n staging: fsl-dpaa2/ethsw: do not force user to bring interface down\n \n Link settings can be changed only when the interface is down. Disable\n and re-enable the interface, if necessary, behind the scenes so that we do\n not force users to an if down/up sequence.\n \n Reported-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Link: https://lore.kernel.org/r/1565700187-16048-11-git-send-email-ioana.ciornei@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 18f1296f067e09f1aba8971a2b76b78734a5eb87\nAuthor: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\nDate: Tue Aug 13 15:43:06 2019 +0300\n\n staging: fsl-dpaa2/ethsw: register_netdev only when ready\n \n The register_netdev() call should be made only when ready to process any\n user request on the interface. Move the call to be the last one issued\n in the probe sequence.\n \n Reported-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Link: https://lore.kernel.org/r/1565700187-16048-10-git-send-email-ioana.ciornei@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 38fc21369eaaac937a3bbb0a445014731dd32566\nAuthor: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\nDate: Tue Aug 13 15:43:05 2019 +0300\n\n staging: fsl-dpaa2/ethsw: reword error message\n \n In the current state, the dpaa2-ethsw driver supports only one bridge\n per DPSW object. Reword the error message so that this information is\n much more clear.\n \n Suggested-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Link: https://lore.kernel.org/r/1565700187-16048-9-git-send-email-ioana.ciornei@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit e76d3d60e3bb918fffb1af3558fe34cd3ea2116b\nAuthor: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\nDate: Tue Aug 13 15:43:04 2019 +0300\n\n staging: fsl-dpaa2/ethsw: remove redundant VLAN check\n \n The ethsw_add_vlan() function is already called only when the VLAN is\n not yet configured on the switch. Remove the redundant check.\n \n Reported-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Link: https://lore.kernel.org/r/1565700187-16048-8-git-send-email-ioana.ciornei@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 36ec62b7e9e6d2096f01e2df38b309d24ade1abd\nAuthor: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\nDate: Tue Aug 13 15:43:03 2019 +0300\n\n staging: fsl-dpaa2/ethsw: remove unnecessary memset\n \n The ethtool core already zeroes the memory before calling\n .get_ethtool_stats() thus making the memset unnecessary. Remove it.\n \n Reported-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Link: https://lore.kernel.org/r/1565700187-16048-7-git-send-email-ioana.ciornei@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 38ec70cc875f1dc9c6afba47dfdd3bea85c7a9ac\nAuthor: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\nDate: Tue Aug 13 15:43:02 2019 +0300\n\n staging: fsl-dpaa2/ethsw: use bool when encoding learning/flooding state\n \n Use a bool instead of an u8 in ethsw_set_learning() and\n ethsw_port_set_flood() to encode an binary type property.\n \n Reported-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Link: https://lore.kernel.org/r/1565700187-16048-6-git-send-email-ioana.ciornei@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 47980a20a4e257c251eee860e5041f16e2313b26\nAuthor: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\nDate: Tue Aug 13 15:43:01 2019 +0300\n\n staging: fsl-dpaa2/ethsw: remove debug message\n \n Since ethtool will be loud enough if the .set_link_ksettings() callback\n fails, remove the debug messages which do not add additional\n information.\n \n Reported-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Link: https://lore.kernel.org/r/1565700187-16048-5-git-send-email-ioana.ciornei@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 73866324771014136699df73a4cd67e2c47f235c\nAuthor: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\nDate: Tue Aug 13 15:43:00 2019 +0300\n\n staging: fsl-dpaa2/ethsw: add line terminator to all formats\n \n Add the \u0027\\n\u0027 line terminator to the string formats missing it.\n \n Reported-by: Joe Perches \u003cjoe@perches.com\u003e\n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Link: https://lore.kernel.org/r/1565700187-16048-4-git-send-email-ioana.ciornei@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f740eb942b2693bf7516026b4c598cb5e4a26e70\nAuthor: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\nDate: Tue Aug 13 15:42:59 2019 +0300\n\n staging: fsl-dpaa2/ethsw: enable switch ports only on dev_open\n \n At probe time, only the DPSW object should be enabled without the\n associated ports, which will get enabled on dev_open. Remove the\n ethsw_open() and ethsw_stop() functions and replace them only with\n dpsw_enable()/_disable().\n \n Reported-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Link: https://lore.kernel.org/r/1565700187-16048-3-git-send-email-ioana.ciornei@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 9d8cfa9d5346b133487a2466b0ebd405c3eb6d07\nAuthor: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\nDate: Tue Aug 13 15:42:58 2019 +0300\n\n staging: fsl-dpaa2/ethsw: remove IGMP default address\n \n Do not add an IGMP multicast address by default since we do not support\n Rx/Tx ar the moment.\n \n Reported-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Link: https://lore.kernel.org/r/1565700187-16048-2-git-send-email-ioana.ciornei@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 5ad773f9a92f6a0b8d7433921e89b9beb567e5d0\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Tue Aug 13 11:30:46 2019 +0200\n\n drm/bridge: dumb-vga-dac: Fix dereferencing -ENODEV DDC channel\n \n If the VGA connector has no DDC channel, an error pointer will be\n dereferenced, e.g. on Salvator-XS:\n \n Unable to handle kernel NULL pointer dereference at virtual address 000000000000017d\n ...\n Call trace:\n sysfs_do_create_link_sd.isra.0+0x40/0x108\n sysfs_create_link+0x20/0x40\n drm_sysfs_connector_add+0xa8/0xc8\n drm_connector_register.part.3+0x54/0xb0\n drm_connector_register_all+0xb0/0xd0\n drm_modeset_register_all+0x54/0x88\n drm_dev_register+0x18c/0x1d8\n rcar_du_probe+0xe4/0x150\n ...\n \n This happens because vga-\u003eddc either contains a valid DDC channel\n pointer, or -ENODEV, and drm_connector_init_with_ddc() expects a valid\n DDC channel pointer, or NULL.\n \n Fix this by resetting vga-\u003eddc to NULL in case of -ENODEV, and replacing\n the existing error checks by non-NULL checks.\n This is similar to what the HDMI connector driver does.\n \n Fixes: a4f9087e85de141e (\"drm/bridge: dumb-vga-dac: Provide ddc symlink in connector sysfs directory\")\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Tested-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813093046.4976-1-geert+renesas@glider.be\n\ncommit 1cd8fa288eb83c1fe0dfa492b09d228a8d802fbf\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Aug 13 12:07:14 2019 -0300\n\n perf ui: No need to set ui_browser to 1 twice\n \n We need to do it only when fallbacking from GTK to the TUI.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-dda0acxqef1k72n9z4myjbjt@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 0a4d8fb229dd78f9e0752817339e19e903b37a60\nAuthor: Tan Xiaojun \u003ctanxiaojun@huawei.com\u003e\nDate: Fri Aug 2 11:48:57 2019 +0800\n\n perf record: Support aarch64 random socket_id assignment\n \n Same as in the commit 01766229533f (\"perf record: Support s390 random\n socket_id assignment\"), aarch64 also have this problem.\n \n Without this fix:\n \n [root@localhost perf]# ./perf report --header -I -v\n ...\n socket_id number is too big.You may need to upgrade the perf tool.\n \n # \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n # captured on : Thu Aug 1 22:58:38 2019\n # header version : 1\n ...\n # Core ID and Socket ID information is not available\n ...\n \n With this fix:\n [root@localhost perf]# ./perf report --header -I -v\n ...\n cpumask list: 0-31\n cpumask list: 32-63\n cpumask list: 64-95\n cpumask list: 96-127\n \n # \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n # captured on : Thu Aug 1 22:58:38 2019\n # header version : 1\n ...\n # CPU 0: Core ID 0, Socket ID 36\n # CPU 1: Core ID 1, Socket ID 36\n ...\n # CPU 126: Core ID 126, Socket ID 8442\n # CPU 127: Core ID 127, Socket ID 8442\n ...\n \n Signed-off-by: Tan Xiaojun \u003ctanxiaojun@huawei.com\u003e\n Acked-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Kan Liang \u003ckan.liang@linux.intel.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Song Liu \u003csongliubraving@fb.com\u003e\n Cc: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n Cc: Tzvetomir Stoyanov (VMware) \u003ctz.stoyanov@gmail.com\u003e\n Link: http://lkml.kernel.org/r/1564717737-21602-1-git-send-email-tanxiaojun@huawei.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 3143906c2770778d89b730e0342b745d1b4a8303\nAuthor: Vince Weaver \u003cvincent.weaver@maine.edu\u003e\nDate: Thu Aug 1 14:30:43 2019 -0400\n\n perf.data documentation: Clarify HEADER_SAMPLE_TOPOLOGY format\n \n The perf.data file format documentation for HEADER_SAMPLE_TOPOLOGY\n specifies the layout in a confusing manner that doesn\u0027t match the rest\n of the document. This patch attempts to describe things consistent with\n the rest of the file.\n \n Signed-off-by: Vince Weaver \u003cvincent.weaver@maine.edu\u003e\n Acked-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Chong Jiang \u003cchongjiang@chromium.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Simon Que \u003csque@chromium.org\u003e\n Link: http://lkml.kernel.org/r/alpine.DEB.2.21.1908011425240.14303@macbook-air\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 38fe26b46f55538c2cb8b96500caed6ae9d59d46\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Fri Jun 28 20:22:09 2019 +0300\n\n tools: Keep list of tools in alphabetical order\n \n When `make help` is executed it lists the possible tools to build,\n though couple of entries is kept unordered. Fix it here.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Cc: Alexei Starovoitov \u003cast@kernel.org\u003e\n Cc: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n Cc: Martin KaFai Lau \u003ckafai@fb.com\u003e\n Cc: Yonghong Song \u003cyhs@fb.com\u003e\n Link: https://lkml.kernel.org/n/tip-0ke3p64ksa0hnbueh52n3v3q@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit acb9f2d4755a70e31343f99791aa43b05401b996\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Aug 13 11:06:38 2019 -0300\n\n perf evsel: Provide meaningful warning when trying to use \u0027aux_output\u0027 on older kernels\n \n Just like we do with the \u0027write_backwards\u0027 feature:\n \n Before:\n \n # perf record -e {intel_pt/branch\u003d0/,cycles/aux-output/ppp} uname\n Error:\n The sys_perf_event_open() syscall returned with 22 (Invalid argument) for event (cycles/aux-output/ppp).\n /bin/dmesg | grep -i perf may provide additional information.\n \n #\n \n After:\n \n # perf record -e {intel_pt/branch\u003d0/,cycles/aux-output/ppp} uname\n Error:\n The \u0027aux_output\u0027 feature is not supported, update the kernel.\n #\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Kan Liang \u003ckan.liang@linux.intel.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/n/tip-wgjsjroe1e150c0metgwmqwd@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 243384dd25c8ea721c5c82a229eaf33cbd1bfd52\nAuthor: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\nDate: Tue Aug 6 11:46:06 2019 +0300\n\n perf intel-pt: Add brief documentation for PEBS via Intel PT\n \n Document how to select PEBS via Intel PT and how to display synthesized\n PEBS samples.\n \n Signed-off-by: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Kan Liang \u003ckan.liang@linux.intel.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190806084606.4021-8-alexander.shishkin@linux.intel.com\n Signed-off-by: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n [ Update the example to use a group with intel_pt// as the group leader, as per Alex comment ]\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 1b9921546a9641aefc4a52c1c635b96b67142993\nAuthor: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\nDate: Tue Aug 6 11:46:05 2019 +0300\n\n perf tools: Add aux-output config term\n \n Expose the aux_output attribute flag to the user to configure, by adding a\n config term \u0027aux-output\u0027. For events that support it, selection of\n \u0027aux-output\u0027 causes the generation of AUX records instead of event records.\n This requires that an AUX area event is also provided.\n \n Signed-off-by: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Kan Liang \u003ckan.liang@linux.intel.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190806084606.4021-7-alexander.shishkin@linux.intel.com\n Signed-off-by: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 9e64cefe4335b0f2799956d3f3cca8bb652d950f\nAuthor: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\nDate: Tue Aug 6 11:46:04 2019 +0300\n\n perf intel-pt: Process options for PEBS event synthesis\n \n Process synth_opts.other_events and attr.aux_output to set up for\n synthesizing PEBs via Intel PT events.\n \n Signed-off-by: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Kan Liang \u003ckan.liang@linux.intel.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190806084606.4021-6-alexander.shishkin@linux.intel.com\n Signed-off-by: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n [ Fixed up libbperf clashes, i.e. some places using perf_evsel (now in libperf)\n need to use instead \u0027evsel\u0027 (a tools/perf only abstraction) ]\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 181ebb5e23a5e480f6d6aa2816a9c4aaa65afa59\nAuthor: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\nDate: Tue Aug 6 11:46:03 2019 +0300\n\n perf tools: Add itrace option \u0027o\u0027 to synthesize aux-output events\n \n Add itrace option \u0027o\u0027 to synthesize events recorded in the AUX area due\n to the use of perf record\u0027s aux-output config term.\n \n Signed-off-by: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Kan Liang \u003ckan.liang@linux.intel.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190806084606.4021-5-alexander.shishkin@linux.intel.com\n Signed-off-by: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 5a4b58e5d64ac7ebca175ffd8d74ca1b5cb0a01f\nAuthor: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\nDate: Tue Aug 6 11:46:02 2019 +0300\n\n perf tools: Add aux_output attribute flag\n \n Add aux_output attribute flag to match the kernel\u0027s perf_event.h file.\n \n Signed-off-by: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Kan Liang \u003ckan.liang@linux.intel.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190806084606.4021-4-alexander.shishkin@linux.intel.com\n Signed-off-by: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit ce7b0e426ef359ee1d4a6126314ee3547a8eed87\nAuthor: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\nDate: Tue Aug 6 17:41:01 2019 +0300\n\n perf record: Add an option to take an AUX snapshot on exit\n \n It is sometimes useful to generate a snapshot when perf record exits;\n I\u0027ve been using a wrapper script around the workload that would do a\n killall -USR2 perf when the workload exits.\n \n This patch makes it easier and also works when perf record is attached\n to a pre-existing task. A new snapshot option \u0027e\u0027 can be specified in\n -S to enable this behavior:\n \n root@elsewhere:~# perf record -e intel_pt// -Se sleep 1\n [ perf record: Woken up 2 times to write data ]\n [ perf record: Captured and wrote 0.085 MB perf.data ]\n \n Co-developed-by: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Signed-off-by: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190806144101.62892-1-alexander.shishkin@linux.intel.com\n [ Fixed up !HAVE_AUXTRACE_SUPPORT build in builtin-record.c, adding 2 missing __maybe_unused ]\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 73e5de70dca00344cb48e018131a4cadec0fabf0\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Aug 12 17:27:11 2019 -0300\n\n perf ftrace: Improve error message about capability to use ftrace\n \n If we link against libcap, then we can state that CAP_SYS_ADMIN is\n needed, if not, fallback to telling the user it needs to be root, as was\n before linking against libcap.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Igor Lubashev \u003cilubashe@akamai.com\u003e\n Cc: James Morris \u003cjmorris@namei.org\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Suzuki Poulouse \u003csuzuki.poulose@arm.com\u003e\n Link: https://lkml.kernel.org/n/tip-hhnbjdo8r67054of9zm2kxtl@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit c766f3df635de14295e410c6dd5410bc416c24a0\nAuthor: Igor Lubashev \u003cilubashe@akamai.com\u003e\nDate: Wed Aug 7 10:44:17 2019 -0400\n\n perf ftrace: Use CAP_SYS_ADMIN instead of euid\u003d\u003d0\n \n The kernel requires CAP_SYS_ADMIN instead of euid\u003d\u003d0 to mount debugfs\n for ftrace. Make perf do the same.\n \n Signed-off-by: Igor Lubashev \u003cilubashe@akamai.com\u003e\n Acked-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: James Morris \u003cjmorris@namei.org\u003e\n Cc: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Suzuki Poulouse \u003csuzuki.poulose@arm.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Link: http://lkml.kernel.org/r/bd8763b72ed4d58d0b42d44fbc7eb474d32e53a3.1565188228.git.ilubashe@akamai.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 083c1359b0e03867a8c7effd21d4c0be3639f336\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Aug 13 11:38:19 2019 -0300\n\n perf tools: Add CAP_SYSLOG define for older systems\n \n Some of the systems I test don\u0027t have that define, provide it\n conditionally since we\u0027ll use it in the kptr_restrict checks in the next\n patch.\n \n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Igor Lubashev \u003cilubashe@akamai.com\u003e\n Cc: James Morris \u003cjmorris@namei.org\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Suzuki Poulouse \u003csuzuki.poulose@arm.com\u003e\n Link: https://lkml.kernel.org/n/tip-dcize2v6jjab7tds5ngz97dk@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 97993bd6eb89bf08649c01d4d57453feca4314f8\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Aug 12 16:43:08 2019 -0300\n\n perf tools: Add NO_LIBCAP\u003d1 to the minimal build test\n \n We need to add these so that we test building without all selectable\n features.\n \n Acked-by: Igor Lubashev \u003cilubashe@akamai.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: James Morris \u003cjmorris@namei.org\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Suzuki Poulouse \u003csuzuki.poulose@arm.com\u003e\n Link: https://lkml.kernel.org/n/tip-eknnvp22elznj0cl5a39hc4v@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit c22e150e3afa6f8db2300bd510e4ac26bbee1bf3\nAuthor: Igor Lubashev \u003cilubashe@akamai.com\u003e\nDate: Wed Aug 7 10:44:14 2019 -0400\n\n perf tools: Add helpers to use capabilities if present\n \n Add utilities to help checking capabilities of the running procss. Make\n perf link with libcap, if it is available. If no libcap-dev[el],\n fallback to the geteuid() \u003d\u003d 0 test used before.\n \n Committer notes:\n \n $ perf test python\n 18: \u0027import perf\u0027 in python : FAILED!\n $ perf test -v python\n Couldn\u0027t bump rlimit(MEMLOCK), failures may take place when creating BPF maps, etc\n 18: \u0027import perf\u0027 in python :\n --- start ---\n test child forked, pid 23288\n Traceback (most recent call last):\n File \"\u003cstdin\u003e\", line 1, in \u003cmodule\u003e\n ImportError: /tmp/build/perf/python/perf.so: undefined symbol: cap_get_flag\n test child finished with -1\n ---- end ----\n \u0027import perf\u0027 in python: FAILED!\n $\n \n This happens because differently from the perf binary generated with\n this patch applied:\n \n $ ldd /tmp/build/perf/perf | grep libcap\n libcap.so.2 \u003d\u003e /lib64/libcap.so.2 (0x00007f724a4ef000)\n $\n \n The python binding isn\u0027t linking with libcap:\n \n $ ldd /tmp/build/perf/python/perf.so | grep libcap\n $\n \n So add \u0027cap\u0027 to the \u0027extra_libraries\u0027 variable in\n tools/perf/util/setup.py, and rebuild:\n \n $ perf test python\n 18: \u0027import perf\u0027 in python : Ok\n $\n \n If we explicitely disable libcap it also continues to work:\n \n $ make NO_LIBCAP\u003d1 -C tools/perf O\u003d/tmp/build/perf install-bin\n $ ldd /tmp/build/perf/perf | grep libcap\n $ ldd /tmp/build/perf/python/perf.so | grep libcap\n $ perf test python\n 18: \u0027import perf\u0027 in python : Ok\n $\n \n Signed-off-by: Igor Lubashev \u003cilubashe@akamai.com\u003e\n Acked-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: James Morris \u003cjmorris@namei.org\u003e\n Cc: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Suzuki Poulouse \u003csuzuki.poulose@arm.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n [ split from a larger patch ]\n Link: http://lkml.kernel.org/r/8a1e76cf5c7c9796d0d4d240fbaa85305298aafa.1565188228.git.ilubashe@akamai.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit a1f487d75c9be17421f1ab10172427dd16e5f56a\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 18:33:21 2019 +0200\n\n ARM: iop32x: merge everything into mach-iop32x/\n \n Various bits of iop32x are now in their traditional locations in plat-iop,\n mach-iop/include/mach/ and in include/asm/mach/hardware. As nothing\n outside of the iop32x mach code references these any more, this can all\n be moved into one place now.\n \n The only remaining things in the include/mach/ directory are now the\n NR_IRQS definition, the entry-macros.S file and the the decompressor\n uart access. After the irqchip code has been converted to SPARSE_IRQ\n and GENERIC_IRQ_MULTI_HANDLER, it can be moved to ARCH_MULTIPLATFORM.\n \n Link: https://lore.kernel.org/r/20190809163334.489360-7-arnd@arndb.de\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit ba9ef6afc45ff4eb2d9c6e26c6a94279ee3b4348\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 18:33:20 2019 +0200\n\n ARM: iop32x: make mach/uncompress.h independent of mach/hardware.h\n \n All supported uarts use the same address: IQ80321_UART and IQ31244_UART\n are both defined to the default value of 0xfe800000. By using that as\n the address unconditionally, all dependencies on other machine headers\n can be avoided.\n \n Link: https://lore.kernel.org/r/20190809163334.489360-6-arnd@arndb.de\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit c7b68049943079550d4e6af0f10aa3aabd64131a\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 18:33:19 2019 +0200\n\n ARM: xscale: fix multi-cpu compilation\n \n Building a combined ARMv4+XScale kernel produces these\n and other build failures:\n \n /tmp/copypage-xscale-3aa821.s: Assembler messages:\n /tmp/copypage-xscale-3aa821.s:167: Error: selected processor does not support `pld [r7,#0]\u0027 in ARM mode\n /tmp/copypage-xscale-3aa821.s:168: Error: selected processor does not support `pld [r7,#32]\u0027 in ARM mode\n /tmp/copypage-xscale-3aa821.s:169: Error: selected processor does not support `pld [r1,#0]\u0027 in ARM mode\n /tmp/copypage-xscale-3aa821.s:170: Error: selected processor does not support `pld [r1,#32]\u0027 in ARM mode\n /tmp/copypage-xscale-3aa821.s:171: Error: selected processor does not support `pld [r7,#64]\u0027 in ARM mode\n /tmp/copypage-xscale-3aa821.s:176: Error: selected processor does not support `ldrd r4,r5,[r7],#8\u0027 in ARM mode\n /tmp/copypage-xscale-3aa821.s:180: Error: selected processor does not support `strd r4,r5,[r1],#8\u0027 in ARM mode\n \n Add an explict .arch armv5 in the inline assembly to allow the ARMv5\n specific instructions regardless of the compiler -march\u003d target.\n \n Link: https://lore.kernel.org/r/20190809163334.489360-5-arnd@arndb.de\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit aad7ad2a01e774f5cd0dde8f0bc6e253ae3f0d35\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 18:33:18 2019 +0200\n\n dma: iop-adma: allow building without platform headers\n \n Now that iop3xx and iop13xx are gone, the iop-adma driver no\n longer needs to deal with incompatible register layout defined\n in machine specific header files.\n \n Move the iop32x specific definitions into drivers/dma/iop-adma.h\n and the platform_data into include/linux/platform_data/dma-iop32x.h,\n and change the machine code to no longer reference those.\n \n The DMA0_ID/DMA1_ID/AAU_ID macros are required as part of the\n platform data interface and still need to be visible, so move\n those from one header to the other.\n \n Link: https://lore.kernel.org/r/20190809163334.489360-4-arnd@arndb.de\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 00c9755524fbaa28117be774d7c92fddb5ca02f3\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 18:33:17 2019 +0200\n\n dmaengine: iop-adma: use correct printk format strings\n \n When compile-testing on other architectures, we get lots of warnings\n about incorrect format strings, like:\n \n drivers/dma/iop-adma.c: In function \u0027iop_adma_alloc_slots\u0027:\n drivers/dma/iop-adma.c:307:6: warning: format \u0027%x\u0027 expects argument of type \u0027unsigned int\u0027, but argument 6 has type \u0027dma_addr_t {aka long long unsigned int}\u0027 [-Wformat\u003d]\n \n drivers/dma/iop-adma.c: In function \u0027iop_adma_prep_dma_memcpy\u0027:\n \u003e\u003e drivers/dma/iop-adma.c:518:40: warning: format \u0027%u\u0027 expects argument of type \u0027unsigned int\u0027, but argument 5 has type \u0027size_t {aka long unsigned int}\u0027 [-Wformat\u003d]\n \n Use %zu for printing size_t as required, and cast the dma_addr_t\n arguments to \u0027u64\u0027 for printing with %llx. Ideally this should use\n the %pad format string, but that requires an lvalue argument that\n doesn\u0027t work here.\n \n Link: https://lore.kernel.org/r/20190809163334.489360-3-arnd@arndb.de\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Acked-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 7f8bf11a4c2c100b8107922ad8f84d0117dd6cb3\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 18:33:16 2019 +0200\n\n dmaengine: iop-adma: include prefetch.h\n \n Compile-testing this driver fails on m68k without the\n extra header inclusion.\n \n Link: https://lore.kernel.org/r/20190809163334.489360-2-arnd@arndb.de\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Acked-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 59d3ae9a5bf60c037e3a6f6e6bcfbd1c048aa313\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 18:33:15 2019 +0200\n\n ARM: remove Intel iop33x and iop13xx support\n \n There are three families of IOP machines we support in Linux: iop32x\n (which includes EP80219), iop33x and iop13xx (aka IOP34x aka WP8134x).\n \n All products we support in the kernel are based on the first of these,\n iop32x, the other families only ever supported the Intel reference\n boards but no actual machine anyone could ever buy.\n \n While one could clearly make them all three work in a single kernel\n with some work, this takes the easy way out, removing the later two\n platforms entirely, under the assumption that there are no remaining\n users.\n \n Earlier versions of OpenWRT and Debian both had support for iop32x\n but not the others, and they both dropped iop32x as well in their 2015\n releases.\n \n Link: https://lore.kernel.org/r/20190809163334.489360-1-arnd@arndb.de\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Acked-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e # for I2C parts\n Acked-by: Dan Williams \u003cdan.j.williams@intel.com\u003e\n Acked-by: Martin Michlmayr \u003ctbm@cyrius.com\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 8c9e465b488b2b8b83b321045f4c07adbd69d2e2\nAuthor: Marek Behún \u003cmarek.behun@nic.cz\u003e\nDate: Mon Aug 12 18:11:18 2019 +0200\n\n dt-bindings: gpio: Document GPIOs via Moxtet bus\n \n This patch adds documentation of the device tree bindings for GPIOs\n on the devices connected via Moxtet bus.\n \n Link: https://lore.kernel.org/r/20190812161118.21476-6-marek.behun@nic.cz\n Signed-off-by: Marek Behún \u003cmarek.behun@nic.cz\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 856ed97af5cfcfdc73cc4178d9670122bca52f6b\nAuthor: Marek Behún \u003cmarek.behun@nic.cz\u003e\nDate: Mon Aug 12 18:11:17 2019 +0200\n\n drivers: gpio: Add support for GPIOs over Moxtet bus\n \n This adds support for interpreting the input and output bits of one\n device on Moxtet bus as GPIOs.\n This is needed for example by the SFP cage module of Turris Mox.\n \n Link: https://lore.kernel.org/r/20190812161118.21476-5-marek.behun@nic.cz\n Signed-off-by: Marek Behún \u003cmarek.behun@nic.cz\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 2091b417aab195a2c35abe780ad15ade1b0932ac\nAuthor: Marek Behún \u003cmarek.behun@nic.cz\u003e\nDate: Mon Aug 12 18:11:16 2019 +0200\n\n bus: moxtet: Add sysfs and debugfs documentation\n \n Add sysfs ABI documentation for the attribute files module_id and\n module_name\n \n Add debugfs ABI documentation for reading input from the shift registers\n and reading last written output or write output to the shift registers.\n \n Link: https://lore.kernel.org/r/20190812161118.21476-4-marek.behun@nic.cz\n Signed-off-by: Marek Behún \u003cmarek.behun@nic.cz\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 20e6cf7e747815461ac8bcb526f7079ce48866b5\nAuthor: Marek Behún \u003cmarek.behun@nic.cz\u003e\nDate: Mon Aug 12 18:11:15 2019 +0200\n\n dt-bindings: bus: Document moxtet bus binding\n \n This adds device tree binding documentation for the Moxtet bus, a bus\n via which the different modules connected to the Turris Mox router can\n be configured.\n \n Link: https://lore.kernel.org/r/20190812161118.21476-3-marek.behun@nic.cz\n Signed-off-by: Marek Behún \u003cmarek.behun@nic.cz\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 5bc7f990cd98f3f7977d62bfe75c50b4b33d13f6\nAuthor: Marek Behún \u003cmarek.behun@nic.cz\u003e\nDate: Mon Aug 12 18:11:14 2019 +0200\n\n bus: Add support for Moxtet bus\n \n On the Turris Mox router different modules can be connected to the main\n CPU board: currently a module with a SFP cage, a module with MiniPCIe\n connector, a PCIe pass-through MiniPCIe connector module, a 4-port\n switch module, an 8-port switch module, and a 4-port USB3 module.\n \n For example:\n [CPU]-[PCIe-pass-through]-[PCIe]-[8-port switch]-[8-port switch]-[SFP]\n \n Each of this modules has an input and output shift register, and these\n are connected via SPI to the CPU board.\n \n Via SPI we are able to discover which modules are connected, in which\n order, and we can also read some information about the modules (eg.\n their interrupt status), and configure them.\n From each module 8 bits can be read (of which low 4 bits identify the\n module) and 8 bits can be written.\n \n For example from the module with a SFP cage we can read the LOS,\n TX-FAULT and MOD-DEF0 signals, while we can write TX-DISABLE and\n RATE-SELECT signals.\n \n This driver creates a new bus type, called \"moxtet\". For each Mox module\n it finds via SPI, it creates a new device on the moxtet bus so that\n drivers can be written for them.\n \n It also implements a virtual interrupt controller for the modules which\n send their interrupt status over the SPI shift register. These modules\n do this in addition to sending their interrupt status via the shared\n interrupt line. When the shared interrupt is triggered, we read from the\n shift register and handle IRQs for all devices which are in interrupt.\n \n The topology of how Mox modules are connected can then be read by\n listing /sys/bus/moxtet/devices.\n \n Link: https://lore.kernel.org/r/20190812161118.21476-2-marek.behun@nic.cz\n Signed-off-by: Marek Behún \u003cmarek.behun@nic.cz\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 232219b9a464c2479c98aa589acb1bd3383ae9d6\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Tue Aug 13 12:03:01 2019 +0200\n\n i2c-cht-wc: Fix lockdep warning\n \n When the kernel is build with lockdep support and the i2c-cht-wc driver is\n used, the following warning is shown:\n \n [ 66.674334] \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n [ 66.674337] WARNING: possible circular locking dependency detected\n [ 66.674340] 5.3.0-rc4+ #83 Not tainted\n [ 66.674342] ------------------------------------------------------\n [ 66.674345] systemd-udevd/1232 is trying to acquire lock:\n [ 66.674349] 00000000a74dab07 (intel_soc_pmic_chtwc:167:(\u0026cht_wc_regmap_cfg)-\u003elock){+.+.}, at: regmap_write+0x31/0x70\n [ 66.674360]\n but task is already holding lock:\n [ 66.674362] 00000000d44a85b7 (i2c_register_adapter){+.+.}, at: i2c_smbus_xfer+0x49/0xf0\n [ 66.674370]\n which lock already depends on the new lock.\n \n [ 66.674371]\n the existing dependency chain (in reverse order) is:\n [ 66.674374]\n -\u003e #1 (i2c_register_adapter){+.+.}:\n [ 66.674381] rt_mutex_lock_nested+0x46/0x60\n [ 66.674384] i2c_smbus_xfer+0x49/0xf0\n [ 66.674387] i2c_smbus_read_byte_data+0x45/0x70\n [ 66.674391] cht_wc_byte_reg_read+0x35/0x50\n [ 66.674394] _regmap_read+0x63/0x1a0\n [ 66.674396] _regmap_update_bits+0xa8/0xe0\n [ 66.674399] regmap_update_bits_base+0x63/0xa0\n [ 66.674403] regmap_irq_update_bits.isra.0+0x3b/0x50\n [ 66.674406] regmap_add_irq_chip+0x592/0x7a0\n [ 66.674409] devm_regmap_add_irq_chip+0x89/0xed\n [ 66.674412] cht_wc_probe+0x102/0x158\n [ 66.674415] i2c_device_probe+0x95/0x250\n [ 66.674419] really_probe+0xf3/0x380\n [ 66.674422] driver_probe_device+0x59/0xd0\n [ 66.674425] device_driver_attach+0x53/0x60\n [ 66.674428] __driver_attach+0x92/0x150\n [ 66.674431] bus_for_each_dev+0x7d/0xc0\n [ 66.674434] bus_add_driver+0x14d/0x1f0\n [ 66.674437] driver_register+0x6d/0xb0\n [ 66.674440] i2c_register_driver+0x45/0x80\n [ 66.674445] do_one_initcall+0x60/0x2f4\n [ 66.674450] kernel_init_freeable+0x20d/0x2b4\n [ 66.674453] kernel_init+0xa/0x10c\n [ 66.674457] ret_from_fork+0x3a/0x50\n [ 66.674459]\n -\u003e #0 (intel_soc_pmic_chtwc:167:(\u0026cht_wc_regmap_cfg)-\u003elock){+.+.}:\n [ 66.674465] __lock_acquire+0xe07/0x1930\n [ 66.674468] lock_acquire+0x9d/0x1a0\n [ 66.674472] __mutex_lock+0xa8/0x9a0\n [ 66.674474] regmap_write+0x31/0x70\n [ 66.674480] cht_wc_i2c_adap_smbus_xfer+0x72/0x240 [i2c_cht_wc]\n [ 66.674483] __i2c_smbus_xfer+0x1a3/0x640\n [ 66.674486] i2c_smbus_xfer+0x67/0xf0\n [ 66.674489] i2c_smbus_read_byte_data+0x45/0x70\n [ 66.674494] bq24190_probe+0x26b/0x410 [bq24190_charger]\n [ 66.674497] i2c_device_probe+0x189/0x250\n [ 66.674500] really_probe+0xf3/0x380\n [ 66.674503] driver_probe_device+0x59/0xd0\n [ 66.674506] device_driver_attach+0x53/0x60\n [ 66.674509] __driver_attach+0x92/0x150\n [ 66.674512] bus_for_each_dev+0x7d/0xc0\n [ 66.674515] bus_add_driver+0x14d/0x1f0\n [ 66.674518] driver_register+0x6d/0xb0\n [ 66.674521] i2c_register_driver+0x45/0x80\n [ 66.674524] do_one_initcall+0x60/0x2f4\n [ 66.674528] do_init_module+0x5c/0x230\n [ 66.674531] load_module+0x2707/0x2a20\n [ 66.674534] __do_sys_init_module+0x188/0x1b0\n [ 66.674537] do_syscall_64+0x5c/0xb0\n [ 66.674541] entry_SYSCALL_64_after_hwframe+0x49/0xbe\n [ 66.674543]\n other info that might help us debug this:\n \n [ 66.674545] Possible unsafe locking scenario:\n \n [ 66.674547] CPU0 CPU1\n [ 66.674548] ---- ----\n [ 66.674550] lock(i2c_register_adapter);\n [ 66.674553] lock(intel_soc_pmic_chtwc:167:(\u0026cht_wc_regmap_cfg)-\u003elock);\n [ 66.674556] lock(i2c_register_adapter);\n [ 66.674559] lock(intel_soc_pmic_chtwc:167:(\u0026cht_wc_regmap_cfg)-\u003elock);\n [ 66.674561]\n *** DEADLOCK ***\n \n The problem is that the CHT Whiskey Cove PMIC\u0027s builtin i2c-adapter is\n itself a part of an i2c-client (the PMIC). This means that transfers done\n through it take adapter-\u003ebus_lock twice, once for the parent i2c-adapter\n and once for its own bus_lock. Lockdep does not like this nested locking.\n \n To make lockdep happy in the case of busses with muxes, the i2c-core\u0027s\n i2c_adapter_lock_bus function calls:\n \n rt_mutex_lock_nested(\u0026adapter-\u003ebus_lock, i2c_adapter_depth(adapter));\n \n But i2c_adapter_depth only works when the direct parent of the adapter is\n another adapter, as it is only meant for muxes. In this case there is an\n i2c-client and MFD instantiated platform_device in the parent-\u003echild chain\n between the 2 devices.\n \n This commit overrides the default i2c_lock_operations, passing a hardcoded\n depth of 1 to rt_mutex_lock_nested, making lockdep happy.\n \n Note that if there were to be a mux attached to the i2c-wc-cht adapter,\n this would break things again since the i2c-mux code expects the\n root-adapter to have a locking depth of 0. But the i2c-wc-cht adapter\n always has only 1 client directly attached in the form of the charger IC\n paired with the CHT Whiskey Cove PMIC.\n \n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit 2e925d23150186962497f6b91f4a6ba6c7c42b16\nAuthor: Patrice Chotard \u003cpatrice.chotard@st.com\u003e\nDate: Wed Jul 31 09:22:04 2019 +0200\n\n ARM: multi_v7_defconfig: Enable SPI_STM32_QSPI support\n \n Enable support for QSPI block on STM32 SoCs.\n \n Link: https://lore.kernel.org/r/20190731072204.32160-1-patrice.chotard@st.com\n Signed-off-by: Patrice Chotard \u003cpatrice.chotard@st.com\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit f63cf88fd88b9d01063338d1f05381800660952e\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 22:27:44 2019 +0200\n\n ARM: remove w90x900 platform\n \n This removes the old Winbond w90x900 platform, also known\n as Nuvoton NUC900. Wan Zongshun originally contributed\n the port and maintained it since then.\n \n From all I can tell, this platform is no longer being used\n with modern kernels, based on various indications:\n \n - The supported chips (nuc910/950/960) are no longer marketed\n by the manufacturer\n \n - Newer chips from the same family (nuc97x, nuc980, n329x)\n that are still marketed have Linux BSPs but those were never\n submitted for upstream inclusion.\n \n - The last patch from the platform maintainer was in 2011.\n \n - All patches to w90x900 platform specific files afterwards\n are cleanups that were apparently done without access to\n test hardware.\n \n - Both the website and the email address listed in the\n MAINTAINERS have become unreachable.\n \n Link: https://lore.kernel.org/r/20190809202749.742267-17-arnd@arndb.de\n Cc: \"Wanzongshun (Vincent)\" \u003cwanzongshun@huawei.com\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit c68b26697d2744d32df621e0ba9a17094bb37d6b\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 22:27:29 2019 +0200\n\n ARM: remove ks8695 platform\n \n ks8695 is an older SoC originally made by Kendin, which was later acquired\n by Micrel, and subsequently by Microchip.\n \n The platform port was originally contributed by Andrew Victor and Ben\n Dooks, and later maintained by Greg Ungerer.\n \n When I recently submitted cleanups, but Greg noted that the platform no\n longer boots and nobody is using it any more, we decided to remove it.\n \n Link: https://lore.kernel.org/r/20190809202749.742267-2-arnd@arndb.de\n Cc: Andrew Victor \u003clinux@maxim.org.za\u003e\n Acked-by: Ben Dooks \u003cben-linux@fluff.org\u003e\n Link: https://wikidevi.com/wiki/Micrel\n Link: https://lore.kernel.org/linux-arm-kernel/2bc41895-d4f9-896c-0726-0b2862fcbf25@kernel.org/\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Acked-by: Greg Ungerer \u003cgerg@kernel.org\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit fc1ca6e01d0a71fd4bd7a5a7cb8309a9dd48ddbc\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Mon Aug 12 14:50:16 2019 +0200\n\n drm/bridge: dw-hdmi-i2s: add .get_eld support\n \n Provide the eld to the generic hdmi-codec driver.\n This will let the driver enforce the maximum channel number and set the\n channel allocation depending on the hdmi sink.\n \n Cc: Jonas Karlman \u003cjonas@kwiboo.se\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Reviewed-by: Jonas Karlman \u003cjonas@kwiboo.se\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812125016.20169-1-jbrunet@baylibre.com\n\ncommit 43e88f670a5e4618609822853a5f659631c32505\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Mon Aug 12 14:07:25 2019 +0200\n\n drm/bridge: dw-hdmi-i2s: enable only the required i2s lanes\n \n Enable the i2s lanes depending on the number of channel in the stream\n \n Reviewed-by: Jonas Karlman \u003cjonas@kwiboo.se\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812120726.1528-8-jbrunet@baylibre.com\n\ncommit 46cecde310bba81243f906955f5fd6f64d5668f0\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Mon Aug 12 14:07:24 2019 +0200\n\n drm/bridge: dw-hdmi-i2s: reset audio fifo before applying new params\n \n When changing the audio hw params, reset the audio fifo to make sure\n any old remaining data is flushed.\n \n The databook mentions that such reset should be followed by a reset of\n the i2s block to make sure the samples stay aligned\n \n Reviewed-by: Jonas Karlman \u003cjonas@kwiboo.se\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812120726.1528-7-jbrunet@baylibre.com\n\ncommit 0c6098859176ffa250b196498722dc7844e41048\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Mon Aug 12 14:07:23 2019 +0200\n\n drm/bridge: dw-hdmi-i2s: set the channel allocation\n \n setup the channel allocation provided by the generic hdmi-codec driver\n \n Reviewed-by: Jonas Karlman \u003cjonas@kwiboo.se\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812120726.1528-6-jbrunet@baylibre.com\n\ncommit 747bee3574040246be18e2e9db709a7cb058b35b\nAuthor: Simon Horman \u003chorms+renesas@verge.net.au\u003e\nDate: Fri Aug 9 14:30:04 2019 -0700\n\n dt-bindings: i2c: i2c-emev2: Rename bindings documentation file\n \n Rename the bindings documentation file for Renesas EMEV2 IIC controller\n from i2c-emev2.txt to renesas,iic-emev2.txt.\n \n This is part of an ongoing effort to name bindings documentation files for\n Renesas IP blocks consistently, in line with the compat strings they\n document.\n \n Signed-off-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit 17a1e555b608a847c903e093718d0a768843c586\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Mon Aug 12 14:07:22 2019 +0200\n\n drm/bridge: dw-hdmi-i2s: enable lpcm multi channels\n \n Properly setup the channel count and layout in dw-hdmi i2s driver so\n we are not limited to 2 channels.\n \n Also correct the maximum channel reported by the DAI from 6 to 8 ch\n \n Reviewed-by: Jonas Karlman \u003cjonas@kwiboo.se\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812120726.1528-5-jbrunet@baylibre.com\n\ncommit 4e4521f76ff96f83561555f76c50d25314320d10\nAuthor: Simon Horman \u003chorms+renesas@verge.net.au\u003e\nDate: Fri Aug 9 14:30:02 2019 -0700\n\n dt-bindings: i2c: rcar: Rename bindings documentation file\n \n Rename the bindings documentation file for R-Car I2C controller\n from i2c-rcar.txt to renesas,i2c.txt.\n \n This is part of an ongoing effort to name bindings documentation files for\n Renesas IP blocks consistently, in line with the compat strings they\n document.\n \n Signed-off-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit da5f5bc92f49f5b3acf8b07d95fb7d8a8f098d25\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Mon Aug 12 14:07:21 2019 +0200\n\n drm/bridge: dw-hdmi: set channel count in the infoframes\n \n Set the number of channel in the infoframes\n \n Reviewed-by: Jonas Karlman \u003cjonas@kwiboo.se\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812120726.1528-4-jbrunet@baylibre.com\n\ncommit 2a2a3d2ff799d62f25a9d222ad0fc73753c8a6c6\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Mon Aug 12 14:07:20 2019 +0200\n\n drm/bridge: dw-hdmi: move audio channel setup out of ahb\n \n Part of the channel count setup done in dw-hdmi ahb should\n actually be done whatever the interface providing the data.\n \n Let\u0027s move it to dw-hdmi driver instead.\n \n Reviewed-by: Jonas Karlman \u003cjonas@kwiboo.se\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812120726.1528-3-jbrunet@baylibre.com\n\ncommit 8067f62bccaff4d5c7e0900431e8ab4372dcb8ab\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Mon Aug 12 14:07:19 2019 +0200\n\n drm/bridge: dw-hdmi-i2s: support more i2s format\n \n The dw-hdmi-i2s supports more formats than just regular i2s.\n Add support for left justified, right justified and dsp modes\n A and B.\n \n Reviewed-by: Jonas Karlman \u003cjonas@kwiboo.se\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812120726.1528-2-jbrunet@baylibre.com\n\ncommit 72362da87348c756361dd6c9e8554fb72ef67746\nAuthor: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\nDate: Wed Aug 14 13:52:39 2019 +0100\n\n ARM: imx_v6_v7_defconfig: Enable the PSCI CPUidle driver\n \n Enable the PSCI CPUidle driver to replace the functionality\n previously provided by the generic ARM CPUidle driver.\n \n Link: https://lore.kernel.org/r/20190814125239.6270-3-lorenzo.pieralisi@arm.com\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Cc: Shawn Guo \u003cshawnguo@kernel.org\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit a227229f62051dc9be0170e6805bd913b3d230d0\nAuthor: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\nDate: Wed Aug 14 13:52:38 2019 +0100\n\n arm64: defconfig: Enable the PSCI CPUidle driver\n \n Enable the PSCI CPUidle driver to replace the functionality\n previously provided by the generic ARM CPUidle driver through\n CPU operations.\n \n Link: https://lore.kernel.org/r/20190814125239.6270-2-lorenzo.pieralisi@arm.com\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Cc: Will Deacon \u003cwill@kernel.org\u003e\n Cc: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Cc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit af80559b4d9cd481c63505edbe425ff6bad4ab8d\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Fri Aug 9 17:40:47 2019 +0200\n\n i2c: replace i2c_new_secondary_device with an ERR_PTR variant\n \n In the general move to have i2c_new_*_device functions which return\n ERR_PTR instead of NULL, this patch converts i2c_new_secondary_device().\n \n There are only few users, so this patch converts the I2C core and all\n users in one go. The function gets renamed to i2c_new_ancillary_device()\n so out-of-tree users will get a build failure to understand they need to\n adapt their error checking code.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Reviewed-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e # adv748x\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e # adv7511 + adv7604\n Reviewed-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e # adv7604\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit d0b3c32ed9220616548ff63808751cf2f6608df1\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Aug 13 17:22:51 2019 +0100\n\n arm64: memory: Cosmetic cleanups\n \n Cleanup memory.h so that the indentation is consistent, remove pointless\n line-wrapping and use consistent parameter names for different versions\n of the same macro.\n \n Reviewed-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Acked-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Reviewed-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 68933aa973740796895e297e7dbf7baf3e9c51b1\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Aug 13 17:06:29 2019 +0100\n\n arm64: memory: Add comments to end of non-trivial #ifdef blocks\n \n Commenting the #endif of a multi-statement #ifdef block with the\n condition which guards it is useful and can save having to scroll back\n through the file to figure out which set of Kconfig options apply to\n a particular piece of code.\n \n Reviewed-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Acked-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Reviewed-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 6bbd497f027332b14cf2a6792c418c32286b66c2\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Aug 13 17:01:05 2019 +0100\n\n arm64: memory: Implement __tag_set() as common function\n \n There\u0027s no need for __tag_set() to be a complicated macro when\n CONFIG_KASAN_SW_TAGS\u003dy and a simple static inline otherwise. Rewrite\n the thing as a common static inline function.\n \n Tested-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Reviewed-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Tested-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Reviewed-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit a5ac40f53bfa5e43d9c76e3c23415ee4afd71932\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Aug 13 16:58:36 2019 +0100\n\n arm64: memory: Simplify _VA_START and _PAGE_OFFSET definitions\n \n Rather than subtracting from -1 and then adding 1, we can simply\n subtract from 0.\n \n Tested-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Reviewed-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Tested-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 9ba33dcc6bef4e56c762b446f0000f27ee737b8b\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Aug 13 16:34:32 2019 +0100\n\n arm64: memory: Simplify virt_to_page() implementation\n \n Build virt_to_page() on top of virt_to_pfn() so we can avoid the need\n for explicit shifting.\n \n Tested-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Reviewed-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Tested-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Reviewed-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 96628f0fb18080a4166fc9eab8f7fd062d860667\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Aug 13 16:46:11 2019 +0100\n\n arm64: memory: Rewrite default page_to_virt()/virt_to_page()\n \n The default implementations of page_to_virt() and virt_to_page() are\n fairly confusing to read and the former evaluates its \u0027page\u0027 parameter\n twice in the macro\n \n Rewrite them so that the computation is expressed as \u0027base + index\u0027 in\n both cases and the parameter is always evaluated exactly once.\n \n Tested-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Reviewed-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Tested-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 577c2b35283fbadcc9ce4b56304ccea3ec8a5ca1\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Aug 13 16:26:54 2019 +0100\n\n arm64: memory: Ensure address tag is masked in conversion macros\n \n When converting a linear virtual address to a physical address, pfn or\n struct page *, we must make sure that the tag bits are masked before the\n calculation otherwise we end up with corrupt pointers when running with\n CONFIG_KASAN_SW_TAGS\u003dy:\n \n | Unable to handle kernel paging request at virtual address 0037fe0007580d08\n | [0037fe0007580d08] address between user and kernel address ranges\n \n Mask out the tag in __virt_to_phys_nodebug() and virt_to_page().\n \n Reported-by: Qian Cai \u003ccai@lca.pw\u003e\n Reported-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n Tested-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Reviewed-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Tested-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Reviewed-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Fixes: 9cb1c5ddd2c4 (\"arm64: mm: Remove bit-masking optimisations for PAGE_OFFSET and VMEMMAP_START\")\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 68dd8ef321626f14ae9ef2039b7a03c707149489\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Aug 13 15:52:23 2019 +0100\n\n arm64: memory: Fix virt_addr_valid() using __is_lm_address()\n \n virt_addr_valid() is intended to test whether or not the passed address\n is a valid linear map address. Unfortunately, it relies on\n _virt_addr_is_linear() which is broken because it assumes the linear\n map is at the top of the address space, which it no longer is.\n \n Reimplement virt_addr_valid() using __is_lm_address() and remove\n _virt_addr_is_linear() entirely. At the same time, ensure we evaluate\n the macro parameter only once and move it within the __ASSEMBLY__ block.\n \n Reported-by: Qian Cai \u003ccai@lca.pw\u003e\n Reported-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n Tested-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Reviewed-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Tested-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Reviewed-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Fixes: 14c127c957c1 (\"arm64: mm: Flip kernel VA space\")\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 759b2e69238dff3f072e1c8e231aecf662082b6c\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Tue Aug 13 13:48:38 2019 +0100\n\n staging: rtl8723bs: remove redundant assignment to ret\n \n Variable ret is initialized to a value that is never read and it is\n re-assigned later. The initialization is redundant and can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190813124838.1317-1-colin.king@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b1198cca2982f02c27459cfd67d081d7c1460cf9\nAuthor: Merwin Trever Ferrao \u003cmerwintf@gmail.com\u003e\nDate: Tue Aug 13 12:28:06 2019 +0530\n\n Staging: rtl8712: rtl8712_recv: fixed 80 character length warning\n \n When the checkpatch.pl script was run, it showed lines with length\n more than 80 characters in rtl8712_recv.c file. Fixed by\n breaking it up into two lines within 80 characters.\n \n Signed-off-by: Merwin Trever Ferrao \u003cmerwintf@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190813065806.GA23606@IoT-COE\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit ca8a0a4934e0d850c978d17371d2c8acd3ec5ff2\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Aug 13 10:16:38 2019 +0530\n\n staging: rtl8712: r8712_dump_aggr_xframe(): Change type to void\n \n Change return type of r8712_dump_aggr_xframe from u8 to void as it\n always returns _SUCCESS and its return value is never used.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190813044638.16348-4-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b849e9a4a09f349e37ab936cc082781098b3f48c\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Aug 13 10:16:37 2019 +0530\n\n staging: rtl8712: r8712_xmitframe_aggr_1st(): Change return type to void\n \n Change return type of r8712_xmitframe_aggr_1st from u8 to void as it\n always returns _SUCCESS and its return value is never used.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190813044638.16348-3-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit bc6132f141b75ea1ee67574bf44adfd0433cad17\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Aug 13 10:16:36 2019 +0530\n\n staging: rtl8712: r8712_append_mpdu_unit(): Change return type\n \n Change return type of r8712_append_mpdu_unit from u8 to void and remove\n its return statement as it always returns only _SUCCESS. Modify call\n sites to simply call this function instead of checking its return value,\n and execute all the statements in the if-block for when the function\n returns _SUCCESS.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190813044638.16348-2-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 6da2422efbb7f7e11b7f81dcf62817891b311b85\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Aug 13 10:16:35 2019 +0530\n\n staging: rtl8712: xmitframe_addmic(): Change return values and type\n \n Change return values of xmitframe_addmic from _SUCCESS and _FAIL to 0\n and -ENOMEM respectively. Modify call sites to check for non-zero values\n instead of _FAIL.\n Also change return type from sint to int.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190813044638.16348-1-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 737d4d9ef6261ffa361fa708fac6047ac720688a\nAuthor: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\nDate: Tue Aug 13 12:24:26 2019 +0800\n\n staging: rtl8723bs: Remove debugging information exposed via procfs\n \n The procfs provides many useful information for debugging, but it may be\n too much for normal usage, routines like proc_get_sec_info() reports\n various security related information.\n \n So let\u0027s remove it.\n \n Signed-off-by: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190813042426.13733-1-kai.heng.feng@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 9ddc7004f6aae68d8c3eee81b4b2a734d25c92e1\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Tue Aug 13 10:30:54 2019 +0800\n\n staging: erofs: xattr.c: avoid BUG_ON\n \n Kill all the remaining BUG_ON in EROFS:\n - one BUG_ON was used to detect xattr on-disk corruption,\n proper error handling should be added for it instead;\n - the other BUG_ONs are used to detect potential issues,\n use DBG_BUGON only in (eng) debugging version.\n \n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190813023054.73126-3-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 753a3c80b6afe55012ab49b076d30bfaf0266c46\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Tue Aug 13 10:30:53 2019 +0800\n\n staging: erofs: remove incomplete cleancache\n \n cleancache was not fully implemented in EROFS.\n In addition, it\u0027s tend to remove the whole cleancache in\n related attempt [1].\n \n [1] https://lore.kernel.org/linux-fsdevel/20190527103207.13287-3-jgross@suse.com/\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190813023054.73126-2-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 2a5f805f30cba3c958511978589f7f79dcfbec6b\nAuthor: Adrian Reber \u003careber@redhat.com\u003e\nDate: Mon Aug 12 22:09:39 2019 +0200\n\n selftests: add tests for clone3()\n \n This tests clone3() with and without set_tid to see if all desired PIDs\n are working as expected. The test tries to clone3() with a set_tid of\n -1, 1, pid_max, a PID which is already in use and an unused PID. The\n same tests are also running in PID namespace.\n \n In addition the clone3 test (without set_tid) tries to call clone3()\n with different sizes of clone_args.\n \n Signed-off-by: Adrian Reber \u003careber@redhat.com\u003e\n Link: https://lore.kernel.org/r/20190812200939.23784-2-areber@redhat.com\n Signed-off-by: Christian Brauner \u003cchristian.brauner@ubuntu.com\u003e\n\ncommit a0611fdee2b959f60ca86b9d83f697249c411d51\nAuthor: Adrian Reber \u003careber@redhat.com\u003e\nDate: Mon Aug 12 22:09:38 2019 +0200\n\n fork: extend clone3() to support setting a PID\n \n The main motivation to add set_tid to clone3() is CRIU.\n \n To restore a process with the same PID/TID CRIU currently uses\n /proc/sys/kernel/ns_last_pid. It writes the desired (PID - 1) to\n ns_last_pid and then (quickly) does a clone(). This works most of the\n time, but it is racy. It is also slow as it requires multiple syscalls.\n \n Extending clone3() to support set_tid makes it possible restore a\n process using CRIU without accessing /proc/sys/kernel/ns_last_pid and\n race free (as long as the desired PID/TID is available).\n \n This clone3() extension places the same restrictions (CAP_SYS_ADMIN)\n on clone3() with set_tid as they are currently in place for ns_last_pid.\n \n Signed-off-by: Adrian Reber \u003careber@redhat.com\u003e\n Reviewed-by: Christian Brauner \u003cchristian.brauner@ubuntu.com\u003e\n Reviewed-by: Oleg Nesterov \u003coleg@redhat.com\u003e\n Acked-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\n Link: https://lore.kernel.org/r/20190812200939.23784-1-areber@redhat.com\n Signed-off-by: Christian Brauner \u003cchristian.brauner@ubuntu.com\u003e\n\ncommit b5b24617987f522abb5c42851ae505557f2c84f8\nAuthor: Xiaowei Bao \u003cxiaowei.bao@nxp.com\u003e\nDate: Wed Aug 14 10:03:30 2019 +0800\n\n PCI: layerscape: Add CONFIG_PCI_LAYERSCAPE_EP to build EP/RC separately\n \n Add CONFIG_PCI_LAYERSCAPE_EP so that endpoint and host controller\n drivers can be built separately.\n \n Signed-off-by: Xiaowei Bao \u003cxiaowei.bao@nxp.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n\ncommit c0900eb8264c989878e475eb02a0d19eaeb62682\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Thu Aug 8 10:55:22 2019 +0200\n\n MAINTAINERS: Update with Amlogic DRM bindings converted as YAML\n \n The amlogic,meson-dw-hdmi.txt and amlogic,meson-vpu.txt has been\n converted to YAML schemas, update MAINTAINERS to match them again.\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Maxime Jourdan \u003cmjourdan@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808085522.21950-4-narmstrong@baylibre.com\n\ncommit 6b9ebf1e0e678b0053cedc2fc7cef5308fe464e4\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Thu Aug 8 10:55:21 2019 +0200\n\n dt-bindings: display: amlogic, meson-vpu: convert to yaml\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for the Amlogic Display Controller over to YAML schemas.\n \n The original example has a leftover \"dmc\" memory cell, that has been\n removed in the yaml rewrite.\n \n The port connection table has been dropped in favor of a description\n of each port.\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808085522.21950-3-narmstrong@baylibre.com\n\ncommit 376bf52deef5e47909f08acdf4456ceccc01edd5\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Thu Aug 8 10:55:20 2019 +0200\n\n dt-bindings: display: amlogic, meson-dw-hdmi: convert to yaml\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for the Amlogic Synopsys DW-HDMI specifics over to YAML schemas.\n \n The original example and usage of clock-names uses a reversed \"isfr\"\n and \"iahb\" clock-names, the rewritten YAML bindings uses the reversed\n instead of fixing the device trees order.\n \n The #sound-dai-cells optional property has been added to match this node\n as a sound dai.\n \n The port connection table has been dropped in favor of a description\n of each port.\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808085522.21950-2-narmstrong@baylibre.com\n\ncommit fd5d16531a39322c3d7433d9f8a36203c9aaeddc\nAuthor: Xiaowei Bao \u003cxiaowei.bao@nxp.com\u003e\nDate: Wed Aug 14 10:03:29 2019 +0800\n\n PCI: layerscape: Add the bar_fixed_64bit property to the endpoint driver\n \n The layerscape PCIe controller have 4 BARs.\n \n BAR0 and BAR1 are 32bit, BAR2 and BAR4 are 64bit and that\u0027s a\n fixed hardware configuration.\n \n Set the bar_fixed_64bit variable accordingly.\n \n Signed-off-by: Xiaowei Bao \u003cxiaowei.bao@nxp.com\u003e\n [lorenzo.pieralisi@arm.com: commit log]\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Acked-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit 3d2969fac698ef6694e653c192d26ea501b7c3d5\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Tue Aug 13 10:30:52 2019 +0800\n\n staging: erofs: inline erofs_inode_is_data_compressed()\n \n As a helper in erofs_fs.h, erofs_inode_is_data_compressed()\n should be inlined.\n \n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190813023054.73126-1-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f130b712ff259f03feaf7a245e3464e3faad89ae\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Aug 13 21:09:05 2019 +0100\n\n drm/i915: Serialise read/write of the barrier\u0027s engine\n \n We use the request pointer inside the i915_active_node as the indicator\n of the barrier\u0027s status; we mark it as used during\n i915_request_add_active_barriers(), and search for an available barrier\n in reuse_idle_barrier(). That check must be carefully serialised to\n ensure we do use an engine for the barrier and not just a random\n pointer. (Along the other reuse path, we are fully serialised by the\n timeline-\u003emutex.) The acquisition of the barrier itself is ordered through\n the strong memory barrier in llist_del_all().\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111397\n Fixes: d8af05ff38ae (\"drm/i915: Allow sharing the idle-barrier from other kernel requests\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813200905.11369-1-chris@chris-wilson.co.uk\n\ncommit b26496ae786d5f352a546ebdd4e54dc416e74f6e\nAuthor: Stuart Summers \u003cstuart.summers@intel.com\u003e\nDate: Tue Aug 13 10:41:20 2019 -0700\n\n drm/i915: Print CCID for all renderCS\n \n Use render class instead of RCS0 when printing CCID.\n \n Signed-off-by: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813174121.129593-2-stuart.summers@intel.com\n\ncommit 7991eb39eedc2f8565568f02877f7e22c4f89f98\nAuthor: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\nDate: Mon Aug 12 12:32:46 2019 +0200\n\n iommu/exynos: Remove __init annotation from exynos_sysmmu_probe()\n \n Exynos SYSMMU driver supports deferred probe. It happens when clocks\n needed for this driver are not yet available. Typically next calls to\n driver -\u003eprobe() happen before init section is free, but this is not\n really guaranteed. To make if safe, remove __init annotation from\n exynos_sysmmu_probe() function.\n \n Signed-off-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 5f889b9a61dd7069e61d0328bc514f0f508328c1\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Aug 13 19:21:12 2019 +0100\n\n drm/i915: Disregard drm_mode_config.fb_base\n \n The fb_base is only used for communicating the GTT BAR from one piece of\n the display code (kms setup) to another (fbdev). What is required in the\n fbdev is just the aperture address which should be derived from the\n bo we allocate for the framebuffer directly.\n \n The same appears true for drm/; it is not used by the core or the uAPI,\n it is merely for conveniently passing a device address from bit of\n display management code to another.\n \n v2: Note that since we only expose enough of a system map to cover our\n single framebuffer, the screen_base/size and the smem are one and the\n same.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Reviewed-by: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813182112.23227-1-chris@chris-wilson.co.uk\n\ncommit a4eb99a1d617542ece50b04070b944bb665b8cf1\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Aug 13 22:57:07 2019 +0100\n\n drm/i915: Include engine-\u003emmio_base in the debug dump\n \n Some IGT would like to know the mmio address of each engine so make it\n available.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813215707.14703-1-chris@chris-wilson.co.uk\n\ncommit b20a6e298bcb8cb8ae18de26baaf462a6418515b\nAuthor: Darius Rad \u003calpha@area49.net\u003e\nDate: Tue Jul 23 13:37:46 2019 -0300\n\n media: rc: imon: Allow iMON RC protocol for ffdc 7e device\n \n Allow selecting the IR protocol, MCE or iMON, for a device that\n identifies as follows (with config id 0x7e):\n \n 15c2:ffdc SoundGraph Inc. iMON PAD Remote Controller\n \n As the driver is structured to default to iMON when both RC\n protocols are supported, existing users of this device (using MCE\n protocol) will need to manually switch to MCE (RC-6) protocol from\n userspace (with ir-keytable, sysfs).\n \n Signed-off-by: Darius Rad \u003calpha@area49.net\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 699bf94114151aae4dceb2d9dbf1a6312839dcae\nAuthor: Sean Young \u003csean@mess.org\u003e\nDate: Tue Aug 13 13:45:09 2019 -0300\n\n media: tm6000: double free if usb disconnect while streaming\n \n The usb_bulk_urb will kfree\u0027d on disconnect, so ensure the pointer is set\n to NULL after each free.\n \n stop stream\n urb killing\n urb buffer free\n tm6000: got start feed request tm6000_start_feed\n tm6000: got start stream request tm6000_start_stream\n tm6000: pipe reset\n tm6000: got start feed request tm6000_start_feed\n tm6000: got start feed request tm6000_start_feed\n tm6000: got start feed request tm6000_start_feed\n tm6000: got start feed request tm6000_start_feed\n tm6000: IR URB failure: status: -71, length 0\n xhci_hcd 0000:00:14.0: ERROR unknown event type 37\n xhci_hcd 0000:00:14.0: ERROR unknown event type 37\n tm6000: error tm6000_urb_received\n usb 1-2: USB disconnect, device number 5\n tm6000: disconnecting tm6000 #0\n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n BUG: KASAN: use-after-free in dvb_fini+0x75/0x140 [tm6000_dvb]\n Read of size 8 at addr ffff888241044060 by task kworker/2:0/22\n \n CPU: 2 PID: 22 Comm: kworker/2:0 Tainted: G W 5.3.0-rc4+ #1\n Hardware name: LENOVO 20KHCTO1WW/20KHCTO1WW, BIOS N23ET65W (1.40 ) 07/02/2019\n Workqueue: usb_hub_wq hub_event\n Call Trace:\n dump_stack+0x9a/0xf0\n print_address_description.cold+0xae/0x34f\n __kasan_report.cold+0x75/0x93\n ? tm6000_fillbuf+0x390/0x3c0 [tm6000_alsa]\n ? dvb_fini+0x75/0x140 [tm6000_dvb]\n kasan_report+0xe/0x12\n dvb_fini+0x75/0x140 [tm6000_dvb]\n tm6000_close_extension+0x51/0x80 [tm6000]\n tm6000_usb_disconnect.cold+0xd4/0x105 [tm6000]\n usb_unbind_interface+0xe4/0x390\n device_release_driver_internal+0x121/0x250\n bus_remove_device+0x197/0x260\n device_del+0x268/0x550\n ? __device_links_no_driver+0xd0/0xd0\n ? usb_remove_ep_devs+0x30/0x3b\n usb_disable_device+0x122/0x400\n usb_disconnect+0x153/0x430\n hub_event+0x800/0x1e40\n ? trace_hardirqs_on_thunk+0x1a/0x20\n ? hub_port_debounce+0x1f0/0x1f0\n ? retint_kernel+0x10/0x10\n ? lock_is_held_type+0xf1/0x130\n ? hub_port_debounce+0x1f0/0x1f0\n ? process_one_work+0x4ae/0xa00\n process_one_work+0x4ba/0xa00\n ? pwq_dec_nr_in_flight+0x160/0x160\n ? do_raw_spin_lock+0x10a/0x1d0\n worker_thread+0x7a/0x5c0\n ? process_one_work+0xa00/0xa00\n kthread+0x1d5/0x200\n ? kthread_create_worker_on_cpu+0xd0/0xd0\n ret_from_fork+0x3a/0x50\n \n Allocated by task 2682:\n save_stack+0x1b/0x80\n __kasan_kmalloc.constprop.0+0xc2/0xd0\n usb_alloc_urb+0x28/0x60\n tm6000_start_feed+0x10a/0x300 [tm6000_dvb]\n dmx_ts_feed_start_filtering+0x86/0x120 [dvb_core]\n dvb_dmxdev_start_feed+0x121/0x180 [dvb_core]\n dvb_dmxdev_filter_start+0xcb/0x540 [dvb_core]\n dvb_demux_do_ioctl+0x7ed/0x890 [dvb_core]\n dvb_usercopy+0x97/0x1f0 [dvb_core]\n dvb_demux_ioctl+0x11/0x20 [dvb_core]\n do_vfs_ioctl+0x5d8/0x9d0\n ksys_ioctl+0x5e/0x90\n __x64_sys_ioctl+0x3d/0x50\n do_syscall_64+0x74/0xe0\n entry_SYSCALL_64_after_hwframe+0x49/0xbe\n \n Freed by task 22:\n save_stack+0x1b/0x80\n __kasan_slab_free+0x12c/0x170\n kfree+0xfd/0x3a0\n xhci_giveback_urb_in_irq+0xfe/0x230\n xhci_td_cleanup+0x276/0x340\n xhci_irq+0x1129/0x3720\n __handle_irq_event_percpu+0x6e/0x420\n handle_irq_event_percpu+0x6f/0x100\n handle_irq_event+0x55/0x84\n handle_edge_irq+0x108/0x3b0\n handle_irq+0x2e/0x40\n do_IRQ+0x83/0x1a0\n \n Cc: stable@vger.kernel.org\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 5cd690a308e86bcb10b8a0d6d42f153e82f695f6\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Wed Aug 14 09:24:03 2019 +0200\n\n mfd: asic3: Include the right header\n \n This is a GPIO driver, use the appropriate header\n \u003clinux/gpio/driver.h\u003e rather than the legacy \u003clinux/gpio.h\u003e\n header.\n \n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 6fb7195811fc9e70007fc37147ca0cbdb34b5e5b\nAuthor: Sean Young \u003csean@mess.org\u003e\nDate: Tue Aug 13 08:19:11 2019 -0300\n\n media: rc: imon-rsc keymap has incorrect mappings\n \n KEY_MAX is not a key but designates the highest value a linux keycode\n can ever have.\n \n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 40667d1bd822f7a9d961bac190cd36505de0de4e\nAuthor: Denis Efremov \u003cefremov@linux.com\u003e\nDate: Tue Aug 13 08:58:41 2019 +0300\n\n MAINTAINERS: altera-sysmgr: Fix typo in a filepath\n \n Fix typo (s/sysgmr/sysmgr/) in the header filepath.\n \n Fixes: f36e789a1f8d (\"mfd: altera-sysmgr: Add SOCFPGA System Manager\")\n Signed-off-by: Denis Efremov \u003cefremov@linux.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 46e4a26615cc7854340e4b69ca59ee78d6f20c8b\nAuthor: Sean Young \u003csean@mess.org\u003e\nDate: Sun Aug 11 02:05:51 2019 -0300\n\n media: em28xx: modules workqueue not inited for 2nd device\n \n syzbot reports an error on flush_request_modules() for the second device.\n This workqueue was never initialised so simply remove the offending line.\n \n usb 1-1: USB disconnect, device number 2\n em28xx 1-1:1.153: Disconnecting em28xx #1\n ------------[ cut here ]------------\n WARNING: CPU: 0 PID: 12 at kernel/workqueue.c:3031\n __flush_work.cold+0x2c/0x36 kernel/workqueue.c:3031\n Kernel panic - not syncing: panic_on_warn set ...\n CPU: 0 PID: 12 Comm: kworker/0:1 Not tainted 5.3.0-rc2+ #25\n Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS\n Google 01/01/2011\n Workqueue: usb_hub_wq hub_event\n Call Trace:\n __dump_stack lib/dump_stack.c:77 [inline]\n dump_stack+0xca/0x13e lib/dump_stack.c:113\n panic+0x2a3/0x6da kernel/panic.c:219\n __warn.cold+0x20/0x4a kernel/panic.c:576\n report_bug+0x262/0x2a0 lib/bug.c:186\n fixup_bug arch/x86/kernel/traps.c:179 [inline]\n fixup_bug arch/x86/kernel/traps.c:174 [inline]\n do_error_trap+0x12b/0x1e0 arch/x86/kernel/traps.c:272\n do_invalid_op+0x32/0x40 arch/x86/kernel/traps.c:291\n invalid_op+0x23/0x30 arch/x86/entry/entry_64.S:1026\n RIP: 0010:__flush_work.cold+0x2c/0x36 kernel/workqueue.c:3031\n Code: 9a 22 00 48 c7 c7 20 e4 c5 85 e8 d9 3a 0d 00 0f 0b 45 31 e4 e9 98 86\n ff ff e8 51 9a 22 00 48 c7 c7 20 e4 c5 85 e8 be 3a 0d 00 \u003c0f\u003e 0b 45 31 e4\n e9 7d 86 ff ff e8 36 9a 22 00 48 c7 c7 20 e4 c5 85\n RSP: 0018:ffff8881da20f720 EFLAGS: 00010286\n RAX: 0000000000000024 RBX: dffffc0000000000 RCX: 0000000000000000\n RDX: 0000000000000000 RSI: ffffffff8128a0fd RDI: ffffed103b441ed6\n RBP: ffff8881da20f888 R08: 0000000000000024 R09: fffffbfff11acd9a\n R10: fffffbfff11acd99 R11: ffffffff88d66ccf R12: 0000000000000000\n R13: 0000000000000001 R14: ffff8881c6685df8 R15: ffff8881d2a85b78\n flush_request_modules drivers/media/usb/em28xx/em28xx-cards.c:3325 [inline]\n em28xx_usb_disconnect.cold+0x280/0x2a6\n drivers/media/usb/em28xx/em28xx-cards.c:4023\n usb_unbind_interface+0x1bd/0x8a0 drivers/usb/core/driver.c:423\n __device_release_driver drivers/base/dd.c:1120 [inline]\n device_release_driver_internal+0x404/0x4c0 drivers/base/dd.c:1151\n bus_remove_device+0x2dc/0x4a0 drivers/base/bus.c:556\n device_del+0x420/0xb10 drivers/base/core.c:2288\n usb_disable_device+0x211/0x690 drivers/usb/core/message.c:1237\n usb_disconnect+0x284/0x8d0 drivers/usb/core/hub.c:2199\n hub_port_connect drivers/usb/core/hub.c:4949 [inline]\n hub_port_connect_change drivers/usb/core/hub.c:5213 [inline]\n port_event drivers/usb/core/hub.c:5359 [inline]\n hub_event+0x1454/0x3640 drivers/usb/core/hub.c:5441\n process_one_work+0x92b/0x1530 kernel/workqueue.c:2269\n process_scheduled_works kernel/workqueue.c:2331 [inline]\n worker_thread+0x7ab/0xe20 kernel/workqueue.c:2417\n kthread+0x318/0x420 kernel/kthread.c:255\n ret_from_fork+0x24/0x30 arch/x86/entry/entry_64.S:352\n Kernel Offset: disabled\n Rebooting in 86400 seconds..\n \n Fixes: be7fd3c3a8c5e (\"media: em28xx: Hauppauge DualHD second tuner functionality)\n Reviewed-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Reviewed-by: Brad Love \u003cbrad@nextdimension.cc\u003e\n Reported-by: syzbot+b7f57261c521087d89bb@syzkaller.appspotmail.com\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit f1409116835a8ae7cd924e2829772c470eb6cba4\nAuthor: Sean Young \u003csean@mess.org\u003e\nDate: Fri Aug 9 19:02:59 2019 -0300\n\n media: selftests: ir: fix ir_loopback test failure\n \n The decoder is called rc-mm, not rcmm. This was renamed late in the cycle\n so this bug crept in.\n \n Acked-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 494fce160f2dac6f78d4282b54942fe03de59641\nAuthor: Sean Young \u003csean@mess.org\u003e\nDate: Fri Aug 9 19:31:08 2019 -0300\n\n media: imon_raw: prevent \"nonsensical timing event of duration 0\"\n \n Sometimes the device sends IR data which is all space, no pulses\n whatsoever. Add the end of this the driver will put the rc device into\n idle mode when it already is in idle mode. The following will be logged:\n \n rc rc0: nonsensical timing event of duration 0\n rc rc0: two consecutive events of type space\n \n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit e70d13f7ac061da3fcb982a049e98f63b7ded3d9\nAuthor: Sean Young \u003csean@mess.org\u003e\nDate: Fri Aug 9 18:47:41 2019 -0300\n\n media: imon_raw: simplify and explain bit operations\n \n This code needs some explanation.\n \n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 1b09a2afa4dca452b28da31d0a6628f1d633807b\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Thu Aug 8 12:53:28 2019 -0300\n\n media: ir-kbd-i2c: convert to i2c_new_dummy_device()\n \n Convert this driver to use the new i2c_new_dummy_device() call and bail\n out if the dummy device cannot be registered to make failure more\n visible to the user.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 0b08ae030154118705b993ff69b3f0c9c450c229\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Tue Aug 13 17:21:45 2019 -0700\n\n drm/i915/guc: Remove client-\u003esubmissions\n \n The engine-\u003eguc_id is GuC FW defined and it is not guaranteed to be\n below I915_NUM_ENGINES, so we shouldn\u0027t use it with the i915-defined\n client-\u003esubmissions, as we might overflow.\n Instead of fixing it, just get rid of client-\u003esubmissions, because the\n information we get from it is not interesting anymore now that we only\n have 1 client.\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190814002145.29056-1-daniele.ceraolospurio@intel.com\n\ncommit 92ffdb61f65d5d27bc750b6249c381c89767aaba\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Tue Aug 6 04:07:38 2019 -0300\n\n media: rc: add include guard to rc-map.h\n \n Add a header include guard just in case.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 247d46b3ef7560c956cb0886f20d65974eb7c9ff\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Jul 5 05:51:32 2019 -0300\n\n media: stv0900_core: remove redundant assignment to variables mclk, div and ad_div\n \n The variables mclk, div and ad_div are being assigned with a values\n that are never read and are being updated later with a new values.\n The assignments are redundant and can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n \n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 5532c6282787124a7fd778d779d19148ae87479e\nAuthor: Luke Nowakowski-Krijger \u003clnowakow@eng.ucsd.edu\u003e\nDate: Thu Jul 4 17:58:12 2019 -0300\n\n media: dvb_frontend.h: Fix shifting signed 32-bit value problem\n \n Fix DVBFE_ALGO_RECOVERY and DVBFE_ALGO_SEARCH_ERROR use BIT macro which\n fixes undefined behavior error by certain compilers.\n \n Also changed all other bit shifted definitions to use macro for better\n readability.\n \n Signed-off-by: Luke Nowakowski-Krijger \u003clnowakow@eng.ucsd.edu\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit f6af820ef1be58c2e4b81aa479b9f109eb6344ce\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Wed Jul 3 13:28:37 2019 -0300\n\n media: media/dvb: Use kmemdup rather than duplicating its implementation\n \n kmemdup is introduced to duplicate a region of memory in a neat way.\n Rather than kmalloc/kzalloc + memcpy, which the programmer needs to\n write the size twice (sometimes lead to mistakes), kmemdup improves\n readability, leads to smaller code and also reduce the chances of mistakes.\n Suggestion to use kmemdup rather than using kmalloc/kzalloc + memcpy.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 48059784ebac75a98e024c29ab301e05a8b8012f\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Wed Jul 3 13:28:48 2019 -0300\n\n media: media/tuners: Use kmemdup rather than duplicating its implementation\n \n kmemdup is introduced to duplicate a region of memory in a neat way.\n Rather than kmalloc/kzalloc + memcpy, which the programmer needs to\n write the size twice (sometimes lead to mistakes), kmemdup improves\n readability, leads to smaller code and also reduce the chances of mistakes.\n Suggestion to use kmemdup rather than using kmalloc/kzalloc + memcpy.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 771560e5d341dc61f9e87bc715f3dbb4efe6ec62\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Wed Jul 3 13:28:57 2019 -0300\n\n media: media/usb: Use kmemdup rather than duplicating its implementation\n \n kmemdup is introduced to duplicate a region of memory in a neat way.\n Rather than kmalloc/kzalloc + memcpy, which the programmer needs to\n write the size twice (sometimes lead to mistakes), kmemdup improves\n readability, leads to smaller code and also reduce the chances of mistakes.\n Suggestion to use kmemdup rather than using kmalloc/kzalloc + memcpy.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 7a637fd32b234d4605fdf3c75c622b0ca1ce8cfb\nAuthor: Hongwei Zhang \u003chongweiz@ami.com\u003e\nDate: Wed Jul 31 16:01:36 2019 -0400\n\n dt-bindings: gpio: aspeed: Add SGPIO support\n \n Add bindings to support SGPIO on AST2400 or AST2500.\n \n Signed-off-by: Hongwei Zhang \u003chongweiz@ami.com\u003e\n Reviewed-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Link: https://lore.kernel.org/r/1564603297-1391-2-git-send-email-hongweiz@ami.com\n [Adjusted when applying in several ways]\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 41b57d1bb8a4084e651c1f9a754fca64952666a0\nAuthor: Mark Rutland \u003cmark.rutland@arm.com\u003e\nDate: Tue Aug 6 17:25:39 2019 +0100\n\n lib: Remove redundant ftrace flag removal\n \n Since architectures can implement ftrace using a variety of mechanisms,\n generic code should always use CC_FLAGS_FTRACE rather than assuming that\n ftrace is built using -pg.\n \n Since commit:\n \n 2464a609ded09420 (\"ftrace: do not trace library functions\")\n \n ... lib/Makefile has removed CC_FLAGS_FTRACE from KBUILD_CFLAGS, so ftrace is\n disabled for all files under lib/.\n \n Given that, we shouldn\u0027t explicitly remove -pg when building\n lib/string.o, as this is redundant and bad form.\n \n Clean things up accordingly.\n \n There should be no functional change as a result of this patch.\n \n Signed-off-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Cc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n Cc: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Cc: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: Coly Li \u003ccolyli@suse.de\u003e\n Cc: Gary R Hook \u003cgary.hook@amd.com\u003e\n Cc: Ingo Molnar \u003cmingo@kernel.org\u003e\n Cc: Kees Cook \u003ckeescook@chromium.org\u003e\n Cc: Kent Overstreet \u003ckent.overstreet@gmail.com\u003e\n Cc: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Cc: Matthew Wilcox \u003cwilly@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190806162539.51918-1-mark.rutland@arm.com\n\ncommit 31664caaf4d8f4f6771ca33744297d420f579ed8\nAuthor: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\nDate: Tue Aug 13 21:38:51 2019 +0800\n\n iio:light:noa1305: Fix missing break statement.\n \n This got caught by the implicit fall through detection but is\n a bug rather than missing marking.\n \n Reported-by: 0-DAY kernel test infrastructure\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n Fixes: 741172d18e8a (\"iio: light: noa1305: Add support for NOA1305\")\n Link: https://lore.kernel.org/r/20190813133851.14345-1-Jonathan.Cameron@huawei.com\n Cc: Gustavo Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit be6133b8917f8131b76728f11db51a65c91b3fec\nAuthor: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\nDate: Tue Aug 13 23:59:53 2019 -0700\n\n drm/i915: Update DRIVER_DATE to 20190813\n \n Signed-off-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n\ncommit 9bb34e9013fce81c1884d903e776ad4e18887cc3\nAuthor: Deepak Rawat \u003cdrawat@vmware.com\u003e\nDate: Thu May 9 12:42:58 2019 -0700\n\n drm/vmwgfx: Use VMW_DEBUG_KMS for vmwgfx mode-setting user errors\n \n For errors during layout change ioctl use VMW_DEBUG_KMS instead of\n DRM_ERROR.\n \n Signed-off-by: Deepak Rawat \u003cdrawat@vmware.com\u003e\n Reviewed-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Signed-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n\ncommit 745adc3f9216de4fbeedccccf5d324588b42f655\nAuthor: Deepak Rawat \u003cdrawat@vmware.com\u003e\nDate: Thu May 9 11:47:40 2019 -0700\n\n drm/vmwgfx: Add debug message for layout change ioctl\n \n Add debug code to check user-space layout change request.\n \n Signed-off-by: Deepak Rawat \u003cdrawat@vmware.com\u003e\n Reviewed-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Signed-off-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n\ncommit 49bddc73d15c25a68e4294d76fc74519fda984cd\nAuthor: Aneesh Kumar K.V \u003caneesh.kumar@linux.ibm.com\u003e\nDate: Wed Aug 7 09:30:29 2019 +0530\n\n libnvdimm/of_pmem: Provide a unique name for bus provider\n \n ndctl binaries, v66 and older, mistakenly require the ndbus to have\n unique names. If not while enumerating the bus in userspace it drops bus\n with similar names. This results in us not listing devices beneath the\n bus.\n \n Signed-off-by: Aneesh Kumar K.V \u003caneesh.kumar@linux.ibm.com\u003e\n Tested-by: Vaibhav Jain \u003cvaibhav@linux.ibm.com\u003e\n Link: https://lore.kernel.org/r/20190807040029.11344-1-aneesh.kumar@linux.ibm.com\n Signed-off-by: Dan Williams \u003cdan.j.williams@intel.com\u003e\n\ncommit 0a1dcf954ece215284e2a7765613c559d20590fc\nAuthor: Ben Pai \u003cBen_Pai@wistron.com\u003e\nDate: Mon Aug 12 19:13:07 2019 +0800\n\n ARM: dts: aspeed: Add Mihawk BMC platform\n \n The Mihawk BMC is an ASPEED ast2500 based BMC that is part of an\n OpenPower Power9 server.\n \n Signed-off-by: Ben Pai \u003cBen_Pai@wistron.com\u003e\n Reviewed-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Signed-off-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n\ncommit d00523097a3723ebc89e7752730e81aac65e549d\nAuthor: John Wang \u003cwangzqbj@inspur.com\u003e\nDate: Thu Aug 1 17:36:07 2019 +0800\n\n ARM: dts: aspeed: fp5280g2: Fix power supply address\n \n There are two PSU on i2c11. PSU0\u0027s address is 0x58, PSU1\u0027s address is\n `0x59`, not `0x5a`.\n \n Signed-off-by: John Wang \u003cwangzqbj@inspur.com\u003e\n Signed-off-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n\ncommit e4a5017ac5b3924384a36a0a043cb65bb41dd5be\nAuthor: Hayes Wang \u003chayeswang@realtek.com\u003e\nDate: Tue Aug 13 11:42:09 2019 +0800\n\n r8152: change rx_copybreak and rx_pending through ethtool\n \n Let the rx_copybreak and rx_pending could be modified by\n ethtool.\n \n Signed-off-by: Hayes Wang \u003chayeswang@realtek.com\u003e\n Signed-off-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n\ncommit 47922fcde5365a2d37e8d4056e537fc8a7213c39\nAuthor: Hayes Wang \u003chayeswang@realtek.com\u003e\nDate: Tue Aug 13 11:42:08 2019 +0800\n\n r8152: support skb_add_rx_frag\n \n Use skb_add_rx_frag() to reduce the memory copy for rx data.\n \n Use a new list of rx_used to store the rx buffer which couldn\u0027t be\n reused yet.\n \n Besides, the total number of rx buffer may be increased or decreased\n dynamically. And it is limited by RTL8152_MAX_RX_AGG.\n \n Signed-off-by: Hayes Wang \u003chayeswang@realtek.com\u003e\n Signed-off-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n\ncommit d55d70894c6d4709b9ae61109a9fa7c319586b53\nAuthor: Hayes Wang \u003chayeswang@realtek.com\u003e\nDate: Tue Aug 13 11:42:07 2019 +0800\n\n r8152: use alloc_pages for rx buffer\n \n Replace kmalloc_node() with alloc_pages() for rx buffer.\n \n Signed-off-by: Hayes Wang \u003chayeswang@realtek.com\u003e\n Signed-off-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n\ncommit 252df8b86667fe4640a2d9fb5cfc705ad285d578\nAuthor: Hayes Wang \u003chayeswang@realtek.com\u003e\nDate: Tue Aug 13 11:42:06 2019 +0800\n\n r8152: replace array with linking list for rx information\n \n The original method uses an array to store the rx information. The\n new one uses a list to link each rx structure. Then, it is possible\n to increase/decrease the number of rx structure dynamically.\n \n Signed-off-by: Hayes Wang \u003chayeswang@realtek.com\u003e\n Signed-off-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n\ncommit ec5791c202aca90c1b3b99dff268a995cf2d6aa1\nAuthor: Hayes Wang \u003chayeswang@realtek.com\u003e\nDate: Tue Aug 13 11:42:05 2019 +0800\n\n r8152: separate the rx buffer size\n \n The different chips may accept different rx buffer sizes. The RTL8152\n supports 16K bytes, and RTL8153 support 32K bytes.\n \n Signed-off-by: Hayes Wang \u003chayeswang@realtek.com\u003e\n Signed-off-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n\ncommit 65b27995a4ab8fc51b4adc6b4dcdca20f7a595bb\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Mon Aug 12 23:52:19 2019 +0200\n\n net: phy: let phy_speed_down/up support speeds \u003e1Gbps\n \n So far phy_speed_down/up can be used up to 1Gbps only. Remove this\n restriction by using new helper __phy_speed_down. New member adv_old\n in struct phy_device is used by phy_speed_up to restore the advertised\n modes before calling phy_speed_down. Don\u0027t simply advertise what is\n supported because a user may have intentionally removed modes from\n advertisement.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n\ncommit 331c56ac73846fa267c04ee6aa9a00bb5fed9440\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Mon Aug 12 23:51:27 2019 +0200\n\n net: phy: add phy_speed_down_core and phy_resolve_min_speed\n \n phy_speed_down_core provides most of the functionality for\n phy_speed_down. It makes use of new helper phy_resolve_min_speed that is\n based on the sorting of the settings[] array. In certain cases it may be\n helpful to be able to exclude legacy half duplex modes, therefore\n prepare phy_resolve_min_speed() for it.\n \n v2:\n - rename __phy_speed_down to phy_speed_down_core\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n\ncommit 7b261e0ef5f88087765df7f079a4bd0d285f7579\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Mon Aug 12 23:50:30 2019 +0200\n\n net: phy: add __set_linkmode_max_speed\n \n We will need the functionality of __set_linkmode_max_speed also for\n linkmode bitmaps other than phydev-\u003esupported. Therefore split it.\n \n v2:\n - remove unused parameter from __set_linkmode_max_speed\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n\ncommit 043b8413e8c0c0ffbf8be268eb73716e05a96064\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Mon Aug 12 20:02:02 2019 +0300\n\n net: devlink: remove redundant rtnl lock assert\n \n It is enough for caller of devlink_compat_switch_id_get() to hold the net\n device to guarantee that devlink port is not destroyed concurrently. Remove\n rtnl lock assertion and modify comment to warn user that they must hold\n either rtnl lock or reference to net device. This is necessary to\n accommodate future implementation of rtnl-unlocked TC offloads driver\n callbacks.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n\ncommit a9a96760165d8781570f2dadeed8e38484b13499\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Mon Aug 12 22:41:56 2019 +0800\n\n net: hns3: Make hclge_func_reset_sync_vf static\n \n Fix sparse warning:\n \n drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c:3190:5:\n warning: symbol \u0027hclge_func_reset_sync_vf\u0027 was not declared. Should it be static?\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n\ncommit 270b9991cdb03df39916afa7184c783fc5b512b5\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Tue Jul 30 15:47:51 2019 -0700\n\n drm/i915: Add _TRANS2()\n \n A new macro that is going to be added in a further patch will need to\n adjust the offset returned by _MMIO_TRANS2(), so here adding\n _TRANS2() and moving most of the implementation of _MMIO_TRANS2() to\n it and while at it taking the opportunity to rename pipe to trans.\n \n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Dhinakaran Pandiyan \u003cdhinakaran.pandiya@intel.com\u003e\n Reviewed-by: Dhinakaran Pandiyan \u003cdhinakaran.pandiya@intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Reviewed-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730224753.14907-2-jose.souza@intel.com\n\ncommit 46c63d2499b12902ed30847100af6c26228e878c\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Tue Jul 30 15:47:50 2019 -0700\n\n drm/i915/bdw+: Move misc display IRQ handling to it own function\n \n Just moving it to reduce the tabs and avoid break code lines.\n No behavior changes intended here.\n \n v2:\n - Reading misc display IRQ outside of gen8_de_misc_irq_handler() as\n other irq handlers (Dhinakaran)\n \n Cc: Dhinakaran Pandiyan \u003cdhinakaran.pandiyan@intel.com\u003e\n Reviewed-by: Dhinakaran Pandiyan \u003cdhinakaran.pandiyan@intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Reviewed-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730224753.14907-1-jose.souza@intel.com\n\ncommit 7aa8dd91da63af535b3b9153257e089254982b7e\nAuthor: Stephen Boyd \u003csboyd@kernel.org\u003e\nDate: Tue Aug 13 11:38:25 2019 -0700\n\n devicetree: Expose dtbs_check and dt_binding_check some more\n \n It wasn\u0027t obvious that this was a command to run based on \u0027make help\u0027,\n so add it to the top-level help for devicetree builds. Also, add an\n example to the documentation to show that db_binding_check can be run\n with DT_SCHEMA_FILES\u003d to only check one schema file instead of all of\n them.\n \n Cc: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Cc: \u003clinux-kbuild@vger.kernel.org\u003e\n Cc: \u003cdevicetree@vger.kernel.org\u003e\n Cc: \u003clinux-doc@vger.kernel.org\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n [robh: fix-up due to .md to .rst conversion]\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 0d530aa6894c149a0a6cd36d63b03738817b745e\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Thu Aug 8 10:51:39 2019 +0200\n\n dt-bindings: watchdog: meson-gxbb-wdt: convert to yaml\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for the Amlogic GXBB Watchdog timer over to a YAML schemas.\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 4df7e9a13953f95c3970d7a1fa97f8bc16f83a37\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Thu Aug 8 10:51:38 2019 +0200\n\n dt-bindings: serial: meson-uart: convert to yaml\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for the Amlogic UART Serial controller over to a YAML schemas.\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 87a55485f2fc3819ad20f6e7f33fd5a252f8e869\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Thu Aug 8 10:51:37 2019 +0200\n\n dt-bindings: phy: meson-g12a-usb3-pcie-phy: convert to yaml\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for the Amlogic G12A USB3 + PCIE Combo PHY over to a YAML schemas.\n \n While the original phy bindings specifies phy-supply as required,\n the examples and implementations makes it optional, thus phy-supply\n is not present in the properties and required lists.\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit da86d286cce83d775066b15f650807c6a285afc2\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Thu Aug 8 10:51:36 2019 +0200\n\n dt-bindings: phy: meson-g12a-usb2-phy: convert to yaml\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for the Amlogic G12A USB2 PHY over to a YAML schemas.\n \n While the original phy bindings specifies phy-supply as required,\n the examples and implementations makes it optional, thus phy-supply\n is not in the required list of attributes.\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit f50ffbc5e9eb463d84073e023ffe07280b30ac50\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Thu Aug 8 10:51:35 2019 +0200\n\n dt-bindings: arm: amlogic: amlogic,meson-gx-ao-secure: convert to yaml\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for the Amlogic Always-On Secure Registers over to a YAML schemas.\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 018562f36f8c6c473354e85c2323d72bd77977f2\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Thu Aug 8 10:51:34 2019 +0200\n\n dt-bindings: reset: amlogic,meson-reset: convert to yaml\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for the Amlogic Reset controller over to a YAML schemas.\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Acked-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 66de150a88896ffa8a954ab09fce91ba116898ef\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Thu Aug 8 10:51:33 2019 +0200\n\n dt-bindings: spi: meson: convert to yaml\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for the Amlogic SPI controllers over to two separate YAML schemas.\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 78db5580af782b4d84279559c75f2fbdc46b3d4b\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Thu Aug 8 10:51:32 2019 +0200\n\n dt-bindings: rng: amlogic,meson-rng: convert to yaml\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for the Amlogic Random Number generator over to a YAML schemas.\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 8d63f97113a7e474b4fbbdc6ccde1091da167211\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Thu Aug 8 10:51:31 2019 +0200\n\n dt-bindings: mailbox: meson-mhu: convert to yaml\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for the Amlogic MHU controller over to a YAML schemas.\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit cdea0121ae803237ce9b09779194f82ca1467994\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Wed Jul 31 17:08:49 2019 -0300\n\n docs: writing-schema.md: convert from markdown to ReST\n \n The documentation standard is ReST and not markdown.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Acked-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 5f9040fd3dff38b9679f11c5ac7f718a829d4233\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Thu Jul 11 14:23:01 2019 +0200\n\n dt-bindings: crypto: Convert Allwinner A10 Security Engine to a schema\n \n The older Allwinner SoCs have a crypto engine that is supported in Linux,\n with a matching Device Tree binding.\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for that controller over to a YAML schemas.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 31ca694e864592b90e4bee2fdce14b24645953b1\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Thu Jul 11 11:40:36 2019 +0200\n\n dt-bindings: bus: Convert Allwinner DE2 bus to a schema\n \n The Allwinner SoCs using the second generation of the display engine have a\n bus for that display engine. The bus is supported in Linux, with a matching\n Device Tree binding.\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for that controller over to a YAML schemas.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n [robh: add \u0027type: object\u0027 for child nodes]\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 5bf2845ece35513971d22b2117f9c3b0f35b6ac7\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Thu Jul 11 11:38:35 2019 +0200\n\n dt-bindings: input: Convert Allwinner LRADC to a schema\n \n The Allwinner SoCs have an LRADC used to report keys and supported in\n Linux, with a matching Device Tree binding.\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for that controller over to a YAML schemas.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 92b49822288134b4cd17a2c02a15ad265b9b327c\nAuthor: Jiri Pirko \u003cjiri@mellanox.com\u003e\nDate: Mon Aug 12 14:28:31 2019 +0200\n\n devlink: send notifications for deleted snapshots on region destroy\n \n Currently the notifications for deleted snapshots are sent only in case\n user deletes a snapshot manually. Send the notifications in case region\n is destroyed too.\n \n Signed-off-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n\ncommit cfcdef5e30469f3f2d6786ad35fc3fdef2a3833f\nAuthor: Eric Dumazet \u003cedumazet@google.com\u003e\nDate: Wed Jul 24 18:07:52 2019 -0700\n\n rcu: Allow rcu_do_batch() to dynamically adjust batch sizes\n \n Bimodal behavior of rcu_do_batch() is not really suited to Google\n applications like gfe servers.\n \n When a process with millions of sockets exits, closing all files\n queues two rcu callbacks per socket.\n \n This eventually reaches the point where RCU enters an emergency\n mode, where rcu_do_batch() do not return until whole queue is flushed.\n \n Each rcu callback lasts at least 70 nsec, so with millions of\n elements, we easily spend more than 100 msec without rescheduling.\n \n Goal of this patch is to avoid the infamous message like following\n \"need_resched set for \u003e 51999388 ns (52 ticks) without schedule\"\n \n We dynamically adjust the number of elements we process, instead\n of 10 / INFINITE choices, we use a floor of ~1 % of current entries.\n \n If the number is above 1000, we switch to a time based limit of 3 msec\n per batch, adjustable with /sys/module/rcutree/parameters/rcu_resched_ns\n \n Signed-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\n [ paulmck: Forward-port and remove debug statements. ]\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit f48fe4c586604c3a09938c6a6e9fd3356dfe8f3c\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue Jul 16 02:17:00 2019 -0700\n\n rcu/nocb: Don\u0027t wake no-CBs GP kthread if timer posted under overload\n \n When under overload conditions, __call_rcu_nocb_wake() will wake the\n no-CBs GP kthread any time the no-CBs CB kthread is asleep or there\n are no ready-to-invoke callbacks, but only after a timer delay. If the\n no-CBs GP kthread has a -\u003enocb_bypass_timer pending, the deferred wakeup\n from __call_rcu_nocb_wake() is redundant. This commit therefore makes\n __call_rcu_nocb_wake() avoid posting the redundant deferred wakeup if\n -\u003enocb_bypass_timer is pending. This requires adding a bit of ordering\n of timer actions.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 296181d78df9892e08e794f2a9a4d2c38f9acedb\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Mon Jul 15 06:06:40 2019 -0700\n\n rcu/nocb: Reduce __call_rcu_nocb_wake() leaf rcu_node -\u003elock contention\n \n Currently, __call_rcu_nocb_wake() advances callbacks each time that it\n detects excessive numbers of callbacks, though only if it succeeds in\n conditionally acquiring its leaf rcu_node structure\u0027s -\u003elock. Despite\n the conditional acquisition of -\u003elock, this does increase contention.\n This commit therefore avoids advancing callbacks unless there are\n callbacks in -\u003ecblist whose grace period has completed and advancing\n has not yet been done during this jiffy.\n \n Note that this decision does not take the presence of new callbacks\n into account. That is because on this code path, there will always be\n at least one new callback, namely the one we just enqueued.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 1d5a81c18dc68fc38a52e8dab1992a043a358927\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Mon Jul 15 01:09:04 2019 -0700\n\n rcu/nocb: Reduce nocb_cb_wait() leaf rcu_node -\u003elock contention\n \n Currently, nocb_cb_wait() advances callbacks on each pass through its\n loop, though only if it succeeds in conditionally acquiring its leaf\n rcu_node structure\u0027s -\u003elock. Despite the conditional acquisition of\n -\u003elock, this does increase contention. This commit therefore avoids\n advancing callbacks unless there are callbacks in -\u003ecblist whose grace\n period has completed.\n \n Note that nocb_cb_wait() doesn\u0027t worry about callbacks that have not\n yet been assigned a grace period. The idea is that the only reason for\n nocb_cb_wait() to advance callbacks is to allow it to continue invoking\n callbacks. Time will tell whether this is the correct choice.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 23651d9b9616060cf86af5e3b15defcf3bcd2642\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Wed Jul 10 12:54:56 2019 -0700\n\n rcu/nocb: Advance CBs after merge in rcutree_migrate_callbacks()\n \n The rcutree_migrate_callbacks() invokes rcu_advance_cbs() on both the\n offlined CPU\u0027s -\u003ecblist and that of the surviving CPU, then merges\n them. However, after the merge, and of the offlined CPU\u0027s callbacks\n that were not ready to be invoked will no longer be associated with a\n grace-period number. This commit therefore invokes rcu_advance_cbs()\n one more time on the merged -\u003ecblist in order to assign a grace-period\n number to these callbacks.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 273f034065002bf9480601d66404c991b243b91e\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue Jul 9 06:54:42 2019 -0700\n\n rcu/nocb: Avoid synchronous wakeup in __call_rcu_nocb_wake()\n \n When callbacks are in full flow, the common case is waiting for a\n grace period, and this grace period will normally take a few jiffies to\n complete. It therefore isn\u0027t all that helpful for __call_rcu_nocb_wake()\n to do a synchronous wakeup in this case. This commit therefore turns this\n into a timer-based deferred wakeup of the no-CBs grace-period kthread.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit f7a81b12d6af42a9d09be1e5f041169f04b0b67a\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue Jun 25 13:32:51 2019 -0700\n\n rcu/nocb: Print no-CBs diagnostics when rcutorture writer unduly delayed\n \n This commit causes locking, sleeping, and callback state to be printed\n for no-CBs CPUs when the rcutorture writer is delayed sufficiently for\n rcutorture to complain.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 6aacd88d1721e12b013ae4ccf4f17609bd5091f3\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Sat Jul 13 12:27:03 2019 -0700\n\n rcu/nocb: EXP Check use and usefulness of -\u003enocb_lock_contended\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit d1b222c6be1f8bfc77099e034219732ecaeaaf96\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue Jul 2 16:03:33 2019 -0700\n\n rcu/nocb: Add bypass callback queueing\n \n Use of the rcu_data structure\u0027s segmented -\u003ecblist for no-CBs CPUs\n takes advantage of unrelated grace periods, thus reducing the memory\n footprint in the face of floods of call_rcu() invocations. However,\n the -\u003ecblist field is a more-complex rcu_segcblist structure which must\n be protected via locking. Even though there are only three entities\n which can acquire this lock (the CPU invoking call_rcu(), the no-CBs\n grace-period kthread, and the no-CBs callbacks kthread), the contention\n on this lock is excessive under heavy stress.\n \n This commit therefore greatly reduces contention by provisioning\n an rcu_cblist structure field named -\u003enocb_bypass within the\n rcu_data structure. Each no-CBs CPU is permitted only a limited\n number of enqueues onto the -\u003ecblist per jiffy, controlled by a new\n nocb_nobypass_lim_per_jiffy kernel boot parameter that defaults to\n about 16 enqueues per millisecond (16 * 1000 / HZ). When that limit is\n exceeded, the CPU instead enqueues onto the new -\u003enocb_bypass.\n \n The -\u003enocb_bypass is flushed into the -\u003ecblist every jiffy or when\n the number of callbacks on -\u003enocb_bypass exceeds qhimark, whichever\n happens first. During call_rcu() floods, this flushing is carried out\n by the CPU during the course of its call_rcu() invocations. However,\n a CPU could simply stop invoking call_rcu() at any time. The no-CBs\n grace-period kthread therefore carries out less-aggressive flushing\n (every few jiffies or when the number of callbacks on -\u003enocb_bypass\n exceeds (2 * qhimark), whichever comes first). This means that the\n no-CBs grace-period kthread cannot be permitted to do unbounded waits\n while there are callbacks on -\u003enocb_bypass. A -\u003enocb_bypass_timer is\n used to provide the needed wakeups.\n \n [ paulmck: Apply Coverity feedback reported by Colin Ian King. ]\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit eda669a6a2c517fd6db41d0fe3c95c1b749c60bd\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Mon Jul 1 17:36:53 2019 -0700\n\n rcu/nocb: Atomic -\u003elen field in rcu_segcblist structure\n \n Upcoming -\u003enocb_lock contention-reduction work requires that the\n rcu_segcblist structure\u0027s -\u003elen field be concurrently manipulated,\n but only if there are no-CBs CPUs in the kernel. This commit\n therefore makes this -\u003elen field be an atomic_long_t, but only\n in CONFIG_RCU_NOCB_CPU\u003dy kernels.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit faca5c250935262f026cac1bb951a0f7672474b8\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Wed Jun 26 09:50:38 2019 -0700\n\n rcu/nocb: Unconditionally advance and wake for excessive CBs\n \n When there are excessive numbers of callbacks, and when either the\n corresponding no-CBs callback kthread is asleep or there is no more\n ready-to-invoke callbacks, and when least one callback is pending,\n __call_rcu_nocb_wake() will advance the callbacks, but refrain from\n awakening the corresponding no-CBs grace-period kthread. However,\n because rcu_advance_cbs_nowake() is used, it is possible (if a bit\n unlikely) that the needed advancement could not happen due to a grace\n period not being in progress. Plus there will always be at least one\n pending callback due to one having just now been enqueued.\n \n This commit therefore attempts to advance callbacks and awakens the\n no-CBs grace-period kthread when there are excessive numbers of callbacks\n posted and when the no-CBs callback kthread is not in a position to do\n anything helpful.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 4fd8c5f153bc41ae847b9ddb1539b34f70c18278\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Sun Jun 2 13:41:08 2019 -0700\n\n rcu/nocb: Reduce -\u003enocb_lock contention with separate -\u003enocb_gp_lock\n \n The sleep/wakeup of the no-CBs grace-period kthreads is synchronized\n using the -\u003enocb_lock of the first CPU corresponding to that kthread.\n This commit provides a separate -\u003enocb_gp_lock for this purpose, thus\n reducing contention on -\u003enocb_lock.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 523bddd553c09a2cf051eb724bffba680424f5ec\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Sat Jun 1 13:33:55 2019 -0700\n\n rcu/nocb: Reduce contention at no-CBs invocation-done time\n \n Currently, nocb_cb_wait() unconditionally acquires the leaf rcu_node\n -\u003elock to advance callbacks when done invoking the previous batch.\n It does this while holding -\u003enocb_lock, which means that contention on\n the leaf rcu_node -\u003elock visits itself on the -\u003enocb_lock. This commit\n therefore makes this lock acquisition conditional, forgoing callback\n advancement when the leaf rcu_node -\u003elock is not immediately available.\n (In this case, the no-CBs grace-period kthread will eventually do any\n needed callback advancement.)\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 6608c3a027bcc0b34cc02bc764ea9f52b9dce46f\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Sat Jun 1 06:16:38 2019 -0700\n\n rcu/nocb: Reduce contention at no-CBs registry-time CB advancement\n \n Currently, __call_rcu_nocb_wake() conditionally acquires the leaf rcu_node\n structure\u0027s -\u003elock, and only afterwards does rcu_advance_cbs_nowake()\n check to see if it is possible to advance callbacks without potentially\n needing to awaken the grace-period kthread. Given that the no-awaken\n check can be done locklessly, this commit reverses the order, so that\n rcu_advance_cbs_nowake() is invoked without holding the leaf rcu_node\n structure\u0027s -\u003elock and rcu_advance_cbs_nowake() checks the grace-period\n state before conditionally acquiring that lock, thus reducing the number\n of needless acquistions of the leaf rcu_node structure\u0027s -\u003elock.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 9fcb09bddd56bae42319b606bae86e85c625f868\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Sat Jun 1 05:14:47 2019 -0700\n\n rcu/nocb: Round down for number of no-CBs grace-period kthreads\n \n Currently, when the square root of the number of CPUs is rounded down\n by int_sqrt(), this round-down is applied to the number of callback\n kthreads per grace-period kthreads. This makes almost no difference\n for large systems, but results in oddities such as three no-CBs\n grace-period kthreads for a five-CPU system, which is a bit excessive.\n This commit therefore causes the round-down to apply to the number of\n no-CBs grace-period kthreads, so that systems with from four to eight\n CPUs have only two no-CBs grace period kthreads.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 81c0b3d724f419c0524f432c1ac22b9f518c2899\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue May 28 07:18:08 2019 -0700\n\n rcu/nocb: Avoid -\u003enocb_lock capture by corresponding CPU\n \n A given rcu_data structure\u0027s -\u003enocb_lock can be acquired very frequently\n by the corresponding CPU and occasionally by the corresponding no-CBs\n grace-period and callbacks kthreads. In particular, these two kthreads\n will have frequent gaps between -\u003enocb_lock acquisitions that are roughly\n a grace period in duration. This means that any excessive -\u003enocb_lock\n contention will be due to the CPU\u0027s acquisitions, and this in turn\n enables a very naive contention-avoidance strategy to be quite effective.\n \n This commit therefore modifies rcu_nocb_lock() to first\n attempt a raw_spin_trylock(), and to atomically increment a\n separate -\u003enocb_lock_contended across a raw_spin_lock(). This new\n -\u003enocb_lock_contended field is checked in __call_rcu_nocb_wake() when\n interrupts are enabled, with a spin-wait for contending acquisitions\n to complete, thus allowing the kthreads a chance to acquire the lock.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 7f36ef82e5cf0b401c2676fb3e56ad0633ed6ad5\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue May 28 05:54:26 2019 -0700\n\n rcu/nocb: Avoid needless wakeups of no-CBs grace-period kthread\n \n Currently, the code provides an extra wakeup for the no-CBs grace-period\n kthread if one of its CPUs is generating excessive numbers of callbacks.\n But satisfying though it is to wake something up when things are going\n south, unless the thing being awakened can actually help solve the\n problem, that extra wakeup does nothing but consume additional CPU time,\n which is exactly what you don\u0027t want during a call_rcu() flood.\n \n This commit therefore avoids doing anything if the corresponding\n no-CBs callback kthread is going full tilt. Otherwise, if advancing\n callbacks immediately might help and if the leaf rcu_node structure\u0027s\n lock is immediately available, this commit invokes a new variant of\n rcu_advance_cbs() that advances callbacks only if doing so won\u0027t require\n awakening the grace-period kthread (not to be confused with any of the\n no-CBs grace-period kthreads).\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit ce0a825e40606d6dbe6dfe90d4d4c0ccc9fa3bde\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Thu May 23 13:56:12 2019 -0700\n\n rcu/nocb: Make __call_rcu_nocb_wake() safe for many callbacks\n \n It might be hard to imagine having more than two billion callbacks\n queued on a single CPU\u0027s -\u003ecblist, but someone will do it sometime.\n This commit therefore makes __call_rcu_nocb_wake() handle this situation\n by upgrading local variable \"len\" from \"int\" to \"long\".\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 383e13328373ae1e17119ff89c86ff5f9413f31c\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Thu May 23 13:49:26 2019 -0700\n\n rcu/nocb: Never downgrade -\u003enocb_defer_wakeup in wake_nocb_gp_defer()\n \n Currently, wake_nocb_gp_defer() simply stores whatever waketype was\n passed in, which can result in a RCU_NOCB_WAKE_FORCE being downgraded\n to RCU_NOCB_WAKE, which could in turn delay callback processing.\n This commit therefore adds a check so that wake_nocb_gp_defer() only\n updates -\u003enocb_defer_wakeup when the update increases the forcefulness,\n thus avoiding downgrades.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit aeeacd9d844b2219d47e9010298b635c68a2a4c9\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Thu May 23 10:43:58 2019 -0700\n\n rcu/nocb: Enable re-awakening under high callback load\n \n The __call_rcu_nocb_wake() function and its predecessors set\n -\u003eqlen_last_fqs_check to zero for the first callback and to LONG_MAX / 2\n for forced reawakenings. The former can result in a too-quick reawakening\n when there are many callbacks ready to invoke and the latter prevents a\n second reawakening. This commit therefore sets -\u003eqlen_last_fqs_check\n to the current number of callbacks in both cases. While in the area,\n this commit also moves both assignments under -\u003enocb_lock.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 0bd55c693617cd2488378d011b66b92e1dd66ecf\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Mon Aug 12 10:28:08 2019 -0700\n\n rcu/nohz: Turn off tick for offloaded CPUs\n \n Historically, no-CBs CPUs allowed the scheduler-clock tick to be\n unconditionally disabled on any transition to idle or nohz_full userspace\n execution (see the rcu_needs_cpu() implementations). Unfortunately,\n the checks used by rcu_needs_cpu() are defeated now that no-CBs CPUs\n use -\u003ecblist, which might make users of battery-powered devices rather\n unhappy. This commit therefore adds explicit rcu_segcblist_is_offloaded()\n checks to return to the historical energy-efficient semantics.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 969974e5c51e717fc9070b00eb2f61ae589ed13d\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Wed May 22 09:35:11 2019 -0700\n\n rcu/nocb: Suppress uninitialized false-positive in nocb_gp_wait()\n \n Some compilers complain that wait_gp_seq might be used uninitialized\n in nocb_gp_wait(). This cannot actually happen because when wait_gp_seq\n is uninitialized, needwait_gp must be false, which prevents wait_gp_seq\n from being used. But this analysis is apparently beyond some compilers,\n so this commit adds a bogus initialization of wait_gp_seq for the sole\n purpose of suppressing the false-positive warning.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 921bb5fad11c0e8ec5f7625547552b252281f4de\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue May 21 13:53:28 2019 -0700\n\n rcu/nocb: Use build-time no-CBs check in rcu_pending()\n \n Currently, rcu_pending() invokes rcu_segcblist_is_offloaded() even\n in CONFIG_RCU_NOCB_CPU\u003dn kernels, which cannot possibly be offloaded.\n Given that rcu_pending() is on a fastpath, it makes sense to check for\n CONFIG_RCU_NOCB_CPU\u003dy before invoking rcu_segcblist_is_offloaded().\n This commit therefore makes this change.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit c1ab99d66ebcebedd9d416a840c488eaf079f3e9\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue May 21 13:39:15 2019 -0700\n\n rcu/nocb: Use build-time no-CBs check in rcu_core()\n \n Currently, rcu_core() invokes rcu_segcblist_is_offloaded() each time it\n needs to know whether the current CPU is a no-CBs CPU. Given that it is\n not possible to change the no-CBs status of a CPU after boot, and given\n that it is not possible to even have no-CBs CPUs in CONFIG_RCU_NOCB_CPU\u003dn\n kernels, this repeated runtime invocation wastes CPU. This commit\n therefore created a const on-stack variable to allow this check to be\n done only once per rcu_core() invocation.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit ec5ef87bac820be8ae9cc0a95108cded039ed8ef\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue May 21 13:03:49 2019 -0700\n\n rcu/nocb: Use build-time no-CBs check in rcu_do_batch()\n \n Currently, rcu_do_batch() invokes rcu_segcblist_is_offloaded() each time\n it needs to know whether the current CPU is a no-CBs CPU. Given that it\n is not possible to change the no-CBs status of a CPU after boot, and given\n that it is not possible to even have no-CBs CPUs in CONFIG_RCU_NOCB_CPU\u003dn\n kernels, this per-callback invocation wastes CPU. This commit therefore\n created a const on-stack variable to allow this check to be done only\n once per rcu_do_batch() invocation.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 4f9c1bc727f917c8c32ee1decc88e89057e0dffc\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue May 21 09:20:10 2019 -0700\n\n rcu/nocb: Remove obsolete nocb_gp_head and nocb_gp_tail fields\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 2a777de757f4c7050997c6232a585eff59c5ea36\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue May 21 09:10:24 2019 -0700\n\n rcu/nocb: Remove obsolete nocb_cb_tail and nocb_cb_head fields\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit c035280f1761b3336f4dad336906c19735d7ba5f\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue May 21 08:28:41 2019 -0700\n\n rcu/nocb: Remove obsolete nocb_q_count and nocb_q_count_lazy fields\n \n This commit removes the obsolete nocb_q_count and nocb_q_count_lazy\n fields, also removing rcu_get_n_cbs_nocb_cpu(), adjusting\n rcu_get_n_cbs_cpu(), and making rcutree_migrate_callbacks() once again\n disable the -\u003ecblist fields of offline CPUs.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit e7f4c5b3998a3cf1bd8dbf110948075b47ac9b78\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue May 21 07:18:00 2019 -0700\n\n rcu/nocb: Remove obsolete nocb_head and nocb_tail fields\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 5d6742b37727e111f4755155e59c5319cf5caa7b\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Wed May 15 09:56:40 2019 -0700\n\n rcu/nocb: Use rcu_segcblist for no-CBs CPUs\n \n Currently the RCU callbacks for no-CBs CPUs are queued on a series of\n ad-hoc linked lists, which means that these callbacks cannot benefit\n from \"drive-by\" grace periods, thus suffering needless delays prior\n to invocation. In addition, the no-CBs grace-period kthreads first\n wait for callbacks to appear and later wait for a new grace period,\n which means that callbacks appearing during a grace-period wait can\n be delayed. These delays increase memory footprint, and could even\n result in an out-of-memory condition.\n \n This commit therefore enqueues RCU callbacks from no-CBs CPUs on the\n rcu_segcblist structure that is already used by non-no-CBs CPUs. It also\n restructures the no-CBs grace-period kthread to be checking for incoming\n callbacks while waiting for grace periods. Also, instead of waiting\n for a new grace period, it waits for the closest grace period that will\n cause some of the callbacks to be safe to invoke. All of these changes\n reduce callback latency and thus the number of outstanding callbacks,\n in turn reducing the probability of an out-of-memory condition.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit e83e73f5b0f8de6a8978ba64185e80fdf48a2a63\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue May 14 09:50:49 2019 -0700\n\n rcu/nocb: Leave -\u003ecblist enabled for no-CBs CPUs\n \n As a first step towards making no-CBs CPUs use the -\u003ecblist, this commit\n leaves the -\u003ecblist enabled for these CPUs. The main reason to make\n no-CBs CPUs use -\u003ecblist is to take advantage of callback numbering,\n which will reduce the effects of missed grace periods which in turn will\n reduce forward-progress problems for no-CBs CPUs.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit e6060b41c9955374079926a7612b857a8458ed1f\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Mon May 13 15:57:50 2019 -0700\n\n rcu/nocb: Allow lockless use of rcu_segcblist_empty()\n \n Currently, rcu_segcblist_empty() assumes that the callback list is not\n being changed by other CPUs, but upcoming changes will require it to\n operate locklessly. This commit therefore adds the needed READ_ONCE()\n call, along with the WRITE_ONCE() calls when updating the callback list\u0027s\n -\u003ehead field.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 76c6927c3ee443e756f2c0c9f992cb04b26c65f2\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Mon May 13 14:36:11 2019 -0700\n\n rcu/nocb: Allow lockless use of rcu_segcblist_restempty()\n \n Currently, rcu_segcblist_restempty() assumes that the callback list\n is not being changed by other CPUs, but upcoming changes will require\n it to operate locklessly. This commit therefore adds the needed\n READ_ONCE() calls, along with the WRITE_ONCE() calls when updating\n the callback list.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit ca5c8258081178c60b27e3532d9ea95b6eaa7040\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue Apr 16 15:15:24 2019 -0700\n\n rcu/nocb: Remove deferred wakeup checks for extended quiescent states\n \n The idea behind the checks for extended quiescent states at the end of\n __call_rcu_nocb() is to handle cases where call_rcu() is invoked directly\n from within an extended quiescent state, for example, from the idle loop.\n However, this will result in a timer-mediated deferred wakeup, which\n will cause the needed wakeup to happen within a jiffy or thereabouts.\n There should be no forward-progress concerns, and if there are, the proper\n response is to exit the extended quiescent state while executing the\n endless blast of call_rcu() invocations, for example, using RCU_NONIDLE().\n Given the more realistic case of an isolated call_rcu() invocation, there\n should be no problem.\n \n This commit therefore removes the checks for invoking call_rcu() within\n an extended quiescent state for on no-CBs CPUs.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 85f69b32126dcf798f2c8d69a7957ba990bc2e02\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue Apr 16 14:48:28 2019 -0700\n\n rcu/nocb: Check for deferred nocb wakeups before nohz_full early exit\n \n In theory, a timer is used to defer wakeups of no-CBs grace-period\n kthreads when the wakeup cannot be done safely directly from the\n call_rcu(). In practice, the one-jiffy delay is not always consistent\n with timely callback invocation under heavy call_rcu() loads. Therefore,\n there are a number of checks for a pending deferred wakeup, including\n from the scheduling-clock interrupt. Unfortunately, this check follows\n the rcu_nohz_full_cpu() early exit, which renders it useless on such CPUs.\n \n This commit therefore moves the check for the pending deferred no-CB\n wakeup to precede the rcu_nohz_full_cpu() early exit.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit c00045be32fe13333ba8c62748ba04747c182838\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue Apr 16 14:09:15 2019 -0700\n\n rcu/nocb: Make rcutree_migrate_callbacks() start at leaf rcu_node structure\n \n Because rcutree_migrate_callbacks() is invoked infrequently and because\n an exact snapshot of the grace-period state might save some callbacks a\n second trip through a grace period, this function has used the root\n rcu_node structure. However, this safe-second-trip optimization\n happens only if rcutree_migrate_callbacks() races with grace-period\n initialization, so it is not worth the added mental load. This commit\n therefore makes rcutree_migrate_callbacks() start with the leaf rcu_node\n structures, as is done elsewhere.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 750d7f6a434ff4640fa825dfb1eccb44e79fb6af\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue Apr 16 08:19:43 2019 -0700\n\n rcu/nocb: Add checks for offloaded callback processing\n \n This commit is a preparatory patch for offloaded callbacks using the\n same -\u003ecblist structure used by non-offloaded callbacks. It therefore\n adds rcu_segcblist_is_offloaded() calls where they will be needed when\n !rcu_segcblist_is_enabled() no longer flags the offloaded case. It also\n adds checks in rcu_do_batch() to ensure that there are no missed checks:\n Currently, it should not be possible for offloaded execution to reach\n rcu_do_batch(), though this will change later in this series.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit ce5215c1342c6c89b3c3c45fea82cddf0b013787\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Fri Apr 12 15:58:34 2019 -0700\n\n rcu/nocb: Use separate flag to indicate offloaded -\u003ecblist\n \n RCU callback processing currently uses rcu_is_nocb_cpu() to determine\n whether or not the current CPU\u0027s callbacks are to be offloaded.\n This works, but it is not so good for cache locality. Plus use of\n -\u003ecblist for offloaded callbacks will greatly increase the frequency\n of these checks. This commit therefore adds a -\u003eoffloaded flag to the\n rcu_segcblist structure to provide a more flexible and cache-friendly\n means of checking for callback offloading.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 1bb5f9b95afe5d9d6b586389ce5e8f461a5b671c\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Fri Apr 12 12:34:41 2019 -0700\n\n rcu/nocb: Use separate flag to indicate disabled -\u003ecblist\n \n NULLing the RCU_NEXT_TAIL pointer was a clever way to save a byte, but\n forward-progress considerations would require that this pointer be both\n NULL and non-NULL, which, absent a quantum-computer port of the Linux\n kernel, simply won\u0027t happen. This commit therefore creates as separate\n -\u003eenabled flag to replace the current NULL checks.\n \n [ paulmck: Add include files per 0day test robot and -next. ]\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 18cd8c93e69e3853eb408980089fb3c58813f922\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Sat Jun 1 05:12:36 2019 -0700\n\n rcu/nocb: Print gp/cb kthread hierarchy if dump_tree\n \n This commit causes the no-CBs grace-period/callback hierarchy to be\n printed to the console when the dump_tree kernel boot parameter is set.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit f7c612b000d7e974826089b5a6f6eecd6805862a\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue Apr 2 08:05:55 2019 -0700\n\n rcu/nocb: Rename rcu_nocb_leader_stride kernel boot parameter\n \n This commit changes the name of the rcu_nocb_leader_stride kernel\n boot parameter to rcu_nocb_gp_stride in order to account for the new\n distinction between callback and grace-period no-CBs kthreads.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit f7c9a9b664fb32a127e8e9a987b52023b92c3a0b\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Mon Apr 1 09:57:01 2019 -0700\n\n rcu/nocb: Rename and document no-CB CB kthread sleep trace event\n \n The nocb_cb_wait() function traces a \"FollowerSleep\" trace_rcu_nocb_wake()\n event, which never was documented and is now misleading. This commit\n therefore changes \"FollowerSleep\" to \"CBSleep\", documents this, and\n updates the documentation for \"Sleep\" as well.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 0bdc33daef96a54f9e5799d84f2fbc05d9e5cae3\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Sun Mar 31 16:20:52 2019 -0700\n\n rcu/nocb: Rename rcu_organize_nocb_kthreads() local variable\n \n This commit renames rdp_leader to rdp_gp in order to account for the\n new distinction between callback and grace-period no-CBs kthreads.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 0d52a6652f15f6b1155297326de85c07ca421d64\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Sun Mar 31 16:19:02 2019 -0700\n\n rcu/nocb: Rename wake_nocb_leader_defer() to wake_nocb_gp_defer()\n \n This commit adjusts naming to account for the new distinction between\n callback and grace-period no-CBs kthreads.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 5f675ba6eb5d1e4be56fc7c28881728373d880c4\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Sun Mar 31 16:11:57 2019 -0700\n\n rcu/nocb: Rename __wake_nocb_leader() to __wake_nocb_gp()\n \n This commit adjusts naming to account for the new distinction between\n callback and grace-period no-CBs kthreads. While in the area, it also\n updates local variables.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 5d62c08c5fe54492899b978d0ddc0bf7fd071317\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Sun Mar 31 16:10:17 2019 -0700\n\n rcu/nocb: Rename wake_nocb_leader() to wake_nocb_gp()\n \n This commit adjusts naming to account for the new distinction between\n callback and grace-period no-CBs kthreads.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 9fa471a881df9ba84e0e0844d918ed1ec55fc567\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Sun Mar 31 16:07:43 2019 -0700\n\n rcu/nocb: Rename nocb_follower_wait() to nocb_cb_wait()\n \n This commit adjusts naming to account for the new distinction between\n callback and grace-period no-CBs kthreads.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 12f54c3a8410102afb96ed437aebe7f1d87f399f\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Fri Mar 29 16:43:51 2019 -0700\n\n rcu/nocb: Provide separate no-CBs grace-period kthreads\n \n Currently, there is one no-CBs rcuo kthread per CPU, and these kthreads\n are divided into groups. The first rcuo kthread to come online in a\n given group is that group\u0027s leader, and the leader both waits for grace\n periods and invokes its CPU\u0027s callbacks. The non-leader rcuo kthreads\n only invoke callbacks.\n \n This works well in the real-time/embedded environments for which it was\n intended because such environments tend not to generate all that many\n callbacks. However, given huge floods of callbacks, it is possible for\n the leader kthread to be stuck invoking callbacks while its followers\n wait helplessly while their callbacks pile up. This is a good recipe\n for an OOM, and rcutorture\u0027s new callback-flood capability does generate\n such OOMs.\n \n One strategy would be to wait until such OOMs start happening in\n production, but similar OOMs have in fact happened starting in 2018.\n It would therefore be wise to take a more proactive approach.\n \n This commit therefore features per-CPU rcuo kthreads that do nothing\n but invoke callbacks. Instead of having one of these kthreads act as\n leader, each group has a separate rcog kthread that handles grace periods\n for its group. Because these rcuog kthreads do not invoke callbacks,\n callback floods on one CPU no longer block callbacks from reaching the\n rcuc callback-invocation kthreads on other CPUs.\n \n This change does introduce additional kthreads, however:\n \n 1. The number of additional kthreads is about the square root of\n the number of CPUs, so that a 4096-CPU system would have only\n about 64 additional kthreads. Note that recent changes\n decreased the number of rcuo kthreads by a factor of two\n (CONFIG_PREEMPT\u003dn) or even three (CONFIG_PREEMPT\u003dy), so\n this still represents a significant improvement on most systems.\n \n 2. The leading \"rcuo\" of the rcuog kthreads should allow existing\n scripting to affinity these additional kthreads as needed, the\n same as for the rcuop and rcuos kthreads. (There are no longer\n any rcuob kthreads.)\n \n 3. A state-machine approach was considered and rejected. Although\n this would allow the rcuo kthreads to continue their dual\n leader/follower roles, it complicates callback invocation\n and makes it more difficult to consolidate rcuo callback\n invocation with existing softirq callback invocation.\n \n The introduction of rcuog kthreads should thus be acceptable.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 6484fe54b5c64e9a388f369001508ab8df85a646\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Thu Mar 28 15:44:18 2019 -0700\n\n rcu/nocb: Update comments to prepare for forward-progress work\n \n This commit simply rewords comments to prepare for leader nocb kthreads\n doing only grace-period work and callback shuffling. This will mean\n the addition of replacement kthreads to invoke callbacks. The \"leader\"\n and \"follower\" thus become less meaningful, so the commit changes no-CB\n comments with these strings to \"GP\" and \"CB\", respectively. (Give or\n take the usual grammatical transformations.)\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 58bf6f77c6fb0abe8e1330d8375dddd52711ef4c\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Thu Mar 28 15:33:59 2019 -0700\n\n rcu/nocb: Rename rcu_data fields to prepare for forward-progress work\n \n This commit simply renames rcu_data fields to prepare for leader\n nocb kthreads doing only grace-period work and callback shuffling.\n This will mean the addition of replacement kthreads to invoke callbacks.\n The \"leader\" and \"follower\" thus become less meaningful, so the commit\n changes no-CB fields with these strings to \"gp\" and \"cb\", respectively.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit bee6f87166e9c6b8d81a7570995bd637e8da485a\nAuthor: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\nDate: Tue Jul 16 18:12:28 2019 -0400\n\n acpi: Use built-in RCU list checking for acpi_ioremaps list\n \n This commit applies the consolidated list_for_each_entry_rcu() support\n for lockdep conditions.\n \n Acked-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Signed-off-by: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 842a56cf3eb00f717f9522766c0e7b71bafd5fc1\nAuthor: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\nDate: Tue Jul 16 18:12:27 2019 -0400\n\n x86/pci: Pass lockdep condition to pcm_mmcfg_list iterator\n \n The pcm_mmcfg_list is traversed by list_for_each_entry_rcu() outside\n of an RCU read-side critical section, which is safe because the\n pci_mmcfg_lock is held. This commit therefore adds a lockdep expression\n to list_for_each_entry_rcu() in order t avoid lockdep warnings.\n \n Acked-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Signed-off-by: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit c2fa1e1bfa5b74558854a70b8afd797d43eb2743\nAuthor: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\nDate: Tue Jul 16 18:12:25 2019 -0400\n\n driver/core: Convert to use built-in RCU list checking\n \n This commit applies the consolidated hlist_for_each_entry_rcu() support\n for lockdep conditions.\n \n Acked-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit a1916a153c254cd0ad13b23bed262ed56922cc7a\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Tue Aug 13 11:54:43 2019 -0700\n\n libbpf: attempt to load kernel BTF from sysfs first\n \n Add support for loading kernel BTF from sysfs (/sys/kernel/btf/vmlinux)\n as a target BTF. Also extend the list of on disk search paths for\n vmlinux ELF image with entries that perf is searching for.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 7fd785685e2243bb639b31557e258d11464c3489\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Tue Aug 13 11:54:42 2019 -0700\n\n btf: rename /sys/kernel/btf/kernel into /sys/kernel/btf/vmlinux\n \n Expose kernel\u0027s BTF under the name vmlinux to be more uniform with using\n kernel module names as file names in the future.\n \n Fixes: 341dfcf8d78e (\"btf: expose BTF info through sysfs\")\n Suggested-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 5e4e6c4d3ae0ccabd99ee6f8f48154fb2f59683b\nAuthor: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\nDate: Tue Jul 23 17:23:38 2019 +0300\n\n interconnect: qcom: Add QCS404 interconnect provider driver\n \n Add driver for the interconnect buses found in Qualcomm QCS404-based\n platforms. The topology consists of three NoCs that are controlled by\n a remote processor. This remote processor collects the aggregated\n bandwidth for each master-slave pairs.\n \n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Signed-off-by: Georgi Djakov \u003cgeorgi.djakov@linaro.org\u003e\n\ncommit 3a0f1604888a6ab4ac60ca3d4716dd7fc18e16d3\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Aug 13 14:29:16 2019 +0100\n\n drm/i915/gt: Save/restore interrupts around breadcrumb disable\n \n Stop assuming we only get called with irqs-on for disarming the\n breadcrumbs, and do a full save/restore spin_lock_irq.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813132916.20382-2-chris@chris-wilson.co.uk\n\ncommit a79ca656b648e6851dd2f1eb0dcef86b278a8908\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Aug 13 20:07:05 2019 +0100\n\n drm/i915: Push the wakeref-\u003ecount deferral to the backend\n \n If the backend wishes to defer the wakeref parking, make it responsible\n for unlocking the wakeref (i.e. bumping the counter). This allows it to\n time the unlock much more carefully in case it happens to needs the\n wakeref to be active during its deferral.\n \n For instance, during engine parking we may choose to emit an idle\n barrier (a request). To do so, we borrow the engine-\u003ekernel_context\n timeline and to ensure exclusive access we keep the\n engine-\u003ewakeref.count as 0. However, to submit that request to HW may\n require a intel_engine_pm_get() (e.g. to keep the submission tasklet\n alive) and before we allow that we have to rewake our wakeref to avoid a\n recursive deadlock.\n \n \u003c4\u003e [257.742916] IRQs not enabled as expected\n \u003c4\u003e [257.742930] WARNING: CPU: 0 PID: 0 at kernel/softirq.c:169 __local_bh_enable_ip+0xa9/0x100\n \u003c4\u003e [257.742936] Modules linked in: vgem snd_hda_codec_hdmi snd_hda_codec_realtek snd_hda_codec_generic i915 btusb btrtl btbcm btintel snd_hda_intel snd_intel_nhlt bluetooth snd_hda_codec coretemp snd_hwdep crct10dif_pclmul snd_hda_core crc32_pclmul ecdh_generic ecc ghash_clmulni_intel snd_pcm r8169 realtek lpc_ich prime_numbers i2c_hid\n \u003c4\u003e [257.742991] CPU: 0 PID: 0 Comm: swapper/0 Tainted: G U W 5.3.0-rc3-g5d0a06cd532c-drmtip_340+ #1\n \u003c4\u003e [257.742998] Hardware name: GIGABYTE GB-BXBT-1900/MZBAYAB-00, BIOS F6 02/17/2015\n \u003c4\u003e [257.743008] RIP: 0010:__local_bh_enable_ip+0xa9/0x100\n \u003c4\u003e [257.743017] Code: 37 5b 5d c3 8b 80 50 08 00 00 85 c0 75 a9 80 3d 0b be 25 01 00 75 a0 48 c7 c7 f3 0c 06 ac c6 05 fb bd 25 01 01 e8 77 84 ff ff \u003c0f\u003e 0b eb 89 48 89 ef e8 3b 41 06 00 eb 98 e8 e4 5c f4 ff 5b 5d c3\n \u003c4\u003e [257.743025] RSP: 0018:ffffa78600003cb8 EFLAGS: 00010086\n \u003c4\u003e [257.743035] RAX: 0000000000000000 RBX: 0000000000000200 RCX: 0000000000010302\n \u003c4\u003e [257.743042] RDX: 0000000080010302 RSI: 0000000000000000 RDI: 00000000ffffffff\n \u003c4\u003e [257.743050] RBP: ffffffffc0494bb3 R08: 0000000000000000 R09: 0000000000000001\n \u003c4\u003e [257.743058] R10: 0000000014c8f0e9 R11: 00000000fee2ff8e R12: ffffa23ba8c38008\n \u003c4\u003e [257.743065] R13: ffffa23bacc579c0 R14: ffffa23bb7db0f60 R15: ffffa23b9cc8c430\n \u003c4\u003e [257.743074] FS: 0000000000000000(0000) GS:ffffa23bbba00000(0000) knlGS:0000000000000000\n \u003c4\u003e [257.743082] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n \u003c4\u003e [257.743089] CR2: 00007fe477b20778 CR3: 000000011f72a000 CR4: 00000000001006f0\n \u003c4\u003e [257.743096] Call Trace:\n \u003c4\u003e [257.743104] \u003cIRQ\u003e\n \u003c4\u003e [257.743265] __i915_request_commit+0x240/0x5d0 [i915]\n \u003c4\u003e [257.743427] ? __i915_request_create+0x228/0x4c0 [i915]\n \u003c4\u003e [257.743584] __engine_park+0x64/0x250 [i915]\n \u003c4\u003e [257.743730] ____intel_wakeref_put_last+0x1c/0x70 [i915]\n \u003c4\u003e [257.743878] i915_sample+0x2ee/0x310 [i915]\n \u003c4\u003e [257.744030] ? i915_pmu_cpu_offline+0xb0/0xb0 [i915]\n \u003c4\u003e [257.744040] __hrtimer_run_queues+0x11e/0x4b0\n \u003c4\u003e [257.744068] hrtimer_interrupt+0xea/0x250\n \u003c4\u003e [257.744079] ? lockdep_hardirqs_off+0x79/0xd0\n \u003c4\u003e [257.744101] smp_apic_timer_interrupt+0x96/0x280\n \u003c4\u003e [257.744114] apic_timer_interrupt+0xf/0x20\n \u003c4\u003e [257.744125] RIP: 0010:__do_softirq+0xb3/0x4ae\n \n v2: Keep the priority_hint assert\n v3: That assert was desperately trying to point out my bug. Sorry, little\n assert.\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111378\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813190705.23869-1-chris@chris-wilson.co.uk\n\ncommit be06f8e7425dcf554ebc1c0f78fb286bbbfbe23a\nAuthor: Georgi Djakov \u003cgeorgi.djakov@linaro.org\u003e\nDate: Tue Jul 23 17:23:37 2019 +0300\n\n interconnect: qcom: Add interconnect RPM over SMD driver\n \n On some Qualcomm SoCs, there is a remote processor, which controls some of\n the Network-On-Chip interconnect resources. Other CPUs express their needs\n by communicating with this processor. Add a driver to handle communication\n with this remote processor.\n \n Reviewed-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Signed-off-by: Georgi Djakov \u003cgeorgi.djakov@linaro.org\u003e\n\ncommit 24f516ebbab8a212a9aa8c3d69f185371f5e200b\nAuthor: Georgi Djakov \u003cgeorgi.djakov@linaro.org\u003e\nDate: Tue Jul 23 17:23:35 2019 +0300\n\n dt-bindings: interconnect: Add Qualcomm QCS404 DT bindings\n \n The Qualcomm QCS404 platform has several buses that could be controlled\n and tuned according to the bandwidth demand.\n \n Reviewed-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Georgi Djakov \u003cgeorgi.djakov@linaro.org\u003e\n\ncommit 9e3ce77c116374556d2fb2728bc9e24c67362dd6\nAuthor: David Dai \u003cdaidavid1@codeaurora.org\u003e\nDate: Tue Aug 13 17:53:41 2019 +0300\n\n interconnect: qcom: Add tagging and wake/sleep support for sdm845\n \n Add support for wake and sleep commands by using a tag to indicate\n whether or not the aggregate and set requests fall into execution\n state specific bucket.\n \n Signed-off-by: David Dai \u003cdaidavid1@codeaurora.org\u003e\n Reviewed-by: Evan Green \u003cevgreen@chromium.org\u003e\n Signed-off-by: Georgi Djakov \u003cgeorgi.djakov@linaro.org\u003e\n\ncommit cbd5a9c28bb5d2560be34fc6a2b6e60197628433\nAuthor: Georgi Djakov \u003cgeorgi.djakov@linaro.org\u003e\nDate: Fri Aug 9 15:13:24 2019 +0300\n\n interconnect: Add pre_aggregate() callback\n \n Introduce an optional callback in interconnect provider drivers. It can be\n used for implementing actions, that need to be executed before the actual\n aggregation of the bandwidth requests has started.\n \n The benefit of this for now is that it will significantly simplify the code\n in provider drivers.\n \n Suggested-by: Evan Green \u003cevgreen@chromium.org\u003e\n Reviewed-by: Evan Green \u003cevgreen@chromium.org\u003e\n Signed-off-by: Georgi Djakov \u003cgeorgi.djakov@linaro.org\u003e\n\ncommit 127ab2cc5f19692efe422935267b9db0845b2b04\nAuthor: Georgi Djakov \u003cgeorgi.djakov@linaro.org\u003e\nDate: Fri Aug 9 15:13:23 2019 +0300\n\n interconnect: Add support for path tags\n \n Consumers may have use cases with different bandwidth requirements based\n on the system or driver state. The consumer driver can append a specific\n tag to the path and pass this information to the interconnect platform\n driver to do the aggregation based on this state.\n \n Introduce icc_set_tag() function that will allow the consumers to append\n an optional tag to each path. The aggregation of these tagged paths is\n platform specific.\n \n Reviewed-by: Evan Green \u003cevgreen@chromium.org\u003e\n Signed-off-by: Georgi Djakov \u003cgeorgi.djakov@linaro.org\u003e\n\ncommit 37143dcc44f8f3348d47616598137b1580468973\nAuthor: Mark Rutland \u003cmark.rutland@arm.com\u003e\nDate: Tue Aug 13 15:16:39 2019 +0100\n\n arm64: constify sys64_hook instances\n \n All instances of struct sys64_hook contain compile-time constant data,\n and are never inentionally modified, so let\u0027s make them all const.\n \n Signed-off-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Cc: Will Deacon \u003cwill.deacon@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 0da23df2ff043c8d39b389e32ee68af64b5f408e\nAuthor: Mark Rutland \u003cmark.rutland@arm.com\u003e\nDate: Tue Aug 13 15:16:38 2019 +0100\n\n arm64: constify aarch64_insn_encoding_class[]\n \n The aarch64_insn_encoding_class[] array contains compile-time constant\n data, and is never intentionally modified, so let\u0027s mark it as const.\n \n Signed-off-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Cc: Will Deacon \u003cwill.deacon@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 0bf136a8cf11502643c27c8de1b1db8f161a31ed\nAuthor: Mark Rutland \u003cmark.rutland@arm.com\u003e\nDate: Tue Aug 13 15:16:37 2019 +0100\n\n arm64: constify icache_policy_str[]\n \n The icache_policy_str[] array contains compile-time constant data, and\n is never intentionally modified, so let\u0027s mark it as const.\n \n Signed-off-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Cc: Will Deacon \u003cwill.deacon@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 80d838122643a09a9f99824adea4b4261e4451e6\nAuthor: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nDate: Mon Aug 12 14:50:45 2019 -0700\n\n arm64: prefer __section from compiler_attributes.h\n \n GCC unescapes escaped string section names while Clang does not. Because\n __section uses the `#` stringification operator for the section name, it\n doesn\u0027t need to be escaped.\n \n This antipattern was found with:\n $ grep -e __section\\(\\\" -e __section__\\(\\\" -r\n \n Reported-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Suggested-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Signed-off-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 1cdd8705c7acad7a49d0546d0f22fcde3ad8dcb9\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Mon Aug 12 10:54:05 2019 -0700\n\n drm/i915/tgl: Fix missing parentheses on TGL_TRANS_DDI_FUNC_CTL_VAL_TO_PORT\n \n In this case we want to apply the mask and then shift so the\n parentheses is needed.\n \n SPANK! SPANK! SPANK! Naughty programmer!\n \n Fixes: 9749a5b6c09f (\"drm/i915/tgl: Fix the read of the DDI that transcoder is attached to\")\n Cc: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812175405.14479-1-jose.souza@intel.com\n\ncommit 77905379e9b28a490051b15239bcaf6458d171ee\nAuthor: Lang Cheng \u003cchenglang@huawei.com\u003e\nDate: Fri Aug 9 17:41:04 2019 +0800\n\n RDMA/hns: Remove unuseful member\n \n For struct hns_roce_cmdq, toggle is unused and\n delete it.\n \n Signed-off-by: Lang Cheng \u003cchenglang@huawei.com\u003e\n Link: https://lore.kernel.org/r/1565343666-73193-8-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit bf8c02f961c89e5ccae5987b7ab28f5592a35101\nAuthor: Xi Wang \u003cwangxi11@huawei.com\u003e\nDate: Fri Aug 9 17:41:03 2019 +0800\n\n RDMA/hns: bugfix for slab-out-of-bounds when loading hip08 driver\n \n kasan will report a BUG when run command \u0027insmod hns_roce_hw_v2.ko\u0027, the\n calltrace is as follows:\n \n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n BUG: KASAN: slab-out-of-bounds in hns_roce_v2_init_eq_table+0x1324/0x1948\n [hns_roce_hw_v2]\n Read of size 8 at addr ffff8020e7a10608 by task insmod/256\n \n CPU: 0 PID: 256 Comm: insmod Tainted: G O 5.2.0-rc4 #1\n Hardware name: Huawei D06 /D06, BIOS Hisilicon D06 UEFI RC0\n Call trace:\n dump_backtrace+0x0/0x1e8\n show_stack+0x14/0x20\n dump_stack+0xc4/0xfc\n print_address_description+0x60/0x270\n __kasan_report+0x164/0x1b8\n kasan_report+0xc/0x18\n __asan_load8+0x84/0xa8\n hns_roce_v2_init_eq_table+0x1324/0x1948 [hns_roce_hw_v2]\n hns_roce_init+0xf8/0xfe0 [hns_roce]\n __hns_roce_hw_v2_init_instance+0x284/0x330 [hns_roce_hw_v2]\n hns_roce_hw_v2_init_instance+0xd0/0x1b8 [hns_roce_hw_v2]\n hclge_init_roce_client_instance+0x180/0x310 [hclge]\n hclge_init_client_instance+0xcc/0x508 [hclge]\n hnae3_init_client_instance.part.3+0x3c/0x80 [hnae3]\n hnae3_register_client+0x134/0x1a8 [hnae3]\n hns_roce_hw_v2_init+0x14/0x10000 [hns_roce_hw_v2]\n do_one_initcall+0x9c/0x3e0\n do_init_module+0xd4/0x2d8\n load_module+0x3284/0x3690\n __se_sys_init_module+0x274/0x308\n __arm64_sys_init_module+0x40/0x50\n el0_svc_handler+0xbc/0x210\n el0_svc+0x8/0xc\n \n Allocated by task 256:\n __kasan_kmalloc.isra.0+0xd0/0x180\n kasan_kmalloc+0xc/0x18\n __kmalloc+0x16c/0x328\n hns_roce_v2_init_eq_table+0x764/0x1948 [hns_roce_hw_v2]\n hns_roce_init+0xf8/0xfe0 [hns_roce]\n __hns_roce_hw_v2_init_instance+0x284/0x330 [hns_roce_hw_v2]\n hns_roce_hw_v2_init_instance+0xd0/0x1b8 [hns_roce_hw_v2]\n hclge_init_roce_client_instance+0x180/0x310 [hclge]\n hclge_init_client_instance+0xcc/0x508 [hclge]\n hnae3_init_client_instance.part.3+0x3c/0x80 [hnae3]\n hnae3_register_client+0x134/0x1a8 [hnae3]\n hns_roce_hw_v2_init+0x14/0x10000 [hns_roce_hw_v2]\n do_one_initcall+0x9c/0x3e0\n do_init_module+0xd4/0x2d8\n load_module+0x3284/0x3690\n __se_sys_init_module+0x274/0x308\n __arm64_sys_init_module+0x40/0x50\n el0_svc_handler+0xbc/0x210\n el0_svc+0x8/0xc\n \n Freed by task 0:\n (stack is not available)\n \n The buggy address belongs to the object at ffff8020e7a10600\n which belongs to the cache kmalloc-128 of size 128\n The buggy address is located 8 bytes inside of\n 128-byte region [ffff8020e7a10600, ffff8020e7a10680)\n The buggy address belongs to the page:\n page:ffff7fe00839e840 refcount:1 mapcount:0 mapping:ffff802340020200 index:0x0\n flags: 0x5fffe00000000200(slab)\n raw: 5fffe00000000200 dead000000000100 dead000000000200 ffff802340020200\n raw: 0000000000000000 0000000081000100 00000001ffffffff 0000000000000000\n page dumped because: kasan: bad access detected\n \n Memory state around the buggy address:\n ffff8020e7a10500: 00 00 00 00 00 00 00 00 fc fc fc fc fc fc fc fc\n ffff8020e7a10580: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc\n \u003effff8020e7a10600: 00 fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc\n ^\n ffff8020e7a10680: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc\n ffff8020e7a10700: 00 fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc\n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n Disabling lock debugging due to kernel taint\n \n Fixes: a5073d6054f7 (\"RDMA/hns: Add eq support of hip08\")\n \n Signed-off-by: Xi Wang \u003cwangxi11@huawei.com\u003e\n Link: https://lore.kernel.org/r/1565343666-73193-7-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 9bba3f0cbfc8abf2e1549ea03c0128186081d7a8\nAuthor: Xi Wang \u003cwangxi11@huawei.com\u003e\nDate: Fri Aug 9 17:41:02 2019 +0800\n\n RDMA/hns: Bugfix for slab-out-of-bounds when unloading hip08 driver\n \n kasan will report a BUG when run command \u0027rmmod hns_roce_hw_v2\u0027, the calltrace\n is as follows:\n \n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n BUG: KASAN: slab-out-of-bounds in hns_roce_table_mhop_put+0x584/0x828\n [hns_roce]\n Read of size 8 at addr ffff802185e08300 by task rmmod/270\n \n Call trace:\n dump_backtrace+0x0/0x1e8\n show_stack+0x14/0x20\n dump_stack+0xc4/0xfc\n print_address_description+0x60/0x270\n __kasan_report+0x164/0x1b8\n kasan_report+0xc/0x18\n __asan_load8+0x84/0xa8\n hns_roce_table_mhop_put+0x584/0x828 [hns_roce]\n hns_roce_table_put+0x174/0x1a0 [hns_roce]\n hns_roce_mr_free+0x124/0x210 [hns_roce]\n hns_roce_dereg_mr+0x90/0xb8 [hns_roce]\n ib_dealloc_pd_user+0x60/0xf0\n ib_mad_port_close+0x128/0x1d8\n ib_mad_remove_device+0x94/0x118\n remove_client_context+0xa0/0xe0\n disable_device+0xfc/0x1c0\n __ib_unregister_device+0x60/0xe0\n ib_unregister_device+0x24/0x38\n hns_roce_exit+0x3c/0x138 [hns_roce]\n __hns_roce_hw_v2_uninit_instance.isra.30+0x28/0x50 [hns_roce_hw_v2]\n hns_roce_hw_v2_uninit_instance+0x44/0x60 [hns_roce_hw_v2]\n hclge_uninit_client_instance+0x15c/0x238 [hclge]\n hnae3_uninit_client_instance+0x84/0xa8 [hnae3]\n hnae3_unregister_client+0x84/0x158 [hnae3]\n hns_roce_hw_v2_exit+0x14/0x20 [hns_roce_hw_v2]\n __arm64_sys_delete_module+0x20c/0x308\n el0_svc_handler+0xbc/0x210\n el0_svc+0x8/0xc\n \n Allocated by task 255:\n __kasan_kmalloc.isra.0+0xd0/0x180\n kasan_kmalloc+0xc/0x18\n __kmalloc+0x16c/0x328\n hns_roce_init_hem_table+0x20c/0x428 [hns_roce]\n hns_roce_init+0x214/0xfe0 [hns_roce]\n __hns_roce_hw_v2_init_instance+0x284/0x330 [hns_roce_hw_v2]\n hns_roce_hw_v2_init_instance+0xd0/0x1b8 [hns_roce_hw_v2]\n hclge_init_roce_client_instance+0x180/0x310 [hclge]\n hclge_init_client_instance+0xcc/0x508 [hclge]\n hnae3_init_client_instance.part.3+0x3c/0x80 [hnae3]\n hnae3_register_client+0x134/0x1a8 [hnae3]\n 0xffff200009c00014\n do_one_initcall+0x9c/0x3e0\n do_init_module+0xd4/0x2d8\n load_module+0x3284/0x3690\n __se_sys_init_module+0x274/0x308\n __arm64_sys_init_module+0x40/0x50\n el0_svc_handler+0xbc/0x210\n el0_svc+0x8/0xc\n \n Freed by task 0:\n (stack is not available)\n \n The buggy address belongs to the object at ffff802185e06300\n which belongs to the cache kmalloc-8k of size 8192\n The buggy address is located 0 bytes to the right of\n 8192-byte region [ffff802185e06300, ffff802185e08300)\n The buggy address belongs to the page:\n page:ffff7fe008617800 refcount:1 mapcount:0 mapping:ffff802340020e00 index:0x0\n compound_mapcount: 0\n flags: 0x5fffe00000010200(slab|head)\n raw: 5fffe00000010200 dead000000000100 dead000000000200 ffff802340020e00\n raw: 0000000000000000 00000000803e003e 00000001ffffffff 0000000000000000\n page dumped because: kasan: bad access detected\n \n Memory state around the buggy address:\n ffff802185e08200: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00\n ffff802185e08280: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00\n \u003effff802185e08300: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc\n ^\n ffff802185e08380: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc\n ffff802185e08400: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00\n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n Disabling lock debugging due to kernel taint\n \n Fixes: a25d13cbe816 (\"RDMA/hns: Add the interfaces to support multi hop addressing for the contexts in hip08\")\n \n Signed-off-by: Xi Wang \u003cwangxi11@huawei.com\u003e\n Link: https://lore.kernel.org/r/1565343666-73193-6-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit d7e5ca88d60c0b5e166e33bfe356e961824ce0f4\nAuthor: Yangyang Li \u003cliyangyang20@huawei.com\u003e\nDate: Fri Aug 9 17:41:01 2019 +0800\n\n RDMA/hns: Modify pi vlaue when cq overflows\n \n When exiting \"for loop\", the actual value of pi will be\n increased by 1, which is compatible with the next calculation.\n But when pi is equal to \"ci + hr_cq-\u003e ib_cq.cqe\", the \"break\"\n was called and the pi is actual value, it will lead one cqe\n still existing, so the \"\u003d\u003d\" should be modify to \"\u003e\".\n \n Signed-off-by: Yangyang Li \u003cliyangyang20@huawei.com\u003e\n Link: https://lore.kernel.org/r/1565343666-73193-5-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 76827087bb3f82f3cc18871633e06a4df36a5ed9\nAuthor: Lijun Ou \u003coulijun@huawei.com\u003e\nDate: Fri Aug 9 17:40:59 2019 +0800\n\n RDMA/hns: Bugfix for creating qp attached to srq\n \n When create a qp and attached to srq, rq will no longer be used\n and the members of rq will be set zero. As a result, the wrid\n of rq will not be allocated and used.\n \n Fixes: 926a01dc000d (\"RDMA/hns: Add QP operations support for hip08 SoC\")\n Signed-off-by: Lijun Ou \u003coulijun@huawei.com\u003e\n Link: https://lore.kernel.org/r/1565343666-73193-3-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 0e1aa6f0959e053651102c69cde9d6c42fea2a91\nAuthor: Weihang Li \u003cliweihang@hisilicon.com\u003e\nDate: Fri Aug 9 17:40:58 2019 +0800\n\n RDMA/hns: Logic optimization of wc_flags\n \n We should set IB_WC_WITH_VLAN only when VLAN is enabled.\n In addition, move setting of IB_WC_WITH_SMAC below\n setting of wc-\u003esmac.\n \n Signed-off-by: Weihang Li \u003cliweihang@hisilicon.com\u003e\n Link: https://lore.kernel.org/r/1565343666-73193-2-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 9dc4cfff115f03e999e5d50a593c83cd0ef4125c\nAuthor: Leon Romanovsky \u003cleonro@mellanox.com\u003e\nDate: Tue Aug 13 13:28:14 2019 +0300\n\n RDMA/mlx5: Annotate lock dependency in bind/unbind slave port\n \n NULL-ing notifier_call is performed under protection\n of mlx5_ib_multiport_mutex lock. Such protection is\n not easily spotted and better to be guarded by lockdep\n annotation.\n \n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Link: https://lore.kernel.org/r/20190813102814.22350-1-leon@kernel.org\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 8293a598feec361bf40c81711e9974139a9dd580\nAuthor: Yishai Hadas \u003cyishaih@mellanox.com\u003e\nDate: Thu Aug 8 11:43:58 2019 +0300\n\n IB/mlx5: Expose XRQ legacy commands over the DEVX interface\n \n Expose missing XRQ legacy commands over the DEVX interface.\n \n Signed-off-by: Yishai Hadas \u003cyishaih@mellanox.com\u003e\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Link: https://lore.kernel.org/r/20190808084358.29517-5-leon@kernel.org\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 972d7560ee37d48a9121ca1b7796d107a16784e7\nAuthor: Yishai Hadas \u003cyishaih@mellanox.com\u003e\nDate: Thu Aug 8 11:43:57 2019 +0300\n\n IB/mlx5: Add legacy events to DEVX list\n \n Add two events that were defined in the device specification but were\n not exposed in the driver list.\n \n Post this patch those events can be read over the DEVX events interface\n once be reported by the firmware.\n \n Signed-off-by: Yishai Hadas \u003cyishaih@mellanox.com\u003e\n Reviewed-by: Edward Srouji \u003cedwards@mellanox.com\u003e\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Link: https://lore.kernel.org/r/20190808084358.29517-4-leon@kernel.org\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit f6545bec969358eace40419aff26a2a236e0b813\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Wed Jul 31 15:13:58 2019 +0900\n\n kbuild: add [M] marker for build log of *.mod.o\n \n This builds module objects, so [M] makes sense.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 4b950bb9ac0c7246dcf75060040577c3de60c166\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Sun Jul 28 20:27:41 2019 +0200\n\n Kbuild: Handle PREEMPT_RT for version string and magic\n \n Update the build scripts and the version magic to reflect when\n CONFIG_PREEMPT_RT is enabled in the same way as CONFIG_PREEMPT is treated.\n \n The resulting version strings:\n \n Linux m 5.3.0-rc1+ #100 SMP Fri Jul 26 ...\n Linux m 5.3.0-rc1+ #101 SMP PREEMPT Fri Jul 26 ...\n Linux m 5.3.0-rc1+ #102 SMP PREEMPT_RT Fri Jul 26 ...\n \n The module vermagic:\n \n 5.3.0-rc1+ SMP mod_unload modversions\n 5.3.0-rc1+ SMP preempt mod_unload modversions\n 5.3.0-rc1+ SMP preempt_rt mod_unload modversions\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit cf8dfd15e5fb280fa1e79e1d373456cb1e701222\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sun Jul 21 01:27:40 2019 +0900\n\n kbuild: move flex and bison rules to Makefile.host\n \n Flex and bison are used for kconfig, dtc, genksyms, all of which are\n host programs. I never imagine the kernel embeds a parser or a lexer.\n \n Move the flex and bison rules to scripts/Makefile.host. This file is\n included only when hostprogs-y etc. is present in the Makefile in the\n directory. So, parsing these rules are skipped in most of directories.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 6ba7dc6616ce69ef667204df29597767c1c9ebcf\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sun Jul 21 01:27:39 2019 +0900\n\n kbuild: make bison create C file and header in a single pattern rule\n \n We generally expect bison to create not only a C file, but also a\n header, which will be included from the lexer.\n \n Currently, Kbuild generates them in separate rules. So, for instance,\n when building Kconfig, you will notice bison is invoked twice:\n \n HOSTCC scripts/kconfig/conf.o\n HOSTCC scripts/kconfig/confdata.o\n HOSTCC scripts/kconfig/expr.o\n LEX scripts/kconfig/lexer.lex.c\n YACC scripts/kconfig/parser.tab.h\n HOSTCC scripts/kconfig/lexer.lex.o\n YACC scripts/kconfig/parser.tab.c\n HOSTCC scripts/kconfig/parser.tab.o\n HOSTCC scripts/kconfig/preprocess.o\n HOSTCC scripts/kconfig/symbol.o\n HOSTLD scripts/kconfig/conf\n \n Make handles such cases nicely in pattern rules [1]. Merge the two\n rules so that one invokcation of bison can generate both of them.\n \n HOSTCC scripts/kconfig/conf.o\n HOSTCC scripts/kconfig/confdata.o\n HOSTCC scripts/kconfig/expr.o\n LEX scripts/kconfig/lexer.lex.c\n YACC scripts/kconfig/parser.tab.[ch]\n HOSTCC scripts/kconfig/lexer.lex.o\n HOSTCC scripts/kconfig/parser.tab.o\n HOSTCC scripts/kconfig/preprocess.o\n HOSTCC scripts/kconfig/symbol.o\n HOSTLD scripts/kconfig/conf\n \n [1] Pattern rule\n \n GNU Make manual says:\n \"Pattern rules may have more than one target. Unlike normal rules,\n this does not act as many different rules with the same prerequisites\n and recipe. If a pattern rule has multiple targets, make knows that\n the rule\u0027s recipe is responsible for making all of the targets. The\n recipe is executed only once to make all the targets. When searching\n for a pattern rule to match a target, the target patterns of a rule\n other than the one that matches the target in need of a rule are\n incidental: make worries only about giving a recipe and prerequisites\n to the file presently in question. However, when this file\u0027s recipe is\n run, the other targets are marked as having been updated themselves.\"\n \n https://www.gnu.org/software/make/manual/html_node/Pattern-Intro.html\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 49d5089d926c2bf0c76410bf32e5c48b296ec6f6\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sun Jul 21 01:27:38 2019 +0900\n\n kbuild: use $(basename ...) for cmd_asn1_compiler\n \n $(basename ...) trims the last suffix. Using it is more intuitive in\n my opinion.\n \n This pattern rule makes %.asn1.c and %.asn1.h at the same time.\n Previously, the short log showed only either of them, depending on\n the target file in question.\n \n To clarify that two files are being generated by the single recipe,\n I changed the log as follows:\n \n Before:\n \n ASN.1 crypto/asymmetric_keys/x509.asn1.c\n \n After:\n \n ASN.1 crypto/asymmetric_keys/x509.asn1.[ch]\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 75959d44f9dc8e44410667009724e4e238515502\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Tue Jul 16 21:47:27 2019 +0200\n\n kbuild: Fail if gold linker is detected\n \n The gold linker has known issues of failing the build both in random and in\n predictible ways:\n \n - The x86/X32 VDSO build fails with:\n \n arch/x86/entry/vdso/vclock_gettime-x32.o:vclock_gettime.c:function do_hres:\n error: relocation overflow: reference to \u0027hvclock_page\u0027\n \n That\u0027s a known issue for years and the usual workaround is to disable\n CONFIG_X86_32\n \n - A recent build failure is caused by turning a relocation into an\n absolute one for unknown reasons. See link below.\n \n - There are a couple of gold workarounds applied already, but reports\n about broken builds with ld.gold keep coming in on a regular base and in\n most cases the root cause is unclear.\n \n In context of the most recent fail H.J. stated:\n \n \"Since building a workable kernel for different kernel configurations\n isn\u0027t a requirement for gold, I don\u0027t recommend gold for kernel.\"\n \n So instead of dealing with attempts to duct tape gold support without\n understanding the root cause and without support from the gold folks, fail\n the build when gold is detected.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Acked-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n Link: https://lore.kernel.org/r/CAMe9rOqMqkQ0LNpm25yE_Yt0FKp05WmHOrwc0aRDb53miFKM+w@mail.gmail.com\n Reviewed-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Tested-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 15bfc2348d549b44bdca266747f71c0d54bc0e5f\nAuthor: Denis Efremov \u003cefremov@linux.com\u003e\nDate: Thu Aug 1 09:06:57 2019 +0300\n\n modpost: check for static EXPORT_SYMBOL* functions\n \n This patch adds a check to warn about static EXPORT_SYMBOL* functions\n during the modpost. In most of the cases, a static symbol marked for\n exporting is an odd combination that should be fixed either by deleting\n the exporting mark or by removing the static attribute and adding the\n appropriate declaration to headers.\n \n This check could help to detect the following problems:\n 1. 550113d4e9f5 (\"i2c: add newly exported functions to the header, too\")\n 2. 54638c6eaf44 (\"net: phy: make exported variables non-static\")\n 3. 98ef2046f28b (\"mm: remove the exporting of totalram_pages\")\n 4. 73df167c819e (\"s390/zcrypt: remove the exporting of ap_query_configuration\")\n 5. a57caf8c527f (\"sunrpc/cache: remove the exporting of cache_seq_next\")\n 6. e4e4730698c9 (\"crypto: skcipher - remove the exporting of skcipher_walk_next\")\n 7. 14b4c48bb1ce (\"gve: Remove the exporting of gve_probe\")\n 8. 9b79ee9773a8 (\"scsi: libsas: remove the exporting of sas_wait_eh\")\n 9. ...\n \n The build time impact is very limited and is almost at the unnoticeable\n level (\u003c 1 sec).\n \n Acked-by: Emil Velikov \u003cemil.l.velikov@gmail.com\u003e\n Signed-off-by: Denis Efremov \u003cefremov@linux.com\u003e\n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n\ncommit 31fa571c01d869c659d6bc0a7960a55ce314923c\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Tue Aug 13 22:43:23 2019 +0800\n\n ALSA: hda/sigmatel - remove unused variable \u0027stac9200_core_init\u0027\n \n sound/pci/hda/patch_sigmatel.c:981:30: warning:\n stac9200_core_init defined but not used [-Wunused-const-variable\u003d]\n \n It is never used, so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 3ca3c4aad2efa2931b663acc4ece7a38b31071d1\nAuthor: Andre Hartmann \u003caha_1980@gmx.de\u003e\nDate: Sat Mar 23 16:04:19 2019 +0100\n\n can: netlink: fix documentation typos\n \n This patch fixes some documentation typos in struct can_bittiming_const.\n \n Signed-off-by: Andre Hartmann \u003caha_1980@gmx.de\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit e83e416cdf4ff7f4815fe039a70688cf7e2d9d58\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Fri Jul 26 09:35:43 2019 +0200\n\n can: vcan: introduce pr_fmt and make use of it\n \n This patch introduces pr_fmt and makes use of it, also it converts a\n printk() to pr_info().\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 4248f5e02f2e23f75f15db15253137923209ad79\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Wed Jul 24 14:28:21 2019 +0200\n\n can: vcan: remove unnecessary blank lines\n \n This patch removes unnecessary blank lines, so that checkpatch doesn\u0027t\n complain anymore.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit d569de814967362bdd69f073b88f1f1321097dee\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Wed Jul 24 14:16:29 2019 +0200\n\n can: vcan: convert block comments to network style comments\n \n This patch converts all block comments to network subsystem style block\n comments.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 456a8a646b2563438c16a9b27decf9aa717f1ebb\nAuthor: Oliver Hartkopp \u003csocketcan@hartkopp.net\u003e\nDate: Sat Aug 10 21:18:10 2019 +0200\n\n can: gw: add support for CAN FD frames\n \n Introduce CAN FD support which needs an extension of the netlink API to\n pass CAN FD type content to the kernel which has a different size to\n Classic CAN. Additionally the struct canfd_frame has a new \u0027flags\u0027 element\n that can now be modified with can-gw.\n \n The new CGW_FLAGS_CAN_FD option flag defines whether the routing job\n handles Classic CAN or CAN FD frames. This setting is very strict at\n reception time and enables the new possibilities, e.g. CGW_FDMOD_* and\n modifying the flags element of struct canfd_frame, only when\n CGW_FLAGS_CAN_FD is set.\n \n Signed-off-by: Oliver Hartkopp \u003csocketcan@hartkopp.net\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit e9dc7c60507c822992e793bd3845f0556ae0ff98\nAuthor: Oliver Hartkopp \u003csocketcan@hartkopp.net\u003e\nDate: Sat Aug 10 21:18:09 2019 +0200\n\n can: gw: use struct canfd_frame as internal data structure\n \n To prepare the CAN FD support this patch implements the first adaptions in\n data structures for CAN FD without changing the current functionality.\n \n Additionally some code at the end of this patch is moved or indented to\n simplify the review of the next implementation step.\n \n Signed-off-by: Oliver Hartkopp \u003csocketcan@hartkopp.net\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 21468e6de80e2e3584024f796e3c0ce8b12474bd\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Wed Jul 24 14:48:14 2019 +0200\n\n can: gw: cgw_parse_attr(): remove unnecessary braces for single statement block\n \n This patch removes some unnecessary braces for a single statement block.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 78f8a326f924b090d5a495049c91aa599796f2aa\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Wed Jul 24 14:34:42 2019 +0200\n\n can: gw: cgw_dump_jobs(): avoid long lines\n \n This patch rewraps the arguments of cgw_put_job() to avoid long lines,\n which also fixes the indention.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 0815c891e52208dc798916650b3496a2a53ee81f\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Wed Jul 24 14:33:13 2019 +0200\n\n can: gw: can_can_gw_rcv(): remove return at end of void function\n \n This patch remove the return at the end of the void function\n can_can_gw_rcv().\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 5dfc8c94a887d3e0553316430c4fb33df85ccc89\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Wed Jul 24 14:31:48 2019 +0200\n\n can: gw: add missing spaces around operators\n \n This patch add missing spaces around the \u0027^\u0027 and \u0027+\u0027 operators.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit f62564f5e3f2463d3b252f7e73b86700c89795e0\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Wed Jul 24 14:28:21 2019 +0200\n\n can: gw: remove unnecessary blank lines, add suggested blank lines\n \n This patch removes unnecessary blank lines, and adds suggested ones, so\n that checkpatch doesn\u0027t complain anymore.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 465c0deb10577e384937970dc99b61cc69212651\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Wed Jul 24 14:16:29 2019 +0200\n\n can: gw: convert block comments to network style comments\n \n This patch converts all block comments to network subsystem style block\n comments.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit bf74aa86e111aa3b2fbb25db37e3a3fab71b5b68\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Aug 12 14:57:14 2019 +0200\n\n can: bcm: switch timer to HRTIMER_MODE_SOFT and remove hrtimer_tasklet\n \n This patch switches the timer to HRTIMER_MODE_SOFT, which executed the\n timer callback in softirq context and removes the hrtimer_tasklet.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Signed-off-by: Anna-Maria Gleixner \u003canna-maria@linutronix.de\u003e\n Acked-by: Oliver Hartkopp \u003csocketcan@hartkopp.net\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 9989f6333c58e5068abf65384b05de3847b07ce1\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Mon Aug 12 15:49:04 2019 +0200\n\n can: bcm: bcm_sock_no_ioctlcmd(): mark function as static\n \n This patch marks the bcm_sock_no_ioctlcmd() function as static as it\u0027s\n only used in this source file.\n \n Fixes: 473d924d7d46 (\"can: fix ioctl function removal\")\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit af0b1470bda9751c4d757a0efd50edda8da49c99\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Mon Aug 12 15:49:04 2019 +0200\n\n can: raw: raw_sock_no_ioctlcmd(): mark function as static\n \n This patch marks the raw_sock_no_ioctlcmd() function as static as it\u0027s\n only used in this source file.\n \n Fixes: 473d924d7d46 (\"can: fix ioctl function removal\")\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit d956b1a87f9bc84aed7bd4f5f5ae52f9ead0381d\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Tue Aug 13 09:07:14 2019 +0200\n\n can: raw: raw_module_init(): use pr_err() instead of printk(KERN_ERR, ...)\n \n This patch converts a printk(KERN_ERR, ...) to a pr_err().\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit bff10040672d41bd027c267d78e967c374d4e04b\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Tue Aug 13 09:03:55 2019 +0200\n\n can: raw: balance braces around else statements\n \n This patch balances the braces around else statements, so that\n checkpatch doesn\u0027t complain anymore.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit d5e4ecac87fc620b642c331d374a5fc4569493ed\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Wed Jul 24 14:28:21 2019 +0200\n\n can: raw: remove unnecessary blank lines, add suggested blank lines\n \n This patch removes unnecessary blank lines, and adds suggested ones, so\n that checkpatch doesn\u0027t complain anymore.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit d6ada83bba9905c436b1ffcbd5fd995f5e38f291\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Wed Jul 24 14:16:29 2019 +0200\n\n can: raw: convert block comments to network style comments\n \n This patch converts all block comments to network subsystem style block\n comments.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 97c334935100e762feb4f083291bb2fc006de7df\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Tue Aug 13 11:25:33 2019 +0200\n\n can: af_can: add missing identifiers to struct receiver::func\n \n This patch adds the missing identifiers to the struct receiver::func\n declaration.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit fbd4665a67f74b6af44f6a9cf335ddc3fc124f75\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Tue Aug 13 09:41:40 2019 +0200\n\n can: af_can: can_pernet_init(): Use preferred style kzalloc(sizeof()) usage\n \n This patch switches can_pernet_init() to the preferred style of using\n the sizeof() operator in kzalloc().\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 1cf571eb81d3f435e6a4ea359a08b35cb7c5f81a\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Tue Aug 13 09:29:10 2019 +0200\n\n can: af_can: avoid splitting quoted string across lines\n \n This patch joins all error message strings in af_can to be in single\n lines, to ease searching for them.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit b11844b6068cf091640e6e43a27906b2c5bf86fa\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Tue Aug 13 09:06:17 2019 +0200\n\n can: af_can: fix alignment\n \n This patch fixes the alignment of find_dev_rcv_lists() and canfd_rcv()\n so that checkpatch doesn\u0027t complain anymore.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 8325ce9bca5522852e6fe9e9bee7ea8a9c30caac\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Tue Aug 13 09:03:55 2019 +0200\n\n can: af_can: balance braces around else statements\n \n This patch balances the braces around else statements, so that\n checkpatch doesn\u0027t complain anymore.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 147d9e9a67e118270bc83ea89cc989f7f1e4edea\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Wed Jul 24 14:16:29 2019 +0200\n\n can: af_can: convert block comments to network style comments\n \n This patch converts all block comments to network subsystem style block\n comments.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 3e994ff28f867a20ae0dd055a5e4a4a34aad9c3c\nAuthor: Srinivas Neeli \u003csrinivas.neeli@xilinx.com\u003e\nDate: Mon Aug 12 15:36:46 2019 +0530\n\n can: xilinx_can: xcan_set_bittiming(): fix the data phase btr1 calculation\n \n While calculating bitrate for the data phase, the driver is using phase\n segment 1 of the arbitration phase instead of the data phase.\n \n Fixes: c223da6 (\"can: xilinx_can: Add support for CANFD FD frames\")\n Signed-off-by: Appana Durga Kedareswara rao \u003cappana.durga.rao@xilinx.com\u003e\n Signed-off-by: Srinivas Neeli \u003csrinivas.neeli@xilinx.com\u003e\n Acked-by: Shubhrajyoti Datta \u003cshubhrajyoti.datta@xilinx.com\u003e\n Signed-off-by: Michal Simek \u003cmichal.simek@xilinx.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 9d06bcb9aa48dc4fc474dc8441a036bc4a090270\nAuthor: Appana Durga Kedareswara rao \u003cappana.durga.rao@xilinx.com\u003e\nDate: Mon Aug 12 15:36:45 2019 +0530\n\n can: xilinx_can: xcan_rx_fifo_get_next_frame(): fix FSR register FL and RI mask values for canfd 2.0\n \n For CANFD 2.0 IP configuration existing driver is using incorrect mask\n values for FSR register FL and RI fields.\n \n Fixes: c223da6 (\"can: xilinx_can: Add support for CANFD FD frames\")\n Signed-off-by: Appana Durga Kedareswara rao \u003cappana.durga.rao@xilinx.com\u003e\n Acked-by: Shubhrajyoti Datta \u003cshubhrajyoti.datta@xilinx.com\u003e\n Signed-off-by: Michal Simek \u003cmichal.simek@xilinx.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit e6997dd2688403bf8f760bd9fd7ddb5c654bc8bd\nAuthor: Appana Durga Kedareswara rao \u003cappana.durga.rao@xilinx.com\u003e\nDate: Mon Aug 12 15:36:44 2019 +0530\n\n can: xilinx_can: fix the data update logic for CANFD FD frames\n \n commit c223da689324 (\"can: xilinx_can: Add support for CANFD FD frames\")\n is writing data to a wrong offset for FD frames.\n \n This patch fixes this issue.\n \n Fixes: c223da6 (\"can: xilinx_can: Add support for CANFD FD frames\")\n Reviewed-by: Radhey Shyam Pandey \u003cradhey.shyam.pandey@xilinx.com\u003e\n Reviewed-by: Shubhrajyoti Datta \u003cShubhrajyoti.datta@xilinx.com\u003e\n Signed-off-by: Appana Durga Kedareswara rao \u003cappana.durga.rao@xilinx.com\u003e\n Signed-off-by: Michal Simek \u003cmichal.simek@xilinx.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 5dae15b21d36a2e6d7e92c6af39c33dea5c39cc3\nAuthor: Vidya Sagar \u003cvidyas@nvidia.com\u003e\nDate: Tue Aug 13 17:06:26 2019 +0530\n\n phy: tegra: Add PCIe PIPE2UPHY support\n \n Synopsys DesignWare core based PCIe controllers in Tegra 194 SoC\n interface with Universal PHY (UPHY) module through a PIPE2UPHY (P2U)\n module. For each PCIe lane of a controller, there is a P2U unit\n instantiated at hardware level. This driver provides support for the\n programming required for each P2U that is going to be used for a PCIe\n controller.\n \n Signed-off-by: Vidya Sagar \u003cvidyas@nvidia.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Acked-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n\ncommit 813c6fbcb7756e96f96c5931d8830970ff50e3b0\nAuthor: Vidya Sagar \u003cvidyas@nvidia.com\u003e\nDate: Tue Aug 13 17:06:25 2019 +0530\n\n dt-bindings: PHY: P2U: Add Tegra194 P2U block\n \n Add support for Tegra194 P2U (PIPE to UPHY) module block which is a glue\n module instantiated once for each PCIe lane between Synopsys DesignWare\n core based PCIe IP and Universal PHY block.\n \n Signed-off-by: Vidya Sagar \u003cvidyas@nvidia.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Acked-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\n\ncommit 42c253ecaa53c0a2fee431018dd18d2fa797580f\nAuthor: Vidya Sagar \u003cvidyas@nvidia.com\u003e\nDate: Tue Aug 13 17:06:24 2019 +0530\n\n dt-bindings: PCI: tegra: Add device tree support for Tegra194\n \n Add support for Tegra194 PCIe controllers. These controllers are based\n on Synopsys DesignWare core IP.\n \n Signed-off-by: Vidya Sagar \u003cvidyas@nvidia.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n\ncommit 7b31a72e7195aebd620d504f28b97ad8398e95ec\nAuthor: Vidya Sagar \u003cvidyas@nvidia.com\u003e\nDate: Tue Aug 13 17:06:23 2019 +0530\n\n dt-bindings: Add PCIe supports-clkreq property\n \n Some host controllers need to know the existence of clkreq signal routing\n to downstream devices to be able to advertise low power features like\n ASPM L1 substates. Without clkreq signal routing being present, enabling\n ASPM L1 substates might lead to downstream devices being disconnected\n from the bus. Hence a new device tree property \u0027supports-clkreq\u0027 is added\n to make such host controllers aware of clkreq signal routing to\n downstream devices.\n \n Signed-off-by: Vidya Sagar \u003cvidyas@nvidia.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n\ncommit 07f123def73e07e398537be7a5b43a244cb0cbf3\nAuthor: Vidya Sagar \u003cvidyas@nvidia.com\u003e\nDate: Tue Aug 13 17:06:22 2019 +0530\n\n PCI: dwc: Add support to enable CDM register check\n \n Add support to enable CDM (Configuration Dependent Module) register\n check for any data corruption based on the DT property\n \u0027snps,enable-cdm-check\u0027.\n \n Signed-off-by: Vidya Sagar \u003cvidyas@nvidia.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Reviewed-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Acked-by: Gustavo Pimentel \u003cgustavo.pimentel@synopsys.com\u003e\n\ncommit 1e5227a3a98b8acf4e322c12453cd26276f186c8\nAuthor: Bob Peterson \u003crpeterso@redhat.com\u003e\nDate: Tue Aug 13 09:25:15 2019 -0400\n\n gfs2: Fix possible fs name overflows\n \n This patch fixes three places in which temporary character buffers\n could overflow due to the addition of the file system id from patch\n 3792ce973f07. Thanks to Dan Carpenter for pointing it out.\n \n Signed-off-by: Bob Peterson \u003crpeterso@redhat.com\u003e\n Signed-off-by: Andreas Gruenbacher \u003cagruenba@redhat.com\u003e\n\ncommit 0dc99e042a4cfbc1e27572d523d2a1dbaf402cbf\nAuthor: Denis Efremov \u003cefremov@linux.com\u003e\nDate: Tue Aug 13 03:16:02 2019 -0300\n\n media: MAINTAINERS: Remove zoran driver\n \n Cleanup MAINTAINERS from zoran record since the driver was removed.\n \n Fixes: 8dce4b265a53 (\"media: zoran: remove deprecated driver\")\n Signed-off-by: Denis Efremov \u003cefremov@linux.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 34994c9d9d4886790604fbf16f3806781039a2d0\nAuthor: Denis Efremov \u003cefremov@linux.com\u003e\nDate: Tue Aug 13 03:03:34 2019 -0300\n\n media: MAINTAINERS: hantro: Fix typo in a filepath\n \n Fix typo (s/platform//) in the directory path.\n \n Cc: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Fixes: a29add8c9bb2 (\"media: rockchip/vpu: rename from rockchip to hantro\")\n Signed-off-by: Denis Efremov \u003cefremov@linux.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 009b37c65df54761a64562da37f919fc44909a16\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Mon Jul 15 00:18:51 2019 -0300\n\n media: exynos4-is: Remove call to memset after dma_alloc_coherent\n \n In commit 518a2f1925c3\n (\"dma-mapping: zero memory returned from dma_alloc_*\"),\n dma_alloc_coherent has already zeroed the memory.\n So memset is not needed\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Reviewed-by: Sylwester Nawrocki \u003cs.nawrocki@samsung.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 6a11f0f588865f7f8bf2841d9a904c01158ada59\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Mon Jul 15 00:18:44 2019 -0300\n\n media: ngene: Remove call to memset after pci_alloc_consistent\n \n pci_alloc_consistent calls dma_alloc_coherent directly.\n In commit 518a2f1925c3\n (\"dma-mapping: zero memory returned from dma_alloc_*\"),\n dma_alloc_coherent has already zeroed the memory.\n So memset is not needed.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit e671499303e4c0e4bee073889de8b90261ad5abb\nAuthor: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\nDate: Wed Jul 10 18:51:49 2019 -0300\n\n media: i2c: Convert to new i2c device probe()\n \n The I2C core framework provides a simplified probe framework from commit\n b8a1a4cd5a98 (\"i2c: Provide a temporary .probe_new() call-back type\").\n \n These drivers do not utilise the i2c_device_id table in the probe, so we\n can easily convert them to utilise the simplified i2c driver\n registration.\n \n Signed-off-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Acked-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Reviewed-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit f4e84a5d9db7dfea96832157bc461177e87a47dd\nAuthor: Vidya Sagar \u003cvidyas@nvidia.com\u003e\nDate: Tue Aug 13 17:06:21 2019 +0530\n\n dt-bindings: PCI: designware: Add binding for CDM register check\n \n Add support to enable CDM (Configuration Dependent Module) registers check\n for any data corruption. CDM registers include standard PCIe configuration\n space registers, Port Logic registers and iATU and DMA registers.\n Refer Section S.4 of Synopsys DesignWare Cores PCI Express Controller Databook\n Version 4.90a.\n \n Signed-off-by: Vidya Sagar \u003cvidyas@nvidia.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Reviewed-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit bec5352d20d1a8c11d482fb079b553c00b7204ac\nAuthor: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\nDate: Wed Jul 10 18:51:48 2019 -0300\n\n media: i2c: et8ek8: Convert to new i2c device probe()\n \n The I2C core framework provides a simplified probe framework from commit\n b8a1a4cd5a98 (\"i2c: Provide a temporary .probe_new() call-back type\").\n \n This driver does not utilise the i2c_device_id table in the probe, so we can\n easily convert it to utilise the simplified i2c driver registration.\n \n Signed-off-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Reviewed-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit feee519ae55c8859a50eca0da83b1ef4ea45f65d\nAuthor: Vidya Sagar \u003cvidyas@nvidia.com\u003e\nDate: Tue Aug 13 17:06:20 2019 +0530\n\n PCI: dwc: Export dw_pcie_wait_for_link() API\n \n Export the dw_pcie_wait_for_link() function to be able to build drivers\n using it as loadable modules.\n \n Signed-off-by: Vidya Sagar \u003cvidyas@nvidia.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n\ncommit 42b66122151810c7b550dc185a2e3488104eaa3f\nAuthor: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\nDate: Wed Jul 10 18:51:47 2019 -0300\n\n media: i2c: s5c73m3: Convert to new i2c device probe()\n \n The I2C core framework provides a simplified probe framework from commit\n b8a1a4cd5a98 (\"i2c: Provide a temporary .probe_new() call-back type\").\n \n This driver does not utilise the i2c_device_id table in the probe, so we can\n easily convert it to utilise the simplified i2c driver registration.\n \n Signed-off-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Acked-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Reviewed-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit c1042d62fe05cffd5fa314790f54ec2f53443107\nAuthor: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\nDate: Wed Jul 10 18:51:46 2019 -0300\n\n media: i2c: smiapp: Convert to new i2c device probe()\n \n The I2C core framework provides a simplified probe framework from commit\n b8a1a4cd5a98 (\"i2c: Provide a temporary .probe_new() call-back type\").\n \n This driver does not utilise the i2c_device_id table in the probe, so we can\n easily convert it to utilise the simplified i2c driver registration.\n \n Signed-off-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Reviewed-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 5e56808900a7bc8838dcc561d464cf5c2c9a2bd1\nAuthor: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\nDate: Wed Jul 10 18:51:45 2019 -0300\n\n media: radio: si470x: Convert to new i2c device probe()\n \n The I2C core framework provides a simplified probe framework from commit\n b8a1a4cd5a98 (\"i2c: Provide a temporary .probe_new() call-back type\").\n \n This driver does not utilise the i2c_device_id table in the probe, so we can\n easily convert it to utilise the simplified i2c driver registration.\n \n Signed-off-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Reviewed-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 5b0841fa653f6c156500ecfe7e996837884363c4\nAuthor: Vidya Sagar \u003cvidyas@nvidia.com\u003e\nDate: Tue Aug 13 17:06:19 2019 +0530\n\n PCI: dwc: Add extended configuration space capability search API\n \n Add extended configuration space capability search API using struct\n dw_pcie* pointer.\n \n Signed-off-by: Vidya Sagar \u003cvidyas@nvidia.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Acked-by: Gustavo Pimentel \u003cgustavo.pimentel@synopsys.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n\ncommit 7bb823a6522afb03acbb117fd0d67ed5b7013bed\nAuthor: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\nDate: Wed Jul 10 18:51:44 2019 -0300\n\n media: radio: si4713: Convert to new i2c device probe()\n \n The I2C core framework provides a simplified probe framework from commit\n b8a1a4cd5a98 (\"i2c: Provide a temporary .probe_new() call-back type\").\n \n This driver does not utilise the i2c_device_id table in the probe, so we can\n easily convert it to utilise the simplified i2c driver registration.\n \n Signed-off-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Reviewed-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 9deff920bd184f167db98635b16668602ec4d2f5\nAuthor: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\nDate: Wed Jul 10 09:37:19 2019 -0300\n\n media: i2c: adv748x: Convert to new i2c device probe()\n \n The I2C core framework provides a simplified probe framework from commit\n b8a1a4cd5a98 (\"i2c: Provide a temporary .probe_new() call-back type\").\n \n Convert the ADV748x to utilise this simplified i2c driver registration.\n \n Signed-off-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Reviewed-by: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\n Acked-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 2dbfc6525046122f35bf5c244c22551fb5566761\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Jul 9 14:26:40 2019 -0300\n\n media: i2c: tvp5150: Add of_node_put() before goto\n \n Each iteration of for_each_available_child_of_node puts the previous\n node, but in the case of a goto from the middle of the loop, there is\n no put, thus causing a memory leak. Hence add an of_node_put before the\n goto in four places.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 22ff7d4ca3f748e3b685fd2385afda18b9fa0258\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Jul 9 14:24:54 2019 -0300\n\n media: platform: mtk-mdp: mtk_mdp_core: Add of_node_put() before goto\n \n Each iteration of for_each_child_of_node puts the previous node, but in\n the case of a goto from the middle of the loop, there is no put, thus\n causing a memory leak. Hence add an of_node_put before the goto in two\n places.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Acked-by: Houlong Wei \u003choulong.wei@mediatek.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 7a6854f6874ff416afa6552706a0011418f07888\nAuthor: Vidya Sagar \u003cvidyas@nvidia.com\u003e\nDate: Tue Aug 13 17:06:18 2019 +0530\n\n PCI: dwc: Move config space capability search API\n \n Move PCIe config space capability search API to common DesignWare file\n as this can be used by both host and EP mode drivers.\n \n Signed-off-by: Vidya Sagar \u003cvidyaos@nvidia.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Reviewed-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Acked-by: Gustavo Pimentel \u003cgustavo.pimentel@synopsys.com\u003e\n\ncommit 4fd22938569c14f6092c05880ca387409d78355f\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Tue Jul 9 06:59:52 2019 -0300\n\n media: fdp1: Reduce FCP not found message level to debug\n \n When support for the IPMMU is not enabled, the FDP driver may be\n probe-deferred multiple times, causing several messages to be printed\n like:\n \n rcar_fdp1 fe940000.fdp1: FCP not found (-517)\n rcar_fdp1 fe944000.fdp1: FCP not found (-517)\n \n Fix this by reducing the message level to debug level, as is done in the\n VSP1 driver.\n \n Fixes: 4710b752e029f3f8 (\"[media] v4l: Add Renesas R-Car FDP1 Driver\")\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 5db8a692fe6959cbbe663a1311e4206d9b0eb752\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Thu Jun 27 23:47:56 2019 -0300\n\n media: pvrusb2: use kzalloc instead of kmalloc and memset\n \n Replace kmalloc followed by a memset with kzalloc.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Acked-by: Mike Isely \u003cisely@pobox.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n [hverkuil-cisco@xs4all.nl: add pvrusb2: prefix in Subject]\n [hverkuil-cisco@xs4all.nl: fix checkpatch whitespace warnings]\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 5cf7e71f7a63c0feb5e306a33942b3af94ef413b\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Thu Jun 13 12:54:20 2019 -0300\n\n media: zd1301_demod: don\u0027t check retval after our own assignemt\n \n No need to check a retval after we assigned a constant to it.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Reviewed-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 0b25167d1cdd42f56c18f8d666a412bbefc819ac\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Thu Jun 13 12:54:19 2019 -0300\n\n media: mn88473: don\u0027t check retval after our own assignemt\n \n No need to check a retval after we assigned a constant to it.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Reviewed-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit e6ec19a25f2183f67e665acbd12fd74efaa6abae\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Thu Jun 13 12:54:18 2019 -0300\n\n media: mn88472: don\u0027t check retval after our own assignemt\n \n No need to check a retval after we assigned a constant to it.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Reviewed-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit ae3af6c43c0dfbded90524a7db5b94fb8fee57c6\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Thu Jun 13 12:54:17 2019 -0300\n\n media: cxd2820r: don\u0027t check retval after our own assignemt\n \n No need to check a retval after we assigned a constant to it.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Reviewed-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 34925d9f019ca5f2bbb2220bfd7bf353c61d8a27\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Fri Aug 9 16:04:06 2019 -0300\n\n media: i2c: adv7842: convert to i2c_new_dummy_device\n \n Move away from the to-be-removed i2c_new_dummy() to\n i2c_new_dummy_device(). So, we now get an ERRPTR which we use in error\n handling by printing the error code. To keep the rest of the driver\n logic as is, internally a NULL ptr is still kept.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 2f822f1da08ac5c93e351e79d22920f08fa51baf\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Fri Aug 9 13:52:15 2019 -0300\n\n media: i2c: tda1997x: prevent potential NULL pointer access\n \n i2c_new_dummy() can fail returning a NULL pointer. This is not checked\n and the returned pointer is blindly used. Convert to\n devm_i2c_new_dummy_client() which returns an ERR_PTR and also add a\n validity check. Using devm_* here also fixes a leak because the dummy\n client was not released in the probe error path.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 53f5d7a46fab01f22a2b4ffed8bd05aa425aed2f\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 14:26:02 2019 -0300\n\n media: usb: go7007: s2250-board: convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 9524da83a50b108119af3c858e438a84e29e39d9\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 14:26:01 2019 -0300\n\n media: i2c: adv7511-v4l2: convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 43aacc52c945f7a0daf03447d96a191912b728c4\nAuthor: Bob Peterson \u003crpeterso@redhat.com\u003e\nDate: Tue Aug 6 13:52:21 2019 -0400\n\n gfs2: untangle the logic in gfs2_drevalidate\n \n Before this patch, function gfs2_drevalidate was a horrific tangle of\n unreadable labels, cases and goto statements. This patch tries to\n simplify the logic and make it more readable.\n \n Signed-off-by: Bob Peterson \u003crpeterso@redhat.com\u003e\n Signed-off-by: Andreas Gruenbacher \u003cagruenba@redhat.com\u003e\n\ncommit 31b9754caf8e27121806df75f030a1f8ff184b2f\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 14:26:00 2019 -0300\n\n media: i2c: adv7180: convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit b13d326ee4a9be18adc7e2d4c998d9016c0b797e\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 14:25:59 2019 -0300\n\n media: i2c: ad9389b: convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 93bbd6c5eeb1dbc1f826b4924c21dc71d14b25c8\nAuthor: Appana Durga Kedareswara rao \u003cappana.durga.rao@xilinx.com\u003e\nDate: Mon Aug 12 15:36:43 2019 +0530\n\n can: xilinx_can: xcanfd_rx(): fix FSR register handling in the RX path\n \n After commit c223da689324 (\"can: xilinx_can: Add support for CANFD FD\n frames\") the driver is updating the FSR IRI index multiple times (i.e in\n xcanfd_rx() and xcan_rx_fifo_get_next_frame()), It should be updated\n once per RX packet. This patch fixes this issue, also this patch removes\n the unnecessary fsr register checks in xcanfd_rx() API.\n \n Fixes: c223da6 (\"can: xilinx_can: Add support for CANFD FD frames\")\n Reviewed-by: Radhey Shyam Pandey \u003cradhey.shyam.pandey@xilinx.com\u003e\n Reviewed-by: Shubhrajyoti Datta \u003cShubhrajyoti.datta@xilinx.com\u003e\n Signed-off-by: Appana Durga Kedareswara rao \u003cappana.durga.rao@xilinx.com\u003e\n Signed-off-by: Michal Simek \u003cmichal.simek@xilinx.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 4cdd515630484158e4b8a3f6cc549af866185dbe\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 14:25:58 2019 -0300\n\n media: dvb-frontends: mn88473: convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit f0b0710c9f7b8ad6c7a3001f903484746c08bb72\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 14:25:57 2019 -0300\n\n media: dvb-frontends: mn88472: convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 0a6ed11f9884d74ab7333a921dafd35a0de10aac\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 14:25:56 2019 -0300\n\n media: dvb-frontends: mn88443x: convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 0a248872e4c50dddcf502dd850037e52016f6d53\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 14:25:55 2019 -0300\n\n media: dvb-frontends: cxd2820r_core: convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 6b0d35891c835319291f2c3b459e666fc809f831\nAuthor: Venkatesh Yadav Abbarapu \u003cvenkatesh.abbarapu@xilinx.com\u003e\nDate: Mon Aug 12 15:36:42 2019 +0530\n\n can: xilinx_can: xcan_probe(): skip error message on deferred probe\n \n When can clock is provided from the clock wizard, clock wizard driver\n may not be available when can driver probes resulting to the error\n message \"Device clock not found error\".\n \n As this error message is not very userful to the end user, skip printing\n it in the case of deferred probe.\n \n Fixes: b1201e44 (\"can: xilinx CAN controller support\")\n Signed-off-by: Appana Durga Kedareswara rao \u003cappana.durga.rao@xilinx.com\u003e\n Signed-off-by: Venkatesh Yadav Abbarapu \u003cvenkatesh.abbarapu@xilinx.com\u003e\n Signed-off-by: Michal Simek \u003cmichal.simek@xilinx.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 864919ea0380e62adb2503b89825fe358acb8216\nAuthor: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\nDate: Fri Jun 21 08:32:50 2019 -0300\n\n media: mtk-mdp: fix reference count on old device tree\n \n of_get_next_child() increments the reference count of the returning\n device_node. Decrement it in the check if we are using the old or the\n new DTB.\n \n Fixes: ba1f1f70c2c0 (\"[media] media: mtk-mdp: Fix mdp device tree\")\n Signed-off-by: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Acked-by: Houlong Wei \u003choulong.wei@mediatek.com\u003e\n [hverkuil-cisco@xs4all.nl: use node instead of parent as temp variable]\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 3924bc2fd1b607faa5cb0bc0655e9853656de31a\nAuthor: Vidya Sagar \u003cvidyas@nvidia.com\u003e\nDate: Tue Aug 13 17:06:17 2019 +0530\n\n PCI: dwc: Group DBI registers writes requiring unlocking\n \n Some of DesignWare core\u0027s DBI registers (a.k.a configuration space\n registers) are write-protected with a lock without enabling which they\n are read-only by default.\n \n These write-protected registers are implementation specific. Tegra194\u0027s\n BAR-0 register which is at offset 0x10 in the configuration space is an\n example. Current implementation in dw_pcie_setup_rc() API attempts to\n unlock those write-protected registers whenever they are updated and\n lock them back again for writing.\n \n Group all write-protected registers writes so that locking and unlocking\n is performed once to avoid bloating the code with multiple unlock/lock\n sequences for all those write-protected registers.\n \n Signed-off-by: Vidya Sagar \u003cvidyas@nvidia.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Reviewed-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Acked-by: Jingoo Han \u003cjingoohan1@gmail.com\u003e\n\ncommit b557b5073194d63bcd2850c009f9326250b4bd97\nAuthor: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\nDate: Thu May 30 18:15:14 2019 -0300\n\n media: cedrus: Don\u0027t set chroma size for scale \u0026 rotation\n \n Scale and rotation are currently not implemented, so it makes no sense to\n set chroma size for it.\n \n Signed-off-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Acked-by: Paul Kocialkowski \u003cpaul.kocialkowski@bootlin.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 3486cc40ccbfe4cbf55feecce88a792e3043c178\nAuthor: Anssi Hannula \u003canssi.hannula@bitwise.fi\u003e\nDate: Wed Apr 3 15:18:53 2019 +0300\n\n can: xilinx_can: xcan_chip_start(): fix failure with invalid bus\n \n Currently the xilinx_can xcan_chip_start() function, called from\n .ndo_open() and via CAN_MODE_START (bus-off restart), waits for the SR\n register to show the wanted operating state, with a 1 sec timeout.\n \n However, that register bit will only be set once the HW has observed 11\n consecutive recessive bits (BusIdle) on the bus.\n \n If the bus will not see the 11 bits (e.g. it is stuck dominant), the\n function will timeout and return an error. If this was done as part of a\n scheduled restart from bus-off, the interface will stay in bus-off state\n forever even if the bus recovers later.\n \n According to M_CAN and FLEXCAN documentation they also wait for 11\n consecutive recessive bits, but their drivers do not seem to wait for\n that.\n \n To make the behavior consistent, modify xilinx_can to also not wait for\n the synchronization to complete.\n \n The only way for users to know for sure that the bus has been joined\n successfully is to see successfully received or transmitted frames. That\n does not seem optimal, but it is consistent with other drivers and we\n should have a properly working restart-ms with xilinx_can.\n \n Tested on ZynqMP with Xilinx CAN-FD 1.0.\n \n Fixes: b1201e44f50b (\"can: xilinx CAN controller support\")\n Signed-off-by: Anssi Hannula \u003canssi.hannula@bitwise.fi\u003e\n Tested-by: Appana Durga Kedareswara Rao \u003cappana.durga.rao@xilinx.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 633eadc9ba1e9a56be09ef94f14578a9839d3634\nAuthor: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\nDate: Thu May 30 18:15:13 2019 -0300\n\n media: cedrus: Remove dst_bufs from context\n \n This array is just duplicated capture buffer queue. Remove it and adjust\n code to look into capture buffer queue instead.\n \n Signed-off-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Acked-by: Paul Kocialkowski \u003cpaul.kocialkowski@bootlin.com\u003e\n Reviewed-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 8c7e96d3fe75bb41fadf23e7a494f37d1cd9906c\nAuthor: Vidya Sagar \u003cvidyas@nvidia.com\u003e\nDate: Tue Aug 13 17:06:16 2019 +0530\n\n PCI: Disable MSI for Tegra root ports\n \n Tegra PCIe rootports don\u0027t generate MSI interrupts for PME and AER events.\n Since PCIe spec (Ref: r4.0 sec 7.7.1.2 and 7.7.2.2) doesn\u0027t support using\n a mix of INTx and MSI/MSI-X, MSI needs to be disabled to avoid root ports\n service drivers registering their respective ISRs with MSI interrupt and\n to let only INTx be used for all events.\n \n Signed-off-by: Vidya Sagar \u003cvidyas@nvidia.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Reviewed-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n\ncommit 448d5a55759a2e60208b75ceed4bd88cd4f5a963\nAuthor: Vidya Sagar \u003cvidyas@nvidia.com\u003e\nDate: Tue Aug 13 17:06:15 2019 +0530\n\n PCI: Add #defines for some of PCIe spec r4.0 features\n \n Add #defines only for the Data Link Feature and Physical Layer 16.0 GT/s\n features as defined in PCIe spec r4.0, sec 7.7.4 for Data Link Feature and\n sec 7.7.5 for Physical Layer 16.0 GT/s.\n \n Signed-off-by: Vidya Sagar \u003cvidyas@nvidia.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Reviewed-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Acked-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit 09c41a23a2e2de7ed347251d7079a42de0dd230b\nAuthor: Helen Koike \u003chelen.koike@collabora.com\u003e\nDate: Thu Aug 8 21:27:41 2019 -0300\n\n media: Revert \"media: vimc: propagate pixel format in the stream\"\n \n This reverts commit b6c61a6c37317efd7327199bfe24770af3d7e799.\n \n The requested pixelformat is being propagated from the capture to the\n tpg in the sensor.\n \n This was a bad design choice, as we start having the following issues:\n \n * We set a pixelformat in the capture;\n * We set matching media bus formats in the subdevices pads;\n * Link validate looks fine (sizes matches, media bus formats matches);\n * Issue: if some of the subdevice doesn\u0027t know how to generate the\n requested pixelformat in the capture, then stream_on fails. This is bad\n because capture says it supports that pixelformat, everything looks\n fine, but it is not, and there is no way to find it out through the\n links.\n \n This patch was implemented so we could request any pixelformat from the\n pipeline regardeless of the media bus format configured between pads.\n Not all pixelformat can be mapped into a media bus code (e.g.\n multiplanar formats), so with this patch we could request those\n pixelformats from the tpg.\n \n Solution: map pixelformats to media bus codes as before, and implement\n conversions to other pixelformats in the capture to support multiplanar.\n \n So first step to this solution is to revert this patch.\n \n Signed-off-by: Helen Koike \u003chelen.koike@collabora.com\u003e\n Signed-off-by: Lucas A. M. Magalhaes \u003clucmaga@gmail.com\u003e\n Tested-by: André Almeida \u003candrealmeid@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 7123e1c7ea29f841819900bcf99763aa51ecda58\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Jul 25 07:00:44 2019 +0000\n\n can: ti_hecc: ti_hecc_mailbox_read(): remove set but not used variable \u0027mbx_mask\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/net/can/ti_hecc.c: In function \u0027ti_hecc_mailbox_read\u0027:\n drivers/net/can/ti_hecc.c:533:12: warning:\n variable \u0027mbx_mask\u0027 set but not used [-Wunused-but-set-variable]\n \n It is never used so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Reviewed-by: Jeroen Hofstee \u003cjhofstee@victronenergy.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 10b1aed6dac01d566b6b3c261a191546c82f0157\nAuthor: Helen Koike \u003chelen.koike@collabora.com\u003e\nDate: Thu Aug 8 21:27:40 2019 -0300\n\n media: Revert \"media: vimc: Remove or modify stream checks\"\n \n This reverts commit 3c1b9ac753e99005d7ee0a883d6e5b577ba32aa9.\n \n Required to Revert \"media: vimc: propagate pixel format in the stream\"\n which introduced the stream object in the vimc_ent_device struct.\n \n Signed-off-by: Helen Koike \u003chelen.koike@collabora.com\u003e\n Signed-off-by: Lucas A. M. Magalhaes \u003clucmaga@gmail.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 84246ae3fde112c2d3e8215dfffadcc39b1362a6\nAuthor: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\nDate: Thu Aug 8 02:10:58 2019 -0300\n\n media: rcar-vin: Report correct image stride\n \n The image stride was adjusted when it was written to hardware and not\n when configuring the format. Calculate the correct stride value and\n report it to userspace.\n \n Signed-off-by: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit babfcda6ef436a529edb5d48e8180c323f1ca391\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Fri Jul 26 09:29:14 2019 +0200\n\n can: ti_hecc: ti_hecc_mailbox_read(): add blank lines to improve readability\n \n This patch adds two blank lines in ti_hecc_mailbox_read() to improve the\n readability of the function.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit c71400cec4389461153798de845a2be0f3827c99\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Fri Jul 26 09:09:33 2019 +0200\n\n can: ti_hecc: ti_hecc_start(): avoid multiple assignments\n \n This patch changes the multiple assignments of HECC_TX_MASK in\n ti_hecc_start() into two single ones.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit dfa9352dbe4e0da468cba0412b83ddbba35cd788\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Thu Jul 25 15:11:32 2019 +0200\n\n can: ti_hecc: fix print formating strings\n \n This patch fixes the print format strings in the driver.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit ab3a78f65983adf1bba53c8506968c25df206ae4\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Thu Jul 25 15:10:07 2019 +0200\n\n can: ti_hecc: avoid long lines\n \n This patch fixes long lines in the driver.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit d5f1a984edd8aebcc3de96905047137931b0f8a5\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Thu Jul 25 15:07:19 2019 +0200\n\n can: ti_hecc: fix indention\n \n This patch fixes the indention in the driver.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 99a3d7c122431c625d368fb959aa0de5fa94a85b\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Wed Jul 24 14:16:29 2019 +0200\n\n can: ti_hecc: convert block comments to network style comments\n \n This patch converts all block comments to network subsystem style block\n comments.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 175a60b73b1789dc624d16b8e24f5c3c9114d448\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Jul 25 06:59:40 2019 +0000\n\n can: sja1000: f81601: remove unused including \u003clinux/version.h\u003e\n \n Remove including \u003clinux/version.h\u003e that don\u0027t need it.\n \n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 5549369cfde0f86799bff110ce46c3721c19e2ca\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Jul 25 07:02:17 2019 +0000\n\n can: kvaser_pciefd: Remove unused including \u003clinux/version.h\u003e\n \n Remove including \u003clinux/version.h\u003e that don\u0027t need it.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 1910cd88539c07abf34a558a0976b506483c377c\nAuthor: Christer Beskow \u003cchbe@kvaser.com\u003e\nDate: Tue Aug 6 09:29:38 2019 +0200\n\n can: kvaser_pciefd: kvaser_pciefd_pwm_stop(): remove unnecessary code when setting pwm duty cycle to zero\n \n To set the duty cycle to zero (i.e. pwm_stop), the trigger value shall\n be equal to the top value.\n \n This is achieved by reading the value of the top bit field from the pwm\n register and then writing back this value to the trigger and top bit\n fields.\n \n Addresses-Coverity: (\"Logically dead code\")\n Reported-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Christer Beskow \u003cchbe@kvaser.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 256acbebdc3b2ddfad4a5b1aa99488fad34b0f0d\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Tue Jul 30 03:38:17 2019 -0300\n\n media: omap_vout: convert to vb2\n \n Convert omap_vout to the vb2 framework. After this change\n \u0027v4l2-compliance -s\u0027 passes all tests.\n \n As usual, this vb2 conversion is a \u0027big bang\u0027 patch and hard to read.\n \n Tested on a Pandaboard and a Beagle XM board.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 839b9d2c59b3b3e74cb58b457615ff61154d8a41\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Tue Jul 30 03:38:16 2019 -0300\n\n media: omap_vout: fix various v4l2-compliance failures\n \n This patch fixes the following v4l2-compliance failures:\n \n - FIELD_ANY could be returned, which is not allowed.\n - JPEG colorspace was set instead of SRGB.\n - No control events.\n - Empty bus_info in QUERYCAP.\n - Overlay format handling wasn\u0027t zeroing bitmap/clips and\n didn\u0027t return the chromakey correctly.\n - G_FBUF didn\u0027t fill in many of the v4l2_framebuffer values.\n Now also return the base address of the corresponding\n framebuffer that this overlays.\n - Missing ENUM/G/S_OUTPUT ioctls.\n - ROTATE/VFLIP controls were added when the HW didn\u0027t support them.\n \n With these changes \u0027v4l2-compliance\u0027 passes all non-streaming tests.\n \n Tested on a Pandaboard and a Beagle XM board.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 9840a4ffcf0b26e08472ed53d176a6a0f1d4c498\nAuthor: Petar Penkov \u003cppenkov@google.com\u003e\nDate: Mon Aug 12 16:30:39 2019 -0700\n\n selftests/bpf: fix race in flow dissector tests\n \n Since the \"last_dissection\" map holds only the flow keys for the most\n recent packet, there is a small race in the skb-less flow dissector\n tests if a new packet comes between transmitting the test packet, and\n reading its keys from the map. If this happens, the test packet keys\n will be overwritten and the test will fail.\n \n Changing the \"last_dissection\" map to a hash map, keyed on the\n source/dest port pair resolves this issue. Additionally, let\u0027s clear the\n last test results from the map between tests to prevent previous test\n cases from interfering with the following test cases.\n \n Fixes: 0905beec9f52 (\"selftests/bpf: run flow dissector tests in skb-less mode\")\n Signed-off-by: Petar Penkov \u003cppenkov@google.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit d66fa3c70e598746a907e5db5ed024035e01817a\nAuthor: Peter Wu \u003cpeter@lekensteyn.nl\u003e\nDate: Tue Aug 13 01:38:33 2019 +0100\n\n tools: bpftool: add feature check for zlib\n \n bpftool requires libelf, and zlib for decompressing /proc/config.gz.\n zlib is a transitive dependency via libelf, and became mandatory since\n elfutils 0.165 (Jan 2016). The feature check of libelf is already done\n in the elfdep target of tools/lib/bpf/Makefile, pulled in by bpftool via\n a dependency on libbpf.a. Add a similar feature check for zlib.\n \n Suggested-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Peter Wu \u003cpeter@lekensteyn.nl\u003e\n Acked-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 341dfcf8d78eaa3a2dc96dea06f0392eb2978364\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Mon Aug 12 11:39:47 2019 -0700\n\n btf: expose BTF info through sysfs\n \n Make .BTF section allocated and expose its contents through sysfs.\n \n /sys/kernel/btf directory is created to contain all the BTFs present\n inside kernel. Currently there is only kernel\u0027s main BTF, represented as\n /sys/kernel/btf/kernel file. Once kernel modules\u0027 BTFs are supported,\n each module will expose its BTF as /sys/kernel/btf/\u003cmodule-name\u003e file.\n \n Current approach relies on a few pieces coming together:\n 1. pahole is used to take almost final vmlinux image (modulo .BTF and\n kallsyms) and generate .BTF section by converting DWARF info into\n BTF. This section is not allocated and not mapped to any segment,\n though, so is not yet accessible from inside kernel at runtime.\n 2. objcopy dumps .BTF contents into binary file and subsequently\n convert binary file into linkable object file with automatically\n generated symbols _binary__btf_kernel_bin_start and\n _binary__btf_kernel_bin_end, pointing to start and end, respectively,\n of BTF raw data.\n 3. final vmlinux image is generated by linking this object file (and\n kallsyms, if necessary). sysfs_btf.c then creates\n /sys/kernel/btf/kernel file and exposes embedded BTF contents through\n it. This allows, e.g., libbpf and bpftool access BTF info at\n well-known location, without resorting to searching for vmlinux image\n on disk (location of which is not standardized and vmlinux image\n might not be even available in some scenarios, e.g., inside qemu\n during testing).\n \n Alternative approach using .incbin assembler directive to embed BTF\n contents directly was attempted but didn\u0027t work, because sysfs_proc.o is\n not re-compiled during link-vmlinux.sh stage. This is required, though,\n to update embedded BTF data (initially empty data is embedded, then\n pahole generates BTF info and we need to regenerate sysfs_btf.o with\n updated contents, but it\u0027s too late at that point).\n \n If BTF couldn\u0027t be generated due to missing or too old pahole,\n sysfs_btf.c handles that gracefully by detecting that\n _binary__btf_kernel_bin_start (weak symbol) is 0 and not creating\n /sys/kernel/btf at all.\n \n v2-\u003ev3:\n - added Documentation/ABI/testing/sysfs-kernel-btf (Greg K-H);\n - created proper kobject (btf_kobj) for btf directory (Greg K-H);\n - undo v2 change of reusing vmlinux, as it causes extra kallsyms pass\n due to initially missing __binary__btf_kernel_bin_{start/end} symbols;\n \n v1-\u003ev2:\n - allow kallsyms stage to re-use vmlinux generated by gen_btf();\n \n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 367f4acc69909bfbbab2839a33cfe66ffa075b0b\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Sat Aug 10 09:50:23 2019 +0200\n\n ARM: mach-nspire: Kill off CLCD auxdata\n \n After transitioning the CLCD to use DRM we can now kill off\n the final piece of auxdata on the NSPIRE.\n \n Link: https://lore.kernel.org/r/20190810075023.7327-1-linus.walleij@linaro.org\n Cc: Daniel Tang \u003cdt.tangr@gmail.com\u003e\n Cc: Fabian Vogt \u003cfabian@ritter-vogt.de\u003e\n Tested-by: Fabian Vogt \u003cfabian@ritter-vogt.de\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit da5fbcb1d03a41c69287595e1364c86a498c8087\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Sat Aug 10 09:42:30 2019 +0200\n\n ARM: dts: Update the NSPIRE DTS files for DRM\n \n The DRM subsystem graphics drivers require more granular\n definition of the connection between display drivers and\n panels, and a proper panel compatible. This utilizes the\n bindings merged to the DRM subsystem to properly define\n the display on the NSPIRE devices.\n \n We also do away with the undocumented DT binding\n \"lcd-type\".\n \n We add both the clocks to the CLCD block so the driver\n have full control over its clocking.\n \n Link: https://lore.kernel.org/r/20190810074230.6492-1-linus.walleij@linaro.org\n Cc: Daniel Tang \u003cdt.tangr@gmail.com\u003e\n Cc: Fabian Vogt \u003cfabian@ritter-vogt.de\u003e\n Tested-by: Fabian Vogt \u003cfabian@ritter-vogt.de\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit c08b598410559c7006f2cea04ba80646eb214212\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Tue Aug 13 09:27:31 2019 +0200\n\n ARM: dts: nomadik: Set up the CS GPIO right\n \n Now that the SPI GPIO driver knows how to handle these\n chip select GPIOs and we get nasty messages about the\n core having to enforce active low on the GPIO, fix this\n up by actually requesting the CS GPIO line as active\n low.\n \n Link: https://lore.kernel.org/r/20190813072731.4558-1-linus.walleij@linaro.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 115f3bad1a36791b5293aeebde342a95a516198e\nAuthor: Matthias Maennich \u003cmaennich@google.com\u003e\nDate: Tue Aug 13 13:17:06 2019 +0100\n\n usb-storage: remove single-use define for debugging\n \n USB_STORAGE was defined as \"usb-storage: \" and used in a single location\n as argument to printk. In order to be able to use the name\n \u0027USB_STORAGE\u0027, drop the definition and use the string directly for the\n printk call.\n \n Signed-off-by: Matthias Maennich \u003cmaennich@google.com\u003e\n Link: https://lore.kernel.org/r/20190813121733.52480-10-maennich@google.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a27401cecf7795cda7e0c17751feb98fedbaa99d\nAuthor: David Lechner \u003cdavid@lechnology.com\u003e\nDate: Wed Aug 7 15:09:34 2019 -0500\n\n ARM: dts: am335x-boneblue: Use of am335x-osd335x-common.dtsi\n \n This makes use of the am335x-osd335x-common.dtsi file that contains the\n common device tree components for Octavo Systems AM335x System-in-\n Package that is used on the BeagleBone Blue.\n \n This has two minor side-effects:\n 1. pinmux_i2c0_pins is renamed to pinmux-i2c0-pins\n 2. the 1000MHz cpufreq operating point is enabled\n \n Cc: Robert Nelson \u003crobertcnelson@gmail.com\u003e\n Signed-off-by: David Lechner \u003cdavid@lechnology.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 26b1c8bed58637d88022ca2a937fc9c031f6dd31\nAuthor: H. Nikolaus Schaller \u003chns@goldelico.com\u003e\nDate: Mon Jul 8 16:46:04 2019 +0200\n\n ARM: dts: gta04: define chosen/stdout-path\n \n This allows to remove the console\u003d entry in the kernel command line.\n \n Signed-off-by: H. Nikolaus Schaller \u003chns@goldelico.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 9fc6bff8078ba350ca335eeab81d568073d44f65\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Sun Jul 21 20:24:55 2019 -0300\n\n ARM: dts: omap3-n950-n9: Remove regulator-boot-off property\n \n This property was never supported upstream. Get rid of it.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Acked-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit dc1b6ca8fa66fac58e133786295fac0a237e6d73\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Sun Jul 21 20:24:54 2019 -0300\n\n ARM: dts: am335x-cm-t335: Remove regulator-boot-off property\n \n This property was never supported upstream. Get rid of it.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 5d2fc542e8c92130b55ddeb81876ec398adf4d13\nAuthor: Raag Jadav \u003craagjadav@gmail.com\u003e\nDate: Tue Aug 13 00:09:54 2019 +0530\n\n regulator: act8945a-regulator: fix ldo register addresses in set_mode hook\n \n According to ACT8945A datasheet[1], operating modes for ldos are\n controlled by BIT(5) of their respective _CTRL registers.\n \n [1] https://active-semi.com/wp-content/uploads/ACT8945A_Datasheet.pdf\n \n Fixes: 7482d6ecc68e (\"regulator: act8945a-regulator: Implement PM functionalities\")\n Signed-off-by: Raag Jadav \u003craagjadav@gmail.com\u003e\n Link: https://lore.kernel.org/r/1565635194-5816-1-git-send-email-raagjadav@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9c6c417d95d83999c16965186f6e755ad8a8b658\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Aug 12 14:05:02 2019 -0500\n\n ASoC: SOF: fix HDA direct MMIO access\n \n The recent change to remove the bus-\u003eio_ops callbacks used an older\n version of the SOF code base, and when merged into Mark\u0027s for-next it\n invalidated changes, resulting in broken compilation identified by\n kbuild and reproduced during the weekly SOF rebase.\n \n Restore SOF code overridden by git merge and apply Takashi\u0027s intended\n change in the \u0027right\u0027 location.\n \n Fixes: c2f16a94a8049 (\"Merge branch \u0027topic/hda-bus-ops-cleanup\u0027\")\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Cc: Takashi Iwai \u003ctiwai@suse.de\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190812190502.30729-1-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 8da2d74c10fd71cc04e2585fe8bb3ced325f8e25\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Aug 12 09:03:05 2019 -0500\n\n ASoC: Intel: hsw: remove i386 build warning w/ size_t argument\n \n Recent changes in the common IPC code introduced a build warning with\n size_t fields, use the correct %zu format.\n \n include/linux/dynamic_debug.h:82:16: warning: format \u0027%lu\u0027 expects\n argument of type \u0027long unsigned int\u0027, but argument 4 has type \u0027size_t\u0027\n [-Wformat\u003d]\n \n Fixes: abf31feea26c0 (\u0027ASoC: Intel: Update request-reply IPC model\u0027)\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Cc: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Acked-By: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190812140305.17570-1-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit ae3a5901dde2ab136ec0cebda2fccc48e810d2ec\nAuthor: Masanari Iida \u003cstandby24x7@gmail.com\u003e\nDate: Tue Aug 13 12:42:35 2019 +0900\n\n ASoC: ti: Fix typos in ti/Kconfig\n \n This patch fixes some spelling typo in Kconfig.\n \n Signed-off-by: Masanari Iida \u003cstandby24x7@gmail.com\u003e\n Acked-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Link: https://lore.kernel.org/r/20190813034235.30673-1-standby24x7@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c8a738f4cfaeccce40b171aca6da5fc45433ce60\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Jun 28 12:10:54 2019 +0800\n\n bus: ti-sysc: remove set but not used variable \u0027quirks\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/bus/ti-sysc.c: In function sysc_reset:\n drivers/bus/ti-sysc.c:1452:50: warning: variable quirks set but not used [-Wunused-but-set-variable]\n \n It is never used since commit e0db94fe87da (\"bus: ti-sysc: Make\n OCP reset work for sysstatus and sysconfig reset bits\")\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 7188f656cdf762d4ea8ce16b6aaf4c6b06e119ec\nAuthor: Cheng-Yi Chiang \u003ccychiang@chromium.org\u003e\nDate: Tue Aug 13 15:44:30 2019 +0800\n\n ASoC: rockchip: rockchip_max98090: Set period size to 240\n \n From stress testing of arecord, we found that period size\n greater than ~900 will bring pl330 to DYING state and\n can not recover within 100 iterations.\n The result is that arecord will stuck and get I/O error,\n and issue can not be recovered until reboot.\n \n This issue does not happen when period size is small.\n Set constraint of period size to 240 to prevent such issue.\n With the constraint, there will be no issue after 2000 iterations.\n \n We can revert this patch once the root cause is found\n in rockchip\u0027s pl330 implementation.\n \n Signed-off-by: Cheng-Yi Chiang \u003ccychiang@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190813074430.191791-1-cychiang@chromium.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit bb88b86c1f623e079968c6b8799989019c5f8413\nAuthor: Tero Kristo \u003ct-kristo@ti.com\u003e\nDate: Wed Aug 7 15:46:05 2019 +0300\n\n bus: ti-sysc: allow reset sharing across devices\n \n Some devices need to share their reset signals, like DSP MMUs, thus drop\n the exclusive notation from reset request. Also, balance the init time\n reset count, otherwise the resets will never be applied post boot.\n \n Signed-off-by: Tero Kristo \u003ct-kristo@ti.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 4345f0dc015da045a775655e6afac66dbaa1f62d\nAuthor: Tero Kristo \u003ct-kristo@ti.com\u003e\nDate: Wed Aug 7 15:46:04 2019 +0300\n\n bus: ti-sysc: rework the reset handling\n \n If reset controllers are assigned to the ti-sysc target-module, only\n ti-sysc is going to be able to control these. Thus, remove all the\n disable_on_idle flag usage, and assert/de-assert the reset always\n in the idle path. Otherwise the reset signal will always just be\n de-asserted.\n \n Signed-off-by: Tero Kristo \u003ct-kristo@ti.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit b6036314436b9f4d8dbb8f1bdea511982de73cf9\nAuthor: Tero Kristo \u003ct-kristo@ti.com\u003e\nDate: Wed Aug 7 15:46:03 2019 +0300\n\n bus: ti-sysc: re-order the clkdm control around reset handling\n \n Parenting clockdomain for the IP should be enabled during the reset\n handling logic, otherwise the reset may not finish properly. Re-order\n the clockdomain control logic to avoid this.\n \n Signed-off-by: Tero Kristo \u003ct-kristo@ti.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit b58056da2ec28e2c1b66096cd5109997f04d3fd1\nAuthor: Suman Anna \u003cs-anna@ti.com\u003e\nDate: Tue Aug 13 04:13:32 2019 -0700\n\n bus: ti-sysc: Add missing kerneldoc comments\n \n A few fields in various structures is missing the corresponding\n kerneldoc comments. Add them. Also, fixed the comment for sidlemodes.\n \n Signed-off-by: Suman Anna \u003cs-anna@ti.com\u003e\n Acked-by: Roger Quadros \u003crogerq@ti.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 54d662227caebaabd25823c6d955ba92dfa361a8\nAuthor: Suman Anna \u003cs-anna@ti.com\u003e\nDate: Tue Aug 13 04:13:32 2019 -0700\n\n bus: ti-sysc: Switch to SPDX license identifier\n \n Use the appropriate SPDX license identifier in the TI sysc\n interconnect target driver source files and drop the previous\n boilerplate license text. Also, add the the SPDX license\n identifier in the associated ti-sysc header files.\n \n Signed-off-by: Suman Anna \u003cs-anna@ti.com\u003e\n Acked-by: Roger Quadros \u003crogerq@ti.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit d220540795ec2682a145c896461a4a742478ac52\nAuthor: Suman Anna \u003cs-anna@ti.com\u003e\nDate: Tue Aug 13 04:13:32 2019 -0700\n\n dt-bindings: ti-sysc: Add SPDX license identifier\n \n Add the appropriate SPDX license identifier to the common\n TI sysc bindings header file.\n \n Signed-off-by: Suman Anna \u003cs-anna@ti.com\u003e\n Acked-by: Roger Quadros \u003crogerq@ti.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 82a9c15d1a91e0d731653dee4158b2587fe27df3\nAuthor: Suman Anna \u003cs-anna@ti.com\u003e\nDate: Tue Aug 13 04:13:31 2019 -0700\n\n MAINTAINERS: Add ti-sysc files under the OMAP2+ entry\n \n Add the ti-sysc source files under the OMAP2+ entry so that\n the get_maintainer script also generates the linux-omap\n list to be cc\u0027d for all patches touching these files.\n \n Signed-off-by: Suman Anna \u003cs-anna@ti.com\u003e\n Acked-by: Roger Quadros \u003crogerq@ti.com\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n\ncommit 105333435b4f3b21ffc325f32fae17719310db64\nAuthor: Florian Westphal \u003cfw@strlen.de\u003e\nDate: Mon Aug 12 13:40:04 2019 +0200\n\n netfilter: connlabels: prefer static lock initialiser\n \n seen during boot:\n BUG: spinlock bad magic on CPU#2, swapper/0/1\n lock: nf_connlabels_lock+0x0/0x60, .magic: 00000000, .owner: \u003cnone\u003e/-1, .owner_cpu: 0\n Call Trace:\n do_raw_spin_lock+0x14e/0x1b0\n nf_connlabels_get+0x15/0x40\n ct_init_net+0xc4/0x270\n ops_init+0x56/0x1c0\n register_pernet_operations+0x1c8/0x350\n register_pernet_subsys+0x1f/0x40\n tcf_register_action+0x7c/0x1a0\n do_one_initcall+0x13d/0x2d9\n \n Problem is that ct action init function can run before\n connlabels_init(). Lock has not been initialised yet.\n \n Fix it by using a static initialiser.\n \n Fixes: b57dc7c13ea9 (\"net/sched: Introduce action ct\")\n Signed-off-by: Florian Westphal \u003cfw@strlen.de\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit 0a30ba509fdeea740bedb81ce3bee4f74a759654\nAuthor: Valdis Klētnieks \u003cvaldis.kletnieks@vt.edu\u003e\nDate: Thu Aug 8 01:43:22 2019 -0400\n\n netfilter: nf_nat_proto: make tables static\n \n Sparse warns about two tables not being declared.\n \n CHECK net/netfilter/nf_nat_proto.c\n net/netfilter/nf_nat_proto.c:725:26: warning: symbol \u0027nf_nat_ipv4_ops\u0027 was not declared. Should it be static?\n net/netfilter/nf_nat_proto.c:964:26: warning: symbol \u0027nf_nat_ipv6_ops\u0027 was not declared. Should it be static?\n \n And in fact they can indeed be static.\n \n Signed-off-by: Valdis Kletnieks \u003cvaldis.kletnieks@vt.edu\u003e\n Acked-by: Florian Westphal \u003cfw@strlen.de\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit 5785cf15fd74ec3b1a076fd39bc67382a8455fe7\nAuthor: Valdis Klētnieks \u003cvaldis.kletnieks@vt.edu\u003e\nDate: Thu Aug 8 01:28:08 2019 -0400\n\n netfilter: nf_tables: add missing prototypes.\n \n Sparse rightly complains about undeclared symbols.\n \n CHECK net/netfilter/nft_set_hash.c\n net/netfilter/nft_set_hash.c:647:21: warning: symbol \u0027nft_set_rhash_type\u0027 was not declared. Should it be static?\n net/netfilter/nft_set_hash.c:670:21: warning: symbol \u0027nft_set_hash_type\u0027 was not declared. Should it be static?\n net/netfilter/nft_set_hash.c:690:21: warning: symbol \u0027nft_set_hash_fast_type\u0027 was not declared. Should it be static?\n CHECK net/netfilter/nft_set_bitmap.c\n net/netfilter/nft_set_bitmap.c:296:21: warning: symbol \u0027nft_set_bitmap_type\u0027 was not declared. Should it be static?\n CHECK net/netfilter/nft_set_rbtree.c\n net/netfilter/nft_set_rbtree.c:470:21: warning: symbol \u0027nft_set_rbtree_type\u0027 was not declared. Should it be static?\n \n Include nf_tables_core.h rather than nf_tables.h to pick up the additional definitions.\n \n Signed-off-by: Valdis Kletnieks \u003cvaldis.kletnieks@vt.edu\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit 2a475c409fe81a76fb26a6b023509d648237bbe6\nAuthor: Jeremy Sowden \u003cjeremy@azazel.net\u003e\nDate: Wed Aug 7 15:17:05 2019 +0100\n\n kbuild: remove all netfilter headers from header-test blacklist.\n \n All the blacklisted NF headers can now be compiled stand-alone, so\n removed them from the blacklist.\n \n Cc: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Jeremy Sowden \u003cjeremy@azazel.net\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit 20a9379d9a03ee0712d225035308973ecc5f6783\nAuthor: Jeremy Sowden \u003cjeremy@azazel.net\u003e\nDate: Wed Aug 7 15:17:04 2019 +0100\n\n netfilter: remove \"#ifdef __KERNEL__\" guards from some headers.\n \n A number of non-UAPI Netfilter header-files contained superfluous\n \"#ifdef __KERNEL__\" guards. Removed them.\n \n Signed-off-by: Jeremy Sowden \u003cjeremy@azazel.net\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit 78458e3e08cda2aacaec9fde8c295dfc2f88618a\nAuthor: Jeremy Sowden \u003cjeremy@azazel.net\u003e\nDate: Wed Aug 7 15:17:03 2019 +0100\n\n netfilter: add missing IS_ENABLED(CONFIG_NETFILTER) checks to some header-files.\n \n linux/netfilter.h defines a number of struct and inline function\n definitions which are only available is CONFIG_NETFILTER is enabled.\n These structs and functions are used in declarations and definitions in\n other header-files. Added preprocessor checks to make sure these\n headers will compile if CONFIG_NETFILTER is disabled.\n \n Signed-off-by: Jeremy Sowden \u003cjeremy@azazel.net\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit 0abc8bf4f2842e409926096f0fa009b468cbd855\nAuthor: Jeremy Sowden \u003cjeremy@azazel.net\u003e\nDate: Wed Aug 7 15:17:02 2019 +0100\n\n netfilter: add missing IS_ENABLED(CONFIG_NF_CONNTRACK) checks to some header-files.\n \n struct nf_conn contains a \"struct nf_conntrack ct_general\" member and\n struct net contains a \"struct netns_ct ct\" member which are both only\n defined in CONFIG_NF_CONNTRACK is enabled. These members are used in a\n number of inline functions defined in other header-files. Added\n preprocessor checks to make sure the headers will compile if\n CONFIG_NF_CONNTRACK is disabled.\n \n Signed-off-by: Jeremy Sowden \u003cjeremy@azazel.net\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit 47e640af2e492cc28778dd6f894d50313f7fba75\nAuthor: Jeremy Sowden \u003cjeremy@azazel.net\u003e\nDate: Wed Aug 7 15:17:01 2019 +0100\n\n netfilter: add missing IS_ENABLED(CONFIG_NF_TABLES) check to header-file.\n \n nf_tables.h defines an API comprising several inline functions and\n macros that depend on the nft member of struct net. However, this is\n only defined is CONFIG_NF_TABLES is enabled. Added preprocessor checks\n to ensure that nf_tables.h will compile if CONFIG_NF_TABLES is disabled.\n \n Signed-off-by: Jeremy Sowden \u003cjeremy@azazel.net\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit 9211bfbff80a7604273086fec5685efcdc10be2b\nAuthor: Jeremy Sowden \u003cjeremy@azazel.net\u003e\nDate: Wed Aug 7 15:17:00 2019 +0100\n\n netfilter: add missing IS_ENABLED(CONFIG_BRIDGE_NETFILTER) checks to header-file.\n \n br_netfilter.h defines inline functions that use an enum constant and\n struct member that are only defined if CONFIG_BRIDGE_NETFILTER is\n enabled. Added preprocessor checks to ensure br_netfilter.h will\n compile if CONFIG_BRIDGE_NETFILTER is disabled.\n \n Signed-off-by: Jeremy Sowden \u003cjeremy@azazel.net\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit a1b2f04ea527397fcacacd09e0d690927feef429\nAuthor: Jeremy Sowden \u003cjeremy@azazel.net\u003e\nDate: Wed Aug 7 15:16:59 2019 +0100\n\n netfilter: add missing includes to a number of header-files.\n \n A number of netfilter header-files used declarations and definitions\n from other headers without including them. Added include directives to\n make those declarations and definitions available.\n \n Signed-off-by: Jeremy Sowden \u003cjeremy@azazel.net\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit bd96b4c75675e616eefef6d85d163530eef9c5e5\nAuthor: Jeremy Sowden \u003cjeremy@azazel.net\u003e\nDate: Wed Aug 7 15:16:58 2019 +0100\n\n netfilter: inline four headers files into another one.\n \n linux/netfilter/ipset/ip_set.h included four other header files:\n \n include/linux/netfilter/ipset/ip_set_comment.h\n include/linux/netfilter/ipset/ip_set_counter.h\n include/linux/netfilter/ipset/ip_set_skbinfo.h\n include/linux/netfilter/ipset/ip_set_timeout.h\n \n Of these the first three were not included anywhere else. The last,\n ip_set_timeout.h, was included in a couple of other places, but defined\n inline functions which call other inline functions defined in ip_set.h,\n so ip_set.h had to be included before it.\n \n Inlined all four into ip_set.h, and updated the other files that\n included ip_set_timeout.h.\n \n Signed-off-by: Jeremy Sowden \u003cjeremy@azazel.net\u003e\n Acked-by: Jozsef Kadlecsik \u003ckadlec@netfilter.org\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit 43dd16efc7f235f153804500a4363769bd2276fc\nAuthor: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\nDate: Thu Aug 1 14:09:26 2019 +0200\n\n netfilter: nf_tables: store data in offload context registers\n \n Store immediate data into offload context register. This allows follow\n up instructions to take it from the corresponding source register.\n \n This patch is required to support for payload mangling, although other\n instructions that take data from source register will benefit from this\n too.\n \n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit bd8699e9e29287b5571b32b68c3dcd05985fa9b1\nAuthor: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\nDate: Tue Jul 30 13:32:01 2019 +0200\n\n netfilter: nft_bitwise: add offload support\n \n Extract mask from bitwise operation and store it into the corresponding\n context register so the cmp instruction can set the mask accordingly.\n \n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit 7e59b3fea2a2510b52761c20ccc71d3e9f6b7db8\nAuthor: yangxingwu \u003cxingwu.yang@gmail.com\u003e\nDate: Tue Jul 16 10:13:01 2019 +0800\n\n netfilter: remove unnecessary spaces\n \n This patch removes extra spaces.\n \n Signed-off-by: yangxingwu \u003cxingwu.yang@gmail.com\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit b1635ee6120cbeb3de6ab270432b2a2b839c9c56\nAuthor: Yishai Hadas \u003cyishaih@mellanox.com\u003e\nDate: Thu Aug 8 11:43:56 2019 +0300\n\n net/mlx5: Add XRQ legacy commands opcodes\n \n Add XRQ legacy commands opcodes, will be used via the DEVX interface.\n \n Signed-off-by: Yishai Hadas \u003cyishaih@mellanox.com\u003e\n Reviewed-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n\ncommit 647d58a989b3b0b788c721a08394aec825e3438c\nAuthor: Yishai Hadas \u003cyishaih@mellanox.com\u003e\nDate: Thu Aug 8 11:43:55 2019 +0300\n\n net/mlx5: Use debug message instead of warn\n \n As QP may be created by DEVX, it may be valid to not find the rsn in\n mlx5 core tree, change the level to be debug.\n \n Signed-off-by: Yishai Hadas \u003cyishaih@mellanox.com\u003e\n Reviewed-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n\ncommit 220b65d8009dff397a112ee560c297299ba5e423\nAuthor: Tina Zhang \u003ctina.zhang@intel.com\u003e\nDate: Mon Aug 12 14:47:51 2019 +0800\n\n drm/i915/gvt: Double check batch buffer size after copy\n \n Double check the end of the privilege buffer to make sure the size\n of the privilege buffer remains unchanged after copy.\n \n v4:\n - Refine the commit message. (Zhenyu)\n v3:\n - To get the right offset of the batch buffer end cmd. (Yan)\n \n v2:\n - Use lightweight way to audit batch buffer end. (Yan)\n \n Reviewed-by: Yan Zhao \u003cyan.y.zhao@intel.com\u003e\n Signed-off-by: Tina Zhang \u003ctina.zhang@intel.com\u003e\n Signed-off-by: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n\ncommit 00a33be406349ae04a676e05c535987750f838ca\nAuthor: Gao, Fred \u003cfred.gao@intel.com\u003e\nDate: Thu Jul 18 09:39:18 2019 +0800\n\n drm/i915/gvt: Add valid length check for MI variable commands\n \n Add valid length check for the commands with variable length.\n \n v2: remove the macro definition. (Zhenyu)\n v3: refine the LRI command. (Zhenyu)\n \n Reviewed-by: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n Signed-off-by: Gao, Fred \u003cfred.gao@intel.com\u003e\n Signed-off-by: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n\ncommit 1e2adc0d037076fb1de6492dd152b368adea0d2d\nAuthor: Gao, Fred \u003cfred.gao@intel.com\u003e\nDate: Thu Jul 18 09:39:09 2019 +0800\n\n drm/i915/gvt: Add MI command valid length check\n \n Add the constant valid length of MI command.\n \n v2: Add F_VAL_CONST flag. (Zhenyu Wang)\n \n Reviewed-by: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n Signed-off-by: Gao, Fred \u003cfred.gao@intel.com\u003e\n Signed-off-by: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n\ncommit 4f870f1fba3443705389d9a2405c10c1cf73f2e0\nAuthor: Gao, Fred \u003cfred.gao@intel.com\u003e\nDate: Thu Jul 18 09:39:01 2019 +0800\n\n drm/i915/gvt: Utility for valid command length check\n \n Add utility for valid command length check.\n \n v2: Add F_VAL_CONST flag to identify the value is const\n although LEN maybe variable. (Zhenyu)\n v3: unused code removal, flag rename/conflict. (Zhenyu)\n v4: redefine F_IP_ADVANCE_CUSTOM and move the check function to\n next patch. (Zhenyu)\n \n Reviewed-by: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n Signed-off-by: Gao, Fred \u003cfred.gao@intel.com\u003e\n Signed-off-by: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n\ncommit 8cfbca7848ffe3f5d49155748814cd68a774e449\nAuthor: Zhi Wang \u003czhi.a.wang@intel.com\u003e\nDate: Mon Jul 22 14:07:07 2019 +0300\n\n drm/i915/gvt: factor out tlb and mocs register offset table\n \n Factor out tlb and mocs register offset table to fix the issues reported\n by klocwork, #512 and #550. Mostly, the reason why the klocwork reports\n these problems is because there can be possbilities for platforms, which\n have more rings than the ring offset table, to take the dirty data from\n the stack as the register offset. It results to a random HW register\n offset writting in this scenairo when doing context switch between vGPUs.\n \n After the factoring, the ring offset table of TLB and MOCS should be per\n platform.\n \n v2:\n \n - Enable TLB register switch for GEN8. (Zhenyu)\n \n Reviewed-by: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n Signed-off-by: Zhi Wang \u003czhi.a.wang@intel.com\u003e\n Signed-off-by: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n\ncommit f8871ec8fc73f57295703a8d61c8c33d7ab4805b\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Thu Jun 13 15:34:19 2019 +0200\n\n drm/i915/gvt: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Because there is no need to check these functions, a number of local\n functions can be made to return void to simplify things as nothing can\n fail.\n \n Cc: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n Cc: Zhi Wang \u003czhi.a.wang@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@linux.intel.com\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: intel-gvt-dev@lists.freedesktop.org\n Cc: intel-gfx@lists.freedesktop.org\n Cc: dri-devel@lists.freedesktop.org\n Reviewed-by: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n\ncommit e14006d939fea6d89ddcbb40a72f5c0089a28c85\nAuthor: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\nDate: Mon Jun 10 17:27:50 2019 +0800\n\n drm/i915/gvt: Fix typo of VBLANK_TIMER_PERIOD\n \n This fixes typo for VBLANK_TIMER_PERIOD.\n \n Reviewed-by: Xiong Zhang \u003cxiong.y.zhang@intel.com\u003e\n Signed-off-by: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n\ncommit abb042f310ff34e88060d618b0fcc9b05baa97e3\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Tue Aug 13 08:15:59 2019 +0000\n\n drm/i915/uc: Log fw status changes only under debug config\n \n We don\u0027t care about internal firmware status changes unless\n we are doing some real debugging. Note that our CI is not\n using DRM_I915_DEBUG_GUC config by default so use it.\n \n v2: protect against accidental overwrites (Chris)\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190813081559.23936-1-michal.wajdeczko@intel.com\n\ncommit faac910201e9beb66530bd8c3fe8a02d907ee2a9\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:39 2019 -0700\n\n rtc: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Alessandro Zummo \u003ca.zummo@towertech.it\u003e\n Cc: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Cc: linux-rtc@vger.kernel.org\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190730181557.90391-40-swboyd@chromium.org\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit baf58858e8b6d853a7a8308901fcdd438e92a522\nAuthor: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nDate: Mon Aug 12 14:50:38 2019 -0700\n\n sh: prefer __section from compiler_attributes.h\n \n Reported-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Suggested-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Signed-off-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Yoshinori Sato \u003cysato@users.sourceforge.jp\u003e\n\ncommit 8619b5a9035aefce160e059b2c814bacae122a39\nAuthor: Guenter Roeck \u003clinux@roeck-us.net\u003e\nDate: Sun Aug 4 19:14:23 2019 -0700\n\n sh: Drop -Werror from kernel Makefile\n \n Since commit a035d552a93b (\"Makefile: Globally enable fall-through\n warning\"), all sh builds fail with errors such as\n \n arch/sh/kernel/disassemble.c: In function \u0027print_sh_insn\u0027:\n arch/sh/kernel/disassemble.c:478:8: error: this statement may fall through\n \n Since this effectively disables all build and boot tests for the\n architecture, let\u0027s drop -Werror from the sh kernel Makefile until\n the problems are fixed.\n \n Cc: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Yoshinori Sato \u003cysato@users.sourceforge.jp\u003e\n Acked-by: Rich Felker \u003cdalias@libc.org\u003e\n\ncommit 52791eeec1d9f4a7e7fe08aaba0b1553149d93bc\nAuthor: Christian König \u003cchristian.koenig@amd.com\u003e\nDate: Sun Aug 11 10:06:32 2019 +0200\n\n dma-buf: rename reservation_object to dma_resv\n \n Be more consistent with the naming of the other DMA-buf objects.\n \n Signed-off-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/323401/\n\ncommit 5d344f58da760b226562e7d5199fb73294eb93fa\nAuthor: Christian König \u003cchristian.koenig@amd.com\u003e\nDate: Mon Aug 5 14:24:55 2019 +0200\n\n dma-buf: nuke reservation_object seq number\n \n The only remaining use for this is to protect against setting a new exclusive\n fence while we grab both exclusive and shared. That can also be archived by\n looking if the exclusive fence has changed or not after completing the\n operation.\n \n v2: switch setting excl fence to rcu_assign_pointer\n \n Signed-off-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/322380/\n\ncommit d3c6dd1fb30d3853c2012549affe75c930f4a2f9\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 12 16:42:47 2019 +0100\n\n dma-buf/sw_sync: Synchronize signal vs syncpt free\n \n During release of the syncpt, we remove it from the list of syncpt and\n the tree, but only if it is not already been removed. However, during\n signaling, we first remove the syncpt from the list. So, if we\n concurrently free and signal the syncpt, the free may decide that it is\n not part of the tree and immediately free itself -- meanwhile the\n signaler goes on to use the now freed datastructure.\n \n In particular, we get struck by commit 0e2f733addbf (\"dma-buf: make\n dma_fence structure a bit smaller v2\") as the cb_list is immediately\n clobbered by the kfree_rcu.\n \n v2: Avoid calling into timeline_fence_release() from under the spinlock\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111381\n Fixes: d3862e44daa7 (\"dma-buf/sw-sync: Fix locking around sync_timeline lists\")\n References: 0e2f733addbf (\"dma-buf: make dma_fence structure a bit smaller v2\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Sumit Semwal \u003csumit.semwal@linaro.org\u003e\n Cc: Sean Paul \u003cseanpaul@chromium.org\u003e\n Cc: Gustavo Padovan \u003cgustavo@padovan.org\u003e\n Cc: Christian König \u003cchristian.koenig@amd.com\u003e\n Cc: \u003cstable@vger.kernel.org\u003e # v4.14+\n Acked-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812154247.20508-1-chris@chris-wilson.co.uk\n\ncommit 5f15c1e6e181636f4b2fe7a67d582356f57cfd2c\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 12 21:36:26 2019 +0100\n\n drm/i915/guc: Use a local cancel_port_requests\n \n Since execlists and the guc have diverged in their port tracking, we\n cannot simply reuse the execlists cancellation code as it leads to\n unbalanced reference counting. Use a local, simpler routine for the guc.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812203626.3948-1-chris@chris-wilson.co.uk\n\ncommit 478ffad6d6902298802b5b8044e542739fcb0ed1\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Mon Aug 12 16:31:51 2019 -0700\n\n drm/i915: drop engine_pin/unpin_breadcrumbs_irq\n \n The last user has been removed, so drop the functions.\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812233152.2172-2-daniele.ceraolospurio@intel.com\n\ncommit ee94e0c4598a34ccbb2927af5700e373e5f96f23\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Mon Aug 12 16:31:50 2019 -0700\n\n drm/i915/guc: keep breadcrumb irq always enabled\n \n We rely on the tasklet to update the GT PM refcount, so we can\u0027t disable\n it even if we\u0027ve processed all the requests for the engine because we\n might have detected the request completion before the interrupt arrived.\n \n Since on all platforms on which we plan to support guc submission we\n don\u0027t allow disabling the breadcrumb interrupts, we can further siplify\n the park/unpark flow by removing the interrupt pin/unpin. A BUG_ON has\n been added to catch changes to this flow that would require us to\n restore some kind of pinning.\n \n v2: split removal of engine_pin/unpin_breadcrumbs_irq to its own\n patch (chris)\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812233152.2172-1-daniele.ceraolospurio@intel.com\n\ncommit 3a3a781246931e110f1590ea99f1fd1d3d7deb72\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Sat Aug 10 00:24:42 2019 -0500\n\n sh: kernel: disassemble: Mark expected switch fall-throughs\n \n Remove logically dead code and mark switch cases where we are expecting\n to fall through.\n \n Fix the following warnings (Building: defconfig sh):\n \n arch/sh/kernel/disassemble.c:478:8: warning: this statement may fall\n through [-Wimplicit-fallthrough\u003d]\n arch/sh/kernel/disassemble.c:487:8: warning: this statement may fall\n through [-Wimplicit-fallthrough\u003d]\n arch/sh/kernel/disassemble.c:496:8: warning: this statement may fall\n through [-Wimplicit-fallthrough\u003d]\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Yoshinori Sato \u003cysato@users.sourceforge.jp\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Tested-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit fb8f77490f55a9d2a526a7a12113485bcec5ef7c\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Fri Aug 9 23:59:44 2019 -0500\n\n sh: kernel: hw_breakpoint: Fix missing break in switch statement\n \n Add missing break statement in order to prevent the code from\n erroneously falling through to case SH_BREAKPOINT_WRITE.\n \n Fixes: 09a072947791 (\"sh: hw-breakpoints: Add preliminary support for SH-4A UBC.\")\n Cc: stable@vger.kernel.org\n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Yoshinori Sato \u003cysato@users.sourceforge.jp\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Tested-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n\ncommit 7607a121f4617840fe645c65f090af6403738031\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Sun Aug 11 19:22:00 2019 -0500\n\n dmaengine: fsldma: Mark expected switch fall-through\n \n Mark switch cases where we are expecting to fall through.\n \n Fix the following warning (Building: powerpc-ppa8548_defconfig powerpc):\n \n drivers/dma/fsldma.c: In function ‘fsl_dma_chan_probe’:\n drivers/dma/fsldma.c:1165:26: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n chan-\u003etoggle_ext_pause \u003d fsl_chan_toggle_ext_pause;\n ~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/dma/fsldma.c:1166:2: note: here\n case FSL_DMA_IP_83XX:\n ^~~~\n \n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Acked-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190812002159.GA26899@embeddedor\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 069e4a19f44d82c43d6a3866499f3d4da6ab3dec\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Mon Aug 12 12:11:43 2019 +0200\n\n dmaengine: ti: unexport filter functions\n \n The two filter functions are now marked static, but still exported,\n which triggers a coming build-time check:\n \n WARNING: \"omap_dma_filter_fn\" [vmlinux] is a static EXPORT_SYMBOL_GPL\n WARNING: \"edma_filter_fn\" [vmlinux] is a static EXPORT_SYMBOL\n \n Remove the unneeded exports as well, as originally intended.\n \n Fixes: 9c71b9eb3cb2 (\"dmaengine: omap-dma: make omap_dma_filter_fn private\")\n Fixes: d2bfe7b5d182 (\"dmaengine: edma: make edma_filter_fn private\")\n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Acked-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Link: https://lore.kernel.org/r/20190812101155.997721-1-arnd@arndb.de\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 84814186da28b5bf5bfca96d59e2275bf910cef5\nAuthor: Steffen Maier \u003cmaier@linux.ibm.com\u003e\nDate: Wed Aug 7 16:49:48 2019 +0200\n\n scsi: core: fix dh and multipathing for SCSI hosts without request batching\n \n This was missing from scsi_device_from_queue() due to the introduction of\n another new scsi_mq_ops_no_commit of linux-next commit 8930a6c20791 (\"scsi:\n core: add support for request batching\") from Martin\u0027s scsi/5.4/scsi-queue\n or James\u0027 scsi/misc.\n \n Only devicehandler code seems to call scsi_device_from_queue():\n *** drivers/scsi/scsi_dh.c:\n scsi_dh_activate[255] sdev \u003d scsi_device_from_queue(q);\n scsi_dh_set_params[302] sdev \u003d scsi_device_from_queue(q);\n scsi_dh_attach[325] sdev \u003d scsi_device_from_queue(q);\n scsi_dh_attached_handler_name[363] sdev \u003d scsi_device_from_queue(q);\n \n Fixes multipath tools follow-on errors:\n \n $ multipath -v6\n ...\n libdevmapper: ioctl/libdm-iface.c(1887): device-mapper: reload ioctl on mpatha failed: No such device\n ...\n mpatha: failed to load map, error 19\n ...\n \n showing also as kernel messages:\n \n device-mapper: table: 252:0: multipath: error attaching hardware handler\n device-mapper: ioctl: error adding target to table\n \n Signed-off-by: Steffen Maier \u003cmaier@linux.ibm.com\u003e\n Fixes: 8930a6c20791 (\"scsi: core: add support for request batching\")\n Cc: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n Reviewed-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n Reviewed-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit e69d4dba64f0602d6e4dd0a0d00ebfe75a3cc430\nAuthor: Steffen Maier \u003cmaier@linux.ibm.com\u003e\nDate: Wed Aug 7 16:49:47 2019 +0200\n\n scsi: core: fix missing .cleanup_rq for SCSI hosts without request batching\n \n This was missing from scsi_mq_ops_no_commit of linux-next commit\n 8930a6c20791 (\"scsi: core: add support for request batching\") from Martin\u0027s\n scsi/5.4/scsi-queue or James\u0027 scsi/misc.\n \n See also linux-next commit b7e9e1fb7a92 (\"scsi: implement .cleanup_rq\n callback\") from block/for-next.\n \n Signed-off-by: Steffen Maier \u003cmaier@linux.ibm.com\u003e\n Fixes: 8930a6c20791 (\"scsi: core: add support for request batching\")\n Cc: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n Cc: Ming Lei \u003cming.lei@redhat.com\u003e\n Reviewed-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n Reviewed-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 95ae251fe82838b85c6d37e5a1775006e2a42ae0\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Mon Jul 22 09:26:24 2019 -0700\n\n f2fs: add fs-verity support\n \n Add fs-verity support to f2fs. fs-verity is a filesystem feature that\n enables transparent integrity protection and authentication of read-only\n files. It uses a dm-verity like mechanism at the file level: a Merkle\n tree is used to verify any block in the file in log(filesize) time. It\n is implemented mainly by helper functions in fs/verity/. See\n Documentation/filesystems/fsverity.rst for the full documentation.\n \n The f2fs support for fs-verity consists of:\n \n - Adding a filesystem feature flag and an inode flag for fs-verity.\n \n - Implementing the fsverity_operations to support enabling verity on an\n inode and reading/writing the verity metadata.\n \n - Updating -\u003ereadpages() to verify data as it\u0027s read from verity files\n and to support reading verity metadata pages.\n \n - Updating -\u003ewrite_begin(), -\u003ewrite_end(), and -\u003ewritepages() to support\n writing verity metadata pages.\n \n - Calling the fs-verity hooks for -\u003eopen(), -\u003esetattr(), and -\u003eioctl().\n \n Like ext4, f2fs stores the verity metadata (Merkle tree and\n fsverity_descriptor) past the end of the file, starting at the first 64K\n boundary beyond i_size. This approach works because (a) verity files\n are readonly, and (b) pages fully beyond i_size aren\u0027t visible to\n userspace but can be read/written internally by f2fs with only some\n relatively small changes to f2fs. Extended attributes cannot be used\n because (a) f2fs limits the total size of an inode\u0027s xattr entries to\n 4096 bytes, which wouldn\u0027t be enough for even a single Merkle tree\n block, and (b) f2fs encryption doesn\u0027t encrypt xattrs, yet the verity\n metadata *must* be encrypted when the file is because it contains hashes\n of the plaintext data.\n \n Acked-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n Acked-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 84fb7ca4b3f864782cb1e66e60b04d4ea1d2930b\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Mon Jul 22 09:26:24 2019 -0700\n\n ext4: update on-disk format documentation for fs-verity\n \n Document the format of verity files on ext4, and the corresponding inode\n and superblock flags.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 22cfe4b48ccb5a3dbb92d6dcb88f396e0f400f74\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Mon Jul 22 09:26:24 2019 -0700\n\n ext4: add fs-verity read support\n \n Make ext4_mpage_readpages() verify data as it is read from fs-verity\n files, using the helper functions from fs/verity/.\n \n To support both encryption and verity simultaneously, this required\n refactoring the decryption workflow into a generic \"post-read\n processing\" workflow which can do decryption, verification, or both.\n \n The case where the ext4 block size is not equal to the PAGE_SIZE is not\n supported yet, since in that case ext4_mpage_readpages() sometimes falls\n back to block_read_full_page(), which does not support fs-verity yet.\n \n Co-developed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit c93d8f88580921c84d2213161ef3c22560511b84\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Mon Jul 22 09:26:24 2019 -0700\n\n ext4: add basic fs-verity support\n \n Add most of fs-verity support to ext4. fs-verity is a filesystem\n feature that enables transparent integrity protection and authentication\n of read-only files. It uses a dm-verity like mechanism at the file\n level: a Merkle tree is used to verify any block in the file in\n log(filesize) time. It is implemented mainly by helper functions in\n fs/verity/. See Documentation/filesystems/fsverity.rst for the full\n documentation.\n \n This commit adds all of ext4 fs-verity support except for the actual\n data verification, including:\n \n - Adding a filesystem feature flag and an inode flag for fs-verity.\n \n - Implementing the fsverity_operations to support enabling verity on an\n inode and reading/writing the verity metadata.\n \n - Updating -\u003ewrite_begin(), -\u003ewrite_end(), and -\u003ewritepages() to support\n writing verity metadata pages.\n \n - Calling the fs-verity hooks for -\u003eopen(), -\u003esetattr(), and -\u003eioctl().\n \n ext4 stores the verity metadata (Merkle tree and fsverity_descriptor)\n past the end of the file, starting at the first 64K boundary beyond\n i_size. This approach works because (a) verity files are readonly, and\n (b) pages fully beyond i_size aren\u0027t visible to userspace but can be\n read/written internally by ext4 with only some relatively small changes\n to ext4. This approach avoids having to depend on the EA_INODE feature\n and on rearchitecturing ext4\u0027s xattr support to support paging\n multi-gigabyte xattrs into memory, and to support encrypting xattrs.\n Note that the verity metadata *must* be encrypted when the file is,\n since it contains hashes of the plaintext data.\n \n This patch incorporates work by Theodore Ts\u0027o and Chandan Rajendra.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 432434c9f8e18cb4cf0fe05bc3eeceada0e10dc6\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Mon Jul 22 09:26:23 2019 -0700\n\n fs-verity: support builtin file signatures\n \n To meet some users\u0027 needs, add optional support for having fs-verity\n handle a portion of the authentication policy in the kernel. An\n \".fs-verity\" keyring is created to which X.509 certificates can be\n added; then a sysctl \u0027fs.verity.require_signatures\u0027 can be set to cause\n the kernel to enforce that all fs-verity files contain a signature of\n their file measurement by a key in this keyring.\n \n See the \"Built-in signature verification\" section of\n Documentation/filesystems/fsverity.rst for the full documentation.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit add890c9f9d2d1d79184ded72f23b37b164fc673\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Mon Jul 22 09:26:23 2019 -0700\n\n fs-verity: add SHA-512 support\n \n Add SHA-512 support to fs-verity. This is primarily a demonstration of\n the trivial changes needed to support a new hash algorithm in fs-verity;\n most users will still use SHA-256, due to the smaller space required to\n store the hashes. But some users may prefer SHA-512.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Reviewed-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 4dd893d832cf4da5409e1fecea9c4f6452a93f2b\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Mon Jul 22 09:26:23 2019 -0700\n\n fs-verity: implement FS_IOC_MEASURE_VERITY ioctl\n \n Add a function for filesystems to call to implement the\n FS_IOC_MEASURE_VERITY ioctl. This ioctl retrieves the file measurement\n that fs-verity calculated for the given file and is enforcing for reads;\n i.e., reads that don\u0027t match this hash will fail. This ioctl can be\n used for authentication or logging of file measurements in userspace.\n \n See the \"FS_IOC_MEASURE_VERITY\" section of\n Documentation/filesystems/fsverity.rst for the documentation.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Reviewed-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 3fda4c617e84c21c8a03b7cc2b3dbfe6c6461592\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Mon Jul 22 09:26:22 2019 -0700\n\n fs-verity: implement FS_IOC_ENABLE_VERITY ioctl\n \n Add a function for filesystems to call to implement the\n FS_IOC_ENABLE_VERITY ioctl. This ioctl enables fs-verity on a file.\n \n See the \"FS_IOC_ENABLE_VERITY\" section of\n Documentation/filesystems/fsverity.rst for the documentation.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Reviewed-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 1799cc573ebbce3000ab7dbe4a0b70ee4235da6d\nAuthor: Ming Lei \u003cming.lei@redhat.com\u003e\nDate: Thu Jul 25 10:05:00 2019 +0800\n\n scsi: implement .cleanup_rq callback\n \n Implement .cleanup_rq() callback for freeing driver private part\n of the request. Then we can avoid to leak this part if the request isn\u0027t\n completed by SCSI, and freed by blk-mq or upper layer(such as dm-rq) finally.\n \n Cc: Ewan D. Milne \u003cemilne@redhat.com\u003e\n Cc: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Cc: Hannes Reinecke \u003chare@suse.com\u003e\n Cc: Christoph Hellwig \u003chch@lst.de\u003e\n Cc: Mike Snitzer \u003csnitzer@redhat.com\u003e\n Cc: dm-devel@redhat.com\n Cc: \u003cstable@vger.kernel.org\u003e\n Fixes: 396eaf21ee17 (\"blk-mq: improve DM\u0027s blk-mq IO merging via blk_insert_cloned_request feedback\")\n Signed-off-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n (cherry picked from commit b7e9e1fb7a9227be34ad4a5e778022c3164494cf)\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit bccd30adf478e152a17d347c4d50177ea94b6114\nAuthor: Ming Lei \u003cming.lei@redhat.com\u003e\nDate: Thu Jul 25 10:04:59 2019 +0800\n\n blk-mq: add callback of .cleanup_rq\n \n SCSI maintains its own driver private data hooked off of each SCSI\n request, and the pridate data won\u0027t be freed after scsi_queue_rq()\n returns BLK_STS_RESOURCE or BLK_STS_DEV_RESOURCE. An upper layer driver\n (e.g. dm-rq) may need to retry these SCSI requests, before SCSI has\n fully dispatched them, due to a lower level SCSI driver\u0027s resource\n limitation identified in scsi_queue_rq(). Currently SCSI\u0027s per-request\n private data is leaked when the upper layer driver (dm-rq) frees and\n then retries these requests in response to BLK_STS_RESOURCE or\n BLK_STS_DEV_RESOURCE returns from scsi_queue_rq().\n \n This usecase is so specialized that it doesn\u0027t warrant training an\n existing blk-mq interface (e.g. blk_mq_free_request) to allow SCSI to\n account for freeing its driver private data -- doing so would add an\n extra branch for handling a special case that all other consumers of\n SCSI (and blk-mq) won\u0027t ever need to worry about.\n \n So the most pragmatic way forward is to delegate freeing SCSI driver\n private data to the upper layer driver (dm-rq). Do so by adding\n new .cleanup_rq callback and calling a new blk_mq_cleanup_rq() method\n from dm-rq. A following commit will implement the .cleanup_rq() hook\n in scsi_mq_ops.\n \n Cc: Ewan D. Milne \u003cemilne@redhat.com\u003e\n Cc: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Cc: Hannes Reinecke \u003chare@suse.com\u003e\n Cc: Christoph Hellwig \u003chch@lst.de\u003e\n Cc: Mike Snitzer \u003csnitzer@redhat.com\u003e\n Cc: dm-devel@redhat.com\n Cc: \u003cstable@vger.kernel.org\u003e\n Fixes: 396eaf21ee17 (\"blk-mq: improve DM\u0027s blk-mq IO merging via blk_insert_cloned_request feedback\")\n Signed-off-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n (cherry picked from commit 226b4fc75c78f9c497c5182d939101b260cfb9f3)\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 075c2b6bf654bdba977795e95bba130113156a35\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sun Jul 21 23:25:02 2019 +0900\n\n scsi: use __u{8,16,32,64} instead of uint{8,16,32,64}_t in uapi headers\n \n When CONFIG_UAPI_HEADER_TEST\u003dy, exported headers are compile-tested to make\n sure they can be included from user-space.\n \n Currently, scsi_bsg_fc.h, scsi_netlink.h, and scsi_netlink_fc.h are\n excluded from the test coverage. To make them join the compile-test, we\n need to fix the build errors attached below.\n \n For a case like this, we decided to use __u{8,16,32,64} variable types in\n this discussion:\n \n https://lkml.org/lkml/2019/6/5/18\n \n Build log:\n \n CC usr/include/scsi/scsi_netlink_fc.h.s\n CC usr/include/scsi/scsi_netlink.h.s\n CC usr/include/scsi/scsi_bsg_fc.h.s\n In file included from ./usr/include/scsi/scsi_netlink_fc.h:10:0,\n from \u003ccommand-line\u003e:32:\n ./usr/include/scsi/scsi_netlink.h:29:2: error: unknown type name uint8_t\n uint8_t version;\n ^~~~~~~\n ./usr/include/scsi/scsi_netlink.h:30:2: error: unknown type name uint8_t\n uint8_t transport;\n ^~~~~~~\n ./usr/include/scsi/scsi_netlink.h:31:2: error: unknown type name uint16_t\n uint16_t magic;\n ^~~~~~~~\n ./usr/include/scsi/scsi_netlink.h:32:2: error: unknown type name uint16_t\n uint16_t msgtype;\n ^~~~~~~~\n CC usr/include/rdma/vmw_pvrdma-abi.h.s\n ./usr/include/scsi/scsi_netlink.h:33:2: error: unknown type name uint16_t\n uint16_t msglen;\n ^~~~~~~~\n ./usr/include/scsi/scsi_netlink.h:34:33: error: uint64_t undeclared here (not in a function); did you mean __uint128_t ?\n } __attribute__((aligned(sizeof(uint64_t))));\n ^~~~~~~~\n __uint128_t\n ./usr/include/scsi/scsi_netlink.h:78:2: error: expected specifier-qualifier-list before uint64_t\n uint64_t vendor_id;\n ^~~~~~~~\n In file included from \u003ccommand-line\u003e:32:0:\n ./usr/include/scsi/scsi_netlink_fc.h:46:2: error: expected specifier-qualifier-list before uint64_t\n uint64_t seconds;\n ^~~~~~~~\n make[2]: *** [scripts/Makefile.build;302: usr/include/scsi/scsi_netlink_fc.h.s] Error 1\n make[2]: *** Waiting for unfinished jobs....\n In file included from \u003ccommand-line\u003e:32:0:\n ./usr/include/scsi/scsi_netlink.h:29:2: error: unknown type name uint8_t\n uint8_t version;\n ^~~~~~~\n ./usr/include/scsi/scsi_netlink.h:30:2: error: unknown type name uint8_t\n uint8_t transport;\n ^~~~~~~\n ./usr/include/scsi/scsi_netlink.h:31:2: error: unknown type name uint16_t\n uint16_t magic;\n ^~~~~~~~\n ./usr/include/scsi/scsi_netlink.h:32:2: error: unknown type name uint16_t\n uint16_t msgtype;\n ^~~~~~~~\n ./usr/include/scsi/scsi_netlink.h:33:2: error: unknown type name uint16_t\n uint16_t msglen;\n ^~~~~~~~\n ./usr/include/scsi/scsi_netlink.h:34:33: error: uint64_t undeclared here (not in a function); did you mean __uint128_t ?\n } __attribute__((aligned(sizeof(uint64_t))));\n ^~~~~~~~\n __uint128_t\n ./usr/include/scsi/scsi_netlink.h:78:2: error: expected specifier-qualifier-list before uint64_t\n uint64_t vendor_id;\n ^~~~~~~~\n make[2]: *** [scripts/Makefile.build;302: usr/include/scsi/scsi_netlink.h.s] Error 1\n In file included from \u003ccommand-line\u003e:32:0:\n ./usr/include/scsi/scsi_bsg_fc.h:69:2: error: unknown type name uint8_t\n uint8_t reserved;\n ^~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:72:2: error: unknown type name uint8_t\n uint8_t port_id[3];\n ^~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:90:2: error: unknown type name uint8_t\n uint8_t reserved;\n ^~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:93:2: error: unknown type name uint8_t\n uint8_t port_id[3];\n ^~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:114:2: error: unknown type name uint8_t\n uint8_t command_code;\n ^~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:117:2: error: unknown type name uint8_t\n uint8_t port_id[3];\n ^~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:154:2: error: unknown type name uint32_t\n uint32_t status; /* See FC_CTELS_STATUS_xxx */\n ^~~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:158:3: error: unknown type name uint8_t\n uint8_t action; /* fragment_id for CT REJECT */\n ^~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:159:3: error: unknown type name uint8_t\n uint8_t reason_code;\n ^~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:160:3: error: unknown type name uint8_t\n uint8_t reason_explanation;\n ^~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:161:3: error: unknown type name uint8_t\n uint8_t vendor_unique;\n ^~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:177:2: error: unknown type name uint8_t\n uint8_t reserved;\n ^~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:180:2: error: unknown type name uint8_t\n uint8_t port_id[3];\n ^~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:185:2: error: unknown type name uint32_t\n uint32_t preamble_word0; /* revision \u0026 IN_ID */\n ^~~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:186:2: error: unknown type name uint32_t\n uint32_t preamble_word1; /* GS_Type, GS_SubType, Options, Rsvd */\n ^~~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:187:2: error: unknown type name uint32_t\n uint32_t preamble_word2; /* Cmd Code, Max Size */\n ^~~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:207:2: error: unknown type name uint64_t\n uint64_t vendor_id;\n ^~~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:210:2: error: unknown type name uint32_t\n uint32_t vendor_cmd[0];\n ^~~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:217:2: error: unknown type name uint32_t\n uint32_t vendor_rsp[0];\n ^~~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:236:2: error: unknown type name uint8_t\n uint8_t els_code;\n ^~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:254:2: error: unknown type name uint32_t\n uint32_t preamble_word0; /* revision \u0026 IN_ID */\n ^~~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:255:2: error: unknown type name uint32_t\n uint32_t preamble_word1; /* GS_Type, GS_SubType, Options, Rsvd */\n ^~~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:256:2: error: unknown type name uint32_t\n uint32_t preamble_word2; /* Cmd Code, Max Size */\n ^~~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:268:2: error: unknown type name uint32_t\n uint32_t msgcode;\n ^~~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:292:2: error: unknown type name uint32_t\n uint32_t result;\n ^~~~~~~~\n ./usr/include/scsi/scsi_bsg_fc.h:295:2: error: unknown type name uint32_t\n uint32_t reply_payload_rcv_len;\n ^~~~~~~~\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit ba13f2c8d7a08362d519c8f91c846a8987b8df5f\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Sun Aug 4 19:35:49 2019 -0700\n\n fscrypt: document the new ioctls and policy version\n \n Update the fscrypt documentation file to catch up to all the latest\n changes, including the new ioctls to manage master encryption keys in\n the filesystem-level keyring and the support for v2 encryption policies.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 62de25927a2558324f2f425545cdc90682168ca9\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Sun Aug 4 19:35:49 2019 -0700\n\n ubifs: wire up new fscrypt ioctls\n \n Wire up the new ioctls for adding and removing fscrypt keys to/from the\n filesystem, and the new ioctl for retrieving v2 encryption policies.\n \n The key removal ioctls also required making UBIFS use\n fscrypt_drop_inode().\n \n For more details see Documentation/filesystems/fscrypt.rst and the\n fscrypt patches that added the implementation of these ioctls.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 8ce589c773180f15d9ef426679a92909e8233dbe\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Sun Aug 4 19:35:48 2019 -0700\n\n f2fs: wire up new fscrypt ioctls\n \n Wire up the new ioctls for adding and removing fscrypt keys to/from the\n filesystem, and the new ioctl for retrieving v2 encryption policies.\n \n The key removal ioctls also required making f2fs_drop_inode() call\n fscrypt_drop_inode().\n \n For more details see Documentation/filesystems/fscrypt.rst and the\n fscrypt patches that added the implementation of these ioctls.\n \n Acked-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 29b3692e6dbf82266ec3c2764c236f8708d7fc89\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Sun Aug 4 19:35:48 2019 -0700\n\n ext4: wire up new fscrypt ioctls\n \n Wire up the new ioctls for adding and removing fscrypt keys to/from the\n filesystem, and the new ioctl for retrieving v2 encryption policies.\n \n The key removal ioctls also required making ext4_drop_inode() call\n fscrypt_drop_inode().\n \n For more details see Documentation/filesystems/fscrypt.rst and the\n fscrypt patches that added the implementation of these ioctls.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 5ab7189a31bad40e4b44020cae6e56c8074721a1\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Sun Aug 4 19:35:48 2019 -0700\n\n fscrypt: require that key be added when setting a v2 encryption policy\n \n By looking up the master keys in a filesystem-level keyring rather than\n in the calling processes\u0027 key hierarchy, it becomes possible for a user\n to set an encryption policy which refers to some key they don\u0027t actually\n know, then encrypt their files using that key. Cryptographically this\n isn\u0027t much of a problem, but the semantics of this would be a bit weird.\n Thus, enforce that a v2 encryption policy can only be set if the user\n has previously added the key, or has capable(CAP_FOWNER).\n \n We tolerate that this problem will continue to exist for v1 encryption\n policies, however; there is no way around that.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 78a1b96bcf7a0721c7852bb1475218c3cbef884a\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Sun Aug 4 19:35:47 2019 -0700\n\n fscrypt: add FS_IOC_REMOVE_ENCRYPTION_KEY_ALL_USERS ioctl\n \n Add a root-only variant of the FS_IOC_REMOVE_ENCRYPTION_KEY ioctl which\n removes all users\u0027 claims of the key, not just the current user\u0027s claim.\n I.e., it always removes the key itself, no matter how many users have\n added it.\n \n This is useful for forcing a directory to be locked, without having to\n figure out which user ID(s) the key was added under. This is planned to\n be used by a command like \u0027sudo fscrypt lock DIR --all-users\u0027 in the\n fscrypt userspace tool (http://github.com/google/fscrypt).\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 23c688b54016eed15d39f4387ca9da241e165922\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Sun Aug 4 19:35:47 2019 -0700\n\n fscrypt: allow unprivileged users to add/remove keys for v2 policies\n \n Allow the FS_IOC_ADD_ENCRYPTION_KEY and FS_IOC_REMOVE_ENCRYPTION_KEY\n ioctls to be used by non-root users to add and remove encryption keys\n from the filesystem-level crypto keyrings, subject to limitations.\n \n Motivation: while privileged fscrypt key management is sufficient for\n some users (e.g. Android and Chromium OS, where a privileged process\n manages all keys), the old API by design also allows non-root users to\n set up and use encrypted directories, and we don\u0027t want to regress on\n that. Especially, we don\u0027t want to force users to continue using the\n old API, running into the visibility mismatch between files and keyrings\n and being unable to \"lock\" encrypted directories.\n \n Intuitively, the ioctls have to be privileged since they manipulate\n filesystem-level state. However, it\u0027s actually safe to make them\n unprivileged if we very carefully enforce some specific limitations.\n \n First, each key must be identified by a cryptographic hash so that a\n user can\u0027t add the wrong key for another user\u0027s files. For v2\n encryption policies, we use the key_identifier for this. v1 policies\n don\u0027t have this, so managing keys for them remains privileged.\n \n Second, each key a user adds is charged to their quota for the keyrings\n service. Thus, a user can\u0027t exhaust memory by adding a huge number of\n keys. By default each non-root user is allowed up to 200 keys; this can\n be changed using the existing sysctl \u0027kernel.keys.maxkeys\u0027.\n \n Third, if multiple users add the same key, we keep track of those users\n of the key (of which there remains a single copy), and won\u0027t really\n remove the key, i.e. \"lock\" the encrypted files, until all those users\n have removed it. This prevents denial of service attacks that would be\n possible under simpler schemes, such allowing the first user who added a\n key to remove it -- since that could be a malicious user who has\n compromised the key. Of course, encryption keys should be kept secret,\n but the idea is that using encryption should never be *less* secure than\n not using encryption, even if your key was compromised.\n \n We tolerate that a user will be unable to really remove a key, i.e.\n unable to \"lock\" their encrypted files, if another user has added the\n same key. But in a sense, this is actually a good thing because it will\n avoid providing a false notion of security where a key appears to have\n been removed when actually it\u0027s still in memory, available to any\n attacker who compromises the operating system kernel.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 5dae460c2292dbbdac3a7a982cd566f470d957a2\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Sun Aug 4 19:35:47 2019 -0700\n\n fscrypt: v2 encryption policy support\n \n Add a new fscrypt policy version, \"v2\". It has the following changes\n from the original policy version, which we call \"v1\" (*):\n \n - Master keys (the user-provided encryption keys) are only ever used as\n input to HKDF-SHA512. This is more flexible and less error-prone, and\n it avoids the quirks and limitations of the AES-128-ECB based KDF.\n Three classes of cryptographically isolated subkeys are defined:\n \n - Per-file keys, like used in v1 policies except for the new KDF.\n \n - Per-mode keys. These implement the semantics of the DIRECT_KEY\n flag, which for v1 policies made the master key be used directly.\n These are also planned to be used for inline encryption when\n support for it is added.\n \n - Key identifiers (see below).\n \n - Each master key is identified by a 16-byte master_key_identifier,\n which is derived from the key itself using HKDF-SHA512. This prevents\n users from associating the wrong key with an encrypted file or\n directory. This was easily possible with v1 policies, which\n identified the key by an arbitrary 8-byte master_key_descriptor.\n \n - The key must be provided in the filesystem-level keyring, not in a\n process-subscribed keyring.\n \n The following UAPI additions are made:\n \n - The existing ioctl FS_IOC_SET_ENCRYPTION_POLICY can now be passed a\n fscrypt_policy_v2 to set a v2 encryption policy. It\u0027s disambiguated\n from fscrypt_policy/fscrypt_policy_v1 by the version code prefix.\n \n - A new ioctl FS_IOC_GET_ENCRYPTION_POLICY_EX is added. It allows\n getting the v1 or v2 encryption policy of an encrypted file or\n directory. The existing FS_IOC_GET_ENCRYPTION_POLICY ioctl could not\n be used because it did not have a way for userspace to indicate which\n policy structure is expected. The new ioctl includes a size field, so\n it is extensible to future fscrypt policy versions.\n \n - The ioctls FS_IOC_ADD_ENCRYPTION_KEY, FS_IOC_REMOVE_ENCRYPTION_KEY,\n and FS_IOC_GET_ENCRYPTION_KEY_STATUS now support managing keys for v2\n encryption policies. Such keys are kept logically separate from keys\n for v1 encryption policies, and are identified by \u0027identifier\u0027 rather\n than by \u0027descriptor\u0027. The \u0027identifier\u0027 need not be provided when\n adding a key, since the kernel will calculate it anyway.\n \n This patch temporarily keeps adding/removing v2 policy keys behind the\n same permission check done for adding/removing v1 policy keys:\n capable(CAP_SYS_ADMIN). However, the next patch will carefully take\n advantage of the cryptographically secure master_key_identifier to allow\n non-root users to add/remove v2 policy keys, thus providing a full\n replacement for v1 policies.\n \n (*) Actually, in the API fscrypt_policy::version is 0 while on-disk\n fscrypt_context::format is 1. But I believe it makes the most sense\n to advance both to \u00272\u0027 to have them be in sync, and to consider the\n numbering to start at 1 except for the API quirk.\n \n Reviewed-by: Paul Crowley \u003cpaulcrowley@google.com\u003e\n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit c1144c9b8ad94d8c11809d75c1f322a853cdfc4a\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Sun Aug 4 19:35:47 2019 -0700\n\n fscrypt: add an HKDF-SHA512 implementation\n \n Add an implementation of HKDF (RFC 5869) to fscrypt, for the purpose of\n deriving additional key material from the fscrypt master keys for v2\n encryption policies. HKDF is a key derivation function built on top of\n HMAC. We choose SHA-512 for the underlying unkeyed hash, and use an\n \"hmac(sha512)\" transform allocated from the crypto API.\n \n We\u0027ll be using this to replace the AES-ECB based KDF currently used to\n derive the per-file encryption keys. While the AES-ECB based KDF is\n believed to meet the original security requirements, it is nonstandard\n and has problems that don\u0027t exist in modern KDFs such as HKDF:\n \n 1. It\u0027s reversible. Given a derived key and nonce, an attacker can\n easily compute the master key. This is okay if the master key and\n derived keys are equally hard to compromise, but now we\u0027d like to be\n more robust against threats such as a derived key being compromised\n through a timing attack, or a derived key for an in-use file being\n compromised after the master key has already been removed.\n \n 2. It doesn\u0027t evenly distribute the entropy from the master key; each 16\n input bytes only affects the corresponding 16 output bytes.\n \n 3. It isn\u0027t easily extensible to deriving other values or keys, such as\n a public hash for securely identifying the key, or per-mode keys.\n Per-mode keys will be immediately useful for Adiantum encryption, for\n which fscrypt currently uses the master key directly, introducing\n unnecessary usage constraints. Per-mode keys will also be useful for\n hardware inline encryption, which is currently being worked on.\n \n HKDF solves all the above problems.\n \n Reviewed-by: Paul Crowley \u003cpaulcrowley@google.com\u003e\n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 5a7e29924dac34f0690b06906aad9d70d3c022a5\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Sun Aug 4 19:35:46 2019 -0700\n\n fscrypt: add FS_IOC_GET_ENCRYPTION_KEY_STATUS ioctl\n \n Add a new fscrypt ioctl, FS_IOC_GET_ENCRYPTION_KEY_STATUS. Given a key\n specified by \u0027struct fscrypt_key_specifier\u0027 (the same way a key is\n specified for the other fscrypt key management ioctls), it returns\n status information in a \u0027struct fscrypt_get_key_status_arg\u0027.\n \n The main motivation for this is that applications need to be able to\n check whether an encrypted directory is \"unlocked\" or not, so that they\n can add the key if it is not, and avoid adding the key (which may\n involve prompting the user for a passphrase) if it already is.\n \n It\u0027s possible to use some workarounds such as checking whether opening a\n regular file fails with ENOKEY, or checking whether the filenames \"look\n like gibberish\" or not. However, no workaround is usable in all cases.\n \n Like the other key management ioctls, the keyrings syscalls may seem at\n first to be a good fit for this. Unfortunately, they are not. Even if\n we exposed the keyring ID of the -\u003es_master_keys keyring and gave\n everyone Search permission on it (note: currently the keyrings\n permission system would also allow everyone to \"invalidate\" the keyring\n too), the fscrypt keys have an additional state that doesn\u0027t map cleanly\n to the keyrings API: the secret can be removed, but we can be still\n tracking the files that were using the key, and the removal can be\n re-attempted or the secret added again.\n \n After later patches, some applications will also need a way to determine\n whether a key was added by the current user vs. by some other user.\n Reserved fields are included in fscrypt_get_key_status_arg for this and\n other future extensions.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit b1c0ec3599f42ad372063b0235a3c33f65eb1e30\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Sun Aug 4 19:35:46 2019 -0700\n\n fscrypt: add FS_IOC_REMOVE_ENCRYPTION_KEY ioctl\n \n Add a new fscrypt ioctl, FS_IOC_REMOVE_ENCRYPTION_KEY. This ioctl\n removes an encryption key that was added by FS_IOC_ADD_ENCRYPTION_KEY.\n It wipes the secret key itself, then \"locks\" the encrypted files and\n directories that had been unlocked using that key -- implemented by\n evicting the relevant dentries and inodes from the VFS caches.\n \n The problem this solves is that many fscrypt users want the ability to\n remove encryption keys, causing the corresponding encrypted directories\n to appear \"locked\" (presented in ciphertext form) again. Moreover,\n users want removing an encryption key to *really* remove it, in the\n sense that the removed keys cannot be recovered even if kernel memory is\n compromised, e.g. by the exploit of a kernel security vulnerability or\n by a physical attack. This is desirable after a user logs out of the\n system, for example. In many cases users even already assume this to be\n the case and are surprised to hear when it\u0027s not.\n \n It is not sufficient to simply unlink the master key from the keyring\n (or to revoke or invalidate it), since the actual encryption transform\n objects are still pinned in memory by their inodes. Therefore, to\n really remove a key we must also evict the relevant inodes.\n \n Currently one workaround is to run \u0027sync \u0026\u0026 echo 2 \u003e\n /proc/sys/vm/drop_caches\u0027. But, that evicts all unused inodes in the\n system rather than just the inodes associated with the key being\n removed, causing severe performance problems. Moreover, it requires\n root privileges, so regular users can\u0027t \"lock\" their encrypted files.\n \n Another workaround, used in Chromium OS kernels, is to add a new\n VFS-level ioctl FS_IOC_DROP_CACHE which is a more restricted version of\n drop_caches that operates on a single super_block. It does:\n \n shrink_dcache_sb(sb);\n invalidate_inodes(sb, false);\n \n But it\u0027s still a hack. Yet, the major users of filesystem encryption\n want this feature badly enough that they are actually using these hacks.\n \n To properly solve the problem, start maintaining a list of the inodes\n which have been \"unlocked\" using each master key. Originally this\n wasn\u0027t possible because the kernel didn\u0027t keep track of in-use master\n keys at all. But, with the -\u003es_master_keys keyring it is now possible.\n \n Then, add an ioctl FS_IOC_REMOVE_ENCRYPTION_KEY. It finds the specified\n master key in -\u003es_master_keys, then wipes the secret key itself, which\n prevents any additional inodes from being unlocked with the key. Then,\n it syncs the filesystem and evicts the inodes in the key\u0027s list. The\n normal inode eviction code will free and wipe the per-file keys (in\n -\u003ei_crypt_info). Note that freeing -\u003ei_crypt_info without evicting the\n inodes was also considered, but would have been racy.\n \n Some inodes may still be in use when a master key is removed, and we\n can\u0027t simply revoke random file descriptors, mmap\u0027s, etc. Thus, the\n ioctl simply skips in-use inodes, and returns -EBUSY to indicate that\n some inodes weren\u0027t evicted. The master key *secret* is still removed,\n but the fscrypt_master_key struct remains to keep track of the remaining\n inodes. Userspace can then retry the ioctl to evict the remaining\n inodes. Alternatively, if userspace adds the key again, the refreshed\n secret will be associated with the existing list of inodes so they\n remain correctly tracked for future key removals.\n \n The ioctl doesn\u0027t wipe pagecache pages. Thus, we tolerate that after a\n kernel compromise some portions of plaintext file contents may still be\n recoverable from memory. This can be solved by enabling page poisoning\n system-wide, which security conscious users may choose to do. But it\u0027s\n very difficult to solve otherwise, e.g. note that plaintext file\n contents may have been read in other places than pagecache pages.\n \n Like FS_IOC_ADD_ENCRYPTION_KEY, FS_IOC_REMOVE_ENCRYPTION_KEY is\n initially restricted to privileged users only. This is sufficient for\n some use cases, but not all. A later patch will relax this restriction,\n but it will require introducing key hashes, among other changes.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 22d94f493bfb408fdd764f7b1d0363af2122fba5\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Sun Aug 4 19:35:46 2019 -0700\n\n fscrypt: add FS_IOC_ADD_ENCRYPTION_KEY ioctl\n \n Add a new fscrypt ioctl, FS_IOC_ADD_ENCRYPTION_KEY. This ioctl adds an\n encryption key to the filesystem\u0027s fscrypt keyring -\u003es_master_keys,\n making any files encrypted with that key appear \"unlocked\".\n \n Why we need this\n ~~~~~~~~~~~~~~~~\n \n The main problem is that the \"locked/unlocked\" (ciphertext/plaintext)\n status of encrypted files is global, but the fscrypt keys are not.\n fscrypt only looks for keys in the keyring(s) the process accessing the\n filesystem is subscribed to: the thread keyring, process keyring, and\n session keyring, where the session keyring may contain the user keyring.\n \n Therefore, userspace has to put fscrypt keys in the keyrings for\n individual users or sessions. But this means that when a process with a\n different keyring tries to access encrypted files, whether they appear\n \"unlocked\" or not is nondeterministic. This is because it depends on\n whether the files are currently present in the inode cache.\n \n Fixing this by consistently providing each process its own view of the\n filesystem depending on whether it has the key or not isn\u0027t feasible due\n to how the VFS caches work. Furthermore, while sometimes users expect\n this behavior, it is misguided for two reasons. First, it would be an\n OS-level access control mechanism largely redundant with existing access\n control mechanisms such as UNIX file permissions, ACLs, LSMs, etc.\n Encryption is actually for protecting the data at rest.\n \n Second, almost all users of fscrypt actually do need the keys to be\n global. The largest users of fscrypt, Android and Chromium OS, achieve\n this by having PID 1 create a \"session keyring\" that is inherited by\n every process. This works, but it isn\u0027t scalable because it prevents\n session keyrings from being used for any other purpose.\n \n On general-purpose Linux distros, the \u0027fscrypt\u0027 userspace tool [1] can\u0027t\n similarly abuse the session keyring, so to make \u0027sudo\u0027 work on all\n systems it has to link all the user keyrings into root\u0027s user keyring\n [2]. This is ugly and raises security concerns. Moreover it can\u0027t make\n the keys available to system services, such as sshd trying to access the\n user\u0027s \u0027~/.ssh\u0027 directory (see [3], [4]) or NetworkManager trying to\n read certificates from the user\u0027s home directory (see [5]); or to Docker\n containers (see [6], [7]).\n \n By having an API to add a key to the *filesystem* we\u0027ll be able to fix\n the above bugs, remove userspace workarounds, and clearly express the\n intended semantics: the locked/unlocked status of an encrypted directory\n is global, and encryption is orthogonal to OS-level access control.\n \n Why not use the add_key() syscall\n ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n \n We use an ioctl for this API rather than the existing add_key() system\n call because the ioctl gives us the flexibility needed to implement\n fscrypt-specific semantics that will be introduced in later patches:\n \n - Supporting key removal with the semantics such that the secret is\n removed immediately and any unused inodes using the key are evicted;\n also, the eviction of any in-use inodes can be retried.\n \n - Calculating a key-dependent cryptographic identifier and returning it\n to userspace.\n \n - Allowing keys to be added and removed by non-root users, but only keys\n for v2 encryption policies; and to prevent denial-of-service attacks,\n users can only remove keys they themselves have added, and a key is\n only really removed after all users who added it have removed it.\n \n Trying to shoehorn these semantics into the keyrings syscalls would be\n very difficult, whereas the ioctls make things much easier.\n \n However, to reuse code the implementation still uses the keyrings\n service internally. Thus we get lockless RCU-mode key lookups without\n having to re-implement it, and the keys automatically show up in\n /proc/keys for debugging purposes.\n \n References:\n \n [1] https://github.com/google/fscrypt\n [2] https://goo.gl/55cCrI#heading\u003dh.vf09isp98isb\n [3] https://github.com/google/fscrypt/issues/111#issuecomment-444347939\n [4] https://github.com/google/fscrypt/issues/116\n [5] https://bugs.launchpad.net/ubuntu/+source/fscrypt/+bug/1770715\n [6] https://github.com/google/fscrypt/issues/128\n [7] https://askubuntu.com/questions/1130306/cannot-run-docker-on-an-encrypted-filesystem\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit feed825861919ac3fac4b2ab83673dd5225f7f0d\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Sun Aug 4 19:35:45 2019 -0700\n\n fscrypt: rename keyinfo.c to keysetup.c\n \n Rename keyinfo.c to keysetup.c since this better describes what the file\n does (sets up the key), and it matches the new file keysetup_v1.c.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 0109ce76dd6e9e69559e56ea9f10d1e12ca4a50b\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Sun Aug 4 19:35:45 2019 -0700\n\n fscrypt: move v1 policy key setup to keysetup_v1.c\n \n In preparation for introducing v2 encryption policies which will find\n and derive encryption keys differently from the current v1 encryption\n policies, move the v1 policy-specific key setup code from keyinfo.c into\n keysetup_v1.c.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 3ec4f2a62927c72607db633f55b0489e3d5b8e48\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Sun Aug 4 19:35:45 2019 -0700\n\n fscrypt: refactor key setup code in preparation for v2 policies\n \n Do some more refactoring of the key setup code, in preparation for\n introducing a filesystem-level keyring and v2 encryption policies:\n \n - Now that ci_inode exists, don\u0027t pass around the inode unnecessarily.\n \n - Define a function setup_file_encryption_key() which handles the crypto\n key setup given an under-construction fscrypt_info. Don\u0027t pass the\n fscrypt_context, since everything is in the fscrypt_info.\n [This will be extended for v2 policies and the fs-level keyring.]\n \n - Define a function fscrypt_set_derived_key() which sets the per-file\n key, without depending on anything specific to v1 policies.\n [This will also be used for v2 policies.]\n \n - Define a function fscrypt_setup_v1_file_key() which takes the raw\n master key, thus separating finding the key from using it.\n [This will also be used if the key is found in the fs-level keyring.]\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit a828daabb2ae917c12f4dc617c04db1185e6221f\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Sun Aug 4 19:35:45 2019 -0700\n\n fscrypt: rename fscrypt_master_key to fscrypt_direct_key\n \n In preparation for introducing a filesystem-level keyring which will\n contain fscrypt master keys, rename the existing \u0027struct\n fscrypt_master_key\u0027 to \u0027struct fscrypt_direct_key\u0027. This is the\n structure in the existing table of master keys that\u0027s maintained to\n deduplicate the crypto transforms for v1 DIRECT_KEY policies.\n \n I\u0027ve chosen to keep this table as-is rather than make it automagically\n add/remove the keys to/from the filesystem-level keyring, since that\n would add a lot of extra complexity to the filesystem-level keyring.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 59dc6a8e1f534cde6f7986f2fd278062e25336c0\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Sun Aug 4 19:35:44 2019 -0700\n\n fscrypt: add -\u003eci_inode to fscrypt_info\n \n Add an inode back-pointer to \u0027struct fscrypt_info\u0027, such that\n inode-\u003ei_crypt_info-\u003eci_inode \u003d\u003d inode.\n \n This will be useful for:\n \n 1. Evicting the inodes when a fscrypt key is removed, since we\u0027ll track\n the inodes using a given key by linking their fscrypt_infos together,\n rather than the inodes directly. This avoids bloating \u0027struct inode\u0027\n with a new list_head.\n \n 2. Simplifying the per-file key setup, since the inode pointer won\u0027t\n have to be passed around everywhere just in case something goes wrong\n and it\u0027s needed for fscrypt_warn().\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 3b6df59bc4d242ac5847592de55d1ff327cd4549\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Sun Aug 4 19:35:44 2019 -0700\n\n fscrypt: use FSCRYPT_* definitions, not FS_*\n \n Update fs/crypto/ to use the new names for the UAPI constants rather\n than the old names, then make the old definitions conditional on\n !__KERNEL__.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 2336d0deb2d4680349de59d6fbdfc338437be191\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Sun Aug 4 19:35:44 2019 -0700\n\n fscrypt: use FSCRYPT_ prefix for uapi constants\n \n Prefix all filesystem encryption UAPI constants except the ioctl numbers\n with \"FSCRYPT_\" rather than with \"FS_\". This namespaces the constants\n more appropriately and makes it clear that they are related specifically\n to the filesystem encryption feature, and to the \u0027fscrypt_*\u0027 structures.\n With some of the old names like \"FS_POLICY_FLAGS_VALID\", it was not\n immediately clear that the constant had anything to do with encryption.\n \n This is also useful because we\u0027ll be adding more encryption-related\n constants, e.g. for the policy version, and we\u0027d otherwise have to\n choose whether to use unclear names like FS_POLICY_V1 or inconsistent\n names like FS_ENCRYPTION_POLICY_V1.\n \n For source compatibility with existing userspace programs, keep the old\n names defined as aliases to the new names.\n \n Finally, as long as new names are being defined anyway, I skipped\n defining new names for the fscrypt mode numbers that aren\u0027t actually\n used: INVALID (0), AES_256_GCM (2), AES_256_CBC (3), SPECK128_256_XTS\n (7), and SPECK128_256_CTS (8).\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 7af0ab0d3aab951518b0d520f95e9f6b1995ec69\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Sun Aug 4 19:35:43 2019 -0700\n\n fs, fscrypt: move uapi definitions to new header \u003clinux/fscrypt.h\u003e\n \n More fscrypt definitions are being added, and we shouldn\u0027t use a\n disproportionate amount of space in \u003clinux/fs.h\u003e for fscrypt stuff.\n So move the fscrypt definitions to a new header \u003clinux/fscrypt.h\u003e.\n \n For source compatibility with existing userspace programs, \u003clinux/fs.h\u003e\n still includes the new header.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 29a98c1caf7b37b12a79eee7f839bf2924593c1c\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Wed Jul 24 11:08:00 2019 -0700\n\n fscrypt: use ENOPKG when crypto API support missing\n \n Return ENOPKG rather than ENOENT when trying to open a file that\u0027s\n encrypted using algorithms not available in the kernel\u0027s crypto API.\n \n This avoids an ambiguity, since ENOENT is also returned when the file\n doesn\u0027t exist.\n \n Note: this is the same approach I\u0027m taking for fs-verity.\n \n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit a4d14e915bcb86e13b45231cd4fe2ce19bd9ba86\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Wed Jul 24 11:07:59 2019 -0700\n\n fscrypt: improve warnings for missing crypto API support\n \n Users of fscrypt with non-default algorithms will encounter an error\n like the following if they fail to include the needed algorithms into\n the crypto API when configuring the kernel (as per the documentation):\n \n Error allocating \u0027adiantum(xchacha12,aes)\u0027 transform: -2\n \n This requires that the user figure out what the \"-2\" error means.\n Make it more friendly by printing a warning like the following instead:\n \n Missing crypto API support for Adiantum (API name: \"adiantum(xchacha12,aes)\")\n \n Also upgrade the log level for *other* errors to KERN_ERR.\n \n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 63f668f0def1b58ed1e480661c4043d2b11abc6c\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Wed Jul 24 11:07:59 2019 -0700\n\n fscrypt: improve warning messages for unsupported encryption contexts\n \n When fs/crypto/ encounters an inode with an invalid encryption context,\n currently it prints a warning if the pair of encryption modes are\n unrecognized, but it\u0027s silent if there are other problems such as\n unsupported context size, format, or flags. To help people debug such\n situations, add more warning messages.\n \n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 886da8b39cf27995836062bb7fe5fd5cb764540a\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Wed Jul 24 11:07:58 2019 -0700\n\n fscrypt: make fscrypt_msg() take inode instead of super_block\n \n Most of the warning and error messages in fs/crypto/ are for situations\n related to a specific inode, not merely to a super_block. So to make\n things easier, make fscrypt_msg() take an inode rather than a\n super_block, and make it print the inode number.\n \n Note: This is the same approach I\u0027m taking for fsverity_msg().\n \n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 1c5100a2aab4143f36ca671c28189a8734d5381a\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Wed Jul 24 11:07:58 2019 -0700\n\n fscrypt: clean up base64 encoding/decoding\n \n Some minor cleanups for the code that base64 encodes and decodes\n encrypted filenames and long name digests:\n \n - Rename \"digest_{encode,decode}()\" \u003d\u003e \"base64_{encode,decode}()\" since\n they are used for filenames too, not just for long name digests.\n - Replace \u0027while\u0027 loops with more conventional \u0027for\u0027 loops.\n - Use \u0027u8\u0027 for binary data. Keep \u0027char\u0027 for string data.\n - Fully constify the lookup table (pointer was not const).\n - Improve comment.\n \n No actual change in behavior.\n \n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 75798f85f2badb04074cc909dbbb93072f4041ff\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Wed Jul 24 11:07:57 2019 -0700\n\n fscrypt: remove loadable module related code\n \n Since commit 643fa9612bf1 (\"fscrypt: remove filesystem specific build\n config option\"), fs/crypto/ can no longer be built as a loadable module.\n Thus it no longer needs a module_exit function, nor a MODULE_LICENSE.\n So remove them, and change module_init to late_initcall.\n \n Reviewed-by: Chandan Rajendra \u003cchandan@linux.ibm.com\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 2b4f4cb93ae32f40dc232f8699b125541f494b3d\nAuthor: Frederick Lawler \u003cfred@fredlawl.com\u003e\nDate: Wed Jul 17 21:07:44 2019 -0500\n\n scsi: esas2r: Prefer pcie_capability_read_word()\n \n Commit 8c0d3a02c130 (\"PCI: Add accessors for PCI Express Capability\") added\n accessors for the PCI Express Capability so that drivers didn\u0027t need to be\n aware of differences between v1 and v2 of the PCI Express Capability.\n \n Replace pci_read_config_word() and pci_write_config_word() calls with\n pcie_capability_read_word() and pcie_capability_write_word().\n \n Signed-off-by: Frederick Lawler \u003cfred@fredlawl.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 3e76ca9561ce6bd245793edebd36e718bff89d3e\nAuthor: Frederick Lawler \u003cfred@fredlawl.com\u003e\nDate: Wed Jul 17 21:07:43 2019 -0500\n\n scsi: csiostor: Prefer pcie_capability_read_word()\n \n Commit 8c0d3a02c130 (\"PCI: Add accessors for PCI Express Capability\") added\n accessors for the PCI Express Capability so that drivers didn\u0027t need to be\n aware of differences between v1 and v2 of the PCI Express Capability.\n \n Replace pci_read_config_word() and pci_write_config_word() calls with\n pcie_capability_read_word() and pcie_capability_write_word().\n \n Signed-off-by: Frederick Lawler \u003cfred@fredlawl.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 1d4f4a5e507f5879af56370380df600dd9119030\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Aug 9 18:59:32 2019 +0100\n\n scsi: sym53c8xx_2: remove redundant assignment to retv\n \n Variable retv is initialized to a value that is never read and it is\n re-assigned later. The initialization is redundant and can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Acked-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 18daf910ed6ff408767b8d35aaa2f9f11119cd6d\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Fri Aug 9 17:02:14 2019 +0200\n\n scsi: pmcraid: Fix a typo - pcmraid --\u003e pmcraid\n \n This should be \u0027pmcraid\u0027, not \u0027pcmraid\u0027\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit dbe447ac18aa99a15f462a63bc8016f3da139f23\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Tue Aug 6 03:29:02 2019 -0500\n\n scsi: fas216: Mark expected switch fall-throughs\n \n Mark switch cases where we are expecting to fall through.\n \n Fix the following warnings (Building: rpc_defconfig arm):\n \n drivers/scsi/arm/fas216.c: In function fas216_disconnect_intr :\n drivers/scsi/arm/fas216.c:913:6: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n if (fas216_get_last_msg(info, info-\u003escsi.msgin_fifo) \u003d\u003d ABORT) {\n ^\n drivers/scsi/arm/fas216.c:919:2: note: here\n default: /* huh? */\n ^~~~~~~\n drivers/scsi/arm/fas216.c: In function fas216_kick :\n drivers/scsi/arm/fas216.c:1959:3: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n fas216_allocate_tag(info, SCpnt);\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/arm/fas216.c:1960:2: note: here\n case TYPE_OTHER:\n ^~~~\n drivers/scsi/arm/fas216.c: In function fas216_busservice_intr :\n drivers/scsi/arm/fas216.c:1413:3: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n fas216_stoptransfer(info);\n ^~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/arm/fas216.c:1414:2: note: here\n case STATE(STAT_STATUS, PHASE_SELSTEPS):/* Sel w/ steps -\u003e Status */\n ^~~~\n drivers/scsi/arm/fas216.c:1424:3: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n fas216_stoptransfer(info);\n ^~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/arm/fas216.c:1425:2: note: here\n case STATE(STAT_MESGIN, PHASE_COMMAND): /* Command -\u003e Message In */\n ^~~~\n drivers/scsi/arm/fas216.c: In function fas216_funcdone_intr :\n drivers/scsi/arm/fas216.c:1573:6: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n if ((stat \u0026 STAT_BUSMASK) \u003d\u003d STAT_MESGIN) {\n ^\n drivers/scsi/arm/fas216.c:1579:2: note: here\n default:\n ^~~~~~~\n drivers/scsi/arm/fas216.c: In function fas216_handlesync :\n drivers/scsi/arm/fas216.c:605:20: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n info-\u003escsi.phase \u003d PHASE_MSGOUT_EXPECT;\n ~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/arm/fas216.c:607:2: note: here\n case async:\n ^~~~\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 9ab089d30bcf09d154c0da78920c13b581d23286\nAuthor: Chandrakanth Patil \u003cchandrakanth.patil@broadcom.com\u003e\nDate: Fri Jul 26 13:02:14 2019 +0530\n\n scsi: megaraid_sas: Introduce module parameter for default queue depth\n \n This patch provides a module parameter and sysfs interface to select\n whether the queue depth for each device should be based on the value\n suggested by firmware (the default) or the maximum supported by the\n controller (can_queue).\n \n Although we have a sysfs interface per sdev to change the queue depth of\n individual scsi devices, this implementation provides a single sysfs entry\n per shost to switch between the controller max and the value reported by\n firmware. The module parameter can provide an interface for one time grub\n settings and provides persistent settings across the boot.\n \n [mkp: tweaked commit desc]\n \n Signed-off-by: Kashyap Desai \u003ckashyap.desai@broadcom.com\u003e\n Signed-off-by: Chandrakanth Patil \u003cchandrakanth.patil@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit d1436e456b840162eb4d162f99055d2adb79a566\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:02:19 2019 -0700\n\n scsi: qla2xxx: Fix a NULL pointer dereference\n \n BUG: KASAN: null-ptr-deref in qla24xx_handle_plogi_done_event+0x134/0x9f0 [qla2xxx]\n Read of size 4 at addr 00000000000000a0 by task swapper/2/0\n \n CPU: 2 PID: 0 Comm: swapper/2 Not tainted 5.2.0-dbg+ #1\n Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011\n Call Trace:\n \u003cIRQ\u003e\n dump_stack+0x8a/0xd6\n __kasan_report.cold+0x5/0x41\n kasan_report+0x16/0x20\n __asan_load4+0x7e/0x80\n qla24xx_handle_plogi_done_event+0x134/0x9f0 [qla2xxx]\n qla2x00_els_dcmd2_sp_done+0x15f/0x230 [qla2xxx]\n qla24xx_els_ct_entry+0x3b3/0x610 [qla2xxx]\n qla24xx_process_response_queue+0x514/0x10e0 [qla2xxx]\n qla24xx_msix_rsp_q+0x80/0x100 [qla2xxx]\n __handle_irq_event_percpu+0x72/0x450\n handle_irq_event_percpu+0x74/0xf0\n handle_irq_event+0x5e/0x8f\n handle_edge_irq+0x13a/0x320\n handle_irq+0x30/0x40\n do_IRQ+0x91/0x190\n common_interrupt+0xf/0xf\n \u003c/IRQ\u003e\n RIP: 0010:default_idle+0x31/0x230\n \n Fixes: 8777e4314d39 (\"scsi: qla2xxx: Migrate NVME N2N handling into state machine\") # v4.19.\n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit fe6b52c9adbcce03c549d07c3db06b1694581225\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:02:18 2019 -0700\n\n scsi: qla2xxx: Simplify qla24xx_async_abort_cmd()\n \n Make the code easier to read by converting \u0027goto\u0027 statements into\n \u0027return\u0027 statements.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 1778a0e7bff8d4468658e7b4e5a16d99048cd7a8\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:02:17 2019 -0700\n\n scsi: qla2xxx: Remove two superfluous if-tests\n \n This patch fixes the following Coverity complaint:\n \n Null-checking sp-\u003eu.iocb_cmd.u.ctarg.rsp suggests that it may be null, but\n it has already been dereferenced on all paths leading to the check.\n \n See also commit e374f9f59281 (\"scsi: qla2xxx: Migrate switch registration commands away from mailbox interface\") # v4.16.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit c6e58160a2c0caf6a8079134f0807bef0ecec6b8\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:02:16 2019 -0700\n\n scsi: qla2xxx: Introduce qla2x00_els_dcmd2_free()\n \n This patch reduces code duplication.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 897def2004213636ffe2e9ee6a75660c5b53b03d\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:02:15 2019 -0700\n\n scsi: qla2xxx: Inline the qla2x00_fcport_event_handler() function\n \n Instead of calling qla2x00_fcport_event_handler() and letting the switch\n statement inside that function decide which other function to call, call\n the latter function directly. Remove the event member from the event_arg\n structure because it is no longer needed. Remove the\n qla_handle_els_plogi_done() function because it is never called.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 724361921f65a40ae5b80641dc1e92c0ff314d89\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:02:14 2019 -0700\n\n scsi: qla2xxx: Report invalid mailbox status codes\n \n It is easy to mix up the QLA_* and the MBS_* status codes. Complain loudly\n if that happens.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 1a20719e074ada3e6d06ee065afd527779a6793b\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:02:13 2019 -0700\n\n scsi: qla2xxx: Remove superfluous sts_entry_* casts\n \n The C language supports implicit casting of void pointers to non-void\n pointers. Remove explicit sts_entry_* casts that are not necessary.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 5ec9f9040f61dfd0412a53fa2fc1e321727a5a61\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:02:12 2019 -0700\n\n scsi: qla2xxx: Let the compiler check the type of the SCSI command context pointer\n \n Split srb_cmd.ctx into two pointers such that the compiler can check the\n type of that pointer.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 88263208dd23327b56fbd7c2858c389c1153b366\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:02:11 2019 -0700\n\n scsi: qla2xxx: Complain if sp-\u003edone() is not called from the completion path\n \n Not calling sp-\u003edone() from the command completion path is a severe bug.\n Hence complain loudly if that happens.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 0dcec41acb85da33841c2ab56dbf337ed00a3914\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:02:10 2019 -0700\n\n scsi: qla2xxx: Make sure that aborted commands are freed\n \n The LIO core requires that the target driver callback functions\n .queue_data_in() and .queue_status() call target_put_sess_cmd() or\n transport_generic_free_cmd(). These calls may happen synchronously or\n asynchronously. Make sure that one of these LIO functions is called in case\n a command has been aborted. This patch avoids that the code for removing a\n session hangs due to commands that do not make progress.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Fixes: 694833ee00c4 (\"scsi: tcm_qla2xxx: Do not allow aborted cmd to advance.\") # v4.13.\n Fixes: a07100e00ac4 (\"qla2xxx: Fix TMR ABORT interaction issue between qla2xxx and TCM\") # v4.5.\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit bcc856574356df2eb5eb2d4491e7205339e55728\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:02:09 2019 -0700\n\n scsi: qla2xxx: Introduce qla2xxx_get_next_handle()\n \n This patch reduces code duplication.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 3cb5f3aeaf07215ba86f0ed45f084a9056968dd5\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:02:08 2019 -0700\n\n scsi: qla2xxx: Modify NVMe include directives\n \n Since struct sg_table is used in nvme-fc-driver.h, include\n \u003clinux/scatterlist.h\u003e from that header file.\n \n Since no definitions or declarations from \u003clinux/blk-mq.h\u003e are used in the\n qla_nvme.h header file, do not include \u003clinux/blk-mq.h\u003e from that header\n file.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit e752a04e1bd14ccff2762605004f515ad9b0d48a\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:02:07 2019 -0700\n\n scsi: qla2xxx: Make qlt_handle_abts_completion() more robust\n \n Avoid that this function crashes if mcmd \u003d\u003d NULL.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 85cffefa09e448906a6f0bc20f422d75a18675bd\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:02:06 2019 -0700\n\n scsi: qla2xxx: Fix a race condition between aborting and completing a SCSI command\n \n Instead of allocating a struct srb dynamically from inside .queuecommand(),\n set qla2xxx_driver_template.cmd_size such that struct scsi_cmnd and struct\n srb are contiguous. Do not call QLA_QPAIR_MARK_BUSY() /\n QLA_QPAIR_MARK_NOT_BUSY() for SRBs associated with SCSI commands. That is\n safe because scsi_remove_host() is called before queue pairs are deleted\n and scsi_remove_host() waits for all outstanding SCSI commands to finish.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit bdb61b9b944d1e5b7cee5a9fe21014363c55b811\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:02:05 2019 -0700\n\n scsi: qla2xxx: Introduce the function qla2xxx_init_sp()\n \n This patch does not change any functionality but makes the next patch\n easier to read.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 6c18a43e3c82b0b67531a1cdec7ba31540fe6424\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:02:04 2019 -0700\n\n scsi: qla2xxx: Enable type checking for the SRB free and done callback functions\n \n Since all pointers passed to the srb_t.done() and srb_t.free() functions\n have type srb_t, change the type of the first argument of these functions\n from void * into struct srb *. This allows the compiler to verify the\n argument types for these functions. This patch does not change any\n functionality.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit aa20e38bf567ff81a7778c769be6f5cea098174b\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:02:03 2019 -0700\n\n scsi: qla2xxx: Rework key encoding in qlt_find_host_by_d_id()\n \n Use the same approach for encoding the destination ID as the approach used\n by qlt_update_vp_map().\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 0f8243e64682b270bad6c58a136a1acd777fcfc3\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:02:02 2019 -0700\n\n scsi: qla2xxx: Set the responder mode if appropriate for ELS pass-through IOCBs\n \n According to the firmware documentation responder mode must be set for ELS\n pass-through IOCBs if a response is expected.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 9933c050fe919ed42408a6e0198e3d7e7d3278a8\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:02:01 2019 -0700\n\n scsi: qla2xxx: Make it explicit that ELS pass-through IOCBs use little endian\n \n According to the firmware documentation the firmware expects all ELS\n pass-through IOCB parameters in little endian format. Make this explicit.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 0597fe601a3a7d103c35b922046251906e0349b3\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:02:00 2019 -0700\n\n scsi: qla2xxx: Check secondary image if reading the primary image fails\n \n This patch fixes several Coverity complaints about reading data that has\n not been initialized.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit c43e7832ee77daadf147876b93f23035b89fa028\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:59 2019 -0700\n\n scsi: qla2xxx: Change the return type of qla24xx_read_flash_data()\n \n This change makes it easier to detect qla24xx_read_flash_data() failures\n and also to handle such failures. This change does not modify the behavior\n of the driver since all callers ignore the qla24xx_read_flash_data()\n return value.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit df95f39ae76474d922d9be9c0260dc263c451b09\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:58 2019 -0700\n\n scsi: qla2xxx: Introduce the be_id_t and le_id_t data types for FC src/dst IDs\n \n Introduce the be_id_t and le_id_t data types for Fibre Channel source and\n destination ID formats supported by the firmware instead of using an\n uint8_t[3] array. Introduce functions for converting from and to the\n port_id_t data types. This patch does not change the behavior of the\n qla2xxx driver but improves source code readability and also allows the\n compiler to verify the endianness of Fibre Channel IDs.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit fb3250974a484882caf10605ebff9ecb23773475\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:57 2019 -0700\n\n scsi: qla2xxx: Complain if a soft reset fails\n \n Failure of a soft reset is a severe failure. Hence report such failures.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 527e9b704c3d46da1cd4061a5ce939456eac8566\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:56 2019 -0700\n\n scsi: qla2xxx: Use memcpy() and strlcpy() instead of strcpy() and strncpy()\n \n This patch makes the string manipulation code easier to verify.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit dc6d6d34e1f674b39d91454fca2b826e190e0351\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:55 2019 -0700\n\n scsi: qla2xxx: Check the PCI info string output buffer size\n \n Pass the output buffer size to the code that generates a PCI info string\n and check the output buffer size while generating a PCI info string.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit fcef08932db7770cc176555ac2f08574b1e43af6\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:54 2019 -0700\n\n scsi: qla2xxx: Complain if waiting for pending commands times out\n \n Such a timeout is a severe issue. Hence complain if waiting for pending\n commands times out. This patch fixes a small bug: it modifies\n qla82xx_chip_reset_cleanup() such that the \"Done waiting\" message is\n reported if qla82xx_chip_reset_cleanup() succeeded instead of if that\n function failed.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit bc97c9484baefe5092120b2b81c3993296ad14ad\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:53 2019 -0700\n\n scsi: qla2xxx: Declare fourth qla2x00_set_model_info() argument const\n \n Make it clear to humans and also to the compiler that the string passed\n as fourth argument is not modified.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit e6803efae5acd109fad9f2f07dab674563441a53\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:52 2019 -0700\n\n scsi: qla2xxx: Always check the qla2x00_wait_for_hba_online() return value\n \n This patch fixes several Coverity complaints about not always checking\n the qla2x00_wait_for_hba_online() return value.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit c4707fccb36412d3a5e58e426071ea2b220aab83\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:51 2019 -0700\n\n scsi: qla2xxx: Suppress multiple Coverity complaint about out-of-bounds accesses\n \n This patch does not change any functionality.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit f91ff36a5731ca41f9009a00bf9f9b8e6c623dae\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:50 2019 -0700\n\n scsi: qla2xxx: Suppress a Coveritiy complaint about integer overflow\n \n This patch fixes the following Coverity complaint:\n \n Unintentional integer overflow (OVERFLOW_BEFORE_WIDEN)\n overflow_before_widen: Potentially overflowing expression dma_eng_num *\n 65536U with type unsigned int (32 bits, unsigned) is evaluated using 32-bit\n arithmetic, and then used in a context that expects an expression of type\n uint64_t (64 bits, unsigned).\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 3a392b4f384a08919c93e5f751e9fffa1ad723e3\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:49 2019 -0700\n\n scsi: qla2xxx: Remove unreachable code from qla83xx_idc_lock()\n \n This was detected by Coverity.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit c29282c65d1cf54daeea63be46243d7f69d72f4d\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:48 2019 -0700\n\n scsi: qla2xxx: Fix qla24xx_process_bidir_cmd()\n \n Set the r??_data_len variables before using these instead of after.\n \n This patch fixes the following Coverity complaint:\n \n const: At condition req_data_len !\u003d rsp_data_len, the value of req_data_len\n must be equal to 0.\n const: At condition req_data_len !\u003d rsp_data_len, the value of rsp_data_len\n must be equal to 0.\n dead_error_condition: The condition req_data_len !\u003d rsp_data_len cannot be\n true.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Fixes: a9b6f722f62d (\"[SCSI] qla2xxx: Implementation of bidirectional.\") # v3.7.\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 282d10565ee9b574c40b1233a7a666034e76aa4c\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:47 2019 -0700\n\n scsi: qla2xxx: Simplify a debug statement\n \n Read the FC port state once instead of twice. This patch fixes the\n following Coverity complaint:\n \n Unchecked return value (CHECKED_RETURN)\n check_return: Calling atomic_read without checking return value (as is\n done elsewhere 80 out of 92 times).\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 258f0752ef363caafb0bce311369d50c87aa0b25\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:46 2019 -0700\n\n scsi: qla2xxx: Remove dead code\n \n Since sess \u003d\u003d NULL before \u0027goto out_term2\u0027 is executed, the code under \u0027if\n (sess)\u0027 cannot be reached. Hence remove that code. This was detected by\n Coverity.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 6f153bcf12f4637673a24ee5393438251b82794c\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:45 2019 -0700\n\n scsi: qla2xxx: Complain if parsing the version string fails\n \n This patch fixes a Coverity complaint about not checking the sscanf()\n return value.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 58e060e11b0b18298208e5d0d1bd9bd98b4238a7\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:44 2019 -0700\n\n scsi: qla2xxx: Complain if a mailbox command times out\n \n This patch fixes the following Coverity complaint:\n \n Unchecked return value (CHECKED_RETURN)\n check_return: Calling wait_for_completion_timeout without checking return\n value (as is done elsewhere 14 out of 17 times).\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit cc74049f35e84b6727c70589750c84e6166963ae\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:43 2019 -0700\n\n scsi: qla2xxx: Use strlcpy() instead of strncpy()\n \n This patch fixes several gcc complaints about string truncation.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 1df627b4f7fa5b28ac10b39869c769373376d8f5\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:42 2019 -0700\n\n scsi: qla2xxx: Do not corrupt vha-\u003eplogi_ack_list\n \n Delete the PLOGIN ACK data structure from the vha-\u003eplogi_ack_list before\n freeing that data structure to avoid that that list gets corrupted.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit b3e9772d9ec5aed081ed06b1431a52945dc37ccf\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:41 2019 -0700\n\n scsi: qla2xxx: Report the firmware status code if a mailbox command fails\n \n It is helpful when debugging this driver to have the firmware status code\n available if a mailbox command fails. Hence report that firmware status\n code.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit ac452b8e79320c9e90c78edf32ba2d42431e4daf\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:40 2019 -0700\n\n scsi: qla2xxx: Fix session lookup in qlt_abort_work()\n \n Pass the correct session ID to find_sess_by_s_id() instead of passing an\n uninitialized variable.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Fixes: 2d70c103fd2a (\"[SCSI] qla2xxx: Add LLD target-mode infrastructure for \u003e\u003d 24xx series\") # v3.5.\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit f1333c48a4ad472683fedae075d2dab6aa8c0273\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:39 2019 -0700\n\n scsi: qla2xxx: Simplify qla24xx_abort_sp_done()\n \n Instead of explicitly checking whether a timeout has occurred, ignore the\n del_timer() return value.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 8b21900b09a55ccee1aaddaabee729484f641997\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:38 2019 -0700\n\n scsi: qla2xxx: Remove two superfluous tests\n \n Since qlt_remove_target() only calls qlt_release() if\n vha-\u003evha_tgt.qla_tgt !\u003d NULL, checking that pointer inside qlt_release()\n is not necessary. This patch fixes the following Coverity complaint:\n \n CID 188348 (#1 of 1): Dereference after null check (FORWARD_NULL)\n var_deref_model: Passing null pointer \u0026vha-\u003evha_tgt.qla_tgt-\u003etgt_list_entry\n to list_del, which dereferences it.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 53be100b36aeca6bbc611edd07ce65fdba8b18ac\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:37 2019 -0700\n\n scsi: qla2xxx: Remove a superfluous pointer check\n \n Checking a pointer after it has been dereferenced is not useful. This was\n detected by Coverity.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 09434912ed4c0b6948fee8b0bb1b058211ef4cac\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:36 2019 -0700\n\n scsi: qla2xxx: Simplify qlt_lport_dump()\n \n Simplify the implementation of this function by using the %phC format\n specifier instead of using explicit for-loops.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 9c17c3b1e0490e436d17b63e8751b2fba030c0f7\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:35 2019 -0700\n\n scsi: qla2xxx: Reduce the number of casts in GID list code\n \n This patch makes the code that parses the GID list easier to read without\n changing the behavior of the code.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 57bf595a6f24370424f625aac123cd12dbff5ffe\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:34 2019 -0700\n\n scsi: qla2xxx: Verify locking assumptions at runtime\n \n Make sure that locking assumptions are verified at runtime if kernel\n debugging is enabled.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 9e75b5e2c8384080842015d26c9272fd55c0d798\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:33 2019 -0700\n\n scsi: qla2xxx: Change data_dsd into an array\n \n This patch does not change any functionality but fixes a Coverity complaint\n about using a scalar as an array.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit bb6efb1ea639ac01184d057afc5c5f94496e348d\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:32 2019 -0700\n\n scsi: qla2xxx: Declare qla_tgt_cmd.cdb const\n \n Make it clear that the CDB is not modified after processing of a SCSI\n command has started.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 6d58ef05fd33f0f48923d8ed0a8a002c268313e2\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:31 2019 -0700\n\n scsi: qla2xxx: Reduce the scope of three local variables in qla2xxx_queuecommand()\n \n This patch makes it clear that the tag, hwq and qpair variables are only\n used in the mq path.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 909c1d1492b634b13a5d960e6803f0d2751509a0\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:30 2019 -0700\n\n scsi: qla2xxx: Change the return type of qla2x00_update_ms_fdmi_iocb() into void\n \n The value returned by this function is not used. Hence change the return\n type of this function into \u0027void\u0027 and remove the return statement.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit c254b52b58b50a8599fefefd8793a64249b15044\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:29 2019 -0700\n\n scsi: qla2xxx: Declare the fourth ql_dump_buffer() argument const\n \n This patch makes it clear to humans and also to the compiler that\n ql_dump_buffer() does not modify the memory the @buf argument points at.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 36645232d9fccc3cea0e91499d1048f27db60f6f\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:28 2019 -0700\n\n scsi: qla2xxx: Remove a superfluous forward declaration\n \n Since qlt_make_local_sess() is defined before it is called, remove the\n forward declaration of that function.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 3f5ee085d37d5e7ddb95e265853a395d289fb4d2\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:27 2019 -0700\n\n scsi: qla2xxx: Remove an include directive from qla_mr.c\n \n There is no bsg code in the qla_mr.c source file. Hence do not include\n the \u003clinux/bsg-lib.h\u003e header file from qla_mr.c.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit ba0cef2532717354d90ea81dc47837d61c98defc\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:26 2019 -0700\n\n scsi: qla2xxx: Include the \u003casm/unaligned.h\u003e header file from qla_dsd.h\n \n Since the put_unaligned_*() macros are used in this header file, include\n the header file that defines these macros.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Fixes: 15b7a68c1d03 (\"scsi: qla2xxx: Introduce the dsd32 and dsd64 data structures\") # v5.2-rc1.\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 0184793df2e860534380a66aa76b8807e71188c2\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:25 2019 -0700\n\n scsi: qla2xxx: Use tabs instead of spaces for indentation\n \n This patch only modifies whitespace.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit c1c7178c26202db0856cb9189ccca043b95d13bc\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:24 2019 -0700\n\n scsi: qla2xxx: Improve Linux kernel coding style conformance\n \n Insert a space where required, surround complex expressions in macros with\n parentheses, use the UL suffix instead of the (unsigned long) cast, do not\n use line continuations when not necessary and do not explicitly initialize\n static variables to zero.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 8dd9593cc07ad7d999bef81b06789ef873a94881\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:23 2019 -0700\n\n scsi: qla2xxx: Really fix qla2xxx_eh_abort()\n \n I\u0027m not sure how this happened but the patch that was intended to fix abort\n handling was incomplete. This patch fixes that patch as follows:\n \n - If aborting the SCSI command failed, wait until the SCSI command\n completes.\n \n - Return SUCCESS instead of FAILED if an abort attempt races with SCSI\n command completion.\n \n - Since qla2xxx_eh_abort() increments the sp reference count by calling\n sp_get(), decrement the sp reference count before returning.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Fixes: 219d27d7147e (\"scsi: qla2xxx: Fix race conditions in the code for aborting SCSI commands\")\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit d2d2b5a5741d317bed1fa38211f1f3b142d8cf7a\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 8 20:01:22 2019 -0700\n\n scsi: qla2xxx: Make qla2x00_abort_srb() again decrease the sp reference count\n \n Since qla2x00_abort_srb() starts with increasing the reference count of\n @sp, decrease that same reference count before returning.\n \n Cc: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Fixes: 219d27d7147e (\"scsi: qla2xxx: Fix race conditions in the code for aborting SCSI commands\") # v5.2.\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Tested-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Reviewed-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit b0eb93cfd516201ccf0e4d36e226cfe1b16cc1fe\nAuthor: Kees Cook \u003ckeescook@chromium.org\u003e\nDate: Thu Aug 8 11:37:45 2019 -0700\n\n lkdtm: Add Control Flow Integrity test\n \n This adds a simple test for forward CFI (indirect function calls) with\n function prototype granularity (as implemented by Clang\u0027s CFI).\n \n Signed-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\n\ncommit ac5656d8a4cdd93cd2c74355ed12e5617817e0e7\nAuthor: Aaron Goidel \u003cacgoide@tycho.nsa.gov\u003e\nDate: Mon Aug 12 11:20:00 2019 -0400\n\n fanotify, inotify, dnotify, security: add security hook for fs notifications\n \n As of now, setting watches on filesystem objects has, at most, applied a\n check for read access to the inode, and in the case of fanotify, requires\n CAP_SYS_ADMIN. No specific security hook or permission check has been\n provided to control the setting of watches. Using any of inotify, dnotify,\n or fanotify, it is possible to observe, not only write-like operations, but\n even read access to a file. Modeling the watch as being merely a read from\n the file is insufficient for the needs of SELinux. This is due to the fact\n that read access should not necessarily imply access to information about\n when another process reads from a file. Furthermore, fanotify watches grant\n more power to an application in the form of permission events. While\n notification events are solely, unidirectional (i.e. they only pass\n information to the receiving application), permission events are blocking.\n Permission events make a request to the receiving application which will\n then reply with a decision as to whether or not that action may be\n completed. This causes the issue of the watching application having the\n ability to exercise control over the triggering process. Without drawing a\n distinction within the permission check, the ability to read would imply\n the greater ability to control an application. Additionally, mount and\n superblock watches apply to all files within the same mount or superblock.\n Read access to one file should not necessarily imply the ability to watch\n all files accessed within a given mount or superblock.\n \n In order to solve these issues, a new LSM hook is implemented and has been\n placed within the system calls for marking filesystem objects with inotify,\n fanotify, and dnotify watches. These calls to the hook are placed at the\n point at which the target path has been resolved and are provided with the\n path struct, the mask of requested notification events, and the type of\n object on which the mark is being set (inode, superblock, or mount). The\n mask and obj_type have already been translated into common FS_* values\n shared by the entirety of the fs notification infrastructure. The path\n struct is passed rather than just the inode so that the mount is available,\n particularly for mount watches. This also allows for use of the hook by\n pathname-based security modules. However, since the hook is intended for\n use even by inode based security modules, it is not placed under the\n CONFIG_SECURITY_PATH conditional. Otherwise, the inode-based security\n modules would need to enable all of the path hooks, even though they do not\n use any of them.\n \n This only provides a hook at the point of setting a watch, and presumes\n that permission to set a particular watch implies the ability to receive\n all notification about that object which match the mask. This is all that\n is required for SELinux. If other security modules require additional hooks\n or infrastructure to control delivery of notification, these can be added\n by them. It does not make sense for us to propose hooks for which we have\n no implementation. The understanding that all notifications received by the\n requesting application are all strictly of a type for which the application\n has been granted permission shows that this implementation is sufficient in\n its coverage.\n \n Security modules wishing to provide complete control over fanotify must\n also implement a security_file_open hook that validates that the access\n requested by the watching application is authorized. Fanotify has the issue\n that it returns a file descriptor with the file mode specified during\n fanotify_init() to the watching process on event. This is already covered\n by the LSM security_file_open hook if the security module implements\n checking of the requested file mode there. Otherwise, a watching process\n can obtain escalated access to a file for which it has not been authorized.\n \n The selinux_path_notify hook implementation works by adding five new file\n permissions: watch, watch_mount, watch_sb, watch_reads, and watch_with_perm\n (descriptions about which will follow), and one new filesystem permission:\n watch (which is applied to superblock checks). The hook then decides which\n subset of these permissions must be held by the requesting application\n based on the contents of the provided mask and the obj_type. The\n selinux_file_open hook already checks the requested file mode and therefore\n ensures that a watching process cannot escalate its access through\n fanotify.\n \n The watch, watch_mount, and watch_sb permissions are the baseline\n permissions for setting a watch on an object and each are a requirement for\n any watch to be set on a file, mount, or superblock respectively. It should\n be noted that having either of the other two permissions (watch_reads and\n watch_with_perm) does not imply the watch, watch_mount, or watch_sb\n permission. Superblock watches further require the filesystem watch\n permission to the superblock. As there is no labeled object in view for\n mounts, there is no specific check for mount watches beyond watch_mount to\n the inode. Such a check could be added in the future, if a suitable labeled\n object existed representing the mount.\n \n The watch_reads permission is required to receive notifications from\n read-exclusive events on filesystem objects. These events include accessing\n a file for the purpose of reading and closing a file which has been opened\n read-only. This distinction has been drawn in order to provide a direct\n indication in the policy for this otherwise not obvious capability. Read\n access to a file should not necessarily imply the ability to observe read\n events on a file.\n \n Finally, watch_with_perm only applies to fanotify masks since it is the\n only way to set a mask which allows for the blocking, permission event.\n This permission is needed for any watch which is of this type. Though\n fanotify requires CAP_SYS_ADMIN, this is insufficient as it gives implicit\n trust to root, which we do not do, and does not support least privilege.\n \n Signed-off-by: Aaron Goidel \u003cacgoide@tycho.nsa.gov\u003e\n Acked-by: Casey Schaufler \u003ccasey@schaufler-ca.com\u003e\n Acked-by: Jan Kara \u003cjack@suse.cz\u003e\n Signed-off-by: Paul Moore \u003cpaul@paul-moore.com\u003e\n\ncommit 4514fe8cd96f8147c18c05c7c32271d95d6ab0c4\nAuthor: Jonathan Neuschäfer \u003cj.neuschaefer@gmx.net\u003e\nDate: Thu Aug 8 18:48:09 2019 +0200\n\n Documentation/arm/samsung-s3c24xx: Remove stray U+FEFF character to fix title\n \n It seems a UTF-8 byte order mark (the least useful kind of BOM...) snuck\n into the file and broke Sphinx\u0027s detection of the title line.\n \n Besides making arm/samsung-s3c24xx/index.html look a little better, this\n patch also confines the non-index pages in arm/samsung-s3c24xx to their\n own table of contents.\n \n Signed-off-by: Jonathan Neuschäfer \u003cj.neuschaefer@gmx.net\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 13afbbf49a34c21085dd6c7a48340ed5d2309729\nAuthor: Jonathan Neuschäfer \u003cj.neuschaefer@gmx.net\u003e\nDate: Thu Aug 8 18:58:56 2019 +0200\n\n Documentation/arm/sa1100/assabet: Fix \u0027make assabet_defconfig\u0027 command\n \n \"make assabet_config\" doesn\u0027t work.\n \n Signed-off-by: Jonathan Neuschäfer \u003cj.neuschaefer@gmx.net\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 198266af3ab934847c840b2442ed66b255723e8a\nAuthor: Jonathan Neuschäfer \u003cj.neuschaefer@gmx.net\u003e\nDate: Thu Aug 8 18:58:55 2019 +0200\n\n Documentation/arm/sa1100: Remove some obsolete documentation\n \n The support for the following boards, among others, was removed in 2004\n with commit \"[ARM] Remove broken SA1100 machine support.\":\n \n - ADS Bitsy\n - Brutus\n - Freebird\n - ADS GraphicsClient Plus\n - ADS GraphicsMaster\n - Höft \u0026 Wessel Webpanel\n - Compaq Itsy\n - nanoEngine\n - Pangolin\n - PLEB\n - Yopy\n \n Tifon support has been removed in 2.4.3.3.\n \n Signed-off-by: Jonathan Neuschäfer \u003cj.neuschaefer@gmx.net\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 6a241a11f9257e34f64db5e1769c77af72593c20\nAuthor: Alex Shi \u003calex.shi@linux.alibaba.com\u003e\nDate: Fri Aug 9 20:04:47 2019 +0800\n\n docs/zh_CN: update Chinese howto.rst for latexdocs making\n \n Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e found a reference\n error in Chinese howto.rst. and further more there more infos of\n latexdocs/epubdocs format doc making in English howto.rst.\n \n So I update this part according to latest howto.rst and settled\n the correct reference.\n \n Signed-off-by: Alex Shi \u003calex.shi@linux.alibaba.com\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 03d36521f5d22e4bd93246a53fd493f275924593\nAuthor: Sheriff Esseson \u003csheriffesseson@gmail.com\u003e\nDate: Fri Aug 9 14:23:49 2019 +0100\n\n Documentation: virt: Fix broken reference to virt tree\u0027s index\n \n Fix broken reference to virt/index.rst.\n \n Fixes: 2f5947dfcaec (\"Documentation: move Documentation/virtual to Documentation/virt\")\n Signed-off-by: Sheriff Esseson \u003csheriffesseson@gmail.com\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 3b2cb439a63fd5bed75bf9e023f4a8ec6623b3c0\nAuthor: Marco Villegas \u003cgit@marvil07.net\u003e\nDate: Fri Aug 9 18:29:07 2019 -0500\n\n docs: Fix typo on pull requests guide\n \n Signed-off-by: Marco Villegas \u003cgit@marvil07.net\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit b0a3fa44659ccee87215600b4235982bc0f3e828\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Wed Aug 7 16:23:10 2019 +0800\n\n rtc: mxc: use spin_lock_irqsave instead of spin_lock_irq in IRQ context\n \n As spin_unlock_irq will enable interrupts.\n mxc_rtc_irq_enable is called from interrupt handler mxc_rtc_interrupt.\n Interrupts are enabled in interrupt handler.\n Use spin_lock_irqsave/spin_unlock_irqrestore instead of spin_(un)lock_irq\n in IRQ context to avoid this.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190807082310.10135-1-huangfq.daxian@gmail.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 41a8e19f47dfe9154c56b35842700ad38a7c33e0\nAuthor: Matti Vaittinen \u003cmatti.vaittinen@fi.rohmeurope.com\u003e\nDate: Mon Aug 12 09:36:11 2019 +0300\n\n rtc: bd70528: fix driver dependencies\n \n With CONFIG_BD70528_WATCHDOG\u003dm, a built-in rtc driver cannot call\n into the low-level functions that are part of the watchdog module:\n \n drivers/rtc/rtc-bd70528.o: In function `bd70528_set_time\u0027:\n rtc-bd70528.c:(.text+0x22c): undefined reference to `bd70528_wdt_lock\u0027\n rtc-bd70528.c:(.text+0x2a8): undefined reference to `bd70528_wdt_unlock\u0027\n drivers/rtc/rtc-bd70528.o: In function `bd70528_set_rtc_based_timers\u0027:\n rtc-bd70528.c:(.text+0x50c): undefined reference to `bd70528_wdt_set\u0027\n \n Add a Kconfig dependency which forces RTC to be a module if watchdog is a\n module. If watchdog is not compiled at all the stub functions for watchdog\n control are used. compiling the RTC without watchdog is fine.\n \n Fixes: 32a4a4ebf768 (\"rtc: bd70528: Initial support for ROHM bd70528 RTC\")\n Suggested-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Matti Vaittinen \u003cmatti.vaittinen@fi.rohmeurope.com\u003e\n Link: https://lore.kernel.org/r/84462e01e43d39024948a3bdd24087ff87dc2255.1565591387.git.matti.vaittinen@fi.rohmeurope.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 15e2544ed38a1e39b702277bdcd2bb1275e78925\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Aug 12 19:06:31 2019 +0300\n\n kernel-doc: Allow anonymous enum\n \n In C is a valid construction to have an anonymous enumerator.\n \n Though we have now:\n \n drivers/pinctrl/intel/pinctrl-intel.c:240: error: Cannot parse enum!\n \n Support it in the kernel-doc script.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit cb3cab06142eaa157bf5666c4e514ef087bf4323\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 22:27:42 2019 +0200\n\n rtc: remove w90x900/nuc900 driver\n \n The ARM w90x900 platform is getting removed, so this driver is obsolete.\n \n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Link: https://lore.kernel.org/r/20190809202749.742267-15-arnd@arndb.de\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 82bf829b692d6bc9c44cd43ddd75a9b04f1bfa8d\nAuthor: Jonathan Neuschäfer \u003cj.neuschaefer@gmx.net\u003e\nDate: Mon Aug 12 18:07:05 2019 +0200\n\n Documentation: sphinx: Don\u0027t parse socket() as identifier reference\n \n With the introduction of Documentation/sphinx/automarkup.py, socket() is\n parsed as a reference to the in-kernel definition of socket. Sphinx then\n decides that struct socket is a good match, which is usually not\n intended, when the syscall is meant instead. This was observed in\n Documentation/networking/af_xdp.rst.\n \n Prevent socket() from being misinterpreted by adding it to the Skipfuncs\n list in automarkup.py.\n \n Signed-off-by: Jonathan Neuschäfer \u003cj.neuschaefer@gmx.net\u003e\n Reviewed-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 11fec009d97e5bd2329ef7d52d71e9f6763f1048\nAuthor: Jonathan Neuschäfer \u003cj.neuschaefer@gmx.net\u003e\nDate: Mon Aug 12 18:07:04 2019 +0200\n\n Documentation: sphinx: Add missing comma to list of strings\n \n In Python, like in C, when a comma is omitted in a list of strings, the\n two strings around the missing comma are concatenated.\n \n Cc: stable@vger.kernel.org # v5.2 only\n Signed-off-by: Jonathan Neuschäfer \u003cj.neuschaefer@gmx.net\u003e\n Reviewed-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit abab1cd6c617762adade199c33034d6d744bd28e\nAuthor: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\nDate: Sun Jul 28 23:04:02 2019 +0200\n\n dt-bindings: add vendor prefix \"acme\" for \"Acme Systems srl\"\n \n Add prefix for Acme Systems srl\n https://www.acmesystems.it\n \n Signed-off-by: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\n Link: https://lore.kernel.org/r/20190728210403.2730-2-uwe@kleine-koenig.org\n Acked-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Ludovic Desroches \u003cludovic.desroches@microchip.com\u003e\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 328e3e369bcee3375a8f48d9ae68daf173fc8453\nAuthor: Stefan Wahren \u003cwahrenst@gmx.net\u003e\nDate: Tue Jul 16 20:13:13 2019 +0200\n\n ARM: dts: bcm283x: Define memory at board level\n \n Now with the varity of several RPi boards, the memory should be defined\n at board level. This step gives us the chance to fix the memory size\n of the RPi 1 B+, Zero (incl. W) and Compute Module 1.\n \n Signed-off-by: Stefan Wahren \u003cwahrenst@gmx.net\u003e\n Acked-by: Eric Anholt \u003ceric@anholt.net\u003e\n\ncommit e7774049ff255c8ba72bcee9c7ab307a95e8d7bc\nAuthor: Stefan Wahren \u003cwahrenst@gmx.net\u003e\nDate: Sun Jul 14 17:55:00 2019 +0200\n\n ARM: dts: bcm283x: Define MMC interfaces at board level\n \n Starting with RPi 4 this is the first board, which doesn\u0027t use sdhost\n as default SD interface. So the MMC interfaces should be defined finally at\n board level. Since all boards using sdhci already does this, we can drop the\n pinctrl part from bcm2835-rpi.dtsi.\n \n Signed-off-by: Stefan Wahren \u003cwahrenst@gmx.net\u003e\n Acked-by: Eric Anholt \u003ceric@anholt.net\u003e\n\ncommit 6b14dd7267126931e9a95c68a442e8f2dabdc3c4\nAuthor: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\nDate: Mon Aug 12 19:50:04 2019 +0200\n\n ARM: dts: meson8b: odroidc1: use the MAC address stored in the eFuse\n \n Odroid-C1 uses the MAC address stored in eFuse at offset 0x1b4 (which is\n defined as a \"standard\" offset for all Meson8 and Meson8b boards, but\n testing shows that MXQ doesn\u0027t have the eFuse values programmed and\n EC-100 stores it\u0027s MAC address in eMMC).\n \n Add the nvmem cell which points to the MAC address and asssign it to the\n Ethernet controller as \"mac-address\".\n As result of this the MAC address which is stored in the eFuse is now\n assigned to the Ethernet controller and consistent across reboots.\n \n Signed-off-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit ddc73f32ba9529e269a043a45979bd95eb594361\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Aug 4 08:55:51 2019 +0200\n\n drm/msm: drop use of drmP.h\n \n Drop the deprecated drmP.h header file, and trim msm_drv.h\n to the relevant include files.\n \n This resulted in a suprisingly many edits as many files relied\n on headers included via msm_drv.h.\n But msm_drv.h is not supposed to carry include files it do not need, so\n the individual files have to include what extra they needs.\n \n v2:\n - Rebased on top of https://gitlab.freedesktop.org/drm/msm.git msm-next\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Rob Clark \u003crobdclark@gmail.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Jordan Crouse \u003cjcrouse@codeaurora.org\u003e\n Cc: Jeykumar Sankaran \u003cjsanka@codeaurora.org\u003e\n Cc: Bruce Wang \u003cbzwang@chromium.org\u003e\n Cc: Shayenne Moura \u003cshayenneluzmoura@gmail.com\u003e\n Cc: Mamta Shukla \u003cmamtashukla555@gmail.com\u003e\n Cc: Jonathan Marek \u003cjonathan@marek.ca\u003e\n Cc: Carsten Behling \u003ccarsten.behling@googlemail.com\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Paul Kocialkowski \u003cpaul.kocialkowski@bootlin.com\u003e\n Cc: Sibi Sankar \u003csibis@codeaurora.org\u003e\n Cc: Todor Tomov \u003ctodor.tomov@linaro.org\u003e\n Cc: linux-arm-msm@vger.kernel.org\n Cc: freedreno@lists.freedesktop.org\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190804065551.GA5211@ravnborg.org\n\ncommit d14734a04a8aea793a5534de39a21ad59b324a11\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Mon Jul 29 15:26:22 2019 +0200\n\n arm64: dts: meson-g12b-odroid-n2: enable DVFS\n \n Enable DVFS for the Odroid-N2 by setting the clock, OPP and supply\n for each cores of each CPU clusters.\n \n The first cluster uses the \"VDDCPU_B\" power supply, and the second\n cluster uses the \"VDDCPU_A\" power supply.\n \n Each power supply can achieve 0.73V to 1.01V using 2 distinct PWM\n outputs clocked at 800KHz with an inverse duty-cycle.\n \n DVFS has been tested by running the arm64 cpuburn at [1] and cycling\n between all the possible cpufreq translations of each cluster and\n checking the final frequency using the clock-measurer, script at [2].\n \n [1] https://github.com/ssvb/cpuburn-arm/blob/master/cpuburn-a53.S\n [2] https://gist.github.com/superna9999/d4de964dbc0f84b7d527e1df2ddea25f\n \n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Tested-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 1c2b93902faf8654d13eac8c86876365d6f0da6f\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Tue Jul 2 12:49:36 2019 -0600\n\n drm/panfrost: Bump driver version to 1.1\n \n Increment the driver version to expose the new BO allocation flags.\n \n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Cc: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Cc: Steven Price \u003csteven.price@arm.com\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808222200.13176-10-robh@kernel.org\n\ncommit 187d2929206e6b098312c174ea873e4cedf5420d\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Fri Jul 26 16:09:43 2019 -0600\n\n drm/panfrost: Add support for GPU heap allocations\n \n The midgard/bifrost GPUs need to allocate GPU heap memory which is\n allocated on GPU page faults and not pinned in memory. The vendor driver\n calls this functionality GROW_ON_GPF.\n \n This implementation assumes that BOs allocated with the\n PANFROST_BO_NOEXEC flag are never mmapped or exported. Both of those may\n actually work, but I\u0027m unsure if there\u0027s some interaction there. It\n would cause the whole object to be pinned in memory which would defeat\n the point of this.\n \n On faults, we map in 2MB at a time in order to utilize huge pages (if\n enabled). Currently, once we\u0027ve mapped pages in, they are only unmapped\n if the BO is freed. Once we add shrinker support, we can unmap pages\n with the shrinker.\n \n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Cc: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808222200.13176-9-robh@kernel.org\n\ncommit b31bdd1389fc765c07ab3d5b341092cb16807d29\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Fri Jul 26 16:06:57 2019 -0600\n\n drm/panfrost: Convert MMU IRQ handler to threaded handler\n \n In preparation to handle mapping of page faults, we need the MMU handler\n to be threaded as code paths take a mutex.\n \n As the IRQ may be shared, we can\u0027t use the default handler and must\n disable the MMU interrupts locally.\n \n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Cc: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808222200.13176-8-robh@kernel.org\n\ncommit c6d29c66e582f9a3ae27d08a488aa89232545a39\nAuthor: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\nDate: Mon Aug 12 09:48:57 2019 +0200\n\n arm64: dts: meson-g12b-khadas-vim3: add initial device-tree\n \n The Khadas VIM3 uses the Amlogic S922X or A311S SoC, both based on the\n Amlogic G12B SoC family, on a board with the same form factor as the\n VIM/VIM2 models. It ships in two variants; basic and\n pro which differ in RAM and eMMC size:\n \n - 2GB (basic) or 4GB (pro) LPDDR4 RAM\n - 16GB (basic) or 32GB (pro) eMMC 5.1 storage\n - 16MB SPI flash\n - 10/100/1000 Base-T Ethernet\n - AP6398S Wireless (802.11 a/b/g/n/ac, BT5.0)\n - HDMI 2.1 video\n - 1x USB 2.0 + 1x USB 3.0 ports\n - 1x USB-C (power) with USB 2.0 OTG\n - 3x LED\u0027s (1x red, 1x blue, 1x white)\n - 3x buttons (power, function, reset)\n - IR receiver\n - M2 socket with PCIe, USB, ADC \u0026 I2C\n - 40pin GPIO Header\n - 1x micro SD card slot\n \n A common meson-g12b-khadas-vim3.dtsi is added to support both S922X and\n A311D SoCs supported by two variants of the board.\n \n Signed-off-by: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 73e467f60acdabd480d1b377a623ba13db0e5dd2\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Thu Aug 8 14:30:39 2019 -0600\n\n drm/panfrost: Consolidate reset handling\n \n Runtime PM resume and job timeouts both call the same sequence of\n functions, so consolidate them to a common function. This will make\n changing the reset related code easier. The MMU also needs some\n re-initialization on reset, so rework its call. In the process, we\n hide the address space details within the MMU code in preparation to\n support multiple address spaces.\n \n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Cc: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808222200.13176-7-robh@kernel.org\n\ncommit 203270c025be02d52feca729d9193860317c2e82\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Thu Jul 11 15:56:14 2019 -0600\n\n drm/panfrost: Add a no execute flag for BO allocations\n \n Executable buffers have an alignment restriction that they can\u0027t cross\n 16MB boundary as the GPU program counter is 24-bits. This restriction is\n currently not handled and we just get lucky. As current userspace\n assumes all BOs are executable, that has to remain the default. So add a\n new PANFROST_BO_NOEXEC flag to allow userspace to indicate which BOs are\n not executable.\n \n There is also a restriction that executable buffers cannot start or end\n on a 4GB boundary. This is mostly avoided as there is only 4GB of space\n currently and the beginning is already blocked out for NULL ptr\n detection. Add support to handle this restriction fully regardless of\n the current constraints.\n \n For existing userspace, all created BOs remain executable, but the GPU\n VA alignment will be increased to the size of the BO. This shouldn\u0027t\n matter as there is plenty of GPU VA space.\n \n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Cc: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808222200.13176-6-robh@kernel.org\n\ncommit eecbb3cdcccac24e1e67e92fec24f9035fe19784\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Mon Jul 1 17:05:43 2019 -0600\n\n drm/panfrost: Split panfrost_mmu_map SG list mapping to its own function\n \n In preparation to create partial GPU mappings of BOs on page faults,\n split out the SG list handling of panfrost_mmu_map().\n \n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Cc: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Reviewed: Steven Price \u003csteven.price@arm.com\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808222200.13176-5-robh@kernel.org\n\ncommit a5efb4c9a562aa1c2ffa27b0477ae7b20c309344\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Mon Jul 1 17:03:30 2019 -0600\n\n drm/panfrost: Restructure the GEM object creation\n \n Setting the GPU VA when creating the GEM object doesn\u0027t allow for any\n conditional adjustments to the mapping. In preparation to support\n adjusting the mapping and per FD address spaces, restructure the GEM\n object creation to map and unmap the GEM object in the GEM object .open()\n and .close() hooks.\n \n While panfrost_gem_free_object() and panfrost_gem_prime_import_sg_table()\n are not really needed after this commit, keep them as we\u0027ll need them in\n subsequent commits.\n \n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Cc: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808222200.13176-4-robh@kernel.org\n\ncommit 3bf5189d93d04519e93e16bf3aa60589e136c666\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Fri Jul 19 08:30:12 2019 -0600\n\n drm/shmem: Put pages independent of a SG table being set\n \n If a driver does its own management of pages, the shmem helper object\u0027s\n pages array could be allocated when a SG table is not. There\u0027s not\n really any good reason to tie putting pages with having a SG table when\n freeing the object, so just put pages if the pages array is populated.\n \n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808222200.13176-3-robh@kernel.org\n\ncommit 930a402485b205214231f1141632afdca8ff89cb\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Fri Jul 19 08:28:51 2019 -0600\n\n drm/gem: Allow sparsely populated page arrays in drm_gem_put_pages\n \n Panfrost has a need for pages allocated on demand via GPU page faults.\n When releasing the pages, the only thing preventing using\n drm_gem_put_pages() is needing to skip over unpopulated pages, so allow\n for skipping over NULL struct page pointers.\n \n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel@lists.freedesktop.org\n Reviewed-by: Steven Price \u003csteven.price@arm.com\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808222200.13176-2-robh@kernel.org\n\ncommit 74d5f3d06f707eb5f7e1908ad88954bde02000ce\nAuthor: Igor Lubashev \u003cilubashe@akamai.com\u003e\nDate: Wed Aug 7 10:44:14 2019 -0400\n\n tools build: Add capability-related feature detection\n \n Add utilities to help checking capabilities of the running procss. Make\n perf link with libcap, if it is available. If no libcap-dev[el], assume\n no capabilities.\n \n Committer testing:\n \n $ make O\u003d/tmp/build/perf -C tools/perf install-bin\n make: Entering directory \u0027/home/acme/git/perf/tools/perf\u0027\n BUILD: Doing \u0027make -j8\u0027 parallel build\n \n Auto-detecting system features:\n \u003cSNIP\u003e\n ... libbfd: [ on ]\n ... libcap: [ OFF ]\n ... libelf: [ on ]\n \u003cSNIP\u003e\n Makefile.config:833: No libcap found, disables capability support, please install libcap-devel/libcap-dev\n \u003cSNIP\u003e\n $ grep libcap /tmp/build/perf/FEATURE-DUMP\n feature-libcap\u003d0\n $ cat /tmp/build/perf/feature/test-libcap.make.output\n test-libcap.c:2:10: fatal error: sys/capability.h: No such file or directory\n 2 | #include \u003csys/capability.h\u003e\n | ^~~~~~~~~~~~~~~~~~\n compilation terminated.\n $\n \n Now install libcap-devel and try again:\n \n $ make O\u003d/tmp/build/perf -C tools/perf install-bin\n make: Entering directory \u0027/home/acme/git/perf/tools/perf\u0027\n BUILD: Doing \u0027make -j8\u0027 parallel build\n Warning: Kernel ABI header at \u0027tools/include/linux/bits.h\u0027 differs from latest version at \u0027include/linux/bits.h\u0027\n diff -u tools/include/linux/bits.h include/linux/bits.h\n Warning: Kernel ABI header at \u0027tools/arch/x86/include/asm/cpufeatures.h\u0027 differs from latest version at \u0027arch/x86/include/asm/cpufeatures.h\u0027\n diff -u tools/arch/x86/include/asm/cpufeatures.h arch/x86/include/asm/cpufeatures.h\n \n Auto-detecting system features:\n \u003cSNIP\u003e\n ... libbfd: [ on ]\n ... libcap: [ on ]\n ... libelf: [ on ]\n \u003cSNIP\u003e\u003e\n CC /tmp/build/perf/jvmti/libjvmti.o\n \u003cSNIP\u003e\u003e\n $ grep libcap /tmp/build/perf/FEATURE-DUMP\n feature-libcap\u003d1\n $ cat /tmp/build/perf/feature/test-libcap.make.output\n $ ldd /tmp/build/perf/feature/test-libcap.make.bin\n ldd: /tmp/build/perf/feature/test-libcap.make.bin: No such file or directory\n $ ldd /tmp/build/perf/feature/test-libcap.bin\n linux-vdso.so.1 (0x00007ffc35bfe000)\n libcap.so.2 \u003d\u003e /lib64/libcap.so.2 (0x00007ff9c62ff000)\n libc.so.6 \u003d\u003e /lib64/libc.so.6 (0x00007ff9c6139000)\n /lib64/ld-linux-x86-64.so.2 (0x00007ff9c6326000)\n $\n \n Signed-off-by: Igor Lubashev \u003cilubashe@akamai.com\u003e\n Acked-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Tested-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: James Morris \u003cjmorris@namei.org\u003e\n Cc: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Suzuki Poulouse \u003csuzuki.poulose@arm.com\u003e\n [ split from a larger patch ]\n Link: http://lkml.kernel.org/r/8a1e76cf5c7c9796d0d4d240fbaa85305298aafa.1565188228.git.ilubashe@akamai.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit ac9fd659ef8cac9a71f0ef943ef6da04f1939775\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Sun Aug 11 16:37:25 2019 +0200\n\n drm: gm12u320: Add -ENODEV to list of errors to ignore\n \n Add -ENODEV to the list of usb-transfer errors which we ignore to\n avoid logging Frame update errors when the device gets unplugged.\n \n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190811143725.5951-2-hdegoede@redhat.com\n\ncommit 9b61db1aed5c62180a374ab4dd774d1625bdd8d4\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Sun Aug 11 16:37:24 2019 +0200\n\n drm: gm12u320: Do not take a mutex from a wait_event condition\n \n I made the condition of the wait_event_timeout call in\n gm12u320_fb_update_work a helper which takes a mutex to make sure\n that any writes to fb_update.run or fb_update.fb from other CPU cores\n are seen before the check is done.\n \n This is not necessary as the wait_event helpers contain the necessary\n barriers for this themselves.\n \n More over it is harmfull since by the time the check is done the task\n is no longer in the TASK_RUNNING state and calling mutex_lock while not\n in task-running is not allowed, leading to this warning when the kernel\n is build with some extra locking checks enabled:\n \n [11947.450011] do not call blocking ops when !TASK_RUNNING; state\u003d2 set at\n [\u003c00000000e4306de6\u003e] prepare_to_wait_event+0x61/0x190\n \n This commit fixes this by dropping the helper and simply directly\n checking the condition (without unnecessary locking) in the\n wait_event_timeout call.\n \n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190811143725.5951-1-hdegoede@redhat.com\n\ncommit 4abfa2e4e74f22fff67d561d01f8e080b5b547be\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Tue Jul 30 15:38:57 2019 +0200\n\n drm: gm12u320: Use DRM_DEV_ERROR everywhere\n \n Previously the driver was using a mix of DRM_ERROR and dev_err, be\n consisent and use DRM_DEV_ERROR everywhere instead.\n \n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Suggested-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730133857.30778-2-hdegoede@redhat.com\n\ncommit 8515090ce51441c29feb51e0634eb9c5a857469f\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Tue Jul 30 15:38:56 2019 +0200\n\n drm: gm12u320: Some minor cleanups\n \n 3 small cleanups:\n \n 1) Drop unused DRIVER_PATCHLEVEL\n 2) We do not set mode_config.preferred_depth, so instead of passing the\n unset mode_config.preferred_depth to drm_fbdev_generic_setup\n simply pass 0\n 3) Use __maybe_unused instead of #ifdef CONFIG_PM around the suspend /\n resume functions\n \n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Suggested-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Suggested-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730133857.30778-1-hdegoede@redhat.com\n\ncommit 40d81772dac45643cecc7add0e95356072265754\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 9 16:44:34 2019 -0300\n\n perf top: Collapse and resort all evsels in a group\n \n And link them, i.e. find the hist entries in the non-leader events and\n link them to the ones in the leader.\n \n This should be the same thing already done for the \u0027perf report\u0027 case,\n but now we do it periodically.\n \n With this in place we get percentages in from the second overhead column\n on, not just on the first (the leader).\n \n Try it using:\n \n perf top --stdio -e \u0027{cycles,instructions}\u0027\n \n You should see something like:\n \n PerfTop: 20776 irqs/sec kernel:68.7% exact: 0.0% lost: 0/0 drop: 0/0 [cycles], (all, 8 CPUs)\n ---------------------------------------------------------------------------------------------------\n \n 4.44% 0.44% [kernel] [k] do_syscall_64\n 2.27% 0.17% [kernel] [k] entry_SYSCALL_64\n 1.73% 0.27% [kernel] [k] syscall_return_via_sysret\n 1.60% 0.91% [kernel] [k] _raw_spin_lock_irqsave\n 1.45% 3.53% libglib-2.0.so.0.6000.4 [.] g_string_insert_unichar\n 1.39% 0.21% [kernel] [k] copy_user_enhanced_fast_string\n 1.26% 1.15% [kernel] [k] psi_task_change\n 1.16% 0.14% libpixman-1.so.0.38.0 [.] 0x000000000006f403\n 1.00% 0.32% [kernel] [k] __sched_text_start\n 0.97% 2.11% [kernel] [k] n_tty_write\n 0.96% 0.04% [kernel] [k] queued_spin_lock_slowpath\n 0.93% 0.88% [kernel] [k] menu_select\n 0.87% 0.14% [kernel] [k] try_to_wake_up\n 0.77% 0.10% libpixman-1.so.0.38.0 [.] 0x000000000006f40b\n 0.73% 0.09% libpixman-1.so.0.38.0 [.] 0x000000000006f413\n 0.69% 0.48% libc-2.29.so [.] __memmove_avx_unaligned_erms\n 0.68% 0.29% [kernel] [k] _raw_spin_lock_irq\n 0.61% 0.04% libpixman-1.so.0.38.0 [.] 0x000000000006f423\n 0.60% 0.37% [kernel] [k] native_sched_clock\n 0.57% 0.23% [kernel] [k] do_idle\n 0.57% 0.23% [kernel] [k] __fget\n 0.56% 0.30% [kernel] [k] __switch_to_asm\n 0.56% 0.00% libc-2.29.so [.] __memset_avx2_erms\n 0.52% 0.32% [kernel] [k] _raw_spin_lock\n 0.49% 0.24% [kernel] [k] n_tty_poll\n 0.49% 0.54% libglib-2.0.so.0.6000.4 [.] g_mutex_lock\n 0.48% 0.62% [kernel] [k] _raw_spin_unlock_irqrestore\n 0.47% 0.27% [kernel] [k] __switch_to\n 0.47% 0.25% [kernel] [k] pick_next_task_fair\n 0.45% 0.17% [kernel] [k] filldir64\n 0.40% 0.16% [kernel] [k] update_rq_clock\n 0.39% 0.19% [kernel] [k] enqueue_task_fair\n #\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-uw8cjeifxvjpkjp6x2iil0ar@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 5f8b4d5d237a3e2e35509da4e63769ae5c82c085\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Fri Aug 9 17:56:06 2019 -0300\n\n perf hist: Remove dummy entries when finding real ones.\n \n When he have an event group we have multiple struct hist instances, one\n per evsel, and in each of these hists we may have hist_entries that\n point to the same thing being observed, say a symbol, i.e. if we\u0027re\n looking at instructions and cycles, then we\u0027ll have one hist_entry in\n the \"instructions\" evsel and another in the \"cycles\" evsel.\n \n We need to link those to then show one column for each. When we\u0027re\n looking at some other pair of events, say instructions and cache misses,\n we may have just the \"instructions\" hist entry and not one for \"cache\n misses\", as instructions not necessarily generate cache misses, as the\n logic expects one hist_entry per evsel, we end up adding \"dummy\"\n hist_entries.\n \n This is enough for \u0027perf report\u0027, that does this matching operation\n (hists__match()) just once after processing all events, but for \u0027perf\n top\u0027, we do this at each refresh, so we may finally find events matching\n and then we need to trow away the dummies and link with the real events.\n \n So if we find a match, traverse the link of matches and trow away\n dummies for that hists.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-dwvtjqqifsbsczeb35q6mqkk@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 3e70008a6021fffd2cd1614734603ea970773060\nAuthor: Leo Yan \u003cleo.yan@linaro.org\u003e\nDate: Fri Aug 9 18:47:52 2019 +0800\n\n perf trace: Fix segmentation fault when access syscall info on arm64\n \n \u0027perf trace\u0027 reports the segmentation fault as below on Arm64:\n \n # perf trace -e string -e augmented_raw_syscalls.c\n LLVM: dumping tools/perf/examples/bpf/augmented_raw_syscalls.o\n perf: Segmentation fault\n Obtained 12 stack frames.\n perf(sighandler_dump_stack+0x47) [0xaaaaac96ac87]\n linux-vdso.so.1(+0x5b7) [0xffffadbeb5b7]\n /lib/aarch64-linux-gnu/libc.so.6(strlen+0x10) [0xfffface7d5d0]\n /lib/aarch64-linux-gnu/libc.so.6(_IO_vfprintf+0x1ac7) [0xfffface49f97]\n /lib/aarch64-linux-gnu/libc.so.6(__vsnprintf_chk+0xc7) [0xffffacedfbe7]\n perf(scnprintf+0x97) [0xaaaaac9ca3ff]\n perf(+0x997bb) [0xaaaaac8e37bb]\n perf(cmd_trace+0x28e7) [0xaaaaac8ec09f]\n perf(+0xd4a13) [0xaaaaac91ea13]\n perf(main+0x62f) [0xaaaaac8a147f]\n /lib/aarch64-linux-gnu/libc.so.6(__libc_start_main+0xe3) [0xfffface22d23]\n perf(+0x57723) [0xaaaaac8a1723]\n Segmentation fault\n \n This issue is introduced by commit 30a910d7d3e0 (\"perf trace:\n Preallocate the syscall table\"), it allocates trace-\u003esyscalls.table[]\n array and the element count is \u0027trace-\u003esctbl-\u003esyscalls.nr_entries\u0027; but\n on Arm64, the system call number is not continuously used; e.g. the\n syscall maximum id is 436 but the real entries is only 281.\n \n So the table is allocated with \u0027nr_entries\u0027 as the element count, but it\n accesses the table with the syscall id, which might be out of the bound\n of the array and cause the segmentation fault.\n \n This patch allocates trace-\u003esyscalls.table[] with the element count is\n \u0027trace-\u003esctbl-\u003esyscalls.max_id + 1\u0027, this allows any id to access the\n table without out of the bound.\n \n Signed-off-by: Leo Yan \u003cleo.yan@linaro.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Martin KaFai Lau \u003ckafai@fb.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Song Liu \u003csongliubraving@fb.com\u003e\n Cc: Yonghong Song \u003cyhs@fb.com\u003e\n Fixes: 30a910d7d3e0 (\"perf trace: Preallocate the syscall table\")\n Link: http://lkml.kernel.org/r/20190809104752.27338-1-leo.yan@linaro.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 7d1a5efa20dbfea97cb93b99c67ce5cd5c4a4dbc\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Aug 7 10:45:30 2019 -0300\n\n perf hists: Do not link a pair if already linked\n \n When we have multiple events in a group we link hist_entries in the\n non-leader evsel hists to the one in the leader that points to the same\n sorting criteria, in hists__match().\n \n For \u0027perf report\u0027 we do this just once and then print the results, but\n for \u0027perf top\u0027 we need to look if this was already done in the previous\n refresh of the screen, so check for that and don\u0027t try to link again.\n \n This is part of having \u0027perf top\u0027 using the hists browser for showing\n multiple events in multiple columns.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-iwvb37rgb7upswhruwpcdnhw@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 1205a2719e52b6b52e0f9c0011554419da0377a0\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Aug 6 11:20:42 2019 -0300\n\n perf top: Set display thread COMM to help with debugging\n \n When we want to attach just to the thread that updates the display it\n helps having its COMM stand out, so change it from the default \"perf\" to\n \"perf-top-UI\".\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-5w0hmlk3zfvysxvpsh763k9w@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit b115df076d337a727017538d11d7d46f5bcbff15\nAuthor: Haiyan Song \u003chaiyanx.song@intel.com\u003e\nDate: Wed Jun 12 16:15:42 2019 +0800\n\n perf vendor events intel: Add Icelake V1.00 event file\n \n Add a Intel event file for perf.\n \n Signed-off-by: Haiyan Song \u003chaiyanx.song@intel.com\u003e\n Reviewed-by: Kan Liang \u003ckan.liang@linux.intel.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Jin Yao \u003cyao.jin@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/8859095e-5b02-d6b7-fbdc-3f42b714bae0@intel.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 2b75863b0845764529e01014a5c90664d8044cbe\nAuthor: Luke Mujica \u003clukemujica@google.com\u003e\nDate: Fri Jul 19 13:22:53 2019 -0700\n\n perf tools: Fix paths in include statements\n \n These paths point to the wrong location but still work because they get\n picked up by a -I flag that happens to direct to the correct file. Fix\n paths to lead to the actual file location without help from include\n flags.\n \n Signed-off-by: Luke Mujica \u003clukemujica@google.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Ian Rogers \u003cirogers@google.com\u003e\n Cc: Jiri Olsa \u003cjolsa@redhat.com\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Stephane Eranian \u003ceranian@google.com\u003e\n Link: http://lkml.kernel.org/r/20190719202253.220261-1-lukemujica@google.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 4fe94ce1c6ba678b5f12b94bb9996eea4fc99e85\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Jul 30 11:37:44 2019 -0300\n\n perf test vfs_getname: Disable ~/.perfconfig to get default output\n \n To get the expected output we have to ignore whatever changes the user\n has in its ~/.perfconfig file, so set PERF_CONFIG to /dev/null to\n achieve that.\n \n Before:\n \n # egrep \u0027trace|show_\u0027 ~/.perfconfig\n [trace]\n show_zeros \u003d yes\n show_duration \u003d no\n show_timestamp \u003d no\n show_arg_names \u003d no\n show_prefix \u003d yes\n # echo $PERF_CONFIG\n \n # perf test \"trace + vfs_getname\"\n 70: Check open filename arg using perf trace + vfs_getname: FAILED!\n # export PERF_CONFIG\u003d/dev/null\n # perf test \"trace + vfs_getname\"\n 70: Check open filename arg using perf trace + vfs_getname: Ok\n #\n \n After:\n \n # egrep \u0027trace|show_\u0027 ~/.perfconfig\n [trace]\n show_zeros \u003d yes\n show_duration \u003d no\n show_timestamp \u003d no\n show_arg_names \u003d no\n show_prefix \u003d yes\n # echo $PERF_CONFIG\n \n # perf test \"trace + vfs_getname\"\n 70: Check open filename arg using perf trace + vfs_getname: Ok\n #\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Taeung Song \u003ctreeze.taeung@gmail.com\u003e\n Link: https://lkml.kernel.org/n/tip-3up27pexg5i3exuzqrvt4m8u@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 5de9e5fda05b580c036e1fec6e2d8bf78eb2ac9d\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Jul 30 11:30:37 2019 -0300\n\n perf config: Document the PERF_CONFIG environment variable\n \n There was a provision for setting this variable, but not the\n getenv(\"PERF_CONFIG\") call to set it, as this was fixed in the previous\n cset, document that it can be used to ask for using an alternative\n .perfconfig file or to disable reading whatever file exists in the\n system or home directory, i.e. using:\n \n export PERF_CONFIG\u003d/dev/null\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Taeung Song \u003ctreeze.taeung@gmail.com\u003e\n Link: https://lkml.kernel.org/n/tip-0u4o967hsk7j0o50zp9ctn89@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 61a461fcbd62d42c29a1ea6a9cc3838ad9f49401\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Jul 30 11:20:55 2019 -0300\n\n perf config: Honour $PERF_CONFIG env var to specify alternate .perfconfig\n \n We had this comment in Documentation/perf_counter/config.c, i.e. since\n when we got this from the git sources, but never really did that\n getenv(\"PERF_CONFIG\"), do it now as I need to disable whatever\n ~/.perfconfig root has so that tests parsing tool output are done for\n the expected default output or that we specify an alternate config file\n that when read will make the tools produce expected output.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Taeung Song \u003ctreeze.taeung@gmail.com\u003e\n Fixes: 078006012401 (\"perf_counter tools: add in basic glue from Git\")\n Link: https://lkml.kernel.org/n/tip-jo209zac9rut0dz1rqvbdlgm@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 57fc032ad643ffd018d66bd4c1bd3a91de4841e8\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Jul 30 10:58:41 2019 -0300\n\n perf session: Avoid infinite loop when seeing invalid header.size\n \n Vince reported that when fuzzing the userland perf tool with a bogus\n perf.data file he got into a infinite loop in \u0027perf report\u0027.\n \n Changing the return of fetch_mmaped_event() to ERR_PTR(-EINVAL) for that\n case gets us out of that infinite loop.\n \n Reported-by: Vince Weaver \u003cvincent.weaver@maine.edu\u003e\n Tested-by: Vince Weaver \u003cvincent.weaver@maine.edu\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190726211415.GE24867@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 7a14826ede1d714f0bb56de8167c0e519041eeda\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Mon Aug 12 14:29:38 2019 -0400\n\n ext4: set error return correctly when ext4_htree_store_dirent fails\n \n Currently when the call to ext4_htree_store_dirent fails the error return\n variable \u0027ret\u0027 is is not being set to the error code and variable count is\n instead, hence the error code is not being returned. Fix this by assigning\n ret to the error return code.\n \n Addresses-Coverity: (\"Unused value\")\n Fixes: 8af0f0822797 (\"ext4: fix readdir error in the case of inline_data+dir_index\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n\ncommit a21ce8ad12d2e429fa393d6e3b79943f2b96e3a4\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 12 18:48:04 2019 +0100\n\n drm/i915/overlay: Switch to using i915_active tracking\n \n Remove the raw i915_active_request tracking in favour of the higher\n level i915_active tracking for the sole purpose of making the lockless\n transition easier in later patches.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812174804.26180-2-chris@chris-wilson.co.uk\n\ncommit 3d6792cf0a93d34e939acda7e35a649d83440274\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 12 18:48:03 2019 +0100\n\n drm/i915: Forgo last_fence active request tracking\n \n We were using the last_fence to track the last request that used this\n vma that might be interpreted by a fence register and forced ourselves\n to wait for this request before modifying any fence register that\n overlapped our vma. Due to requirement that we need to track any XY_BLT\n command, linear or tiled, this in effect meant that we have to track the\n vma for its active lifespan anyway, so we can forgo the explicit\n last_fence tracking and just use the whole vma-\u003eactive.\n \n Another solution would be to pipeline the register updates, and would\n help resolve some long running stalls for gen3 (but only gen 2 and 3!)\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812174804.26180-1-chris@chris-wilson.co.uk\n\ncommit ba31ebfa7b749906e0befcc1e0c0db5e7463d55e\nAuthor: Andrea Parri \u003cparri.andrea@gmail.com\u003e\nDate: Mon Aug 5 14:15:17 2019 +0200\n\n MAINTAINERS: Update e-mail address for Andrea Parri\n \n My @amarulasolutions.com address stopped working this July, so update\n to my @gmail.com address where you\u0027ll still be able to reach me.\n \n Signed-off-by: Andrea Parri \u003cparri.andrea@gmail.com\u003e\n Cc: Alan Stern \u003cstern@rowland.harvard.edu\u003e\n Cc: Will Deacon \u003cwill@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Boqun Feng \u003cboqun.feng@gmail.com\u003e\n Cc: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Cc: David Howells \u003cdhowells@redhat.com\u003e\n Cc: Jade Alglave \u003cj.alglave@ucl.ac.uk\u003e\n Cc: Luc Maranget \u003cluc.maranget@inria.fr\u003e\n Cc: \"Paul E. McKenney\" \u003cpaulmck@linux.ibm.com\u003e\n Cc: Akira Yokosawa \u003cakiyks@gmail.com\u003e\n Cc: Daniel Lustig \u003cdlustig@nvidia.com\u003e\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 511b44f7598ce602f9efce687ca9eec013967d9b\nAuthor: Mukesh Ojha \u003cmojha@codeaurora.org\u003e\nDate: Mon Jul 29 13:25:57 2019 +0530\n\n rcu: Fix spelling mistake \"greate\"-\u003e\"great\"\n \n This commit fixes a spelling mistake in file tree_exp.h.\n \n Signed-off-by: Mukesh Ojha \u003cmojha@codeaurora.org\u003e\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 1d5087ab964d84e5a0cfe5059cf5e929127d573f\nAuthor: Paul E. McKenney \u003cpaulmck@linux.vnet.ibm.com\u003e\nDate: Tue May 12 14:50:06 2015 -0700\n\n arm: Use common outgoing-CPU-notification code\n \n This commit removes the open-coded CPU-offline notification with new\n common code. In particular, this change avoids calling scheduler code\n using RCU from an offline CPU that RCU is ignoring. This is a minimal\n change. A more intrusive change might invoke the cpu_check_up_prepare()\n and cpu_set_state_online() functions at CPU-online time, which would\n allow onlining throw an error if the CPU did not go offline properly.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.vnet.ibm.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: Russell King \u003clinux@arm.linux.org.uk\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Dietmar Eggemann \u003cdietmar.eggemann@arm.com\u003e\n\ncommit b823cafa7501f946a37dce5aa1e576a0b2f31ed9\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Fri Jul 5 08:05:10 2019 -0700\n\n rcu: Remove redundant \"if\" condition from rcu_gp_is_expedited()\n \n Because rcu_expedited_nesting is initialized to 1 and not decremented\n until just before init is spawned, rcu_expedited_nesting is guaranteed\n to be non-zero whenever rcu_scheduler_active \u003d\u003d RCU_SCHEDULER_INIT.\n This commit therefore removes this redundant \"if\" equality test.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n Reviewed-by: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\n\ncommit e78a7614f3876ac649b3df608789cb6ef74d0480\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Wed Jun 5 07:46:43 2019 -0700\n\n idle: Prevent late-arriving interrupts from disrupting offline\n \n Scheduling-clock interrupts can arrive late in the CPU-offline process,\n after idle entry and the subsequent call to cpuhp_report_idle_dead().\n Once execution passes the call to rcu_report_dead(), RCU is ignoring\n the CPU, which results in lockdep complaints when the interrupt handler\n uses RCU:\n \n ------------------------------------------------------------------------\n \n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n WARNING: suspicious RCU usage\n 5.2.0-rc1+ #681 Not tainted\n -----------------------------\n kernel/sched/fair.c:9542 suspicious rcu_dereference_check() usage!\n \n other info that might help us debug this:\n \n RCU used illegally from offline CPU!\n rcu_scheduler_active \u003d 2, debug_locks \u003d 1\n no locks held by swapper/5/0.\n \n stack backtrace:\n CPU: 5 PID: 0 Comm: swapper/5 Not tainted 5.2.0-rc1+ #681\n Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS Bochs 01/01/2011\n Call Trace:\n \u003cIRQ\u003e\n dump_stack+0x5e/0x8b\n trigger_load_balance+0xa8/0x390\n ? tick_sched_do_timer+0x60/0x60\n update_process_times+0x3b/0x50\n tick_sched_handle+0x2f/0x40\n tick_sched_timer+0x32/0x70\n __hrtimer_run_queues+0xd3/0x3b0\n hrtimer_interrupt+0x11d/0x270\n ? sched_clock_local+0xc/0x74\n smp_apic_timer_interrupt+0x79/0x200\n apic_timer_interrupt+0xf/0x20\n \u003c/IRQ\u003e\n RIP: 0010:delay_tsc+0x22/0x50\n Code: ff 0f 1f 80 00 00 00 00 65 44 8b 05 18 a7 11 48 0f ae e8 0f 31 48 89 d6 48 c1 e6 20 48 09 c6 eb 0e f3 90 65 8b 05 fe a6 11 48 \u003c41\u003e 39 c0 75 18 0f ae e8 0f 31 48 c1 e2 20 48 09 c2 48 89 d0 48 29\n RSP: 0000:ffff8f92c0157ed0 EFLAGS: 00000212 ORIG_RAX: ffffffffffffff13\n RAX: 0000000000000005 RBX: ffff8c861f356400 RCX: ffff8f92c0157e64\n RDX: 000000321214c8cc RSI: 00000032120daa7f RDI: 0000000000260f15\n RBP: 0000000000000005 R08: 0000000000000005 R09: 0000000000000000\n R10: 0000000000000001 R11: 0000000000000001 R12: 0000000000000000\n R13: 0000000000000000 R14: ffff8c861ee18000 R15: ffff8c861ee18000\n cpuhp_report_idle_dead+0x31/0x60\n do_idle+0x1d5/0x200\n ? _raw_spin_unlock_irqrestore+0x2d/0x40\n cpu_startup_entry+0x14/0x20\n start_secondary+0x151/0x170\n secondary_startup_64+0xa4/0xb0\n \n ------------------------------------------------------------------------\n \n This happens rarely, but can be forced by happen more often by\n placing delays in cpuhp_report_idle_dead() following the call to\n rcu_report_dead(). With this in place, the following rcutorture\n scenario reproduces the problem within a few minutes:\n \n tools/testing/selftests/rcutorture/bin/kvm.sh --cpus 8 --duration 5 --kconfig \"CONFIG_DEBUG_LOCK_ALLOC\u003dy CONFIG_PROVE_LOCKING\u003dy\" --configs \"TREE04\"\n \n This commit uses the crude but effective expedient of moving the disabling\n of interrupts within the idle loop to precede the cpu_is_offline()\n check. It also invokes tick_nohz_idle_stop_tick() instead of\n tick_nohz_idle_stop_tick_protected() to shut off the scheduling-clock\n interrupt.\n \n Signed-off-by: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Frederic Weisbecker \u003cfweisbec@gmail.com\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Ingo Molnar \u003cmingo@kernel.org\u003e\n [ paulmck: Revert tick_nohz_idle_stop_tick_protected() removal, new callers. ]\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 03d570e1a4dc669457af2888999ecc9548fc0d2a\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Sat Aug 10 14:36:20 2019 +0200\n\n clk: ingenic: Use CLK_OF_DECLARE_DRIVER macro\n \n By using CLK_OF_DECLARE_DRIVER instead of the CLK_OF_DECLARE macro, we\n allow the driver to probe also as a platform driver.\n \n While this driver does not have code to probe as a platform driver, this\n is still useful for probing children devices in the case where the\n device node is compatible with \"simple-mfd\".\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Link: https://lkml.kernel.org/r/20190810123620.27238-1-paul@crapouillou.net\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit b8cf3219ccd5c0f05f6265fff26cce0de8061e38\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Fri Aug 9 10:38:32 2019 -0500\n\n drm/amdgpu: flag renoir as experimental for now\n \n The current code won\u0027t likely work on production hw when\n it ships so leave it as experimental until it\u0027s ready.\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c9d0ca8528b334e87f9edf08623a2623bfb99a8c\nAuthor: Huang Rui \u003cray.huang@amd.com\u003e\nDate: Sun Jul 21 22:27:50 2019 +0800\n\n drm/amdgpu: skip mec2 jump table loading for renoir\n \n Renoir need not load mec2 jump table with psp.\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 444a0fea5107e9ad7e3cbbafed78678489e31713\nAuthor: Huang Rui \u003cray.huang@amd.com\u003e\nDate: Sun Jul 21 20:58:31 2019 +0800\n\n drm/amdgpu: use direct loading on renoir vcn for the moment\n \n PSP has issue for renoir, that will cause VCN fw failed to be loaded. So use\n direct loading for the moment till the issue is addressed.\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8deac2363638f65d08742c571b6bc26cf8ca7ed8\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Fri Aug 9 10:46:36 2019 -0500\n\n drm/amdgpu: set fw default loading by psp for renoir\n \n By default, set amdgpu ucode type to AMDGPU_FW_LOAD_PSP.\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 40c8a3293b179557e341a540cdcd1c350246a60d\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Tue Jul 16 17:36:43 2019 +0800\n\n drm/amdgpu: update lbpw for renoir\n \n enable gfx_v9_0_init_lbpw for renoir\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 95f9e74c3acc7a6edb5f9a2517df23b2df46d053\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Tue Jul 16 17:09:47 2019 +0800\n\n drm/amdgpu: enable power gating for renoir\n \n enable gfx power gating for renoir\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f78e007f76bd8be25d038dc01850984a3d551e49\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Mon Aug 12 11:32:56 2019 -0500\n\n drm/amdgpu: enable clock gating for renoir\n \n enable gfx\u0026common clock gating for renoir\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 279ba48e1f7677008e6e04d3eeb37b81b7496ed0\nAuthor: Leo Liu \u003cleo.liu@amd.com\u003e\nDate: Mon Jul 15 09:21:57 2019 -0400\n\n drm/amdgpu: add VCN2.0 to Renoir IP blocks\n \n Thus enable VCN2.0 for Renoir\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 0c6b391d68be81bbf54bc326e5b6c7a18740067b\nAuthor: Leo Liu \u003cleo.liu@amd.com\u003e\nDate: Mon Jul 15 10:14:17 2019 -0400\n\n drm/amdgpu: enable Doorbell support for Renoir (v2)\n \n Add VCN range aperture to NBIO 7.0\n \n v2: rebase (Alex)\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit dc9b6e934b42d2cea40ffabd611bde650e59b75d\nAuthor: Leo Liu \u003cleo.liu@amd.com\u003e\nDate: Mon Jul 15 09:01:51 2019 -0400\n\n drm/amdgpu: enable Renoir VCN firmware loading\n \n By adding new Renoir VCN firmware\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a46e1716f37dbaac20597e16ad40a63c1afc32bf\nAuthor: Huang Rui \u003cray.huang@amd.com\u003e\nDate: Wed Jul 24 14:03:25 2019 -0500\n\n drm/amdgpu: add sdma golden settings for renoir\n \n This patch adds sdma golden settings for renoir asic.\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 33294eb8cb6732aa72afea84eb9434c506136500\nAuthor: Huang Rui \u003cray.huang@amd.com\u003e\nDate: Sun Jun 23 02:51:57 2019 +0800\n\n drm/amdgpu: add gfx golden settings for renoir (v2)\n \n This patch adds gfx golden settings for renoir real asic.\n \n v2: update settings (Alex)\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6a7a0bdbfa0c2453fb6c59c41a767d0e5307f793\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Fri Aug 9 10:32:15 2019 -0500\n\n drm/amdgpu: add psp_v12_0 for renoir (v2)\n \n 1. Add psp ip block\n 2. Use direct loading type by default and it can also config psp\n loading type.\n 3. Bypass sos fw loading and xgmi\u0026ras interface\n \n v2: drop TA loading\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6b3ad3b2da2861cf1e7649d1f6dc699aa8e4910b\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Wed Jul 24 13:56:27 2019 -0500\n\n drm/amdgpu: set rlc funcs for renoir\n \n add gfx_v9_0_rlc_funcs for renoir\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e09ce48182e5c7e3832bf1f63edf24c7bc1cbaaa\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Mon Apr 8 13:14:28 2019 +0800\n\n drm/amdgpu: add asic funcs for renoir\n \n add asic funcs for renoir, init soc15_asic_funcs\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b1326bbc63165e538b29f0dd1f283a082c034706\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Wed Jul 24 13:55:38 2019 -0500\n\n drm/amdgpu: enable dce virtual ip module for Renoir\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 0126abd4d13f71e41fd5b08bea0b44a6b2c186a9\nAuthor: Aaron Liu \u003caaron.liu@amd.com\u003e\nDate: Wed Jul 24 13:53:17 2019 -0500\n\n drm/amdgpu: fix no interrupt issue for renoir emu\n \n In renoir\u0027s ih model, there\u0027s a change in mmIH_CHICKEN\n register, that limits IH to use physical address directly.\n Those chicken bits need to be programmed first.\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Aaron Liu \u003caaron.liu@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 61bdb39c913fcd998faef7664a5f2f0170000335\nAuthor: Huang Rui \u003cray.huang@amd.com\u003e\nDate: Tue Oct 30 11:43:02 2018 +0800\n\n drm/amdgpu: add renoir pci id\n \n Add Renoir PCI id support.\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 05e1f0e0aba621cdf95cb46309a71d13c6f56c37\nAuthor: Huang Rui \u003cray.huang@amd.com\u003e\nDate: Wed Jul 24 13:50:22 2019 -0500\n\n drm/amdgpu: set ip blocks for renoir\n \n Enable ip blocks for renoir.\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2d49738ae1b4a74a48814c62d8fbd5370b1c3a79\nAuthor: Huang Rui \u003cray.huang@amd.com\u003e\nDate: Thu Aug 8 14:58:51 2019 -0500\n\n drm/amdgpu: add sdma support for renoir\n \n Add renoir checks to appropriate places.\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1aafd447bc795be9cd43704dcfebfb570d3e728c\nAuthor: Huang Rui \u003cray.huang@amd.com\u003e\nDate: Wed Jul 24 13:47:52 2019 -0500\n\n drm/amdgpu: add gfx support for renoir\n \n Add Renoir checks to gfx9 code.\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 378d53898a8dabda2dd7fe49c72082439f7d74e6\nAuthor: Huang Rui \u003cray.huang@amd.com\u003e\nDate: Wed Jul 24 13:43:16 2019 -0500\n\n drm/amdgpu: set fw load type for renoir\n \n This patch sets fw load type as direct for renoir for the moment.\n Will switch to psp when psp is ready.\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8787ee01458f12959d2936af447e8af59b16f284\nAuthor: Huang Rui \u003cray.huang@amd.com\u003e\nDate: Wed Jul 24 13:42:16 2019 -0500\n\n drm/amdgpu: add gmc v9 supports for renoir\n \n Add gfx memory controller support for renoir.\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 080deab66d4f766ce83cb8348070ac14fba92302\nAuthor: Huang Rui \u003cray.huang@amd.com\u003e\nDate: Wed Jul 24 13:39:36 2019 -0500\n\n drm/amdgpu: add soc15 common ip block support for renoir\n \n This patch adds common ip support for renoir.\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b51a26a02abe921e599971659e2177fd2c6a1e81\nAuthor: Huang Rui \u003cray.huang@amd.com\u003e\nDate: Wed Jul 24 13:38:15 2019 -0500\n\n drm/amdgpu: add renoir support for gpu_info and ip block setting\n \n This patch adds renoir support for gpu_info firmware and ip block setting.\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1eee4228a583b20082367e5cdce403bbdad2d7f1\nAuthor: Huang Rui \u003cray.huang@amd.com\u003e\nDate: Wed Jul 24 13:36:09 2019 -0500\n\n drm/amdgpu: add renoir asic_type enum\n \n This patch adds renoir to amd_asic_type enum and amdgpu_asic_name[].\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d8a46257c22906db38590eb7986c07770bee002c\nAuthor: Huang Rui \u003cray.huang@amd.com\u003e\nDate: Thu Oct 25 19:49:10 2018 +0800\n\n drm/amdgpu: add renoir header files (v2)\n \n This patch add all renoir header files.\n \n v2: clean up headers (Alex)\n \n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8b94d0517902d179879fdb8a96edc5c0ab2e7c62\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Mon Aug 5 11:29:40 2019 +0100\n\n drm/amd/powerplay: remove redundant duplicated return check\n \n The check on ret is duplicated in two places, it is redundant code.\n Remove it.\n \n Addresses-Coverity: (\"Logically dead code\")\n Fixes: b94afb61cdae (\"drm/amd/powerplay: honor hw limit on fetching metrics data for navi10\")\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 54a9bcb0a4ff6ee401955a1e2e657b97b9391eb7\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Fri Aug 9 22:12:19 2019 +0200\n\n drm/amd/display: Fix a typo - dce_aduio_mask --\u003e dce_audio_mask\n \n This should be \u0027dce_audio_mask\u0027, not \u0027dce_aduio_mask\u0027.\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 44f3dd09a8a103018df199345c669978a90c10c1\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Thu Aug 8 00:29:23 2019 -0500\n\n drm/amd/display: use kvmalloc for dc_state (v2)\n \n It\u0027s large and doesn\u0027t need contiguous memory. Fixes\n allocation failures in some cases.\n \n v2: kvfree the memory.\n \n Reviewed-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 62cfcb9e23d0e1c4457b27732c6daee3364508da\nAuthor: Pierre-Eric Pelloux-Prayer \u003cpierre-eric.pelloux-prayer@amd.com\u003e\nDate: Tue Aug 6 18:27:26 2019 +0200\n\n drm/amdgpu: fix gfx9 soft recovery\n \n The SOC15_REG_OFFSET() macro wasn\u0027t used, making the soft recovery fail.\n \n v2: use WREG32_SOC15 instead of WREG32 + SOC15_REG_OFFSET\n \n Signed-off-by: Pierre-Eric Pelloux-Prayer \u003cpierre-eric.pelloux-prayer@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e34640e247beaa56665cd563208a47dd24406c2f\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Fri Aug 9 15:44:22 2019 +0800\n\n drm/amdgpu/powerplay: update Arcturus smu version in new place\n \n Follow patch below:\n drm/amd/powerplay: re-define smu interface version for smu v11\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a840159c82c7e7fa954f5b9e4ac7721d7e094513\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Fri Aug 9 18:58:42 2019 +0800\n\n drm/amdgpu: enable mmhub clock gating for Arcturus\n \n Init MC_MGCG/LS flag. Also apply to athub CG.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit cb15e8046d0df90deeec9ed79dc432043f11d07b\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Fri Aug 9 18:57:15 2019 +0800\n\n drm/amdgpu: add mmhub clock gating for Arcturus\n \n Add 2 mmhub instances CG\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 15e2f43a72cfe2c122281e7872ad8eaa71387d98\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Fri Aug 9 15:13:38 2019 +0800\n\n drm/amdgpu: increase CGCG gfx idle threshold for Arcturus\n \n Follow the hw spec, and no need to consider gfxoff on Arcturus\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f9da7c43842474afa8a81ca9404884e830ad30e3\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Fri Aug 9 15:24:56 2019 +0800\n\n drm/amdgpu: add GFX_CP_LS flag to Arcturus\n \n Missed AMD_CG_SUPPORT_GFX_CP_LS accidently when commit patch before\n drm/amdgpu: enable gfx clock gating for Arcturus\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5212a3bdf03f24a5ed3eabf8fc4833aa3c5287a3\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Fri Aug 9 17:39:06 2019 +0800\n\n drm/amdgpu: remove ras block\u0027s feature status info in sysfs\n \n feature mask info is enough for rocm tool,\n \"cat /sys/class/drm/card0/device/ras/features\" will get the\n info like this:\n \n feature mask: 0x3ffb\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9b4e63f4513bbd626b2490ce81435a38d7bc3902\nAuthor: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\nDate: Tue Jul 23 12:16:25 2019 +0800\n\n drm/amd/powerplay: change smu_read_sensor sequence in smu\n \n change the smu_read_sensor sequence to:\n \n asic specific sensor read -\u003e smu v11 specific sensor read -\u003e smu v11 common sensor read\n \n Signed-off-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit bee7b51ac93a67c54b77cfa17c87ddde7c661988\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Thu Aug 8 14:54:12 2019 +0800\n\n drm/amdgpu: split athub clock gating from mmhub\n \n Untie the bind of get/set athub CG state from mmhub, for cosmetic fix and Asic\n not using mmhub 1.0. Besides, also fix wrong athub CG state in amdgpu_pm_info.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f7ee1995282c804cef69ead5d3a8ba79b371dde2\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Wed Aug 7 15:48:44 2019 +0800\n\n drm/amdgpu: enable sdma clock gating for Arcturus\n \n Init sdma MGCG/LS flag\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8dc7e07cfff544640d4606165bbd27b48aff6538\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Wed Aug 7 15:47:34 2019 +0800\n\n drm/amdgpu: add sdma clock gating for Arcturus\n \n Add ARCTURUS case in sdma set clockgating function\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 78864760c2e3bc5f878bea1ec911cf8087729de6\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Wed Aug 7 15:45:25 2019 +0800\n\n drm/amdgpu: support sdma clock gating for more instances\n \n Shorten the code with RREG32_SDMA/WREG32_SDMA macro in CG part.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5d111f5b3a318bb98be640ad5e9198b6366fd6e9\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Wed Aug 7 15:17:38 2019 +0800\n\n drm/amdgpu: enable hdp clock gating for Arcturus\n \n Init hdp MGCG/LS flag as Vega20\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6acb87acefd763a9aa2e2ec5ceb6840ff3229fc4\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Wed Aug 7 15:16:19 2019 +0800\n\n drm/amdgpu: add hdp clock gating for Arcturus\n \n Add hdp CGLS for Arcturus in set common clockgating function\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6b76ce62bfcb6a3d03eb695cd739b8db4d089f20\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Wed Aug 7 14:52:38 2019 +0800\n\n drm/amdgpu: enable gfx clock gating for Arcturus\n \n Init gfx MGCG/LS and CGCG/LS flag.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f60481a94529b0052f5593c1047ddb8bf325c722\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Wed Aug 7 14:59:07 2019 +0800\n\n drm/amdgpu: add gfx clock gating for Arcturus\n \n Add ARCTURUS case in gfx set clockgating function. No 3d clock on Arcturus.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 145b03eb738e5a83b746c78cc571d9dc656b4b2c\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Wed Aug 7 12:21:22 2019 +0800\n\n drm/amdgpu: create mmhub ras framework\n \n enable mmhub ras feature and create sysfs/debugfs node for mmhub\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9fb2d8de4a43e7d92d1d53c0d4ebdda0a0ddc8d7\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Tue Aug 6 20:22:49 2019 +0800\n\n drm/amdgpu: support mmhub ras in amdgpu ras\n \n call mmhub ras query/inject in amdgpu ras\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 3d093da0989ef2b25ed7613ff7d5f5caf9033117\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Tue Aug 6 20:15:55 2019 +0800\n\n drm/amdgpu: add amdgpu_mmhub_funcs definition\n \n add amdgpu_mmhub_funcs definition and initialize it,\n prepare for mmhub ras enablement\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 44494f96ba962c6e0a0d8b1c0db5e7b54c75ab3a\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Wed Aug 7 14:27:42 2019 +0800\n\n drm/amdgpu: add sub block parameter in ras inject command\n \n ras sub block index could be passed from shell command\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Guchun Chen \u003cguchun.chen@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a2b459947b1e8e9b8d8cb13640bc019ee932ea30\nAuthor: Guchun Chen \u003cguchun.chen@amd.com\u003e\nDate: Thu Aug 8 14:54:41 2019 +0800\n\n drm/amdgpu: add check to avoid array bound issue\n \n Sub_block_index can be passed from user level, so\n add one check before accessing the array first to\n prevent array index out of bound problem.\n \n Signed-off-by: Guchun Chen \u003cguchun.chen@amd.com\u003e\n Reviewed-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 3c0b547da0d7f9011c87f081ef3cd743b367b088\nAuthor: tiancyin \u003ctianci.yin@amd.com\u003e\nDate: Thu Aug 8 13:20:40 2019 +0800\n\n drm/amd/powerplay: update smu11_driver_if_navi10.h\n \n update the smu11_driver_if_navi10.h since navi14 smu fw\n update to 53.12\n \n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: tiancyin \u003ctianci.yin@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1b41b769e137222eb003395ab2c76db50c799edd\nAuthor: tiancyin \u003ctianci.yin@amd.com\u003e\nDate: Thu Aug 8 11:57:28 2019 +0800\n\n drm/amd/powerplay: re-define smu interface version for smu v11\n \n [why]\n navi14 share same defination of smu interface version with navi10,\n anyone of them update the version may break the other one\u0027s\n version checking.\n \n [how]\n create different version defination, so that they can\n update their version separately.\n \n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: tiancyin \u003ctianci.yin@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 26051720321039329fb6f330cef7e88874a5127e\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Wed Aug 7 14:37:26 2019 -0500\n\n drm/amdgpu: add navi14 PCI ID\n \n Add the navi14 PCI device id.\n \n Reviewed-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 965ebe3d5d641925e0014a5a8c967fe579929b1d\nAuthor: Michel Dänzer \u003cmichel.daenzer@amd.com\u003e\nDate: Wed Jul 24 17:56:28 2019 +0200\n\n drm/amdgpu: Update pitch on page flips without DC as well\n \n DC already handles this correctly since amdgpu minor version 31. Bump\n the minor version again so that xf86-video-amdgpu can take advantage of\n this working without DC as well now.\n \n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Michel Dänzer \u003cmichel.daenzer@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 29c9d8eadc7a7e2fbfe4632c9050c814c260923b\nAuthor: David Francis \u003cDavid.Francis@amd.com\u003e\nDate: Wed Aug 7 10:25:48 2019 -0400\n\n drm/amd/display: Remove drm_dsc_dc.c\n \n This file was accidentally added to the driver during\n Navi promotion\n \n Nothing includes it. No makefile attempts to compile it, and\n it would fail compilation if they tried\n \n Remove it\n \n Signed-off-by: David Francis \u003cDavid.Francis@amd.com\u003e\n Reviewed-by: Harry Wentland \u003charry.wentland@amd.com\u003ew\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 65872e59d67e67a343b555734abe2247dd6f7902\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu Aug 1 15:22:59 2019 +0800\n\n drm/amdgpu: enable vcn clock gating for navi12\n \n enables vcn medium grained clock gating\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ca51678db44fbc7d4018da47015cb5d9317de282\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu Aug 1 15:19:10 2019 +0800\n\n drm/amdgpu: enable athub clock gating for navi12\n \n enables athub medium grained clock gating and memory light sleep\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 89b8d6da24b0c89445fd53ad991c58bf85304cbd\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Aug 6 13:42:03 2019 +0800\n\n drm/amdgpu/athub2: set clock gating for navi12\n \n add navi12 define\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit fbe0bc5794edfe74c05d91965c8b32577e5bb5f1\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu Aug 1 15:01:23 2019 +0800\n\n drm/amdgpu: enable ih clock gating for navi12\n \n enables ih clock gating\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8b797b3d30c4e738d289f830e5733029819d7b29\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu Aug 1 15:39:59 2019 +0800\n\n drm/amdgpu: enable mmhub clock gating for navi12\n \n enables mmhub medium grained clock gating and memory light sleep\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit cf5a95e5b80d58ad262797a7b49e80f8edfc6815\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu Aug 1 15:47:15 2019 +0800\n\n drm/amdgpu/mmhub2: set clock gating for navi12\n \n add navi12 define\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 358ab97f5366dfefa2645f38cf542bf29a393fa3\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Jul 30 12:18:55 2019 +0800\n\n drm/amdgpu: enable sdma clock gating for navi12\n \n enables sdma medium grained clock gating and memory light sleep\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e2f9726ee9b7fa54b685fb30e1e119c11f039645\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Jul 30 12:16:02 2019 +0800\n\n drm/amdgpu/sdma5: set sdma clock gating for navi12\n \n add navi12 define\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5211c37a3445c5cde08f1f6f95d36cefaf3ad818\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu Aug 1 15:00:28 2019 +0800\n\n drm/amdgpu: enable hdp clock gating for navi12\n \n enables hdp medium grained clock gating and memory light sleep\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit dca009e71c72866f99a4e3a8acaef9315adc3e13\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Jul 30 11:28:20 2019 +0800\n\n drm/amdgpu: enable gfx clock gatings for navi12\n \n enables following gfx clock gating features:\n \n - medium grained clock gating\n - medium grained light sleep\n - coarse grained clock gating\n - cp memory light sleep\n - rlc memory light sleep\n \n CGLS (Coarse Grained Light Sleep) will break s3, so don\u0027t enable it.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 05677c958ad2e852e94966e420d09745bc04b707\nAuthor: Marek Olšák \u003cmarek.olsak@amd.com\u003e\nDate: Fri Aug 2 17:44:06 2019 -0400\n\n Revert \"drm/amdgpu: fix transform feedback GDS hang on gfx10 (v2)\"\n \n This reverts commit 9ed2c993d723129f85101e51b2ccc36ef5400a67.\n \n SET_CONFIG_REG writes to memory if register shadowing is enabled,\n causing a VM fault.\n \n NGG streamout is unstable anyway, so all UMDs should use legacy\n streamout. I think Mesa is the only driver using NGG streamout.\n \n Signed-off-by: Marek Olšák \u003cmarek.olsak@amd.com\u003e\n Reviewed-by: Le Ma \u003cLe.Ma@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit cd2d99229dc96219547e6349841e1aad851c6acc\nAuthor: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\nDate: Mon Aug 12 13:44:49 2019 -0400\n\n ext4: drop legacy pre-1970 encoding workaround\n \n Originally, support for expanded timestamps had a bug in that pre-1970\n times were erroneously encoded as being in the the 24th century. This\n was fixed in commit a4dad1ae24f8 (\"ext4: Fix handling of extended\n tv_sec\") which landed in 4.4. Starting with 4.4, pre-1970 timestamps\n were correctly encoded, but for backwards compatibility those\n incorrectly encoded timestamps were mapped back to the pre-1970 dates.\n \n Given that backwards compatibility workaround has been around for 4\n years, and given that running e2fsck from e2fsprogs 1.43.2 and later\n will offer to fix these timestamps (which has been released for 3\n years), it\u0027s past time to drop the legacy workaround from the kernel.\n \n Signed-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n\ncommit 68b953aeb50d9206de27d6c216e301f01dfac34d\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Aug 12 11:06:23 2019 -0500\n\n ASoC: SOF: Intel: hda: fixup HDaudio topology name with DMIC number\n \n The SOF project maintains 6 topologies for HDaudio (iDisp or\n HDaudio+iDisp, no DMIC, 2 DMICs, 4 DMICs). The user is currently\n required to manually rename the topology file used in\n /lib/firmware/intel/sof-tplg. We can do better to avoid such\n renames and use logic to select the relevant file.\n \n The NHLT information can be used to figure out which topology file\n should be used.\n \n Alternatively, when NHLT is not present in ACPI tables or is possibly\n incorrect, a module parameter can provide that information, e.g. on\n Up^2 board with the test DMIC kit.\n \n Tested on Up^2 board and Acer Swift-SF314-55\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190812160623.20821-1-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit f7b08ea5f53bd92dc30671712d555b7a7ebb6e7e\nAuthor: Andrew Murray \u003candrew.murray@arm.com\u003e\nDate: Tue Jul 30 13:51:55 2019 +0100\n\n coresight: etm4x: improve clarity of etm4_os_unlock comment\n \n To improve clarity, let\u0027s update the comment for etm4_os_unlock\n to use the name of the register as per the ETM architecture\n specification.\n \n The existing comment is also misleading as it suggests any value\n written to TRCOSLAR unlocks the trace registers, however it must\n be \u00270\u0027 - let\u0027s also correct this.\n \n Signed-off-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n Reviewed-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n\ncommit ac6cb440cbd0af224a7b714bcff40a92f65930fb\nAuthor: Andrew Murray \u003candrew.murray@arm.com\u003e\nDate: Tue Jul 30 13:51:54 2019 +0100\n\n coresight: etm4x: use module_param instead of module_param_named\n \n Given that the user-exposed module parameter for \u0027boot_enable\u0027 matches\n the variable that it sets, let\u0027s use module_param instead of\n module_param_named.\n \n Let\u0027s also use octal permissions (checkpatch recommends this) and\n provide a module parameter description.\n \n Signed-off-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n Reviewed-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n\ncommit 8f86a5b4ad679e4836733b47414226074eee4e4d\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Mon Aug 12 10:23:31 2019 +0200\n\n gpio: merrifield: Pass irqchip when adding gpiochip\n \n We need to convert all old gpio irqchips to pass the irqchip\n setup along when adding the gpio_chip. For more info see\n drivers/gpio/TODO.\n \n For chained irqchips this is a pretty straight-forward\n conversion.\n \n Cc: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Cc: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n Cc: David Cohen \u003cdavid.a.cohen@linux.intel.com\u003e\n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 7b1e889436a156c2c88953fbc5602a64a75889bd\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Mon Aug 12 10:13:51 2019 +0200\n\n gpio: lynxpoint: Pass irqchip when adding gpiochip\n \n We need to convert all old gpio irqchips to pass the irqchip\n setup along when adding the gpio_chip. For more info see\n drivers/gpio/TODO.\n \n For chained irqchips this is a pretty straight-forward\n conversion.\n \n Cc: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Cc: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n Cc: David Cohen \u003cdavid.a.cohen@linux.intel.com\u003e\n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 8069e69a9792402b52eff61e5bb874fa94798fc9\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Sun Aug 11 10:05:39 2019 +0200\n\n gpio: intel-mid: Pass irqchip when adding gpiochip\n \n We need to convert all old gpio irqchips to pass the irqchip\n setup along when adding the gpio_chip. For more info see\n drivers/gpio/TODO.\n \n For chained irqchips this is a pretty straight-forward\n conversion.\n \n Cc: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Cc: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n Cc: David Cohen \u003cdavid.a.cohen@linux.intel.com\u003e\n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit a5ff2130a4d98d22377e5fdab87d713f4b17f4f2\nAuthor: Helge Deller \u003cdeller@gmx.de\u003e\nDate: Mon Aug 12 19:13:20 2019 +0200\n\n parisc: speed up flush_tlb_all_local with qemu\n \n When started in qemu, we know that qemu will drop all local TLB entries\n on any pxtlbe instruction. So, if we detect qemu, replace the whole\n flush_tlb_all_local function by one pdtlbe instruction.\n \n Signed-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n\ncommit 82992fc70f98dee091faa926eb5cecadda5c84f4\nAuthor: Helge Deller \u003cdeller@gmx.de\u003e\nDate: Mon Aug 12 19:11:06 2019 +0200\n\n parisc: Add ALTERNATIVE_CODE() and ALT_COND_RUN_ON_QEMU\n \n The macro ALTERNATIVE_CODE() allows assembly code to patch in a series\n of new assembler statements given at a specific start address.\n The ALT_COND_RUN_ON_QEMU condition is true if the kernel is started in a\n qemu emulation.\n \n Signed-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n\ncommit bfb77169306d5d560a8b62eebaf6d69d02e8d152\nAuthor: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\nDate: Fri Aug 9 15:09:58 2019 +0200\n\n ARM: dts: exynos: Add CAM power domain to Exynos5422/5800\n \n CAM power domain contains CAMIF, 3AA and FIMC LITE devices. It is present\n only in Exynos 5422/5800 SoCs. Currently there are no drivers nor the\n device nodes for those devices, but instantiating its power domain allows\n to turn it off and save some energy.\n \n Signed-off-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 8686764fc0711f409b8fcaaa76e7c5a816e9160b\nAuthor: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\nDate: Fri Aug 9 15:09:57 2019 +0200\n\n ARM: dts: exynos: Add G3D power domain to Exynos542x\n \n Add a power domain for G3D/Mali device present in Exynos542x/5800 SoCs.\n Node for the Mali device will be added by a separate patch.\n \n Signed-off-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 791aa2150b296059b017cf458075c2f90a652b2b\nAuthor: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\nDate: Fri Aug 9 15:09:56 2019 +0200\n\n ARM: dts: exynos: Move MSC power domain to the right (sorted) place\n \n DT nodes should be sorted by \u0027reg\u0027 property, so move MSC power domain\n node in exynos5420.dtsi to the end of power domains to keep them sorted.\n \n Signed-off-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 98dc19902a0b2e5348e43d6a2c39a0a7d0fc639e\nAuthor: Jeremy Linton \u003cjeremy.linton@arm.com\u003e\nDate: Thu Aug 8 15:40:07 2019 -0500\n\n arm64: topology: Use PPTT to determine if PE is a thread\n \n ACPI 6.3 adds a thread flag to represent if a CPU/PE is\n actually a thread. Given that the MPIDR_MT bit may not\n represent this information consistently on homogeneous machines\n we should prefer the PPTT flag if its available.\n \n Signed-off-by: Jeremy Linton \u003cjeremy.linton@arm.com\u003e\n Reviewed-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Reviewed-by: Robert Richter \u003crrichter@marvell.com\u003e\n [will: made acpi_cpu_is_threaded() return \u0027bool\u0027]\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 4189ff23489e6688ef4c7f5109df0cebbad425b1\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Aug 12 08:55:24 2019 +0200\n\n kernel: only define task_struct_whitelist conditionally\n \n If CONFIG_ARCH_TASK_STRUCT_ALLOCATOR is set task_struct_whitelist is\n never called, and thus generates a compiler warning.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190812065524.19959-5-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit d0d82d24cdfd289d755c87448dbc27c2dd0559ec\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Aug 12 08:55:23 2019 +0200\n\n ia64/kprobes: remove the unused ia64_get_bsp_cfm function\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190812065524.19959-4-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 94707d90b7d6ea5552078e2c3b78b7f1e29ee528\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Aug 12 08:55:22 2019 +0200\n\n ia64: annotate switch fallthroughs in ia64_handle_unaligned\n \n Replace the \"no break\" comments with something that the compiler\n recognizes.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190812065524.19959-3-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 782d7a217eb63353dbfd81fb88b24e600bc6bd9c\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Aug 12 08:55:21 2019 +0200\n\n ia64: annotate a switch fallthrough in ia64_do_signal\n \n Also reindent the switch statement to use the normal kernel style\n while at it.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Link: https://lkml.kernel.org/r/20190812065524.19959-2-hch@lst.de\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 0bf63e2c2913391dbf7e49ed8df2c89c180e58d9\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Tue Jul 30 03:17:59 2019 -0300\n\n media: mach-omap2/devices.c: set dma mask\n \n The dma_mask and coherent_dma_mask values were never set.\n \n This prevented the media omap_vout driver from loading successfully.\n \n Tested on a Pandaboard and Beagle XM board.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Acked-by: Tony Lindgren \u003ctony@atomide.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 7da20788d3ff59506aebd946d8a424f01f2c1c66\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Mon Aug 12 18:01:47 2019 +0200\n\n ALSA: hda: Set fifo_size for both playback and capture streams\n \n Currently we set hdac_stream.fifo_size field only for the playback\n stream by some odd reason I forgot, while this field isn\u0027t referred in\n any places. Actually this fifo_size field would have been required in\n the position report correction for VIA chipset, but due to the lack of\n the fifo_size set for capture streams, snd-hda-intel driver fetches\n the register by itself.\n \n This patch straightens and simplifies the code by setting the\n fifo_size field for both playback and capture streams, and use it in\n the HD-audio controller driver.\n \n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit f246ff5c232596dbd5e7a9d227113031fb0bfebe\nAuthor: Jason Ekstrand \u003cjason@jlekstrand.net\u003e\nDate: Mon Aug 12 09:22:11 2019 -0500\n\n drm/syncobj: Add better overview documentation for syncobj (v2)\n \n This patch only brings the syncobj documentation up-to-date for the\n original form of syncobj. It does not contain any information about the\n design of timeline syncobjs.\n \n v2: Incorporate feedback from Lionel and Christian:\n - Mention actual ioctl and flag names\n - Better language around reference counting\n - Misc. language cleanups\n \n Signed-off-by: Jason Ekstrand \u003cjason@jlekstrand.net\u003e\n Reviewed-by: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Acked-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812142211.15885-1-jason@jlekstrand.net\n\ncommit db50077b953072b420943e89fcf06d5922898b89\nAuthor: Lijun Ou \u003coulijun@huawei.com\u003e\nDate: Thu Aug 8 22:53:54 2019 +0800\n\n RDMA/hns: Use the new APIs for printing log\n \n Here uses the new APIs instead of some dev print interfaces in\n some functions.\n \n Signed-off-by: Lijun Ou \u003coulijun@huawei.com\u003e\n Link: https://lore.kernel.org/r/1565276034-97329-15-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit d967e2625faa43da9e3be3228b8c57d3cf404953\nAuthor: Weihang Li \u003cliweihang@hisilicon.com\u003e\nDate: Thu Aug 8 22:53:53 2019 +0800\n\n RDMA/hns: Disable alw_lcl_lpbk of SSU\n \n If we enabled alw_lcl_lpbk in promiscuous mode, packet whose source\n and destination mac address is equal will be handled in both inner\n loopback and outer loopback. This will halve performance of roce in\n promiscuous mode.\n \n Signed-off-by: Weihang Li \u003cliweihang@hisilicon.com\u003e\n Link: https://lore.kernel.org/r/1565276034-97329-14-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 249f2f921f24db878b0cc5da80cb15d531587d79\nAuthor: Weihang Li \u003cliweihang@hisilicon.com\u003e\nDate: Thu Aug 8 22:53:52 2019 +0800\n\n RDMA/hns: Remove redundant print in hns_roce_v2_ceq_int()\n \n There is no need to tell users when eq-\u003econs_index is overflow, we\n just set it back to zero.\n \n Signed-off-by: Weihang Li \u003cliweihang@hisilicon.com\u003e\n Link: https://lore.kernel.org/r/1565276034-97329-13-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 260c3b3499198b2cfa911ea7fb4f28bb129ef1e8\nAuthor: Yangyang Li \u003cliyangyang20@huawei.com\u003e\nDate: Thu Aug 8 22:53:51 2019 +0800\n\n RDMA/hns: Refactor hns_roce_v2_set_hem for hip08\n \n In order to reduce the complexity of hns_roce_v2_set_hem, extract\n the implementation of op as a function.\n \n Signed-off-by: Yangyang Li \u003cliyangyang20@huawei.com\u003e\n Link: https://lore.kernel.org/r/1565276034-97329-12-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 4b42d05d0b2ca56bef42976d304f478f567966ec\nAuthor: Lang Cheng \u003cchenglang@huawei.com\u003e\nDate: Thu Aug 8 22:53:50 2019 +0800\n\n RDMA/hns: Remove unnecessary kzalloc\n \n For hns_roce_v2_query_qp and hns_roce_v2_modify_qp,\n we can use stack memory to create qp context data.\n Make the code simpler.\n \n Signed-off-by: Lang Cheng \u003cchenglang@huawei.com\u003e\n Link: https://lore.kernel.org/r/1565276034-97329-11-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit bebdb83f97ee725a8327d36d7cedb85eb0492726\nAuthor: Lang Cheng \u003cchenglang@huawei.com\u003e\nDate: Thu Aug 8 22:53:49 2019 +0800\n\n RDMA/hns: Refactor irq request code\n \n Remove unnecessary if...else..., to make the code look simpler.\n \n Signed-off-by: Lang Cheng \u003cchenglang@huawei.com\u003e\n Link: https://lore.kernel.org/r/1565276034-97329-10-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit e7f40440afb829594c3ae96187404bc0ecf15ce3\nAuthor: Lang Cheng \u003cchenglang@huawei.com\u003e\nDate: Thu Aug 8 22:53:48 2019 +0800\n\n RDMA/hns: Split bool statement and assign statement\n \n Assign statement can not be contained in bool statement or\n function param.\n \n Signed-off-by: Lang Cheng \u003cchenglang@huawei.com\u003e\n Link: https://lore.kernel.org/r/1565276034-97329-9-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 0e20ebf8d48e02846e202cad23bbfd6850b0b003\nAuthor: Lang Cheng \u003cchenglang@huawei.com\u003e\nDate: Thu Aug 8 22:53:47 2019 +0800\n\n RDMA/hns: Handling the error return value of hem function\n \n Handling the error return value of hns_roce_calc_hem_mhop.\n \n Signed-off-by: Lang Cheng \u003cchenglang@huawei.com\u003e\n Link: https://lore.kernel.org/r/1565276034-97329-8-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 6def7de6d450893852802794ab44d8053672cf88\nAuthor: Lang Cheng \u003cchenglang@huawei.com\u003e\nDate: Thu Aug 8 22:53:46 2019 +0800\n\n RDMA/hns: Update some comments style\n \n Here removes some useless comments and adds necessary spaces to\n another comments.\n \n Signed-off-by: Lang Cheng \u003cchenglang@huawei.com\u003e\n Link: https://lore.kernel.org/r/1565276034-97329-7-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit b5c229dc1585adcbe158ecaebb5e6993c9c220b6\nAuthor: Lang Cheng \u003cchenglang@huawei.com\u003e\nDate: Thu Aug 8 22:53:45 2019 +0800\n\n RDMA/hns: Clean up unnecessary initial assignment\n \n Here remove some unncessary initialization for some valiables.\n \n Signed-off-by: Lang Cheng \u003cchenglang@huawei.com\u003e\n Signed-off-by: Lijun Ou \u003coulijun@huawei.com\u003e\n Link: https://lore.kernel.org/r/1565276034-97329-6-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 2288b3b3b187520d7fef37197c07ca035d9b8111\nAuthor: Yixian Liu \u003cliuyixian@huawei.com\u003e\nDate: Thu Aug 8 22:53:44 2019 +0800\n\n RDMA/hns: Remove unnessary init for cmq reg\n \n There is no need to init the enable bit of cmq.\n \n Signed-off-by: Yixian Liu \u003cliuyixian@huawei.com\u003e\n Link: https://lore.kernel.org/r/1565276034-97329-5-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit ece9c205f707fc0fda832d5da18b745d30c7c6bc\nAuthor: Yixian Liu \u003cliuyixian@huawei.com\u003e\nDate: Thu Aug 8 22:53:43 2019 +0800\n\n RDMA/hns: Update the prompt message for creating and destroy qp\n \n Current prompt message is uncorrect when destroying qp, add qpn\n information when creating qp.\n \n Signed-off-by: Yixian Liu \u003cliuyixian@huawei.com\u003e\n Signed-off-by: Lijun Ou \u003coulijun@huawei.com\u003e\n Link: https://lore.kernel.org/r/1565276034-97329-4-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 8ea417ffc2db52a69b16712ab6ee07e8f74aa740\nAuthor: Lijun Ou \u003coulijun@huawei.com\u003e\nDate: Thu Aug 8 22:53:42 2019 +0800\n\n RDMA/hns: Optimize hns_roce_modify_qp function\n \n Here mainly packages some code into some new functions in order to\n reduce code compelexity.\n \n Signed-off-by: Lijun Ou \u003coulijun@huawei.com\u003e\n Link: https://lore.kernel.org/r/1565276034-97329-3-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit cc95b23c250005873e4a117fccc53a8378fb3aed\nAuthor: Lijun Ou \u003coulijun@huawei.com\u003e\nDate: Thu Aug 8 22:53:41 2019 +0800\n\n RDMA/hns: Encapsulate some lines for setting sq size in user mode\n \n It needs to check the sq size with integrity when configures\n the relatived parameters of sq. Here moves the relatived code\n into a special function.\n \n Signed-off-by: Lijun Ou \u003coulijun@huawei.com\u003e\n Link: https://lore.kernel.org/r/1565276034-97329-2-git-send-email-oulijun@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 1aee09b8fa5b872d5fde8f23a26bd775aff908f7\nAuthor: Stephen Brennan \u003cstephen@brennan.io\u003e\nDate: Sun Aug 11 15:51:20 2019 -0700\n\n staging: rtl8192u: fix spacing errors\n \n Used checkpatch\u0027s --fix-inplace option for types SPACING, OPEN_BRACE,\n ELSE_AFTER_BRACE. Manually edited the resulting changes to correct for\n mistaken fixes and complete fixes that were only partially applied.\n \n Signed-off-by: Stephen Brennan \u003cstephen@brennan.io\u003e\n Link: https://lore.kernel.org/r/20190811225120.7308-1-stephen@brennan.io\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 70deda9dfa6678bcd88020f8fb6614f13ccfb451\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 9 10:53:50 2019 +0530\n\n staging: rtl8712: r8712_construct_txaggr_cmd_hdr(): Change return type\n \n Change return type of r8712_construct_txaggr_cmd_hdr from u8 to void as\n it always returns _SUCCESS and its return value is never used.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190809052353.5308-8-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit e7df51e00828c81fe22225da92223eb6a233e91e\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 9 10:53:49 2019 +0530\n\n staging: rtl8712: r8712_construct_txaggr_cmd_desc(): Change return type\n \n Change return type of r8712_construct_txaggr_cmd_desc from u8 to void\n (and remove its return statement) as it always returns _SUCCESS and its\n return value is never stored, checked or otherwise used.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190809052353.5308-7-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 9958e0e53ba49521323b5ffeafcf996683f6172f\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 9 10:53:48 2019 +0530\n\n staging: rtl8712: r8712_xmit_direct(): Change return type\n \n Change return type of r8712_xmit_direct from int to void as its return\n value is never used. Remove return statement accordingly.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190809052353.5308-6-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 670b673fa0d0f0abd1f95e5dd26dabdcff5a7be1\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 9 10:53:47 2019 +0530\n\n staging: rtl8712: r8712_free_xmitbuf(): Change return type\n \n Change return type of r8712_free_xmitbuf from int to void (and remove\n its return values) as its return value is never used.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190809052353.5308-5-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 0d478943e3c5b3f07e48cfa6e9a0d6a0e546dab2\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 9 10:53:46 2019 +0530\n\n staging: rtl8712: make_wlanhdr(): Change return values and type\n \n Change return values of make_wlanhdr from _SUCCESS/_FAIL to 0/-EINVAL.\n Modify call site to check for non-zero return values instead of _FAIL.\n Change return type from sint to int.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190809052353.5308-4-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b013c5b8ddacdf966392b6197289721a6e4d4ac9\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 9 10:53:44 2019 +0530\n\n staging: rtl8712: r8712_update_attrib(): Change return values and type\n \n Change return values of r8712_update_attrib from _SUCCESS and _FAIL to 0\n and -ENOMEM or -EINVAL respectively. Modify call site to check for the\n new failure conditions. Also modify the return type from sint to int.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190809052353.5308-2-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit ed5d2eaf5f5d59aac17343acebe27b1929e5dcac\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 9 10:53:43 2019 +0530\n\n staging: rtl8712: _r8712_init_xmit_priv(): Change return values and type\n \n Change the return values in _r8712_init_xmit_priv from _SUCCESS/_FAIL to\n 0/-ENOMEM respectively. Change return type from sint to int.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190809052353.5308-1-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit cf1c97dcb96cb2622a0b0524b6931bd7bd7d2344\nAuthor: Andi Shyti \u003candi.shyti@intel.com\u003e\nDate: Sun Aug 11 22:06:33 2019 +0100\n\n drm/i915: Extract general GT interrupt handlers\n \n i915_irq.c is large. It serves as the central dispatch and handler for\n all of our device interrupts. Lets break it up by pulling out the GT\n interrupt handlers.\n \n Based on a patch by Chris Wilson.\n \n Signed-off-by: Andi Shyti \u003candi.shyti@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190811210633.18417-1-chris@chris-wilson.co.uk\n\ncommit d762043f7ab142291be07f15ac7ff83b7485e3e6\nAuthor: Andi Shyti \u003candi.shyti@intel.com\u003e\nDate: Sun Aug 11 15:28:00 2019 +0100\n\n drm/i915: Extract GT powermanagement interrupt handling\n \n i915_irq.c is large. It serves as the central dispatch and handler for\n all of our device interrupts. Pull out the GT pm interrupt handling\n (leaving the central dispatch) so that we can encapsulate the logic a\n little better.\n \n Based on a patch by Chris Wilson.\n \n Signed-off-by: Andi Shyti \u003candi.shyti@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190811142801.2460-1-chris@chris-wilson.co.uk\n\ncommit 133716f4ece320691f0052b296df44ed88307da4\nAuthor: Adham Abozaeid \u003cadham.abozaeid@microchip.com\u003e\nDate: Fri Aug 9 18:25:19 2019 +0000\n\n staging: wilc1000: return kernel error codes from wilc_wlan_stop\n \n return -EIO for bus failures, 0 otherwise.\n \n Signed-off-by: Adham Abozaeid \u003cadham.abozaeid@microchip.com\u003e\n Link: https://lore.kernel.org/r/20190809182510.22443-3-adham.abozaeid@microchip.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 0d1b57c1f04cad448eec54dc8de6da3777d3a28d\nAuthor: Adham Abozaeid \u003cadham.abozaeid@microchip.com\u003e\nDate: Fri Aug 9 18:25:18 2019 +0000\n\n staging: wilc1000: Don\u0027t reset WILC CPU disgracefully\n \n Send abort request to WILC from wilc_wlan_stop instead of resetting the\n CPU.\n The abort request was being sent from wilc_wlan_cleanup after the CPU\n was reset which wasn\u0027t the correct order. The abort request handler\n in the chip will take care of resetting the CPU.\n \n Signed-off-by: Adham Abozaeid \u003cadham.abozaeid@microchip.com\u003e\n Link: https://lore.kernel.org/r/20190809182510.22443-2-adham.abozaeid@microchip.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a3e2d4c7e76653d7e7bb34b6ca1692e3ca462d80\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 7 11:22:28 2019 +0800\n\n RDMA/hns: remove obsolete Kconfig comment\n \n Since commit a07fc0bb483e (\"RDMA/hns: Fix build error\")\n these kconfig comment is obsolete, so just remove it.\n \n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190807032228.6788-1-yuehaibing@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit d8d5cfac45dbbaacc4950b29c7b071547e5edc31\nAuthor: Kamal Heib \u003ckamalheib1@gmail.com\u003e\nDate: Wed Aug 7 13:31:38 2019 +0300\n\n RDMA/{cxgb3, cxgb4, i40iw}: Remove common code\n \n Now that we have a common iWARP query port function we can remove the\n common code from the iWARP drivers.\n \n Signed-off-by: Kamal Heib \u003ckamalheib1@gmail.com\u003e\n Acked-by: Potnuri Bharat Teja \u003cbharat@chelsio.com\u003e\n Link: https://lore.kernel.org/r/20190807103138.17219-5-kamalheib1@gmail.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 4929116bdf72995c0eebc74d48667262be1080ac\nAuthor: Kamal Heib \u003ckamalheib1@gmail.com\u003e\nDate: Wed Aug 7 13:31:37 2019 +0300\n\n RDMA/core: Add common iWARP query port\n \n Add support for a common iWARP query port function, the new function\n includes a common code that is used by the iWARP devices to update the\n port attributes like max_mtu, active_mtu, state, and phys_state, the\n function also includes a call for the driver-specific query_port callback\n to query the device-specific port attributes.\n \n Signed-off-by: Kamal Heib \u003ckamalheib1@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190807103138.17219-4-kamalheib1@gmail.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 691f380df242a6725cb4cc706887c293cfba6afc\nAuthor: Kamal Heib \u003ckamalheib1@gmail.com\u003e\nDate: Wed Aug 7 13:31:36 2019 +0300\n\n RDMA/cxgb3: Use ib_device_set_netdev()\n \n This change is required to associate the cxgb3 ib_dev with the\n underlying net_device, so in the upcoming patch we can call\n ib_device_get_netdev().\n \n Signed-off-by: Kamal Heib \u003ckamalheib1@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190807103138.17219-3-kamalheib1@gmail.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 72a7720fca37fec0daf295923f17ac5d88a613e1\nAuthor: Kamal Heib \u003ckamalheib1@gmail.com\u003e\nDate: Wed Aug 7 13:31:35 2019 +0300\n\n RDMA: Introduce ib_port_phys_state enum\n \n In order to improve readability, add ib_port_phys_state enum to replace\n the use of magic numbers.\n \n Signed-off-by: Kamal Heib \u003ckamalheib1@gmail.com\u003e\n Reviewed-by: Andrew Boyer \u003caboyer@tobark.org\u003e\n Acked-by: Michal Kalderon \u003cmichal.kalderon@marvell.com\u003e\n Acked-by: Bernard Metzler \u003cbmt@zurich.ibm.com\u003e\n Link: https://lore.kernel.org/r/20190807103138.17219-2-kamalheib1@gmail.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 12db36bc3cec76aae1c214d787eb0a54bc40fceb\nAuthor: Sean Paul \u003cseanpaul@chromium.org\u003e\nDate: Mon Aug 12 10:01:03 2019 -0400\n\n drm: Fix kerneldoc warns in connector-related docs\n \n Fixes the following warnings:\n ../drivers/gpu/drm/drm_connector.c:989: WARNING: Unexpected indentation.\n ../drivers/gpu/drm/drm_connector.c:993: WARNING: Unexpected indentation.\n ../include/drm/drm_connector.h:544: WARNING: Inline interpreted text or phrase reference start-string without end-string.\n ../include/drm/drm_connector.h:544: WARNING: Inline interpreted text or phrase reference start-string without end-string.\n \n Changes in v2:\n - Use () instead of \u0026 for functions (Sam)\n \n Fixes: 1b27fbdde1df (\"drm: Add drm_atomic_get_(old|new)_connector_for_encoder() helpers\")\n Fixes: bb5a45d40d50 (\"drm/hdcp: update content protection property with uevent\")\n Cc: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Pekka Paalanen \u003cpekka.paalanen@collabora.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Cc: Sean Paul \u003cseanpaul@chromium.org\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: dri-devel@lists.freedesktop.org\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812140112.6702-1-sean@poorly.run\n\ncommit e8760d8a698c3d61d6267f32acc2b845ac1de15e\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Tue Aug 6 14:46:14 2019 +0800\n\n clk: imx8mq: Unregister clks when of_clk_add_provider failed\n \n When of_clk_add_provider failed, all clks should be unregistered.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 73d9c8d4c0017e21e1ff519474ceb1450484dc9a\nAuthor: zhengbin \u003czhengbin13@huawei.com\u003e\nDate: Tue Jul 23 22:10:42 2019 +0800\n\n blk-mq: Fix memory leak in blk_mq_init_allocated_queue error handling\n \n If blk_mq_init_allocated_queue-\u003eelevator_init_mq fails, need to release\n the previously requested resources.\n \n Fixes: d34849913819 (\"blk-mq-sched: allow setting of default IO scheduler\")\n Signed-off-by: zhengbin \u003czhengbin13@huawei.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit ef7e6a1284374f22ca7016ae3f7c3dfccf7a3de2\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Tue Aug 6 14:46:13 2019 +0800\n\n clk: imx8mm: Unregister clks when of_clk_add_provider failed\n \n When of_clk_add_provider failed, all clks should be unregistered.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit b13431ed6eab808affbd796a7c8caf05c7ae4cdd\nAuthor: Yoshihiro Shimoda \u003cyoshihiro.shimoda.uh@renesas.com\u003e\nDate: Thu Aug 8 15:19:02 2019 +0900\n\n pinctrl: sh-pfc: Remove incomplete flag \"cfg-\u003etype\"\n \n The old commit c58d9c1b26e3 (\"sh-pfc: Implement generic pinconf\n support\") broke the cfg-\u003etype flag to PINMUX_TYPE_FUNCTION because\n sh_pfc_pinconf_set() didn\u0027t call sh_pfc_reconfig_pin().\n Now if we fix the cfg-\u003etype condition, it gets worse because:\n - Some drivers might be deferred so that .set_mux() will be called\n multiple times.\n - In such the case, the sh-pfc driver returns -EBUSY even if\n the group is the same, and then that driver fails to probe.\n \n Since the pinctrl subsystem already has such conditions according\n to @set_mux and @gpio_request_enable, this patch just remove\n the incomplete flag from sh-pfc/pinctrl.c.\n \n Fixes: c58d9c1b26e3 (\"sh-pfc: Implement generic pinconf support\")\n Signed-off-by: Yoshihiro Shimoda \u003cyoshihiro.shimoda.uh@renesas.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 1d4ba593d02e8ec41c2baaabdce9f62b71062dc7\nAuthor: Yoshihiro Shimoda \u003cyoshihiro.shimoda.uh@renesas.com\u003e\nDate: Thu Aug 8 15:19:01 2019 +0900\n\n pinctrl: sh-pfc: Add new flags into struct sh_pfc_pin_config\n \n To clean/modify the code up later, this patch just adds new flags\n \"mux_set\" and \"gpio_enabled\" into the struct sh_pfc_pin_config.\n \n Signed-off-by: Yoshihiro Shimoda \u003cyoshihiro.shimoda.uh@renesas.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit f32f77814925e8325cbca836328b2f467bc24996\nAuthor: Martin Kaiser \u003cmartin@kaiser.cx\u003e\nDate: Thu Aug 1 17:46:28 2019 +0200\n\n ARM: dts: imx25: mbimxsd25: native-mode is part of display-timings\n \n Move the native-mode property inside the display-timings node.\n \n According to\n Documentation/devicetree/bindings/display/panel/display-timing.txt.\n native-mode is a property of the display-timings node.\n \n If it\u0027s located outside of display-timings, the native-mode setting is\n ignored and the first display timing is used (which is a problem only if\n someone adds another display timing).\n \n Signed-off-by: Martin Kaiser \u003cmartin@kaiser.cx\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 7244c49dd36d3787a8867cce3b39e8db3403f7ef\nAuthor: Martin Kaiser \u003cmartin@kaiser.cx\u003e\nDate: Thu Aug 1 17:46:27 2019 +0200\n\n ARM: dts: apf27dev: native-mode is part of display-timings\n \n Move the native-mode property inside the display-timings node.\n \n According to\n Documentation/devicetree/bindings/display/panel/display-timing.txt.\n native-mode is a property of the display-timings node.\n \n If it\u0027s located outside of display-timings, the native-mode setting is\n ignored and the first display timing is used (which is a problem only if\n someone adds another display timing).\n \n Signed-off-by: Martin Kaiser \u003cmartin@kaiser.cx\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 001d4793734716b416bbb97ee3184f4d677b01c0\nAuthor: Martin Kaiser \u003cmartin@kaiser.cx\u003e\nDate: Thu Aug 1 17:46:26 2019 +0200\n\n ARM: dts: edb7211: native-mode is part of display-timings\n \n Move the native-mode property inside the display-timings node.\n \n According to\n Documentation/devicetree/bindings/display/panel/display-timing.txt.\n native-mode is a property of the display-timings node.\n \n If it\u0027s located outside of display-timings, the native-mode setting is\n ignored and the first display timing is used (which is a problem only if\n someone adds another display timing).\n \n Signed-off-by: Martin Kaiser \u003cmartin@kaiser.cx\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 558ea43a7d8347dedd6d585d1da4c38de0852018\nAuthor: Martin Kaiser \u003cmartin@kaiser.cx\u003e\nDate: Thu Aug 1 17:46:25 2019 +0200\n\n ARM: dts: imx27-phytec-phycore-rdk: native-mode is part of display-timings\n \n Move the native-mode property inside the display-timings node.\n \n According to\n Documentation/devicetree/bindings/display/panel/display-timing.txt.\n native-mode is a property of the display-timings node.\n \n If it\u0027s located outside of display-timings, the native-mode setting is\n ignored and the first display timing is used (which is a problem only if\n someone adds another display timing).\n \n Signed-off-by: Martin Kaiser \u003cmartin@kaiser.cx\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 38910680d5cb632013f056629e15706e26b1109a\nAuthor: Martin Kaiser \u003cmartin@kaiser.cx\u003e\nDate: Thu Aug 1 17:46:24 2019 +0200\n\n ARM: dts: mbimxsd25: native-mode is part of display-timings\n \n Move the native-mode property inside the display-timings node.\n \n According to\n Documentation/devicetree/bindings/display/panel/display-timing.txt.\n native-mode is a property of the display-timings node.\n \n If it\u0027s located outside of display-timings, the native-mode setting is\n ignored and the first display timing is used (which is a problem only if\n someone adds another display timing).\n \n Signed-off-by: Martin Kaiser \u003cmartin@kaiser.cx\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 6fbeef213c4189231b950ddaeeb3d3ce867d573b\nAuthor: Martin Kaiser \u003cmartin@kaiser.cx\u003e\nDate: Thu Aug 1 17:46:23 2019 +0200\n\n ARM: dts: eukrea-mbimxsd27: native-mode is part of display-timings\n \n Move the native-mode property inside the display-timings node.\n \n According to\n Documentation/devicetree/bindings/display/panel/display-timing.txt.\n native-mode is a property of the display-timings node.\n \n If it\u0027s located outside of display-timings, the native-mode setting is\n ignored and the first display timing is used (which is a problem only if\n someone adds another display timing).\n \n Signed-off-by: Martin Kaiser \u003cmartin@kaiser.cx\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 28992ae05cc64363623cb75ee3de95425690ede0\nAuthor: Martin Kaiser \u003cmartin@kaiser.cx\u003e\nDate: Thu Aug 1 17:46:22 2019 +0200\n\n ARM: dts: imx25: mbimxsd25: native-mode is part of display-timings\n \n Move the native-mode property inside the display-timings node.\n \n According to\n Documentation/devicetree/bindings/display/panel/display-timing.txt.\n native-mode is a property of the display-timings node.\n \n If it\u0027s located outside of display-timings, the native-mode setting is\n ignored and the first display timing is used (which is a problem only if\n someone adds another display timing).\n \n Signed-off-by: Martin Kaiser \u003cmartin@kaiser.cx\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit aae15b1453a22dde4a3623d01cfeaf5e39ac424f\nAuthor: Martin Kaiser \u003cmartin@kaiser.cx\u003e\nDate: Thu Aug 1 17:46:21 2019 +0200\n\n ARM: dts: imx27 phyCARD-S: native-mode is part of display-timings\n \n Move the native-mode property inside the display-timings node.\n \n According to\n Documentation/devicetree/bindings/display/panel/display-timing.txt.\n native-mode is a property of the display-timings node.\n \n If it\u0027s located outside of display-timings, the native-mode setting is\n ignored and the first display timing is used (which is a problem only if\n someone adds another display timing).\n \n Signed-off-by: Martin Kaiser \u003cmartin@kaiser.cx\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit ffbf23d50353915dc2622a3b7b4ddc678165f92d\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Thu Aug 1 12:56:36 2019 +0300\n\n firmware: imx: Add DSP IPC protocol interface\n \n Some of i.MX8 processors (e.g i.MX8QM, i.MX8QXP) contain\n the Tensilica HiFi4 DSP for advanced pre- and post-audio\n processing.\n \n The communication between Host CPU and DSP firmware is\n taking place using a shared memory area for message passing\n and a dedicated Messaging Unit for notifications.\n \n DSP IPC protocol offers a doorbell interface using\n imx-mailbox API.\n \n We use 4 MU channels (2 x TXDB, 2 x RXDB) to implement a\n request-reply protocol.\n \n Connection 0 (txdb0, rxdb0):\n - Host writes messasge to shared memory [SHMEM]\n - Host sends a request [MU]\n - DSP handles request [SHMEM]\n - DSP sends reply [MU]\n \n Connection 1 (txdb1, rxdb1):\n - DSP writes a message to shared memory [SHMEM]\n - DSP sends a request [MU]\n - Host handles request [SHMEM]\n - Host sends reply [MU]\n \n The protocol interface will be used by a Host client to\n communicate with the DSP. First client will be the i.MX8\n part from Sound Open Firmware infrastructure.\n \n The protocol offers the following interface:\n \n On Tx:\n - imx_dsp_ring_doorbell, will be called to notify the DSP\n that it needs to handle a request.\n \n On Rx:\n - clients need to provide two callbacks:\n .handle_reply\n .handle_request\n - the callbacks will be used by the protocol on\n notification arrival from DSP.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Reviewed-by: Oleksij Rempel \u003co.rempel@pengutronix.de\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 533ca1feed98b0bf024779a14760694c7cb4d431\nAuthor: Dexuan Cui \u003cdecui@microsoft.com\u003e\nDate: Fri Aug 2 22:50:20 2019 +0000\n\n PCI: hv: Avoid use of hv_pci_dev-\u003epci_slot after freeing it\n \n The slot must be removed before the pci_dev is removed, otherwise a panic\n can happen due to use-after-free.\n \n Fixes: 15becc2b56c6 (\"PCI: hv: Add hv_pci_remove_slots() when we unload the driver\")\n Signed-off-by: Dexuan Cui \u003cdecui@microsoft.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Cc: stable@vger.kernel.org\n\ncommit 6fa4e0cae684d268d309d1b1f929d52e3df5649c\nAuthor: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\nDate: Mon Aug 12 12:52:26 2019 +0300\n\n ASoC: pcm3168a: Allow reconfiguration of tdm_slots and slot_width\n \n When using right_j format and the codec is slave it can support 16bit\n format, but only if slot_width \u003d\u003d 16, in the same DAI mode the 24 bit\n audio can work with 24 or 32 slot_width.\n \n Because of this, the codec and CPU needs to be reconfigured when the sample\n format changes.\n \n Signed-off-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Link: https://lore.kernel.org/r/20190812095226.18870-3-peter.ujfalusi@ti.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit abe51c351827e0086dad079dfe02918fecdf4830\nAuthor: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\nDate: Mon Aug 12 12:52:25 2019 +0300\n\n ASoC: pcm3168a: Retain the independence of DAC and ADC side of the codec\n \n The DAC and ADC path of the codec is independent, have dedicated LRCK (FS)\n and BCK for DAC/ADC.\n \n They can be configured to use different format, TDM slots and slot_width if\n needed.\n \n Move these parameters under dedicated io_params structure and manage them\n independently based on the dai.\n \n Signed-off-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Link: https://lore.kernel.org/r/20190812095226.18870-2-peter.ujfalusi@ti.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 52718908c3f44b939a9934ef624e68457b21704f\nAuthor: Jarkko Nikula \u003cjarkko.nikula@linux.intel.com\u003e\nDate: Mon Aug 12 13:13:44 2019 +0300\n\n spi: dw-pci: Add support for Intel Elkhart Lake PSE SPI\n \n Add support for Intel(R) Programmable Services Engine (Intel(R) PSE) SPI\n controller in Intel Elkhart Lake when interface is assigned to the host\n processor.\n \n Signed-off-by: Jarkko Nikula \u003cjarkko.nikula@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190812101344.3975-1-jarkko.nikula@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit fba5b1e9ab527bd46bc529370d52a7b2b552dce1\nAuthor: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\nDate: Mon Aug 12 10:24:09 2019 +0100\n\n regmap-irq: Correct error paths in regmap_irq_thread for pm_runtime\n \n Some error paths in regmap_irq_thread put the pm_runtime others do not,\n there is no reason to leave the pm_runtime enabled in some cases so\n update those paths to also put the pm_runtime.\n \n Signed-off-by: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n Link: https://lore.kernel.org/r/20190812092409.21593-1-ckeepax@opensource.cirrus.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 79631210fc413546d0ed73632ff904ded5192cc9\nAuthor: Keyon Jie \u003cyang.jie@linux.intel.com\u003e\nDate: Fri Aug 9 18:22:36 2019 -0500\n\n ASoC: Intel: skl-hda-dsp-generic: add dmic dapm widget and route\n \n Adding DAPM MIC endpoint widget \"SoC DMIC\" and route, to enable\n DMIC DAPM support with hda generic machine.\n \n Signed-off-by: Keyon Jie \u003cyang.jie@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190809232236.21182-5-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 0181d2853126db0717de6bd5b3287f1c87faf7c3\nAuthor: Keyon Jie \u003cyang.jie@linux.intel.com\u003e\nDate: Fri Aug 9 18:22:35 2019 -0500\n\n ASoC: Intel: skl-hda-dsp-generic: add dependency to dmic driver\n \n The hda generic machine actually has dependency on the dmic driver,\n select SND_SOC_DMIC at the machine selected to fix it.\n \n Signed-off-by: Keyon Jie \u003cyang.jie@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190809232236.21182-4-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 1a3fb5d3f420d5f022768fcff091e5a9862f5877\nAuthor: Sathya Prakash M R \u003csathya.prakash.m.r@intel.com\u003e\nDate: Fri Aug 9 18:22:33 2019 -0500\n\n ASoC: Intel: boards: Match Product Family instead of product\n \n The generic machine driver of sof_rt5682\n supports more platforms of same product family.\n hence match the product family instead of product name.\n \n Signed-off-by: Sathya Prakash M R \u003csathya.prakash.m.r@intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190809232236.21182-2-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 872f3ac583cb00c14a4dc86b1511cf98618dd671\nAuthor: Szymon Mielczarek \u003cszymonx.mielczarek@linux.intel.com\u003e\nDate: Fri Aug 9 10:40:34 2019 +0200\n\n ASoC: dapm: Invalidate only paths reachable for a given stream\n \n By resetting the cached number of endpoints for all card\u0027s widgets we may\n overwrite previously cached values for other streams. The situation may\n happen especially when running streams simultaneously.\n \n Signed-off-by: Szymon Mielczarek \u003cszymonx.mielczarek@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190809084034.26220-1-szymonx.mielczarek@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit cac974a51ebb955f3be260a9633836e1413c8672\nAuthor: Jaska Uimonen \u003cjaska.uimonen@intel.com\u003e\nDate: Fri Aug 9 18:17:14 2019 -0500\n\n ASoC: SOF: topology: use set_get_data in process load\n \n Currently when loading sof process components there\u0027s a check if binary\n control data is associated with it. If found the data is extracted to be\n part of component loading and initialization. If binary data exceeds the\n ipc max size, loading fails with error as large message support is only\n implemented in set_get_data method. So make the process loading use\n set_get_data to enable large parameters in component initialization.\n \n Also refactor the process component loading function as it digs out 3\n times almost identical information of related controls. This is\n redundant, looks ugly and makes it difficult to understand the\n mechanism. So make a function out of fetching the control data and use\n it in process loading.\n \n Signed-off-by: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Signed-off-by: Jaska Uimonen \u003cjaska.uimonen@intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190809231714.20874-1-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9ea08f2a6d27b6a26d33dae5c58e4099672d6bb3\nAuthor: Shengjiu Wang \u003cshengjiu.wang@nxp.com\u003e\nDate: Fri Aug 9 18:27:47 2019 +0800\n\n ASoC: fsl_esai: Add new compatible string for imx6ull\n \n Add new compatible string \"fsl,imx6ull-esai\" in the binding document.\n \n Signed-off-by: Shengjiu Wang \u003cshengjiu.wang@nxp.com\u003e\n Acked-by: Nicolin Chen \u003cnicoleotsuka@gmail.com\u003e\n Link: https://lore.kernel.org/r/1565346467-5769-2-git-send-email-shengjiu.wang@nxp.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9c2806c4941641a6c75736f8c4303c89d2013cc4\nAuthor: Shengjiu Wang \u003cshengjiu.wang@nxp.com\u003e\nDate: Fri Aug 9 18:27:46 2019 +0800\n\n ASoC: fsl_esai: Add compatible string for imx6ull\n \n Add compatible string for imx6ull, from imx6ull platform,\n the issue of channel swap after xrun is fixed in hardware.\n \n Signed-off-by: Shengjiu Wang \u003cshengjiu.wang@nxp.com\u003e\n Acked-by: Nicolin Chen \u003cnicoleotsuka@gmail.com\u003e\n Link: https://lore.kernel.org/r/1565346467-5769-1-git-send-email-shengjiu.wang@nxp.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit ae032156ec4142b7a1d644eba652806785de1777\nAuthor: Curtis Malainey \u003ccujomalainey@chromium.org\u003e\nDate: Fri Aug 9 14:59:51 2019 -0700\n\n ASoC: rt5677: Revert remove superfluous set\n \n Commit 4dc057a786dc23 (\"ASoC: rt5677: remove superfluous set\") was an\n attempted code cleanup but was incorrectly tested before sent and\n actually breaks the interrupt since it never resets the value on each\n loop now. The breakage is most testable when hotwording code is added\n and also uses the interrupt.\n \n Signed-off-by: Curtis Malainey \u003ccujomalainey@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190809215952.155660-1-cujomalainey@chromium.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 8b209cdb696f49f7f5d6c1922bed435e54631ddd\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 22:27:36 2019 +0200\n\n ASoC: remove w90x900/nuc900 platform drivers\n \n The ARM w90x900 platform is getting removed, so this driver is obsolete.\n \n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Acked-by: Mark Brown \u003cbroonie@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190809202749.742267-9-arnd@arndb.de\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 8afd1a99d713ccc7cf1bd509b205c3bb60cf0b90\nAuthor: Keyon Jie \u003cyang.jie@linux.intel.com\u003e\nDate: Fri Aug 9 18:22:34 2019 -0500\n\n ASoC: Intel: sof-rt5682: add dmic dapm widget to support dmic PCM\n \n We need add DAPM MIC endpoint widget \"SoC DMIC\" and route, to enable\n DMIC PCM DAPM support.\n \n Signed-off-by: Keyon Jie \u003cyang.jie@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190809232236.21182-3-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit b9ee5e04fd77898208c51b1395fa0b5e8536f9b6\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Thu Aug 8 12:48:26 2019 +0000\n\n powerpc/64e: Drop stale call to smp_processor_id() which hangs SMP startup\n \n Commit ebb9d30a6a74 (\"powerpc/mm: any thread in one core can be the\n first to setup TLB1\") removed the need to know the cpu_id in\n early_init_this_mmu(), but the call to smp_processor_id() which was\n marked __maybe_used remained.\n \n Since commit ed1cd6deb013 (\"powerpc: Activate CONFIG_THREAD_INFO_IN_TASK\")\n thread_info cannot be reached before MMU is properly set up.\n \n Drop this stale call to smp_processor_id() which makes SMP hang when\n CONFIG_PREEMPT is set.\n \n Fixes: ebb9d30a6a74 (\"powerpc/mm: any thread in one core can be the first to setup TLB1\")\n Fixes: ed1cd6deb013 (\"powerpc: Activate CONFIG_THREAD_INFO_IN_TASK\")\n Cc: stable@vger.kernel.org # v5.1+\n Reported-by: Chris Packham \u003cChris.Packham@alliedtelesis.co.nz\u003e\n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Tested-by: Chris Packham \u003cchris.packham@alliedtelesis.co.nz\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/bef479514f4c08329fa649f67735df8918bc0976.1565268248.git.christophe.leroy@c-s.fr\n\ncommit 2e1da13fba4cb529c2c8c1d9f657690d1e853d7d\nAuthor: Vlastimil Babka \u003cvbabka@suse.cz\u003e\nDate: Wed Aug 7 15:02:58 2019 +0200\n\n x86/kconfig: Remove X86_DIRECT_GBPAGES dependency on !DEBUG_PAGEALLOC\n \n These days CONFIG_DEBUG_PAGEALLOC just compiles in the code that has to be\n enabled on boot time, or with an extra config option, and only then are the\n large page based direct mappings disabled.\n \n Therefore remove the config dependency, allowing 1GB direct mappings with\n debug_pagealloc compiled in but not enabled.\n \n Signed-off-by: Vlastimil Babka \u003cvbabka@suse.cz\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190807130258.22185-1-vbabka@suse.cz\n\ncommit a46d14eca7b75fffe35603aa8b81df654353d80f\nAuthor: Phil Auld \u003cpauld@redhat.com\u003e\nDate: Thu Aug 1 09:37:49 2019 -0400\n\n sched/fair: Use rq_lock/unlock in online_fair_sched_group\n \n Enabling WARN_DOUBLE_CLOCK in /sys/kernel/debug/sched_features causes\n warning to fire in update_rq_clock. This seems to be caused by onlining\n a new fair sched group not using the rq lock wrappers.\n \n [] rq-\u003eclock_update_flags \u0026 RQCF_UPDATED\n [] WARNING: CPU: 5 PID: 54385 at kernel/sched/core.c:210 update_rq_clock+0xec/0x150\n \n [] Call Trace:\n [] online_fair_sched_group+0x53/0x100\n [] cpu_cgroup_css_online+0x16/0x20\n [] online_css+0x1c/0x60\n [] cgroup_apply_control_enable+0x231/0x3b0\n [] cgroup_mkdir+0x41b/0x530\n [] kernfs_iop_mkdir+0x61/0xa0\n [] vfs_mkdir+0x108/0x1a0\n [] do_mkdirat+0x77/0xe0\n [] do_syscall_64+0x55/0x1d0\n [] entry_SYSCALL_64_after_hwframe+0x44/0xa9\n \n Using the wrappers in online_fair_sched_group instead of the raw locking\n removes this warning.\n \n [ tglx: Use rq_*lock_irq() ]\n \n Signed-off-by: Phil Auld \u003cpauld@redhat.com\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: Vincent Guittot \u003cvincent.guittot@linaro.org\u003e\n Cc: Ingo Molnar \u003cmingo@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190801133749.11033-1-pauld@redhat.com\n\ncommit 73feb4d0f8f14c5102bd46ce2255ff55d3d52db7\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Tue Jul 2 15:45:45 2019 +0800\n\n soc: imx-scu: Add SoC UID(unique identifier) support\n \n Add i.MX SCU SoC\u0027s UID(unique identifier) support, user\n can read it from sysfs:\n \n root@imx8qxpmek:~# cat /sys/devices/soc0/soc_uid\n 7B64280B57AC1898\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit d3b682c49e2223f86553c9c10f7fbba073d28703\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jun 19 10:21:42 2019 +0800\n\n dt-bindings: arm: imx: Add the soc binding for i.MX8MN\n \n This patch adds the soc \u0026 board binding for i.MX8MN.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 4ecd20c929b86e87ee4a74379678d1c0ef90c68d\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 12 10:10:44 2019 +0100\n\n drm/i915/gt: Use the local engine wakeref when checking RING registers\n \n Now that we can atomically acquire the engine wakeref, make use of it\n when check whether the RING registers are idle.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812091045.29587-7-chris@chris-wilson.co.uk\n\ncommit acb9488dcad866987406aefc4f8298f2bf971575\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 12 10:10:39 2019 +0100\n\n drm/i915/selftests: Prevent the timeslice expiring during suppression tests\n \n When testing whether we prevent suppressing preemption, it helps to\n avoid a time slice expiring prematurely.\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111108\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812091045.29587-2-chris@chris-wilson.co.uk\n\ncommit f597625d12ba3528658faf5b0f0aae657430a88a\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 12 10:10:38 2019 +0100\n\n drm/i915/execlists: Avoid sync calls during park\n \n Since we allow ourselves to use non-process context during parking, we\n cannot allow ourselves to sleep and in particular cannot call\n del_timer_sync() -- but we can use a plain del_timer().\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111375\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812091045.29587-1-chris@chris-wilson.co.uk\n\ncommit 3ea5802910545d11180630d41dda0a81c91539f4\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Mon Aug 12 09:29:35 2019 +0000\n\n drm/i915/uc: Update copyright and license\n \n Include 2019 in copyright years and start using SPDX tag.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812092935.21048-1-michal.wajdeczko@intel.com\n\ncommit bbd1b70639f785a970d998f35155c713f975e3ac\nAuthor: Jeremy Linton \u003cjeremy.linton@arm.com\u003e\nDate: Thu Aug 8 15:40:06 2019 -0500\n\n ACPI/PPTT: Add support for ACPI 6.3 thread flag\n \n ACPI 6.3 adds a flag to the CPU node to indicate whether\n the given PE is a thread. Add a function to return that\n information for a given linux logical CPU.\n \n Signed-off-by: Jeremy Linton \u003cjeremy.linton@arm.com\u003e\n Reviewed-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Reviewed-by: Robert Richter \u003crrichter@marvell.com\u003e\n Acked-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit a4872e80ce7d2a1844328176dbf279d0a2b89bdb\nAuthor: Hsin-Hsiung Wang \u003chsin-hsiung.wang@mediatek.com\u003e\nDate: Mon Aug 5 13:21:50 2019 +0800\n\n mfd: mt6397: Extract IRQ related code from core driver\n \n In order to support different types of irq design, we decide to add\n separate irq drivers for different design and keep mt6397 mfd core\n simple and reusable to all generations of PMICs so far.\n \n Signed-off-by: Hsin-Hsiung Wang \u003chsin-hsiung.wang@mediatek.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit c550b60c647941b581e4b0f4f2b1526b34f4a8e5\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Sun Aug 11 19:08:01 2019 -0500\n\n drm/komeda: Fix potential integer overflow in komeda_crtc_update_clock_ratio\n \n Add suffix ULL to constant 1000 in order to avoid a potential integer\n overflow and give the compiler complete information about the proper\n arithmetic to use. Notice that this constant is being used in a context\n that expects an expression of type u64, but it\u0027s currently evaluated\n using 32-bit arithmetic.\n \n Addresses-Coverity-ID: 1485796 (\"Unintentional integer overflow\")\n Fixes: ed22c6d9304d (\"drm/komeda: Use drm_display_mode \"crtc_\" prefixed hardware timings\")\n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Reviewed-by: James Qian Wang (Arm Technology China) \u003cjames.qian.wang@arm.com\u003e\n Signed-off-by: james qian wang (Arm Technology China) \u003cjames.qian.wang@arm.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812000801.GA29204@embeddedor\n\ncommit 708cb5cc3fde63a33afa0aaf7a26fd7ea9015dd3\nAuthor: Hsin-Hsiung Wang \u003chsin-hsiung.wang@mediatek.com\u003e\nDate: Mon Aug 5 13:21:49 2019 +0800\n\n mfd: mt6397: Rename macros to something more readable\n \n Signed-off-by: Hsin-Hsiung Wang \u003chsin-hsiung.wang@mediatek.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit c8ae9c2da1cc5d18b6d51d10160508a3dc3436bf\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Mon Jul 8 09:41:08 2019 +0100\n\n reset: Add support for resets provided by SCMI\n \n On some ARM based systems, a separate Cortex-M based System Control\n Processor(SCP) provides the overall power, clock, reset and system\n control. System Control and Management Interface(SCMI) Message Protocol\n is defined for the communication between the Application Cores(AP)\n and the SCP.\n \n Adds support for the resets provided using SCMI protocol for performing\n reset management of various devices present on the SoC. Various reset\n functionalities are achieved by the means of different ARM SCMI device\n operations provided by the ARM SCMI framework.\n \n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit 95a15d80aa0de938299acfcbc6aa6f2b16f5d7e5\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Mon Jul 8 09:41:06 2019 +0100\n\n firmware: arm_scmi: Add RESET protocol in SCMI v2.0\n \n SCMIv2.0 adds a new Reset Management Protocol to manage various reset\n states a given device or domain can enter. Device(s) that can be\n collectively reset through a common reset signal constitute a reset\n domain for the firmware.\n \n A reset domain can be reset autonomously or explicitly through assertion\n and de-assertion of the signal. When autonomous reset is chosen, the\n firmware is responsible for taking the necessary steps to reset the\n domain and to subsequently bring it out of reset. When explicit reset is\n chosen, the caller has to specifically assert and then de-assert the\n reset signal by issuing two separate RESET commands.\n \n Add the basic SCMI reset infrastructure that can be used by Linux\n reset controller driver.\n \n Reviewed-by: Peng Fan \u003cpeng.fan@nxp.com\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit ae39913ce4095682d1a6e5734a5558062a59d29f\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Mon Jul 8 09:41:04 2019 +0100\n\n dt-bindings: arm: Extend SCMI to support new reset protocol\n \n SCMIv2.0 adds a new Reset Management Protocol to manage various reset\n states a given device or domain can enter. Extend the existing SCMI\n bindings to add reset protocol support by re-using the reset bindings\n for both reset providers and consumers.\n \n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: devicetree@vger.kernel.org\n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit 823839571d76d465c6e454495864b085d699adb5\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Mon Jul 8 09:41:17 2019 +0100\n\n firmware: arm_scmi: Make use SCMI v2.0 fastchannel for performance protocol\n \n SCMI v2.0 adds support for \"FastChannel\" which do not use a message\n header as they are specialized for a single message.\n \n Only PERFORMANCE_LIMITS_{SET,GET} and PERFORMANCE_LEVEL_{SET,GET}\n commands are supported over fastchannels. As they are optional, they\n need to be discovered by PERFORMANCE_DESCRIBE_FASTCHANNEL command.\n Further {LIMIT,LEVEL}_SET commands can have optional doorbell support.\n \n Add support for making use of these fastchannels.\n \n Cc: Ionela Voinescu \u003cIonela.Voinescu@arm.com\u003e\n Cc: Chris Redpath \u003cChris.Redpath@arm.com\u003e\n Cc: Quentin Perret \u003cQuentin.Perret@arm.com\u003e\n Reviewed-by: Peng Fan \u003cpeng.fan@nxp.com\u003e\n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit ac8aaf348cf54a07aff8e709329ef82ecfa230cc\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Mon Jul 8 09:41:12 2019 +0100\n\n firmware: arm_scmi: Add discovery of SCMI v2.0 performance fastchannels\n \n SCMI v2.0 adds support for \"FastChannel\", a lightweight unidirectional\n channel that is dedicated to a single SCMI message type for controlling\n a specific platform resource. They do not use a message header as they\n are specialized for a single message.\n \n Only PERFORMANCE_LIMITS_{SET,GET} and PERFORMANCE_LEVEL_{SET,GET}\n commands are supported over fastchannels. As they are optional, they\n need to be discovered by PERFORMANCE_DESCRIBE_FASTCHANNEL command.\n Further {LIMIT,LEVEL}_SET commands can have optional doorbell support.\n \n Add support for discovery of these fastchannels.\n \n Cc: Ionela Voinescu \u003cIonela.Voinescu@arm.com\u003e\n Cc: Chris Redpath \u003cChris.Redpath@arm.com\u003e\n Cc: Quentin Perret \u003cQuentin.Perret@arm.com\u003e\n Reviewed-by: Peng Fan \u003cpeng.fan@nxp.com\u003e\n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit aa90ac45bc88e6a8efb199e45dab5d5a6b01c3d2\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Wed Aug 7 13:46:27 2019 +0100\n\n firmware: arm_scmi: Use {get,put}_unaligned_le{32,64} accessors\n \n Instead of type-casting the {tx,rx}.buf all over the place while\n accessing them to read/write __le{32,64} from/to the firmware, let\u0027s\n use the existing {get,put}_unaligned_le{32,64} accessors to hide all\n the type cast ugliness.\n \n Suggested-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit 2bc06ffa0635e99cd4d88d6f6fe5bb4e4b9e50f1\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Mon Jul 8 14:42:16 2019 +0100\n\n firmware: arm_scmi: Use asynchronous CLOCK_RATE_SET when possible\n \n CLOCK_PROTOCOL_ATTRIBUTES provides attributes to indicate the maximum\n number of pending asynchronous clock rate changes supported by the\n platform. If it\u0027s non-zero, then we should be able to use asynchronous\n clock rate set for any clocks until the maximum limit is reached.\n \n Tracking the current count of pending asynchronous clock set rate\n requests, we can decide if the incoming/new request for clock set rate\n can be handled asynchronously or not until the maximum limit is\n reached.\n \n Cc: linux-clk@vger.kernel.org\n Reviewed-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit d0aba11614552d43a61c3acdf36876b00060286e\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Mon Jul 8 09:42:22 2019 +0100\n\n firmware: arm_scmi: Drop config flag in clk_ops-\u003erate_set\n \n CLOCK_PROTOCOL_ATTRIBUTES provides attributes to indicate the maximum\n number of pending asynchronous clock rate changes supported by the\n platform. If it\u0027s non-zero, then we should be able to use asynchronous\n clock rate set for any clocks until the maximum limit is reached.\n \n In order to add that support, let\u0027s drop the config flag passed to\n clk_ops-\u003erate_set and handle the asynchronous requests dynamically.\n \n Cc: Stephen Boyd \u003csboyd@kernel.org\u003e\n Cc: linux-clk@vger.kernel.org\n Acked-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit d09aac0eb17c6ce2b66095e1e324f60ec9dd8988\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Mon Jul 8 09:41:01 2019 +0100\n\n firmware: arm_scmi: Add asynchronous sensor read if it supports\n \n SENSOR_DESCRIPTION_GET provides attributes to indicate if the sensor\n supports asynchronous read. We can read that flag and use asynchronous\n reads for any sensors with that attribute set.\n \n Let\u0027s use the new scmi_do_xfer_with_response to support asynchronous\n sensor reads.\n \n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit 6a55331c87d86a7406d8126ae75bdd07244a91b1\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Mon Jul 8 09:40:57 2019 +0100\n\n firmware: arm_scmi: Drop async flag in sensor_ops-\u003ereading_get\n \n SENSOR_DESCRIPTION_GET provides attributes to indicate if the sensor\n supports asynchronous read. Ideally we should be able to read that flag\n and use asynchronous reads for any sensors with that attribute set.\n \n In order to add that support, let\u0027s drop the async flag passed to\n sensor_ops-\u003ereading_get and dynamically switch between sync and async\n flags based on the attributes as provided by the firmware.\n \n Cc: linux-hwmon@vger.kernel.org\n Acked-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit 58ecdf03dbb9ce8f130bc1ac63fcd8f4d72f9c06\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Mon Jul 8 09:40:54 2019 +0100\n\n firmware: arm_scmi: Add support for asynchronous commands and delayed response\n \n Messages that are sent to platform, also known as commands and can be:\n \n 1. Synchronous commands that block the channel until the requested work\n has been completed. The platform responds to these commands over the\n same channel and hence can\u0027t be used to send another command until the\n previous command has completed.\n \n 2. Asynchronous commands on the other hand, the platform schedules the\n requested work to complete later in time and returns almost immediately\n freeing the channel for new commands. The response indicates the success\n or failure in the ability to schedule the requested work. When the work\n has completed, the platform sends an additional delayed response message.\n \n Using the same transmit buffer used for sending the asynchronous command\n even for the delayed response corresponding to it simplifies handling of\n the delayed response. It\u0027s the caller of asynchronous command that is\n responsible for allocating the completion flag that scmi driver can\n complete to indicate the arrival of delayed response.\n \n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit 22d1f76109f74b716a5f4d46bade61f544f40f81\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Mon Jul 8 09:40:52 2019 +0100\n\n firmware: arm_scmi: Add mechanism to unpack message headers\n \n In order to identify the message type when a response arrives, we need\n a mechanism to unpack the message header similar to packing. Let\u0027s\n add one.\n \n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit 38c927fbebb3d69c561ae77600f64bb1c308d0c2\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Mon Jul 8 09:40:48 2019 +0100\n\n firmware: arm_scmi: Separate out tx buffer handling and prepare to add rx\n \n Currently we pre-allocate transmit buffers only and use the first free\n slot in that pre-allocated buffer for transmitting any new message that\n are generally originated from OS to the platform firmware.\n \n Notifications or the delayed responses on the other hand are originated\n from the platform firmware and consumes by the OS. It\u0027s better to have\n separate and dedicated pre-allocated buffers to handle the notifications.\n We can still use the transmit buffers for the delayed responses.\n \n In addition, let\u0027s prepare existing scmi_xfer_{get,put} for acquiring\n and releasing a slot to identify the right(tx/rx) buffers.\n \n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit 46cc7c286ce33f729055f28c114d84e6340c08dd\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Mon Jul 8 09:40:46 2019 +0100\n\n firmware: arm_scmi: Add receive channel support for notifications\n \n With scmi_mbox_chan_setup enabled to identify and setup both Tx and Rx,\n let\u0027s consolidate setting up of both the channels under the function\n scmi_mbox_txrx_setup.\n \n Since some platforms may opt not to support notifications or delayed\n response, they may not need support for Rx. Hence Rx is optional and\n failure of setting one up is not considered fatal.\n \n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit 3748daf7fb6b5459c7a62d4976b9698b82ee9b48\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Mon Jul 8 09:40:44 2019 +0100\n\n firmware: arm_scmi: Segregate tx channel handling and prepare to add rx\n \n The transmit(Tx) channels are specified as the first entry and the\n receive(Rx) channels are the second entry as per the device tree\n bindings. Since we currently just support Tx, index 0 is hardcoded at\n all required callsites.\n \n In order to prepare for adding Rx support, let\u0027s remove those hardcoded\n index and add boolean parameter to identify Tx/Rx channels when setting\n them up.\n \n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit 2747a967c85d1745c497599c632f1319a58aed5f\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Mon Jul 8 09:40:42 2019 +0100\n\n firmware: arm_scmi: Reorder some functions to avoid forward declarations\n \n Re-shuffling few functions to keep definitions and their usages close.\n This is also needed to avoid too many unnecessary forward declarations\n while adding new features(delayed response and notifications).\n \n Keeping this separate to avoid mixing up of these trivial change that\n doesn\u0027t affect functionality into the ones that does.\n \n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit 9dc34d635c67e57051853855c43249408641a5ab\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Mon Jul 8 15:48:36 2019 +0100\n\n firmware: arm_scmi: Check if platform has released shmem before using\n \n Sometimes platfom may take too long to respond to the command and OS\n might timeout before platform transfer the ownership of the shared\n memory region to the OS with the response.\n \n Since the mailbox channel associated with the channel is freed and new\n commands are dispatch on the same channel, OS needs to wait until it\n gets back the ownership. If not, either OS may end up overwriting the\n platform response for the last command(which is fine as OS timed out\n that command) or platform might overwrite the payload for the next\n command with the response for the old.\n \n The latter is problematic as platform may end up interpretting the\n response as the payload. In order to avoid such race, let\u0027s wait until\n the OS gets back the ownership before we prepare the shared memory with\n the payload for the next command.\n \n Reported-by: Jim Quinlan \u003cjames.quinlan@broadcom.com\u003e\n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit 5b65af8f60f580188c0be55b25cd45908024b985\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Mon Jul 8 09:40:40 2019 +0100\n\n firmware: arm_scmi: Use the term \u0027message\u0027 instead of \u0027command\u0027\n \n In preparation to adding support for other two types of messages that\n SCMI specification mentions, let\u0027s replace the term \u0027command\u0027 with the\n correct term \u0027message\u0027.\n \n As per the specification the messages are of 3 types:\n commands(synchronous or asynchronous), delayed responses and notifications.\n \n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit c29a628976b39ecf2f9ced376ceeef8e9f6ccde0\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Mon Jul 8 09:40:38 2019 +0100\n\n firmware: arm_scmi: Fix few trivial typos in comments\n \n While adding new comments found couple of typos that are better fixed.\n \n s/informfation/information/\n s/statues/status/\n \n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit 37bbffcb19a7c98ed0a8c0b2dd2e11bfc05d5964\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Mon Jul 8 09:40:35 2019 +0100\n\n firmware: arm_scmi: Remove extra check for invalid length message responses\n \n scmi_xfer_get_init ensures both transmit and receive buffer lengths are\n within the maximum limits. If receive buffer length is not supplied by\n the caller, it\u0027s set to the maximum limit value. Receive buffer length\n is never modified after that. So there\u0027s no need for the extra check\n when receive transmit completion for a command essage.\n \n Further, if the response header length is greater than the prescribed\n receive buffer length, the response buffer is truncated to the latter.\n \n Reported-by: Jim Quinlan \u003cjames.quinlan@broadcom.com\u003e\n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit 9eefa43a1a03960c7458e1463132f893702741be\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Mon Jul 8 09:40:33 2019 +0100\n\n firmware: arm_scmi: Align few names in sensors protocol with SCMI specification\n \n Looks like more code developed during the draft versions of the\n specification slipped through and they don\u0027t match the final\n released version. This seem to have happened only with sensor\n protocol.\n \n Renaming few command and function names here to match exactly with\n the released version of SCMI specification for ease of maintenance.\n \n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit 4752544a447b52b9949eb648a3b7719357853f91\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Tue Jul 23 14:00:09 2019 +0100\n\n firmware: arm_scmi: Use the correct style for SPDX License Identifier\n \n Fix to correct the SPDX License Identifier style in header file related\n to firmware frivers for ARM SCMI message protocol.\n \n For C header files Documentation/process/license-rules.rst mandates\n C-like comments(opposed to C source files where C++ style should be\n used).\n \n While at it, change GPL-2.0 to GPL-2.0-only similar to the ones in\n psci.h and scpi_protocol.h\n \n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit 194108705c9ec9a96efb1da69c0d0b932e1bf11a\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 22:27:34 2019 +0200\n\n watchdog: remove w90x900 driver\n \n The ARM w90x900 platform is getting removed, so this driver is obsolete\n \n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Acked-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Link: https://lore.kernel.org/r/20190809202749.742267-7-arnd@arndb.de\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Wim Van Sebroeck \u003cwim@linux-watchdog.org\u003e\n\ncommit 4d4ad783c387030aae7539e5812e3f3ba80a9d06\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 22:27:32 2019 +0200\n\n watchdog: remove ks8695 driver\n \n The platform is getting removed, so there are no remaining\n users of this driver.\n \n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Acked-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Link: https://lore.kernel.org/r/20190809202749.742267-5-arnd@arndb.de\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Wim Van Sebroeck \u003cwim@linux-watchdog.org\u003e\n\ncommit 819ce4a99596d850f424d95b83d8653e9fa3686a\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 16:40:29 2019 +0200\n\n watchdog: pnx4008_wdt: allow compile-testing\n \n The only thing that prevents building this driver on other\n platforms is the mach/hardware.h include, which is not actually\n used here at all, so remove the line and allow CONFIG_COMPILE_TEST.\n \n Acked-by: Sylvain Lemieux \u003cslemieux.tyco@gmail.com\u003e\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Link: https://lore.kernel.org/r/20190809144043.476786-4-arnd@arndb.de\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Wim Van Sebroeck \u003cwim@linux-watchdog.org\u003e\n\ncommit b196f0cfac93cdef0fac237b0e7ad74fa42ec6f7\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Wed Jul 31 10:42:52 2019 -0700\n\n watchdog: ziirave_wdt: Log bootloader/firmware info during probe\n \n Log bootloader/firmware info during probe. This information is\n available via sysfs already, but it\u0027s really helpful to have this in\n kernel log during startup as well.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Cc: Rick Ramstetter \u003crick@anteaterllc.com\u003e\n Cc: linux-watchdog@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Link: https://lore.kernel.org/r/20190731174252.18041-6-andrew.smirnov@gmail.com\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Wim Van Sebroeck \u003cwim@linux-watchdog.org\u003e\n\ncommit 79da9fef7d04383695c5916264292933bfb8993e\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Wed Jul 31 10:42:51 2019 -0700\n\n watchdog: ziirave_wdt: Don\u0027t bail out on unexpected timeout value\n \n Reprogramming bootloader on watchdog MCU will result in reported\n default timeout value of \"0\". That in turn will be unnecessarily\n rejected by the driver as invalid device (-ENODEV). Simplify probe to\n just read stored timeout value, clamp it to an acceptable range and\n program the value unconditionally to fix the above.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Cc: Rick Ramstetter \u003crick@anteaterllc.com\u003e\n Cc: linux-watchdog@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Link: https://lore.kernel.org/r/20190731174252.18041-5-andrew.smirnov@gmail.com\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Wim Van Sebroeck \u003cwim@linux-watchdog.org\u003e\n\ncommit a5d693f9023aab8a326bd86be8e9deb017d4c6b6\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Wed Jul 31 10:42:50 2019 -0700\n\n watchdog: ziirave_wdt: Be more verbose during firmware update\n \n Add more error logging to ziirave_firm_upload() for diagnostics.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Cc: Rick Ramstetter \u003crick@anteaterllc.com\u003e\n Cc: linux-watchdog@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Link: https://lore.kernel.org/r/20190731174252.18041-4-andrew.smirnov@gmail.com\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Wim Van Sebroeck \u003cwim@linux-watchdog.org\u003e\n\ncommit 97c590d164dd6f426bd8eef8970cc60afdc5e13d\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Wed Jul 31 10:42:49 2019 -0700\n\n watchdog: ziirave_wdt: Be verbose about errors in probe()\n \n The driver is quite silent in case of probe failure, which makes it\n more difficult to diagnose problem from the kernel log. Add logging to\n all of the silent error paths ziirave_wdt_probe() to improve that.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Cc: Rick Ramstetter \u003crick@anteaterllc.com\u003e\n Cc: linux-watchdog@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Link: https://lore.kernel.org/r/20190731174252.18041-3-andrew.smirnov@gmail.com\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Wim Van Sebroeck \u003cwim@linux-watchdog.org\u003e\n\ncommit ba7cc81c9ffe206419915eefc51527104661ca7c\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Wed Jul 31 10:42:48 2019 -0700\n\n watchdog: ziirave_wdt: Add missing newline\n \n Add missing newline.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Cc: Rick Ramstetter \u003crick@anteaterllc.com\u003e\n Cc: linux-watchdog@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Link: https://lore.kernel.org/r/20190731174252.18041-2-andrew.smirnov@gmail.com\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Wim Van Sebroeck \u003cwim@linux-watchdog.org\u003e\n\ncommit b18f22d02ad1a4cfc0ca348766da651f9119cf44\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Tue Aug 6 02:39:53 2019 -0500\n\n watchdog: jz4740: Fix unused variable warning in jz4740_wdt_probe\n \n Fix the following warning (Building: ci20_defconfig mips):\n \n drivers/watchdog/jz4740_wdt.c: In function ‘jz4740_wdt_probe’:\n drivers/watchdog/jz4740_wdt.c:165:6: warning: unused variable ‘ret’ [-Wunused-variable]\n int ret;\n ^~~\n Fixes: 9ee644c9326c (\"watchdog: jz4740_wdt: drop warning after registering device\")\n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Link: https://lore.kernel.org/r/20190806073953.GA13685@embeddedor\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Wim Van Sebroeck \u003cwim@linux-watchdog.org\u003e\n\ncommit 26ae6a8e9b09a94a38e5351502bee0e801d4d8c6\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:48 2019 -0700\n\n watchdog: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Wim Van Sebroeck \u003cwim@linux-watchdog.org\u003e\n Cc: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Cc: linux-watchdog@vger.kernel.org\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Wim Van Sebroeck \u003cwim@linux-watchdog.org\u003e\n\ncommit 4b4b8b03458ef33dc2545c038d3134ba668038a4\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Jul 29 20:46:50 2019 -0500\n\n watchdog: riowd: Mark expected switch fall-through\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warnings (Building: sparc64):\n \n drivers/watchdog/riowd.c: In function ‘riowd_ioctl’:\n drivers/watchdog/riowd.c:136:3: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n riowd_writereg(p, riowd_timeout, WDTO_INDEX);\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/watchdog/riowd.c:139:2: note: here\n case WDIOC_GETTIMEOUT:\n ^~~~\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Link: https://lore.kernel.org/r/20190730014650.GA31309@embeddedor\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Wim Van Sebroeck \u003cwim@linux-watchdog.org\u003e\n\ncommit 2c017640826a16eab385b756431ce560e44a7054\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Jul 29 17:31:59 2019 -0500\n\n watchdog: wdt977: Mark expected switch fall-through\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warning (Building: arm):\n \n drivers/watchdog/wdt977.c: In function ‘wdt977_ioctl’:\n LD [M] drivers/media/platform/vicodec/vicodec.o\n drivers/watchdog/wdt977.c:400:3: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n wdt977_keepalive();\n ^~~~~~~~~~~~~~~~~~\n drivers/watchdog/wdt977.c:403:2: note: here\n case WDIOC_GETTIMEOUT:\n ^~~~\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Link: https://lore.kernel.org/r/20190729223159.GA20878@embeddedor\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Wim Van Sebroeck \u003cwim@linux-watchdog.org\u003e\n\ncommit ca58397c53ddba9544748dc0cbaef4ff34282466\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Jul 29 15:06:02 2019 -0500\n\n watchdog: scx200_wdt: Mark expected switch fall-through\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warning (Building: i386):\n \n drivers/watchdog/scx200_wdt.c: In function ‘scx200_wdt_ioctl’:\n drivers/watchdog/scx200_wdt.c:188:3: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n scx200_wdt_ping();\n ^~~~~~~~~~~~~~~~~\n drivers/watchdog/scx200_wdt.c:189:2: note: here\n case WDIOC_GETTIMEOUT:\n ^~~~\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Link: https://lore.kernel.org/r/20190729200602.GA6854@embeddedor\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Wim Van Sebroeck \u003cwim@linux-watchdog.org\u003e\n\ncommit 630ee1a50c40210ef021435faa0e80d17a904883\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Jul 29 10:10:33 2019 -0500\n\n watchdog: Mark expected switch fall-throughs\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warnings:\n \n drivers/watchdog/ar7_wdt.c: warning: this statement may fall\n through [-Wimplicit-fallthrough\u003d]: \u003d\u003e 237:3\n drivers/watchdog/pcwd.c: warning: this statement may fall\n through [-Wimplicit-fallthrough\u003d]: \u003d\u003e 653:3\n drivers/watchdog/sb_wdog.c: warning: this statement may fall\n through [-Wimplicit-fallthrough\u003d]: \u003d\u003e 204:3\n drivers/watchdog/wdt.c: warning: this statement may fall\n through [-Wimplicit-fallthrough\u003d]: \u003d\u003e 391:3\n \n Reported-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Link: https://lore.kernel.org/r/20190729151033.GA10143@embeddedor\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Wim Van Sebroeck \u003cwim@linux-watchdog.org\u003e\n\ncommit 5fbdac150d4813765fd515ffad6c7f5e63be14aa\nAuthor: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\nDate: Wed Jul 17 08:48:06 2019 +0000\n\n mtd: spi-nor: fix description for int (*flash_is_locked)()\n \n The description was interleaved.\n \n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n\ncommit 9e38e690ace3e7a22a81fc02652fc101efb340cf\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Wed Jul 24 13:54:12 2019 +0530\n\n PCI: tegra: Fix OF node reference leak\n \n Each iteration of for_each_child_of_node() executes of_node_put() on the\n previous node, but in some return paths in the middle of the loop\n of_node_put() is missing thus causing a reference leak.\n \n Hence stash these mid-loop return values in a variable \u0027err\u0027 and add a\n new label err_node_put which executes of_node_put() on the previous node\n and returns \u0027err\u0027 on failure.\n \n Change mid-loop return statements to point to jump to this label to\n fix the reference leak.\n \n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n [lorenzo.pieralisi@arm.com: rewrote commit log]\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n\ncommit 802d9bd4fac70be2ea61fa83660a87a57d06bab0\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:27 2019 -0700\n\n mfd: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit b620c17672b9c162bbc1e480eaf43a825345cb2a\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Sun Jul 28 18:56:14 2019 -0500\n\n mfd: db8500-prcmu: Mark expected switch fall-throughs\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warnings:\n \n drivers/mfd/db8500-prcmu.c: In function \u0027dsiclk_rate\u0027:\n drivers/mfd/db8500-prcmu.c:1592:7: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n div *\u003d 2;\n ~~~~^~~~\n drivers/mfd/db8500-prcmu.c:1593:2: note: here\n case PRCM_DSI_PLLOUT_SEL_PHI_2:\n ^~~~\n drivers/mfd/db8500-prcmu.c:1594:7: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n div *\u003d 2;\n ~~~~^~~~\n drivers/mfd/db8500-prcmu.c:1595:2: note: here\n case PRCM_DSI_PLLOUT_SEL_PHI:\n ^~~~\n \n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit ec65b56046d27a21a5ae02eb7fcb321e1942a541\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Thu Aug 1 16:28:41 2019 +0300\n\n mfd: intel-lpss: Add Intel Tiger Lake PCI IDs\n \n Intel Tiger Lake has the same LPSS than Intel Broxton.\n Add the new IDs to the list of supported devices.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit ff71266aa490a9f8ed761d47945f300ba19e7c93\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Thu Jul 25 18:02:14 2019 -0400\n\n mfd: Drop obsolete JZ4740 driver\n \n It has been replaced with the ingenic-iio driver for the ADC.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit ea1acf11ee7ac2d63d93e78c638ec8abc710a1a7\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Sun Jul 28 18:58:58 2019 -0500\n\n mfd: omap-usb-host: Mark expected switch fall-throughs\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warnings:\n \n drivers/mfd/omap-usb-host.c: In function \u0027usbhs_runtime_resume\u0027:\n drivers/mfd/omap-usb-host.c:303:7: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n if (!IS_ERR(omap-\u003ehsic480m_clk[i])) {\n ^\n drivers/mfd/omap-usb-host.c:313:3: note: here\n case OMAP_EHCI_PORT_MODE_TLL:\n ^~~~\n drivers/mfd/omap-usb-host.c: In function \u0027usbhs_runtime_suspend\u0027:\n drivers/mfd/omap-usb-host.c:345:7: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n if (!IS_ERR(omap-\u003ehsic480m_clk[i]))\n ^\n drivers/mfd/omap-usb-host.c:349:3: note: here\n case OMAP_EHCI_PORT_MODE_TLL:\n ^~~~\n \n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 76380a607ba0b28627c9b4b55cd47a079a59624b\nAuthor: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\nDate: Fri Jul 5 12:55:03 2019 +0800\n\n mfd: intel-lpss: Remove D3cold delay\n \n Goodix touchpad may drop its first couple input events when\n i2c-designware-platdrv and intel-lpss it connects to took too long to\n runtime resume from runtime suspended state.\n \n This issue happens becuase the touchpad has a rather small buffer to\n store up to 13 input events, so if the host doesn\u0027t read those events in\n time (i.e. runtime resume takes too long), events are dropped from the\n touchpad\u0027s buffer.\n \n The bottleneck is D3cold delay it waits when transitioning from D3cold\n to D0, hence remove the delay to make the resume faster. I\u0027ve tested\n some systems with intel-lpss and haven\u0027t seen any regression.\n \n Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id\u003d202683\n Signed-off-by: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\n Reviewed-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 7891c30a3e30cbe5a6195fcdf6b36ab7a92c74ae\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Aug 12 10:12:03 2019 +0100\n\n dma-fence: Report the composite sync_file status\n \n Same as for the individual fences, we want to report the actual status\n of the fence when queried.\n \n Reported-by: Petri Latvala \u003cpetri.latvala@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Sumit Semwal \u003csumit.semwal@linaro.org\u003e\n Cc: Gustavo Padovan \u003cgustavo@padovan.org\u003e\n Cc: Petri Latvala \u003cpetri.latvala@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812091203.29871-1-chris@chris-wilson.co.uk\n\ncommit a664a834579ae8a6166ac9e5a3232976cab2c24d\nAuthor: Peter Wu \u003cpeter@lekensteyn.nl\u003e\nDate: Fri Aug 9 01:39:11 2019 +0100\n\n tools: bpftool: fix reading from /proc/config.gz\n \n /proc/config has never existed as far as I can see, but /proc/config.gz\n is present on Arch Linux. Add support for decompressing config.gz using\n zlib which is a mandatory dependency of libelf anyway. Replace existing\n stdio functions with gzFile operations since the latter transparently\n handles uncompressed and gzip-compressed files.\n \n Cc: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Signed-off-by: Peter Wu \u003cpeter@lekensteyn.nl\u003e\n Reviewed-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit b194a77fcc4001dc40aecdd15d249648e8a436d1\nAuthor: Xiaofei Tan \u003ctanxiaofei@huawei.com\u003e\nDate: Fri Jul 26 09:43:37 2019 +0800\n\n efi: cper: print AER info of PCIe fatal error\n \n AER info of PCIe fatal error is not printed in the current driver.\n Because APEI driver will panic directly for fatal error, and can\u0027t\n run to the place of printing AER info.\n \n An example log is as following:\n {763}[Hardware Error]: Hardware error from APEI Generic Hardware Error Source: 11\n {763}[Hardware Error]: event severity: fatal\n {763}[Hardware Error]: Error 0, type: fatal\n {763}[Hardware Error]: section_type: PCIe error\n {763}[Hardware Error]: port_type: 0, PCIe end point\n {763}[Hardware Error]: version: 4.0\n {763}[Hardware Error]: command: 0x0000, status: 0x0010\n {763}[Hardware Error]: device_id: 0000:82:00.0\n {763}[Hardware Error]: slot: 0\n {763}[Hardware Error]: secondary_bus: 0x00\n {763}[Hardware Error]: vendor_id: 0x8086, device_id: 0x10fb\n {763}[Hardware Error]: class_code: 000002\n Kernel panic - not syncing: Fatal hardware error!\n \n This issue was imported by the patch, \u002737448adfc7ce (\"aerdrv: Move\n cper_print_aer() call out of interrupt context\")\u0027. To fix this issue,\n this patch adds print of AER info in cper_print_pcie() for fatal error.\n \n Here is the example log after this patch applied:\n {24}[Hardware Error]: Hardware error from APEI Generic Hardware Error Source: 10\n {24}[Hardware Error]: event severity: fatal\n {24}[Hardware Error]: Error 0, type: fatal\n {24}[Hardware Error]: section_type: PCIe error\n {24}[Hardware Error]: port_type: 0, PCIe end point\n {24}[Hardware Error]: version: 4.0\n {24}[Hardware Error]: command: 0x0546, status: 0x4010\n {24}[Hardware Error]: device_id: 0000:01:00.0\n {24}[Hardware Error]: slot: 0\n {24}[Hardware Error]: secondary_bus: 0x00\n {24}[Hardware Error]: vendor_id: 0x15b3, device_id: 0x1019\n {24}[Hardware Error]: class_code: 000002\n {24}[Hardware Error]: aer_uncor_status: 0x00040000, aer_uncor_mask: 0x00000000\n {24}[Hardware Error]: aer_uncor_severity: 0x00062010\n {24}[Hardware Error]: TLP Header: 000000c0 01010000 00000001 00000000\n Kernel panic - not syncing: Fatal hardware error!\n \n Fixes: 37448adfc7ce (\"aerdrv: Move cper_print_aer() call out of interrupt context\")\n Signed-off-by: Xiaofei Tan \u003ctanxiaofei@huawei.com\u003e\n Reviewed-by: James Morse \u003cjames.morse@arm.com\u003e\n [ardb: put parens around terms of \u0026\u0026 operator]\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n\ncommit 015341da9888b58fbdeafd369c43f7a95fe4e763\nAuthor: Anshuman Gupta \u003canshuman.gupta@intel.com\u003e\nDate: Sun Aug 11 15:32:32 2019 +0530\n\n drm/i915/tgl: Fixing up list of PG3 power domains.\n \n The DDI-IO power wells (PWR_WELL_CTL_DDI) are backing\n the IO/PHY functionality, which doesn\u0027t need the PG3\n power power well. Accordingly fixing up the list of\n PG3 power domains.\n \n Cc: Imre Deak \u003cimre.deak@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Anshuman Gupta \u003canshuman.gupta@intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190811100232.27964-1-anshuman.gupta@intel.com\n\ncommit 2e04dbceed682a97c44cf50fd73c7cfc79e208ea\nAuthor: Anshuman Gupta \u003canshuman.gupta@intel.com\u003e\nDate: Sun Aug 11 13:49:08 2019 +0530\n\n drm/i915/icl: Remove DDI IO power domain from PG3 power domains\n \n The DDI-IO power wells (PWR_WELL_CTL_DDI) are backing\n the IO/PHY functionality, which doesn\u0027t need the PG3\n power power well. Accordingly fixing up the list of\n PG3 power domains.\n \n v2: Removed \"DDI E/F IO\"power domain as well [Imre]\n \n Cc: Imre Deak \u003cimre.deak@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Anshuman Gupta \u003canshuman.gupta@intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190811081908.9114-1-anshuman.gupta@intel.com\n\ncommit 5c7ed4385424b1c2ce0c27c41448e080e44ebb09\nAuthor: Dan Williams \u003cdan.j.williams@intel.com\u003e\nDate: Mon Aug 5 08:27:06 2019 -0600\n\n HMAT: Skip publishing target info for nodes with no online memory\n \n There are multiple scenarios where the HMAT may contain information\n about proximity domains that are not currently online. Rather than fail\n to report any HMAT data just elide those offline domains.\n \n If and when those domains are later onlined they can be added to the\n HMEM reporting at that point.\n \n This was found while testing EFI_MEMORY_SP support which reserves\n \"specific purpose\" memory from the general allocation pool. If that\n reservation results in an empty numa-node then the node is not marked\n online leading a spurious:\n \n \"acpi/hmat: Ignoring HMAT: Invalid table\"\n \n ...result for HMAT parsing.\n \n Reviewed-by: Dave Hansen \u003cdave.hansen@linux.intel.com\u003e\n Signed-off-by: Dan Williams \u003cdan.j.williams@intel.com\u003e\n Signed-off-by: Keith Busch \u003ckeith.busch@intel.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit b630f62bc57cb2f6364cb1199983361d2a664abd\nAuthor: Keith Busch \u003ckeith.busch@intel.com\u003e\nDate: Mon Aug 5 08:27:05 2019 -0600\n\n HMAT: Register attributes for memory hot add\n \n Some of the memory nodes described in HMAT may not be online at the\n time the HMAT subsystem parses their nodes\u0027 attributes. Should the node be\n set to online later, as can happen when using PMEM as RAM after boot, the\n nodes will be missing their initiator links and performance attributes.\n \n Regsiter a memory notifier callback and register the memory attributes\n the first time its node is brought online if it wasn\u0027t registered.\n \n Signed-off-by: Keith Busch \u003ckeith.busch@intel.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 06188d713885d2a469067637240605b657a4b5a7\nAuthor: Keith Busch \u003ckeith.busch@intel.com\u003e\nDate: Mon Aug 5 08:27:04 2019 -0600\n\n HMAT: Register memory-side cache after parsing\n \n Instead of registering the HMAT cache attributes in line with parsing\n the table, save the attributes in the memory target and register them\n after parsing completes. This will make it easier to register the\n attributes later when hot add is supported.\n \n Tested-by: Brice Goglin \u003cBrice.Goglin@inria.fr\u003e\n Signed-off-by: Keith Busch \u003ckeith.busch@intel.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit a5f978c3609f02ae0ccff1bf010b32e41707ef66\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Sun Aug 11 19:51:32 2019 +0000\n\n drm/i915/uc: Use -EIO code for GuC initialization failures\n \n Since commit 6ca9a2beb54a (\"drm/i915: Unwind i915_gem_init() failure\")\n we believed that we correctly handle all errors encountered during\n GuC initialization, including special one that indicates request to\n run driver with disabled GPU submission (-EIO).\n \n Unfortunately since commit 121981fafe69 (\"drm/i915/guc: Combine\n enable_guc_loading|submission modparams\") we stopped using that\n error code to avoid unwanted fallback to execlist submission mode.\n \n In result any GuC initialization failure was treated as non-recoverable\n error leading to driver load abort, so we could not even read related\n GuC error log to investigate cause of the problem.\n \n For now always return -EIO on any uC hardware related failure.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190811195132.9660-5-michal.wajdeczko@intel.com\n\ncommit 07587c3435110e172051c3a1fa0764027a1815f8\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Sun Aug 11 19:51:31 2019 +0000\n\n drm/i915/uc: Update messages from fw upload step\n \n Our old messages were redundant or misleading (as loaded is\n not the same as running). Keep only one message for debug.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190811195132.9660-4-michal.wajdeczko@intel.com\n\ncommit d8d6e5eb2bad48a6a649e3d38fe30c15d965e149\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Mon Aug 12 07:39:49 2019 +0000\n\n drm/i915/uc: Include HuC firmware version in summary\n \n After successful uC initialization we are reporting GuC\n firmware version and status of GuC submission and HuC.\n Add HuC fw version to this report to make it complete,\n but also skip all HuC info if HuC is not supported.\n \n v2: squeeze to one line (Chris)\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190812073949.24076-1-michal.wajdeczko@intel.com\n\ncommit ae7a3166a708bee439c7f683327f7381bdfd08ec\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Sun Aug 11 19:51:29 2019 +0000\n\n drm/i915/uc: Fail early if there is no GuC fw available\n \n We don\u0027t want to rely on misleading WOPCM partitioning error.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190811195132.9660-2-michal.wajdeczko@intel.com\n\ncommit 7afe9a4e56658bd659660ea10671831664469e5f\nAuthor: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\nDate: Sun Aug 11 13:33:06 2019 -0500\n\n i3c: master: fix a memory leak bug\n \n In i3c_master_getmwl_locked(), the buffer used for the dest payload data is\n allocated using kzalloc() in i3c_ccc_cmd_dest_init(). Later on, the length\n of the dest payload data is checked against \u0027sizeof(*mwl)\u0027. If they are not\n equal, -EIO is returned to indicate the error. However, the allocated\n buffer is not deallocated on this path, leading to a memory leak.\n \n To fix the above issue, free the buffer before returning the error.\n \n Signed-off-by: Wenwen Wang \u003cwenwen@cs.uga.edu\u003e\n Signed-off-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n\ncommit 30f644fd6ec9a0a241adb91579aaca7404132d7b\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Wed Jul 24 23:38:28 2019 +0200\n\n backlight: lms283gf05: Fix a typo in the description passed to \u0027devm_gpio_request_one()\u0027\n \n The description passed to \u0027devm_gpio_request_one()\u0027 should be related to\n LMS283GF05, not LMS285GF05.\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Reviewed-by: Daniel Thompson \u003cdaniel.thompson@linaro.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 072e2c8192cfc6ae1de1a2aca02d4512b69bdeed\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 22:34:00 2019 +0300\n\n backlight: lm3630a: Switch to use fwnode_property_count_uXX()\n \n Use use fwnode_property_count_uXX() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Daniel Thompson \u003cdaniel.thompson@linaro.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 865cac14c2dac2fa5f16d1781ca2746b0d323796\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Mon Jul 8 14:41:29 2019 +0200\n\n backlight: rave-sp: Leave initial state and register with correct device\n \n This way the backlight can be referenced through its device node and\n enabling/disabling can be managed through the panel driver.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Reviewed-by: Daniel Thompson \u003cdaniel.thompson@linaro.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 4c8b4c3855b0049f31be5bbfb19042005e98e822\nAuthor: Ondrej Jirman \u003cmegous@megous.com\u003e\nDate: Tue Aug 6 17:57:42 2019 +0200\n\n drm: sun4i: Add support for enabling DDC I2C bus to sun8i_dw_hdmi glue\n \n Orange Pi 3 board requires enabling a voltage shifting circuit via GPIO\n for the DDC bus to be usable.\n \n Add support for hdmi-connector node\u0027s optional ddc-en-gpios property to\n support this use case.\n \n Signed-off-by: Ondrej Jirman \u003cmegous@megous.com\u003e\n Reviewed-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Signed-off-by: Maxime Ripard \u003cmripard@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190806155744.10263-4-megous@megous.com\n\ncommit f6700060f53f0764fb970fa1cb716a7f671f0581\nAuthor: Ondrej Jirman \u003cmegous@megous.com\u003e\nDate: Tue Aug 6 17:57:41 2019 +0200\n\n dt-bindings: display: hdmi-connector: Support DDC bus enable\n \n Some Allwinner SoC using boards (Orange Pi 3 for example) need to enable\n on-board voltage shifting logic for the DDC bus using a gpio to be able\n to access DDC bus. Use ddc-en-gpios property on the hdmi-connector to\n model this.\n \n Add binding documentation for optional ddc-en-gpios property.\n \n Signed-off-by: Ondrej Jirman \u003cmegous@megous.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Maxime Ripard \u003cmripard@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190806155744.10263-3-megous@megous.com\n\ncommit 0ed4c252bf80b35fe768ec6506b2e58986f99687\nAuthor: Icenowy Zheng \u003cicenowy@aosc.io\u003e\nDate: Sun Jul 28 11:12:24 2019 +0800\n\n clk: sunxi-ng: v3s: add Allwinner V3 support\n \n Allwinner V3 has the same main die with V3s, but with more pins wired.\n There\u0027s a I2S bus on V3 that is not available on V3s.\n \n Add the V3-only peripheral\u0027s clocks and reset to the V3s CCU driver,\n bound to a new V3 compatible string. The driver name is not changed\n because it\u0027s part of the device tree binding (the header file name).\n \n Signed-off-by: Icenowy Zheng \u003cicenowy@aosc.io\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 720099603d1f62e37b789366d7e89824b009ca28\nAuthor: Icenowy Zheng \u003cicenowy@aosc.io\u003e\nDate: Sun Jul 28 11:12:23 2019 +0800\n\n clk: sunxi-ng: v3s: add missing clock slices for MMC2 module clocks\n \n The MMC2 clock slices are currently not defined in V3s CCU driver, which\n makes MMC2 not working.\n \n Fix this issue.\n \n Fixes: d0f11d14b0bc (\"clk: sunxi-ng: add support for V3s CCU\")\n Signed-off-by: Icenowy Zheng \u003cicenowy@aosc.io\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit c76f5089796a948e8daaeda348040a3f98b18748\nAuthor: Boris Brezillon \u003cboris.brezillon@bootlin.com\u003e\nDate: Tue Aug 6 10:40:41 2019 +0530\n\n mtd: spi-nor: Rework hwcaps selection for the spi-mem case\n \n The spi-mem layer provides a spi_mem_supports_op() function to check\n whether a specific operation is supported by the controller or not.\n This is much more accurate than the hwcaps selection logic based on\n SPI_{RX,TX}_ flags.\n \n Rework the hwcaps selection logic to use spi_mem_supports_op() when\n nor-\u003espimem !\u003d NULL.\n \n Signed-off-by: Boris Brezillon \u003cboris.brezillon@bootlin.com\u003e\n Signed-off-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n\ncommit b35b9a10362d203451d920d01ab8d6cd55cbaf2a\nAuthor: Boris Brezillon \u003cboris.brezillon@bootlin.com\u003e\nDate: Tue Aug 6 10:40:40 2019 +0530\n\n mtd: spi-nor: Move m25p80 code in spi-nor.c\n \n The m25p80 driver is actually a generic wrapper around the spi-mem\n layer. Not only the driver name is misleading, but we\u0027d expect such a\n common logic to be directly available in the core. Another reason for\n moving this code is that SPI NOR controller drivers should\n progressively be replaced by SPI controller drivers implementing the\n spi_mem_ops interface, and when the conversion is done, we should have\n a single spi-nor driver directly interfacing with the spi-mem layer.\n \n While moving the code we also fix a longstanding issue when\n non-DMA-able buffers are passed by the MTD layer.\n \n Signed-off-by: Boris Brezillon \u003cboris.brezillon@bootlin.com\u003e\n Signed-off-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n\ncommit ba972dac9854a76202c4e798fa618e6d56ca2f81\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:21 2019 +0200\n\n mfd: twl-core: Convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit ad9fc1f4229ed390590c3ea5be23addf84d54672\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:20 2019 +0200\n\n mfd: palmas: Convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 7a99c8f3310b13c87ec87114e6fa3e915005b187\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:19 2019 +0200\n\n mfd: max8998: Convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 4e32bff681fb4a7eb552af7187735a68c5afeab1\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:18 2019 +0200\n\n mfd: max8997: Convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit ddbf6ffeb63cf06cfc9707a550ece773bbd2a925\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:17 2019 +0200\n\n mfd: max8925-i2c: Convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit b8afcd54db8aae362095c8b490f6240520c4e9a4\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:16 2019 +0200\n\n mfd: max8907: Convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 0005a9e1bab7d2d46538d3db16ef1fde3433c710\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:15 2019 +0200\n\n mfd: max77843: Convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit ad28edcb87337a1c9e52cae4694eb419883b16a9\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:14 2019 +0200\n\n mfd: max77693: Convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Reviewed-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit f6ae8129631f6a3e072ec8d3ab60a65aa825f5f0\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:13 2019 +0200\n\n mfd: max14577: Convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Reviewed-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit f173f26a4d543fa57e6ed9505bbbbf9bec61f544\nAuthor: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\nDate: Tue Aug 6 10:40:39 2019 +0530\n\n mtd: spi-nor: always use bounce buffer for register read/writes\n \n spi-mem layer expects all buffers passed to it to be DMA\u0027able. But\n spi-nor layer mostly allocates buffers on stack for reading/writing to\n registers and therefore are not DMA\u0027able. Introduce bounce buffer to be\n used to read/write to registers. This ensures that buffer passed to\n spi-mem layer during register read/writes is DMA\u0027able. With this change\n nor-\u003ecmd-buf is no longer used, so drop it.\n \n Signed-off-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Signed-off-by: Tudor Ambarus \u003ctudor.ambarus@microchip.com\u003e\n\ncommit e310ee86f9efb92104ac84c178fd11bdf83216eb\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:12 2019 +0200\n\n mfd: da9150-core: Convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 98f0c05f409e4fbf186b11dcb2baed84a253c224\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:11 2019 +0200\n\n mfd: bcm590xx: Convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit f5d5d193c5f777fb9c03298992668a34c575abdd\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:10 2019 +0200\n\n mfd: ab3100-core: Convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 9520b835ffda85cdc1c84b1946c4dd3b6d00184c\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:09 2019 +0200\n\n mfd: 88pm860x-core: Convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 83215897356fb54fdb7b9e399d5256421d4cfe4a\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:08 2019 +0200\n\n mfd: 88pm800: Convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit aee36174b22d57775b3445399491777843f294ed\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 19:51:03 2019 +0800\n\n mfd: timberdale: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit e406b832d89d63b9eb525fa8fea18eb7a1598c60\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 20:26:28 2019 +0200\n\n mfd: da9063: Remove now unused platform_data\n \n All preparational patches have been applied, we can now remove the\n include file for platform_data. Yiha!\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit a604e5b29ce6cf5ec2950df7a5562ac2dd8d70e2\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:35 2019 +0200\n\n mfd: tps80031: Convert to devm_i2c_new_dummy_device\n \n Move from i2c_new_dummy() to devm_i2c_new_dummy_device(). So, we now get\n an ERRPTR which we use in error handling and we can skip removal of the\n created devices.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 624e3fceb533b33927535aaec713bd91ec80e99b\nAuthor: Yicheng Li \u003cyichengli@chromium.org\u003e\nDate: Mon Jul 8 11:15:37 2019 -0700\n\n mfd: cros_ec: Update cros_ec_commands.h\n \n Update cros_ec_commands.h to match the fingerprint MCU section in\n the current ec_commands.h\n \n Signed-off-by: Yicheng Li \u003cyichengli@chromium.org\u003e\n Reviewed-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Reviewed-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 1f70b8b812f36c24112b3e56a32412522673e50e\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Sun Aug 11 22:09:02 2019 +0100\n\n dma-fence: Propagate errors to dma-fence-array container\n \n When one of the array of fences is signaled, propagate its errors to the\n parent fence-array (keeping the first error to be raised).\n \n v2: Opencode cmpxchg_local to avoid compiler freakout.\n v3: Be careful not to flag an error if we race against signal-on-any.\n v4: Same applies to installing the signal cb.\n v5: Use cmpxchg to only set the error once before using a nifty idea by\n Christian to avoid changing the status after emitting the signal.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Sumit Semwal \u003csumit.semwal@linaro.org\u003e\n Cc: Gustavo Padovan \u003cgustavo@padovan.org\u003e\n Cc: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190811210902.22112-1-chris@chris-wilson.co.uk\n\ncommit f5d4c647d0ddcf4ffcb9ff584fba87a976addcbd\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Mon Aug 12 00:06:38 2019 -0700\n\n Input: applespi - no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 042cc4c778e1aa56f83ebe44b6f959c204861f08\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Mon Aug 12 00:05:21 2019 -0700\n\n Input: edt-ft5x06 - no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit f2ee73147a3f23cc4b032a76b5677b4b8441ba74\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 16:40:31 2019 +0200\n\n gpio: lpc32xx: allow building on non-lpc32xx targets\n \n The driver uses hardwire MMIO addresses instead of the data\n that is passed in device tree. Change it over to only\n hardcode the register offset values and allow compile-testing.\n \n Acked-by: Sylvain Lemieux \u003cslemieux.tyco@gmail.com\u003e\n Tested-by: Sylvain Lemieux \u003cslemieux.tyco@gmail.com\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Signed-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n\ncommit 656d29ce0f44163dbb12ff30b5a1a50644040500\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Aug 12 00:00:55 2019 -0700\n\n Input: matrix-keymap - switch to use device_property_count_u32()\n \n Use use device_property_count_u32() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 9ed05c94f20d504b6f0653ae981d171a801ae707\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Aug 12 00:00:35 2019 -0700\n\n Input: atmel_mxt_ts - switch to use device_property_count_u32()\n \n Use use device_property_count_u32() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 104c995fd6ab7c2bdb275a54334123184fd2efb2\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Aug 12 00:00:16 2019 -0700\n\n Input: hideep - switch to use device_property_count_u32()\n \n Use use device_property_count_u32() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 6e57ce76d0622aa4d790edd8e49f3704c48707c9\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Sun Aug 11 23:59:51 2019 -0700\n\n Input: mpr121 - switch to use device_property_count_u32()\n \n Use use device_property_count_u32() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit d99995a4f0f189731695efad4135daf04df53075\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sun Aug 11 23:42:31 2019 -0700\n\n Input: axp20x-pek - convert driver to use dev_groups\n \n Drivers now have the option to have the driver core create and remove any\n needed sysfs attribute files. So take advantage of that and do not\n register \"by hand\" a sysfs group of attributes.\n \n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit cf1b2a208d2e1b4c436a56672d09d12c34c06384\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sun Aug 11 23:41:23 2019 -0700\n\n Input: gpio_keys - convert driver to use dev_groups\n \n Drivers now have the option to have the driver core create and remove any\n needed sysfs attribute files. So take advantage of that and do not\n register \"by hand\" a bunch of sysfs files.\n \n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 9b1e6ce34a1ab6d205effa7da38ea520f04d4e03\nAuthor: Clément Péron \u003cperon.clem@gmail.com\u003e\nDate: Sun Aug 11 22:31:44 2019 +0200\n\n arm64: defconfig: Enable Sun4i SPDIF module\n \n Allwinner A64 and H6 use the Sun4i SPDIF driver.\n \n Enable this to allow a proper support.\n \n Signed-off-by: Clément Péron \u003cperon.clem@gmail.com\u003e\n Signed-off-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\n\ncommit ed075453d527a2635f245e0a91b92252ad3d1c8a\nAuthor: Jonathan Neuschäfer \u003cj.neuschaefer@gmx.net\u003e\nDate: Tue Jan 29 14:59:17 2019 +0100\n\n dt-bindings: mfd: rn5t618: Document optional property system-power-controller\n \n The RN5T618 family of PMICs can be used as system management\n controllers, in which case they handle poweroff and restart. Document\n this capability by referring to the corresponding generic DT binding.\n \n Signed-off-by: Jonathan Neuschäfer \u003cj.neuschaefer@gmx.net\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit cbfe612d471fc3eda048a6a70c5c8f5ee34026a4\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Jul 6 18:47:22 2019 +0200\n\n mfd: aat2870: No need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 64e8a9bacadb58e04ab517d8fb5735302b96db5f\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Jul 6 18:47:21 2019 +0200\n\n mfd: ab8500: No need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit efecc3b531a30004d0993907897ab977747af108\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Jul 6 18:47:20 2019 +0200\n\n mfd: ab3100: No need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit aa31332f67ac5418d0ca448153cf49ad09dba360\nAuthor: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nDate: Fri Jul 26 15:05:39 2019 +1000\n\n usb: gadget: aspeed: Implement dummy hub TT requests\n \n We just accept them instead of stalling and return\n zeros on GetTTState.\n \n Signed-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 22a5d3ce0d032d843e36d431e78e6a7dd0efa193\nAuthor: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nDate: Fri Jul 26 15:05:38 2019 +1000\n\n usb: Add definitions for the USB2.0 hub TT requests\n \n Signed-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit cca1754c1d004370b86f41ad346130589fd41152\nAuthor: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nDate: Fri Jul 26 15:05:37 2019 +1000\n\n usb: gadget: aspeed: Improve debugging when nuking\n \n When nuking requests, it\u0027s useful to display how many were\n actually nuked. It has proven handy when debugging issues\n where EP0 went in a wrong state.\n \n Signed-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 8016759894089ef79415852d253eb54217b44539\nAuthor: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nDate: Fri Jul 26 15:05:36 2019 +1000\n\n usb: gadget: aspeed: Remove unused \"suspended\" flag\n \n The state bit in the hub is sufficient\n \n Signed-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 5f0625aaba5e4c0db75e942ff3c76516465a77d9\nAuthor: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nDate: Fri Jul 26 15:05:35 2019 +1000\n\n usb: gadget: aspeed: Rework the reset logic\n \n We had some dodgy code using the speed setting to decide whether a\n port reset would reset the device or just enable it.\n \n Instead, if the device is disabled and has a gadget attached, a\n reset will enable it. If it\u0027s already enabled, a reset will\n reset it.\n \n Signed-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 155940a16d688b5376f7c955a5f53259c6fe5689\nAuthor: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nDate: Fri Jul 26 15:05:34 2019 +1000\n\n usb: gadget: aspeed: Check suspend/resume callback existence\n \n .. before calling them\n \n Signed-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit ef9d34686bae33de38b0198a4c6c1a50daab31d7\nAuthor: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nDate: Fri Jul 26 15:05:33 2019 +1000\n\n usb: gadget: aspeed: Don\u0027t reject requests on suspended devices\n \n A disconnect may just suspend the hub in absence of a physical\n disconnect detection. If we start rejecting requests, the mass\n storage function gets into a spin trying to requeue the same\n request for ever and hangs.\n \n Signed-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 7e420cef9047dc945b019dc235ec91c792c30f2d\nAuthor: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nDate: Fri Jul 26 15:05:32 2019 +1000\n\n usb: gadget: aspeed: Fix EP0 stall handling\n \n When stalling EP0, we need to wait for an ACK interrupt,\n otherwise we may get out of sync on the next setup packet\n data phase. Also we need to ignore the direction when\n processing that interrupt as the HW reports a potential\n mismatch.\n \n Implement this by adding a stall state to EP0. This fixes\n some reported issues with mass storage and some hosts.\n \n Signed-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit d042c35eb86a8ef13c4b5180fd6953a51d9e1efc\nAuthor: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nDate: Fri Jul 26 15:05:31 2019 +1000\n\n usb: gadget: aspeed: Cleanup EP0 state on port reset\n \n Otherwise, we can have a stale state after a disconnect and reconnect\n causing errors on the first SETUP packet to the device.\n \n Signed-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 9fbbeb4eff076859891c022c151ad20eda83d983\nAuthor: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nDate: Fri Jul 26 15:05:30 2019 +1000\n\n usb: gadget: aspeed: Don\u0027t set port enable change bit on reset\n \n This bit should be only set when the port enable goes down, for\n example, on errors. Not when it gets set after a port reset. Some\n USB stacks seem to be sensitive to this and fails enumeration.\n \n Signed-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit c59ae0a1055127dd3828a88e111a0db59b254104\nAuthor: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\nDate: Sun Aug 11 20:31:20 2019 -0700\n\n MIPS: tlbex: Explicitly cast _PAGE_NO_EXEC to a boolean\n \n clang warns:\n \n arch/mips/mm/tlbex.c:634:19: error: use of logical \u0027\u0026\u0026\u0027 with constant\n operand [-Werror,-Wconstant-logical-operand]\n if (cpu_has_rixi \u0026\u0026 _PAGE_NO_EXEC) {\n ^ ~~~~~~~~~~~~~\n arch/mips/mm/tlbex.c:634:19: note: use \u0027\u0026\u0027 for a bitwise operation\n if (cpu_has_rixi \u0026\u0026 _PAGE_NO_EXEC) {\n ^~\n \u0026\n arch/mips/mm/tlbex.c:634:19: note: remove constant to silence this\n warning\n if (cpu_has_rixi \u0026\u0026 _PAGE_NO_EXEC) {\n ~^~~~~~~~~~~~~~~~\n 1 error generated.\n \n Explicitly cast this value to a boolean so that clang understands we\n intend for this to be a non-zero value.\n \n Fixes: 00bf1c691d08 (\"MIPS: tlbex: Avoid placing software PTE bits in Entry* PFN fields\")\n Link: https://github.com/ClangBuiltLinux/linux/issues/609\n Signed-off-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: clang-built-linux@googlegroups.com\n\ncommit 077ff3be06e8de2657075dd2738b9a21f8ebd890\nAuthor: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\nDate: Sun Aug 11 20:31:17 2019 -0700\n\n MIPS/ptrace: Update mips_get_syscall_arg\u0027s return type\n \n clang warns:\n \n arch/mips/include/asm/syscall.h:136:3: error: variable \u0027ret\u0027 is\n uninitialized when used here [-Werror,-Wuninitialized]\n ret |\u003d mips_get_syscall_arg(args++, task, regs, i++);\n ^~~\n arch/mips/include/asm/syscall.h:129:9: note: initialize the variable\n \u0027ret\u0027 to silence this warning\n int ret;\n ^\n \u003d 0\n 1 error generated.\n \n It\u0027s not wrong; however, it\u0027s not an issue in practice because ret is\n only assigned to, not read from. ret could just be initialized to zero\n but looking into it further, ret has been unused since it was first\n added in 2012 so just get rid of it and update mips_get_syscall_arg\u0027s\n return type since none of the return values are ever checked. If it is\n ever needed again, this commit can be reverted and ret can be properly\n initialized.\n \n Fixes: c0ff3c53d4f9 (\"MIPS: Enable HAVE_ARCH_TRACEHOOK.\")\n Link: https://github.com/ClangBuiltLinux/linux/issues/604\n Signed-off-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: clang-built-linux@googlegroups.com\n\ncommit c2869aafe7191d366d74c55cb8a93c6d0baba317\nAuthor: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\nDate: Sun Aug 11 20:31:16 2019 -0700\n\n MIPS: Don\u0027t use bc_false uninitialized in __mm_isBranchInstr\n \n clang warns:\n \n arch/mips/kernel/branch.c:148:8: error: variable \u0027bc_false\u0027 is used\n uninitialized whenever switch case is taken\n [-Werror,-Wsometimes-uninitialized]\n case mm_bc2t_op:\n ^~~~~~~~~~\n arch/mips/kernel/branch.c:157:8: note: uninitialized use occurs here\n if (bc_false)\n ^~~~~~~~\n arch/mips/kernel/branch.c:149:8: error: variable \u0027bc_false\u0027 is used\n uninitialized whenever switch case is taken\n [-Werror,-Wsometimes-uninitialized]\n case mm_bc1t_op:\n ^~~~~~~~~~\n arch/mips/kernel/branch.c:157:8: note: uninitialized use occurs here\n if (bc_false)\n ^~~~~~~~\n arch/mips/kernel/branch.c:142:4: note: variable \u0027bc_false\u0027 is declared\n here\n int bc_false \u003d 0;\n ^\n 2 errors generated.\n \n When mm_bc1t_op and mm_bc2t_op are taken, the bc_false initialization\n does not happen, which leads to a garbage value upon use, as illustrated\n below with a small sample program.\n \n $ mipsel-linux-gnu-gcc --version | head -n1\n mipsel-linux-gnu-gcc (Debian 8.3.0-2) 8.3.0\n \n $ clang --version | head -n1\n ClangBuiltLinux clang version 9.0.0 (git://github.com/llvm/llvm-project\n 544315b4197034a3be8acd12cba56a75fb1f08dc) (based on LLVM 9.0.0svn)\n \n $ cat test.c\n #include \u003cstdio.h\u003e\n \n static void switch_scoped(int opcode)\n {\n switch (opcode) {\n case 1:\n case 2: {\n int bc_false \u003d 0;\n \n bc_false \u003d 4;\n case 3:\n case 4:\n printf(\"\\t* switch scoped bc_false \u003d %d\\n\", bc_false);\n }\n }\n }\n \n static void function_scoped(int opcode)\n {\n int bc_false \u003d 0;\n \n switch (opcode) {\n case 1:\n case 2: {\n bc_false \u003d 4;\n case 3:\n case 4:\n printf(\"\\t* function scoped bc_false \u003d %d\\n\", bc_false);\n }\n }\n }\n \n int main(void)\n {\n int opcode;\n \n for (opcode \u003d 1; opcode \u003c 5; opcode++) {\n printf(\"opcode \u003d %d:\\n\", opcode);\n switch_scoped(opcode);\n function_scoped(opcode);\n printf(\"\\n\");\n }\n \n return 0;\n }\n \n $ mipsel-linux-gnu-gcc -std\u003dgnu89 -static test.c \u0026\u0026 \\\n qemu-mipsel a.out\n opcode \u003d 1:\n * switch scoped bc_false \u003d 4\n * function scoped bc_false \u003d 4\n \n opcode \u003d 2:\n * switch scoped bc_false \u003d 4\n * function scoped bc_false \u003d 4\n \n opcode \u003d 3:\n * switch scoped bc_false \u003d 2147483004\n * function scoped bc_false \u003d 0\n \n opcode \u003d 4:\n * switch scoped bc_false \u003d 2147483004\n * function scoped bc_false \u003d 0\n \n $ clang -std\u003dgnu89 --target\u003dmipsel-linux-gnu -m32 -static test.c \u0026\u0026 \\\n qemu-mipsel a.out\n opcode \u003d 1:\n * switch scoped bc_false \u003d 4\n * function scoped bc_false \u003d 4\n \n opcode \u003d 2:\n * switch scoped bc_false \u003d 4\n * function scoped bc_false \u003d 4\n \n opcode \u003d 3:\n * switch scoped bc_false \u003d 2147483004\n * function scoped bc_false \u003d 0\n \n opcode \u003d 4:\n * switch scoped bc_false \u003d 2147483004\n * function scoped bc_false \u003d 0\n \n Move the definition up so that we get the right behavior and mark it\n __maybe_unused as it will not be used when CONFIG_MIPS_FP_SUPPORT\n isn\u0027t enabled.\n \n Fixes: 6a1cc218b9cc (\"MIPS: branch: Remove FP branch handling when CONFIG_MIPS_FP_SUPPORT\u003dn\")\n Link: https://github.com/ClangBuiltLinux/linux/issues/603\n Signed-off-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: clang-built-linux@googlegroups.com\n\ncommit 53f6f391786e01bf2050c03d8a36d9defdcc2831\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Aug 10 12:42:43 2019 +0200\n\n caif: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Cc: Richard Fontana \u003crfontana@redhat.com\u003e\n Cc: Steve Winslow \u003cswinslow@gmail.com\u003e\n Cc: netdev@vger.kernel.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6f20a697e487dea9e054daa18651fef04d97fe3d\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Aug 10 12:31:08 2019 +0200\n\n xen-netback: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Cc: Wei Liu \u003cwei.liu@kernel.org\u003e\n Cc: Paul Durrant \u003cpaul.durrant@citrix.com\u003e\n Cc: xen-devel@lists.xenproject.org\n Cc: netdev@vger.kernel.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Acked-by: Wei Liu \u003cwei.liu@kernel.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit eede236112b7ea4d9a7c2f2af2ac6648699b88f3\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Fri Aug 9 18:47:59 2019 -0400\n\n net: dsa: mv88e6xxx: add delay in direct SMI wait\n \n The mv88e6xxx_smi_direct_wait routine is used to wait on indirect\n registers access. It is of no exception and must delay between read\n attempts, like other wait routines.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1c6463b6fc36f2261df1a68ee13a5ebd980914b1\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Fri Aug 9 18:47:58 2019 -0400\n\n net: dsa: mv88e6xxx: fix SMI bit checking\n \n The current mv88e6xxx_smi_direct_wait function is only used to check\n the 16th bit of the (16-bit) SMI Command register. But the bit shift\n operation is not enough if we eventually use this function to check\n other bits, thus replace it with a mask.\n \n Fixes: e7ba0fad9c53 (\"net: dsa: mv88e6xxx: refine SMI support\")\n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 2ad4da776bdbad92e876aefe2f7bae3e691bbda4\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Fri Aug 9 18:47:57 2019 -0400\n\n net: dsa: mv88e6xxx: remove wait and update routines\n \n Now that we have proper Wait Bit and Wait Mask routines, remove the\n unused mv88e6xxx_wait routine and its Global 1 and Global 2 variants.\n \n The indirect tables such as the Device Mapping Table or Priority\n Override Table make use of an Update bit to distinguish reading (0)\n from writing (1) operations. After a write operation occurs, the bit\n self clears right away so there\u0027s no need to wait on it. Thus keep\n things simple and remove the mv88e6xxx_update helper as well.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 28ae1e9662b2b29a004adbb11ab0bd033bae296a\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Fri Aug 9 18:47:56 2019 -0400\n\n net: dsa: mv88e6xxx: wait for AVB Busy bit\n \n The AVB is not an indirect table using an Update bit, but a unit using\n a Busy bit. This means that we must ensure that this bit is cleared\n before setting it and wait until it gets cleared again after writing\n an operation. Reflect that.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 19fb7f69da2a1d46bb11cce467b3a9c0b2244202\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Fri Aug 9 18:47:55 2019 -0400\n\n net: dsa: mv88e6xxx: introduce wait bit routine\n \n Many portions of the driver need to wait until a given bit is set\n or cleared. Some busses even have a specific implementation for this\n operation. In preparation for such variant, implement a generic Wait\n Bit routine that can be used by the driver core functions.\n \n This allows us to get rid of the custom implementations we may find\n in the driver. Note that for the EEPROM bits, BUSY and RUNNING bits\n are independent, thus it is more efficient to wait independently for\n each bit instead of waiting for their mask.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 683f2244c5a3fb61c0e01999d43b4775775ef4e3\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Fri Aug 9 18:47:54 2019 -0400\n\n net: dsa: mv88e6xxx: introduce wait mask routine\n \n The current mv88e6xxx_wait routine is used to wait for a given mask\n to be cleared to zero. However in some cases, the driver may have\n to wait for a given mask to be of a certain non-zero value.\n \n Thus provide a generic wait mask routine that will be used to implement\n the current mv88e6xxx_wait function, and use it to wait for 88E6185\n PPU states.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 929938536fc0b1628fec8aff62a756e1c4f506d2\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Fri Aug 9 18:47:53 2019 -0400\n\n net: dsa: mv88e6xxx: wait for 88E6185 PPU disabled\n \n The PPU state of 88E6185 can be either \"Disabled at Reset\" or\n \"Disabled after Initialization\". Because we intentionally clear the\n PPU Enabled bit before checking its state, it is safe to wait for the\n MV88E6185_G1_STS_PPU_STATE_DISABLED state explicitly instead of waiting\n for any state different than MV88E6185_G1_STS_PPU_STATE_POLLING.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit eb2e7f092271091f774a20c0c3b8df3c8c52e060\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Fri Aug 9 22:59:07 2019 +0200\n\n r8169: inline rtl8169_free_rx_databuff\n \n rtl8169_free_rx_databuff is used in only one place, so let\u0027s inline it.\n We can improve the loop because rtl8169_init_ring zero\u0027s RX_databuff\n before calling rtl8169_rx_fill, and rtl8169_rx_fill fills\n Rx_databuff starting from index 0.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 087f5b8758ae9f1b1968bc469bb3f5fae53e639b\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Fri Aug 9 20:45:14 2019 +0200\n\n net: phy: realtek: add support for the 2.5Gbps PHY in RTL8125\n \n This adds support for the integrated 2.5Gbps PHY in Realtek RTL8125.\n Advertisement of 2.5Gbps mode is done via a vendor-specific register.\n Same applies to reading NBase-T link partner advertisement.\n Unfortunately this 2.5Gbps PHY shares the PHY ID with the integrated\n 1Gbps PHY\u0027s in other Realtek network chips and so far no method is\n known to differentiate them. As a workaround use a dedicated fake PHY ID\n that is set by the network driver by intercepting the MDIO PHY ID read.\n \n v2:\n - Create dedicated PHY driver and use a fake PHY ID that is injected by\n the network driver. Suggested by Andrew Lunn.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit bf22b343ca800aac076ccf986e762b28545aa6bb\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Fri Aug 9 20:44:22 2019 +0200\n\n net: phy: add phy_modify_paged_changed\n \n Add helper function phy_modify_paged_changed, behavios is the same\n as for phy_modify_changed.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f4069cd7fa6583e7094001c6fce6f426d17a4c76\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Fri Aug 9 20:43:50 2019 +0200\n\n net: phy: prepare phylib to deal with PHY\u0027s extending Clause 22\n \n The integrated PHY in 2.5Gbps chip RTL8125 is the first (known to me)\n PHY that uses standard Clause 22 for all modes up to 1Gbps and adds\n 2.5Gbps control using vendor-specific registers. To use phylib for\n the standard part little extensions are needed:\n - Move most of genphy_config_aneg to a new function\n __genphy_config_aneg that takes a parameter whether restarting\n auto-negotiation is needed (depending on whether content of\n vendor-specific advertisement register changed).\n - Don\u0027t clear phydev-\u003elp_advertising in genphy_read_status so that\n we can set non-C22 mode flags before.\n \n Basically both changes mimic the behavior of the equivalent Clause 45\n functions.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 3eef8689325e6358a0c51d2ef8c6a689b2f22d48\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Fri Aug 9 20:43:04 2019 +0200\n\n net: phy: simplify genphy_config_advert by using the linkmode_adv_to_xxx_t functions\n \n Using linkmode_adv_to_mii_adv_t and linkmode_adv_to_mii_ctrl1000_t\n allows to simplify the code. In addition avoiding the conversion to\n the legacy u32 advertisement format allows to remove the warning.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Suggested-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 150e8f8a1bae1dea75f4217f483ac95ca48c979a\nAuthor: Jiri Pirko \u003cjiri@mellanox.com\u003e\nDate: Fri Aug 9 13:05:12 2019 +0200\n\n netdevsim: register couple of devlink params\n \n Register couple of devlink params, one generic, one driver-specific.\n Make the values available over debugfs.\n \n Example:\n $ echo \"111\" \u003e /sys/bus/netdevsim/new_device\n $ devlink dev param\n netdevsim/netdevsim111:\n name max_macs type generic\n values:\n cmode driverinit value 32\n name test1 type driver-specific\n values:\n cmode driverinit value true\n $ cat /sys/kernel/debug/netdevsim/netdevsim111/max_macs\n 32\n $ cat /sys/kernel/debug/netdevsim/netdevsim111/test1\n Y\n $ devlink dev param set netdevsim/netdevsim111 name max_macs cmode driverinit value 16\n $ devlink dev param set netdevsim/netdevsim111 name test1 cmode driverinit value false\n $ devlink dev reload netdevsim/netdevsim111\n $ cat /sys/kernel/debug/netdevsim/netdevsim111/max_macs\n 16\n $ cat /sys/kernel/debug/netdevsim/netdevsim111/test1\n \n Signed-off-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit e272d4fb74d6306a1ec7309f7e8ee09011ff7f3c\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Sat Aug 3 11:51:18 2019 -0400\n\n configfs: fix a deadlock in configfs_symlink()\n \n Configfs abuses symlink(2). Unlike the normal filesystems, it\n wants the target resolved at symlink(2) time, like link(2) would\u0027ve\n done. The problem is that -\u003esymlink() is called with the parent\n directory locked exclusive, so resolving the target inside the\n -\u003esymlink() is easily deadlocked.\n \n Short of really ugly games in sys_symlink() itself, all we can\n do is to unlock the parent before resolving the target and\n relock it after. However, that invalidates the checks done\n by the caller of -\u003esymlink(), so we have to\n * check that dentry is still where it used to be\n (it couldn\u0027t have been moved, but it could\u0027ve been unhashed)\n * recheck that it\u0027s still negative (somebody else\n might\u0027ve successfully created a symlink with the same name\n while we were looking the target up)\n * recheck the permissions on the parent directory.\n \n Cc: stable@vger.kernel.org\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit bb5835edcdf8bf78bbe51cff13e332c439bc0567\nAuthor: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\nDate: Sun Aug 11 16:32:41 2019 -0400\n\n ext4: add new ioctl EXT4_IOC_GET_ES_CACHE\n \n For debugging reasons, it\u0027s useful to know the contents of the extent\n cache. Since the extent cache contains much of what is in the fiemap\n ioctl, use an fiemap-style interface to return this information.\n \n Signed-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n\ncommit 1ad3ea6e0a694b0486eb2cbe60378ad0fbf23642\nAuthor: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\nDate: Sun Aug 11 16:31:41 2019 -0400\n\n ext4: add a new ioctl EXT4_IOC_GETSTATE\n \n The new ioctl EXT4_IOC_GETSTATE returns some of the dynamic state of\n an ext4 inode for debugging purposes.\n \n Signed-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n\ncommit b0c013e2928d3696ceb6401311dbc1d7fcccd6dd\nAuthor: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\nDate: Sun Aug 11 16:30:41 2019 -0400\n\n ext4: add a new ioctl EXT4_IOC_CLEAR_ES_CACHE\n \n The new ioctl EXT4_IOC_CLEAR_ES_CACHE will force an inode\u0027s extent\n status cache to be cleared out. This is intended for use for\n debugging.\n \n Signed-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n\ncommit 547b9ad698b434eadca46319cb47e5875b55ef03\nAuthor: Chandan Rajendra \u003cchandan@linux.ibm.com\u003e\nDate: Sun Aug 11 16:29:41 2019 -0400\n\n jbd2: flush_descriptor(): Do not decrease buffer head\u0027s ref count\n \n When executing generic/388 on a ppc64le machine, we notice the following\n call trace,\n \n VFS: brelse: Trying to free free buffer\n WARNING: CPU: 0 PID: 6637 at /root/repos/linux/fs/buffer.c:1195 __brelse+0x84/0xc0\n \n Call Trace:\n __brelse+0x80/0xc0 (unreliable)\n invalidate_bh_lru+0x78/0xc0\n on_each_cpu_mask+0xa8/0x130\n on_each_cpu_cond_mask+0x130/0x170\n invalidate_bh_lrus+0x44/0x60\n invalidate_bdev+0x38/0x70\n ext4_put_super+0x294/0x560\n generic_shutdown_super+0xb0/0x170\n kill_block_super+0x38/0xb0\n deactivate_locked_super+0xa4/0xf0\n cleanup_mnt+0x164/0x1d0\n task_work_run+0x110/0x160\n do_notify_resume+0x414/0x460\n ret_from_except_lite+0x70/0x74\n \n The warning happens because flush_descriptor() drops bh reference it\n does not own. The bh reference acquired by\n jbd2_journal_get_descriptor_buffer() is owned by the log_bufs list and\n gets released when this list is processed. The reference for doing IO is\n only acquired in write_dirty_buffer() later in flush_descriptor().\n \n Reported-by: Harish Sriram \u003charish@linux.ibm.com\u003e\n Reviewed-by: Jan Kara \u003cjack@suse.cz\u003e\n Signed-off-by: Chandan Rajendra \u003cchandan@linux.ibm.com\u003e\n Signed-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n\ncommit 991f52306ab8b5d8427f6f953162b5ab92e88e51\nAuthor: Shi Siyuan \u003cshisiyuan@xiaomi.com\u003e\nDate: Sun Aug 11 16:28:41 2019 -0400\n\n ext4: remove unnecessary error check\n \n Remove unnecessary error check in ext4_file_write_iter(),\n because this check will be done in upcoming later function --\n ext4_write_checks() -\u003e generic_write_checks()\n \n Change-Id: I7b0ab27f693a50765c15b5eaa3f4e7c38f42e01e\n Signed-off-by: shisiyuan \u003cshisiyuan@xiaomi.com\u003e\n Signed-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n\ncommit 4e34323135ec0752af152ac588a4f96495074849\nAuthor: yangerkun \u003cyangerkun@huawei.com\u003e\nDate: Sun Aug 11 16:27:41 2019 -0400\n\n ext4: fix warning when turn on dioread_nolock and inline_data\n \n mkfs.ext4 -O inline_data /dev/vdb\n mount -o dioread_nolock /dev/vdb /mnt\n echo \"some inline data...\" \u003e\u003e /mnt/test-file\n echo \"some inline data...\" \u003e\u003e /mnt/test-file\n sync\n \n The above script will trigger \"WARN_ON(!io_end-\u003ehandle \u0026\u0026 sbi-\u003es_journal)\"\n because ext4_should_dioread_nolock() returns false for a file with inline\n data. Move the check to a place after we have already removed the inline\n data and prepared inode to write normal pages.\n \n Reviewed-by: Jan Kara \u003cjack@suse.cz\u003e\n Signed-off-by: yangerkun \u003cyangerkun@huawei.com\u003e\n Signed-off-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n\ncommit e9feb58020f952f7d9de785ede9a7d54ab1eda5c\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sun Aug 11 10:35:55 2019 +0300\n\n drop_monitor: Expose tail drop counter\n \n Previous patch made the length of the per-CPU skb drop list\n configurable. Expose a counter that shows how many packets could not be\n enqueued to this list.\n \n This allows users determine the desired queue length.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 30328d46af593dcf24582f2a431d84ea0cf4bdef\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sun Aug 11 10:35:54 2019 +0300\n\n drop_monitor: Make drop queue length configurable\n \n In packet alert mode, each CPU holds a list of dropped skbs that need to\n be processed in process context and sent to user space. To avoid\n exhausting the system\u0027s memory the maximum length of this queue is\n currently set to 1000.\n \n Allow users to tune the length of this queue according to their needs.\n The configured length is reported to user space when drop monitor\n configuration is queried.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 444be061d012f1a8ebf95292a648a4e0e2afa83f\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sun Aug 11 10:35:53 2019 +0300\n\n drop_monitor: Add a command to query current configuration\n \n Users should be able to query the current configuration of drop monitor\n before they start using it. Add a command to query the existing\n configuration which currently consists of alert mode and packet\n truncation length.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 57986617a736aec2980c1c78a9dd8dcdf477ee6e\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sun Aug 11 10:35:52 2019 +0300\n\n drop_monitor: Allow truncation of dropped packets\n \n When sending dropped packets to user space it is not always necessary to\n copy the entire packet as usually only the headers are of interest.\n \n Allow user to specify the truncation length and add the original length\n of the packet as additional metadata to the netlink message.\n \n By default no truncation is performed.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ca30707dee2bc8bc81cfd8b4277fe90f7ca6df1f\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sun Aug 11 10:35:51 2019 +0300\n\n drop_monitor: Add packet alert mode\n \n So far drop monitor supported only one alert mode in which a summary of\n locations in which packets were recently dropped was sent to user space.\n \n This alert mode is sufficient in order to understand that packets were\n dropped, but lacks information to perform a more detailed analysis.\n \n Add a new alert mode in which the dropped packet itself is passed to\n user space along with metadata: The drop location (as program counter\n and resolved symbol), ingress netdevice and drop timestamp. More\n metadata can be added in the future.\n \n To avoid performing expensive operations in the context in which\n kfree_skb() is invoked (can be hard IRQ), the dropped skb is cloned and\n queued on per-CPU skb drop list. Then, in process context the netlink\n message is allocated, prepared and finally sent to user space.\n \n The per-CPU skb drop list is limited to 1000 skbs to prevent exhausting\n the system\u0027s memory. Subsequent patches will make this limit\n configurable and also add a counter that indicates how many skbs were\n tail dropped.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 28315f7999870bb56da236f6b4ffce63efcc7897\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sun Aug 11 10:35:50 2019 +0300\n\n drop_monitor: Add alert mode operations\n \n The next patch is going to add another alert mode in which the dropped\n packet is notified to user space, instead of only a summary of recent\n drops.\n \n Abstract the differences between the modes by adding alert mode\n operations. The operations are selected based on the currently\n configured mode and associated with the probes and the work item just\n before tracing starts.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c5ab9b1c41f6d89d84fe147e51fe623f90bd026c\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sun Aug 11 10:35:49 2019 +0300\n\n drop_monitor: Require CAP_NET_ADMIN for drop monitor configuration\n \n Currently, the configure command does not do anything but return an\n error. Subsequent patches will enable the command to change various\n configuration options such as alert mode and packet truncation.\n \n Similar to other netlink-based configuration channels, make sure only\n users with the CAP_NET_ADMIN capability set can execute this command.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 44075f563793880d17d1a2f3a17e06835b157d07\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sun Aug 11 10:35:48 2019 +0300\n\n drop_monitor: Reset per-CPU data before starting to trace\n \n The function reset_per_cpu_data() allocates and prepares a new skb for\n the summary netlink alert message (\u0027NET_DM_CMD_ALERT\u0027). The new skb is\n stored in the per-CPU \u0027data\u0027 variable and the old is returned.\n \n The function is invoked during module initialization and from the\n workqueue, before an alert is sent. This means that it is possible to\n receive an alert with stale data, if we stopped tracing when the\n hysteresis timer (\u0027data-\u003esend_timer\u0027) was pending.\n \n Instead of invoking the function during module initialization, invoke it\n just before we start tracing and ensure we get a fresh skb.\n \n This also allows us to remove the calls to initialize the timer and the\n work item from the module initialization path, since both could have\n been triggered by the error paths of reset_per_cpu_data().\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 70c69274f354ecb905cd46449111e73b17dedaad\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sun Aug 11 10:35:47 2019 +0300\n\n drop_monitor: Initialize timer and work item upon tracing enable\n \n The timer and work item are currently initialized once during module\n init, but subsequent patches will need to associate different functions\n with the work item, based on the configured alert mode.\n \n Allow subsequent patches to make that change by initializing and\n de-initializing these objects during tracing enable and disable.\n \n This also guarantees that once the request to disable tracing returns,\n no more netlink notifications will be generated.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7c747838a55818fd0cdbe48afd98fba726aa898d\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Sun Aug 11 10:35:46 2019 +0300\n\n drop_monitor: Split tracing enable / disable to different functions\n \n Subsequent patches will need to enable / disable tracing based on the\n configured alerting mode.\n \n Reduce the nesting level and prepare for the introduction of this\n functionality by splitting the tracing enable / disable operations into\n two different functions.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9f4dd137b20e4e09ada4087b03eb7cb347c35a75\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Sat Aug 10 10:03:28 2019 +0100\n\n drm/i915: Remove unused debugfs/i915_emon_status\n \n Before we start upon our great GT interrupt refactor, throw out the\n cruft! In this case, it is an unloved debugfs showing the current ips\n status, a fairly meaningless bunch of numbers that we are not checking.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Andi Shyti \u003candi.shyti@intel.com\u003e\n Acked-by: Andi Shyti \u003candi.shyti@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190810090329.6966-1-chris@chris-wilson.co.uk\n\ncommit ef19ee60f9cf713518c1d5e15946466d51e9f00b\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Sat Aug 10 22:45:45 2019 -0700\n\n iio: hi8435: Drop hi8435_remove() by using devres for remaining elements\n \n Convert the remainder of hi8435_probe() to use devres and get rid of\n hi8435_remove().\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: linux-kernel@vger.kernel.org\n Cc: linux-iio@vger.kernel.org\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit b1ac3a4b9aa2f68d498824f1235788e67b51b486\nAuthor: Przemyslaw Gaj \u003cpgaj@cadence.com\u003e\nDate: Sat Jun 22 21:54:59 2019 +0100\n\n i3c: add addr and lvr to i2c_dev_desc structure\n \n I need to store address and lvr value for I2C devices without static definition\n in DT. This allows secondary master to transmit DEFSLVS command properly.\n \n Main changes between v4 and v5:\n - Change in defslvs to use addr and lvr from i2c_dev_desc structure\n - Change in CDNS and DW drivers to use addr and lvr from i2c_dev_desc structure\n \n Signed-off-by: Przemyslaw Gaj \u003cpgaj@cadence.com\u003e\n Signed-off-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n\ncommit 6822dc9daa26102161e58a96b2ca938d14696006\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Sat Aug 10 22:45:11 2019 -0700\n\n iio: hi8435: Use gpiod_set_value_cansleep()\n \n Use gpiod_set_value_cansleep() instead of gpiod_set_value() to support\n the case when reset pin is connected to a GPIO expander. See ZII VF610\n SCU4 AIB for one such example.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: linux-kernel@vger.kernel.org\n Cc: linux-iio@vger.kernel.org\n Cc: Jonathan Cameron \u003cjic23@kernel.org\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit fd3f8f3118c66c492c5fe05568569d26db258fbd\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Thu Jul 25 23:51:45 2019 +0300\n\n i3c: master: cdns: Use for_each_set_bit()\n \n This simplifies and standardizes slot manipulation code\n by using for_each_set_bit() library function.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n\ncommit 91227632fc7a047965dcc41da89a40dc982caae8\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Jul 23 16:07:25 2019 +0530\n\n i3c: master: Add of_node_put() before return\n \n Each iteration of for_each_available_childe_of_node puts the previous\n node, but in the case of a return from the middle of the loop, there\n is no put, thus causing a memory leak. Add an of_node_put before the\n return.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n\ncommit e031d5f558f1535968df1ecae8d734b84c17f78d\nAuthor: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\nDate: Mon Aug 5 11:57:11 2019 -0700\n\n iio:st_sensors: remove buffer allocation at each buffer enable\n \n This patch is removing the buffer allocation at each buffer enable.\n We just allocate enough memory in the main structure during probe\n to cover maximum size needed (that anyway is pretty small) [16bytes].\n \n Signed-off-by: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 7e174a49bb80f32c20a963b12aa514394666cc3c\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Aug 10 12:17:32 2019 +0200\n\n ieee802154: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Cc: Alexander Aring \u003calex.aring@gmail.com\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: Harry Morris \u003ch.morris@cascoda.com\u003e\n Cc: linux-wpan@vger.kernel.org\n Cc: netdev@vger.kernel.org\n Acked-by: Stefan Schmidt \u003cstefan@datenfreihafen.org\u003e\n Acked-by: Michael Hennerich \u003cmichael.hennerich@analog.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 35dc61ebfc10849121ce059af7343bf470d1fc1b\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Aug 10 12:17:31 2019 +0200\n\n ixgbe: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Cc: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: intel-wired-lan@lists.osuosl.org\n Cc: netdev@vger.kernel.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 43c4eb0381198bdef655461d745c38bfaf3e4b06\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Aug 10 12:17:30 2019 +0200\n\n i40e: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Cc: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: intel-wired-lan@lists.osuosl.org\n Cc: netdev@vger.kernel.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ecc557075166e751305e7448d9436d35dbce2e54\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Aug 10 12:17:29 2019 +0200\n\n fm10k: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Cc: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: intel-wired-lan@lists.osuosl.org\n Cc: netdev@vger.kernel.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit e6882aa623f6fe0d80fa82ebf3ee78c353bffbe1\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Aug 10 12:17:28 2019 +0200\n\n mvpp2: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: Maxime Chevallier \u003cmaxime.chevallier@bootlin.com\u003e\n Cc: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Cc: Nathan Huckleberry \u003cnhuck@google.com\u003e\n Cc: netdev@vger.kernel.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 2f62f8e6c379c32cc1d629043b67feb06cda7661\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Aug 10 12:17:27 2019 +0200\n\n skge: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Cc: Mirko Lindner \u003cmlindner@marvell.com\u003e\n Cc: Stephen Hemminger \u003cstephen@networkplumber.org\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: netdev@vger.kernel.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 687236b07a672ad10a0432270c7b19217580a0b1\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Aug 10 12:17:26 2019 +0200\n\n qca: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: Stefan Wahren \u003cstefan.wahren@i2se.com\u003e\n Cc: Michael Heimpold \u003cmichael.heimpold@i2se.com\u003e\n Cc: Yangtao Li \u003ctiny.windzz@gmail.com\u003e\n Cc: netdev@vger.kernel.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 92aff5b4678745660259035a5467bc8870003900\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Aug 10 12:17:25 2019 +0200\n\n dpaa2: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Because we don\u0027t care about the individual files, we can remove the\n stored dentry for the files, as they are not needed to be kept track of\n at all.\n \n Cc: Ioana Radulescu \u003cruxandra.radulescu@nxp.com\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: netdev@vger.kernel.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 8d72ab119f42f25abb393093472ae0ca275088b6\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Aug 10 12:17:24 2019 +0200\n\n stmmac: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Because we don\u0027t care about the individual files, we can remove the\n stored dentry for the files, as they are not needed to be kept track of\n at all.\n \n Cc: Giuseppe Cavallaro \u003cpeppe.cavallaro@st.com\u003e\n Cc: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n Cc: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: Maxime Coquelin \u003cmcoquelin.stm32@gmail.com\u003e\n Cc: netdev@vger.kernel.org\n Cc: linux-stm32@st-md-mailman.stormreply.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 16e9b481e988b1f7e6df2243bb510e1c9b581272\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Aug 10 12:17:23 2019 +0200\n\n nfp: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: Alexei Starovoitov \u003cast@kernel.org\u003e\n Cc: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n Cc: Jesper Dangaard Brouer \u003chawk@kernel.org\u003e\n Cc: John Fastabend \u003cjohn.fastabend@gmail.com\u003e\n Cc: Edwin Peer \u003cedwin.peer@netronome.com\u003e\n Cc: Yangtao Li \u003ctiny.windzz@gmail.com\u003e\n Cc: Simon Horman \u003csimon.horman@netronome.com\u003e\n Cc: oss-drivers@netronome.com\n Cc: netdev@vger.kernel.org\n Acked-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 11ab11e69d63f4f7af92e9466f19255eff877945\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Aug 10 12:17:22 2019 +0200\n\n hns3: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Cc: Yisen Zhuang \u003cyisen.zhuang@huawei.com\u003e\n Cc: Salil Mehta \u003csalil.mehta@huawei.com\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: netdev@vger.kernel.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9dac1e8eeaa2fafa0ce2931a3abf5644c5ab61eb\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Aug 10 12:17:21 2019 +0200\n\n cxgb4: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n If a debugfs call fails, it will properly warn in the syslog, there\u0027s no\n need for all individual drivers to also print a message, so that is one\n more reason to not care about checking the return values.\n \n Cc: Vishal Kulkarni \u003cvishal@chelsio.com\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: Casey Leedom \u003cleedom@chelsio.com\u003e\n Cc: netdev@vger.kernel.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 3a131e85043cf538d5e70c0f23f9d69a4dd642b9\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Aug 10 12:17:20 2019 +0200\n\n bnxt: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n This cleans up a lot of unneeded code and logic around the debugfs\n files, making all of this much simpler and easier to understand.\n \n Cc: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: netdev@vger.kernel.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9e3926df8779a69f8541c9983c6d19b097671abd\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Aug 10 12:17:19 2019 +0200\n\n xgbe: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n This cleans up a lot of unneeded code and logic around the debugfs\n files, making all of this much simpler and easier to understand.\n \n Cc: Tom Lendacky \u003cthomas.lendacky@amd.com\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: netdev@vger.kernel.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9f818c8a7388ad1a5c60ace50be6f658c058a5f2\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Aug 10 12:17:18 2019 +0200\n\n mlx5: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n This cleans up a lot of unneeded code and logic around the debugfs\n files, making all of this much simpler and easier to understand as we\n don\u0027t need to keep the dentries saved anymore.\n \n Cc: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n Cc: Leon Romanovsky \u003cleon@kernel.org\u003e\n Cc: netdev@vger.kernel.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit fedcc6da10f33098d76b45adefa1b4e92a261748\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Aug 10 12:17:17 2019 +0200\n\n bonding: no need to print a message if debugfs_create_dir() fails\n \n The debugfs core now will print a message if this function fails, so\n don\u0027t duplicate that logic. Also, no need to change the code logic if\n the call fails either, as no debugfs calls should interrupt normal\n kernel code for any reason.\n \n Cc: Jay Vosburgh \u003cj.vosburgh@gmail.com\u003e\n Cc: Veaceslav Falico \u003cvfalico@gmail.com\u003e\n Cc: Andy Gospodarek \u003candy@greyhouse.net\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: netdev@vger.kernel.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a62052ba2aecb9269a32efeb3e22f96b83a13304\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Aug 10 12:17:16 2019 +0200\n\n wimax: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n This cleans up a lot of unneeded code and logic around the debugfs wimax\n files, making all of this much simpler and easier to understand.\n \n Cc: Inaky Perez-Gonzalez \u003cinaky.perez-gonzalez@intel.com\u003e\n Cc: linux-wimax@intel.com\n Cc: netdev@vger.kernel.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 14d1b9a6247c4548015d940ca92dbea05c0245bb\nAuthor: Matthew Auld \u003cmatthew.auld@intel.com\u003e\nDate: Fri Aug 9 21:29:24 2019 +0100\n\n drm/i915: buddy allocator\n \n Simple buddy allocator. We want to allocate properly aligned\n power-of-two blocks to promote usage of huge-pages for the GTT, so 64K,\n 2M and possibly even 1G. While we do support allocating stuff at a\n specific offset, it is more intended for preallocating portions of the\n address space, say for an initial framebuffer, for other uses drm_mm is\n probably a much better fit. Anyway, hopefully this can all be thrown\n away if we eventually move to having the core MM manage device memory.\n \n Signed-off-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190809202926.14545-2-matthew.auld@intel.com\n\ncommit 05f219d709ec5775b631b383860d3b4c8691a45f\nAuthor: Matthew Auld \u003cmatthew.auld@intel.com\u003e\nDate: Sat Aug 10 18:43:38 2019 +0100\n\n drm/i915/blt: support copying objects\n \n We can already clear an object with the blt, so try to do the same to\n support copying from one object backing store to another. Really this is\n just object -\u003e object, which is not that useful yet, what we really want\n is two backing stores, but that will require some vma rework first,\n otherwise we are stuck with \"tmp\" objects.\n \n Signed-off-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Abdiel Janulgue \u003cabdiel.janulgue@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190810174338.19810-1-chris@chris-wilson.co.uk\n\ncommit e0d409ff16201e76e07ebb5ec942abfbaf21c661\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Aug 4 22:16:34 2019 +0200\n\n drm/panel: drop return code from drm_panel_detach()\n \n There are no errors that can be reported by this function,\n so drop the return code.\n Fix the only bridge driver that checked the return result.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart+renesas@ideasonboard.com\u003e\n Cc: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Cc: Gwan-gyeong Mun \u003cgwan-gyeong.mun@intel.com\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190804201637.1240-14-sam@ravnborg.org\n\ncommit 8b0c6e2faa9749e219ba0f5e811fccd0ea754577\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Aug 4 22:16:33 2019 +0200\n\n drm/panel: use inline comments in drm_panel.h\n \n Inline comments provide better space for additional comments.\n Comments was slightly edited to follow the normal style,\n but no change to actual content.\n Used the opportuniy to change the order in drm_panel_funcs\n to follow the order they will be used by a panel.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190804201637.1240-13-sam@ravnborg.org\n\ncommit 7a833d307cd828e5f214f8d490f16fefabf8dbff\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Aug 4 22:16:32 2019 +0200\n\n drm/panel: move drm_panel functions to .c file\n \n Move inline functions from include/drm/drm_panel.h to drm_panel.c.\n This is in preparation for follow-up patches that will add extra\n logic to the functions.\n As they are no longer static inline, EXPORT them.\n \n v2:\n - align order of functions in drm_panel.h and drm_panel.c (Laurent)\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190804201637.1240-12-sam@ravnborg.org\n\ncommit 9d027b8e5a5754571a6b803f3635442fbc2abcdc\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Aug 4 22:16:31 2019 +0200\n\n drm/panel: ili9322: move bus_flags to get_modes()\n \n To prepare the driver to receive drm_connector only in the get_modes()\n callback, move bus_flags handling to ili9322_get_modes().\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190804201637.1240-11-sam@ravnborg.org\n\ncommit ce7d93aa2a080d94cca16948b9e3bef54f7ba031\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Aug 4 22:16:28 2019 +0200\n\n drm/mxsfb: fix opencoded use of drm_panel_*\n \n Use the drm_panel_get_modes() function.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Marek Vasut \u003cmarex@denx.de\u003e\n Acked-by: Stefan Agner \u003cstefan@agner.ch\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Shawn Guo \u003cshawnguo@kernel.org\u003e\n Cc: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\n Cc: Pengutronix Kernel Team \u003ckernel@pengutronix.de\u003e\n Cc: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Cc: NXP Linux Team \u003clinux-imx@nxp.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190804201637.1240-8-sam@ravnborg.org\n\ncommit 24bc3e9ce68a41c73b32af8cd077fdf26b31516c\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Aug 4 22:16:26 2019 +0200\n\n drm/fsl-dcu: fix opencoded use of drm_panel_*\n \n Use drm_panel_get_modes() to access modes.\n This has a nice side effect to simplify the code.\n \n drm_panel_get_modes() may return a negative value if\n for example panel is NULL. This is a small change\n compared to before, but really what we want.\n \n v2:\n - Add more info to changelog (Stefan)\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Stefan Agner \u003cstefan@agner.ch\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Alison Wang \u003calison.wang@nxp.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190804201637.1240-6-sam@ravnborg.org\n\ncommit 90fbc5105653c1c757eb5889b32106cdd754e69f\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Aug 4 22:16:25 2019 +0200\n\n drm/imx: fix opencoded use of drm_panel_*\n \n Use the drm_panel_get_modes() function to get the modes.\n \n This patch leave one test for the function pointer:\n panel-\u003efuncs-\u003eget_modes\n \n This is used to check if the panel may have any modes.\n There is no direct replacement.\n We may be able to just check that drm_panel_get_modes() return \u003e 0,\n but as this is not the same functionality it is left for later.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Shawn Guo \u003cshawnguo@kernel.org\u003e\n Cc: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\n Cc: Pengutronix Kernel Team \u003ckernel@pengutronix.de\u003e\n Cc: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Cc: NXP Linux Team \u003clinux-imx@nxp.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190804201637.1240-5-sam@ravnborg.org\n\ncommit 7bb0a60aa57818a9c49c5b4ec64e5f75581c369d\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Aug 4 22:16:22 2019 +0200\n\n drm/bridge: tc358767: fix opencoded use of drm_panel_*\n \n Replace open coded version with call to drm_panel_get_modes().\n \n Include change to deal with the possible negative\n return values from drm_panel_get_modes()\n \n v2:\n - Added more info to changelog (Philipp)\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Cc: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Cc: Laurent Pinchart \u003cLaurent.pinchart@ideasonboard.com\u003e\n Cc: Jonas Karlman \u003cjonas@kwiboo.se\u003e\n Cc: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190804201637.1240-2-sam@ravnborg.org\n\ncommit b7e7fffd3e8c81aa0654ed4c9f7a142b4c4dab1a\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Tue Jul 30 12:11:08 2019 +0200\n\n cpuidle: teo: Get rid of redundant check in teo_update()\n \n Notice that setting measured_us to UINT_MAX in teo_update() earlier\n doesn\u0027t change the behavior of the following code, so do that and\n eliminate a redundant check used for setting measured_us to UINT_MAX.\n \n This change is not expected to alter functionality.\n \n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit aaa914cd1c6332950523c942988a2ccd7cc2f7d9\nAuthor: Matthew Auld \u003cmatthew.auld@intel.com\u003e\nDate: Fri Aug 9 20:34:56 2019 +0100\n\n drm/i915/gtt: disable 2M pages for pre-gen11\n \n We currently disable THP(Transparent-Huge-Pages) for our shmem objects\n due to a performance regression with read BW in some internal\n benchmarks. Given that this is our main source of 2M pages, there really\n isn\u0027t much point in enabling 2M GTT pages, especially as that comes at\n the cost of disabling the GTT cache. However from gen11 it looks like we\n should hopefully see the HW issue resolved. Given this opt for only\n enabling 2M GTT pages from gen11 onwards.\n \n Signed-off-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190809193456.3836-2-matthew.auld@intel.com\n\ncommit 1feb7864b2cbd874a677ce04d61834d5664133ab\nAuthor: Matthew Auld \u003cmatthew.auld@intel.com\u003e\nDate: Fri Aug 9 20:34:55 2019 +0100\n\n drm/i915/gtt: enable GTT cache by default\n \n For some platforms the GTT cache is by default not enabled, and\n currently where we explicitly enable it, we make it conditional on 2M GTT\n page support, since the BSpec states that we must disable it if we\n enable 2M/1G pages. To make this more consistent opt for blanket\n enabling the GTT cache for all relevant gens in a single place, while\n still keeping the same behaviour of checking for 2M support.\n \n BSpec: 9314\n BSpec: 423\n Signed-off-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190809193456.3836-1-matthew.auld@intel.com\n\ncommit 18851edfa7dbe156cfc5a73bf009d8a4def84bb5\nAuthor: Matthew Auld \u003cmatthew.auld@intel.com\u003e\nDate: Sat Aug 10 11:50:08 2019 +0100\n\n drm/i915/selftests: move gpu-write-dw into utils\n \n Using the gpu to write to some dword over a number of pages is rather\n useful, and we already have two copies of such a thing, and we don\u0027t\n want a third so move it to utils. There is probably some other stuff\n also...\n \n Signed-off-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190810105008.14320-1-chris@chris-wilson.co.uk\n\ncommit 5130802ddbb10a73cf17d1117e2044c8d5ba7d65\nAuthor: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\nDate: Tue Jul 23 11:44:02 2019 +0530\n\n thermal: cpu_cooling: Switch to QoS requests for freq limits\n \n The cpufreq core now takes the min/max frequency constraints via QoS\n requests and the CPUFREQ_ADJUST notifier shall get removed later on.\n \n Switch over to using the QoS request for maximum frequency constraint\n for cpu_cooling driver.\n \n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n [ rjw: Subject ]\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 6a1490367c44f94614e39e8b98ff7114ff8a6449\nAuthor: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\nDate: Tue Jul 23 11:44:01 2019 +0530\n\n cpufreq: Add policy create/remove notifiers back\n \n This effectively reverts some changes made by commit f9f41e3ef99\n (\"cpufreq: Remove policy create/remove notifiers\").\n \n We have a new use case for policy create/remove notifiers (for\n allocating/freeing QoS requests per policy), so add them back.\n \n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n [ rjw: Subject \u0026 changelog ]\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 11f26633cccb7243217370837cbb066a73f678a5\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Sat Aug 10 13:18:06 2019 +0200\n\n PM: suspend: Fix platform_suspend_prepare_noirq()\n \n After commit ac9eafbe930a (\"ACPI: PM: s2idle: Execute LPS0 _DSM\n functions with suspended devices\"), a NULL pointer may be dereferenced\n if suspend-to-idle is attempted on a platform without \"traditional\"\n suspend support due to invalid fall-through in\n platform_suspend_prepare_noirq().\n \n Fix that and while at it add missing braces in platform_resume_noirq().\n \n Fixes: ac9eafbe930a (\"ACPI: PM: s2idle: Execute LPS0 _DSM functions with suspended devices\")\n Reported-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 92b8608691bf129b1137be46a3a0058bcacc97a9\nAuthor: Simon Horman \u003chorms+renesas@verge.net.au\u003e\nDate: Fri Aug 9 14:37:10 2019 -0700\n\n dt-bindings: usb: renesas_gen3: Rename bindings documentation file to reflect IP block\n \n For consistency with the naming of (most) other documentation files for DT\n bindings for Renesas IP blocks rename the Renesas USB3.0 peripheral\n documentation file from renesas,usb3.txt to renesas,usb3-peri.txt\n \n This refines a recent rename from renesas_usb3.txt to renesas,usb3.txt.\n The motivation is to more accurately reflect the IP block documented in\n this file.\n \n Signed-off-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Niklas Sderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\n Link: https://lore.kernel.org/r/20190809213710.31783-1-horms+renesas@verge.net.au\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 0e1d8083bddb38b7169f6240905422f95d3c31b9\nAuthor: Christian König \u003cchristian.koenig@amd.com\u003e\nDate: Mon Aug 5 11:16:12 2019 +0200\n\n dma-buf: further relax reservation_object_add_shared_fence\n \n Other cores don\u0027t busy wait any more and we removed the last user of checking\n the seqno for changes. Drop updating the number for shared fences altogether.\n \n Signed-off-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/322379/?series\u003d64837\u0026rev\u003d1\n\ncommit dd7a7d1ff2f199a8a80ee233480922d4f17adc6d\nAuthor: Christian König \u003cchristian.koenig@amd.com\u003e\nDate: Tue Aug 6 16:30:33 2019 +0200\n\n drm/i915: use new reservation_object_fences helper\n \n Instead of open coding the sequence loop use the new helper.\n \n Signed-off-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/322377/?series\u003d64837\u0026rev\u003d1\n\ncommit 67c97fb79a7f8621d4514275691d75f5ff158c46\nAuthor: Christian König \u003cchristian.koenig@amd.com\u003e\nDate: Tue Aug 6 14:19:33 2019 +0200\n\n dma-buf: add reservation_object_fences helper\n \n Add a new helper to get a consistent set of pointers from the reservation\n object. While at it group all access helpers together in the header file.\n \n v2: correctly return shared_count as well\n \n Signed-off-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/322378/?series\u003d64837\u0026rev\u003d1\n\ncommit 0e2f733addbf7702065cc18d38b67a1284d0c2be\nAuthor: Christian König \u003cchristian.koenig@amd.com\u003e\nDate: Wed Aug 7 12:31:48 2019 +0200\n\n dma-buf: make dma_fence structure a bit smaller v2\n \n We clear the callback list on kref_put so that by the time we\n release the fence it is unused. No one should be adding to the cb_list\n that they don\u0027t themselves hold a reference for.\n \n This small change is actually making the structure 16% smaller.\n \n v2: add the comment to the code as well.\n \n Signed-off-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/322916/\n\ncommit 554e330ceb9f00204bb692974c490ad50fc104cc\nAuthor: Matthew Auld \u003cmatthew.auld@intel.com\u003e\nDate: Sat Aug 10 10:29:45 2019 +0100\n\n drm/i915/blt: bump the size restriction\n \n As pointed out by Chris, with our current approach we are actually\n limited to S16_MAX * PAGE_SIZE for our size when using the blt to clear\n pages. Keeping things simple try to fix this by reducing the copy to a\n sequence of S16_MAX * PAGE_SIZE blocks.\n \n Reported-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n [ickle: hide the details of the engine pool inside emit_vma]\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190810092945.2762-1-chris@chris-wilson.co.uk\n\ncommit 963ad1285b7c7c5e4c75c2aa108ace866010e9bc\nAuthor: Matthew Auld \u003cmatthew.auld@intel.com\u003e\nDate: Sat Aug 10 10:17:47 2019 +0100\n\n drm/i915/blt: don\u0027t assume pinned intel_context\n \n Currently we just pass in bcs0-\u003eengine_context so it matters not, but in\n the future we may want to pass in something that is not a\n kernel_context, so try to be a bit more generic.\n \n Suggested-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190810091748.10972-1-chris@chris-wilson.co.uk\n\ncommit b772d7aeae093a120c737c0183df532262229efe\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 22:27:31 2019 +0200\n\n gpio: remove ks8695 driver\n \n The platform is getting removed, so there are no remaining\n users of this driver.\n \n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Link: https://lore.kernel.org/r/20190809202749.742267-4-arnd@arndb.de\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit bf4b87b0d06aa3860e5b23d7b144a42c657d58c2\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 8 13:17:20 2019 +0530\n\n pinctrl: freescale: imx: Add of_node_put() before return\n \n Each iteration of for_each_child_of_node() puts the previous node;\n however, in the case of a return from the middle of the loop, there is no\n put, thus causing a memory leak. Hence put of_node_put() statements as\n required before two mid-loop return statements.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190808074720.15754-1-nishkadg.linux@gmail.com\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 5a6bc29032700bcc1fb752dd58f73179a78d115b\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 8 13:13:29 2019 +0530\n\n pinctrl: sprd: Add of_node_put() before return to prevent memory leak\n \n Each iteration of for_each_child_of_node puts the previous node, but in\n the case of a return from the middle of the loop, there is no put, thus\n causing a memory leak. Hence add an of_node_put before the return in\n two places.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190808074329.15579-1-nishkadg.linux@gmail.com\n Reviewed-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 474137c9111a533aeaa46c1630bb9fb106a46e30\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 8 13:24:56 2019 +0530\n\n pinctrl: stm32: stm32: Add of_node_put() before return\n \n Each iteration of for_each_child_of_node and\n for_each_available_child_of_node puts the previous node, but in\n the case of a return from the middle of the loop, there is no put, thus\n causing a memory leak. Hence add an of_node_put before the return in\n two places.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190808075457.16109-1-nishkadg.linux@gmail.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 21b2920fb587e570b43973300a11b921c3a61d3e\nAuthor: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\nDate: Tue Aug 6 17:30:37 2019 -0700\n\n pinctrl: aspeed: g6: Remove const specifier from aspeed_g6_sig_expr_set\u0027s ctx parameter\n \n clang errors:\n \n drivers/pinctrl/aspeed/pinctrl-aspeed-g6.c:2325:9: error: incompatible\n pointer types initializing \u0027int (*)(struct aspeed_pinmux_data *, const\n struct aspeed_sig_expr *, bool)\u0027 with an expression of type \u0027int (const\n struct aspeed_pinmux_data *, const struct aspeed_sig_expr *, bool)\u0027\n [-Werror,-Wincompatible-pointer-types]\n .set \u003d aspeed_g6_sig_expr_set,\n ^~~~~~~~~~~~~~~~~~~~~~\n 1 error generated.\n \n Commit 674fa8daa8c9 (\"pinctrl: aspeed-g5: Delay acquisition of regmaps\")\n changed the set function pointer declaration and the g6 one wasn\u0027t\n updated (I assume because it wasn\u0027t merged yet).\n \n Fixes: 2eda1cdec49f (\"pinctrl: aspeed: Add AST2600 pinmux support\")\n Link: https://github.com/ClangBuiltLinux/linux/issues/632\n Signed-off-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190807003037.48457-1-natechancellor@gmail.com\n Reviewed-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit a190d94830c30e41bd5fbdf803a636263208fa2e\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 22:27:43 2019 +0200\n\n usb: remove ehci-w90x900 driver\n \n The ARM w90x900 platform is getting removed, so this driver is obsolete.\n \n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Link: https://lore.kernel.org/r/20190809202749.742267-16-arnd@arndb.de\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 040eb7ba19839accd5889fe249107c9573c23197\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Aug 9 18:40:42 2019 +0100\n\n tty/serial: atmel: remove redundant assignment to ret\n \n Variable ret is initialized to a value that is never read and it\n is re-assigned later. The initialization is redundant and can be\n removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190809174042.6276-1-colin.king@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 605b5b4c90ecc31c5fea234f882bf4248dae4fc8\nAuthor: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\nDate: Sat Aug 10 03:01:30 2019 +0800\n\n parport: parport_serial: Add support for Sunix Multi I/O boards\n \n Sunix Multi I/O boards are different to Timedia\u0027s.\n \n This patch adds proper support for Sunix MIO boards with 1 parallel and\n up to 4 serial ports.\n \n Cc: Morris Ku \u003cmorris_ku@sunix.com\u003e\n Cc: Debbie Liu \u003cdebbie_liu@sunix.com\u003e\n Signed-off-by: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190809190130.30773-2-kai.heng.feng@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a4b7aca30609899a925cbf24b08bfea79883254c\nAuthor: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\nDate: Sat Aug 10 03:01:29 2019 +0800\n\n serial: 8250_pci: Add support for Sunix serial boards\n \n Add support to Sunix serial boards with up to 16 ports.\n \n Sunix board need its own setup callback instead of using Timedia\u0027s, to\n properly support more than 4 ports.\n \n Cc: Morris Ku \u003cmorris_ku@sunix.com\u003e\n Cc: Debbie Liu \u003cdebbie_liu@sunix.com\u003e\n Signed-off-by: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190809190130.30773-1-kai.heng.feng@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7e54c250b4fe561051c684a69a7e840f5724f9e7\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 22:27:30 2019 +0200\n\n serial: remove ks8695 driver\n \n The platform is getting removed, so there are no more users\n of this driver.\n \n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Link: https://lore.kernel.org/r/20190809202749.742267-3-arnd@arndb.de\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 62ad42ec9c4933f8973e5a28d9abdb63f8f901a0\nAuthor: Roman Mashak \u003cmrv@mojatatu.com\u003e\nDate: Fri Aug 9 18:46:40 2019 -0400\n\n tc-testing: added tdc tests for matchall filter\n \n Signed-off-by: Roman Mashak \u003cmrv@mojatatu.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f887427b2cecfb57165f0207b33aed89dd29ab61\nAuthor: David Ahern \u003cdsahern@gmail.com\u003e\nDate: Fri Aug 9 16:13:38 2019 -0700\n\n selftests: Fix detection of nettest command in fcnal-test\n \n Most of the tests run by fcnal-test.sh relies on the nettest command.\n Rather than trying to cover all of the individual tests, check for the\n binary only at the beginning.\n \n Also removes the need for log_error which is undefined.\n \n Fixes: 6f9d5cacfe07 (\"selftests: Setup for functional tests for fib and socket lookups\")\n Signed-off-by: David Ahern \u003cdsahern@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6bd744b723d229ac38eeea2cf0717d8cd2858032\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 14:56:02 2019 -0700\n\n Input: remove w90x900 touchscreen driver\n \n The ARM w90x900 platform is getting removed, so this driver is obsolete.\n \n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 060157e1dbc133075a2e20786d6ff6d4b41909f9\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Aug 9 14:55:43 2019 -0700\n\n Input: remove w90x900 keyboard driver\n \n The ARM w90x900 platform is getting removed, so this driver is obsolete.\n \n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit b51c225e6c4e987e131b8b1332f66969382bf328\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Sat Aug 3 00:48:28 2019 +0800\n\n net/mlx5e: Use refcount_t for refcount\n \n refcount_t is better for reference counters since its\n implementation can prevent overflows.\n So convert atomic_t ref counters to refcount_t.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit c938451f6b9ccbf25eceb27fe1d1c24fd98af923\nAuthor: Parav Pandit \u003cparav@mellanox.com\u003e\nDate: Sat Jul 20 22:20:58 2019 -0500\n\n net/mlx5e: Use vhca_id in generating representor port_index\n \n It is desired to use unique port indices when multiple pci devices\u0027\n devlink instance have the same switch-id.\n \n Make use of vhca-id to generate such unique devlink port indices.\n \n Signed-off-by: Parav Pandit \u003cparav@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 724ee17912c9a8e5362751a22fd955166ed0550a\nAuthor: Parav Pandit \u003cparav@mellanox.com\u003e\nDate: Fri Jul 26 13:42:04 2019 -0500\n\n net/mlx5e: Simplify querying port representor parent id\n \n System image GUID doesn\u0027t depend on eswitch switchdev mode.\n \n Hence, remove the check which simplifies the code.\n \n Signed-off-by: Parav Pandit \u003cparav@mellanox.com\u003e\n Reviewed-by: Vu Pham \u003cvuhuong@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit ef2e4094e076858343ea1202046443f642a245cd\nAuthor: Parav Pandit \u003cparav@mellanox.com\u003e\nDate: Fri Jul 26 08:26:52 2019 -0500\n\n net/mlx5: E-switch, Removed unused hwid\n \n Currently mlx5_eswitch_rep stores same hw ID for all representors.\n However it is never used from this structure.\n It is always used from mlx5_vport.\n \n Hence, remove unused field.\n \n Signed-off-by: Parav Pandit \u003cparav@mellanox.com\u003e\n Reviewed-by: Vu Pham \u003cvuhuong@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit d589e785baf5e48ee80a4fbfed96661a4c7c8c8d\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Thu Aug 8 17:01:33 2019 +0300\n\n net/mlx5e: Allow concurrent creation of encap entries\n \n Encap entries creation is fully synchronized by encap_tbl_lock. In order to\n allow concurrent allocation of hardware resources used to offload\n encapsulation, extend mlx5e_encap_entry with \u0027res_ready\u0027 completion. Move\n call to mlx5e_tc_tun_create_header_ipv{4|6}() out of encap_tbl_lock\n critical section. Modify code that attaches new flows to existing encap to\n wait for \u0027res_ready\u0027 completion before using the entry. Insert encap entry\n to table before provisioning it to hardware and modify all users of the\n encap table to verify that encap was fully initialized by checking\n completion result for non-zero value (and to wait for \u0027res_ready\u0027\n completion, if necessary).\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 61086f391044fd587af9d70a9b8f6f800dd474ba\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Fri Aug 2 22:21:56 2019 +0300\n\n net/mlx5e: Protect encap hash table with mutex\n \n To remove dependency on rtnl lock, protect encap hash table from concurrent\n modifications with new \"encap_tbl_lock\" mutex. Use the mutex to protect\n internal encap entry state from concurrent modification. This is necessary\n because a flow can be attached to multiple encap entries simultaneously,\n which significantly complicates using finer grained per-entry lock.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 948993f2beebbdc1c9d926cfdad9827cf6bb67c0\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Sun Jun 3 20:31:47 2018 +0300\n\n net/mlx5e: Extend encap entry with reference counter\n \n List of flows attached to encap entry is used as implicit reference\n counter (encap entry is deallocated when list becomes free) and as a\n mechanism to obtain encap entry that flow is attached to (through list\n head). This is not safe when concurrent modification of list of flows\n attached to encap entry is possible. Proper atomic reference counter is\n required to support concurrent access.\n \n As a preparation for extending encap with reference counting, extract code\n that lookups and deletes encap entry into standalone put/get helpers. In\n order to remove this dependency on external locking, extend encap entry\n with reference counter to manage its lifetime and extend flow structure\n with direct pointer to encap entry that flow is attached to.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Jianbo Liu \u003cjianbol@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit a734d007173119fe8e7bde1689ee4123c529e238\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Thu Aug 8 16:53:15 2019 +0300\n\n net/mlx5e: Allow concurrent creation of mod_hdr entries\n \n Mod_hdr entries creation is fully synchronized by mod_hdr_tbl-\u003elock. In\n order to allow concurrent allocation of hardware resources used to offload\n header rewrite, extend mlx5e_mod_hdr_entry with \u0027res_ready\u0027 completion.\n Move call to mlx5_modify_header_alloc() out of mod_hdr_tbl-\u003elock critical\n section. Modify code that attaches new flows to existing mh to wait for\n \u0027res_ready\u0027 completion before using the entry. Insert mh to mod_hdr table\n before provisioning it to hardware and modify all users of mod_hdr table to\n verify that mh was fully initialized by checking completion result for\n negative value (and to wait for \u0027res_ready\u0027 completion, if necessary).\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit d2faae25c3050a87c8ff965a7939e999e3154b62\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Fri Aug 9 13:20:48 2019 +0300\n\n net/mlx5e: Protect mod_hdr hash table with mutex\n \n To remove dependency on rtnl lock, protect mod_hdr hash table from\n concurrent modifications with new mutex.\n \n Implement helper function to get flow namespace to prevent code\n duplication.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 83a52f0d525587de65b2bc979a12b4dfad9ea82a\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Fri Jun 8 22:10:09 2018 +0300\n\n net/mlx5e: Protect mod header entry flows list with spinlock\n \n To remove dependency on rtnl lock, extend mod header entry with spinlock\n and use it to protect list of flows attached to mod header entry from\n concurrent modifications.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Jianbo Liu \u003cjianbol@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit dd58edc328cec1a0d837f3f2f41e9955ec623e3e\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Fri Jun 1 21:47:43 2018 +0300\n\n net/mlx5e: Extend mod header entry with reference counter\n \n List of flows attached to mod header entry is used as implicit reference\n counter (mod header entry is deallocated when list becomes free) and as a\n mechanism to obtain mod header entry that flow is attached to (through list\n head). This is not safe when concurrent modification of list of flows\n attached to mod header entry is possible. Proper atomic reference counter\n is required to support concurrent access.\n \n As a preparation for extending mod header with reference counting, extract\n code that lookups and deletes mod header entry into standalone put/get\n helpers. In order to remove this dependency on external locking, extend mod\n header entry with reference counter to manage its lifetime and extend flow\n structure with direct pointer to mod header entry that flow is attached to.\n \n To remove code duplication between legacy and switchdev mode\n implementations that both support mod_hdr functionality, store mod_hdr\n table in dedicated structure used by both fdb and kernel namespaces. New\n table structure is extended with table lock by one of the following patches\n in this series. Implement helper function to get correct mod_hdr table\n depending on flow namespace.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Jianbo Liu \u003cjianbol@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit db76ca2424fe28923aaec5e2187e886b025a914c\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Thu Aug 1 16:54:54 2019 +0300\n\n net/mlx5e: Allow concurrent creation of hairpin entries\n \n Hairpin entries creation is fully synchronized by hairpin_tbl_lock. In\n order to allow concurrent initialization of mlx5e_hairpin structure\n instances and provisioning of hairpin entries to hardware, extend\n mlx5e_hairpin_entry with \u0027res_ready\u0027 completion. Move call to\n mlx5e_hairpin_create() out of hairpin_tbl_lock critical section. Modify\n code that attaches new flows to existing hpe to wait for \u0027res_ready\u0027\n completion before using the hpe. Insert hpe to hairpin table before\n provisioning it to hardware and modify all users of hairpin table to verify\n that hpe was fully initialized by checking hpe-\u003ehp pointer (and to wait for\n \u0027res_ready\u0027 completion, if necessary).\n \n Modify dead peer update event handling function to save hpe\u0027s to temporary\n list with their reference counter incremented. Wait for completion of hpe\u0027s\n in temporary list and update their \u0027peer_gone\u0027 flag outside of\n hairpin_tbl_lock critical section.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit b32accda8afd36d7de1581375467fd3642d3c73e\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Wed Jul 31 18:19:06 2019 +0300\n\n net/mlx5e: Protect hairpin hash table with mutex\n \n To remove dependency on rtnl lock, protect hairpin hash table from\n concurrent modifications with new \"hairpin_tbl_lock\" mutex.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 73edca736e581a685c98d3928b4cc029932d2d5a\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Thu Jun 7 23:01:40 2018 +0300\n\n net/mlx5e: Protect hairpin entry flows list with spinlock\n \n To remove dependency on rtnl lock, extend hairpin entry with spinlock and\n use it to protect list of flows attached to hairpin entry from concurrent\n modifications.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Jianbo Liu \u003cjianbol@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit e4f9abbd38e468c73710ad3678005ad05f79c818\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Fri Jun 8 19:26:50 2018 +0300\n\n net/mlx5e: Extend hairpin entry with reference counter\n \n List of flows attached to hairpin entry is used as implicit reference\n counter (hairpin entry is deallocated when list becomes free) and as a\n mechanism to obtain hairpin entry that flow is attached to (through list\n head). This is not safe when concurrent modification of list of flows\n attached to hairpin entry is possible. Proper atomic reference counter is\n required to support concurrent access.\n \n As a preparation for extending hairpin with reference counting, extract\n code that deletes hairpin entry into standalone function. In order to\n remove this dependency on external locking, extend hairpin entry with\n reference counter to manage its lifetime and extend flow structure with\n direct pointer to hairpin entry that flow is attached to.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Jianbo Liu \u003cjianbol@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit e2fb6f7c0df1cb9d5f89c1b6a9d3caf434c040bb\nAuthor: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\nDate: Tue Aug 6 09:55:20 2019 +0200\n\n dt-bindings: arm: amlogic: fix x96-max/sei510 section in amlogic.yaml\n \n Move amediatech,x96-max and seirobotics,sei510 to the S905D2 section and\n update the S905D2 description to S905D2/X2/Y2.\n \n Signed-off-by: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Acked-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit eddd98600b382d80cd71e44150eec5de4a0830a4\nAuthor: Guojia Liao \u003cliaoguojia@huawei.com\u003e\nDate: Fri Aug 9 10:31:18 2019 +0800\n\n net: hns3: refine some macro definitions\n \n Macro arguments should be enclosed in parentheses, in case of\n expression argument, but parentheses of pure number in macro\n definition should be removed for simplicity.\n \n Signed-off-by: Guojia Liao \u003cliaoguojia@huawei.com\u003e\n Signed-off-by: Guangbin Huang \u003chuangguangbin2@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 427a7bff66c8a9f8cf68384e24093fc4bf7763a1\nAuthor: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\nDate: Fri Aug 9 10:31:17 2019 +0800\n\n net: hns3: add handshake with VF for PF reset\n \n Before PF asserting function reset, it should make sure\n that all its VFs have been ready, otherwise, it will cause\n some hardware errors.\n \n So this patch adds function hclge_func_reset_sync_vf() to\n synchronize VF before asserting PF function reset. For new\n firmware which supports command HCLGE_OPC_QUERY_VF_RST_RDY,\n we will try to query VFs\u0027 ready status within 30 seconds.\n And keep the old implementation for compatible with firmware\n which does not support this command.\n \n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Reviewed-by: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 615466ce415a8a94d3c2cbd5ac35aa29bd189ebe\nAuthor: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\nDate: Fri Aug 9 10:31:16 2019 +0800\n\n net: hns3: refine MAC pause statistics querying function\n \n This patch refines the interface for querying MAC pause\n statistics, and adds structure hns3_mac_stats to keep the\n count of TX \u0026 RX.\n \n Signed-off-by: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit dec8466001f740868f01a3cb75b21174197be85e\nAuthor: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\nDate: Fri Aug 9 10:31:15 2019 +0800\n\n net: hns3: add function display NCL_CONFIG info\n \n This adds a new function hclge_ncl_config_data_print()\n to print the data of NCL_CONFIG, to make the code more\n readable. Also, using macro replaces some magic number.\n \n Signed-off-by: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 42611b70f8be12c46906c869f5d819ac70dfd1c7\nAuthor: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\nDate: Fri Aug 9 10:31:14 2019 +0800\n\n net: hns3: add check for max TX BD num for tso and non-tso case\n \n Hardware supports up to 8 TX BD for non-TSO skb and 63 TX\n BD for TSO skb. Currently hns3 driver does not check the max\n BD num that required by a skb before filling desc, which may\n cause the hardware to issue a RAS error throug PCIe AER.\n \n This patch adds the max BD num check before filling desc,\n if the bd num is not within the hardware limit, it will\n record the error by ring-\u003estats.sw_err_cnt counter and\n free the skb.\n \n This patch also cleans up the hns3_nic_bd_num function by\n changing the return type and removing an unnecessary check.\n \n Signed-off-by: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b20d7fe51e0de28982e115b8ec0647066d73d206\nAuthor: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\nDate: Fri Aug 9 10:31:13 2019 +0800\n\n net: hns3: add some statitics info to tx process\n \n This patch adds tx_vlan_err, tx_l4_proto_err, tx_l2l3l4_err\n and tx_tso_err counter to tx process, in order to better\n debug the desc filling error.\n \n This patch also adds a missing u64_stats_update_* around\n ring-\u003estats.sw_err_cnt and adds hns3_rl_err to limit the\n error printing in the IO patch.\n \n Signed-off-by: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ddb54554fa517be1639f10c6f2828429a871368b\nAuthor: Guangbin Huang \u003chuangguangbin2@huawei.com\u003e\nDate: Fri Aug 9 10:31:12 2019 +0800\n\n net: hns3: add DFX registers information for ethtool -d\n \n Now we can use ethtool -d command to dump some registers. However,\n these registers information is not enough to find out where the problem is.\n \n This patch adds DFX registers information after original registers\n when use ethtool -d commmand to dump registers. Also, using macro\n replaces some related magic number.\n \n Signed-off-by: Guangbin Huang \u003chuangguangbin2@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit aacbe27e82f01f6e0e3c2ab859d51d0823604ec8\nAuthor: Yonglong Liu \u003cliuyonglong@huawei.com\u003e\nDate: Fri Aug 9 10:31:11 2019 +0800\n\n net: hns3: modify how pause options is displayed\n \n Currently, the pause options of HNS3 shown like this:\n \"RX/TX\" is always the same with \"RX negotiated/TX negotiated\".\n Because of the driver covered the value of \"RX/TX\" with the value\n of \"RX negotiated/TX negotiated\" after adjust link.\n \n This patch records the pause configurations of the user, and never\n covered them in adjust link.\n \n Signed-off-by: Yonglong Liu \u003cliuyonglong@huawei.com\u003e\n Reviewed-by: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7ac243f99d61f4580a72f62e638ed4e04aa52aa1\nAuthor: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\nDate: Fri Aug 9 10:31:10 2019 +0800\n\n net: hns3: add input length check for debugfs write function\n \n If the input length reaches the maximum value of size_t, the reverse is\n triggered when 1 is added. In addition, there is no need to have such a\n large length. Therefore, the input length should be checked and the value\n should be less than or equal to 1024.\n \n Signed-off-by: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit eb977d996e9316c66e4d9908fd12d682e4db14e9\nAuthor: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\nDate: Fri Aug 9 10:31:09 2019 +0800\n\n net: hns3: clean up for vlan handling in hns3_fill_desc_vtags\n \n This patch refactors the hns3_fill_desc_vtags function\n by avoiding passing too many parameters, reducing indent\n level and some other clean up.\n \n This patch also adds the hns3_fill_skb_desc function to\n fill the first desc of a skb.\n \n Signed-off-by: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 130509213baeb3cada40c19dd87341a6787f8b23\nAuthor: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\nDate: Fri Aug 9 10:31:08 2019 +0800\n\n net: hns3: fix interrupt clearing error for VF\n \n Currently, VF driver has two kinds of interrupts, reset \u0026 CMDQ RX.\n For revision 0x21, according to the UM, each interrupt should be\n cleared by write 0 to the corresponding bit, but the implementation\n writes 0 to the whole register in fact, it will clear other\n interrupt at the same time, then the VF will loss the interrupt.\n But for revision 0x20, this interrupt clear register is a read \u0026\n write register, for compatible, we just keep the old implementation\n for 0x20.\n \n This patch fixes it, also, adds a new register for reading the interrupt\n status according to hardware user manual.\n \n Fixes: e2cb1dec9779 (\"net: hns3: Add HNS3 VF HCL(Hardware Compatibility Layer) Support\")\n Fixes: b90fcc5bd904 (\"net: hns3: add reset handling for VF when doing Core/Global/IMP reset\")\n \n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Reviewed-by: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9e6717af61dfe6de304562602364a19d70c886d4\nAuthor: Zhongzhu Liu \u003cliuzhongzhu@huawei.com\u003e\nDate: Fri Aug 9 10:31:07 2019 +0800\n\n net: hns3: fix GFP flag error in hclge_mac_update_stats()\n \n When CONFIG_DEBUG_ATOMIC_SLEEP on, calling kzalloc with\n GFP_KERNEL in hclge_mac_update_stats() will get below warning:\n \n [ 52.514677] BUG: sleeping function called from invalid context at mm/slab.h:501\n [ 52.522051] in_atomic(): 0, irqs_disabled(): 0, pid: 1015, name: ifconfig\n [ 52.528827] 2 locks held by ifconfig/1015:\n [ 52.532921] #0: (____ptrval____) (\u0026p-\u003elock){....}, at: seq_read+0x54/0x748\n [ 52.539878] #1: (____ptrval____) (rcu_read_lock){....}, at: dev_seq_start+0x0/0x140\n [ 52.547610] CPU: 16 PID: 1015 Comm: ifconfig Not tainted 5.3.0-rc3-00697-g20b80be #98\n [ 52.555408] Hardware name: Huawei TaiShan 2280 V2/BC82AMDC, BIOS 2280-V2 CS V3.B050.01 08/08/2019\n [ 52.564242] Call trace:\n [ 52.566687] dump_backtrace+0x0/0x1f8\n [ 52.570338] show_stack+0x14/0x20\n [ 52.573646] dump_stack+0xb4/0xec\n [ 52.576950] ___might_sleep+0x178/0x198\n [ 52.580773] __might_sleep+0x74/0xe0\n [ 52.584338] __kmalloc+0x244/0x2d8\n [ 52.587744] hclge_mac_update_stats+0xc8/0x1f8 [hclge]\n [ 52.592870] hclge_update_stats+0xe0/0x170 [hclge]\n [ 52.597651] hns3_nic_get_stats64+0xa0/0x458 [hns3]\n [ 52.602514] dev_get_stats+0x58/0x138\n [ 52.606165] dev_seq_printf_stats+0x8c/0x280\n [ 52.610420] dev_seq_show+0x14/0x40\n [ 52.613898] seq_read+0x574/0x748\n [ 52.617205] proc_reg_read+0xb4/0x108\n [ 52.620857] __vfs_read+0x54/0xa8\n [ 52.624162] vfs_read+0xa0/0x190\n [ 52.627380] ksys_read+0xc8/0x178\n [ 52.630685] __arm64_sys_read+0x40/0x50\n [ 52.634509] el0_svc_common.constprop.0+0x120/0x1e0\n [ 52.639369] el0_svc_handler+0x50/0x90\n [ 52.643106] el0_svc+0x8/0xc\n \n So this patch uses GFP_ATOMIC instead of GFP_KERNEL to fix it.\n \n Fixes: d174ea75c96a (\"net: hns3: add statistics for PFC frames and MAC control frames\")\n Signed-off-by: Zhongzhu Liu \u003cliuzhongzhu@huawei.com\u003e\n Reviewed-by: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ca497fb6aa9fbd3b0a87fd0a71e9e1df2600ac30\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 9 09:49:23 2019 +0800\n\n taprio: remove unused variable \u0027entry_list_policy\u0027\n \n net/sched/sch_taprio.c:680:32: warning:\n entry_list_policy defined but not used [-Wunused-const-variable\u003d]\n \n One of the points of commit a3d43c0d56f1 (\"taprio: Add support adding\n an admin schedule\") is that it removes support (it now returns \"not\n supported\") for schedules using the TCA_TAPRIO_ATTR_SCHED_SINGLE_ENTRY\n attribute (which were never used), the parsing of those types of schedules\n was the only user of this policy. So removing this policy should be fine.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Suggested-by: Vinicius Costa Gomes \u003cvinicius.gomes@intel.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a7eb6a4f2560d5ae64bfac98d79d11378ca2de6c\nAuthor: Holger Hoffstätte \u003cholger@applied-asynchrony.com\u003e\nDate: Fri Aug 9 00:02:40 2019 +0200\n\n r8169: fix performance issue on RTL8168evl\n \n Disabling TSO but leaving SG active results is a significant\n performance drop. Therefore disable also SG on RTL8168evl.\n This restores the original performance.\n \n Fixes: 93681cd7d94f (\"r8169: enable HW csum and TSO\")\n Signed-off-by: Holger Hoffstätte \u003cholger@applied-asynchrony.com\u003e\n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1555e6fdf062e2ae2514c67f46d475c7ebcce10e\nAuthor: Josh Hunt \u003cjohunt@akamai.com\u003e\nDate: Wed Aug 7 19:52:30 2019 -0400\n\n tcp: Update TCP_BASE_MSS comment\n \n TCP_BASE_MSS is used as the default initial MSS value when MTU probing is\n enabled. Update the comment to reflect this.\n \n Suggested-by: Neal Cardwell \u003cncardwell@google.com\u003e\n Signed-off-by: Josh Hunt \u003cjohunt@akamai.com\u003e\n Signed-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\n Acked-by: Neal Cardwell \u003cncardwell@google.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c04b79b6cfd714144f6a2cf359603d82ee631e62\nAuthor: Josh Hunt \u003cjohunt@akamai.com\u003e\nDate: Wed Aug 7 19:52:29 2019 -0400\n\n tcp: add new tcp_mtu_probe_floor sysctl\n \n The current implementation of TCP MTU probing can considerably\n underestimate the MTU on lossy connections allowing the MSS to get down to\n 48. We have found that in almost all of these cases on our networks these\n paths can handle much larger MTUs meaning the connections are being\n artificially limited. Even though TCP MTU probing can raise the MSS back up\n we have seen this not to be the case causing connections to be \"stuck\" with\n an MSS of 48 when heavy loss is present.\n \n Prior to pushing out this change we could not keep TCP MTU probing enabled\n b/c of the above reasons. Now with a reasonble floor set we\u0027ve had it\n enabled for the past 6 months.\n \n The new sysctl will still default to TCP_MIN_SND_MSS (48), but gives\n administrators the ability to control the floor of MSS probing.\n \n Signed-off-by: Josh Hunt \u003cjohunt@akamai.com\u003e\n Signed-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\n Acked-by: Neal Cardwell \u003cncardwell@google.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 0a9b26306d6a10573cdc4dbbc804bfff264ae878\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Fri Aug 9 07:31:16 2019 +0100\n\n drm/i915: split out uncore_mmio_debug\n \n Multiple uncore structures will share the debug infrastructure, so\n move it to a common place and add extra locking around it.\n Also, since we now have a separate object, it is cleaner to have\n dedicated functions working on the object to stop and restart the\n mmio debug. Apart from the cosmetic changes, this patch introduces\n 2 functional updates:\n \n - All calls to check_for_unclaimed_mmio will now return false when\n the debug is suspended, not just the ones that are active only when\n i915_modparams.mmio_debug is set. If we don\u0027t trust the result of the\n check while a user is doing mmio access then we shouldn\u0027t attempt the\n check anywhere.\n \n - i915_modparams.mmio_debug is not save/restored anymore around user\n access. The value is now never touched by the kernel while debug is\n disabled so no need for save/restore.\n \n v2: squash mmio_debug patches, restrict mmio_debug lock usage (Chris)\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190809063116.7527-1-chris@chris-wilson.co.uk\n\ncommit 72e67f04637432f91e4cc5e8e4f7eb4e38461e8e\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 8 18:22:26 2019 +0100\n\n drm/i915: Stop reconfiguring our shmemfs mountpoint\n \n The filesystem reconfigure API is undergoing a transition, breaking our\n current code. As we only set the default options, we can simply remove\n the call to s_op-\u003eremount_fs(). In the future, when HW permits, we can\n try re-enabling huge page support, albeit as suggested with new per-file\n controls.\n \n Reported-by: Sergey Senozhatsky \u003csergey.senozhatsky@gmail.com\u003e\n Reported-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Suggested-by: Hugh Dickins \u003chughd@google.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Cc: Hugh Dickins \u003chughd@google.com\u003e\n Cc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n Cc: Sergey Senozhatsky \u003csergey.senozhatsky@gmail.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808172226.18306-1-chris@chris-wilson.co.uk\n\ncommit 75d0a7f31eec8ec4a53b4485905800e09dc5091f\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 9 19:25:18 2019 +0100\n\n drm/i915: Lift timeline into intel_context\n \n Move the timeline from being inside the intel_ring to intel_context\n itself. This saves much pointer dancing and makes the relations of the\n context to its timeline much clearer.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190809182518.20486-4-chris@chris-wilson.co.uk\n\ncommit 48ae397b6b935c6733f15476c338df27eac9293c\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 9 19:25:17 2019 +0100\n\n drm/i915: Push the ring creation flags to the backend\n \n Push the ring creation flags from the outer GEM context to the inner\n intel_context to avoid an unsightly back-reference from inside the\n backend.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Andi Shyti \u003candi.shyti@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190809182518.20486-3-chris@chris-wilson.co.uk\n\ncommit 4c60b1aaa255207eea2892d2d55b718db90d1ad8\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 9 19:25:16 2019 +0100\n\n drm/i915/gt: Make deferred context allocation explicit\n \n Refactor the backends to handle the deferred context allocation in a\n consistent manner, and allow calling it as an explicit first step in\n pinning a context for the first time. This should make it easier for\n backends to keep track of partially constructed contexts from\n initialisation.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190809182518.20486-2-chris@chris-wilson.co.uk\n\ncommit 72e277759340dacdf3d68b0b2b4eae267e601d55\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 9 19:25:15 2019 +0100\n\n drm/i915: Remove i915_gem_context_create_gvt()\n \n As we are phasing out using the GEM context for internal clients that\n need to manipulate logical context state directly, remove the\n constructor for the GVT context. We are not using it for anything other\n than default setup and allocation of an i915_ppgtt.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190809182518.20486-1-chris@chris-wilson.co.uk\n\ncommit 45f5d5a9e34d3fe4140a9a3b5f7ebe86c252440a\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Jul 31 09:48:01 2019 +0200\n\n arm64: dts: renesas: r8a77995: draak: Fix backlight regulator name\n \n Currently there are two nodes named \"regulator1\" in the Draak DTS: a\n 3.3V regulator for the eMMC and the LVDS decoder, and a 12V regulator\n for the backlight. This causes the former to be overwritten by the\n latter.\n \n Fix this by renaming all regulators with numerical suffixes to use named\n suffixes, which are less likely to conflict.\n \n Fixes: 4fbd4158fe8967e9 (\"arm64: dts: renesas: r8a77995: draak: Add backlight\")\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n\ncommit 3a5e523479c49b082b8ac291a1a9fbd035c06df5\nAuthor: Jiri Pirko \u003cjiri@mellanox.com\u003e\nDate: Fri Aug 9 15:27:15 2019 +0200\n\n devlink: remove pointless data_len arg from region snapshot create\n \n The size of the snapshot has to be the same as the size of the region,\n therefore no need to pass it again during snapshot creation. Remove the\n arg and use region-\u003esize instead.\n \n Signed-off-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a55022706068c461f13552fbbb4e0f8cf056933f\nAuthor: Kevin Hilman \u003ckhilman@baylibre.com\u003e\nDate: Thu Aug 8 09:44:48 2019 -0700\n\n arm64: dts: amlogic: g12 CPU timers stop in suspend\n \n The Arm per-CPU architected timers stop ticking in suspend, when the\n SCP powers down the CPUs. Flag that in the DT.\n \n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit b96d4e92709b5c4a0ab1494be76415a466c9c0c3\nAuthor: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\nDate: Wed Jul 31 14:39:59 2019 +0200\n\n arm64: dts: meson-g12b: support a311d and s922x cpu operating points\n \n Meson g12b ships with a low-speed (S922X) and high-speed (A311D) variant\n so remove cpu_opp_table nodes in meson-g12b.dtsi and create two new dtsi\n that can be included in device-specific dts files. Opp points were taken\n from the vendor BSP kernel.\n \n Also make meson-g12b-odroid-n2.dts include the new meson-g12b-s922x.dtsi.\n \n Signed-off-by: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Tested-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 8f920256e9b661c85a01e7334eada2c7c8970493\nAuthor: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\nDate: Wed Jul 31 14:39:58 2019 +0200\n\n dt-bindings: arm: amlogic: add support for the Khadas VIM3\n \n The Khadas VIM3 uses the Amlogic S922X or A311S SoC, both based on the\n Amlogic G12B SoC family, on a board with the same form factor as the\n VIM/VIM2 models. It ships in two variants; basic and\n pro which differ in RAM and eMMC size:\n \n - 2GB (basic) or 4GB (pro) LPDDR4 RAM\n - 16GB (basic) or 32GB (pro) eMMC 5.1 storage\n - 16MB SPI flash\n - 10/100/1000 Base-T Ethernet\n - AP6398S Wireless (802.11 a/b/g/n/ac, BT5.0)\n - HDMI 2.1 video\n - 1x USB 2.0 + 1x USB 3.0 ports\n - 1x USB-C (power) with USB 2.0 OTG\n - 3x LED\u0027s (1x red, 1x blue, 1x white)\n - 3x buttons (power, function, reset)\n - IR receiver\n - M2 socket with PCIe, USB, ADC \u0026 I2C\n - 40pin GPIO Header\n - 1x micro SD card slot\n \n Signed-off-by: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit a0250352e32b961397241b41771c8e629b5bbc61\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Wed Jul 31 14:39:57 2019 +0200\n\n dt-bindings: arm: amlogic: add bindings for the Amlogic G12B based A311D SoC\n \n Add a specific compatible for the Amlogic G12B bases A311D SoC used\n in the Khadas VIM3.\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit a742eda915e36732628e1a546e2deaeceb735c68\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Wed Jul 31 14:39:56 2019 +0200\n\n dt-bindings: arm: amlogic: add bindings for G12B based S922X SoC\n \n Add a specific compatible for the Amlogic G12B family based S922X SoC\n to differentiate with the A311D SoC from the same family.\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 4be247f79ff4e21ba2ae8f94e8a8b4d6956e6be6\nAuthor: Maxime Jourdan \u003cmjourdan@baylibre.com\u003e\nDate: Fri Jul 26 14:46:39 2019 +0200\n\n arm64: dts: meson: add video decoder entries\n \n This enables the video decoder for GXBB, GXL and GXM chips\n \n Signed-off-by: Maxime Jourdan \u003cmjourdan@baylibre.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 1f11d61182aeb97473490ce5d781d5f083ba33a1\nAuthor: Maxime Jourdan \u003cmjourdan@baylibre.com\u003e\nDate: Fri Jul 26 14:46:38 2019 +0200\n\n arm64: dts: meson-gx: add video decoder entry\n \n Add the base video decoder node compatible with the meson vdec driver,\n for GX* chips.\n \n Signed-off-by: Maxime Jourdan \u003cmjourdan@baylibre.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 0b34189b0875040c80bfb99f71554411fadfb89f\nAuthor: Maxime Jourdan \u003cmjourdan@baylibre.com\u003e\nDate: Fri Jul 26 14:46:37 2019 +0200\n\n dt-bindings: media: amlogic,vdec: add default compatible\n \n The first version of the bindings is missing a generic compatible that\n is used by the base node (GX), and then extended by the SoC device trees\n (GXBB, GXL, GXM)\n \n Also change the example to use \"video-codec\" instead of \"video-decoder\",\n as the former is the one used in almost all cases when it comes to video\n decode/encode accelerators.\n \n Signed-off-by: Maxime Jourdan \u003cmjourdan@baylibre.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit ef68984eab6bbe30a2a921c9440213cf5dded75b\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Thu Jul 18 11:03:01 2019 +0200\n\n arm64: dts: meson: add ethernet fifo sizes\n \n If unspecified in DT, the fifo sizes are not automatically detected by\n the dwmac1000 dma driver and the reported fifo sizes default to 0.\n Because of this, flow control will be turned off on the device.\n \n Add the fifo sizes provided by the datasheets in the SoC in DT so\n flow control may be enabled if necessary.\n \n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 53fbee339a9524787f698f10bd676c26643198d5\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Mon Jul 29 15:26:21 2019 +0200\n\n arm64: dts: meson-g12b: add cpus OPP tables\n \n Add the OPP table taken from the HardKernel Odroid-N2 DTS.\n \n The Amlogic G12B SoC seems to available in 2 types :\n - low-speed: Cortex-A73 Cluster up to 1,704GHz\n - high-speed: Cortex-A73 Cluster up to 2.208GHz\n \n The Cortex-A73 Cluster can be clocked up to 1,896GHz for both types.\n \n The Vendor Amlogic A311D OPP table are slighly different, with lower\n voltages than the HardKernel S922X tables but seems to be high-speed type.\n \n This adds the conservative OPP table with the S922X higher voltages\n and the maximum low-speed OPP frequency.\n \n The values were tested to be stable on an HardKernel Odroid-N2 board\n running the arm64 cpuburn at [1] and cycling between all the possible\n cpufreq translations for both clusters and checking the final frequency\n using the clock-measurer, script at [2].\n \n [1] https://github.com/ssvb/cpuburn-arm/blob/master/cpuburn-a53.S\n [2] https://gist.github.com/superna9999/d4de964dbc0f84b7d527e1df2ddea25f\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Tested-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit e9bc0765cc1274b7221360f462a3a993b41cac92\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Mon Jul 29 15:26:20 2019 +0200\n\n arm64: dts: meson-g12a: enable DVFS on G12A boards\n \n Enable DVFS for the U200, SEI520 and X96-Max Amlogic G12A based board\n by setting the clock, OPP and supply for each CPU cores.\n \n The CPU cluster power supply can achieve 0.73V to 1.01V using a PWM\n output clocked at 800KHz with an inverse duty-cycle.\n \n DVFS has been tested by running the arm64 cpuburn at [1] and cycling\n between all the possible cpufreq translations and checking the final\n frequency using the clock-measurer, script at [2].\n \n [1] https://github.com/ssvb/cpuburn-arm/blob/master/cpuburn-a53.S\n [2] https://gist.github.com/superna9999/d4de964dbc0f84b7d527e1df2ddea25f\n \n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Tested-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit b190056fa9eecc016b6a6e41f89a04da14b78bf7\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Mon Jul 29 15:26:19 2019 +0200\n\n arm64: dts: meson-g12a: add cpus OPP table\n \n Add the OPP table taken from the vendor u200 and u211 DTS.\n \n The Amlogic G12A SoC seems to available in 3 types :\n - low-speed: up to 1,8GHz\n - mid-speed: up to 1,908GHz\n - high-speed: up to 2.1GHz\n \n And the S905X2 opp voltages are slightly higher than the S905D2\n OPP voltages for the low-speed table.\n \n This adds the conservative OPP table with the S905X2 higher voltages\n and the maximum low-speed OPP frequency.\n \n The values were tested to be stable on an Amlogic U200 Reference Board,\n SeiRobotics SEI510 and X96 Max Set-Top-Boxes running the arm64 cpuburn\n at [1] and cycling between all the possible cpufreq translations and\n checking the final frequency using the clock-measurer, script at [2].\n \n [1] https://github.com/ssvb/cpuburn-arm/blob/master/cpuburn-a53.S\n [2] https://gist.github.com/superna9999/d4de964dbc0f84b7d527e1df2ddea25f\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Tested-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit cca30c891247fc35655bc24836f46094771d3d47\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Mon Jul 29 15:26:18 2019 +0200\n\n arm64: dts: meson-g12-common: add pwm_a on GPIOE_2 pinmux\n \n Add the ao_pinctrl subnode for the pwm_a function on GPIOE_2.\n \n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Tested-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 1499218c80c99ae73c937c370142c8f6048002d5\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Mon Jul 29 15:26:17 2019 +0200\n\n arm64: dts: move common G12A \u0026 G12B modes to meson-g12-common.dtsi\n \n To simplify the representation of differences betweem the G12A and G12B\n SoCs, move the common nodes into a meson-g12-common.dtsi file and\n express the CPU nodes and differences in meson-g12a.dtsi and meson-g12b.dtsi.\n \n This separation will help for DVFS and future Amlogic SM1 Family support.\n \n The sd_emmc_a quirk is added in the g12a/g12b since since it\u0027s already\n known the sd_emmc_a controller is fixed in the next SM1 SoC family.\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Tested-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 1a9914884db5138682032cf69f2d55739f236c80\nAuthor: Eric Dumazet \u003cedumazet@google.com\u003e\nDate: Fri Aug 9 05:04:47 2019 -0700\n\n tcp: batch calls to sk_flush_backlog()\n \n Starting from commit d41a69f1d390 (\"tcp: make tcp_sendmsg() aware of socket backlog\")\n loopback flows got hurt, because for each skb sent, the socket receives an\n immediate ACK and sk_flush_backlog() causes extra work.\n \n Intent was to not let the backlog grow too much, but we went a bit too far.\n \n We can check the backlog every 16 skbs (about 1MB chunks)\n to increase TCP over loopback performance by about 15 %\n \n Note that the call to sk_flush_backlog() handles a single ACK,\n thanks to coalescing done on backlog, but cleans the 16 skbs\n found in rtx rb-tree.\n \n Reported-by: Soheil Hassas Yeganeh \u003csoheil@google.com\u003e\n Signed-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\n Acked-by: Soheil Hassas Yeganeh \u003csoheil@google.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7fd69b0ba48a2b2d8e5b4f0945b28d3839a7705a\nAuthor: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\nDate: Tue Jul 16 18:12:24 2019 -0400\n\n ipv4: Add lockdep condition to fix for_each_entry()\n \n This commit applies the consolidated list_for_each_entry_rcu() support\n for lockdep conditions.\n \n Acked-by: David S. Miller \u003cdavem@davemloft.net\u003e\n Signed-off-by: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit fbab8d6735e2643365040bd9e1057addc0d9b4cf\nAuthor: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\nDate: Tue Jul 16 18:12:23 2019 -0400\n\n rcu/sync: Remove custom check for RCU readers\n \n The rcu/sync code currently does a special check for being in an RCU\n read-side critical section. With RCU consolidating flavors and the\n generic helper added earlier in this series, this check is no longer need.\n This commit switches to the generic helper, saving a couple of lines\n of code.\n \n Cc: Oleg Nesterov \u003coleg@redhat.com\u003e\n Acked-by: Oleg Nesterov \u003coleg@redhat.com\u003e\n Signed-off-by: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 28875945ba98d1b47a8a706812b6494d165bb0a0\nAuthor: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\nDate: Tue Jul 16 18:12:22 2019 -0400\n\n rcu: Add support for consolidated-RCU reader checking\n \n This commit adds RCU-reader checks to list_for_each_entry_rcu() and\n hlist_for_each_entry_rcu(). These checks are optional, and are indicated\n by a lockdep expression passed to a new optional argument to these two\n macros. If this optional lockdep expression is omitted, these two macros\n act as before, checking for an RCU read-side critical section.\n \n Signed-off-by: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\n [ paulmck: Update to eliminate return within macro and update comment. ]\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 73499ad21d595638213f2a5f8b9b58259fa0cae2\nAuthor: Joerg Roedel \u003cjroedel@suse.de\u003e\nDate: Fri Aug 9 18:09:13 2019 +0200\n\n iommu/omap: Fix compilation warnings\n \n A recent patch introduced a new compiler warning because\n two functions with non-void return type have no return\n statement in omap-iommu.h for CONFIG_OMAP_IOMMU\u003dn.\n \n Fix this by adding return statements to these functions.\n \n Fixes: d9c4d8a6cc0f8 (\u0027iommu/omap: introduce new API for runtime suspend/resume control\u0027)\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 6738ff85c3ee8073d5b030cb26241d0009d4ce29\nAuthor: Andrea Parri \u003candrea.parri@amarulasolutions.com\u003e\nDate: Sat Jun 29 23:10:44 2019 +0200\n\n tools/memory-model: Update the informal documentation\n \n The formal memory consistency model has added support for plain accesses\n (and data races). While updating the informal documentation to describe\n this addition to the model is highly desirable and important future work,\n update the informal documentation to at least acknowledge such addition.\n \n Signed-off-by: Andrea Parri \u003candrea.parri@amarulasolutions.com\u003e\n Cc: Will Deacon \u003cwill.deacon@arm.com\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Boqun Feng \u003cboqun.feng@gmail.com\u003e\n Cc: Nicholas Piggin \u003cnpiggin@gmail.com\u003e\n Cc: David Howells \u003cdhowells@redhat.com\u003e\n Cc: Jade Alglave \u003cj.alglave@ucl.ac.uk\u003e\n Cc: Luc Maranget \u003cluc.maranget@inria.fr\u003e\n Cc: \"Paul E. McKenney\" \u003cpaulmck@linux.ibm.com\u003e\n Cc: Akira Yokosawa \u003cakiyks@gmail.com\u003e\n Cc: Daniel Lustig \u003cdlustig@nvidia.com\u003e\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n Acked-by: Alan Stern \u003cstern@rowland.harvard.edu\u003e\n\ncommit 6240973e5661a83df24e35a9a9c2013496931e2b\nAuthor: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\nDate: Mon Jul 29 08:36:05 2019 -0400\n\n tools/memory-model: Use cumul-fence instead of fence in -\u003eprop example\n \n To reduce ambiguity in the more exotic -\u003eprop ordering example, this\n commit uses the term cumul-fence instead of the term fence for the two\n fences, so that the implict -\u003erfe on loads/stores to Y are covered by\n the description.\n \n Link: https://lore.kernel.org/lkml/20190729121745.GA140682@google.com\n \n Suggested-by: Alan Stern \u003cstern@rowland.harvard.edu\u003e\n Signed-off-by: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\n Acked-by: Alan Stern \u003cstern@rowland.harvard.edu\u003e\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 29a3388bfcce7a6d087051376ea02bf8326a957b\nAuthor: Stephen Douthit \u003cstephend@silicom-usa.com\u003e\nDate: Fri Aug 9 14:18:02 2019 +0000\n\n EDAC, pnd2: Fix ioremap() size in dnv_rd_reg()\n \n Depending on how BIOS has marked the reserved region containing the 32KB\n MCHBAR you can get warnings like:\n \n resource sanity check: requesting [mem 0xfed10000-0xfed1ffff], which spans more than reserved [mem 0xfed10000-0xfed17fff]\n caller dnv_rd_reg+0xc8/0x240 [pnd2_edac] mapping multiple BARs\n \n Not all of the mmio regions used in dnv_rd_reg() are the same size. The\n MCHBAR window is 32KB and the sideband ports are 64KB. Pass the correct\n size to ioremap() depending on which resource we\u0027re reading from.\n \n Signed-off-by: Stephen Douthit \u003cstephend@silicom-usa.com\u003e\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit c2b71c71037bea7765aa6ff37824520d19108769\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Thu Aug 8 14:54:44 2019 +0900\n\n ASoC: soc-core: add for_each_xxx macro for aux_dev\n \n To be more readable code, this patch adds\n new for_each_xxx() macro for aux_dev.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87ftmc6w8s.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit b812cd5864548bd9718879896fdd2822d3cb3d76\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Thu Aug 8 14:53:52 2019 +0900\n\n ASoC: mediatek: mt8183-mt6358-ts3a227-max98357: use snd_soc_dai_link_component for aux_dev\n \n We can use snd_soc_dai_link_component to specify aux_dev.\n Let\u0027s use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87sgqc6wa8.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c86102a333f77dcd0f7ef20ba836c6f13f1a077a\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 8 22:35:07 2019 +0800\n\n ASoC: max9850: remove unused variable \u0027max9850_reg\u0027\n \n sound/soc/codecs/max9850.c:31:33: warning:\n max9850_reg defined but not used [-Wunused-const-variable\u003d]\n \n It is not used since commit 068416620c0d (\"ASoC:\n max9850: Convert to direct regmap API usage\")\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190808143507.66788-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit dbf0649f4340c8bb7d36b8d6255dba03ed6981e7\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 8 22:32:15 2019 +0800\n\n ASoC: max98926: remove two unused variables\n \n sound/soc/codecs/max98926.c:28:26: warning:\n max98926_dai_txt defined but not used [-Wunused-const-variable\u003d]\n sound/soc/codecs/max98926.c:23:27: warning:\n max98926_boost_current_txt defined but not used [-Wunused-const-variable\u003d]\n \n They are never used, so can be removd.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190808143215.65904-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit a48b561d873d1d9fda55782d275eff94ec647863\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Thu Aug 8 14:54:39 2019 +0900\n\n ASoC: soc-core: remove legacy style of aux_dev\n \n Now all drivers are using snd_soc_dai_link_component for aux_dev.\n Let\u0027s remove legacy style\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87h86s6w8x.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 77b21d2822fdbfad19170516256c4c7f236054a9\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Thu Aug 8 14:54:30 2019 +0900\n\n ASoC: ti: rx51: use snd_soc_dai_link_component for aux_dev\n \n We can use snd_soc_dai_link_component to specify aux_dev.\n Let\u0027s use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Acked-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Link: https://lore.kernel.org/r/87imr86w96.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 3d0d2d64b7eb4f2a451fa184829b749851c14f55\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Thu Aug 8 14:54:25 2019 +0900\n\n ASoC: sunxi: sun4i-codec: use snd_soc_dai_link_component for aux_dev\n \n We can use snd_soc_dai_link_component to specify aux_dev.\n Let\u0027s use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87k1bo6w9b.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 124749194ad2e08fbea9c34fa5e2728310bf4486\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Thu Aug 8 14:54:20 2019 +0900\n\n ASoC: samsung: tm2_wm5110: use snd_soc_dai_link_component for aux_dev\n \n We can use snd_soc_dai_link_component to specify aux_dev.\n Let\u0027s use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Acked-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Link: https://lore.kernel.org/r/87lfw46w9g.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 2d946aaa80c79452c700381b4c1f06f11dfd2bdd\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Thu Aug 8 14:54:15 2019 +0900\n\n ASoC: samsung: speyside: use snd_soc_dai_link_component for aux_dev\n \n We can use snd_soc_dai_link_component to specify aux_dev.\n Let\u0027s use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Acked-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Link: https://lore.kernel.org/r/87mugk6w9l.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit d20bb789d7283bcbacd3b12c60668c473bc1dac0\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Thu Aug 8 14:54:10 2019 +0900\n\n ASoC: samsung: neo1973_wm8753: use snd_soc_dai_link_component for aux_dev\n \n We can use snd_soc_dai_link_component to specify aux_dev.\n Let\u0027s use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Acked-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Link: https://lore.kernel.org/r/87o9106w9p.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 70a6b7bfba9bfd781b8c137f059487017dd385d8\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Thu Aug 8 14:54:05 2019 +0900\n\n ASoC: rockchip: rockchip_max98090: use snd_soc_dai_link_component for aux_dev\n \n We can use snd_soc_dai_link_component to specify aux_dev.\n Let\u0027s use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87pnlg6w9v.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit a73b522d4b570ee41754990e81f6f4e849af1af9\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Thu Aug 8 14:53:59 2019 +0900\n\n ASoC: meson: axg-card: use snd_soc_dai_link_component for aux_dev\n \n We can use snd_soc_dai_link_component to specify aux_dev.\n Let\u0027s use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87r25w6wa1.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 5c9e38cb57ce07d780dc05d34ae140271c2da159\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Thu Aug 8 14:53:45 2019 +0900\n\n ASoC: mediatek: mt8183-da7219-max98357: use snd_soc_dai_link_component for aux_dev\n \n We can use snd_soc_dai_link_component to specify aux_dev.\n Let\u0027s use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87tvas6waf.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit aa1afd92f374a895ab74cb96dacd25a3485a08c4\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Thu Aug 8 14:53:33 2019 +0900\n\n ASoC: intel: cht_bsw_max98090_ti: use snd_soc_dai_link_component for aux_dev\n \n We can use snd_soc_dai_link_component to specify aux_dev.\n Let\u0027s use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87v9v86war.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 14fdfc058d812f683d9aa51083d1a920aa6a1eb9\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Thu Aug 8 14:52:47 2019 +0900\n\n ASoC: simple-card: use snd_soc_dai_link_component for aux_dev\n \n We can use snd_soc_dai_link_component to specify aux_dev.\n Let\u0027s use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87wofo6wc1.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 3dc29b8b2062075602c7aff1514a120b4ed0187f\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Thu Aug 8 14:52:33 2019 +0900\n\n ASoC: soc-core: support snd_soc_dai_link_component for aux_dev\n \n To find aux_dev, ASoC is using .name, codec_name, codec_of_node.\n Here, .name is used to fallback in case of no codec.\n \n But, we already have this kind of component finding method by\n snd_soc_dai_link_component and soc_find_component().\n We shouldn\u0027t have duplicated implementation to do same things.\n This patch adds snd_soc_dai_link_component support to finding aux_dev.\n \n Now, no driver is using only .name.\n All drivers are using codec_name and/or codec_of_node.\n This means no driver is finding component from .name so far.\n (Actually almost all drivers are using .name as just \"device name\",\n not for finding component...)\n \n This patch\n 1) add snd_soc_dai_link_component support for aux_dev. legacy style will\n be removed if all drivers are switched to new style.\n 2) try to find component via snd_soc_dai_link_component.\n Then, it doesn\u0027t try to find via .name, because no driver is using\n it so far.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87y3046wcf.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit d9df0187b6edd5871255cee29128c63f134e599c\nAuthor: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\nDate: Thu Jul 25 11:45:42 2019 +0200\n\n regulator: twl6030: workaround the VMMC reset behavior\n \n During reset the VMMC regulator doesn\u0027t reach 0V and only drops to\n 1.8V, furthermore the pulse width is under 200us whereas the SD\n specification expect 1ms.\n \n The WR_S bit allows the TWL6030 to no reset at all the VMMC during warm\n reset and keep the current voltage. Thanks to this workaround the SD\n card doesn\u0027t reach a undefined reset stage.\n \n Actually this behavior is available for all the LDO regulator, so the\n driver will also allow to use it with any of these regulator.\n \n Signed-off-by: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190725094542.16547-4-gregory.clement@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 621d3ce830649d56ef1c9611e370d05ceaffd7fc\nAuthor: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\nDate: Thu Jul 25 11:45:41 2019 +0200\n\n regulator: twl6030: use variable for device node\n \n Instead of refering the full pdev-\u003edev.of_node use a local variable.\n \n Signed-off-by: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190725094542.16547-3-gregory.clement@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 906c4d5c939b4d8ebe7be357121271df7b8c9582\nAuthor: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\nDate: Thu Jul 25 11:45:40 2019 +0200\n\n dt-bindings: regulator: twl6030: Add retain-on-reset property\n \n During reset the VMMC regulator doesn\u0027t reach 0V and only drops to\n 1.8V, furthermore the pulse width is under 200us whereas the SD\n specification expect 1ms.\n \n For this 2 reasons being able to no reset at all the VMMC during warm\n reset and keep the current voltage is a good workaround. The TWL6030\n allows this but needs to be aware of it and this configuration should\n also be shared with the bootloader.\n \n This is the purpose of this new property: ti,retain-on-reset\n \n Signed-off-by: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190725094542.16547-2-gregory.clement@bootlin.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 332ccf00bf85adbf48015084be0e60f5cc57a055\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Fri Aug 9 17:15:31 2019 +0200\n\n ASoC: hdac_hdmi: Offload dapm update at jack detection\n \n hdac_hdmi_present_sense() calls the audio component to get ELD update,\n then it reports the jack status change and updates DAPM graph\n accordingly. This works when it\u0027s called from the normal code paths.\n \n However, it may lead to a dead lock when it\u0027s called from the audio\n component notifier. Namely, the DAPM update involves with the runtime\n PM, and it eventually calls again the audio component get_power()\n ops. Since i915 driver already takes a mutex around the audio\n component ops calls, we\u0027ll eventually get the mutex doubly.\n \n As a workaround, in this patch, only the jack state is updated in the\n code path from hdac_hdmi_eld_notify_cb(), and the DAPM update is\n deferred to a work so that it\u0027s processed in another context.\n \n Reported-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n Link: https://lore.kernel.org/r/20190809151531.24359-1-tiwai@suse.de\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9ffeb6d08c3a4bbd7b1e33711b241f511e2ded79\nAuthor: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\nDate: Fri Aug 9 12:03:12 2019 +0100\n\n PSCI: cpuidle: Refactor CPU suspend power_state parameter handling\n \n Current PSCI code handles idle state entry through the\n psci_cpu_suspend_enter() API, that takes an idle state index as a\n parameter and convert the index into a previously initialized\n power_state parameter before calling the PSCI.CPU_SUSPEND() with it.\n \n This is unwieldly, since it forces the PSCI firmware layer to keep track\n of power_state parameter for every idle state so that the\n index-\u003epower_state conversion can be made in the PSCI firmware layer\n instead of the CPUidle driver implementations.\n \n Move the power_state handling out of drivers/firmware/psci\n into the respective ACPI/DT PSCI CPUidle backends and convert\n the psci_cpu_suspend_enter() API to get the power_state\n parameter as input, which makes it closer to its firmware\n interface PSCI.CPU_SUSPEND() API.\n \n A notable side effect is that the PSCI ACPI/DT CPUidle backends\n now can directly handle (and if needed update) power_state\n parameters before handing them over to the PSCI firmware\n interface to trigger PSCI.CPU_SUSPEND() calls.\n \n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Acked-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Reviewed-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n Reviewed-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Cc: Will Deacon \u003cwill@kernel.org\u003e\n Cc: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n Cc: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: \"Rafael J. Wysocki\" \u003crjw@rjwysocki.net\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 788961462f3471617749edf10d0fcafad410d2bb\nAuthor: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\nDate: Fri Aug 9 12:03:11 2019 +0100\n\n ARM: psci: cpuidle: Enable PSCI CPUidle driver\n \n Allow selection of the PSCI CPUidle in the kernel by updating\n the respective Kconfig entry.\n \n Remove PSCI callbacks from ARM/ARM64 generic CPU ops\n to prevent the PSCI idle driver from clashing with the generic\n ARM CPUidle driver initialization, that relies on CPU ops\n to initialize and enter idle states.\n \n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Reviewed-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n Cc: Will Deacon \u003cwill@kernel.org\u003e\n Cc: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n Cc: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: \"Rafael J. Wysocki\" \u003crjw@rjwysocki.net\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 81d549e0c810773bf003a25f59fa5509857bf9b2\nAuthor: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\nDate: Fri Aug 9 12:03:10 2019 +0100\n\n ARM: psci: cpuidle: Introduce PSCI CPUidle driver\n \n PSCI firmware is the standard power management control for\n all ARM64 based platforms and it is also deployed on some\n ARM 32 bit platforms to date.\n \n Idle state entry in PSCI is currently achieved by calling\n arm_cpuidle_init() and arm_cpuidle_suspend() in a generic\n idle driver, which in turn relies on ARM/ARM64 CPUidle back-end\n to relay the call into PSCI firmware if PSCI is the boot method.\n \n Given that PSCI is the standard idle entry method on ARM64 systems\n (which means that no other CPUidle driver are expected on ARM64\n platforms - so PSCI is already a generic idle driver), in order to\n simplify idle entry and code maintenance, it makes sense to have a PSCI\n specific idle driver so that idle code that it is currently living in\n drivers/firmware directory can be hoisted out of it and moved\n where it belongs, into a full-fledged PSCI driver, leaving PSCI code\n in drivers/firmware as a pure firmware interface, as it should be.\n \n Implement a PSCI CPUidle driver. By default it is a silent Kconfig entry\n which is left unselected, since it selection would clash with the\n generic ARM CPUidle driver that provides a PSCI based idle driver\n through the arm/arm64 arches back-ends CPU operations.\n \n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Acked-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Reviewed-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n Reviewed-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Cc: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n Cc: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: \"Rafael J. Wysocki\" \u003crjw@rjwysocki.net\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit e76d8b7027d9451cfdcb8cd605148ff5e0a1bf5d\nAuthor: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\nDate: Fri Aug 9 12:03:09 2019 +0100\n\n drivers: firmware: psci: Decouple checker from generic ARM CPUidle\n \n The PSCI checker currently relies on the generic ARM CPUidle\n infrastructure to enter an idle state, which in turn creates\n a dependency that is not really needed.\n \n The PSCI checker code to test PSCI CPU suspend is built on\n top of the CPUidle framework and can easily reuse the\n struct cpuidle_state.enter() function (previously initialized\n by an idle driver, with a PSCI back-end) to trigger an entry\n into an idle state, decoupling the PSCI checker from the\n generic ARM CPUidle infrastructure and simplyfing the code\n in the process.\n \n Convert the PSCI checker suspend entry function to use\n the struct cpuidle_state.enter() function callback.\n \n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Acked-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Reviewed-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n Reviewed-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Cc: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 6460d7ba488419dd5bc2cbfd989d5d7930ac1f9f\nAuthor: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\nDate: Fri Aug 9 12:03:08 2019 +0100\n\n ARM: cpuidle: Remove overzealous error logging\n \n CPUidle back-end operations are not implemented in some platforms\n but this should not be considered an error serious enough to be\n logged. Check the arm_cpuidle_init() return value to detect whether\n the failure must be reported or not in the kernel log and do\n not log it if the platform does not support CPUidle operations.\n \n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Acked-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Reviewed-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n Reviewed-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Cc: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n Cc: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: \"Rafael J. Wysocki\" \u003crjw@rjwysocki.net\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 63e3ee61549595719145d23fdacd1d8744051108\nAuthor: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\nDate: Fri Aug 9 12:03:07 2019 +0100\n\n ARM: cpuidle: Remove useless header include\n \n The generic ARM CPUidle driver includes \u003clinux/topology.h\u003e by mistake.\n \n Remove the topology header include.\n \n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Acked-by: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Reviewed-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n Reviewed-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Cc: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n Cc: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: \"Rafael J. Wysocki\" \u003crjw@rjwysocki.net\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 26332247fb3fe919a3e51589601ba3de566cc355\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Fri Aug 9 09:45:27 2019 -0700\n\n Input: synaptics - fix a typo\n \n This should be \u0027synaptics\u0027, not \u0027synpatics\u0027\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 6ca523d7eb198b78f79df882b7dc4a8c795d8777\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Fri Aug 9 15:57:51 2019 +0800\n\n drm/amdgpu: remove RREG64/WREG64\n \n atomic 64 bits REG operations are useless currently\n \n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit dd21a572c9068e9a59b46dea67e8a65a44aee90b\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Fri Aug 9 15:57:50 2019 +0800\n\n drm/amdgpu: implement UMC 64 bits REG operations\n \n implement 64 bits operations via 32 bits interface\n \n v2: make use of lower_32_bits() and upper_32_bits() macros\n \n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c6dddf45402caeadc49dc859fa497cfb98841af4\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Wed Aug 7 10:28:54 2019 +0800\n\n drm/amdgpu: replace readq/writeq with atomic64 operations\n \n what we really want is a read or write that is guaranteed to be 64 bits\n at a time, atomic64 operations are supported on all architectures\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit abcce733adb71013997fcb84142a5454ef133616\nAuthor: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\nDate: Thu Aug 8 18:17:47 2019 +0200\n\n samples/bpf: xdp_fwd explain bpf_fib_lookup return codes\n \n Make it clear that this XDP program depend on the network\n stack to do the ARP resolution. This is connected with the\n BPF_FIB_LKUP_RET_NO_NEIGH return code from bpf_fib_lookup().\n \n Another common mistake (seen via XDP-tutorial) is that users\n don\u0027t realize that sysctl net.ipv{4,6}.conf.all.forwarding\n setting is honored by bpf_fib_lookup.\n \n Reported-by: Anton Protopopov \u003ca.s.protopopov@gmail.com\u003e\n Signed-off-by: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\n Reviewed-by: David Ahern \u003cdsahern@gmail.com\u003e\n Acked-by: Yonghong Song \u003cyhs@fb.com\u003e\n Reviewed-by: Toke Høiland-Jørgensen \u003ctoke@redhat.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit a32a32cb26eb6291125e4eb49b569874ca9a53b5\nAuthor: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\nDate: Thu Aug 8 18:17:42 2019 +0200\n\n samples/bpf: make xdp_fwd more practically usable via devmap lookup\n \n This address the TODO in samples/bpf/xdp_fwd_kern.c, which points out\n that the chosen egress index should be checked for existence in the\n devmap. This can now be done via taking advantage of Toke\u0027s work in\n commit 0cdbb4b09a06 (\"devmap: Allow map lookups from eBPF\").\n \n This change makes xdp_fwd more practically usable, as this allows for\n a mixed environment, where IP-forwarding fallback to network stack, if\n the egress device isn\u0027t configured to use XDP.\n \n Signed-off-by: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\n Reviewed-by: David Ahern \u003cdsahern@gmail.com\u003e\n Reviewed-by: Toke Høiland-Jørgensen \u003ctoke@redhat.com\u003e\n Acked-by: Yonghong Song \u003cyhs@fb.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 3783d43752eab247ed296ac8d5022484ed969151\nAuthor: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\nDate: Thu Aug 8 18:17:37 2019 +0200\n\n samples/bpf: xdp_fwd rename devmap name to be xdp_tx_ports\n \n The devmap name \u0027tx_port\u0027 came from a copy-paste from xdp_redirect_map\n which only have a single TX port. Change name to xdp_tx_ports\n to make it more descriptive.\n \n Signed-off-by: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\n Reviewed-by: David Ahern \u003cdsahern@gmail.com\u003e\n Acked-by: Yonghong Song \u003cyhs@fb.com\u003e\n Reviewed-by: Toke Høiland-Jørgensen \u003ctoke@redhat.com\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit d9973cec9d578b381235bb872a2d378c69c54915\nAuthor: Ivan Khoronzhuk \u003civan.khoronzhuk@linaro.org\u003e\nDate: Thu Aug 8 12:38:03 2019 +0300\n\n xdp: xdp_umem: fix umem pages mapping for 32bits systems\n \n Use kmap instead of page_address as it\u0027s not always in low memory.\n \n Acked-by: Björn Töpel \u003cbjorn.topel@intel.com\u003e\n Signed-off-by: Ivan Khoronzhuk \u003civan.khoronzhuk@linaro.org\u003e\n Signed-off-by: Daniel Borkmann \u003cdaniel@iogearbox.net\u003e\n\ncommit 00d20c6907a379a46cbc0f29dbcd3ded94d75e7a\nAuthor: Andreas Gruenbacher \u003cagruenba@redhat.com\u003e\nDate: Thu Aug 8 19:29:54 2019 +0100\n\n gfs2: Always mark inode dirty in fallocate\n \n When allocating space with fallocate, always update the file timestamps\n and mark the inode dirty, no matter if the FALLOC_FL_KEEP_SIZE flag set\n or not. The inode needs to be marked dirty so that a subsequent fsync\n will pick up the inode, and any new allocations will make it to disk.\n Filesystems like xfs and ext4 always update the timestamps, so make gfs2\n behave the same way.\n \n Fixes xfstest generic/483.\n \n Signed-off-by: Andreas Gruenbacher \u003cagruenba@redhat.com\u003e\n\ncommit d40312598d534c17c17f41c2bb7ce9541a5f786e\nAuthor: Andreas Gruenbacher \u003cagruenba@redhat.com\u003e\nDate: Wed Jul 24 13:05:38 2019 +0200\n\n gfs2: Minor gfs2_alloc_inode cleanup\n \n In gfs2_alloc_inode, when kmem_cache_alloc cannot allocate a new object, return\n NULL immediately. The code currently relies on the fact that i_inode is the\n first member in struct gfs2_inode and so ip and \u0026ip-\u003ei_inode evaluate to the\n same address, but that isn\u0027t immediately obvious.\n \n Signed-off-by: Andreas Gruenbacher \u003cagruenba@redhat.com\u003e\n Reviewed-by: Bob Peterson \u003crpeterso@redhat.com\u003e\n\ncommit 2257e468a63b6d35a77e884ef032c54f9be65c92\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Jul 1 23:54:37 2019 +0200\n\n gfs2: implement gfs2_block_zero_range using iomap_zero_range\n \n iomap handles all the nitty-gritty details of zeroing a file\n range for us, so use the proper helper.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Andreas Gruenbacher \u003cagruenba@redhat.com\u003e\n Reviewed-by: Bob Peterson \u003crpeterso@redhat.com\u003e\n\ncommit 72d36d0529c6d9ee8b950c819508b1e344d8cc4f\nAuthor: Andreas Gruenbacher \u003cagruenba@redhat.com\u003e\nDate: Fri Jul 12 15:57:52 2019 +0200\n\n gfs2: Add support for IOMAP_ZERO\n \n Add support for the IOMAP_ZERO iomap operation so that iomap_zero_range will\n work as expected. In the IOMAP_ZERO case, the caller of iomap_zero_range is\n responsible for taking an exclusive glock on the inode, so we need no\n additional locking in gfs2_iomap_begin.\n \n Signed-off-by: Andreas Gruenbacher \u003cagruenba@redhat.com\u003e\n Reviewed-by: Bob Peterson \u003crpeterso@redhat.com\u003e\n\ncommit 34aad20bc3fff3ce4e3fa8cb1d0265a1df5cded5\nAuthor: Andreas Gruenbacher \u003cagruenba@redhat.com\u003e\nDate: Fri Jul 5 17:38:35 2019 +0200\n\n gfs2: gfs2_iomap_begin cleanup\n \n Following commit d0a22a4b03b8 (\"gfs2: Fix iomap write page reclaim deadlock\"),\n gfs2_iomap_begin and gfs2_iomap_begin_write can be further cleaned up and the\n split between those two functions can be improved.\n \n With suggestions from Christoph Hellwig.\n \n Signed-off-by: Andreas Gruenbacher \u003cagruenba@redhat.com\u003e\n Reviewed-by: Bob Peterson \u003crpeterso@redhat.com\u003e\n\ncommit abd627f8031b1199608a7d5d92c19d61960d552d\nAuthor: Rahul Tanwar \u003crahul.tanwar@linux.intel.com\u003e\nDate: Thu Aug 8 18:02:08 2019 +0800\n\n serial: lantiq: Add support for Lightning Mountain SoC\n \n This patch adds IRQ \u0026 ISR support in the driver for Lightning Mountain SoC.\n \n Signed-off-by: Rahul Tanwar \u003crahul.tanwar@linux.intel.com\u003e\n Reviewed-by: Andy Shevchenko \u003candriy.shevchenko@intel.com\u003e\n Link: https://lore.kernel.org/r/0df20f6e4bbf9de09c85a5c92c92e642f62f441f.1565257887.git.rahul.tanwar@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 18201bc4416854cf48ff5c179de223974cd87b47\nAuthor: Rahul Tanwar \u003crahul.tanwar@linux.intel.com\u003e\nDate: Thu Aug 8 18:02:07 2019 +0800\n\n serial: lantiq: Make IRQ \u0026 ISR assignment dynamic\n \n This driver/IP is reused across multiple SoCs. Older SoCs supported three\n separate IRQs for tx, rx \u0026 err interrupts. Newer Lightning Mountain SoC\n supports single IRQ for all of tx/rx/err interrupts. This patch modifies\n the driver design to support dynamic assignment of IRQ resources \u0026 ISRs\n based on devicetree node compatible entries.\n \n Suggested-by: Andy Shevchenko \u003candriy.shevchenko@intel.com\u003e\n Signed-off-by: Rahul Tanwar \u003crahul.tanwar@linux.intel.com\u003e\n Reviewed-by: Andy Shevchenko \u003candriy.shevchenko@intel.com\u003e\n Link: https://lore.kernel.org/r/b166a0593bee191fcd77b5bdf8fedc6f6330a371.1565257887.git.rahul.tanwar@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 753062c9b72f478741ff0a1b6cd7625a31103cf8\nAuthor: Rahul Tanwar \u003crahul.tanwar@linux.intel.com\u003e\nDate: Thu Aug 8 18:02:06 2019 +0800\n\n serial: lantiq: Use proper DT compatible string\n \n Use explicit string instead of a macro for devicetree compatible string.\n \n This series of patches is to add support for multiple SoCs which reuse the same\n serial controller IP. The following patches will add another compatible string\n to support new Lightning Mountain(LGM) SoC. So it makes sense to have the\n compatible strings explicitly mentioned instead of a fixed macro.\n \n Suggested-by: Andy Shevchenko \u003candriy.shevchenko@intel.com\u003e\n Signed-off-by: Rahul Tanwar \u003crahul.tanwar@linux.intel.com\u003e\n Reviewed-by: Andy Shevchenko \u003candriy.shevchenko@intel.com\u003e\n Link: https://lore.kernel.org/r/57e2b69e9fbd93328a477b4c7dd2dcc78784ecb1.1565257887.git.rahul.tanwar@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a0d993e8c143542941d02267f15bf319a86e6c78\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Aug 6 12:43:22 2019 +0300\n\n serial: 8250_lpss: Enable HS UART on Elkhart Lake\n \n Intel Elkhart Lake may use High Speed UART from OSE IP block.\n This is different to what we have in main LPSS, though compatible\n with older version of it, which is handled by this driver.\n \n Enable OSE HS UART on Intel Elkhart Lake by adding PCI IDs.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806094322.64987-9-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a9c69a4973dfbe35a47a9ffdf91ed455ff16e8ed\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Aug 6 12:43:21 2019 +0300\n\n serial: 8250_lpss: Get rid of custom LPSS_DEVICE() macro\n \n Since PCI core provides a generic PCI_DEVICE_DATA() macro,\n replace LPSS_DEVICE() with former one.\n \n No functional change intended.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806094322.64987-8-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 3ff7eb5566cb48c17d1ce7b439f0a6106c8330c0\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Aug 6 12:43:20 2019 +0300\n\n serial: 8250_lpss: register DMA IRQ and pool with instance ID\n \n It is really useful not only for debugging to have an DMA IRQ line and\n pool being mapped to the corresponding IP by using its instance ID.\n \n Provide PCI device and function as instance ID for Intel Quark UART DMA.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806094322.64987-7-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit c2f204f8dc0689cd829575d664b4f8ed8162bbc8\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Aug 6 12:43:19 2019 +0300\n\n serial: 8250_lpss: add fractional divisor support\n \n For Synopsys DesignWare 8250 uart which version \u003e\u003d 4.00a, there\u0027s a\n valid divisor latch fraction register.\n \n Now the preparation is done, it\u0027s easy to add the feature support.\n This patch firstly tries to get the fractional divisor width during\n probe, then setups specific get_divisor() and set_divisor() hook.\n \n Among other changes the FIFO size is now retrieved from the hardware.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806094322.64987-6-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit e3bfe787693775b1bd07210690fdf7ac33b544a8\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Aug 6 12:43:18 2019 +0300\n\n serial: 8250_lpss: switch to use 8250_dwlib library\n \n Since we have a common library module for Synopsys DesignWare UART,\n let us use it.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806094322.64987-5-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit bec2eadf2f7459e202f71bd84bb939cc4dfc41e0\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Aug 6 12:43:17 2019 +0300\n\n serial: 8250_dw: switch to use 8250_dwlib library\n \n Since we have a common library module for Synopsys DesignWare UART,\n let us use it.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806094322.64987-4-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit bc7b8e8f68ac71713798b1d4e72980a54be36a0a\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Aug 6 12:43:16 2019 +0300\n\n serial: 8250_dw: split Synopsys DesignWare 8250 common functions\n \n We would like to use same functions in the couple of drivers for\n Synopsys DesignWare 8250 UART. Split them from 8250_dw into new brand\n library module which users will select explicitly.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806094322.64987-3-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7ed6935fd1bef2b2155854ab17bbe690ac4cb650\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Aug 6 12:43:15 2019 +0300\n\n serial: 8250_dw: use pointer to uart local variable\n \n The use of pointer will simplify enabling runtime PM for the driver.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806094322.64987-2-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 1432ebbd60033f48ae24302524f9dca592103804\nAuthor: Tero Kristo \u003ct-kristo@ti.com\u003e\nDate: Wed Aug 7 11:26:52 2019 +0300\n\n iommu/omap: remove pm_runtime_irq_safe flag for OMAP IOMMUs\n \n This is not needed for anything, and prevents proper PM transitions for\n parent devices which is bad in case of ti-sysc; this effectively kills\n PM completely. Thus, remove the flag.\n \n Signed-off-by: Tero Kristo \u003ct-kristo@ti.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 604629bcb5057d778839652f8f38d23734f2fe1d\nAuthor: Tero Kristo \u003ct-kristo@ti.com\u003e\nDate: Wed Aug 7 11:26:51 2019 +0300\n\n iommu/omap: add support for late attachment of iommu devices\n \n Current implementation of OMAP IOMMU enforces strict ordering of device\n probe, initiated by iommu and followed by remoteproc later. This doesn\u0027t\n work too well with the new setup done with ti-sysc changes which may\n have the devices probed at pretty much any order. To overcome this limitation,\n if iommu has not been probed yet when a consumer tries to attach to it,\n add the device to orphan device list which will be parsed during iommu\n probe to see if any orphan devices should be attached.\n \n Signed-off-by: Tero Kristo \u003ct-kristo@ti.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit d9c4d8a6cc0f852adf3829fbe40e2e3f6213b0c6\nAuthor: Suman Anna \u003cs-anna@ti.com\u003e\nDate: Wed Aug 7 11:26:50 2019 +0300\n\n iommu/omap: introduce new API for runtime suspend/resume control\n \n This patch adds the support for the OMAP IOMMUs to be suspended\n during the auto suspend/resume of the OMAP remoteproc devices. The\n remote processors are auto suspended after a certain time of idle\n or inactivity period. This is done by introducing two new API,\n omap_iommu_domain_deactivate() and omap_iommu_domain_activate()\n to allow the client users/master devices of the IOMMU devices to\n deactivate \u0026 activate the IOMMU devices from their runtime\n suspend/resume operations. There is no API exposed by the IOMMU\n layer at present, and so these new API are added directly in the\n OMAP IOMMU driver to minimize framework changes.\n \n The API simply decrements and increments the runtime usage count\n of the IOMMU devices and let the context be saved/restored using\n the existing runtime pm callbacks.\n \n Signed-off-by: Suman Anna \u003cs-anna@ti.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit c4206c4e190bd272e4e86389613f2cded4609ba1\nAuthor: Suman Anna \u003cs-anna@ti.com\u003e\nDate: Wed Aug 7 11:26:49 2019 +0300\n\n iommu/omap: Add system suspend/resume support\n \n The MMU registers for the remote processors lose their context\n in Open Switch Retention (OSWR) or device OFF modes. Hence, the\n context of the IOMMU needs to be saved before it is put into any\n of these lower power state (OSWR/OFF) and restored before it is\n powered up to ON again. The IOMMUs need to be active as long as\n the client devices that are present behind the IOMMU are active.\n \n This patch adds the dev_pm_ops callbacks to provide the system\n suspend/resume functionality through the appropriate runtime\n PM callbacks. The PM runtime_resume and runtime_suspend callbacks\n are already used to enable, configure and disable the IOMMUs during\n the attaching and detaching of the client devices to the IOMMUs,\n and the new PM callbacks reuse the same code by invoking the\n pm_runtime_force_suspend() and pm_runtime_force_resume() API. The\n functionality in dev_pm_ops .prepare() checks if the IOMMU device\n was already runtime suspended, and skips invoking the suspend/resume\n PM callbacks. The suspend/resume PM callbacks are plugged in through\n the \u0027late\u0027 pm ops to ensure that the IOMMU devices will be suspended\n only after its master devices (remoteproc devices) are suspended and\n restored before them.\n \n NOTE:\n There are two other existing API, omap_iommu_save_ctx() and\n omap_iommu_restore_ctx(). These are left as is to support\n suspend/resume of devices on legacy OMAP3 SoC.\n \n Signed-off-by: Suman Anna \u003cs-anna@ti.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 3846a3b9511c5166082a93536d919a9c42abcd91\nAuthor: Suman Anna \u003cs-anna@ti.com\u003e\nDate: Wed Aug 7 11:26:45 2019 +0300\n\n iommu/omap: fix boot issue on remoteprocs with AMMU/Unicache\n \n Support has been added to the OMAP IOMMU driver to fix a boot hang\n issue on OMAP remoteprocs with AMMU/Unicache, caused by an improper\n AMMU/Unicache state upon initial deassertion of the processor reset.\n The issue is described in detail in the next three paragraphs.\n \n All the Cortex M3/M4 IPU processor subsystems in OMAP SoCs have a\n AMMU/Unicache IP that dictates the memory attributes for addresses\n seen by the processor cores. The AMMU/Unicache is configured/enabled\n by the SCACHE_CONFIG.BYPASS bit - a value of 1 enables the cache and\n mandates all addresses accessed by M3/M4 be defined in the AMMU. This\n bit is not programmable from the host processor. The M3/M4 boot\n sequence starts out with the AMMU/Unicache in disabled state, and\n SYS/BIOS programs the AMMU regions and enables the Unicache during\n one of its initial boot steps. This SCACHE_CONFIG.BYPASS bit is\n however enabled by default whenever a RET reset is applied to the IP,\n irrespective of whether it was previously enabled or not. The AMMU\n registers lose their context whenever this reset is applied. The reset\n is effective as long as the MMU portion of the subsystem is enabled\n and clocked. This behavior is common to all the IPU and DSP subsystems\n that have an AMMU/Unicache.\n \n The IPU boot sequence involves enabling and programming the MMU, and\n loading the processor and releasing the reset(s) for the processor.\n The PM setup code currently sets the target state for most of the\n power domains to RET. The L2 MMU can be enabled, programmed and\n accessed properly just fine with the domain in hardware supervised\n mode, while the power domain goes through a RET-\u003eON-\u003eRET transition\n during the programming sequence. However, the ON-\u003eRET transition\n asserts a RET reset, and the SCACHE_CONFIG.BYPASS bit gets auto-set.\n An AMMU fault is thrown immediately when the M3/M4 core\u0027s reset is\n released since the first instruction address itself will not be\n defined in any valid AMMU regions. The ON-\u003eRET transition happens\n automatically on the power domain after enabling the iommu due to\n the hardware supervised mode.\n \n This patch adds and invokes the .set_pwrdm_constraint pdata ops, if\n present, during the OMAP IOMMU enable and disable functions to resolve\n the above boot hang issue. The ops will allow to invoke a mach-omap2\n layer API pwrdm_set_next_pwrst() in a multi-arch kernel environment.\n The ops also returns the current power domain state while enforcing\n the constraint so that the driver can store it and use it to set back\n the power domain state while releasing the constraint. The pdata ops\n implementation restricts the target power domain to ON during enable,\n and back to the original power domain state during disable, and thereby\n eliminating the conditions for the boot issue. The implementation is\n effective only when the original power domain state is either RET or\n OFF, and is a no-op when it is ON or INACTIVE.\n \n The .set_pwrdm_constraint ops need to be plugged in pdata-quirks\n for the affected remote processors to be able to boot properly.\n \n Note that the current issue is seen only on kernels with the affected\n power domains programmed to enter RET. For eg., IPU1 on DRA7xx is in a\n separate domain and is susceptible to this bug, while the IPU2 subsystem\n is within CORE power domain, and CORE RET is not supported on this SoC.\n IPUs on OMAP4 and OMAP5 are also susceptible since they are in CORE power\n domain, and CORE RET is a valid power target on these SoCs.\n \n Signed-off-by: Suman Anna \u003cs-anna@ti.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit c3b44a063d3b4d9b5afabaf5ab2e3fc11518dc84\nAuthor: Suman Anna \u003cs-anna@ti.com\u003e\nDate: Wed Aug 7 11:26:48 2019 +0300\n\n iommu/omap: add logic to save/restore locked TLBs\n \n The MMUs provide a mechanism to lock TLB entries to avoid\n eviction and fetching of frequently used page table entries.\n These TLBs lose context when the MMUs are turned OFF. Add the\n logic to save and restore these locked TLBS during suspend\n and resume respectively. There are no locked TLBs during\n initial power ON, and they need not be saved during final\n shutdown.\n \n Signed-off-by: Suman Anna \u003cs-anna@ti.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit db8918f61d51b87aaf7554ab7647ccdb7dbd739f\nAuthor: Suman Anna \u003cs-anna@ti.com\u003e\nDate: Wed Aug 7 11:26:47 2019 +0300\n\n iommu/omap: streamline enable/disable through runtime pm callbacks\n \n The OMAP IOMMU devices are typically present within the respective\n client processor subsystem and have their own dedicated hard-reset\n line. Enabling an IOMMU requires the reset line to be deasserted\n and the clocks to be enabled before programming the necessary IOMMU\n registers. The IOMMU disable sequence follow the reverse order of\n enabling. The OMAP IOMMU driver programs the reset lines through\n pdata ops to invoke the omap_device_assert/deassert_hardreset API.\n The clocks are managed through the pm_runtime framework, and the\n callbacks associated with the device\u0027s pm_domain, implemented in\n the omap_device layer.\n \n Streamline the enable and disable sequences in the OMAP IOMMU\n driver by implementing all the above operations within the\n runtime pm callbacks. All the OMAP devices have device pm_domain\n callbacks plugged in the omap_device layer for automatic runtime\n management of the clocks. Invoking the reset management functions\n within the runtime pm callbacks in OMAP IOMMU driver therefore\n requires that the default device\u0027s pm domain callbacks in the\n omap_device layer be reset, as the ordering sequence for managing\n the reset lines and clocks from the pm_domain callbacks don\u0027t gel\n well with the implementation in the IOMMU driver callbacks. The\n omap_device_enable/omap_device_idle functions are invoked through\n the newly added pdata ops.\n \n Consolidating all the device management sequences within the\n runtime pm callbacks allows the driver to easily support both\n system suspend/resume and runtime suspend/resume using common\n code.\n \n Signed-off-by: Suman Anna \u003cs-anna@ti.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 74c116df66d25a9fb48d44ce545a505edc5fbbba\nAuthor: Suman Anna \u003cs-anna@ti.com\u003e\nDate: Wed Aug 7 11:26:46 2019 +0300\n\n iommu/omap: add pdata ops for omap_device_enable/idle\n \n Add two new platform data ops to allow the OMAP iommu driver to\n be able to invoke the omap_device_enable and omap_device_idle\n from within the driver. These are being added to streamline the\n sequence between managing the hard reset lines and the clocks\n during the suspend path, as the default device pm_domain callback\n sequences in omap_device layer are not conducive for the OMAP\n IOMMU driver.\n \n This could have been done by expanding the existing pdata ops\n for reset management (like in the OMAP remoteproc driver), but\n this was chosen to avoid adding additional code in the separate\n file in the mach-omap2 layer.\n \n Signed-off-by: Suman Anna \u003cs-anna@ti.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit 086f9efae7e96ac2c829d33ca3d513caa3bbb225\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:22 2019 -0700\n\n iommu: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Joerg Roedel \u003cjoro@8bytes.org\u003e\n Cc: iommu@lists.linux-foundation.org\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit b9c6ff94e43a0ee053e0c1d983fba1ac4953b762\nAuthor: Suthikulpanit, Suravee \u003cSuravee.Suthikulpanit@amd.com\u003e\nDate: Tue Jul 23 19:00:37 2019 +0000\n\n iommu/amd: Re-factor guest virtual APIC (de-)activation code\n \n Re-factore the logic for activate/deactivate guest virtual APIC mode (GAM)\n into helper functions, and export them for other drivers (e.g. SVM).\n to support run-time activate/deactivate of SVM AVIC.\n \n Cc: Joerg Roedel \u003cjoro@8bytes.org\u003e\n Signed-off-by: Suravee Suthikulpanit \u003csuravee.suthikulpanit@amd.com\u003e\n Signed-off-by: Joerg Roedel \u003cjroedel@suse.de\u003e\n\ncommit a2ef03fe617a8365fb7794531b11ba587509a9b9\nAuthor: Tomas Espeleta \u003ctomas.espeleta@gmail.com\u003e\nDate: Fri Aug 9 16:37:54 2019 +0200\n\n ALSA: hda - Add a quirk model for fixing Huawei Matebook X right speaker\n \n [ This is rather a revival of the patch Tomas sent in months ago, but\n applying only with the quirk model option -- tiwai ]\n \n Hard coded coefficients to make Huawuei Matebook X right speaker\n work. The Matebook X has a ALC298, please refer to bug 197801 on\n how these numbers were reverse engineered from the Windows driver\n \n The reversed engineered sequence represents a repeating pattern\n of verbs, and the only values that are changing periodically are\n written on indexes 0x23 and 0x25:\n \n 0x500, 0x23\n 0x400, VALUE1\n 0x500, 0x25\n 0x400, VALUE2\n \n * skipped reading sequences (0x500 - 0xc00 sequences are ignored)\n * static values from reverse engineering are used\n \n NOTE: since a significant risk is still considered, this is provided\n as an experimental fix that isn\u0027t applied as default for now. For\n enabling the fix, you\u0027ll have to choose huawei-mbx-stereo via model\n option of snd-hda-intel module.\n \n If we get feedback from users that this works stably, we may apply it\n per default.\n \n [ Some coding style fixes and replacement with AC_VERB_* by tiwai ]\n \n BugLink: https://bugzilla.kernel.org/show_bug.cgi?id\u003d197801\n Signed-off-by: Tomas Espeleta \u003ctomas.espeleta@gmail.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit f43ec25aa5480515a39a405084ca8cb5e2c408aa\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Fri Aug 9 16:13:16 2019 +0200\n\n ALSA: hda - Inform too slow responses\n \n For the sake of debugging, show each response that took too long time\n and its corresponding last verb. Since such an error happens in\n series once when started, use the ratelimited variant for suppressing\n the flood.\n \n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit a2f99330c50bb89fe2625d374fdf22f021163b55\nAuthor: Marco Felsch \u003cm.felsch@pengutronix.de\u003e\nDate: Fri Aug 9 07:40:00 2019 -0700\n\n Input: ads7846 - add support for general touchscreen bindings\n \n A few vendor specific bindings are now covered by common bindings.\n \n Let the driver parse the common bindings to make use of common\n inverting and swapping mechnism. Aslo make use of\n touchscreen_report_pos() to ensure the correct inverting-swapping\n order.\n \n The vendor specific properties are used as default (backward\n compatibility) and gets overwritten by common bindings.\n \n Signed-off-by: Marco Felsch \u003cm.felsch@pengutronix.de\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit f0fbeaa82028eaa4e2481ab49534c2e1465c99ce\nAuthor: Marco Felsch \u003cm.felsch@pengutronix.de\u003e\nDate: Fri Aug 9 07:37:32 2019 -0700\n\n dt-bindings: input: ads7846: fix property description\n \n The ti,y-max is used for the maximum value of the Y axis.\n \n Signed-off-by: Marco Felsch \u003cm.felsch@pengutronix.de\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 9c1cac424c93d2b3122014e07a54b003ddedc168\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Fri Aug 9 15:39:37 2019 +0100\n\n arm64: mm: Really fix sparse warning in untagged_addr()\n \n untagged_addr() can be called with a \u0027__user\u0027 pointer parameter and must\n therefore use \u0027__force\u0027 casts both when passing this parameter through\n to sign_extend64() as a \u0027u64\u0027, but also when casting the \u0027s64\u0027 return\n value back to the \u0027__user\u0027 pointer type.\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit d2d73d2fef421ca0d447946cc430fdf5c4c5b06a\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Fri Aug 9 15:27:32 2019 +0100\n\n arm64: mm: Simplify definition of virt_addr_valid()\n \n _virt_addr_valid() is defined as the same value in two places and rolls\n its own version of virt_to_pfn() in both cases.\n \n Consolidate these definitions by inlining a simplified version directly\n into virt_addr_valid().\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 3148310792b64c1b79e8ce399afccad369d4b17a\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 8 08:41:52 2019 +0100\n\n drm/i915: Drop the fudge warning on ring restart for ctg/elk\n \n Since we have already stopped the ring, cleared the ring, disabled the\n ring (and verifying the ring is clear), a later debug message that the\n ring is no longer clear serves no function. It appears it restarts\n anyway, and we verify that the ring started correctly afterwards.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Acked-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808074207.18274-4-chris@chris-wilson.co.uk\n\ncommit 52f6f9d74f31078964ca1574f7bb612da7877ac8\nAuthor: Jann Horn \u003cjannh@google.com\u003e\nDate: Tue Mar 26 23:03:48 2019 +0100\n\n floppy: fix usercopy direction\n \n As sparse points out, these two copy_from_user() should actually be\n copy_to_user().\n \n Fixes: 229b53c9bf4e (\"take floppy compat ioctls to sodding floppy.c\")\n Cc: stable@vger.kernel.org\n Acked-by: Alexander Popov \u003calex.popov@linux.com\u003e\n Reviewed-by: Mukesh Ojha \u003cmojha@codeaurora.org\u003e\n Signed-off-by: Jann Horn \u003cjannh@google.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 1a07e86cceac170bede143937efd29cbc7d358a6\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 9 10:10:10 2019 +0100\n\n drm/i915: Generalise BSD default selection\n \n For the default I915_EXEC_BSD round robin selector, it may select any\n available VCS engine. Make it so.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Reviewed-by: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190809091010.23281-3-chris@chris-wilson.co.uk\n\ncommit 6b86f90019cf6286d0dc944f4bfd02c76db3b409\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 9 10:10:09 2019 +0100\n\n drm/i915: Replace global bsd_dispatch_index with random seed\n \n We keep a global seed for the legacy BSD round-robin selector, but in\n our testing of multiple simultaneous client workloads, a random seed\n spreads the load more evenly. (As even as an initial round-robin selector\n can be!) Removing the global is one less variable we have to find a home\n for!\n \n We can simulate multi-client (both same and mixed workloads) using\n igt/gem_wsim to work out optimal strategies and then compare our\n simulation with the actual transcoder on multi-engine machines. This\n fixed round-robin turns out to be one of the worst methods.\n \n No user is advised to use this method; the current suggestion is to use\n a virtual engine for agnostic batches, randomised submission or using\n the busyness tracking to select the most idle engine at the time of\n dispatch. At the present time, intel-media is explicit, but libva still\n seems to use it, with the exception of batches that must execute on vcs0.\n Oh well.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Reviewed-by: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190809091010.23281-2-chris@chris-wilson.co.uk\n\ncommit d5b2a3a4f85ad135123db8fc97ccdbf893546330\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 9 13:31:53 2019 +0100\n\n drm/i915: Check for a second VCS engine more carefully\n \n To use the legacy BSD selector, you must have a second VCS engine, or\n else the ABI simply maps the request for another engine onto VCS0.\n However, we only checked a single VCS1 location and overlooking the\n possibility of a sparse VCS set being mapped to the dense ABI.\n \n v2: num_vcs_engines() turns out to be reusable and futureproof it so we\n never have to worry about this silly bit of ABI again!\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Reviewed-by: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190809123153.20574-1-chris@chris-wilson.co.uk\n\ncommit 1f573cce48a2ebb35953034062ef01056b7d6a58\nAuthor: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\nDate: Fri Aug 9 11:20:33 2019 +0100\n\n device.h: Fix warnings for mismatched parameter names in comments\n \n Fix the warnings for parameter named as \"driver\" instead of the actual \"drv\"\n in the comments as reported by the kbuild robot.\n \n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Link: https://lore.kernel.org/r/20190809102033.28463-1-suzuki.poulose@arm.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 894efaf5236d3abf0512d8a71ae927ea9f99b509\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 9 18:21:50 2019 +0800\n\n staging: wusbcore: Fix build error without CONFIG_USB\n \n USB_WUSB should depends on CONFIG_USB, otherwise building fails\n \n drivers/staging/wusbcore/wusbhc.o: In function `wusbhc_giveback_urb\u0027:\n wusbhc.c:(.text+0xa28): undefined reference to `usb_hcd_giveback_urb\u0027\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Fixes: 71ed79b0e4be (\"USB: Move wusbcore and UWB to staging as it is obsolete\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190809102150.66896-1-yuehaibing@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 40073a9d53eac3f3ebf569aa9afd61301eff9190\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 9 19:01:00 2019 +0800\n\n ASoC: SOF: Intel: Add missing include file hdac_hda.h\n \n Building with SND_SOC_SOF_HDA_AUDIO_CODEC fails:\n \n sound/soc/sof/intel/hda-bus.c: In function sof_hda_bus_init:\n sound/soc/sof/intel/hda-bus.c:16:25: error: implicit declaration of function\n snd_soc_hdac_hda_get_ops; did you mean snd_soc_jack_add_gpiods? [-Werror\u003dimplicit-function-declaration]\n #define sof_hda_ext_ops snd_soc_hdac_hda_get_ops()\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Suggested-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n Fixes: d4ff1b3917a5 (\"ASoC: SOF: Intel: Initialize hdaudio bus properly\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 5cbdaeefb655072d304744812708b3f3a31c6b51\nAuthor: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\nDate: Tue Aug 6 01:49:19 2019 -0300\n\n s390/mm: Remove sev_active() function\n \n All references to sev_active() were moved to arch/x86 so we don\u0027t need to\n define it for s390 anymore.\n \n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Halil Pasic \u003cpasic@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190806044919.10622-7-bauerman@linux.ibm.com\n\ncommit ae7eb82a92fae4d255a0caa9f7c0f99e3babfec1\nAuthor: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\nDate: Tue Aug 6 01:49:18 2019 -0300\n\n fs/core/vmcore: Move sev_active() reference to x86 arch code\n \n Secure Encrypted Virtualization is an x86-specific feature, so it shouldn\u0027t\n appear in generic kernel code because it forces non-x86 architectures to\n define the sev_active() function, which doesn\u0027t make a lot of sense.\n \n To solve this problem, add an x86 elfcorehdr_read() function to override\n the generic weak implementation. To do that, it\u0027s necessary to make\n read_from_oldmem() public so that it can be used outside of vmcore.c.\n \n Also, remove the export for sev_active() since it\u0027s only used in files that\n won\u0027t be built as modules.\n \n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Lianbo Jiang \u003clijiang@redhat.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190806044919.10622-6-bauerman@linux.ibm.com\n\ncommit 284e21fab2cfcf90dacce565e0b12f29e5df00c1\nAuthor: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\nDate: Tue Aug 6 01:49:17 2019 -0300\n\n x86, s390/mm: Move sme_active() and sme_me_mask to x86-specific header\n \n Now that generic code doesn\u0027t reference them, move sme_active() and\n sme_me_mask to x86\u0027s \u003casm/mem_encrypt.h\u003e.\n \n Also remove the export for sme_active() since it\u0027s only used in files that\n won\u0027t be built as modules. sme_me_mask on the other hand is used in\n arch/x86/kvm/svm.c (via __sme_set() and __psp_pa()) which can be built as a\n module so its export needs to stay.\n \n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Tom Lendacky \u003cthomas.lendacky@amd.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190806044919.10622-5-bauerman@linux.ibm.com\n\ncommit e740815a97e2b6d6446792f4328378e66de166d1\nAuthor: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\nDate: Tue Aug 6 01:49:16 2019 -0300\n\n dma-mapping: Remove dma_check_mask()\n \n sme_active() is an x86-specific function so it\u0027s better not to call it from\n generic code. Christoph Hellwig mentioned that \"There is no reason why we\n should have a special debug printk just for one specific reason why there\n is a requirement for a large DMA mask.\", so just remove dma_check_mask().\n \n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Tom Lendacky \u003cthomas.lendacky@amd.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190806044919.10622-4-bauerman@linux.ibm.com\n\ncommit 47e5d8f9ed34a5310f45f48bbc15f1e61d83b6db\nAuthor: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\nDate: Tue Aug 6 01:49:15 2019 -0300\n\n swiotlb: Remove call to sme_active()\n \n sme_active() is an x86-specific function so it\u0027s better not to call it from\n generic code.\n \n There\u0027s no need to mention which memory encryption feature is active, so\n just use a more generic message. Besides, other architectures will have\n different names for similar technology.\n \n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Tom Lendacky \u003cthomas.lendacky@amd.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190806044919.10622-3-bauerman@linux.ibm.com\n\ncommit 0c9c1d56397518eb823d458b00b06bcccd956794\nAuthor: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\nDate: Tue Aug 6 01:49:14 2019 -0300\n\n x86, s390: Move ARCH_HAS_MEM_ENCRYPT definition to arch/Kconfig\n \n powerpc is also going to use this feature, so put it in a generic location.\n \n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Reviewed-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190806044919.10622-2-bauerman@linux.ibm.com\n\ncommit 6cd34b10cd642c15a8e4b2ed5fc4815c12c41d52\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 9 08:37:23 2019 +0100\n\n drm/i915/execlists: Backtrack along timeline\n \n After a preempt-to-busy, we may find an active request that is caught\n between execution states. Walk back along the timeline instead of the\n execution list to be safe.\n \n [ 106.417541] i915 0000:00:02.0: Resetting rcs0 for preemption time out\n [ 106.417659] \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n [ 106.418041] BUG: KASAN: slab-out-of-bounds in __execlists_reset+0x2f2/0x440 [i915]\n [ 106.418123] Read of size 8 at addr ffff888703506b30 by task swapper/1/0\n [ 106.418194]\n [ 106.418267] CPU: 1 PID: 0 Comm: swapper/1 Tainted: G U 5.3.0-rc3+ #5\n [ 106.418344] Hardware name: Intel Corporation NUC7i5BNK/NUC7i5BNB, BIOS BNKBL357.86A.0052.2017.0918.1346 09/18/2017\n [ 106.418434] Call Trace:\n [ 106.418508] \u003cIRQ\u003e\n [ 106.418585] dump_stack+0x5b/0x90\n [ 106.418941] ? __execlists_reset+0x2f2/0x440 [i915]\n [ 106.419022] print_address_description+0x67/0x32d\n [ 106.419376] ? __execlists_reset+0x2f2/0x440 [i915]\n [ 106.419731] ? __execlists_reset+0x2f2/0x440 [i915]\n [ 106.419810] __kasan_report.cold.6+0x1a/0x3c\n [ 106.419888] ? __trace_bprintk+0xc0/0xd0\n [ 106.420239] ? __execlists_reset+0x2f2/0x440 [i915]\n [ 106.420318] check_memory_region+0x144/0x1c0\n [ 106.420671] __execlists_reset+0x2f2/0x440 [i915]\n [ 106.421029] execlists_reset+0x3d/0x50 [i915]\n [ 106.421387] intel_engine_reset+0x203/0x3a0 [i915]\n [ 106.421744] ? igt_reset_nop+0x2b0/0x2b0 [i915]\n [ 106.421825] ? _raw_spin_trylock_bh+0xe0/0xe0\n [ 106.421901] ? rcu_core+0x1b9/0x6a0\n [ 106.422251] preempt_reset+0x9a/0xf0 [i915]\n [ 106.422333] tasklet_action_common.isra.15+0xc0/0x1e0\n [ 106.422685] ? execlists_submit_request+0x200/0x200 [i915]\n [ 106.422764] __do_softirq+0x106/0x3cf\n [ 106.422840] irq_exit+0xdc/0xf0\n [ 106.422914] smp_apic_timer_interrupt+0x81/0x1c0\n [ 106.422988] apic_timer_interrupt+0xf/0x20\n [ 106.423059] \u003c/IRQ\u003e\n [ 106.423144] RIP: 0010:cpuidle_enter_state+0xc3/0x620\n [ 106.423222] Code: 24 0f 1f 44 00 00 31 ff e8 da 87 9c ff 80 7c 24 10 00 74 12 9c 58 f6 c4 02 0f 85 33 05 00 00 31 ff e8 c1 77 a3 ff fb 45 85 e4 \u003c0f\u003e 89 bf 02 00 00 48 8d 7d 10 e8 4e 45 b9 ff c7 45 10 00 00 00 00\n [ 106.423311] RSP: 0018:ffff88881c30fda8 EFLAGS: 00000202 ORIG_RAX: ffffffffffffff13\n [ 106.423390] RAX: 0000000000000000 RBX: ffffffff825b4c80 RCX: ffffffff810c8a00\n [ 106.423465] RDX: dffffc0000000000 RSI: 0000000039f89620 RDI: ffff88881f6b00a8\n [ 106.423540] RBP: ffff88881f6b5bf8 R08: 0000000000000002 R09: 000000000002ed80\n [ 106.423616] R10: 0000003fdd956146 R11: ffff88881c2d1e47 R12: 0000000000000008\n [ 106.423691] R13: 0000000000000008 R14: ffffffff825b4f80 R15: ffffffff825b4fc0\n [ 106.423772] ? sched_idle_set_state+0x20/0x30\n [ 106.423851] ? cpuidle_enter_state+0xa6/0x620\n [ 106.423874] ? tick_nohz_idle_stop_tick+0x1d1/0x3f0\n [ 106.423896] cpuidle_enter+0x37/0x60\n [ 106.423919] do_idle+0x246/0x280\n [ 106.423941] ? arch_cpu_idle_exit+0x30/0x30\n [ 106.423964] ? __wake_up_common+0x46/0x240\n [ 106.423986] cpu_startup_entry+0x14/0x20\n [ 106.424009] start_secondary+0x1b0/0x200\n [ 106.424031] ? set_cpu_sibling_map+0x990/0x990\n [ 106.424054] secondary_startup_64+0xa4/0xb0\n [ 106.424075]\n [ 106.424096] Allocated by task 626:\n [ 106.424119] save_stack+0x19/0x80\n [ 106.424143] __kasan_kmalloc.constprop.7+0xc1/0xd0\n [ 106.424165] kmem_cache_alloc+0xb2/0x1d0\n [ 106.424277] i915_sched_lookup_priolist+0x1ab/0x320 [i915]\n [ 106.424385] execlists_submit_request+0x73/0x200 [i915]\n [ 106.424498] submit_notify+0x59/0x60 [i915]\n [ 106.424600] __i915_sw_fence_complete+0x9b/0x330 [i915]\n [ 106.424713] __i915_request_commit+0x4bf/0x570 [i915]\n [ 106.424818] intel_engine_pulse+0x213/0x310 [i915]\n [ 106.424925] context_close+0x22f/0x470 [i915]\n [ 106.425033] i915_gem_context_destroy_ioctl+0x7b/0xa0 [i915]\n [ 106.425058] drm_ioctl_kernel+0x131/0x170\n [ 106.425081] drm_ioctl+0x2d9/0x4f1\n [ 106.425104] do_vfs_ioctl+0x115/0x890\n [ 106.425126] ksys_ioctl+0x35/0x70\n [ 106.425147] __x64_sys_ioctl+0x38/0x40\n [ 106.425169] do_syscall_64+0x66/0x220\n [ 106.425191] entry_SYSCALL_64_after_hwframe+0x44/0xa9\n [ 106.425213]\n [ 106.425234] Freed by task 0:\n [ 106.425255] (stack is not available)\n [ 106.425276]\n [ 106.425297] The buggy address belongs to the object at ffff888703506a40\n [ 106.425297] which belongs to the cache i915_priolist of size 104\n [ 106.425321] The buggy address is located 136 bytes to the right of\n [ 106.425321] 104-byte region [ffff888703506a40, ffff888703506aa8)\n [ 106.425345] The buggy address belongs to the page:\n [ 106.425367] page:ffffea001c0d4180 refcount:1 mapcount:0 mapping:ffff88873e1cf740 index:0xffff888703506e40 compound_mapcount: 0\n [ 106.425391] flags: 0x8000000000010200(slab|head)\n [ 106.425415] raw: 8000000000010200 ffffea0020192b88 ffff8888174b5450 ffff88873e1cf740\n [ 106.425439] raw: ffff888703506e40 000000000010000e 00000001ffffffff 0000000000000000\n [ 106.425464] page dumped because: kasan: bad access detected\n [ 106.425486]\n [ 106.425506] Memory state around the buggy address:\n [ 106.425528] ffff888703506a00: fc fc fc fc fc fc fc fc 00 00 00 00 00 00 00 00\n [ 106.425551] ffff888703506a80: 00 00 00 00 00 fc fc fc fc fc fc fc fc fc fc fc\n [ 106.425573] \u003effff888703506b00: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc\n [ 106.425597] ^\n [ 106.425619] ffff888703506b80: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc\n [ 106.425642] ffff888703506c00: fc fc fc fc fc fc fc fc 00 00 00 00 00 00 00 00\n [ 106.425664] \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n \n Fixes: 22b7a426bbe1 (\"drm/i915/execlists: Preempt-to-busy\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190809073723.6593-1-chris@chris-wilson.co.uk\n\ncommit 3aaf84662d3193018a5f4e36dd7090297053cfc4\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 9 12:07:52 2019 +0100\n\n drm/i915: Free the imported shmemfs file for phys objects\n \n Matthew spotted that we lost the fput() for phys objects now that we are\n not relying on the core to cleanup the GEM object. (For the record, phys\n objects import the shmemfs from their original set of pages and keep it\n to provide swap space, but we never transform back into a shmem object.)\n \n Reported-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Fixes: 0c159ffef628 (\"drm/i915/gem: Defer obj-\u003ebase.resv fini until RCU callback\")\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190809110752.19763-1-chris@chris-wilson.co.uk\n\ncommit 610f29e5cc0e8d5864dd049b0b3576d9437ae7b4\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Fri Aug 9 13:06:16 2019 +0530\n\n regulator: qcom-rpmh: Update PMIC modes for PMIC5\n \n Add the PMIC5 modes and use them pmic5 ldo and smps\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190809073616.1235-4-vkoul@kernel.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 553c452d6093d66e7617ed6c68cc93547d07075f\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Fri Aug 9 13:06:15 2019 +0530\n\n regulator: qcom-rpmh: Fix pmic5_bob voltage count\n \n pmic5_bob voltages count is 136 [0,135] so update it\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190809073616.1235-3-vkoul@kernel.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 7172fb7f3abea4787ca01dda7297241c4b0f0af5\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Fri Aug 9 13:06:14 2019 +0530\n\n regulator: qcom-rpmh: Sort the compatibles\n \n It helps to keep sorted order for compatibles, so sort them\n \n Suggested-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190809073616.1235-2-vkoul@kernel.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c6e20fa49818381dfa7288fad4c33b84408aab54\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Fri Aug 9 13:06:13 2019 +0530\n\n regulator: dt-bindings: Sort the compatibles and nodes\n \n It helps to keep sorted order for compatibles and nodes, so sort them\n \n Suggested-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190809073616.1235-1-vkoul@kernel.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 83ee7ab1627b7597bbb9acc26b37e9f96fff40c9\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Wed Aug 7 10:02:01 2019 -0500\n\n ASoC: SOF: Intel: byt: Refactor fw ready / mem windows creation\n \n There is a lot of duplicate code when processing IPC firmware ready\n notification and creating memory windows.\n \n First step in reducing the code duplication is to introduce generic\n functions:\n * sof_get_windows\n * sof_fw_ready\n that will replace, in the first step, the specific implementation related\n to baytrail related platforms:\n * byt_get_windows\n * byt_fw_ready\n \n So we are basically moving code from intel/byt.c to loader.c keeping\n in mind that mbox_offset is a per platform constant so we need to\n use newly introduced snd_sof_dsp_get_mailbox_offset /\n snd_sof_dsp_get_window_offset in order to get the correct\n mbox offset / window offset value.\n \n Also, bar is a per platform constant so we use snd_sof_dsp_get_bar_index\n instead of the hardcoded BYT_DSP_BAR.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190807150203.26359-5-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit e17422cda9d60339d71e117d3fdfd444e83669a4\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Wed Aug 7 10:02:00 2019 -0500\n\n ASoC: SOF: Introduce snd_sof_dsp_get_window_offset\n \n This will allow us to export the offset for a memory window.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190807150203.26359-4-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit bb9c93f58cb66049cb5a5570cc476d9a8a97f491\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Wed Aug 7 10:01:59 2019 -0500\n\n ASoC: SOF: Introduce snd_sof_dsp_get_mailbox_offset\n \n This will allow us to export mailbox offset in order to\n read the fw_ready message from.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190807150203.26359-3-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 6eebd3907527764618ee250e1106b9e809c46143\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Wed Aug 7 10:02:03 2019 -0500\n\n ASoC: SOF: Intel: hda: Use generic function for fw ready / mem windows creation\n \n We can use generic sof_fw_ready function and reduce code duplication.\n Careful here that we need to provide the implementation for\n get_mailbox_offset and get_window_offset.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190807150203.26359-7-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit ddf14b640b05839cfe44d686d5a5cc7c55076094\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Wed Aug 7 10:02:02 2019 -0500\n\n ASoC: SOF: Intel: bdw: Use generic function for fw ready / mem windows creation\n \n bdw_get_windows / bdw_fw_ready is identical with the generic\n implementation introduced in a previous patch.\n \n So remove bdw_get_windows / bdw_fw_ready and use the generic\n sof_get_windows version.\n \n Do not forget to implement get_mailbox_offset/get_window_offset\n so that we export the correct mailbox/memory window offset to\n the outside world.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190807150203.26359-6-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 0b76f512c206dd7c3f7761b24826887af80204c6\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Wed Aug 7 10:01:58 2019 -0500\n\n ASoC: SOF: Remove call to snd_sof_dsp_mailbox_init\n \n This is reserved for some historical reason, we didn\u0027t enable memory\n windows for byt/bdw at the beginning, to make it compatible, we get\n those mailbox offsets from fw_ready struct firstly, and then update them\n if they existed in the following memory windows, to make sure the\n mailbox still can be used if no memory windows are created.\n \n With this change all platforms have the same implementation for\n xxx_fw_ready function so that we can refactor it in a common file.\n \n Suggested-by: Keyon Jie \u003cyang.jie@linux.intel.com\u003e\n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190807150203.26359-2-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 0fd70e22a0ffebd13028bf2c7da6b747070475bf\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 9 16:24:40 2019 +0800\n\n ASoC: ml26124: remove unused variable \u0027ngth\u0027\n \n In file included from ./include/sound/tlv.h:10:0,\n from sound/soc/codecs/ml26124.c:19:\n sound/soc/codecs/ml26124.c:59:35: warning: ngth defined but not used [-Wunused-const-variable\u003d]\n static const DECLARE_TLV_DB_SCALE(ngth, -7650, 150, 0);\n ^\n ./include/uapi/sound/tlv.h:64:15: note: in definition of macro SNDRV_CTL_TLVD_DECLARE_DB_SCALE\n unsigned int name[] \u003d { \\\n ^~~~\n sound/soc/codecs/ml26124.c:59:14: note: in expansion of macro DECLARE_TLV_DB_SCALE\n static const DECLARE_TLV_DB_SCALE(ngth, -7650, 150, 0);\n ^~~~~~~~~~~~~~~~~~~~\n \n It is never used, so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190809082440.67412-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit bc8d9f737fc01cce913f1cc215b7e66f01697e52\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 9 16:02:34 2019 +0800\n\n ASoC: mt6351: remove unused variable \u0027mt_lineout_control\u0027\n \n sound/soc/codecs/mt6351.c:1070:38: warning:\n mt_lineout_control defined but not used [-Wunused-const-variable\u003d]\n \n It is never used, so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190809080234.23332-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit a62bd63893027bfa32fccbba0e0ac067824c362c\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 9 19:01:00 2019 +0800\n\n ASoC: SOF: Intel: Add missing include file hdac_hda.h\n \n Building with SND_SOC_SOF_HDA_AUDIO_CODEC fails:\n \n sound/soc/sof/intel/hda-bus.c: In function sof_hda_bus_init:\n sound/soc/sof/intel/hda-bus.c:16:25: error: implicit declaration of function\n snd_soc_hdac_hda_get_ops; did you mean snd_soc_jack_add_gpiods? [-Werror\u003dimplicit-function-declaration]\n #define sof_hda_ext_ops snd_soc_hdac_hda_get_ops()\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Suggested-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n Fixes: d4ff1b3917a5 (\u0027ASoC: SOF: Intel: Initialize hdaudio bus properly\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190809110100.71236-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 15ff4a7b584163b12b118a2c381529f05ff3a94d\nAuthor: Christian Gmeiner \u003cchristian.gmeiner@gmail.com\u003e\nDate: Wed Jul 31 23:30:34 2019 +0200\n\n etnaviv: perfmon: fix total and idle HI cyleces readout\n \n As seen at CodeAurora\u0027s linux-imx git repo in imx_4.19.35_1.0.0 branch.\n \n Signed-off-by: Christian Gmeiner \u003cchristian.gmeiner@gmail.com\u003e\n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n\ncommit 1b53591a77d9c1a8c68cf2ea94682dcdb471199b\nAuthor: Christian Gmeiner \u003cchristian.gmeiner@gmail.com\u003e\nDate: Fri Aug 2 13:26:56 2019 +0200\n\n etnaviv: fix whitespace errors\n \n Changes in V2:\n - use indentation as suggested by Philipp Zabel.\n \n Signed-off-by: Christian Gmeiner \u003cchristian.gmeiner@gmail.com\u003e\n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n\ncommit 240b65dc1e89dd83d0c4946f58e49e7ee669b25d\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Fri Jul 12 00:26:34 2019 -0700\n\n usb: dwc3: Use clk_bulk_prepare_enable()\n \n Use clk_bulk_prepare_enable() and clk_bulk_disable_unprepare() to\n simplify code a bit. No functional change intended.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Felipe Balbi \u003cbalbi@kernel.org\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: linux-usb@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 03bf32bbb607e07bf9ac6ea126088d1dc88ea948\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Fri Jul 12 00:26:33 2019 -0700\n\n usb: dwc3: Use devres to get clocks\n \n Use devres to get clocks and drop explicit clock freeing. No\n functional change intended.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Felipe Balbi \u003cbalbi@kernel.org\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: linux-usb@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 3cd703f483d1d8cfcd7fd6f304ab6fc74e29cf15\nAuthor: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\nDate: Fri Jul 19 11:30:37 2019 +0200\n\n usb: dwc3: remove generic PHYs forwarding for XHCI device\n \n Commit 08f871a3aca2 (\"usb: dwc3: host: convey the PHYs to xhci\") added\n forwarding of the generic PHYs from DWC3 core to the instantiated XHCI-plat\n device. However XHCI(-plat) driver never gained support for generic PHYs,\n thus the lookup added by that commit is never used. In meantime the commit\n d64ff406e51e (\"usb: dwc3: use bus-\u003esysdev for DMA configuration\")\n incorrectly changed the device used for creating lookup, making the lookup\n useless and generic PHYs inaccessible from XHCI device.\n \n However since commit 178a0bce05cb (\"usb: core: hcd: integrate the PHY\n wrapper into the HCD core\") USB HCD already handles generic PHYs acquired\n from the HCD\u0027s \u0027sysdev\u0027, which in this case is DWC3 core device. This means\n that creating any custom lookup entries for XHCI driver is no longer needed\n and can be simply removed.\n \n Signed-off-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit d2c68de192cfb90f607a80c6b10c41ebd8a3de6a\nAuthor: Steve Capper \u003csteve.capper@arm.com\u003e\nDate: Wed Aug 7 16:55:24 2019 +0100\n\n docs: arm64: Add layout and 52-bit info to memory document\n \n As the kernel no longer prints out the memory layout on boot, this patch\n adds this information back to the memory document.\n \n Also, as the 52-bit support introduces some subtle changes to the arm64\n memory, the rationale behind these changes are also added to the memory\n document.\n \n Signed-off-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 2c624fe68715e76eba1a7089f91e122310dc663c\nAuthor: Steve Capper \u003csteve.capper@arm.com\u003e\nDate: Wed Aug 7 16:55:23 2019 +0100\n\n arm64: mm: Remove vabits_user\n \n Previous patches have enabled 52-bit kernel + user VAs and there is no\n longer any scenario where user VA !\u003d kernel VA size.\n \n This patch removes the, now redundant, vabits_user variable and replaces\n usage with vabits_actual where appropriate.\n \n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit b6d00d47e81a49f6cf462518c10408f37a3e6785\nAuthor: Steve Capper \u003csteve.capper@arm.com\u003e\nDate: Wed Aug 7 16:55:22 2019 +0100\n\n arm64: mm: Introduce 52-bit Kernel VAs\n \n Most of the machinery is now in place to enable 52-bit kernel VAs that\n are detectable at boot time.\n \n This patch adds a Kconfig option for 52-bit user and kernel addresses\n and plumbs in the requisite CONFIG_ macros as well as sets TCR.T1SZ,\n physvirt_offset and vmemmap at early boot.\n \n To simplify things this patch also removes the 52-bit user/48-bit kernel\n kconfig option.\n \n Signed-off-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit ce3aaed87344c83c77135f80e7b76e1da9c92ee6\nAuthor: Steve Capper \u003csteve.capper@arm.com\u003e\nDate: Wed Aug 7 16:55:21 2019 +0100\n\n arm64: mm: Modify calculation of VMEMMAP_SIZE\n \n In a later patch we will need to have a slightly larger VMEMMAP region\n to accommodate boot time selection between 48/52-bit kernel VAs.\n \n This patch modifies the formula for computing VMEMMAP_SIZE to depend\n explicitly on the PAGE_OFFSET and start of kernel addressable memory.\n (This allows for a slightly larger direct linear map in future).\n \n Signed-off-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit c8b6d2ccf9b10ce872cdea037f9685804440bb7e\nAuthor: Steve Capper \u003csteve.capper@arm.com\u003e\nDate: Wed Aug 7 16:55:20 2019 +0100\n\n arm64: mm: Separate out vmemmap\n \n vmemmap is a preprocessor definition that depends on a variable,\n memstart_addr. In a later patch we will need to expand the size of\n the VMEMMAP region and optionally modify vmemmap depending upon\n whether or not hardware support is available for 52-bit virtual\n addresses.\n \n This patch changes vmemmap to be a variable. As the old definition\n depended on a variable load, this should not affect performance\n noticeably.\n \n Signed-off-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit c812026c54cfaec23fa1d78cdbfd0e56e787470a\nAuthor: Steve Capper \u003csteve.capper@arm.com\u003e\nDate: Wed Aug 7 16:55:19 2019 +0100\n\n arm64: mm: Logic to make offset_ttbr1 conditional\n \n When running with a 52-bit userspace VA and a 48-bit kernel VA we offset\n ttbr1_el1 to allow the kernel pagetables with a 52-bit PTRS_PER_PGD to\n be used for both userspace and kernel.\n \n Moving on to a 52-bit kernel VA we no longer require this offset to\n ttbr1_el1 should we be running on a system with HW support for 52-bit\n VAs.\n \n This patch introduces conditional logic to offset_ttbr1 to query\n SYS_ID_AA64MMFR2_EL1 whenever 52-bit VAs are selected. If there is HW\n support for 52-bit VAs then the ttbr1 offset is skipped.\n \n We choose to read a system register rather than vabits_actual because\n offset_ttbr1 can be called in places where the kernel data is not\n actually mapped.\n \n Calls to offset_ttbr1 appear to be made from rarely called code paths so\n this extra logic is not expected to adversely affect performance.\n \n Signed-off-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 5383cc6efed13784ddb3cff2cc183b6b8c50c8db\nAuthor: Steve Capper \u003csteve.capper@arm.com\u003e\nDate: Wed Aug 7 16:55:18 2019 +0100\n\n arm64: mm: Introduce vabits_actual\n \n In order to support 52-bit kernel addresses detectable at boot time, one\n needs to know the actual VA_BITS detected. A new variable vabits_actual\n is introduced in this commit and employed for the KVM hypervisor layout,\n KASAN, fault handling and phys-to/from-virt translation where there\n would normally be compile time constants.\n \n In order to maintain performance in phys_to_virt, another variable\n physvirt_offset is introduced.\n \n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 90ec95cda91a021d82351c976896a63aa364ebf1\nAuthor: Steve Capper \u003csteve.capper@arm.com\u003e\nDate: Wed Aug 7 16:55:17 2019 +0100\n\n arm64: mm: Introduce VA_BITS_MIN\n \n In order to support 52-bit kernel addresses detectable at boot time, the\n kernel needs to know the most conservative VA_BITS possible should it\n need to fall back to this quantity due to lack of hardware support.\n \n A new compile time constant VA_BITS_MIN is introduced in this patch and\n it is employed in the KASAN end address, KASLR, and EFI stub.\n \n For Arm, if 52-bit VA support is unavailable the fallback is to 48-bits.\n \n In other words: VA_BITS_MIN \u003d min (48, VA_BITS)\n \n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 99426e5e8c9f11b9de65e7c1200868e8a9ceaa47\nAuthor: Steve Capper \u003csteve.capper@arm.com\u003e\nDate: Wed Aug 7 16:55:16 2019 +0100\n\n arm64: dump: De-constify VA_START and KASAN_SHADOW_START\n \n The kernel page table dumper assumes that the placement of VA regions is\n constant and determined at compile time. As we are about to introduce\n variable VA logic, we need to be able to determine certain regions at\n boot time.\n \n Specifically the VA_START and KASAN_SHADOW_START will depend on whether\n or not the system is booted with 52-bit kernel VAs.\n \n This patch adds logic to the kernel page table dumper s.t. these regions\n can be computed at boot time.\n \n Signed-off-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 6bd1d0be0e97936d15cdacc71f5c232fbf71293e\nAuthor: Steve Capper \u003csteve.capper@arm.com\u003e\nDate: Wed Aug 7 16:55:15 2019 +0100\n\n arm64: kasan: Switch to using KASAN_SHADOW_OFFSET\n \n KASAN_SHADOW_OFFSET is a constant that is supplied to gcc as a command\n line argument and affects the codegen of the inline address sanetiser.\n \n Essentially, for an example memory access:\n *ptr1 \u003d val;\n The compiler will insert logic similar to the below:\n shadowValue \u003d *(ptr1 \u003e\u003e KASAN_SHADOW_SCALE_SHIFT + KASAN_SHADOW_OFFSET)\n if (somethingWrong(shadowValue))\n flagAnError();\n \n This code sequence is inserted into many places, thus\n KASAN_SHADOW_OFFSET is essentially baked into many places in the kernel\n text.\n \n If we want to run a single kernel binary with multiple address spaces,\n then we need to do this with KASAN_SHADOW_OFFSET fixed.\n \n Thankfully, due to the way the KASAN_SHADOW_OFFSET is used to provide\n shadow addresses we know that the end of the shadow region is constant\n w.r.t. VA space size:\n KASAN_SHADOW_END \u003d ~0 \u003e\u003e KASAN_SHADOW_SCALE_SHIFT + KASAN_SHADOW_OFFSET\n \n This means that if we increase the size of the VA space, the start of\n the KASAN region expands into lower addresses whilst the end of the\n KASAN region is fixed.\n \n Currently the arm64 code computes KASAN_SHADOW_OFFSET at build time via\n build scripts with the VA size used as a parameter. (There are build\n time checks in the C code too to ensure that expected values are being\n derived). It is sufficient, and indeed is a simplification, to remove\n the build scripts (and build time checks) entirely and instead provide\n KASAN_SHADOW_OFFSET values.\n \n This patch removes the logic to compute the KASAN_SHADOW_OFFSET in the\n arm64 Makefile, and instead we adopt the approach used by x86 to supply\n offset values in kConfig. To help debug/develop future VA space changes,\n the Makefile logic has been preserved in a script file in the arm64\n Documentation folder.\n \n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 14c127c957c1c6070647c171e72f06e0db275ebf\nAuthor: Steve Capper \u003csteve.capper@arm.com\u003e\nDate: Wed Aug 7 16:55:14 2019 +0100\n\n arm64: mm: Flip kernel VA space\n \n In order to allow for a KASAN shadow that changes size at boot time, one\n must fix the KASAN_SHADOW_END for both 48 \u0026 52-bit VAs and \"grow\" the\n start address. Also, it is highly desirable to maintain the same\n function addresses in the kernel .text between VA sizes. Both of these\n requirements necessitate us to flip the kernel address space halves s.t.\n the direct linear map occupies the lower addresses.\n \n This patch puts the direct linear map in the lower addresses of the\n kernel VA range and everything else in the higher ranges.\n \n We need to adjust:\n *) KASAN shadow region placement logic,\n *) KASAN_SHADOW_OFFSET computation logic,\n *) virt_to_phys, phys_to_virt checks,\n *) page table dumper.\n \n These are all small changes, that need to take place atomically, so they\n are bundled into this commit.\n \n As part of the re-arrangement, a guard region of 2MB (to preserve\n alignment for fixed map) is added after the vmemmap. Otherwise the\n vmemmap could intersect with IS_ERR pointers.\n \n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 9cb1c5ddd2c432834dd4f40c0170d6b639e8e5c3\nAuthor: Steve Capper \u003csteve.capper@arm.com\u003e\nDate: Wed Aug 7 16:55:13 2019 +0100\n\n arm64: mm: Remove bit-masking optimisations for PAGE_OFFSET and VMEMMAP_START\n \n Currently there are assumptions about the alignment of VMEMMAP_START\n and PAGE_OFFSET that won\u0027t be valid after this series is applied.\n \n These assumptions are in the form of bitwise operators being used\n instead of addition and subtraction when calculating addresses.\n \n This patch replaces these bitwise operators with addition/subtraction.\n \n Signed-off-by: Steve Capper \u003csteve.capper@arm.com\u003e\n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 85ab9d954698961960240622de4fad85c7d8a61e\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Wed Jul 31 10:40:19 2019 +0200\n\n clk: meson: g12a: expose CPUB clock ID for G12B\n \n Expose the CPUB clock id to add DVFS to the second CPU cluster of\n the Amlogic G12B SoC.\n \n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n\ncommit ffae8475b90c045ca508675794c08a0c8c12d202\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Wed Jul 31 10:40:18 2019 +0200\n\n clk: meson: g12a: add notifiers to handle cpu clock change\n \n In order to implement clock switching for the CLKID_CPU_CLK and\n CLKID_CPUB_CLK, notifiers are added on specific points of the\n clock tree :\n \n cpu_clk / cpub_clk\n | \\- cpu_clk_dyn\n | | \\- cpu_clk_premux0\n | | |- cpu_clk_postmux0\n | | | |- cpu_clk_dyn0_div\n | | | \\- xtal/fclk_div2/fclk_div3\n | | \\- xtal/fclk_div2/fclk_div3\n | \\- cpu_clk_premux1\n | |- cpu_clk_postmux1\n | | |- cpu_clk_dyn1_div\n | | \\- xtal/fclk_div2/fclk_div3\n | \\- xtal/fclk_div2/fclk_div3\n \\ sys_pll / sys1_pll\n \n This for each cluster, a single one for G12A, two for G12B.\n \n Each cpu_clk_premux1 tree is marked as read-only and CLK_SET_RATE_NO_REPARENT,\n to be used as \"parking\" clock in a safe clock frequency.\n \n A notifier is added on each cpu_clk_premux0 to detech when CCF want to\n change the frequency of the cpu_clk_dyn tree.\n In this notifier, the cpu_clk_premux1 tree is configured to use the xtal\n clock and then the cpu_clk_dyn is switch to cpu_clk_premux1 while CCF\n updates the cpu_clk_premux0 tree.\n \n A notifier is added on each sys_pll/sys1_pll to detect when CCF wants to\n change the PLL clock source of the cpu_clk.\n In this notifier, the cpu_clk is switched to cpu_clk_dyn while CCF\n updates the sys_pll/sys1_pll frequency.\n \n A third small notifier is added on each cpu_clk / cpub_clk and cpu_clk_dyn,\n add a small delay at PRE_RATE_CHANGE/POST_RATE_CHANGE to let the other\n notofiers change propagate before changing the cpu_clk_premux0 and sys_pll\n clock trees.\n \n This notifier set permits switching the cpu_clk / cpub_clk without any\n glitches and using a safe parking clock while switching between sub-GHz\n clocks using the cpu_clk_dyn tree.\n \n This setup has been tested and validated on the Amlogic G12A and G12B\n SoCs running the arm64 cpuburn at [1] and cycling between all the possible\n cpufreq translations of each cluster and checking the final frequency using\n the clock-measurer, script at [2].\n \n [1] https://github.com/ssvb/cpuburn-arm/blob/master/cpuburn-a53.S\n [2] https://gist.github.com/superna9999/d4de964dbc0f84b7d527e1df2ddea25f\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n\ncommit 26d34431add04a98a60b8935c25765914fa773f7\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Wed Jul 31 10:40:17 2019 +0200\n\n clk: meson: add g12a cpu dynamic divider driver\n \n Add a clock driver for the cpu dynamic divider, this divider needs\n to have a flag set before setting the divider value then removed\n while writing the new value to the register.\n \n This drivers implements this behavior and will be used essentially\n on the Amlogic G12A and G12B SoCs for cpu clock trees.\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n\ncommit 3567894b6914813299300019e028874927210880\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Wed Jul 31 10:40:16 2019 +0200\n\n clk: core: introduce clk_hw_set_parent()\n \n Introduce the clk_hw_set_parent() provider call to change parent of\n a clock by using the clk_hw pointers.\n \n This eases the clock reparenting from clock rate notifiers and\n implementing DVFS with simpler code avoiding the boilerplates\n functions as __clk_lookup(clk_hw_get_name()) then clk_set_parent().\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Acked-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Acked-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n\ncommit a84ddb83806ed43a67f6fe6a03b0b4695cc314c3\nAuthor: Julien Masson \u003cjmasson@baylibre.com\u003e\nDate: Mon Jun 24 16:49:12 2019 +0200\n\n drm: meson: venc: set the correct macrovision max amplitude value\n \n According to the register description of ENCI_MACV_MAX_AMP, the\n macrovision max amplitude value should be:\n - hdmi 480i \u003d\u003e 0xb\n - hdmi 576i \u003d\u003e 0x7\n \n The max value is 0x7ff (10 bits).\n \n Signed-off-by: Julien Masson \u003cjmasson@baylibre.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/86mui782dt.fsf@baylibre.com\n\ncommit 0703146060786f972aaa22cad9c877a6067fd47d\nAuthor: Julien Masson \u003cjmasson@baylibre.com\u003e\nDate: Mon Jun 24 16:49:04 2019 +0200\n\n drm: meson: add macro used to enable HDMI PLL\n \n This patch add new macro HHI_HDMI_PLL_CNTL_EN which is used to enable\n HDMI PLL.\n \n Signed-off-by: Julien Masson \u003cjmasson@baylibre.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/86o92n82e1.fsf@baylibre.com\n\ncommit e1012141242db89259175eeaac36b3de04377664\nAuthor: Julien Masson \u003cjmasson@baylibre.com\u003e\nDate: Mon Jun 24 16:48:57 2019 +0200\n\n drm: meson: global clean-up\n \n This patch aims to:\n - Add general and TODO comments\n - Respect coding style for multi-line comments\n - Align macro definitions\n - Remove useless macro\n \n Signed-off-by: Julien Masson \u003cjmasson@baylibre.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/86pnn382e8.fsf@baylibre.com\n\ncommit 7eef9e6104545e3aed75ac84129ab332e71b6557\nAuthor: Julien Masson \u003cjmasson@baylibre.com\u003e\nDate: Mon Jun 24 16:48:50 2019 +0200\n\n drm: meson: venc: use proper macros instead of magic constants\n \n This patch add new macros which are used to set the following\n registers:\n - ENCI_CFILT_CTRL\n - ENCI_CFILT_CTRL2\n - ENCI_MACV_MAX_AMP\n - ENCI_VIDEO_MODE_ADV\n - ENCI_VFIFO2VD_CTL\n - ENCI_VIDEO_EN\n - ENCP_VIDEO_MODE\n - VPU_HDMI_SETTING\n - VENC_UPSAMPLE_CTRL0\n - VENC_UPSAMPLE_CTRL1\n - VENC_UPSAMPLE_CTRL2\n - VENC_VDAC_FIFO_CTRL\n - VENC_VDAC_DAC0_FILT_CTRL0\n - VENC_INTCTRL\n \n Signed-off-by: Julien Masson \u003cjmasson@baylibre.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/86r27j82ef.fsf@baylibre.com\n\ncommit 147ae1cbaa18429b9450fd47136a29653294aaad\nAuthor: Julien Masson \u003cjmasson@baylibre.com\u003e\nDate: Mon Jun 24 16:48:43 2019 +0200\n\n drm: meson: viu: use proper macros instead of magic constants\n \n This patch add new macros which are used to set the following\n registers:\n - VIU_SW_RESET\n - VIU_OSD1_CTRL_STAT\n - VIU_OSD2_CTRL_STAT\n - VIU_OSD1_FIFO_CTRL_STAT\n - VIU_OSD2_FIFO_CTRL_STAT\n - VIU_MISC_CTRL0\n - VIU_OSD_BLEND_CTRL\n - OSD1_BLEND_SRC_CTRL\n - OSD2_BLEND_SRC_CTRL\n - DOLBY_PATH_CTRL\n \n Signed-off-by: Julien Masson \u003cjmasson@baylibre.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n [narmstrong: fix OSD1_BLEND_SRC_CTRL register init value for G12A]\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/86sgrz82em.fsf@baylibre.com\n\ncommit 0ce266d018f4749c7049df76d7f670a675bfa9c6\nAuthor: Julien Masson \u003cjmasson@baylibre.com\u003e\nDate: Mon Jun 24 16:48:35 2019 +0200\n\n drm: meson: vpp: use proper macros instead of magic constants\n \n This patch add new macros which are used to set the following\n registers:\n - VPP_OSD_SCALE_COEF_IDX\n - VPP_DOLBY_CTRL\n - VPP_OFIFO_SIZE\n - VPP_HOLD_LINES\n - VPP_SC_MISC\n - VPP_VADJ_CTRL\n \n Signed-off-by: Julien Masson \u003cjmasson@baylibre.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n [narmstrong: put back 0x1020080 in VPP_DUMMY_DATA1 for GXM]\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/86tvcf82eu.fsf@baylibre.com\n\ncommit bfb86819829e5ed5eaba4b9449ecfe6db4f9f91e\nAuthor: Julien Masson \u003cjmasson@baylibre.com\u003e\nDate: Mon Jun 24 16:48:27 2019 +0200\n\n drm: meson: drv: use macro when initializing vpu\n \n This patch add new macro which is used to set WRARB/RDARB mode of\n the VPU.\n \n Signed-off-by: Julien Masson \u003cjmasson@baylibre.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/86v9wv82f1.fsf@baylibre.com\n\ncommit 39bf9985b8598f20a3bf49844d2ac538a0d5697f\nAuthor: Julien Masson \u003cjmasson@baylibre.com\u003e\nDate: Mon Jun 24 16:48:12 2019 +0200\n\n drm: meson: crtc: use proper macros instead of magic constants\n \n This patch add new macros which describe couple bits field of the\n following registers:\n - VD1_BLEND_SRC_CTRL\n - VPP_SC_MISC\n \n Signed-off-by: Julien Masson \u003cjmasson@baylibre.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/86wohb82fa.fsf@baylibre.com\n\ncommit f237bf2de82eafd224eb981c6c0bca8a9e039af6\nAuthor: Julien Masson \u003cjmasson@baylibre.com\u003e\nDate: Mon Jun 24 16:47:56 2019 +0200\n\n drm: meson: mask value when writing bits relaxed\n \n The value used in the macro writel_bits_relaxed has to be masked since\n we don\u0027t want change the bits outside the mask.\n \n Signed-off-by: Julien Masson \u003cjmasson@baylibre.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/86y31r82fo.fsf@baylibre.com\n\ncommit cc911c2251f07470f8e085ec8ff670e6bdd56add\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 9 17:01:13 2019 +0800\n\n ALSA: ac97: remove unused variable \u0027snd_ac97_controls_master_mono\u0027\n \n sound/pci/ac97/ac97_codec.c:599:38: warning:\n snd_ac97_controls_master_mono defined but not used [-Wunused-const-variable\u003d]\n \n It is never used, so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 5e93ec4d47e29a6f37b36c822a5c8832f8f36cd9\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Fri Jul 5 19:15:36 2019 +0200\n\n drm/etnaviv: remove unused function etnaviv_gem_mapping_reference\n \n Hasn\u0027t been used for quite a while. There is no point in keeping\n unused code around.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Reviewed-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n\ncommit c53ab61312ceb2c2c8fdf50137361a77ef0fc155\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Fri Jul 5 19:15:35 2019 +0200\n\n drm/etnaviv: fix etnaviv_cmdbuf_suballoc_new return value\n \n The call site expects to get either a valid suballoc or an error\n pointer, so a NULL return will not be treated as an error. Make\n sure to always return a proper error pointer in case something goes\n wrong.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Reviewed-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n\ncommit be80bc3658e811f9880fbb73e03a1834d0a2ba0f\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Thu Aug 8 16:42:49 2019 +0300\n\n drm/i915: extract i915_gem_shrinker.h from i915_drv.h\n \n It used to be handy that we only had a couple of headers, but over time\n i915_drv.h has become unwieldy. Extract declarations to a separate\n header file corresponding to the implementation module, clarifying the\n modularity of the driver.\n \n Ensure the new header is self-contained, and do so with minimal further\n includes, using forward declarations as needed. Include the new header\n from i915_drv.h to avoid sprinkling includes all over the place; this\n can be changed as a follow-up if necessary.\n \n No functional changes.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/b8406f72ce5bfb8863a54003b756ebae8b17c9cb.1565271681.git.jani.nikula@intel.com\n\ncommit 6401fafbec5a3363b7313669818b468a2dd79903\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Thu Aug 8 16:42:48 2019 +0300\n\n drm/i915: extract gem/i915_gem_stolen.h from i915_drv.h\n \n It used to be handy that we only had a couple of headers, but over time\n i915_drv.h has become unwieldy. Extract declarations to a separate\n header file corresponding to the implementation module, clarifying the\n modularity of the driver.\n \n Ensure the new header is self-contained, and do so with minimal further\n includes, using forward declarations as needed. Include the new header\n from i915_drv.h to avoid sprinkling includes all over the place; this\n can be changed as a follow-up if necessary.\n \n No functional changes.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/0141b4e1f1bf2deb65730ce6973863a3a16ab38f.1565271681.git.jani.nikula@intel.com\n\ncommit 9c9082b98228b0fdb9ef3dec6d67169b05017859\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Thu Aug 8 16:42:47 2019 +0300\n\n drm/i915: extract i915_memcpy.h from i915_drv.h\n \n It used to be handy that we only had a couple of headers, but over time\n i915_drv.h has become unwieldy. Extract declarations to a separate\n header file corresponding to the implementation module, clarifying the\n modularity of the driver.\n \n Ensure the new header is self-contained, and do so with minimal further\n includes, using forward declarations as needed. Include the new header\n only where needed, and sort the modified include directives while at it\n and as needed.\n \n No functional changes.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/f2b887002150acdf218385ea846f7aa617aa5f15.1565271681.git.jani.nikula@intel.com\n\ncommit bdd1510cc7c2940bdd12687559b24b04043dff9f\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Thu Aug 8 16:42:46 2019 +0300\n\n drm/i915: extract i915_suspend.h from i915_drv.h\n \n It used to be handy that we only had a couple of headers, but over time\n i915_drv.h has become unwieldy. Extract declarations to a separate\n header file corresponding to the implementation module, clarifying the\n modularity of the driver.\n \n Ensure the new header is self-contained, and do so with minimal further\n includes, using forward declarations as needed. Include the new header\n only where needed, and sort the modified include directives while at it\n and as needed.\n \n No functional changes.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/94f2884a3e5611c3e1f015104afb965e47bd8992.1565271681.git.jani.nikula@intel.com\n\ncommit be68261d815f97940c0b776dcc42c2c5253acee3\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Thu Aug 8 16:42:45 2019 +0300\n\n drm/i915: extract i915_sysfs.h from i915_drv.h\n \n It used to be handy that we only had a couple of headers, but over time\n i915_drv.h has become unwieldy. Extract declarations to a separate\n header file corresponding to the implementation module, clarifying the\n modularity of the driver.\n \n Ensure the new header is self-contained, and do so with minimal further\n includes, using forward declarations as needed. Include the new header\n only where needed, and sort the modified include directives while at it\n and as needed.\n \n No functional changes.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/2276d0401a52389fe3aafe7e62b07a198353045e.1565271681.git.jani.nikula@intel.com\n\ncommit db94e9f133a04169a0e119f5b85a76c365296ac2\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Thu Aug 8 16:42:44 2019 +0300\n\n drm/i915: extract i915_perf.h from i915_drv.h\n \n It used to be handy that we only had a couple of headers, but over time\n i915_drv.h has become unwieldy. Extract declarations to a separate\n header file corresponding to the implementation module, clarifying the\n modularity of the driver.\n \n Ensure the new header is self-contained, and do so with minimal further\n includes, using forward declarations as needed. Include the new header\n only where needed, and sort the modified include directives while at it\n and as needed.\n \n No functional changes.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/d7826e365695f691a3ac69a69ff6f2bbdb62700d.1565271681.git.jani.nikula@intel.com\n\ncommit 358c855cb60a27a2e91e002604d2c9ed03bdcb96\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Thu Aug 8 16:42:43 2019 +0300\n\n drm/i915: move printing and load error inject to i915_utils.[ch]\n \n Seems like a better fit. Reduce clutter in i915_drv.[ch].\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/bc458fa5e62fc8dae46216666f64ed6976fafaee.1565271681.git.jani.nikula@intel.com\n\ncommit 2c2b005f549544c13ef4cfb0e4842949066889bc\nAuthor: Jiri Slaby \u003cjslaby@suse.cz\u003e\nDate: Wed Aug 7 13:10:37 2019 +0200\n\n ACPI / processor: don\u0027t print errors for processorIDs \u003d\u003d 0xff\n \n Some platforms define their processors in this manner:\n Device (SCK0)\n {\n Name (_HID, \"ACPI0004\" /* Module Device */) // _HID: Hardware ID\n Name (_UID, \"CPUSCK0\") // _UID: Unique ID\n Processor (CP00, 0x00, 0x00000410, 0x06){}\n Processor (CP01, 0x02, 0x00000410, 0x06){}\n Processor (CP02, 0x04, 0x00000410, 0x06){}\n Processor (CP03, 0x06, 0x00000410, 0x06){}\n Processor (CP04, 0x01, 0x00000410, 0x06){}\n Processor (CP05, 0x03, 0x00000410, 0x06){}\n Processor (CP06, 0x05, 0x00000410, 0x06){}\n Processor (CP07, 0x07, 0x00000410, 0x06){}\n Processor (CP08, 0xFF, 0x00000410, 0x06){}\n Processor (CP09, 0xFF, 0x00000410, 0x06){}\n Processor (CP0A, 0xFF, 0x00000410, 0x06){}\n Processor (CP0B, 0xFF, 0x00000410, 0x06){}\n ...\n \n The processors marked as 0xff are invalid, there are only 8 of them in\n this case.\n \n So do not print an error on ids \u003d\u003d 0xff, just print an info message.\n Actually, we could return ENODEV even on the first CPU with ID 0xff, but\n ACPI spec does not forbid the 0xff value to be a processor ID. Given\n 0xff could be a correct one, we would break working systems if we\n returned ENODEV.\n \n Signed-off-by: Jiri Slaby \u003cjslaby@suse.cz\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 0a2ecbe5a9a8169375fe7fb6a0453aef94ece464\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Thu Aug 8 16:42:42 2019 +0300\n\n drm/i915: move I915_STATE_WARN() and _ON() to intel_display.h\n \n It\u0027s for display. Seems like a better fit. Reduce clutter in i915_drv.h.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/8fefcf69b6dcdb7b9c920aeded35c5f8928b9602.1565271681.git.jani.nikula@intel.com\n\ncommit e3adffe838ffe0f400cc69248ee605493d083dc2\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Thu Aug 8 16:42:41 2019 +0300\n\n drm/i915: move add_taint_for_CI() to i915_utils.h\n \n Seems like a better fit. Reduce clutter in i915_drv.h.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/a82d8c4e95496b3b4a9a251b655ea56a35c9d752.1565271681.git.jani.nikula@intel.com\n\ncommit 1c780b617cc04c58e5d87091eaba2dc987d1eafa\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Thu Aug 8 16:42:40 2019 +0300\n\n drm/i915: remove unused dev_priv-\u003eno_aux_handshake\n \n The last user of dev_priv-\u003eno_aux_handshake was removed in commit\n 3cf2efb1a7c6 (\"Revert \"drm/i915/dp: use VBT provided eDP params if\n available\"\"). Finally remove the leftovers.\n \n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/f1f0830bb43ddc6857d6a43e51c14b2f0c58c4e0.1565271681.git.jani.nikula@intel.com\n\ncommit d19bdb876bece27187d4ffbc272672e1239cd313\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Thu Aug 8 11:39:23 2019 +0200\n\n intel-hid: Disable button array during suspend-to-idle\n \n Notice that intel_button_array_enable() never disables the power\n button which is the only one needed to wake up the system from\n suspend-to-idle, so it can be safely called during suspend-to-idle\n as well as during \"regular\" system suspend, and rearrange the\n code in the driver\u0027s \"suspend\" and \"resume\" callbacks accordingly.\n \n While at it, use pm_suspend_no_platform() to check if the current\n suspend-resume cycle is suspend-to-idle, as that is the only\n case when the device should be enabled while suspended.\n \n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Acked-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit cd3e05836fc66539f2b387a0697ce1262ad121ce\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Fri May 24 12:14:21 2019 -0500\n\n drm/i915/kvmgt: Use struct_size() helper\n \n Make use of the struct_size() helper instead of an open-coded version\n in order to avoid any potential type mistakes, in particular in the\n context in which this code is being used.\n \n So, replace the following form:\n \n sizeof(*sparse) + (nr_areas * sizeof(*sparse-\u003eareas)\n \n with:\n \n struct_size(sparse, areas, sparse-\u003enr_areas)\n \n and so on...\n \n Also, notice that variable size is unnecessary, hence it is removed.\n \n This code was detected with the help of Coccinelle.\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n\ncommit 31eb845718398f9bc9f6fbe1aca675f4e6284392\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Thu Aug 8 11:39:17 2019 +0200\n\n intel-hid: intel-vbtn: Avoid leaking wakeup_mode set\n \n Both intel-hid and intel-vbtn set a wakeup_mode flag causing them\n to behave in a special way during system suspend and while suspended\n in their \"prepare\" PM callbacks and clear it in their \"resume\" PM\n callbacks. That may cause the wakeup_mode flag to remain set after\n a system suspend failure (if some other driver\u0027s \"suspend\" callback\n returns an error before the \"suspend\" callback of either intel-hid\n or intel-vbtn is invoked).\n \n After commit 10a08fd65ec1 (\"ACPI: PM: Set up EC GPE for system wakeup\n from drivers that need it\") that also affects the \"wakeup mask\" bit\n of the EC GPE, which may not be cleared after a failing system\n suspend.\n \n Fix this issue by adding \"complete\" PM callbacks to intel-hid and\n intel-vbtn to clear the wakeup_mode flag and the \"wakeup mask\" bit\n of the EC GPE if they have not been cleared earlier.\n \n Fixes: 10a08fd65ec1 (\"ACPI: PM: Set up EC GPE for system wakeup from drivers that need it\")\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Acked-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 625efea83a7c37d281c6a90526813a1366929d24\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Aug 7 09:15:02 2019 +0200\n\n pinctrl: rza1: Use devm_platform_ioremap_resource() helper\n \n Use the devm_platform_ioremap_resource() helper instead of open-coding\n the same operation.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit d8a0d6a3b7810fe3de723505c75f7f34ee6891df\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Mon Aug 5 10:58:46 2019 +0200\n\n drm/panel: simple: Support TI nspire panels\n \n This adds support for the TI nspire panels to the simple panel\n roster. This code is based on arch/arm/mach-nspire/clcd.c.\n This includes likely the first grayscale panel supported.\n \n These panels will be used with the PL11x DRM driver.\n \n Cc: Daniel Tang \u003cdt.tangr@gmail.com\u003e\n Cc: Fabian Vogt \u003cfabian@ritter-vogt.de\u003e\n Tested-by: Fabian Vogt \u003cfabian@ritter-vogt.de\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190805085847.25554-4-linus.walleij@linaro.org\n\ncommit 0c76620e574557aea993fc88b39dc40de394f144\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Tue Aug 6 15:54:37 2019 +0200\n\n drm/panel: simple: Add TI nspire panel bindings\n \n Add bindings for the TI NSPIRE simple display panels.\n \n Cc: devicetree@vger.kernel.org\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190806135437.7451-1-linus.walleij@linaro.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 43f847d0b9540dd1de4cfd626448fb4294422874\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Mon Aug 5 10:58:44 2019 +0200\n\n drm/pl111: Support grayscale\n \n Migrating the TI nspire calculators to use the PL111 driver for\n framebuffer requires grayscale support for the elder panel\n which uses 8bit grayscale only.\n \n DRM does not support 8bit grayscale framebuffers in memory,\n but by defining the bus format to be MEDIA_BUS_FMT_Y8_1X8 we\n can get the hardware to turn on a grayscaling feature and\n convert the RGB framebuffer to grayscale for us.\n \n Cc: Daniel Tang \u003cdt.tangr@gmail.com\u003e\n Cc: Fabian Vogt \u003cfabian@ritter-vogt.de\u003e\n Tested-by: Fabian Vogt \u003cfabian@ritter-vogt.de\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190805085847.25554-2-linus.walleij@linaro.org\n\ncommit 7794f486ed0b1fa8022dd0a27b9babf86a46d1cf\nAuthor: Alan Stern \u003cstern@rowland.harvard.edu\u003e\nDate: Wed Aug 7 10:29:50 2019 -0400\n\n usbfs: Add ioctls for runtime power management\n \n It has been requested that usbfs should implement runtime power\n management, instead of forcing the device to remain at full power as\n long as the device file is open. This patch introduces that new\n feature.\n \n It does so by adding three new usbfs ioctls:\n \n USBDEVFS_FORBID_SUSPEND: Prevents the device from going into\n runtime suspend (and causes a resume if the device is already\n suspended).\n \n USBDEVFS_ALLOW_SUSPEND: Allows the device to go into runtime\n suspend. Some time may elapse before the device actually is\n suspended, depending on things like the autosuspend delay.\n \n USBDEVFS_WAIT_FOR_RESUME: Blocks until the call is interrupted\n by a signal or at least one runtime resume has occurred since\n the most recent ALLOW_SUSPEND ioctl call (which may mean\n immediately, even if the device is currently suspended). In\n the latter case, the device is prevented from suspending again\n just as if FORBID_SUSPEND was called before the ioctl returns.\n \n For backward compatibility, when the device file is first opened\n runtime suspends are forbidden. The userspace program can then allow\n suspends whenever it wants, and either resume the device directly (by\n forbidding suspends again) or wait for a resume from some other source\n (such as a remote wakeup). URBs submitted to a suspended device will\n fail or will complete with an appropriate error code.\n \n This combination of ioctls is sufficient for user programs to have\n nearly the same degree of control over a device\u0027s runtime power\n behavior as kernel drivers do.\n \n Still lacking is documentation for the new ioctls. I intend to add it\n later, after the existing documentation for the usbfs userspace API is\n straightened out into a reasonable form.\n \n Suggested-by: Mayuresh Kulkarni \u003cmkulkarni@opensource.cirrus.com\u003e\n Signed-off-by: Alan Stern \u003cstern@rowland.harvard.edu\u003e\n \n Link: https://lore.kernel.org/r/Pine.LNX.4.44L0.1908071013220.1514-100000@iolanthe.rowland.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a21350feb9ae4d03062eda1fe344ec2a93d6bccc\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 6 16:45:01 2019 +0200\n\n USB: usbsevseg: convert to use dev_groups\n \n USB drivers now support the ability for the driver core to handle the\n creation and removal of device-specific sysfs files in a race-free\n manner. Take advantage of that by converting the driver to use this by\n moving the sysfs attributes into a group and assigning the dev_groups\n pointer to it.\n \n Link: https://lore.kernel.org/r/20190806144502.17792-12-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f9bbcbef0c50b766cf5931ab7d8f3c97a784257e\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 6 16:45:00 2019 +0200\n\n USB: trancevibrator: convert to use dev_groups\n \n USB drivers now support the ability for the driver core to handle the\n creation and removal of device-specific sysfs files in a race-free\n manner. Take advantage of that by converting the driver to use this by\n moving the sysfs attributes into a group and assigning the dev_groups\n pointer to it.\n \n Cc: Ding Xiang \u003cdingxiang@cmss.chinamobile.com\u003e\n Link: https://lore.kernel.org/r/20190806144502.17792-11-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit c5501d23e38df1f1e7afd373a5d924a18936930c\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 6 16:45:02 2019 +0200\n\n USB: usbip: convert to use dev_groups\n \n USB drivers now support the ability for the driver core to handle the\n creation and removal of device-specific sysfs files in a race-free\n manner. Take advantage of that by converting the driver to use this by\n moving the sysfs attributes into a group and assigning the dev_groups\n pointer to it.\n \n Cc: Valentina Manea \u003cvalentina.manea.m@gmail.com\u003e\n Cc: Shuah Khan \u003cshuah@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190806144502.17792-13-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit e605c30977bb5f48af6d14149a252b1619851f33\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 6 16:44:53 2019 +0200\n\n USB: atm: cxacru: convert to use dev_groups\n \n USB drivers now support the ability for the driver core to handle the\n creation and removal of device-specific sysfs files in a race-free\n manner. Take advantage of that by converting the driver to use this by\n moving the sysfs attributes into a group and assigning the dev_groups\n pointer to it.\n \n Link: https://lore.kernel.org/r/20190806144502.17792-4-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit ec935791ec52ae5098de8fcbf5ac484745b5bd79\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 6 16:44:54 2019 +0200\n\n USB: ueagle-atm: convert to use dev_groups\n \n USB drivers now support the ability for the driver core to handle the\n creation and removal of device-specific sysfs files in a race-free\n manner. Take advantage of that by converting the driver to use this by\n moving the sysfs attributes into a group and assigning the dev_groups\n pointer to it.\n \n Cc: Matthieu CASTET \u003ccastet.matthieu@free.fr\u003e\n Cc: Stanislaw Gruszka \u003cstf_xl@wp.pl\u003e\n Link: https://lore.kernel.org/r/20190806144502.17792-5-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 02c4d45d838957971ec5b7ee359380a0ccfdb963\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 6 16:44:55 2019 +0200\n\n USB: usblp: convert to use dev_groups\n \n USB drivers now support the ability for the driver core to handle the\n creation and removal of device-specific sysfs files in a race-free\n manner. Take advantage of that by converting the driver to use this by\n moving the sysfs attributes into a group and assigning the dev_groups\n pointer to it.\n \n Cc: Pete Zaitcev \u003czaitcev@redhat.com\u003e\n Link: https://lore.kernel.org/r/20190806144502.17792-6-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 04a0625e3b374a418139547431eadb0d8676bcba\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 6 16:44:56 2019 +0200\n\n USB: usbtmc: convert to use dev_groups\n \n USB drivers now support the ability for the driver core to handle the\n creation and removal of device-specific sysfs files in a race-free\n manner. Take advantage of that by converting the driver to use this by\n moving the sysfs attributes into a group and assigning the dev_groups\n pointer to it.\n \n Cc: Guido Kiener \u003cguido.kiener@rohde-schwarz.com\u003e\n Cc: Steve Bayless \u003csteve_bayless@keysight.com\u003e\n Link: https://lore.kernel.org/r/20190806144502.17792-7-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit c847a36521b08dbdaa4a23cb0795c0ee01241f8f\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 6 16:44:58 2019 +0200\n\n USB: cytherm: convert to use dev_groups\n \n USB drivers now support the ability for the driver core to handle the\n creation and removal of device-specific sysfs files in a race-free\n manner. Take advantage of that by converting the driver to use this by\n moving the sysfs attributes into a group and assigning the dev_groups\n pointer to it.\n \n Link: https://lore.kernel.org/r/20190806144502.17792-9-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 761ef1e4bc20e7b47e17be3022a2eae87de30560\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 6 16:44:57 2019 +0200\n\n USB: cypress_cy7c63: convert to use dev_groups\n \n USB drivers now support the ability for the driver core to handle the\n creation and removal of device-specific sysfs files in a race-free\n manner. Take advantage of that by converting the driver to use this by\n moving the sysfs attributes into a group and assigning the dev_groups\n pointer to it.\n \n Link: https://lore.kernel.org/r/20190806144502.17792-8-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 899ad6d68ec8751d07f537346fbdead53cd43bca\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 6 16:44:59 2019 +0200\n\n USB: lvstest: convert to use dev_groups\n \n USB drivers now support the ability for the driver core to handle the\n creation and removal of device-specific sysfs files in a race-free\n manner. Take advantage of that by converting the driver to use this by\n moving the sysfs attributes into a group and assigning the dev_groups\n pointer to it.\n \n Link: https://lore.kernel.org/r/20190806144502.17792-10-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 524f3ac1542c294a9337e4b82c6a22a9e7362290\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Mon Aug 5 21:36:31 2019 +0200\n\n USB: chipidea: convert platform driver to use dev_groups\n \n Platform drivers now have the option to have the platform core create\n and remove any needed sysfs attribute files. So take advantage of that\n and do not register \"by hand\" any sysfs files.\n \n Cc: Peter Chen \u003cPeter.Chen@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190805193636.25560-1-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 3e2cb866b2b1497a2246c8d222cd672694ac9f15\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Mon Aug 5 21:36:33 2019 +0200\n\n USB: phy: mv-usb: convert platform driver to use dev_groups\n \n Platform drivers now have the option to have the platform core create\n and remove any needed sysfs attribute files. So take advantage of that\n and do not register \"by hand\" any sysfs files.\n \n Cc: Felipe Balbi \u003cbalbi@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190805193636.25560-3-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f4d09e9f53ae2e45a9abb85227fc5f3731184da5\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Mon Aug 5 21:36:34 2019 +0200\n\n USB: phy: tahvo: convert platform driver to use dev_groups\n \n Platform drivers now have the option to have the platform core create\n and remove any needed sysfs attribute files. So take advantage of that\n and do not register \"by hand\" any sysfs files.\n \n Cc: Felipe Balbi \u003cbalbi@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190805193636.25560-4-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 891c1da6ceff10a500e56790e7272e4ac521dc26\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 6 09:32:34 2019 +0200\n\n USB: phy: twl6030: convert platform driver to use dev_groups\n \n Platform drivers now have the option to have the platform core create\n and remove any needed sysfs attribute files. So take advantage of that\n and do not register \"by hand\" any sysfs files.\n \n Cc: Felipe Balbi \u003cbalbi@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190806073235.25140-1-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f0104b0c0fd34f176b305e339e614023b6e27610\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Aug 5 14:14:26 2019 -0500\n\n USB: gadget: udc: s3c2410_udc: Mark expected switch fall-throughs\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warning (Building: tct_hammer_defconfig arm):\n \n drivers/usb/gadget/udc/s3c2410_udc.c:314:7: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n drivers/usb/gadget/udc/s3c2410_udc.c:418:7: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Link: https://lore.kernel.org/r/20190805191426.GA12414@embeddedor\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 938a76ed64860b361d73a6d80f78865ffdc7e0ea\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Aug 5 13:48:42 2019 -0500\n\n usb: gadget: atmel_usba_udc: Mark expected switch fall-through\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warning (Building: at91_dt_defconfig arm):\n \n drivers/usb/gadget/udc/atmel_usba_udc.c:329:13: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Link: https://lore.kernel.org/r/20190805184842.GA8627@embeddedor\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit fe4010667741df8c8d6bef1780ec34ebf43cf007\nAuthor: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nDate: Fri Aug 9 12:54:58 2019 +1000\n\n ALSA: hda: readl/writel need linux/io.h\n \n Fixes: 19abfefd4c76 (\"ALSA: hda: Direct MMIO accesses\")\n Signed-off-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit fcc32a21655e26d30c746b4828b33a5fd4ccfb11\nAuthor: Denis Efremov \u003cefremov@linux.com\u003e\nDate: Thu Aug 8 07:57:53 2019 +0300\n\n liquidio: Use pcie_flr() instead of reimplementing it\n \n octeon_mbox_process_cmd() directly writes the PCI_EXP_DEVCTL_BCR_FLR\n bit, which bypasses timing requirements imposed by the PCIe spec.\n This patch fixes the function to use the pcie_flr() interface instead.\n \n Signed-off-by: Denis Efremov \u003cefremov@linux.com\u003e\n Reviewed-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n Reviewed-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 32879f000120fead7cbd4a9aacbda95c84c6b287\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Wed Aug 7 21:38:22 2019 +0200\n\n r8169: allocate rx buffers using alloc_pages_node\n \n We allocate 16kb per rx buffer, so we can avoid some overhead by using\n alloc_pages_node directly instead of bothering kmalloc_node. Due to\n this change buffers are page-aligned now, therefore the alignment check\n can be removed.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Acked-by: Hayes Wang \u003chayeswang@realtek.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 018e5b458723d04cbeadbc3f0e06eba85798f427\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 7 21:10:55 2019 +0800\n\n fq_codel: remove set but not used variables \u0027prev_ecn_mark\u0027 and \u0027prev_drop_count\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n net/sched/sch_fq_codel.c: In function fq_codel_dequeue:\n net/sched/sch_fq_codel.c:288:23: warning: variable prev_ecn_mark set but not used [-Wunused-but-set-variable]\n net/sched/sch_fq_codel.c:288:6: warning: variable prev_drop_count set but not used [-Wunused-but-set-variable]\n \n They are not used since commit 77ddaff218fc (\"fq_codel: Kill\n useless per-flow dropped statistic\")\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4749e0e61241cc121de572520a39dab365b9ea1d\nAuthor: Thinh Nguyen \u003cThinh.Nguyen@synopsys.com\u003e\nDate: Thu Aug 8 16:39:42 2019 -0700\n\n usb: dwc3: Update soft-reset wait polling rate\n \n Starting from DWC_usb31 version 1.90a and later, the DCTL.CSFRST bit\n will not be cleared until after all the internal clocks are synchronized\n during soft-reset. This may take a little more than 50ms. Set the\n polling rate at 20ms instead.\n \n Signed-off-by: Thinh Nguyen \u003cthinhn@synopsys.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit b2a3974253d32374af556541141d7fdad8fe2ce0\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sun Jul 21 16:36:49 2019 +0900\n\n usb: dwc3: omap: squash include/linux/platform_data/dwc3-omap.h\n \n This enum is only used in drivers/usb/dwc3/dwc3-omap3.c\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit a6e5e6794a538534e797f422e6f31888ff5b6b64\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 22:17:04 2019 +0300\n\n usb: dwc3: Switch to use device_property_count_u32()\n \n Use use device_property_count_u32() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 8484aa0cb36d8fd751db77e46763719c78a03b22\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Wed Jul 24 21:18:49 2019 +0800\n\n usb: gadget: pch_udc: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 18425e5eb029eadbf82407d4fc224ced52fde986\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 22:16:39 2019 +0300\n\n usb: dwc2: Switch to use device_property_count_u32()\n \n Use use device_property_count_u32() directly, that makes code neater.\n \n Acked-by: Minas Harutyunyan \u003chminas@synopsys.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 5b76f6a0b0f868d8f4fae4e79a92e433a250a2eb\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 2 21:01:04 2019 +0800\n\n usb: dwc3: keystone: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 58dd0bad25543f95ffb5627d183eb4d01f38e633\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 2 21:06:16 2019 +0800\n\n usb: dwc3: omap: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit c6e4999cd930b8bd11dd8d4767e871b47f502845\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 2 21:04:08 2019 +0800\n\n usb: dwc3: meson-g12a: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Acked-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 37bea42fec575203c1b49d64322f717c29e82b90\nAuthor: John Keeping \u003cjohn@metanate.com\u003e\nDate: Mon Aug 5 17:01:21 2019 +0100\n\n usb: dwc2: gadget: Fix kill_all_requests race\n \n When a gadget is disabled, kill_all_requests() can be called\n simultaneously from both a user process via dwc2_hsotg_pullup() and from\n the interrupt handler if the hardware detects disconnection.\n \n Since we drop the lock in dwc2_hsotg_complete_request() in order to call\n the completion handler, this means that the list is modified\n concurrently and leads to an infinite loop in kill_all_requests().\n \n Replace the foreach loop with a while-not-empty loop in order to remove\n the danger of this concurrent modification.\n \n Note: I observed this with threadirqs, I\u0027m not sure if it can be\n triggered without threaded interrupts.\n \n Signed-off-by: John Keeping \u003cjohn@metanate.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit 27125cf8b5aee9f62dd108051203ccbd2e426b51\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Aug 5 13:48:42 2019 -0500\n\n usb: gadget: atmel_usba_udc: Mark expected switch fall-through\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warning (Building: at91_dt_defconfig arm):\n \n drivers/usb/gadget/udc/atmel_usba_udc.c:329:13: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit dccce749ac66aa01593273d86cf3c55542ce6fc0\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Aug 5 14:14:26 2019 -0500\n\n USB: gadget: udc: s3c2410_udc: Mark expected switch fall-throughs\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warning (Building: tct_hammer_defconfig arm):\n \n drivers/usb/gadget/udc/s3c2410_udc.c:314:7: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n drivers/usb/gadget/udc/s3c2410_udc.c:418:7: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Felipe Balbi \u003cfelipe.balbi@linux.intel.com\u003e\n\ncommit da382875c616856b234790206fc80d51f66b3bc7\nAuthor: Jiri Pirko \u003cjiri@mellanox.com\u003e\nDate: Wed Aug 7 13:42:31 2019 +0300\n\n mlxsw: spectrum: Extend to support Spectrum-3 ASIC\n \n Extend existing driver for Spectrum and Spectrum-2 ASICs\n to support Spectrum-3 ASIC as well.\n \n Signed-off-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Reviewed-by: Petr Machata \u003cpetrm@mellanox.com\u003e\n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ccfc639a94f25eb8639e8ffbecad2f6b60d22eb1\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Wed Aug 7 10:03:18 2019 +0200\n\n net: stmmac: selftests: Add a selftest for Flexible RX Parser\n \n Add a selftest for the Flexible RX Parser feature.\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d6e1c12cf9bc18759b3f02b08286a6f53248cb55\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Wed Aug 7 10:03:17 2019 +0200\n\n net: stmmac: Add Flexible RX Parser support in XGMAC\n \n XGMAC cores also support the Flexible RX Parser feature. Add the support\n for it in the XGMAC core.\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 56e58d6c8a5640eb708e85866e9d243d0357ee54\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Wed Aug 7 10:03:16 2019 +0200\n\n net: stmmac: Implement Safety Features in XGMAC core\n \n XGMAC also supports Safety Features. This patch implements the\n configuration and handling of this feature in XGMAC core.\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 74043f6b22c3201c63ae6ef37c4d3a96b0850e91\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Wed Aug 7 10:03:15 2019 +0200\n\n net: stmmac: selftests: Add test for VLAN and Double VLAN Filtering\n \n Add a selftest for VLAN and Double VLAN Filtering in stmmac.\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 3cd1cfcba26e2ebbadcd84ba118e785b8963e11b\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Wed Aug 7 10:03:14 2019 +0200\n\n net: stmmac: Implement VLAN Hash Filtering in XGMAC\n \n Implement the VLAN Hash Filtering feature in XGMAC core.\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1fbdad00055c59adb853935606468cf218c28dad\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Wed Aug 7 10:03:13 2019 +0200\n\n net: stmmac: selftests: Add RSS test\n \n Add a test for RSS in the stmmac selftests.\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 76067459c686c4fc6352613e5a6a54e4ffef2861\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Wed Aug 7 10:03:12 2019 +0200\n\n net: stmmac: Implement RSS and enable it in XGMAC core\n \n Implement the RSS functionality and add the corresponding callbacks in\n XGMAC core.\n \n Changes from v1:\n - Do not use magic constants (Jakub)\n - Use ethtool_rxfh_indir_default() (Jakub)\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7035aad875ba83331ab243173d2775f59880164f\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Wed Aug 7 10:03:11 2019 +0200\n\n net: stmmac: xgmac: Implement tx_queue_prio()\n \n Implement the TX Queue Priority callback in XGMAC core.\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 5656ac5542df5842d9e581546aab19cd83222b32\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Wed Aug 7 10:03:10 2019 +0200\n\n net: stmmac: xgmac: Implement set_mtl_tx_queue_weight()\n \n Implement the TX Queue Weight callback. In order for this to be active\n we also need to set ETS algorithm when configuring Queue.\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b6cdf09f51c20a25b7952773b61116452de66189\nAuthor: Jose Abreu \u003cJose.Abreu@synopsys.com\u003e\nDate: Wed Aug 7 10:03:09 2019 +0200\n\n net: stmmac: xgmac: Implement MMC counters\n \n Implement the MMC counters feature in XGMAC core.\n \n Signed-off-by: Jose Abreu \u003cjoabreu@synopsys.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ec9c7d19336ee98ecba8de80128aa405c45feebb\nAuthor: Daniel Jordan \u003cdaniel.m.jordan@oracle.com\u003e\nDate: Thu Aug 8 12:05:35 2019 -0400\n\n padata: initialize pd-\u003ecpu with effective cpumask\n \n Exercising CPU hotplug on a 5.2 kernel with recent padata fixes from\n cryptodev-2.6.git in an 8-CPU kvm guest...\n \n # modprobe tcrypt alg\u003d\"pcrypt(rfc4106(gcm(aes)))\" type\u003d3\n # echo 0 \u003e /sys/devices/system/cpu/cpu1/online\n # echo c \u003e /sys/kernel/pcrypt/pencrypt/parallel_cpumask\n # modprobe tcrypt mode\u003d215\n \n ...caused the following crash:\n \n BUG: kernel NULL pointer dereference, address: 0000000000000000\n #PF: supervisor read access in kernel mode\n #PF: error_code(0x0000) - not-present page\n PGD 0 P4D 0\n Oops: 0000 [#1] SMP PTI\n CPU: 2 PID: 134 Comm: kworker/2:2 Not tainted 5.2.0-padata-base+ #7\n Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-\u003csnip\u003e\n Workqueue: pencrypt padata_parallel_worker\n RIP: 0010:padata_reorder+0xcb/0x180\n ...\n Call Trace:\n padata_do_serial+0x57/0x60\n pcrypt_aead_enc+0x3a/0x50 [pcrypt]\n padata_parallel_worker+0x9b/0xe0\n process_one_work+0x1b5/0x3f0\n worker_thread+0x4a/0x3c0\n ...\n \n In padata_alloc_pd, pd-\u003ecpu is set using the user-supplied cpumask\n instead of the effective cpumask, and in this case cpumask_first picked\n an offline CPU.\n \n The offline CPU\u0027s reorder-\u003elist.next is NULL in padata_reorder because\n the list wasn\u0027t initialized in padata_init_pqueues, which only operates\n on CPUs in the effective mask.\n \n Fix by using the effective mask in padata_alloc_pd.\n \n Fixes: 6fc4dbcf0276 (\"padata: Replace delayed timer with immediate workqueue in padata_reorder\")\n Signed-off-by: Daniel Jordan \u003cdaniel.m.jordan@oracle.com\u003e\n Cc: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n Cc: Steffen Klassert \u003csteffen.klassert@secunet.com\u003e\n Cc: linux-crypto@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 3a646b6ed49cc64f0dc0a1bba24e62f7b557e7d7\nAuthor: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\nDate: Fri Aug 2 18:20:12 2019 -0500\n\n crypto: ccp - Remove unnecessary linux/pci.h include\n \n Remove unused includes of linux/pci.h.\n \n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Acked-by: Gary R Hook \u003cgary.hook@amd.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit f6b0b78a5c6c198b31967bfc2c7d17150980a0f6\nAuthor: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\nDate: Fri Aug 2 18:20:11 2019 -0500\n\n crypto: ccp - Include DMA declarations explicitly\n \n ccp-dev.h uses dma_direction, which is defined in linux/dma-direction.h.\n Include that explicitly instead of relying on it being included via\n linux/pci.h, since ccp-dev.h requires nothing else from linux/pci.h.\n \n Similarly, ccp-dmaengine.c uses dma_get_mask(), which is defined in\n linux/dma-mapping.h, so include that explicitly since it requires nothing\n else from linux/pci.h.\n \n A future patch will remove the includes of linux/pci.h where it is not\n needed.\n \n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Acked-by: Gary R Hook \u003cgary.hook@amd.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 2229c74079285c8f99754450e56d4d9814558e57\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 2 21:28:09 2019 +0800\n\n crypto: qcom-rng - use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit a54d83d42890e986018d0957fdd683df536505aa\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 2 21:28:08 2019 +0800\n\n crypto: qce - use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 473b4d995963b158f1713274c7992aed9240eb45\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 2 21:28:07 2019 +0800\n\n crypto: stm32 - use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 721744730ab06079df8197a700dc8929dadfcebc\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 2 21:28:06 2019 +0800\n\n crypto: rockchip - use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Reviewed-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit cf68528f9aaae3e5022bf9a4c71769ad48636c89\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 2 21:28:05 2019 +0800\n\n crypto: sunxi-ss - use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 9a8e0a513b2b07aad90182cce102032437795778\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 2 21:28:04 2019 +0800\n\n crypto: picoxcell - use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Acked-by: Jamie Iles \u003cjamie@jamieiles.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit b26120fdb9326052cc0bffe6a52c19a7145d3851\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 2 21:28:03 2019 +0800\n\n crypto: mediatek - use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit f8dab5575b76ea01b300e4b3a145955763ed61a9\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 2 21:28:02 2019 +0800\n\n crypto: inside-secure - use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Acked-by: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit f78c7123ff4cfb90572633d79d85e20f29f87112\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 2 21:28:01 2019 +0800\n\n crypto: img-hash - use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 6d1c0186f392e77b9e873490bed4d27b2fe580d7\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 2 21:28:00 2019 +0800\n\n crypto: exynos - use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Reviewed-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 17729e56f9b8f9f0a057ea16571cdea0457900f0\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 2 21:27:59 2019 +0800\n\n crypto: ccp - use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 0866ba23b7efcc6837d6b4231bf91b79647b81ea\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 2 21:27:58 2019 +0800\n\n crypto: artpec6 - use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 74bf81d0968c2262926dc2a690e671ebb768f2ec\nAuthor: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\nDate: Fri Aug 2 11:47:33 2019 +0300\n\n crypto: gcm - restrict assoclen for rfc4543\n \n Based on seqiv, IPsec ESP and rfc4543/rfc4106 the assoclen can be 16 or\n 20 bytes.\n \n From esp4/esp6, assoclen is sizeof IP Header. This includes spi, seq_no\n and extended seq_no, that is 8 or 12 bytes.\n In seqiv, to asscolen is added the IV size (8 bytes).\n Therefore, the assoclen, for rfc4543, should be restricted to 16 or 20\n bytes, as for rfc4106.\n \n Signed-off-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Reviewed-by: Horia Geanta \u003choria.geanta@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit ad3f0a93b639c342abbe8982cc34a3370169c464\nAuthor: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\nDate: Fri Aug 2 15:57:56 2019 +0800\n\n MAINTAINERS: add maintainer for HiSilicon QM and ZIP controller driver\n \n Add Zhou Wang as a maintainer for HiSilicon QM and ZIP controller driver.\n \n Signed-off-by: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\n Reviewed-by: John Garry \u003cjohn.garry@huawei.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 72c7a68d2ea34803e9c4ef948261ec6744fc72fc\nAuthor: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\nDate: Fri Aug 2 15:57:55 2019 +0800\n\n crypto: hisilicon - add debugfs for ZIP and QM\n \n HiSilicon ZIP engine driver uses debugfs to provide debug information,\n the usage can be found in /Documentation/ABI/testing/debugfs-hisi-zip.\n \n Signed-off-by: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 8201fdf49ff0950fa7a0c55a4aeb1ba3d747d404\nAuthor: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\nDate: Fri Aug 2 15:57:54 2019 +0800\n\n Documentation: Add debugfs doc for hisi_zip\n \n Add debugfs descriptions for HiSilicon ZIP and QM driver.\n \n Signed-off-by: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\n Reviewed-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 79e09f30eeba857b09832209bfc66bd689c58328\nAuthor: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\nDate: Fri Aug 2 15:57:53 2019 +0800\n\n crypto: hisilicon - add SRIOV support for ZIP\n \n HiSilicon ZIP engine supports PCI SRIOV. This patch enable this feature.\n User can enable VFs and pass through them to VM, same ZIP driver can work\n in VM to provide ZLIB and GZIP algorithm by crypto acomp interface.\n \n Signed-off-by: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 62c455ca853e3e352e465d66a6cc39f1f88caa60\nAuthor: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\nDate: Fri Aug 2 15:57:52 2019 +0800\n\n crypto: hisilicon - add HiSilicon ZIP accelerator support\n \n The HiSilicon ZIP accelerator implements the zlib and gzip algorithm. It\n uses Hisilicon QM as the interface to the CPU.\n \n This patch provides PCIe driver to the accelerator and registers it to\n crypto acomp interface. It also uses sgl as data input/output interface.\n \n Signed-off-by: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\n Signed-off-by: Shiju Jose \u003cshiju.jose@huawei.com\u003e\n Signed-off-by: Kenneth Lee \u003cliguozhu@hisilicon.com\u003e\n Signed-off-by: Hao Fang \u003cfanghao11@huawei.com\u003e\n Reviewed-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n Reviewed-by: John Garry \u003cjohn.garry@huawei.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit dfed0098ab91f647b5720ab6f1e03b5b55139408\nAuthor: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\nDate: Fri Aug 2 15:57:51 2019 +0800\n\n crypto: hisilicon - add hardware SGL support\n \n HiSilicon accelerators in Hip08 use same hardware scatterlist for data format.\n We support it in this module.\n \n Specific accelerator drivers can use hisi_acc_create_sgl_pool to allocate\n hardware SGLs ahead. Then use hisi_acc_sg_buf_map_to_hw_sgl to get one\n hardware SGL and pass related information to hardware SGL.\n \n The DMA address of mapped hardware SGL can be passed to SGL src/dst field\n in QM SQE.\n \n Signed-off-by: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 263c9959c9376ec0217d6adc61222a53469eed3c\nAuthor: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\nDate: Fri Aug 2 15:57:50 2019 +0800\n\n crypto: hisilicon - add queue management driver for HiSilicon QM module\n \n QM is a general IP used by HiSilicon accelerators. It provides a general\n PCIe interface for the CPU and the accelerator to share a group of queues.\n \n A QM integrated in an accelerator provides queue management service.\n Queues can be assigned to PF and VFs, and queues can be controlled by\n unified mailboxes and doorbells. Specific task request are descripted by\n specific description buffer, which will be controlled and pass to related\n accelerator IP by QM.\n \n This patch adds a QM driver used by the accelerator driver to access\n the QM hardware.\n \n Signed-off-by: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\n Signed-off-by: Kenneth Lee \u003cliguozhu@hisilicon.com\u003e\n Signed-off-by: Shiju Jose \u003cshiju.jose@huawei.com\u003e\n Signed-off-by: Hao Fang \u003cfanghao11@huawei.com\u003e\n Reviewed-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n Reviewed-by: John Garry \u003cjohn.garry@huawei.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit d13dfae3cedd60072d448b7e382c86143a29cedc\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Thu Aug 1 13:13:51 2019 +0200\n\n crypto: engine - Reduce default RT priority\n \n The crypto engine initializes its kworker thread to FIFO-99 (when\n requesting RT priority), reduce this to FIFO-50.\n \n FIFO-99 is the very highest priority available to SCHED_FIFO and\n it not a suitable default; it would indicate the crypto work is the\n most important work on the machine.\n \n Cc: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: linux-crypto@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 0435d47e2627b76b6dfb59e2e95ffed7232bd0ed\nAuthor: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\nDate: Wed Jul 31 16:08:15 2019 +0300\n\n crypto: caam - change return value in case CAAM has no MDHA\n \n To be consistent with other CAAM modules, caamhash should return 0\n instead of -ENODEV in case CAAM has no MDHA.\n \n Based on commit 1b46c90c8e00 (\"crypto: caam - convert top level drivers to libraries\")\n the value returned by entry point is never checked and\n the exit point is always executed.\n \n Signed-off-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Reviewed-by: Horia Geanta \u003choria.geanta@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 58068cfc810c3b5cc8c25b0fab73c02625aeadf1\nAuthor: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\nDate: Wed Jul 31 16:08:14 2019 +0300\n\n crypto: caam - unregister algorithm only if the registration succeeded\n \n To know if a registration succeeded added a new struct,\n caam_akcipher_alg, that keeps, also, the registration status.\n This status is updated in caam_pkc_init and verified in\n caam_pkc_exit to unregister an algorithm.\n \n Fixes: 1b46c90c8e00 (\"crypto: caam - convert top level drivers to libraries\")\n Signed-off-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Reviewed-by: Horia Geanta \u003choria.geanta@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 4e3a61c55b8c5db9bfaaf1fc1f448f8f874f66eb\nAuthor: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\nDate: Wed Jul 31 16:08:13 2019 +0300\n\n crypto: caam - execute module exit point only if necessary\n \n Commit 1b46c90c8e00 (\"crypto: caam - convert top level drivers to libraries\")\n changed entry and exit points behavior for caamalg,\n caamalg_qi, caamalg_qi2, caamhash, caampkc, caamrng.\n \n For example, previously caam_pkc_init() and caam_pkc_exit() were\n module entry/exit points. This means that if an error would happen\n in caam_pkc_init(), then caam_pkc_exit() wouldn\u0027t have been called.\n After the mentioned commit, caam_pkc_init() and caam_pkc_exit()\n are manually called - from jr.c. caam_pkc_exit() is called\n unconditionally, even if caam_pkc_init() failed.\n \n Added a global variable to keep the status of the algorithm\n registration and free of resources.\n The exit point of caampkc/caamrng module is executed only if the\n registration was successful. Therefore we avoid double free of\n resources in case the algorithm registration failed.\n \n Fixes: 1b46c90c8e00 (\"crypto: caam - convert top level drivers to libraries\")\n Signed-off-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Reviewed-by: Horia Geanta \u003choria.geanta@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit c59a1d41672a89b5cac49db1a472ff889e35a2d2\nAuthor: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\nDate: Wed Jul 31 16:08:12 2019 +0300\n\n crypto: caam - free resources in case caam_rng registration failed\n \n Check the return value of the hardware registration for caam_rng and free\n resources in case of failure.\n \n Fixes: e24f7c9e87d4 (\"crypto: caam - hwrng support\")\n Signed-off-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Reviewed-by: Horia Geanta \u003choria.geanta@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit e9b4913a5f944b23d6109c44b6f3fc6e092e30ce\nAuthor: Horia Geantă \u003choria.geanta@nxp.com\u003e\nDate: Wed Jul 31 16:08:11 2019 +0300\n\n crypto: caam - fix MDHA key derivation for certain user key lengths\n \n Fuzz testing uncovered an issue when |user key| \u003e |derived key|.\n Derived key generation has to be fixed in two cases:\n \n 1. Era \u003e\u003d 6 (DKP is available)\n DKP cannot be used with immediate input key if |user key| \u003e |derived key|,\n since the resulting descriptor (after DKP execution) would be invalid -\n having a few bytes from user key left in descriptor buffer\n as incorrect opcodes.\n \n Fix DKP usage both in standalone hmac and in authenc algorithms.\n For authenc the logic is simplified, by always storing both virtual\n and dma key addresses.\n \n 2. Era \u003c 6\n The same case (|user key| \u003e |derived key|) fails when DKP\n is not available.\n Make sure gen_split_key() dma maps max(|user key|, |derived key|),\n since this is an in-place (bidirectional) operation.\n \n Signed-off-by: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Reviewed-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit a2fb864c042b00debc7696d4459d8058ec7c8013\nAuthor: Horia Geantă \u003choria.geanta@nxp.com\u003e\nDate: Wed Jul 31 16:08:10 2019 +0300\n\n crypto: caam - keep both virtual and dma key addresses\n \n Update alginfo struct to keep both virtual and dma key addresses,\n so that descriptors have them at hand.\n One example where this is needed is in the xcbc(aes) shared descriptors,\n which are updated in current patch.\n Another example is the upcoming fix for DKP.\n \n Signed-off-by: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Reviewed-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit d28a43edf3caff971b58d75c83ad00487152e448\nAuthor: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\nDate: Wed Jul 31 16:08:09 2019 +0300\n\n crypto: caam - update rfc4106 sh desc to support zero length input\n \n Update share descriptor for rfc4106 to skip instructions in case\n cryptlen is zero. If no instructions are jumped the DECO hangs and a\n timeout error is thrown.\n \n Signed-off-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Reviewed-by: Horia Geanta \u003choria.geanta@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 31bb2f0da1b5099732337d9ecef792ab67970ddf\nAuthor: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\nDate: Wed Jul 31 16:08:08 2019 +0300\n\n crypto: caam - check zero-length input\n \n Check zero-length input, for skcipher algorithm, to solve the extra\n tests. This is a valid operation, therefore the API will return no error.\n \n Signed-off-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Reviewed-by: Horia Geanta \u003choria.geanta@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit fcd23ed57c4cccda17b0be736cdfbd7970b842c8\nAuthor: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\nDate: Wed Jul 31 16:08:07 2019 +0300\n\n crypto: caam - check assoclen\n \n Check assoclen to solve the extra tests that expect -EINVAL to be\n returned when the associated data size is not valid.\n \n Validated assoclen for RFC4106 and RFC4543 which expects an assoclen\n of 16 or 20.\n Based on seqiv, IPsec ESP and RFC4543/RFC4106 the assoclen is sizeof IP\n Header (spi, seq_no, extended seq_no) and IV len. This can be 16 or 20\n bytes.\n \n Signed-off-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Reviewed-by: Horia Geanta \u003choria.geanta@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 68a51394f37ae8c6eb51209efc406355a9861874\nAuthor: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\nDate: Wed Jul 31 16:08:06 2019 +0300\n\n crypto: caam - check authsize\n \n Check authsize to solve the extra tests that expect -EINVAL to be\n returned when the authentication tag size is not valid.\n \n Validated authsize for GCM, RFC4106 and RFC4543.\n \n Signed-off-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Reviewed-by: Horia Geanta \u003choria.geanta@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 836d8f43c5e529935bb3140edc710c23ffd1d54a\nAuthor: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\nDate: Wed Jul 31 16:08:05 2019 +0300\n\n crypto: caam - check key length\n \n Check key length to solve the extra tests that expect -EINVAL to be\n returned when the key size is not valid.\n \n Validated AES keylen for skcipher, ahash and aead.\n \n Signed-off-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Reviewed-by: Horia Geanta \u003choria.geanta@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 1ccb39eb5a9822543273c35f5114a9dc101e7315\nAuthor: Horia Geantă \u003choria.geanta@nxp.com\u003e\nDate: Wed Jul 31 16:08:04 2019 +0300\n\n crypto: caam - update IV only when crypto operation succeeds\n \n skcipher encryption might fail and in some cases, like (invalid) input\n length smaller then block size, updating the IV would lead to a useless\n IV copy in case hardware issued an error.\n \n Signed-off-by: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Signed-off-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 1984aaeec372fbfb597883074253d290cbd543d4\nAuthor: Horia Geantă \u003choria.geanta@nxp.com\u003e\nDate: Wed Jul 31 16:08:03 2019 +0300\n\n crypto: caam - fix return code in completion callbacks\n \n Modify drive to provide a valid errno (and not the HW error ID)\n to the user, via completion callbacks.\n \n A \"valid errno\" is currently not explicitly mentioned in the docs,\n however the error code is expected to match the one returned by the\n generic SW implementation.\n \n Note: in most error cases caam/qi and caam/qi2 returned -EIO; align all\n caam drivers to return -EINVAL.\n \n While here, ratelimit prints triggered by fuzz testing, such that\n console is not flooded.\n \n Signed-off-by: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Signed-off-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 51fab3d73054ca5b06b26e20edac0486b052c6f4\nAuthor: Horia Geantă \u003choria.geanta@nxp.com\u003e\nDate: Wed Jul 31 16:08:02 2019 +0300\n\n crypto: caam/qi - fix error handling in ERN handler\n \n ERN handler calls the caam/qi frontend \"done\" callback with a status\n of -EIO. This is incorrect, since the callback expects a status value\n meaningful for the crypto engine - hence the cryptic messages\n like the one below:\n platform caam_qi: 15: unknown error source\n \n Fix this by providing the callback with:\n -the status returned by the crypto engine (fd[status]) in case\n it contains an error, OR\n -a QI \"No error\" code otherwise; this will trigger the message:\n platform caam_qi: 50000000: Queue Manager Interface: No error\n which is fine, since QMan driver provides details about the cause of\n failure\n \n Cc: \u003cstable@vger.kernel.org\u003e # v5.1+\n Fixes: 67c2315def06 (\"crypto: caam - add Queue Interface (QI) backend support\")\n Signed-off-by: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Reviewed-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit bc67d04e75260942fb534fb91673103dcad7ca96\nAuthor: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\nDate: Wed Jul 31 16:05:55 2019 +0300\n\n crypto: aes - helper function to validate key length for AES algorithms\n \n Add inline helper function to check key length for AES algorithms.\n The key can be 128, 192 or 256 bits size.\n This function is used in the generic aes implementation.\n \n Signed-off-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 65526f638efc6435b9658d45a3ddd07e6fdbbb7d\nAuthor: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\nDate: Wed Jul 31 16:05:54 2019 +0300\n\n crypto: gcm - helper functions for assoclen/authsize check\n \n Added inline helper functions to check authsize and assoclen for\n gcm, rfc4106 and rfc4543.\n These are used in the generic implementation of gcm, rfc4106 and\n rfc4543.\n \n Signed-off-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 6c9081a3915dc0782a8f1424343b794f2cf53d9c\nAuthor: John Rutherford \u003cjohn.rutherford@dektech.com.au\u003e\nDate: Wed Aug 7 12:52:29 2019 +1000\n\n tipc: add loopback device tracking\n \n Since node internal messages are passed directly to the socket, it is not\n possible to observe those messages via tcpdump or wireshark.\n \n We now remedy this by making it possible to clone such messages and send\n the clones to the loopback interface. The clones are dropped at reception\n and have no functional role except making the traffic visible.\n \n The feature is enabled if network taps are active for the loopback device.\n pcap filtering restrictions require the messages to be presented to the\n receiving side of the loopback device.\n \n v3 - Function dev_nit_active used to check for network taps.\n - Procedure netif_rx_ni used to send cloned messages to loopback device.\n \n Signed-off-by: John Rutherford \u003cjohn.rutherford@dektech.com.au\u003e\n Acked-by: Jon Maloy \u003cjon.maloy@ericsson.com\u003e\n Acked-by: Ying Xue \u003cying.xue@windriver.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1bfaac7c53f5cfe9c44d2093263fa50630a81b63\nAuthor: Pascal van Leeuwen \u003cpascalvanl@gmail.com\u003e\nDate: Wed Jul 31 08:10:54 2019 +0200\n\n crypto: inside-secure - Remove redundant DES ECB \u0026 CBC keysize check\n \n This patch removes a DES key size check that is redundant as it is already\n performed by the crypto API itself due to min_keysize \u003d max_keysize.\n \n Signed-off-by: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Acked-by: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 114e2ccd9eff5859760cfa3bcbb707c3eda63fc3\nAuthor: Gary R Hook \u003cgary.hook@amd.com\u003e\nDate: Wed Jul 31 00:03:25 2019 +0000\n\n crypto: ccp - Clean up and exit correctly on allocation failure\n \n Return and fail driver initialization if a DMA pool or coherent memory\n can\u0027t be allocated. Be sure to clean up allocated memory.\n \n Fixes: 4b394a232df7 (\"crypto: ccp - Let a v5 CCP provide the same function as v3\")\n \n Signed-off-by: Gary R Hook \u003cgary.hook@amd.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 514838e920d06f5c2ba56e9656107dfcbbb90219\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:05 2019 -0700\n\n crypto: drivers - Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: \u003clinux-crypto@vger.kernel.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit f26882a3475eb7b2c35fd3b1e291924161b1327d\nAuthor: Pascal van Leeuwen \u003cpascalvanl@gmail.com\u003e\nDate: Tue Jul 30 15:27:11 2019 +0200\n\n crypto: inside-secure - Use defines instead of some constants (cosmetic)\n \n This patch replaces some hard constants regarding key, IV and nonce sizes\n with appropriate defines from the crypto header files.\n \n Signed-off-by: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Acked-by: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 48f89d2a2920166c35b1c0b69917dbb0390ebec7\nAuthor: Horia Geantă \u003choria.geanta@nxp.com\u003e\nDate: Tue Jul 30 08:48:33 2019 +0300\n\n crypto: caam - fix concurrency issue in givencrypt descriptor\n \n IV transfer from ofifo to class2 (set up at [29][30]) is not guaranteed\n to be scheduled before the data transfer from ofifo to external memory\n (set up at [38]:\n \n [29] 10FA0004 ld: ind-nfifo (len\u003d4) imm\n [30] 81F00010 \u003cnfifo_entry: ofifo-\u003eclass2 type\u003dmsg len\u003d16\u003e\n [31] 14820004 ld: ccb2-datasz len\u003d4 offs\u003d0 imm\n [32] 00000010 data:0x00000010\n [33] 8210010D operation: cls1-op aes cbc init-final enc\n [34] A8080B04 math: (seqin + math0)-\u003evseqout len\u003d4\n [35] 28000010 seqfifold: skip len\u003d16\n [36] A8080A04 math: (seqin + math0)-\u003evseqin len\u003d4\n [37] 2F1E0000 seqfifold: both msg1-\u003e2-last2-last1 len\u003dvseqinsz\n [38] 69300000 seqfifostr: msg len\u003dvseqoutsz\n [39] 5C20000C seqstr: ccb2 ctx len\u003d12 offs\u003d0\n \n If ofifo -\u003e external memory transfer happens first, DECO will hang\n (issuing a Watchdog Timeout error, if WDOG is enabled) waiting for\n data availability in ofifo for the ofifo -\u003e c2 ififo transfer.\n \n Make sure IV transfer happens first by waiting for all CAAM internal\n transfers to end before starting payload transfer.\n \n New descriptor with jump command inserted at [37]:\n \n [..]\n [36] A8080A04 math: (seqin + math0)-\u003evseqin len\u003d4\n [37] A1000401 jump: jsl1 all-match[!nfifopend] offset\u003d[01] local-\u003e[38]\n [38] 2F1E0000 seqfifold: both msg1-\u003e2-last2-last1 len\u003dvseqinsz\n [39] 69300000 seqfifostr: msg len\u003dvseqoutsz\n [40] 5C20000C seqstr: ccb2 ctx len\u003d12 offs\u003d0\n \n [Note: the issue is present in the descriptor from the very beginning\n (cf. Fixes tag). However I\u0027ve marked it v4.19+ since it\u0027s the oldest\n maintained kernel that the patch applies clean against.]\n \n Cc: \u003cstable@vger.kernel.org\u003e # v4.19+\n Fixes: 1acebad3d8db8 (\"crypto: caam - faster aead implementation\")\n Signed-off-by: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 15b2455ab4e6434a53c639bd2a168cd2a3e1eb47\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Jul 29 17:38:19 2019 -0500\n\n crypto: ux500/crypt - Mark expected switch fall-throughs\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warning (Building: arm):\n \n drivers/crypto/ux500/cryp/cryp.c: In function ‘cryp_save_device_context’:\n drivers/crypto/ux500/cryp/cryp.c:316:16: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n ctx-\u003ekey_4_r \u003d readl_relaxed(\u0026src_reg-\u003ekey_4_r);\n drivers/crypto/ux500/cryp/cryp.c:318:2: note: here\n case CRYP_KEY_SIZE_192:\n ^~~~\n drivers/crypto/ux500/cryp/cryp.c:320:16: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n ctx-\u003ekey_3_r \u003d readl_relaxed(\u0026src_reg-\u003ekey_3_r);\n drivers/crypto/ux500/cryp/cryp.c:322:2: note: here\n case CRYP_KEY_SIZE_128:\n ^~~~\n drivers/crypto/ux500/cryp/cryp.c:324:16: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n ctx-\u003ekey_2_r \u003d readl_relaxed(\u0026src_reg-\u003ekey_2_r);\n drivers/crypto/ux500/cryp/cryp.c:326:2: note: here\n default:\n ^~~~~~~\n In file included from ./include/linux/io.h:13:0,\n from drivers/crypto/ux500/cryp/cryp_p.h:14,\n from drivers/crypto/ux500/cryp/cryp.c:15:\n drivers/crypto/ux500/cryp/cryp.c: In function ‘cryp_restore_device_context’:\n ./arch/arm/include/asm/io.h:92:22: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n #define __raw_writel __raw_writel\n ^\n ./arch/arm/include/asm/io.h:299:29: note: in expansion of macro ‘__raw_writel’\n #define writel_relaxed(v,c) __raw_writel((__force u32) cpu_to_le32(v),c)\n ^~~~~~~~~~~~\n drivers/crypto/ux500/cryp/cryp.c:363:3: note: in expansion of macro ‘writel_relaxed’\n writel_relaxed(ctx-\u003ekey_4_r, \u0026reg-\u003ekey_4_r);\n ^~~~~~~~~~~~~~\n drivers/crypto/ux500/cryp/cryp.c:365:2: note: here\n case CRYP_KEY_SIZE_192:\n ^~~~\n In file included from ./include/linux/io.h:13:0,\n from drivers/crypto/ux500/cryp/cryp_p.h:14,\n from drivers/crypto/ux500/cryp/cryp.c:15:\n ./arch/arm/include/asm/io.h:92:22: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n #define __raw_writel __raw_writel\n ^\n ./arch/arm/include/asm/io.h:299:29: note: in expansion of macro ‘__raw_writel’\n #define writel_relaxed(v,c) __raw_writel((__force u32) cpu_to_le32(v),c)\n ^~~~~~~~~~~~\n drivers/crypto/ux500/cryp/cryp.c:367:3: note: in expansion of macro ‘writel_relaxed’\n writel_relaxed(ctx-\u003ekey_3_r, \u0026reg-\u003ekey_3_r);\n ^~~~~~~~~~~~~~\n drivers/crypto/ux500/cryp/cryp.c:369:2: note: here\n case CRYP_KEY_SIZE_128:\n ^~~~\n In file included from ./include/linux/io.h:13:0,\n from drivers/crypto/ux500/cryp/cryp_p.h:14,\n from drivers/crypto/ux500/cryp/cryp.c:15:\n ./arch/arm/include/asm/io.h:92:22: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n #define __raw_writel __raw_writel\n ^\n ./arch/arm/include/asm/io.h:299:29: note: in expansion of macro ‘__raw_writel’\n #define writel_relaxed(v,c) __raw_writel((__force u32) cpu_to_le32(v),c)\n ^~~~~~~~~~~~\n drivers/crypto/ux500/cryp/cryp.c:371:3: note: in expansion of macro ‘writel_relaxed’\n writel_relaxed(ctx-\u003ekey_2_r, \u0026reg-\u003ekey_2_r);\n ^~~~~~~~~~~~~~\n drivers/crypto/ux500/cryp/cryp.c:373:2: note: here\n default:\n ^~~~~~~\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit f6ebfd7826a8b69ad6691695a97409bd35d146a9\nAuthor: Gary R Hook \u003cgary.hook@amd.com\u003e\nDate: Mon Jul 29 12:56:08 2019 +0000\n\n crypto: ccp - Log an error message when ccp-crypto fails to load\n \n If there are no CCP devices on the system, ccp-crypto will not load.\n Write a message to the system log clarifying the reason for the failure\n of the modprobe operation\n \n Signed-off-by: Gary R Hook \u003cgary.hook@amd.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit e88b27c8eaa8f3126791778803dd2fdd81828f5c\nAuthor: Gilad Ben-Yossef \u003cgilad@benyossef.com\u003e\nDate: Mon Jul 29 13:40:19 2019 +0300\n\n crypto: ccree - use std api sg_zero_buffer\n \n Replace internal cc_zero_sgl() with kernel API of the same function\n sg_zero_buffer().\n \n Signed-off-by: Gilad Ben-Yossef \u003cgilad@benyossef.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 7a4be6c113c1f721818d1e3722a9015fe393295c\nAuthor: Gilad Ben-Yossef \u003cgilad@benyossef.com\u003e\nDate: Mon Jul 29 13:40:18 2019 +0300\n\n crypto: ccree - use the full crypt length value\n \n In case of AEAD decryption verifcation error we were using the\n wrong value to zero out the plaintext buffer leaving the end of\n the buffer with the false plaintext.\n \n Signed-off-by: Gilad Ben-Yossef \u003cgilad@benyossef.com\u003e\n Fixes: ff27e85a85bb (\"crypto: ccree - add AEAD support\")\n CC: stable@vger.kernel.org # v4.17+\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit f0e6f41669d9e07f45b472e4de33d7c233a847bd\nAuthor: Jens Axboe \u003caxboe@kernel.dk\u003e\nDate: Thu Aug 8 22:11:28 2019 -0600\n\n lightnvm: remove unused \u0027geo\u0027 variable\n \n A previous commit correctly removed set-but-not-read variables, but\n this left two new variables now unused. Kill them.\n \n Fixes: ba6f7da99aaf (\"lightnvm: remove set but not used variables \u0027data_len\u0027 and \u0027rq_len\u0027\")\n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit fdbe4eeeb1aac219b14f10c0ed31ae5d1123e9b8\nAuthor: Alessio Balsini \u003cbalsini@android.com\u003e\nDate: Wed Aug 7 01:48:28 2019 +0100\n\n loop: Add LOOP_SET_DIRECT_IO to compat ioctl\n \n Enabling Direct I/O with loop devices helps reducing memory usage by\n avoiding double caching. 32 bit applications running on 64 bits systems\n are currently not able to request direct I/O because is missing from the\n lo_compat_ioctl.\n \n This patch fixes the compatibility issue mentioned above by exporting\n LOOP_SET_DIRECT_IO as additional lo_compat_ioctl() entry.\n The input argument for this ioctl is a single long converted to a 1-bit\n boolean, so compatibility is preserved.\n \n Cc: Jens Axboe \u003caxboe@kernel.dk\u003e\n Signed-off-by: Alessio Balsini \u003cbalsini@android.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 2a9af0273c1cd6647df4f6475b45d823494bf13a\nAuthor: Wesley Terpstra \u003cwesley@sifive.com\u003e\nDate: Thu Jul 25 14:28:07 2019 -0700\n\n PCI/MSI: Enable PCI_MSI_IRQ_DOMAIN support for RISC-V\n \n Add RISC-V as an arch that supports PCI_MSI_IRQ_DOMAIN. The related change\n to generate asm/msi.h is 251a44888183 (\"riscv: include generic support for\n MSI irqdomains\").\n \n Link: https://lore.kernel.org/r/alpine.DEB.2.21.9999.1907251426450.32766@viisi.sifive.com\n Signed-off-by: Wesley Terpstra \u003cwesley@sifive.com\u003e\n [paul.walmsley@sifive.com: wrote patch description; split this\n patch from the arch/riscv patch]\n Signed-off-by: Paul Walmsley \u003cpaul.walmsley@sifive.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Bin Meng \u003cbmeng.cn@gmail.com\u003e\n\ncommit 9a32669fecfb484a1f78fe48d0e42a5efccb0675\nAuthor: wenxu \u003cwenxu@ucloud.cn\u003e\nDate: Wed Aug 7 09:13:54 2019 +0800\n\n netfilter: nf_tables_offload: support indr block call\n \n nftable support indr-block call. It makes nftable an offload vlan\n and tunnel device.\n \n nft add table netdev firewall\n nft add chain netdev firewall aclout { type filter hook ingress offload device mlx_pf0vf0 priority - 300 \\; }\n nft add rule netdev firewall aclout ip daddr 10.0.0.1 fwd to vlan0\n nft add chain netdev firewall aclin { type filter hook ingress device vlan0 priority - 300 \\; }\n nft add rule netdev firewall aclin ip daddr 10.0.0.7 fwd to mlx_pf0vf0\n \n Signed-off-by: wenxu \u003cwenxu@ucloud.cn\u003e\n Acked-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1150ab0f1b333ca310431dac65d8fa403b8471da\nAuthor: wenxu \u003cwenxu@ucloud.cn\u003e\nDate: Wed Aug 7 09:13:53 2019 +0800\n\n flow_offload: support get multi-subsystem block\n \n It provide a callback list to find the blocks of tc\n and nft subsystems\n \n Signed-off-by: wenxu \u003cwenxu@ucloud.cn\u003e\n Acked-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4e481908c51bf02457aecdedc2d80e1be22e0146\nAuthor: wenxu \u003cwenxu@ucloud.cn\u003e\nDate: Wed Aug 7 09:13:52 2019 +0800\n\n flow_offload: move tc indirect block to flow offload\n \n move tc indirect block to flow_offload and rename\n it to flow indirect block.The nf_tables can use the\n indr block architecture.\n \n Signed-off-by: wenxu \u003cwenxu@ucloud.cn\u003e\n Acked-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit e4da91021153b78b9f2972212610bc71263925f4\nAuthor: wenxu \u003cwenxu@ucloud.cn\u003e\nDate: Wed Aug 7 09:13:51 2019 +0800\n\n cls_api: add flow_indr_block_call function\n \n This patch make indr_block_call don\u0027t access struct tc_indr_block_cb\n and tc_indr_block_dev directly\n \n Signed-off-by: wenxu \u003cwenxu@ucloud.cn\u003e\n Acked-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f8436988574e50fe69f6f193ee76c3185b17a5c7\nAuthor: wenxu \u003cwenxu@ucloud.cn\u003e\nDate: Wed Aug 7 09:13:50 2019 +0800\n\n cls_api: remove the tcf_block cache\n \n Remove the tcf_block in the tc_indr_block_dev for muti-subsystem\n support.\n \n Signed-off-by: wenxu \u003cwenxu@ucloud.cn\u003e\n Acked-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 242453c227d14751fed0a8809a58f1bf3c7d837a\nAuthor: wenxu \u003cwenxu@ucloud.cn\u003e\nDate: Wed Aug 7 09:13:49 2019 +0800\n\n cls_api: modify the tc_indr_block_ing_cmd parameters.\n \n This patch make tc_indr_block_ing_cmd can\u0027t access struct\n tc_indr_block_dev and tc_indr_block_cb.\n \n Signed-off-by: wenxu \u003cwenxu@ucloud.cn\u003e\n Acked-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 323ebb61e32b478e2432c5a3cbf9e2ca678a9609\nAuthor: Edward Cree \u003cecree@solarflare.com\u003e\nDate: Tue Aug 6 14:53:55 2019 +0100\n\n net: use listified RX for handling GRO_NORMAL skbs\n \n When GRO decides not to coalesce a packet, in napi_frags_finish(), instead\n of passing it to the stack immediately, place it on a list in the napi\n struct. Then, at flush time (napi_complete_done(), napi_poll(), or\n napi_busy_loop()), call netif_receive_skb_list_internal() on the list.\n We\u0027d like to do that in napi_gro_flush(), but it\u0027s not called if\n !napi-\u003egro_bitmask, so we have to do it in the callers instead. (There are\n a handful of drivers that call napi_gro_flush() themselves, but it\u0027s not\n clear why, or whether this will affect them.)\n Because a full 64 packets is an inefficiently large batch, also consume the\n list whenever it exceeds gro_normal_batch, a new net/core sysctl that\n defaults to 8.\n \n Signed-off-by: Edward Cree \u003cecree@solarflare.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 67270136949e1d55e1a614b0a2e053b7762384ef\nAuthor: Edward Cree \u003cecree@solarflare.com\u003e\nDate: Tue Aug 6 14:53:31 2019 +0100\n\n sfc: falcon: don\u0027t score irq moderation points for GRO\n \n Same rationale as for sfc, except that this wasn\u0027t performance-tested.\n \n Signed-off-by: Edward Cree \u003cecree@solarflare.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 5e040d4b1a440b832c7b4cf8116eebcdff91909c\nAuthor: Edward Cree \u003cecree@solarflare.com\u003e\nDate: Tue Aug 6 14:53:20 2019 +0100\n\n sfc: don\u0027t score irq moderation points for GRO\n \n We already scored points when handling the RX event, no-one else does this,\n and looking at the history it appears this was originally meant to only\n score on merges, not on GRO_NORMAL. Moreover, it gets in the way of\n changing GRO to not immediately pass GRO_NORMAL skbs to the stack.\n Performance testing with four TCP streams received on a single CPU (where\n throughput was line rate of 9.4Gbps in all tests) showed a 13.7% reduction\n in RX CPU usage (n\u003d6, p\u003d0.03).\n \n Signed-off-by: Edward Cree \u003cecree@solarflare.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 5e6d9fc76190aa70db9cbfb18a6f44f4ee83b7f5\nAuthor: Rahul Verma \u003crahulv@marvell.com\u003e\nDate: Mon Aug 5 23:59:50 2019 -0700\n\n qed: Add new ethtool supported port types based on media.\n \n Supported ports in ethtool \u003ceth1\u003e are displayed based on media type.\n For media type fibre and twinaxial, port type is \"FIBRE\". Media type\n Base-T is \"TP\" and media KR is \"Backplane\".\n \n V1-\u003eV2:\n Corrected the subject.\n \n Signed-off-by: Rahul Verma \u003crahulv@marvell.com\u003e\n Signed-off-by: Michal Kalderon \u003cmichal.kalderon@marvell.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ad2dcba008a4a24a39ba39300a460bec2f73fd04\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Aug 6 10:58:54 2019 +0800\n\n cxgb4: smt: Use normal int for refcount\n \n All refcount operations are protected by spinlocks now.\n Then the atomic counter can be replaced by a normal int.\n \n This patch depends on PATCH 1/2.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4a8937b83892cb69524291cae6cdabad4a8be033\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Aug 6 10:58:46 2019 +0800\n\n cxgb4: smt: Add lock for atomic_dec_and_test\n \n The atomic_dec_and_test() is not safe because it is\n outside of locks.\n Move the locks of t4_smte_free() to its caller,\n cxgb4_smt_release() to protect the atomic decrement.\n \n Fixes: 3bdb376e6944 (\"cxgb4: introduce SMT ops to prepare for SMAC rewrite support\")\n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit e858ef1cd4bc1bdfcd18114a8195236e336cee42\nAuthor: David Ahern \u003cdsahern@gmail.com\u003e\nDate: Mon Aug 5 15:41:37 2019 -0700\n\n selftests: Add l2tp tests\n \n Add IPv4 and IPv6 l2tp tests. Current set is over IP and with\n IPsec.\n \n v2\n - add l2tp.sh to TEST_PROGS in Makefile\n \n Signed-off-by: David Ahern \u003cdsahern@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9d2f11238398793be11830a6f41908652b661395\nAuthor: Alexey Dobriyan \u003cadobriyan@gmail.com\u003e\nDate: Mon Aug 5 23:34:41 2019 +0300\n\n net: delete \"register\" keyword\n \n Delete long obsoleted \"register\" keyword.\n \n Signed-off-by: Alexey Dobriyan \u003cadobriyan@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4b4de39850474350bfc2f3df33ee1c23d0475c59\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Sat Aug 3 00:48:21 2019 +0800\n\n mkiss: Use refcount_t for refcount\n \n refcount_t is better for reference counters since its\n implementation can prevent overflows.\n So convert atomic_t ref counters to refcount_t.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 31168a6d129aebc02f92d4b7cc9946c0b6364c2b\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Sat Aug 3 00:47:59 2019 +0800\n\n dpaa_eth: Use refcount_t for refcount\n \n refcount_t is better for reference counters since its\n implementation can prevent overflows.\n So convert atomic_t ref counters to refcount_t.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit abc552284f6b1e8e6f153771dac1dff72e9d6d66\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Wed Jul 24 13:16:15 2019 -0400\n\n MIPS: jz4740: Drop obsolete code\n \n The old clocksource/timer platform code is now obsoleted by the newly\n introduced TCU drivers.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Mathieu Malaterre \u003cmalat@debian.org\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n Cc: Lee Jones \u003clee.jones@linaro.org\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Michael Turquette \u003cmturquette@baylibre.com\u003e\n Cc: Stephen Boyd \u003csboyd@kernel.org\u003e\n Cc: Jason Cooper \u003cjason@lakedaemon.net\u003e\n Cc: Marc Zyngier \u003cmarc.zyngier@arm.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: devicetree@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: linux-doc@vger.kernel.org\n Cc: linux-mips@vger.kernel.org\n Cc: linux-clk@vger.kernel.org\n Cc: od@zcrc.me\n\ncommit 967a7100400ab8fa06752f4791934353bae67ced\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Wed Jul 24 13:16:14 2019 -0400\n\n MIPS: GCW0: Reduce system timer and clocksource to 750 kHz\n \n The default clock (12 MHz) is too fast for the system timer.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Mathieu Malaterre \u003cmalat@debian.org\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n Cc: Lee Jones \u003clee.jones@linaro.org\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Michael Turquette \u003cmturquette@baylibre.com\u003e\n Cc: Stephen Boyd \u003csboyd@kernel.org\u003e\n Cc: Jason Cooper \u003cjason@lakedaemon.net\u003e\n Cc: Marc Zyngier \u003cmarc.zyngier@arm.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: devicetree@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: linux-doc@vger.kernel.org\n Cc: linux-mips@vger.kernel.org\n Cc: linux-clk@vger.kernel.org\n Cc: od@zcrc.me\n\ncommit 157c887aff5236b2ee15d4e7d2719537f14dae49\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Wed Jul 24 13:16:13 2019 -0400\n\n MIPS: CI20: Reduce system timer and clocksource to 3 MHz\n \n The default clock (48 MHz) is too fast for the system timer.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Mathieu Malaterre \u003cmalat@debian.org\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n Cc: Lee Jones \u003clee.jones@linaro.org\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Michael Turquette \u003cmturquette@baylibre.com\u003e\n Cc: Stephen Boyd \u003csboyd@kernel.org\u003e\n Cc: Jason Cooper \u003cjason@lakedaemon.net\u003e\n Cc: Marc Zyngier \u003cmarc.zyngier@arm.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: devicetree@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: linux-doc@vger.kernel.org\n Cc: linux-mips@vger.kernel.org\n Cc: linux-clk@vger.kernel.org\n Cc: od@zcrc.me\n\ncommit a68d3b052b57a223a6842883f6a2f71621d2e8cb\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Wed Jul 24 13:16:12 2019 -0400\n\n MIPS: qi_lb60: Reduce system timer and clocksource to 750 kHz\n \n The default clock (12 MHz) is too fast for the system timer, which fails\n to report time accurately.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Mathieu Malaterre \u003cmalat@debian.org\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n Cc: Lee Jones \u003clee.jones@linaro.org\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Michael Turquette \u003cmturquette@baylibre.com\u003e\n Cc: Stephen Boyd \u003csboyd@kernel.org\u003e\n Cc: Jason Cooper \u003cjason@lakedaemon.net\u003e\n Cc: Marc Zyngier \u003cmarc.zyngier@arm.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: devicetree@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: linux-doc@vger.kernel.org\n Cc: linux-mips@vger.kernel.org\n Cc: linux-clk@vger.kernel.org\n Cc: od@zcrc.me\n\ncommit 36aafdbd52881eda9073c4d03d65438a16b87a92\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Wed Jul 24 13:16:11 2019 -0400\n\n MIPS: jz4740: Add DTS nodes for the TCU drivers\n \n Add DTS nodes for the JZ4780, JZ4770 and JZ4740 devicetree files.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Mathieu Malaterre \u003cmalat@debian.org\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n Cc: Lee Jones \u003clee.jones@linaro.org\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Michael Turquette \u003cmturquette@baylibre.com\u003e\n Cc: Stephen Boyd \u003csboyd@kernel.org\u003e\n Cc: Jason Cooper \u003cjason@lakedaemon.net\u003e\n Cc: Marc Zyngier \u003cmarc.zyngier@arm.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: devicetree@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: linux-doc@vger.kernel.org\n Cc: linux-mips@vger.kernel.org\n Cc: linux-clk@vger.kernel.org\n Cc: od@zcrc.me\n\ncommit 73dd11dc1a883d4c994d729dc9984f4890001157\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Wed Jul 24 13:16:10 2019 -0400\n\n clk: jz4740: Add TCU clock\n \n Add the missing TCU clock to the list of clocks supplied by the CGU for\n the JZ4740 SoC.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Mathieu Malaterre \u003cmalat@debian.org\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Acked-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Acked-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n Cc: Lee Jones \u003clee.jones@linaro.org\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Michael Turquette \u003cmturquette@baylibre.com\u003e\n Cc: Jason Cooper \u003cjason@lakedaemon.net\u003e\n Cc: Marc Zyngier \u003cmarc.zyngier@arm.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: devicetree@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: linux-doc@vger.kernel.org\n Cc: linux-mips@vger.kernel.org\n Cc: linux-clk@vger.kernel.org\n Cc: od@zcrc.me\n\ncommit 34e9368301d534a9c5cded879f277791194fdb31\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Wed Jul 24 13:16:09 2019 -0400\n\n clocksource: Add a new timer-ingenic driver\n \n This driver handles the TCU (Timer Counter Unit) present on the Ingenic\n JZ47xx SoCs, and provides the kernel with a system timer, a clocksource\n and a sched_clock.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Mathieu Malaterre \u003cmalat@debian.org\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Reviewed-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n Cc: Lee Jones \u003clee.jones@linaro.org\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: Michael Turquette \u003cmturquette@baylibre.com\u003e\n Cc: Stephen Boyd \u003csboyd@kernel.org\u003e\n Cc: Jason Cooper \u003cjason@lakedaemon.net\u003e\n Cc: Marc Zyngier \u003cmarc.zyngier@arm.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: devicetree@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: linux-doc@vger.kernel.org\n Cc: linux-mips@vger.kernel.org\n Cc: linux-clk@vger.kernel.org\n Cc: od@zcrc.me\n\ncommit 9536eba03ec7f64fc65144b1323aef7c5e9aafcd\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Wed Jul 24 13:16:08 2019 -0400\n\n irqchip: Add irq-ingenic-tcu driver\n \n This driver handles the interrupt controller built in the Timer/Counter\n Unit (TCU) of the JZ47xx SoCs from Ingenic.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Mathieu Malaterre \u003cmalat@debian.org\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Reviewed-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n Cc: Lee Jones \u003clee.jones@linaro.org\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: Michael Turquette \u003cmturquette@baylibre.com\u003e\n Cc: Stephen Boyd \u003csboyd@kernel.org\u003e\n Cc: Jason Cooper \u003cjason@lakedaemon.net\u003e\n Cc: Marc Zyngier \u003cmarc.zyngier@arm.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: devicetree@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: linux-doc@vger.kernel.org\n Cc: linux-mips@vger.kernel.org\n Cc: linux-clk@vger.kernel.org\n Cc: od@zcrc.me\n\ncommit 4f89e4b8f1215c113d87081efc99f28b3fcb6292\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Wed Jul 24 13:16:07 2019 -0400\n\n clk: ingenic: Add driver for the TCU clocks\n \n Add driver to support the clocks provided by the Timer/Counter Unit\n (TCU) of the JZ47xx SoCs from Ingenic.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Mathieu Malaterre \u003cmalat@debian.org\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Acked-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n Cc: Lee Jones \u003clee.jones@linaro.org\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Michael Turquette \u003cmturquette@baylibre.com\u003e\n Cc: Jason Cooper \u003cjason@lakedaemon.net\u003e\n Cc: Marc Zyngier \u003cmarc.zyngier@arm.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: devicetree@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: linux-doc@vger.kernel.org\n Cc: linux-mips@vger.kernel.org\n Cc: linux-clk@vger.kernel.org\n Cc: od@zcrc.me\n\ncommit 39233b7c611248c0d05209b4854bc63e26485655\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Wed Jul 24 13:16:06 2019 -0400\n\n mfd/syscon: Add device_node_to_regmap()\n \n device_node_to_regmap() is exactly like syscon_node_to_regmap(), but it\n does not check that the node is compatible with \"syscon\", and won\u0027t\n attach the first clock it finds to the regmap.\n \n The rationale behind this, is that one device node with a standard\n compatible string \"foo,bar\" can be covered by multiple drivers sharing a\n regmap, or by a single driver doing all the job without a regmap, but\n these are implementation details which shouldn\u0027t reflect on the\n devicetree.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Acked-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n Cc: Lee Jones \u003clee.jones@linaro.org\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Michael Turquette \u003cmturquette@baylibre.com\u003e\n Cc: Stephen Boyd \u003csboyd@kernel.org\u003e\n Cc: Jason Cooper \u003cjason@lakedaemon.net\u003e\n Cc: Marc Zyngier \u003cmarc.zyngier@arm.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: devicetree@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: linux-doc@vger.kernel.org\n Cc: linux-mips@vger.kernel.org\n Cc: linux-clk@vger.kernel.org\n Cc: od@zcrc.me\n Cc: Mathieu Malaterre \u003cmalat@debian.org\u003e\n\ncommit 2e8722a5255e6596246c0feee221902f11178ee5\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Wed Jul 24 13:16:05 2019 -0400\n\n dt-bindings: Add doc for the Ingenic TCU drivers\n \n Add documentation about how to properly use the Ingenic TCU\n (Timer/Counter Unit) drivers from devicetree.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Tested-by: Mathieu Malaterre \u003cmalat@debian.org\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n Cc: Lee Jones \u003clee.jones@linaro.org\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Michael Turquette \u003cmturquette@baylibre.com\u003e\n Cc: Stephen Boyd \u003csboyd@kernel.org\u003e\n Cc: Jason Cooper \u003cjason@lakedaemon.net\u003e\n Cc: Marc Zyngier \u003cmarc.zyngier@arm.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: devicetree@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: linux-doc@vger.kernel.org\n Cc: linux-mips@vger.kernel.org\n Cc: linux-clk@vger.kernel.org\n Cc: od@zcrc.me\n\ncommit 97689a1a3fdad101d41488f9da0b1d99aa0b6aa2\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Wed Jul 24 13:16:04 2019 -0400\n\n doc: Add doc for the Ingenic TCU hardware\n \n Add documentation about the Timer/Counter Unit (TCU) present in the\n Ingenic JZ47xx SoCs.\n \n The Timer/Counter Unit (TCU) in Ingenic JZ47xx SoCs is a multi-function\n hardware block. It features up to to eight channels, that can be used as\n counters, timers, or PWM.\n \n - JZ4725B, JZ4750, JZ4755 only have six TCU channels. The other SoCs all\n have eight channels.\n \n - JZ4725B introduced a separate channel, called Operating System Timer\n (OST). It is a 32-bit programmable timer. On JZ4770 and above, it is\n 64-bit.\n \n - Each one of the TCU channels has its own clock, which can be reparented\n to three different clocks (pclk, ext, rtc), gated, and reclocked, through\n their TCSR register.\n * The watchdog and OST hardware blocks also feature a TCSR register with\n the same format in their register space.\n * The TCU registers used to gate/ungate can also gate/ungate the watchdog\n and OST clocks.\n \n - Each TCU channel works in one of two modes:\n * mode TCU1: channels cannot work in sleep mode, but are easier to\n operate.\n * mode TCU2: channels can work in sleep mode, but the operation is a bit\n more complicated than with TCU1 channels.\n \n - The mode of each TCU channel depends on the SoC used:\n * On the oldest SoCs (up to JZ4740), all of the eight channels operate in\n TCU1 mode.\n * On JZ4725B, channel 5 operates as TCU2, the others operate as TCU1.\n * On newest SoCs (JZ4750 and above), channels 1-2 operate as TCU2, the\n others operate as TCU1.\n \n - Each channel can generate an interrupt. Some channels share an interrupt\n line, some don\u0027t, and this changes between SoC versions:\n * on older SoCs (JZ4740 and below), channel 0 and channel 1 have their\n own interrupt line; channels 2-7 share the last interrupt line.\n * On JZ4725B, channel 0 has its own interrupt; channels 1-5 share one\n interrupt line; the OST uses the last interrupt line.\n * on newer SoCs (JZ4750 and above), channel 5 has its own interrupt;\n channels 0-4 and (if eight channels) 6-7 all share one interrupt line;\n the OST uses the last interrupt line.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Mathieu Malaterre \u003cmalat@debian.org\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n Cc: Lee Jones \u003clee.jones@linaro.org\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Michael Turquette \u003cmturquette@baylibre.com\u003e\n Cc: Stephen Boyd \u003csboyd@kernel.org\u003e\n Cc: Jason Cooper \u003cjason@lakedaemon.net\u003e\n Cc: Marc Zyngier \u003cmarc.zyngier@arm.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: devicetree@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: linux-doc@vger.kernel.org\n Cc: linux-mips@vger.kernel.org\n Cc: linux-clk@vger.kernel.org\n Cc: od@zcrc.me\n\ncommit 4bc3c420246e70088c1d09f7bc30ab5583fbd5f9\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Wed Jul 24 13:16:03 2019 -0400\n\n dt-bindings: ingenic: Add DT bindings for TCU clocks\n \n This header provides clock numbers for the ingenic,tcu\n DT binding.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Mathieu Malaterre \u003cmalat@debian.org\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Acked-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n Cc: Lee Jones \u003clee.jones@linaro.org\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Michael Turquette \u003cmturquette@baylibre.com\u003e\n Cc: Jason Cooper \u003cjason@lakedaemon.net\u003e\n Cc: Marc Zyngier \u003cmarc.zyngier@arm.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: devicetree@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: linux-doc@vger.kernel.org\n Cc: linux-mips@vger.kernel.org\n Cc: linux-clk@vger.kernel.org\n Cc: od@zcrc.me\n\ncommit f6326fa4855b0f8173af822fcc139afdbf4d4b71\nAuthor: Mark Brown \u003cbroonie@kernel.org\u003e\nDate: Thu Aug 8 23:15:54 2019 +0100\n\n ASoC: sof: Fix warning when IPC flood test is not enabled\n \n dentry is only used when the flood test is done so move the declaration\n of the variable inside the ifdef for the flood test.\n \n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 013b6510131568ce4e01856d5360bfdfe9c3632f\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Mon Aug 5 08:33:58 2019 -0600\n\n drm/panfrost: Add madvise and shrinker support\n \n Add support for madvise and a shrinker similar to other drivers. This\n allows userspace to mark BOs which can be freed when there is memory\n pressure.\n \n Unlike other implementations, we don\u0027t depend on struct_mutex. The\n driver maintains a list of BOs which can be freed when the shrinker\n is called. Access to the list is serialized with the shrinker_lock.\n \n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190805143358.21245-2-robh@kernel.org\n\ncommit 17acb9f35ed736c177f85b8ee711f278e7aff421\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Mon Aug 5 08:33:57 2019 -0600\n\n drm/shmem: Add madvise state and purge helpers\n \n Add support to the shmem GEM helpers for tracking madvise state and\n purging pages. This is based on the msm implementation.\n \n The BO provides a list_head, but the list management is handled outside\n of the shmem helpers as there are different locking requirements.\n \n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Acked-by: Acked-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190805143358.21245-1-robh@kernel.org\n\ncommit 3551a9fa2c5d3f3b0a34af67d33a0f2a5fecd94c\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Fri Aug 2 13:57:27 2019 -0600\n\n drm/panfrost: Remove completed features still in TODO\n \n There\u0027s a few features the driver supports which we forgot to remove, so\n remove them now.\n \n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190802195727.1963-1-robh@kernel.org\n\ncommit 5b5efdf79abffb959cfbdf1dad063358d23069aa\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 8 17:24:07 2019 +0100\n\n drm/i915: Make debugfs/per_file_stats scale better\n \n Currently we walk the entire list of obj-\u003evma for each obj within a file\n to find the matching vma of this context. Since we know we are searching\n for a particular vma bound to a user context, we can use the rbtree to\n search for it rather than repeatedly walk everything.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808162407.28121-1-chris@chris-wilson.co.uk\n\ncommit c990b4c3597d6c75dbc2937401ce32838adc77fc\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 8 15:45:11 2019 +0100\n\n drm/i915: Only include active engines in the capture state\n \n Skip printing out idle engines that did not contribute to the GPU hang.\n As the number of engines gets ever larger, we have increasing noise in\n the error state where typically there is only one guilty request on one\n engine that we need to inspect.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808144511.32269-1-chris@chris-wilson.co.uk\n\ncommit c7302f204490f3eb4ef839bec228315bcd3ba43f\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 8 21:27:58 2019 +0100\n\n drm/i915: Defer final intel_wakeref_put to process context\n \n As we need to acquire a mutex to serialise the final\n intel_wakeref_put, we need to ensure that we are in process context at\n that time. However, we want to allow operation on the intel_wakeref from\n inside timer and other hardirq context, which means that need to defer\n that final put to a workqueue.\n \n Inside the final wakeref puts, we are safe to operate in any context, as\n we are simply marking up the HW and state tracking for the potential\n sleep. It\u0027s only the serialisation with the potential sleeping getting\n that requires careful wait avoidance. This allows us to retain the\n immediate processing as before (we only need to sleep over the same\n races as the current mutex_lock).\n \n v2: Add a selftest to ensure we exercise the code while lockdep watches.\n v3: That test was extremely loud and complained about many things!\n v4: Not a whale!\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111295\n References: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111245\n References: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111256\n Fixes: 18398904ca9e (\"drm/i915: Only recover active engines\")\n Fixes: 51fbd8de87dc (\"drm/i915/pmu: Atomically acquire the gt_pm wakeref\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808202758.10453-1-chris@chris-wilson.co.uk\n\ncommit 36004c42c761a6d95b867a4fbb9a955034e53351\nAuthor: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\nDate: Thu Aug 8 20:15:49 2019 +0200\n\n ASoC: Intel: Skylake: large_config_get overhaul\n \n LARGE_CONFIG_GET is mainly used to retrieve requested module parameters\n but it may also carry TX payload with them. Update its implementation to\n account for both TX and RX data.\n First reply.header carries total payload size within data_off_sizefield.\n Make use of reply.header to realloc returned buffer with correct size.\n \n Failure of IPC request is permissive - error-payload may be returned, an\n informative data why GET for given param failed - and thus function\n should not collapse before entire processing is finished. Caller is\n responsible for checking returned payload and bytes parameters.\n \n Signed-off-by: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190808181549.12521-3-cezary.rojewski@intel.com\n Reviewed-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 63e45324c284d240562a5b435afc4b23ae7d923c\nAuthor: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\nDate: Thu Aug 8 20:15:48 2019 +0200\n\n ASoC: Intel: Skylake: Limit large_config_get to single frame\n \n Reply for the very first LARGE_CONFIG_GET request contains total size of\n payload to be retrieved by host.\n From then on, each subsequent reply carries buffer offset instead. As\n looping is not covered by any real-life example, remove it and cleanup\n the function for followup overhaul.\n \n Signed-off-by: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190808181549.12521-2-cezary.rojewski@intel.com\n Reviewed-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 5a2e340690f2908371eb9c1b4cfb82ba0b235b51\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Thu Aug 1 20:22:48 2019 -0500\n\n PCI: Mark expected switch fall-through\n \n Mark switch cases where we are expecting to fall through.\n \n This fixes the following warning (Building: allmodconfig i386):\n \n drivers/pci/hotplug/ibmphp_res.c: In function ‘update_bridge_ranges’:\n drivers/pci/hotplug/ibmphp_res.c:1943:16: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n function \u003d 0x8;\n ~~~~~~~~~^~~~~\n drivers/pci/hotplug/ibmphp_res.c:1944:6: note: here\n case PCI_HEADER_TYPE_MULTIBRIDGE:\n ^~~~\n \n Link: https://lore.kernel.org/r/20190802012248.GA22622@embeddedor\n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n\ncommit 3b1b1ce3596458de0d5c28309954886a9cb6d5cc\nAuthor: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\nDate: Thu Jun 6 13:25:57 2019 +1000\n\n PCI: Correct pci\u003dresource_alignment parameter example\n \n The \"pci\u003dresource_alignment\" parameter is described as requiring an order\n (not a size) and the code in pci_specified_resource_alignment() expects an\n order.\n \n But the example wrongly shows a size. Convert the example to an order.\n \n Fixes: 8b078c603249 (\"PCI: Update \"pci\u003dresource_alignment\" documentation\")\n Link: https://lore.kernel.org/r/20190606032557.107542-1-aik@ozlabs.ru\n Signed-off-by: Alexey Kardashevskiy \u003caik@ozlabs.ru\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit d2182b2d4b71ff0549a07f414d921525fade707b\nAuthor: Sumit Saxena \u003csumit.saxena@broadcom.com\u003e\nDate: Fri Jul 26 00:55:52 2019 +0530\n\n PCI: Restore Resizable BAR size bits correctly for 1MB BARs\n \n In a Resizable BAR Control Register, bits 13:8 control the size of the BAR.\n The encoded values of these bits are as follows (see PCIe r5.0, sec\n 7.8.6.3):\n \n Value BAR size\n 0 1 MB (2^20 bytes)\n 1 2 MB (2^21 bytes)\n 2 4 MB (2^22 bytes)\n ...\n 43 8 EB (2^63 bytes)\n \n Previously we incorrectly set the BAR size bits for a 1 MB BAR to 0x1f\n instead of 0, so devices that support that size, e.g., new megaraid_sas and\n mpt3sas adapters, fail to initialize during resume from S3 sleep.\n \n Correctly calculate the BAR size bits for Resizable BAR control registers.\n \n Link: https://lore.kernel.org/r/20190725192552.24295-1-sumit.saxena@broadcom.com\n Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id\u003d203939\n Fixes: d3252ace0bc6 (\"PCI: Restore resized BAR state on resume\")\n Signed-off-by: Sumit Saxena \u003csumit.saxena@broadcom.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Cc: stable@vger.kernel.org # v4.19+\n\ncommit 39098edbd79e5c9a4357eb924cb259d1c8a11346\nAuthor: Denis Efremov \u003cefremov@linux.com\u003e\nDate: Tue Aug 6 17:07:15 2019 +0300\n\n PCI: Use PCI_SRIOV_NUM_BARS in loops instead of PCI_IOV_RESOURCE_END\n \n Writing loop conditions as \"i \u003c NUM\" is a common C idiom; using \"i \u003c\u003d END\"\n is unusual and thus prone to errors. Change loops to use the former.\n \n Link: https://lore.kernel.org/r/20190806140715.19847-1-efremov@linux.com\n Signed-off-by: Denis Efremov \u003cefremov@linux.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Kuppuswamy Sathyanarayanan \u003csathyanarayanan.kuppuswamy@linux.intel.com\u003e\n\ncommit b8074aa2460b535915e8f65bf83c4bcb4220f804\nAuthor: Denis Efremov \u003cefremov@linux.com\u003e\nDate: Mon Jul 29 13:13:57 2019 +0300\n\n PCI: Convert pci_resource_to_user() to a weak function\n \n Convert pci_resource_to_user() to a weak function so the existing\n architecture-specific implementations will automatically override the\n generic one. This allows us to remove HAVE_ARCH_PCI_RESOURCE_TO_USER\n definitions and avoid the conditional compilation for this single function.\n \n Link: https://lore.kernel.org/r/20190729101401.28068-1-efremov@linux.com\n Link: https://lore.kernel.org/r/20190729101401.28068-2-efremov@linux.com\n Link: https://lore.kernel.org/r/20190729101401.28068-3-efremov@linux.com\n Link: https://lore.kernel.org/r/20190729101401.28068-4-efremov@linux.com\n Link: https://lore.kernel.org/r/20190729101401.28068-5-efremov@linux.com\n Link: https://lore.kernel.org/r/20190729101401.28068-6-efremov@linux.com\n Signed-off-by: Denis Efremov \u003cefremov@linux.com\u003e\n [bhelgaas: squash into one commit]\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Acked-by: Paul Burton \u003cpaul.burton@mips.com\u003e # MIPS\n\ncommit 0e4cc44f5066b22938f384cd57db99d66311de90\nAuthor: Libin Yang \u003clibin.yang@intel.com\u003e\nDate: Thu Aug 8 14:27:34 2019 -0500\n\n ASoC: intel: skl_hda_dsp_common: create HDMI jack kctl\n \n This patch call snd_jack_add_new_kctl() to create the HDMI jack kctls.\n Userspace needs these kctls to detect the hdmi monitor hotplug.\n \n In /usr/share/alsa/ucm, the config file needs to assign a jack kctl to\n \"JackControl\" to let PA get the jack hotplug status.\n \n Signed-off-by: Libin Yang \u003clibin.yang@intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190808192734.18286-1-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 8e082d8f42fabf9a4a0708d8012f4995765478fc\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 8 11:25:52 2019 +0800\n\n ASoC: tscs454: remove unused variable \u0027PLL_48K_RATE\u0027\n \n The global variable \u0027PLL_48K_RATE\u0027 is never used\n so just remove it.\n \n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190808032552.45360-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 0e36f36b04e7677c6b74349c9d4baea1ed5668f2\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Wed Aug 7 21:51:31 2019 -0500\n\n ASoC: soc-core: fix module_put() warning in soc_cleanup_component\n \n The recent changes introduce warnings in the SOF load/unload module\n tests. The code does not seem balanced with a confusion between\n _close() and _remove() macros. Using _remove() fixes the issue and\n removes the warning.\n \n Suggested-by: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Fixes: 4a81e8f30d0b4 (\u0027ASoC: soc-component: add snd_soc_component_get/put()\u0027)\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Acked-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/20190808025131.32482-1-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 596becd3f82a7b7e091b0f5c380bc9a0e6758126\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Wed Aug 7 10:31:36 2019 +0900\n\n ASoC: soc-core: dai_link check under soc_dpcm_debugfs_add()\n \n soc_dpcm_debugfs_add(rtd) is checking rtd-\u003edai_link pointer,\n but, rtd-\u003edai_link-\u003edynamic have been already checked before calling it.\n \n static int soc_probe_link_dais(...) {\n dai_link \u003d rtd-\u003edai_link;\n ...\n \u003d\u003e if (dai_link-\u003edynamic)\n \u003d\u003e soc_dpcm_debugfs_add(rtd);\n ...\n }\n \n void soc_dpcm_debugfs_add(rtd)\n {\n \u003d\u003e if (!rtd-\u003edai_link)\n return;\n ...\n }\n \n These pointer checks are strange/pointless.\n This patch checks dai_link-\u003edynamic under soc_dpcm_debugfs_add().\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/874l2tahnq.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit ee5b3f11416d1ba69e919b2fe86aae0b46f9a83e\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Wed Aug 7 10:31:31 2019 +0900\n\n ASoC: soc-core: define soc_dpcm_debugfs_add() for non CONFIG_DEBUG_FS\n \n soc_dpcm_debugfs_add() is implemented at soc-pcm.c under CONFIG_DEBUG_FS.\n Thus, soc-core.c which is only user of it need to use CONFIG_DEBUG_FS, too.\n \n This patch defines soc_dpcm_debugfs_add() for non CONFIG_DEBUG_FS case.\n Then, we can remove #ifdef CONFIG_DEBUG_FS from soc-core.c\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/875zn9ahnv.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit b3da42519c3e6ad977af844d61c31d0e5f874f1c\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Wed Aug 7 10:31:24 2019 +0900\n\n ASoC: soc-core: tidyup for card-\u003edeferred_resume_work\n \n card-\u003edeferred_resume_work is used if CONFIG_PM_SLEEP was defined.\n but\n 1) It is defined even though CONFIG_PM_SLEEP was not defined\n 2) random ifdef code is difficult to read.\n This patch tidyup these issues.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/877e7paho1.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit b245d273cbcd64eeaa93305f99c4ea8a6baf9c89\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Wed Aug 7 10:31:19 2019 +0900\n\n ASoC: soc-core: remove unneeded list_empty() check for snd_soc_try_rebind_card()\n \n list_for_each_entry_safe() will do nothing if it was empty list.\n This patch removes unneeded list_empty() check for\n list_for_each_entry_safe().\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/878ss5aho6.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit d8ca7a0a8583fc491b625450580c4092879af3dd\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Wed Aug 7 10:31:14 2019 +0900\n\n ASoC: soc-core: call snd_soc_dapm_debugfs_init() at soc_init_card_debugfs()\n \n We have 2 soc_init_card_debugfs() implementations for with/without DEBUG_FS.\n But, snd_soc_instantiate_card() calls snd_soc_dapm_debugfs_init() under\n ifdef DEBUG_FS after soc_init_card_debugfs(). This is very strange.\n We can call snd_soc_dapm_debugfs_init() under soc_init_card_debugfs().\n \n #ifdef CONFIG_DEBUG_FS\n \u003d\u003e static void soc_init_card_debugfs(...)\n {\n ...\n }\n ...\n #else\n \u003d\u003e static inline void soc_init_card_debugfs(...)\n {\n ...\n }\n #endif\n \n static int snd_soc_instantiate_card(struct snd_soc_card *card)\n {\n ...\n \u003d\u003e soc_init_card_debugfs(card);\n \n * #ifdef CONFIG_DEBUG_FS\n * snd_soc_dapm_debugfs_init(\u0026card-\u003edapm, card-\u003edebugfs_card_root);\n * #endif\n }\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87a7clahob.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9b98c7c2a0599084c7ed629b5f88f474d2578307\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Wed Aug 7 10:31:08 2019 +0900\n\n ASoC: soc-core: tidyup for snd_soc_add_card_controls()\n \n snd_soc_add_card_controls() registers controls by using\n for(... i \u003c num; ...). If controls was NULL, num should be zero.\n Thus, we don\u0027t need to check about controls pointer.\n This patch also cares missing return value.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87blx1ahoi.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit daa480bde6b3a9b728965e52efffc329ccee3f77\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Wed Aug 7 10:31:03 2019 +0900\n\n ASoC: soc-core: tidyup for snd_soc_dapm_add_routes()\n \n snd_soc_dapm_add_routes() registers routes by using\n for(... i \u003c num; ...). If routes was NULL, num should be zero.\n Thus, we don\u0027t need to check about route pointer.\n This patch also cares missing return value.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87d0hhahon.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit e6d7020c2946bef2efab7c70339eee6a6b3cb6a6\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Wed Aug 7 10:30:58 2019 +0900\n\n ASoC: soc-core: tidyup for snd_soc_add_component_controls()\n \n snd_soc_add_component_controls() registers controls by using\n for(... i \u003c num; ...). If controls was NULL, num should be zero.\n Thus, we don\u0027t need to check about controls pointer.\n This patch also cares missing return value.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87ef1xahor.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit cbb153c50ebe4f635d970a1685302288e3f573fe\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 8 20:45:25 2019 +0100\n\n drm/i915/selftests: Fixup a missing legacy_idx\n \n Grr, missed one*. For using the legacy engine map, we should use\n engine-\u003elegacy_idx. Ideally, we should know the intel_context in the\n selftest and avoid all the fiddling around with unwanted GEM contexts.\n \n * In my defence, the conflict was added in another patch after it was\n tested by CI.\n \n v2: mock engines needs legacy love as well\n \n Fixes: f1c4d157ab9b (\"drm/i915: Fix up the inverse mapping for default ctx-\u003eengines[]\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808194525.9410-2-chris@chris-wilson.co.uk\n\ncommit 43004f31ebf2705905460a6d9a77da4182170c38\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Thu Aug 8 17:03:21 2019 +0200\n\n spi: Rename of_spi_register_master() function\n \n Rename this function to of_spi_get_gpio_numbers() as this\n is what the function does, it does not register a master,\n it is called in the path of registering a master so the\n name is logical in a convoluted way, but it is better to\n follow Rusty Russell\u0027s ABI level no 7:\n \"The obvious use is (probably) the correct one\"\n \n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190808150321.23319-1-linus.walleij@linaro.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 5375f1efd70b5adbbbaded22889d50c07f6f89a4\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 7 21:38:22 2019 +0800\n\n regulator: act8865: Fix build error without CONFIG_POWER_SUPPLY\n \n Building without CONFIG_POWER_SUPPLY will fail:\n \n drivers/regulator/act8865-regulator.o: In function `act8865_pmic_probe\u0027:\n act8865-regulator.c:(.text+0x357): undefined reference to `devm_power_supply_register\u0027\n drivers/regulator/act8865-regulator.o: In function `act8600_charger_get_property\u0027:\n act8865-regulator.c:(.text+0x3f1): undefined reference to `power_supply_get_drvdata\u0027\n \n Add POWER_SUPPLY dependency to Kconfig.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Fixes: 2d09a79bf637 (\"regulator: act8865: Add support for act8600 charger\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190807133822.67124-1-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 06369bcc15a1620930a17fb35e26477504cd2255\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Thu Aug 8 15:03:43 2019 +0530\n\n regulator: qcom-rpmh: Add support for SM8150\n \n Add support from RPMH regulators found in SM8150 platform\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190808093343.5600-2-vkoul@kernel.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit d566aae1c80d9be2276057b3236c68bdcc5b3254\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Thu Aug 8 15:03:42 2019 +0530\n\n regulator: dt-bindings: Add PM8150x compatibles\n \n Add PM8150, PM8150L and PM8009 compatibles for these PMICs found\n in some Qualcomm platforms.\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190808093343.5600-1-vkoul@kernel.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 7352e72a513fd2757b2fda695a349d86faa4c94e\nAuthor: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nDate: Thu Aug 8 13:33:35 2019 +0300\n\n regulator: slg51000: Fix a couple NULL vs IS_ERR() checks\n \n The devm_gpiod_get_from_of_node() function never returns NULL, it\n returns error pointers on error.\n \n Fixes: a867bde3dd03 (\"regulator: slg51000: add slg51000 regulator driver\")\n Signed-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Link: https://lore.kernel.org/r/20190808103335.GD30506@mwanda\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 804cbf4bb063204ca6c2471baa694548aab02ce3\nAuthor: Keyon Jie \u003cyang.jie@linux.intel.com\u003e\nDate: Wed Aug 7 09:50:30 2019 -0500\n\n ASoC: hdac_hda: fix page fault issue by removing race\n \n There is a race between hda codec device removing and the\n jack-detecting work, which will lead to a page fault issue as the\n latter work is accessing codec device which could be already removed.\n \n Here add the cancellation of jack-detecting work before codecs are actually\n removed to avoid the race and fix the issue.\n \n Bug: https://github.com/thesofproject/linux/issues/1067\n Signed-off-by: Keyon Jie \u003cyang.jie@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190807145030.26117-1-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9749a5b6c09f1ad1a791ba702e5b1a4c57a50a6e\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Wed Aug 7 17:49:35 2019 -0700\n\n drm/i915/tgl: Fix the read of the DDI that transcoder is attached to\n \n On TGL this register do not map directly to port, it was already\n handled when setting it(TGL_TRANS_DDI_SELECT_PORT()) but not when\n reading it.\n \n To make it consisntent adding a macro for the older gens too.\n \n v2:\n Adding TGL_PORT_TRANS_DDI_SELECT() so all future users can reuse it\n (Lucas)\n \n v3:\n Missed parentheses arround val (Jose)\n \n v4:\n Renamed TGL_PORT_TRANS_DDI_SELECT to TGL_TRANS_DDI_FUNC_CTL_VAL_TO_PORT\n (Lucas)\n Added TRANS_DDI_FUNC_CTL_VAL_TO_PORT (Lucas)\n \n Reviewed-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Cc: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808004935.1787-2-jose.souza@intel.com\n\ncommit 6aa3bef15d6525198f584358055a4176eeca01af\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Wed Aug 7 17:49:34 2019 -0700\n\n drm/i915: Get transcoder power domain before reading its register\n \n When getting the pipes attached to encoder if it is not a eDP encoder\n it iterates over all pipes and read a transcoder register.\n But it should not read a transcoder register before get its power\n domain.\n \n It was not a issue in gens older than 12 because if it only had\n port A connected it would be attached to EDP and it would skip all\n the transcoders readout, if it had more than one port connected,\n pipe B would cause PG3 to be on and it contains all other\n transcoders.\n \n But on gen 12 there is no EDP transcoder so it is always iterating\n over all pipes and if only one sink is connected, PG3 is kept off\n and reading other transcoders registers would cause a\n unclaimed read warning.\n \n So here getting the power domain of the transcoder only if it is\n enabled, otherwise it is not connected to the DDI.\n \n Reviewed-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Cc: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808004935.1787-1-jose.souza@intel.com\n\ncommit 3881a72f8d0657ca998b217d0a183974ea800401\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Thu Aug 8 15:20:20 2019 +0200\n\n leds: pca953x: Include the right header\n \n This LED chip provides a GPIO driver, so include\n \u003clinux/gpio/driver.h\u003e rather than the legacy header\n \u003clinux/gpio.h\u003e.\n \n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 3f411e8c99e3354adc10350d0e2ee14914fd1961\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 8 12:10:12 2019 +0530\n\n staging: rtl8712: r8712_xmit_classifier(): Change return values and type\n \n Change return values of r8712_xmit_classifier from _SUCCESS and _FAIL to\n 0 and -EINVAL respectively.\n Modify call site to check for non-zero return values instead of _FAIL.\n Change return type from sint to int.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Reviewed-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Link: https://lore.kernel.org/r/20190808064012.12661-10-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 1c8a4f7c84dcb187c076a87e716912053f6ad6ba\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 8 12:10:11 2019 +0530\n\n staging: rtl8712: Remove while loop and check_cmd_fifo()\n \n Remove while loop as it only runs while the function check_cmd_fifo\n returns _FAIL, whereas check_cmd_fifo always returns _SUCCESS (and hence\n this loop will never run).\n Remove now unused function check_cmd_fifo.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190808064012.12661-9-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 02433a24fe71ba31b8392c2bf98ff9e8a2bce3b6\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 8 12:10:10 2019 +0530\n\n staging: rtl8712: r8712_free_drv_sw(): Change return type\n \n Change return type of r8712_free_drv_sw from u8 to void and remove its\n return statement as it always returns the same value (_SUCCESS) and even\n this value is never stored, checked or otherwise used.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190808064012.12661-8-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 690407fdc9dd8103a78c251c59513646f1bed50d\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 8 12:10:09 2019 +0530\n\n staging: rtl8712: init_drv_sw(): Change return values\n \n Change return values of init_drv_sw from _SUCCESS and _FAIL to 0 and\n -ENOMEM. Change return type from u8 to int to allow this.\n Add a return variable to streamline return of error codes of called\n functions.\n Modify call site to check for non-zero return value instead of _FAIL.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190808064012.12661-7-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 30d83c801d068abbfbb8f6f47e0c1f2a4a36e4b9\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 8 12:10:08 2019 +0530\n\n staging: rtl8712: init_default_value(): Change return type\n \n Change return type of init_default_value from u8 to void as it always\n returns _SUCCESS. Remove now-unnecessary return statement.\n Modify call site to simply call the function.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190808064012.12661-6-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 9dca5e687fb05e1b968e5151fd868ed4722863f6\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 8 12:10:07 2019 +0530\n\n staging: rtl8712: recvbuf2recvframe(): Change return type\n \n Change return type of recvbuf2recvframe from int to void as its return\n value is never used. Modify or remove return statements accordingly.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190808064012.12661-5-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit af556d30d3387f8b01495b1645d237121b6624d6\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 8 12:10:06 2019 +0530\n\n staging: rtl8712: r8712_recv_entry(): Change return type\n \n Change return type of r8712_recv_entry from s32 to void as its return\n value is never used. Modify or remove return statements accordingly.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190808064012.12661-4-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 1bdcb3666a83e533feb35a5a86e4e8e01b44dc18\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 8 12:10:05 2019 +0530\n\n staging: rtl8712: r8712_wlanhdr_to_ethhdr(): Change return values\n \n Change return values of r8712_wlanhdr_to_ethhdr from _SUCCESS/_FAIL to\n 0/-ENOMEM. Modify the function containing a call site of\n r8712_wlanhdr_to_ethhdr so that it compares the return value to\n 0/-ENOMEM instead of _SUCCESS/_FAIL.\n Change the return type of r8712_wlanhdr_to_ethhdr from sint to int to\n match its prototype in corresponding .h file.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190808064012.12661-3-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 4dbadfaf6a77e3d9ffe7069a8f1a13ae0df00d56\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 8 12:10:04 2019 +0530\n\n staging: rtl8712: recv_indicatepkt_reorder(): Change return values\n \n Change return values of recv_indicatepkt_reorder from _SUCCESS/_FAIL to\n 0/-ENOMEM and -EINVAL.\n Within recv_indicatepkt_reorder, modify comparison to true and use the\n respective value directly.\n Modify call site of recv_indicatepkt_reorder to check for the new return\n values.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190808064012.12661-2-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit fc898688aa37c08b872a2228d1f0397d01111652\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Aug 8 12:10:03 2019 +0530\n\n staging: rtl8712: amsdu_to_msdu(): Change return type\n \n Change return type of amsdu_to_msdu from int to void as it always\n returns _SUCCESS. Remove return statement as well.\n As this return value was never anything other than _SUCCESS, remove the\n if-branch that depends on the return value not being _SUCCESS.\n Remove local variable that was only used in the now-removed if-branch\n Modify call site to call the function amsdu_to_msdu without expecting\n a return value.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190808064012.12661-1-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a9ca0c5951ff1cb782cd78dd3eb57afb35f2f841\nAuthor: Jose Carlos Cazarin Filho \u003cjoseespiriki@gmail.com\u003e\nDate: Tue Aug 6 20:45:39 2019 -0300\n\n staging: rtl8723bs: fix brace position in enum declaration\n \n Fix two checkpath errors of type:\n \"open brace \u0027{\u0027 following enum go on the same line\"\n \n Signed-off-by: Jose Carlos Cazarin Filho \u003cjoseespiriki@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190806234539.7513-1-joseespiriki@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 5222a8fe6bd0135bd87a629d8d182f951b4ffeae\nAuthor: Giridhar Prasath R \u003ccristianoprasath@gmail.com\u003e\nDate: Wed Aug 7 07:33:31 2019 +0530\n\n staging: isdn: hysdn_procconf_init() remove parantheses from return value\n \n ERROR: return is not a function, parentheses are not required\n FILE: git/kernels/staging/drivers/staging/isdn/hysdn/hysdn_procconf.c:385\n + return (0);\n \n Signed-off-by: Giridhar Prasath R \u003ccristianoprasath@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190807020331.19729-1-cristianoprasath@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 4f4139e0213516e74cc22fc9787671f325c2b66b\nAuthor: Merwin Trever Ferrao \u003cmerwintf@gmail.com\u003e\nDate: Tue Aug 6 17:58:49 2019 +0530\n\n Staging: rtl8188eu: core: rtw_security: tidy up crc32_init()\n \n This code generates checkpatch warning:\n \n WARNING: else is not generally useful after a break or return\n \n Moving the declaration to the top of the function we can pull the\n code back one tab and it makes it more readable.\n \n Signed-off-by: Merwin Trever Ferrao \u003cmerwintf@gmail.com\u003e\n Reviewed-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Link: https://lore.kernel.org/r/20190806122849.GA25628@IoT-COE\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b0d525a007b9c954bd2f777174a378eb76cb89e2\nAuthor: Ajay Singh \u003cajay.kathat@microchip.com\u003e\nDate: Tue Aug 6 10:33:49 2019 +0000\n\n staging: wilc1000: use helper API to pack data for firmware\n \n As discussed in [1] use put_unaligned_le16() \u0026 put_unaligned_le32()\n helper to pack the data before sending to firmware.\n \n [1]. https://lore.kernel.org/linux-wireless/86bc79ccd379497d56bade79ec8f717603110ef7.camel@sipsolutions.net/\n \n Signed-off-by: Ajay Singh \u003cajay.kathat@microchip.com\u003e\n Link: https://lore.kernel.org/r/20190806103325.14063-1-ajay.kathat@microchip.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit baf4c10f8878d44912070561d542a74c09f05adf\nAuthor: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\nDate: Wed Jul 10 15:43:44 2019 +0200\n\n clk: mvebu: ap806: Fix clock name for the cluster\n \n Actually, the clocks exposed for the cluster are not the CPU clocks, but\n the PLL clock used as entry clock for the CPU clocks. The CPU clock will\n be managed by a driver submitting in the following patches.\n \n Signed-off-by: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n Link: https://lkml.kernel.org/r/20190710134346.30239-5-gregory.clement@bootlin.com\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit f756e362d938410a1143485cfcc44d7a8e6e3caf\nAuthor: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\nDate: Wed Jul 10 15:43:43 2019 +0200\n\n clk: mvebu: add CPU clock driver for Armada 7K/8K\n \n The CPU frequency is managed at the AP level for the Armada 7K/8K. The\n CPU frequency is modified by cluster: the CPUs of the same cluster have\n the same frequency.\n \n This patch adds the clock driver that will be used by CPUFreq, it is\n based on the work of Omri Itach \u003comrii@marvell.com\u003e.\n \n Signed-off-by: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n Link: https://lkml.kernel.org/r/20190710134346.30239-4-gregory.clement@bootlin.com\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit 33c0259092c805dc1cee9dd7bf66a955124702d9\nAuthor: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\nDate: Wed Jul 10 15:43:42 2019 +0200\n\n clk: mvebu: add helper file for Armada AP and CP clocks\n \n Clock drivers for Armada AP and Armada CP use the same function to\n generate unique clock name. A third drivers is coming with the same\n need, so it\u0027s time to move this function in a common file.\n \n Signed-off-by: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n Link: https://lkml.kernel.org/r/20190710134346.30239-3-gregory.clement@bootlin.com\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit 096f4597406bc86cd93947eb0e96a8188f934ebf\nAuthor: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\nDate: Wed Jul 10 15:43:41 2019 +0200\n\n dt-bindings: ap806: add the cluster clock node in the syscon file\n \n Document the device tree binding for the cluster clock controllers found\n in the Armada 7K/8K SoCs.\n \n Signed-off-by: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\n Link: https://lkml.kernel.org/r/20190710134346.30239-2-gregory.clement@bootlin.com\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit 82413e562ea6eadfb6de946dcc6f74af31d64e7f\nAuthor: Shravan Kumar Ramani \u003csramani@mellanox.com\u003e\nDate: Tue Jun 25 15:13:59 2019 -0400\n\n EDAC, mellanox: Add ECC support for BlueField DDR4\n \n Add ECC support for Mellanox BlueField SoC DDR controller.\n This requires SMC to the running Arm Trusted Firmware to report\n what is the current memory configuration.\n \n Reviewed-by: James Morse \u003cjames.morse@arm.com\u003e\n Signed-off-by: Shravan Kumar Ramani \u003csramani@mellanox.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit b6ff24f7b5101101ff897dfdde3f37924e676bc2\nAuthor: Valdis Kletnieks \u003cvaldis.kletnieks@vt.edu\u003e\nDate: Thu Aug 8 16:32:27 2019 +0200\n\n RAS: Build debugfs.o only when enabled in Kconfig\n \n In addition, the 0day bot reported this build error:\n \n \u003e\u003e drivers/ras/debugfs.c:10:5: error: redefinition of \u0027ras_userspace_consumers\u0027\n int ras_userspace_consumers(void)\n ^~~~~~~~~~~~~~~~~~~~~~~\n In file included from drivers/ras/debugfs.c:3:0:\n include/linux/ras.h:14:19: note: previous definition of \u0027ras_userspace_consumers\u0027 was here\n static inline int ras_userspace_consumers(void) { return 0; }\n ^~~~~~~~~~~~~~~~~~~~~~~\n \n for a riscv-specific .config where CONFIG_DEBUG_FS is not set. Fix all\n that by making debugfs.o depend on that define.\n \n [ bp: Rewrite commit message. ]\n \n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Valdis Kletnieks \u003cvaldis.kletnieks@vt.edu\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Cc: Tony Luck \u003ctony.luck@intel.com\u003e\n Cc: linux-edac@vger.kernel.org\n Cc: x86@kernel.org\n Link: http://lkml.kernel.org/r/7053.1565218556@turing-police\n\ncommit 0a54b809a3a2c31e1055b45b03708eb730222be1\nAuthor: Valdis Klētnieks \u003cvaldis.kletnieks@vt.edu\u003e\nDate: Wed Aug 7 18:59:29 2019 -0400\n\n RAS: Fix prototype warnings\n \n When building with C\u003d2 and/or W\u003d1, legitimate warnings are issued about\n missing prototypes:\n \n CHECK drivers/ras/debugfs.c\n drivers/ras/debugfs.c:4:15: warning: symbol \u0027ras_debugfs_dir\u0027 was not declared. Should it be static?\n drivers/ras/debugfs.c:8:5: warning: symbol \u0027ras_userspace_consumers\u0027 was not declared. Should it be static?\n drivers/ras/debugfs.c:38:12: warning: symbol \u0027ras_add_daemon_trace\u0027 was not declared. Should it be static?\n drivers/ras/debugfs.c:54:13: warning: symbol \u0027ras_debugfs_init\u0027 was not declared. Should it be static?\n CC drivers/ras/debugfs.o\n drivers/ras/debugfs.c:8:5: warning: no previous prototype for \u0027ras_userspace_consumers\u0027 [-Wmissing-prototypes]\n 8 | int ras_userspace_consumers(void)\n | ^~~~~~~~~~~~~~~~~~~~~~~\n drivers/ras/debugfs.c:38:12: warning: no previous prototype for \u0027ras_add_daemon_trace\u0027 [-Wmissing-prototypes]\n 38 | int __init ras_add_daemon_trace(void)\n | ^~~~~~~~~~~~~~~~~~~~\n drivers/ras/debugfs.c:54:13: warning: no previous prototype for \u0027ras_debugfs_init\u0027 [-Wmissing-prototypes]\n 54 | void __init ras_debugfs_init(void)\n | ^~~~~~~~~~~~~~~~\n \n Provide the proper includes.\n \n [ bp: Take care of the same warnings for cec.c too. ]\n \n Signed-off-by: Valdis Kletnieks \u003cvaldis.kletnieks@vt.edu\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Cc: Tony Luck \u003ctony.luck@intel.com\u003e\n Cc: linux-edac@vger.kernel.org\n Cc: x86@kernel.org\n Link: http://lkml.kernel.org/r/7168.1565218769@turing-police\n\ncommit 75e0a1e301913b2b5627a69818f2521f48e6be07\nAuthor: Govind Singh \u003cgovinds@codeaurora.org\u003e\nDate: Fri Jul 26 14:53:28 2019 +0530\n\n clk: qcom: define probe by index API as common API\n \n Extend the probe by index API in common code to be used\n by other qcom clock controller.\n \n Signed-off-by: Govind Singh \u003cgovinds@codeaurora.org\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit 64ebb57a3df6c7126532f8acd22b0612867ad3e0\nAuthor: yong.liang \u003cyong.liang@mediatek.com\u003e\nDate: Fri Jul 26 15:01:35 2019 +0800\n\n clk: reset: Modify reset-controller driver\n \n Set reset signal by a register and\n clear reset signal by another register for 8183.\n \n Signed-off-by: yong.liang \u003cyong.liang@mediatek.com\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit 7d0c76bdf227774154e191636be739d7dc0247ba\nAuthor: Govind Singh \u003cgovinds@codeaurora.org\u003e\nDate: Fri Jul 26 14:53:26 2019 +0530\n\n clk: qcom: Add WCSS gcc clock control for QCS404\n \n Add support for the WCSS QDSP gcc clock control used on qcs404\n based devices. This would allow wcss remoteproc driver to control\n the required gcc clocks to bring the subsystem out of reset.\n \n Signed-off-by: Govind Singh \u003cgovinds@codeaurora.org\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit 1ccc0ddf046a0197f2f9acca02a64da10aa6112d\nAuthor: Markus Elfring \u003celfring@users.sourceforge.net\u003e\nDate: Mon Jul 1 22:20:40 2019 +0200\n\n clk: Use seq_puts() in possible_parent_show()\n \n A string which did not contain a data format specification should be put\n into a sequence. Thus use the corresponding function “seq_puts”.\n \n This issue was detected by using the Coccinelle software.\n \n Signed-off-by: Markus Elfring \u003celfring@users.sourceforge.net\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit 9f7767226083d51c0674f1b71ac52daea6778b84\nAuthor: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\nDate: Tue Jul 2 16:27:10 2019 +0300\n\n clk: Assert prepare_lock in clk_core_get_boundaries\n \n This function iterates the clk consumer list on clk_core so it must be\n called under prepare_lock. This is already done by all callers but add a\n lockdep assert to check anyway.\n \n Signed-off-by: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Link: https://lkml.kernel.org/r/29453ee8e820457d87a8faf9d496390e59c6826f.1562073871.git.leonard.crestez@nxp.com\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit 1bd37a46770017e89943769112c5f09e5a7b24c1\nAuthor: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\nDate: Tue Jul 2 16:27:09 2019 +0300\n\n clk: Add clk_min/max_rate entries in debugfs\n \n Add two files to expose min/max clk rates as determined by\n clk_core_get_boundaries, taking all consumer requests into account.\n \n This information does not appear to be otherwise exposed to userspace.\n \n Signed-off-by: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Link: https://lkml.kernel.org/r/68e96af2df96512300604d797ade2088d7e6e496.1562073871.git.leonard.crestez@nxp.com\n [sboyd@kernel.org: Drop if statements for JSON printing]\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit f1c4d157ab9b13c1228f3b8ad7747126785460d0\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 8 12:06:12 2019 +0100\n\n drm/i915: Fix up the inverse mapping for default ctx-\u003eengines[]\n \n The order in which we store the engines inside default_engines() for the\n legacy ctx-\u003eengines[] has to match the legacy I915_EXEC_RING selector\n mapping in execbuf::user_map. If we present VCS2 as being the second\n instance of the video engine, legacy userspace calls that I915_EXEC_BSD2\n and so we need to insert it into the second video slot.\n \n v2: Record the legacy mapping (hopefully we can remove this need in the\n future)\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111328\n Fixes: 2edda80db3d0 (\"drm/i915: Rename engines to match their user interface\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e #v1\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808110612.23539-2-chris@chris-wilson.co.uk\n\ncommit 387758298bfdec9fbb74406207eb648bb6391670\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 8 12:06:11 2019 +0100\n\n drm/i915: Allocate kernel_contexts directly\n \n Ignore the central i915-\u003ekernel_context for allocating an engine, as\n that GEM context is being phased out. For internal clients, we just need\n the per-engine logical state, so allocate it at the point of use.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808110612.23539-1-chris@chris-wilson.co.uk\n\ncommit d4ff1b3917a529bdc75592af6b1504ad6c4029f7\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Wed Aug 7 20:50:50 2019 +0200\n\n ASoC: SOF: Intel: Initialize hdaudio bus properly\n \n The SOF HD-audio bus has its house-made initialization code. It\u0027s\n supposedly for making the code independent from HD-audio bus drivers.\n However, this is error-prone, and above all, the SOF driver has\n already dependency on HD-audio bus driver when CONFIG_SND_SOF_HDA is\n set. That is, if this Kconfig is set, there is no reason to avoid the\n call to the proper bus init function.\n \n Also, the ext_ops that is set at bus initialization can be better\n handled inside sof_hda_bus_init(). We don\u0027t need to refer this\n outside the bus initialization.\n \n So this patch addresses these issues:\n - sof_hda_bus_init() calls nothing but snd_hdac_ext_bus_init()\n when CONFIG_SND_SOF_HDA is set. Otherwise some fields are\n initialized locally like before for avoiding the dependency.\n - ext_ops is referred inside sof_hda_bus_init(). The ext_ops argument\n of snd_hda_bus_init() is dropped.\n \n Acked-by: Mark Brown \u003cbroonie@kernel.org\u003e\n Reviewed-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 19abfefd4c7604993d1c31e098a3f48bdafe334d\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Wed Aug 7 20:32:08 2019 +0200\n\n ALSA: hda: Direct MMIO accesses\n \n HD-audio drivers access to the mmio registers indirectly via the\n corresponding bus-\u003eio_ops callbacks. This is because some platform\n (notably Tegra SoC) requires the word-aligned access. But it\u0027s rather\n a rare case, and other platforms suffer from the penalties by indirect\n calls unnecessarily.\n \n This patch is an attempt to optimize and cleanup for this situation.\n Now the special aligned access is used only when a new kconfig\n CONFIG_SND_HDA_ALIGNED_MMIO is set. And the HD-audio core itself\n provides the aligned MMIO access helpers instead of the driver side.\n If Kconfig isn\u0027t set (as default), the standard helpers like readl()\n or writel() are used directly.\n \n A couple of places in ASoC Intel drivers have the access via io_ops\n reg_writel(), and they are replaced with the direct writel() calls.\n \n And now with this patch, the whole bus-\u003eio_ops becomes empty, so it\u0027s\n dropped completely. The bus initialization functions are changed\n accordingly as well to drop the whole bus-\u003eio_ops.\n \n Reviewed-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 619a1f195f93276dc8c6e33fe057e007adc9c288\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Wed Aug 7 20:02:31 2019 +0200\n\n ALSA: hda: Remove page allocation redirection\n \n The HD-audio core allocates and releases pages via driver\u0027s specific\n dma_alloc_pages and dma_free_pages ops defined in bus-\u003eio_ops. This\n was because some platforms require the uncached pages and the handling\n of page flags had to be done locally in the driver code.\n \n Since the recent change in ALSA core memory allocator, we can simply\n pass SNDRV_DMA_TYPE_DEV_UC for the uncached pages, and the only\n difference became about this type to be passed to the core allocator.\n That is, it\u0027s good time for cleaning up the mess.\n \n This patch changes the allocation code in HD-audio core to call the\n core allocator directly so that we get rid of dma_alloc_pages and\n dma_free_pages io_ops. If a driver needs the uncached pages, it has\n to set bus-\u003edma_type right after the bus initialization.\n \n This is merely a code refactoring and shouldn\u0027t bring any behavior\n changes.\n \n Reviewed-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 79e178f4383a3a645f76bc2dd44c477b361c6a98\nAuthor: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\nDate: Wed Jul 24 11:54:23 2019 +0800\n\n lib: scatterlist: Fix to support no mapped sg\n \n In function sg_split, the second sg_calculate_split will return -EINVAL\n when in_mapped_nents is 0.\n \n Indeed there is no need to do second sg_calculate_split and sg_split_mapped\n when in_mapped_nents is 0, as in_mapped_nents indicates no mapped entry in\n original sgl.\n \n Signed-off-by: Zhou Wang \u003cwangzhou1@hisilicon.com\u003e\n Acked-by: Robert Jarzmik \u003crobert.jarzmik@free.fr\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit ba6f7da99aaf00042b4ed2d8850ed7534492009e\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Aug 7 21:18:47 2019 +0800\n\n lightnvm: remove set but not used variables \u0027data_len\u0027 and \u0027rq_len\u0027\n \n drivers/lightnvm/pblk-read.c: In function pblk_submit_read_gc:\n drivers/lightnvm/pblk-read.c:423:6: warning: variable data_len set but not used [-Wunused-but-set-variable]\n drivers/lightnvm/pblk-recovery.c: In function pblk_recov_scan_oob:\n drivers/lightnvm/pblk-recovery.c:368:15: warning: variable rq_len set but not used [-Wunused-but-set-variable]\n \n They are not used since commit 48e5da725581 (\"lightnvm:\n move metadata mapping to lower level driver\")\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit ba1cab79cfc629f7238a26e040545ca394cd695f\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Wed Jul 31 10:36:59 2019 -0700\n\n dmaengine: fsl-edma: implement .device_synchronize callback\n \n Implement .device_synchronize callback in order to be able to use\n dmaengine_terminate_sync() and other primitives relying on said\n callback.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Vinod Koul \u003cvkoul@kernel.org\u003e\n Cc: linux-imx@nxp.com\n Cc: dmaengine@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190731173659.14778-1-andrew.smirnov@gmail.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit d1b622f68daf6c9f1abed360428e5a8f22063f99\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Fri Jul 26 18:59:47 2019 +0800\n\n dmaengine: pl330: use the same attributes when freeing pl330-\u003emcode_cpu\n \n In function dmac_alloc_resources(), pl330-\u003emcode_cpu is allocated using\n dma_alloc_attrs() but freed with dma_free_coherent().\n Use the correct dma_free_attrs() function to free pl330-\u003emcode_cpu.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190726105947.25342-1-huangfq.daxian@gmail.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit ca883c304f54a5b2fc83c2729691e9eae3a1226e\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Aug 8 12:56:40 2019 +0100\n\n drm/i915/selftests: Pass intel_context to mock_request\n \n Modernise the mock_request factory to take intel_context not a (GEM\n context, intel_engine_cs) tuple.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190808115640.20552-1-chris@chris-wilson.co.uk\n\ncommit 32d38e6cf007dcb298f4d66c9fd51fab0fd26c5e\nAuthor: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\nDate: Tue Jul 30 13:06:48 2019 +0530\n\n drm/i915/tgl/dsi: Enable blanking packets during BLLP for video mode\n \n Blanking packet bit will control whether the transcoder allows the link\n to enter the LP state during BLLP regions (assuming there is enough time),\n or whether it will keep the link in the HS state with a Blanking Packet\n \n Signed-off-by: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\n Reviewed-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Signed-off-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730073648.5157-7-vandita.kulkarni@intel.com\n\ncommit 33365feca49e624a3461f7c1cc8986567fca8f83\nAuthor: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\nDate: Tue Jul 30 13:06:47 2019 +0530\n\n drm/i915/tgl: Add mipi dsi support for TGL\n \n Most of the functions and mipi dsi sequence remains\n same as of ICL for TGL. Hence extending the support\n to TGL.\n \n Signed-off-by: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\n Reviewed-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Signed-off-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730073648.5157-6-vandita.kulkarni@intel.com\n\ncommit 991d9557b0c457fb92bc49ddde24a7d9ce6144a8\nAuthor: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\nDate: Tue Jul 30 13:06:46 2019 +0530\n\n drm/i915/tgl/dsi: Gate the ddi clocks after pll mapping\n \n For TGL, there is no need to keep DDI clock on till IO enabling\n for mipi dsi.\n \n Signed-off-by: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\n Reviewed-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Signed-off-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730073648.5157-5-vandita.kulkarni@intel.com\n\ncommit 7b864f95886b9772cc43d2c176eb7f585569a1b2\nAuthor: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\nDate: Tue Jul 30 13:06:45 2019 +0530\n\n drm/i915/tgl/dsi: Do not override TA_SURE\n \n Do not override TA_SURE timing parameter to\n zero for DSI 8X frequency 800MHz or below on\n TGL.\n \n Signed-off-by: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\n Reviewed-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Signed-off-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730073648.5157-4-vandita.kulkarni@intel.com\n\ncommit 960e9836f7217c682ef6cf4038c7271ab401cc7d\nAuthor: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\nDate: Tue Jul 30 13:06:44 2019 +0530\n\n drm/i915/tgl/dsi: Set latency PCS_DW1 for tgl\n \n Latency programming remains same as that of ICL and\n setting latency otimization for PCS_DW1 lanes is same as\n that of EHL, hence extending it to TGL.\n \n Signed-off-by: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\n Reviewed-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Signed-off-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730073648.5157-3-vandita.kulkarni@intel.com\n\ncommit 3522a33a2746b519b27a675639ac976c9189d1de\nAuthor: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\nDate: Tue Jul 30 13:06:43 2019 +0530\n\n drm/i915/tgl/dsi: Program TRANS_VBLANK register\n \n Program vblank register for mipi dsi in video mode\n on TGL.\n \n Signed-off-by: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\n Reviewed-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Signed-off-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730073648.5157-2-vandita.kulkarni@intel.com\n\ncommit 2af2e72b18b499fa36d3f7379fd010ff25d2a984\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Jul 2 17:16:33 2019 +0100\n\n iommu/arm-smmu-v3: Defer TLB invalidation until -\u003eiotlb_sync()\n \n Update the iommu_iotlb_gather structure passed to -\u003etlb_add_page() and\n use this information to defer all TLB invalidation until -\u003eiotlb_sync().\n This drastically reduces contention on the command queue, since we can\n insert our commands in batches rather than one-by-one.\n \n Tested-by: Ganapatrao Kulkarni \u003cgkulkarni@marvell.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 587e6c10a7ce89a5924fdbeff2ec524fbd6a124b\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Jul 2 17:16:25 2019 +0100\n\n iommu/arm-smmu-v3: Reduce contention during command-queue insertion\n \n The SMMU command queue is a bottleneck in large systems, thanks to the\n spin_lock which serialises accesses from all CPUs to the single queue\n supported by the hardware.\n \n Attempt to improve this situation by moving to a new algorithm for\n inserting commands into the queue, which is lock-free on the fast-path.\n \n Tested-by: Ganapatrao Kulkarni \u003cgkulkarni@marvell.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 057b05d5ac4745e7999656223bc6426c0129ba86\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Wed Jul 24 13:46:09 2019 +0530\n\n dmaengine: qcom: hidma_mgmt: Add of_node_put() before goto\n \n Each iteration of for_each_available_child_of_node puts the previous\n node, but in the case of a goto from the middle of the loop, there is\n no put, thus causing a memory leak.\n Hence add an of_node_put under the label that the gotos point to.\n In order to avoid decrementing an already-decremented refcount, copy the\n original contents of the label (including the return statement) to just\n above the label, so that the code under the label is executed only when\n a goto exit from the loop occurs.\n Additionally, remove an unnecessary get/put pair from the loop, as the\n loop itself already keeps track of refcount.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Acked-by: Sinan Kaya \u003cokaya@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190724081609.9724-1-nishkadg.linux@gmail.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 232a7f18cf8ecbd92a8cc6ca7feee4f6aab668fe\nAuthor: Robin Gong \u003cyibin.gong@nxp.com\u003e\nDate: Wed Jul 24 15:20:34 2019 +0800\n\n dmaengine: fsl-edma: add i.mx7ulp edma2 version support\n \n Add edma2 for i.mx7ulp by version v3, since v2 has already\n been used by mcf-edma.\n The big changes based on v1 are belows:\n 1. only one dmamux.\n 2. another clock dma_clk except dmamux clk.\n 3. 16 independent interrupts instead of only one interrupt for\n all channels.\n \n Signed-off-by: Robin Gong \u003cyibin.gong@nxp.com\u003e\n Link: https://lore.kernel.org/r/1563952834-7731-1-git-send-email-yibin.gong@nxp.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 719e25dba4430a7f2b1e77fa962f0527ebc9371d\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Mon Jul 15 11:17:23 2019 +0800\n\n dmaengine: qcom_hidma: Remove call to memset after dmam_alloc_coherent\n \n In commit 518a2f1925c3\n (\"dma-mapping: zero memory returned from dma_alloc_*\"),\n dma_alloc_coherent has already zeroed the memory.\n So memset is not needed.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Acked-by: Sinan Kaya \u003cokaya@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190715031723.6375-1-huangfq.daxian@gmail.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 9603a7ab60989f586786f082a8b1e3bf6fbcc749\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Mon Jul 15 11:17:16 2019 +0800\n\n dmaengine: imx-sdma: Remove call to memset after dma_alloc_coherent\n \n In commit 518a2f1925c3\n (\"dma-mapping: zero memory returned from dma_alloc_*\"),\n dma_alloc_coherent has already zeroed the memory.\n So memset is not needed.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190715031716.6328-1-huangfq.daxian@gmail.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit bf5ab1bdedac068a90c23c84d9b3495cf1190b33\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Wed Jul 31 16:39:58 2019 +0300\n\n pinctrl: denverton: Update pin names according to v1.08\n \n Version 1.08 of pin list has some changes in pin names for Intel Denverton.\n \n Update the driver accordingly.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n\ncommit b6fd948832496f01750d6c87dfdcef713043da4c\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Thu Aug 8 11:25:09 2019 +0200\n\n MAINTAINERS: mark wusbcore and UWB as obsolete\n \n Joe rightly points out that we should be using the \"Obsolete\" status for\n these two subsystems.\n \n Also I got the path name wrong for the wusbcore tree.\n \n Reported-by: Joe Perches \u003cjoe@perches.com\u003e\n Fixes: 71ed79b0e4be (\"USB: Move wusbcore and UWB to staging as it is obsolete\")\n Link: https://lore.kernel.org/r/20190808092509.GA20173@kroah.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit ee8193ee96c7802b5ba766243c4c950059a43323\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Fri Aug 2 12:26:02 2019 +0200\n\n ACPI: PM: Print debug messages on device power state changes\n \n Add an acpi_handle_debug() statement to acpi_device_set_power() to\n allow ACPI device power state changes to be tracked.\n \n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit ac9eafbe930abb589e9289842a99cc575cadb854\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Thu Aug 1 19:31:10 2019 +0200\n\n ACPI: PM: s2idle: Execute LPS0 _DSM functions with suspended devices\n \n According to Section 3.5 of the \"Intel Low Power S0 Idle\" document [1],\n Function 5 of the LPS0 _DSM is expected to be invoked when the system\n configuration matches the criteria for entering the target low-power\n state of the platform. In particular, this means that all devices\n should be suspended and in low-power states already when that function\n is invoked.\n \n This is not the case currently, however, because Function 5 of the\n LPS0 _DSM is invoked by it before the \"noirq\" phase of device suspend,\n which means that some devices may not have been put into low-power\n states yet at that point. That is a consequence of the previous\n design of the suspend-to-idle flow that allowed the \"noirq\" phase of\n device suspend and the \"noirq\" phase of device resume to be carried\n out for multiple times while \"suspended\" (if any spurious wakeup\n events were detected) and the point of the LPS0 _DSM Function 5\n invocation was chosen so as to call it (and LPS0 _DSM Function 6\n analogously) once per suspend-resume cycle (regardless of how many\n times the \"noirq\" phases of device suspend and resume were carried\n out while \"suspended\").\n \n Now that the suspend-to-idle flow has been redesigned to carry out\n the \"noirq\" phases of device suspend and resume once in each cycle,\n the code can be reordered to follow the specification that it is\n based on more closely.\n \n For this purpose, add -\u003eprepare_late and -\u003erestore_early platform\n callbacks for suspend-to-idle, to be executed, respectively, after\n the \"noirq\" phase of suspending devices and before the \"noirq\"\n phase of resuming them and make ACPI use them for the invocation\n of LPS0 _DSM functions as appropriate.\n \n While at it, move the LPS0 entry requirements check to be made\n before invoking Functions 3 and 5 of the LPS0 _DSM (also once\n per cycle) as follows from the specification [1].\n \n Link: https://uefi.org/sites/default/files/resources/Intel_ACPI_Low_Power_S0_Idle.pdf # [1]\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Tested-by: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\n\ncommit 29113f2f0a7d8d5332bfdfdfca995c06d0896e83\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Wed Jul 31 11:06:00 2019 +0200\n\n ACPI: EC: PM: Make acpi_ec_dispatch_gpe() print debug message\n \n Add a pm_pr_dbg() debug statement to acpi_ec_dispatch_gpe() to print\n a message when the EC GPE has been dispatched (because its status\n was set).\n \n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Tested-by: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\n\ncommit d7589404932be148fabe696b56b7c391bad6bdb1\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Wed Jul 31 11:05:52 2019 +0200\n\n ACPI: EC: PM: Consolidate some code depending on PM_SLEEP\n \n Move some routines, including acpi_ec_dispatch_gpe(), that are only\n used if CONFIG_PM_SLEEP is set to the #ifdef block containing the EC\n suspend and resume callbacks, to make the \"full EC PM picture\" easier\n to follow.\n \n While at it, move the header of acpi_ec_dispatch_gpe() in the\n header file to a CONFIG_PM_SLEEP #ifdef block.\n \n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Tested-by: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\n\ncommit 6e86633a791fdf631617ef3a9af3263141d34bc9\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Wed Jul 31 11:05:42 2019 +0200\n\n ACPI: PM: s2idle: Eliminate acpi_sleep_no_ec_events()\n \n Change acpi_ec_suspend() to use pm_suspend_no_platform() instead of\n acpi_sleep_no_ec_events(), which allows the latter to be eliminated\n along with the s2idle_in_progress variable which is only used by it.\n \n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Tested-by: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\n\ncommit fcd0a04267ac7c5d5f9a27d2af824270f2091760\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Wed Jul 31 11:05:33 2019 +0200\n\n ACPI: PM: s2idle: Switch EC over to polling during \"noirq\" suspend\n \n Since the ACPI SCI is set up for system wakeup before the \"noirq\"\n suspend of devices, it is better to make suspend-to-idle follow\n suspend-to-RAM (S3) and switch over the EC to polling during \"noirq\"\n suspend (and back to interrupt-based flow during \"noirq\" resume).\n \n The frequency of spurious wakeup interrupts from the EC may be\n reduced this way.\n \n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Tested-by: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\n\ncommit 068b47d0984b8756ae71702a1a87aa226cb72fe8\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Wed Jul 31 11:05:25 2019 +0200\n\n ACPI: PM: s2idle: Add acpi.sleep_no_lps0 module parameter\n \n Add a module parameter to prevent the ACPI LPS0 _DSM functions\n from being invoked (if need be) and rework the suspend-to-idle\n blacklist entries in acpisleep_dmi_table[] to make them simply\n prevent suspend-to-idle from being used by default on the systems\n in question (which really is the original purpose of those entries).\n \n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Tested-by: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\n\ncommit 2e2c2fdc53437beffd2cf26aaf6187e602d565bc\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Wed Jul 31 11:05:15 2019 +0200\n\n ACPI: PM: s2idle: Rearrange lps0_device_attach()\n \n To allow a subsequent change to be simpler, rearrange the code in\n lps0_device_attach() to reduce the indentation level and (while\n at it) make it avoid calling lpi_device_get_constraints() when\n lps0_device_handle is not going to be set.\n \n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Tested-by: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\n\ncommit ddc8cd7d92b333e787f34a5ee3f40ba2f1a31087\nAuthor: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\nDate: Thu Aug 8 03:03:30 2019 +0200\n\n dt-bindings: rcar-{csi2,vin}: Rename bindings documentation files\n \n Renesas media binding documentation files use a naming schema of\n \u0027renesas,\u003cmodule\u003e.txt\u0027. Rename the VIN and CSI-2 files to match this\n pattern.\n \n Signed-off-by: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\n Reviewed-by: Ulrich Hecht \u003culi+renesas@fpond.eu\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 72a83ded7f4c30cfcdcf4e6500ff4da03970d9a4\nAuthor: Simon Horman \u003chorms+renesas@verge.net.au\u003e\nDate: Mon Jul 29 09:47:57 2019 +0200\n\n dt-bindings: rcar-imr: Rename bindings documentation file\n \n Renesas media binding documentation files use a naming schema of\n \u0027renesas,\u003cmodule\u003e.txt\u0027. Rename the IMR file to match this pattern.\n \n Signed-off-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit f32d0c636371788684ea14c3d05c98a7818e4d04\nAuthor: Simon Horman \u003chorms+renesas@verge.net.au\u003e\nDate: Wed Jul 3 10:41:06 2019 +0200\n\n dt-bindings: Rename file of DT bindings for Renesas memory controllers\n \n For consistency with the naming of (most) other documentation files for DT\n bindings for Renesas IP blocks rename the Renesas R-Mobile and SH-Mobile\n memory controllers documentation file from renesas-memory-controllers.txt\n to renesas,dbsc.txt.\n \n Signed-off-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 707d26dcc1765172031b3779bb03a9db249ad949\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Wed Aug 7 15:04:15 2019 +0300\n\n drm/i915: split out intel_pch.[ch] from i915_drv.[ch]\n \n Abstract the rather self-contained piece of code from i915_drv.[ch]. No\n functional changes.\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Acked-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807120415.17917-1-jani.nikula@intel.com\n\ncommit 6da53cfa0f8300c85c7b63c9a6951400474a142c\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Jul 31 11:41:16 2019 +0200\n\n pinctrl: sh-pfc: Use dev_notice_once() instead of open-coding\n \n At the time of commit 9a643c9a11259955 (\"sh-pfc: Convert message\n printing from pr_* to dev_*\"), the dev_*_once() variants didn\u0027t exist\n yet, so the once behavior was open-coded.\n \n Since commit e135303bd5bebcd2 (\"device: Add dev_\u003clevel\u003e_once variants\")\n they do, so \"revert\" to the good practice of using a helper.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Ulrich Hecht \u003culi+renesas@fpond.eu\u003e\n Reviewed-by: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\n\ncommit 1c5fecb61255aa12a16c4c06335ab68979865914\nAuthor: Narendra K \u003cNarendra.K@dell.com\u003e\nDate: Wed Jul 10 18:59:15 2019 +0000\n\n efi: Export Runtime Configuration Interface table to sysfs\n \n System firmware advertises the address of the \u0027Runtime\n Configuration Interface table version 2 (RCI2)\u0027 via\n an EFI Configuration Table entry. This code retrieves the RCI2\n table from the address and exports it to sysfs as a binary\n attribute \u0027rci2\u0027 under /sys/firmware/efi/tables directory.\n The approach adopted is similar to the attribute \u0027DMI\u0027 under\n /sys/firmware/dmi/tables.\n \n RCI2 table contains BIOS HII in XML format and is used to populate\n BIOS setup page in Dell EMC OpenManage Server Administrator tool.\n The BIOS setup page contains BIOS tokens which can be configured.\n \n Signed-off-by: Narendra K \u003cNarendra.K@dell.com\u003e\n Reviewed-by: Mario Limonciello \u003cmario.limonciello@dell.com\u003e\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n\ncommit 5828efb95bc43ad6a59f05458d3aed9649dd5a63\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jun 25 16:28:53 2019 +0200\n\n efi: ia64: move SAL systab handling out of generic EFI code\n \n The SAL systab is an Itanium specific EFI configuration table, so\n move its handling into arch/ia64 where it belongs.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n\ncommit ec7e1605d79d1d469b25e396f2056e42386f512f\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jun 25 15:48:35 2019 +0200\n\n efi/x86: move UV_SYSTAB handling into arch/x86\n \n The SGI UV UEFI machines are tightly coupled to the x86 architecture\n so there is no need to keep any awareness of its existence in the\n generic EFI layer, especially since we already have the infrastructure\n to handle arch-specific configuration tables, and were even already\n using it to some extent.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n\ncommit e55f31a599478fb06a5a5d95e019e963322535cb\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jun 25 15:36:45 2019 +0200\n\n efi: x86: move efi_is_table_address() into arch/x86\n \n The function efi_is_table_address() and the associated array of table\n pointers is specific to x86. Since we will be adding some more x86\n specific tables, let\u0027s move this code out of the generic code first.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n\ncommit 5c3ceef9ad7b340b0acee6c26d0c9e6429decb2c\nAuthor: Qais Yousef \u003cqais.yousef@arm.com\u003e\nDate: Fri Aug 2 11:46:28 2019 +0100\n\n cpufreq: schedutil: fix equation in comment\n \n scale_irq_capacity() call in schedutil_cpu_util() does\n \n util *\u003d (max - irq)\n util /\u003d max\n \n But the comment says\n \n util *\u003d (1 - irq)\n util /\u003d max\n \n Fix the comment to match what the scaling function does.\n \n Signed-off-by: Qais Yousef \u003cqais.yousef@arm.com\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Acked-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Acked-by: Vincent Guittot \u003cvincent.guittot@linaro.org\u003e\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: \"Rafael J . Wysocki\" \u003crjw@rjwysocki.net\u003e\n Link: https://lkml.kernel.org/r/20190802104628.8410-1-qais.yousef@arm.com\n\ncommit 67692435c411e5c53a1c588ecca2037aebd81f2e\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Wed May 29 20:36:44 2019 +0000\n\n sched: Rework pick_next_task() slow-path\n \n Avoid the RETRY_TASK case in the pick_next_task() slow path.\n \n By doing the put_prev_task() early, we get the rt/deadline pull done,\n and by testing rq-\u003enr_running we know if we need newidle_balance().\n \n This then gives a stable state to pick a task from.\n \n Since the fast-path is fair only; it means the other classes will\n always have pick_next_task(.prev\u003dNULL, .rf\u003dNULL) and we can simplify.\n \n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Aaron Lu \u003caaron.lwe@gmail.com\u003e\n Cc: Valentin Schneider \u003cvalentin.schneider@arm.com\u003e\n Cc: mingo@kernel.org\n Cc: Phil Auld \u003cpauld@redhat.com\u003e\n Cc: Julien Desfossez \u003cjdesfossez@digitalocean.com\u003e\n Cc: Nishanth Aravamudan \u003cnaravamudan@digitalocean.com\u003e\n Link: https://lkml.kernel.org/r/aa34d24b36547139248f32a30138791ac6c02bd6.1559129225.git.vpillai@digitalocean.com\n\ncommit 5f2a45fc9e89e022233085e6f0f352eb6ff770bb\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Wed May 29 20:36:43 2019 +0000\n\n sched: Allow put_prev_task() to drop rq-\u003elock\n \n Currently the pick_next_task() loop is convoluted and ugly because of\n how it can drop the rq-\u003elock and needs to restart the picking.\n \n For the RT/Deadline classes, it is put_prev_task() where we do\n balancing, and we could do this before the picking loop. Make this\n possible.\n \n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Valentin Schneider \u003cvalentin.schneider@arm.com\u003e\n Cc: Aaron Lu \u003caaron.lwe@gmail.com\u003e\n Cc: mingo@kernel.org\n Cc: Phil Auld \u003cpauld@redhat.com\u003e\n Cc: Julien Desfossez \u003cjdesfossez@digitalocean.com\u003e\n Cc: Nishanth Aravamudan \u003cnaravamudan@digitalocean.com\u003e\n Link: https://lkml.kernel.org/r/e4519f6850477ab7f3d257062796e6425ee4ba7c.1559129225.git.vpillai@digitalocean.com\n\ncommit 5ba553eff0c3a7c099b1e29a740277a82c0c3314\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Wed May 29 20:36:42 2019 +0000\n\n sched/fair: Expose newidle_balance()\n \n For pick_next_task_fair() it is the newidle balance that requires\n dropping the rq-\u003elock; provided we do put_prev_task() early, we can\n also detect the condition for doing newidle early.\n \n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Aaron Lu \u003caaron.lwe@gmail.com\u003e\n Cc: Valentin Schneider \u003cvalentin.schneider@arm.com\u003e\n Cc: mingo@kernel.org\n Cc: Phil Auld \u003cpauld@redhat.com\u003e\n Cc: Julien Desfossez \u003cjdesfossez@digitalocean.com\u003e\n Cc: Nishanth Aravamudan \u003cnaravamudan@digitalocean.com\u003e\n Link: https://lkml.kernel.org/r/9e3eb1859b946f03d7e500453a885725b68957ba.1559129225.git.vpillai@digitalocean.com\n\ncommit 03b7fad167efca3b7abbbb39733933f9df56e79c\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Wed May 29 20:36:41 2019 +0000\n\n sched: Add task_struct pointer to sched_class::set_curr_task\n \n In preparation of further separating pick_next_task() and\n set_curr_task() we have to pass the actual task into it, while there,\n rename the thing to better pair with put_prev_task().\n \n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Aaron Lu \u003caaron.lwe@gmail.com\u003e\n Cc: Valentin Schneider \u003cvalentin.schneider@arm.com\u003e\n Cc: mingo@kernel.org\n Cc: Phil Auld \u003cpauld@redhat.com\u003e\n Cc: Julien Desfossez \u003cjdesfossez@digitalocean.com\u003e\n Cc: Nishanth Aravamudan \u003cnaravamudan@digitalocean.com\u003e\n Link: https://lkml.kernel.org/r/a96d1bcdd716db4a4c5da2fece647a1456c0ed78.1559129225.git.vpillai@digitalocean.com\n\ncommit 10e7071b2f491b0fb981717ea0a585c441906ede\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Tue Aug 6 15:13:17 2019 +0200\n\n sched: Rework CPU hotplug task selection\n \n The CPU hotplug task selection is the only place where we used\n put_prev_task() on a task that is not current. While looking at that,\n it occured to me that we can simplify all that by by using a custom\n pick loop.\n \n Since we don\u0027t need to put current, we can do away with the fake task\n too.\n \n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Aaron Lu \u003caaron.lwe@gmail.com\u003e\n Cc: Valentin Schneider \u003cvalentin.schneider@arm.com\u003e\n Cc: mingo@kernel.org\n Cc: Phil Auld \u003cpauld@redhat.com\u003e\n Cc: Julien Desfossez \u003cjdesfossez@digitalocean.com\u003e\n Cc: Nishanth Aravamudan \u003cnaravamudan@digitalocean.com\u003e\n\ncommit f95d4eaee6d0207bff2dc93371133d31227d4cfb\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Wed May 29 20:36:40 2019 +0000\n\n sched/{rt,deadline}: Fix set_next_task vs pick_next_task\n \n Because pick_next_task() implies set_curr_task() and some of the\n details haven\u0027t mattered too much, some of what _should_ be in\n set_curr_task() ended up in pick_next_task, correct this.\n \n This prepares the way for a pick_next_task() variant that does not\n affect the current state; allowing remote picking.\n \n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Aaron Lu \u003caaron.lwe@gmail.com\u003e\n Cc: Valentin Schneider \u003cvalentin.schneider@arm.com\u003e\n Cc: mingo@kernel.org\n Cc: Phil Auld \u003cpauld@redhat.com\u003e\n Cc: Julien Desfossez \u003cjdesfossez@digitalocean.com\u003e\n Cc: Nishanth Aravamudan \u003cnaravamudan@digitalocean.com\u003e\n Link: https://lkml.kernel.org/r/38c61d5240553e043c27c5e00b9dd0d184dd6081.1559129225.git.vpillai@digitalocean.com\n\ncommit 5feeb7837a448f659e0aaa19fb446b1d9a4b323a\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Wed May 29 20:36:38 2019 +0000\n\n sched: Fix kerneldoc comment for ia64_set_curr_task\n \n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Aaron Lu \u003caaron.lwe@gmail.com\u003e\n Cc: Valentin Schneider \u003cvalentin.schneider@arm.com\u003e\n Cc: mingo@kernel.org\n Cc: Phil Auld \u003cpauld@redhat.com\u003e\n Cc: Julien Desfossez \u003cjdesfossez@digitalocean.com\u003e\n Cc: Nishanth Aravamudan \u003cnaravamudan@digitalocean.com\u003e\n Link: https://lkml.kernel.org/r/fde3a65ea3091ec6b84dac3c19639f85f452c5d1.1559129225.git.vpillai@digitalocean.com\n\ncommit 99d84bf8c65a7a0dbc9e166ca0a58ed949ac4f37\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Wed May 29 20:36:37 2019 +0000\n\n stop_machine: Fix stop_cpus_in_progress ordering\n \n Make sure the entire for loop has stop_cpus_in_progress set.\n \n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Aaron Lu \u003caaron.lwe@gmail.com\u003e\n Cc: Valentin Schneider \u003cvalentin.schneider@arm.com\u003e\n Cc: mingo@kernel.org\n Cc: Phil Auld \u003cpauld@redhat.com\u003e\n Cc: Julien Desfossez \u003cjdesfossez@digitalocean.com\u003e\n Cc: Nishanth Aravamudan \u003cnaravamudan@digitalocean.com\u003e\n Link: https://lkml.kernel.org/r/0fd8fd4b99b9b9aa88d8b2dff897f7fd0d88f72c.1559129225.git.vpillai@digitalocean.com\n\ncommit de53fd7aedb100f03e5d2231cfce0e4993282425\nAuthor: Dave Chiluk \u003cchiluk+linux@indeed.com\u003e\nDate: Tue Jul 23 11:44:26 2019 -0500\n\n sched/fair: Fix low cpu usage with high throttling by removing expiration of cpu-local slices\n \n It has been observed, that highly-threaded, non-cpu-bound applications\n running under cpu.cfs_quota_us constraints can hit a high percentage of\n periods throttled while simultaneously not consuming the allocated\n amount of quota. This use case is typical of user-interactive non-cpu\n bound applications, such as those running in kubernetes or mesos when\n run on multiple cpu cores.\n \n This has been root caused to cpu-local run queue being allocated per cpu\n bandwidth slices, and then not fully using that slice within the period.\n At which point the slice and quota expires. This expiration of unused\n slice results in applications not being able to utilize the quota for\n which they are allocated.\n \n The non-expiration of per-cpu slices was recently fixed by\n \u0027commit 512ac999d275 (\"sched/fair: Fix bandwidth timer clock drift\n condition\")\u0027. Prior to that it appears that this had been broken since\n at least \u0027commit 51f2176d74ac (\"sched/fair: Fix unlocked reads of some\n cfs_b-\u003equota/period\")\u0027 which was introduced in v3.16-rc1 in 2014. That\n added the following conditional which resulted in slices never being\n expired.\n \n if (cfs_rq-\u003eruntime_expires !\u003d cfs_b-\u003eruntime_expires) {\n /* extend local deadline, drift is bounded above by 2 ticks */\n cfs_rq-\u003eruntime_expires +\u003d TICK_NSEC;\n \n Because this was broken for nearly 5 years, and has recently been fixed\n and is now being noticed by many users running kubernetes\n (https://github.com/kubernetes/kubernetes/issues/67577) it is my opinion\n that the mechanisms around expiring runtime should be removed\n altogether.\n \n This allows quota already allocated to per-cpu run-queues to live longer\n than the period boundary. This allows threads on runqueues that do not\n use much CPU to continue to use their remaining slice over a longer\n period of time than cpu.cfs_period_us. However, this helps prevent the\n above condition of hitting throttling while also not fully utilizing\n your cpu quota.\n \n This theoretically allows a machine to use slightly more than its\n allotted quota in some periods. This overflow would be bounded by the\n remaining quota left on each per-cpu runqueueu. This is typically no\n more than min_cfs_rq_runtime\u003d1ms per cpu. For CPU bound tasks this will\n change nothing, as they should theoretically fully utilize all of their\n quota in each period. For user-interactive tasks as described above this\n provides a much better user/application experience as their cpu\n utilization will more closely match the amount they requested when they\n hit throttling. This means that cpu limits no longer strictly apply per\n period for non-cpu bound applications, but that they are still accurate\n over longer timeframes.\n \n This greatly improves performance of high-thread-count, non-cpu bound\n applications with low cfs_quota_us allocation on high-core-count\n machines. In the case of an artificial testcase (10ms/100ms of quota on\n 80 CPU machine), this commit resulted in almost 30x performance\n improvement, while still maintaining correct cpu quota restrictions.\n That testcase is available at https://github.com/indeedeng/fibtest.\n \n Fixes: 512ac999d275 (\"sched/fair: Fix bandwidth timer clock drift condition\")\n Signed-off-by: Dave Chiluk \u003cchiluk+linux@indeed.com\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Reviewed-by: Phil Auld \u003cpauld@redhat.com\u003e\n Reviewed-by: Ben Segall \u003cbsegall@google.com\u003e\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: John Hammond \u003cjhammond@indeed.com\u003e\n Cc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n Cc: Kyle Anderson \u003ckwa@yelp.com\u003e\n Cc: Gabriel Munos \u003cgmunoz@netflix.com\u003e\n Cc: Peter Oskolkov \u003cposk@posk.io\u003e\n Cc: Cong Wang \u003cxiyou.wangcong@gmail.com\u003e\n Cc: Brendan Gregg \u003cbgregg@netflix.com\u003e\n Link: https://lkml.kernel.org/r/1563900266-19734-2-git-send-email-chiluk+linux@indeed.com\n\ncommit 139d025cda1da5484e7287b35c019fe1dcf9b650\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Mon Jul 29 16:05:15 2019 +0200\n\n sched: Clean up active_mm reference counting\n \n The current active_mm reference counting is confusing and sub-optimal.\n \n Rewrite the code to explicitly consider the 4 separate cases:\n \n user -\u003e user\n \n When switching between two user tasks, all we need to consider\n is switch_mm().\n \n user -\u003e kernel\n \n When switching from a user task to a kernel task (which\n doesn\u0027t have an associated mm) we retain the last mm in our\n active_mm. Increment a reference count on active_mm.\n \n kernel -\u003e kernel\n \n When switching between kernel threads, all we need to do is\n pass along the active_mm reference.\n \n kernel -\u003e user\n \n When switching between a kernel and user task, we must switch\n from the last active_mm to the next mm, hoping of course that\n these are the same. Decrement a reference on the active_mm.\n \n The code keeps a different order, because as you\u0027ll note, both \u0027to\n user\u0027 cases require switch_mm().\n \n And where the old code would increment/decrement for the \u0027kernel -\u003e\n kernel\u0027 case, the new code observes this is a neutral operation and\n avoids touching the reference count.\n \n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Reviewed-by: Rik van Riel \u003criel@surriel.com\u003e\n Reviewed-by: Mathieu Desnoyers \u003cmathieu.desnoyers@efficios.com\u003e\n Cc: luto@kernel.org\n\ncommit 130d9c331bc59a8733b47c58ef197a2b1fa3ed43\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Thu Aug 1 12:42:06 2019 +0200\n\n rcu/tree: Fix SCHED_FIFO params\n \n A rather embarrasing mistake had us call sched_setscheduler() before\n initializing the parameters passed to it.\n \n Fixes: 1a763fd7c633 (\"rcu/tree: Call setschedule() gp ktread to SCHED_FIFO outside of atomic region\")\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Reviewed-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n Cc: Juri Lelli \u003cjuri.lelli@redhat.com\u003e\n\ncommit e57d143091f1c0b1a98140a4d2e63e113afb62c0\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Thu Aug 8 08:47:14 2019 +0200\n\n mutex: Fix up mutex_waiter usage\n \n The patch moving bits into mutex.c was a little too much; by also\n moving struct mutex_waiter a few less common CONFIGs would no longer\n build.\n \n Fixes: 5f35d5a66b3e (\"locking/mutex: Make __mutex_owner static to mutex.c\")\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n\ncommit 2e18b14e164213e608cb609430c17a6fe2ba894c\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Mon Aug 5 21:36:36 2019 +0200\n\n USB: typec: ucsi_ccg: convert i2c driver to use dev_groups\n \n The driver core now supports the option to automatically create and\n remove any needed sysfs attribute files for a driver when the device is\n bound/removed from it. Convert the uscsi_ccg code to use that instead\n of trying to create sysfs files \"by hand\".\n \n Cc: Ajay Gupta \u003cajayg@nvidia.com\u003e\n Cc: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n Cc: Wei Yongjun \u003cweiyongjun1@huawei.com\u003e\n Reviewed-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190805193636.25560-6-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d3b5e319a19e6c6b612528c57c4ad55cb10273a9\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Mon Aug 5 21:36:32 2019 +0200\n\n USB: musb: convert platform driver to use dev_groups\n \n Platform drivers now have the option to have the platform core create\n and remove any needed sysfs attribute files. So take advantage of that\n and do not register \"by hand\" any sysfs files.\n \n Acked-by: Bin Liu \u003cb-liu@ti.com\u003e\n Link: https://lore.kernel.org/r/20190805193636.25560-2-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 71ed79b0e4be0db254640c3beb9a1a0316eb5f61\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 6 12:15:09 2019 +0200\n\n USB: Move wusbcore and UWB to staging as it is obsolete\n \n The UWB and wusbcore code is long obsolete, so let us just move the code\n out of the real part of the kernel and into the drivers/staging/\n location with plans to remove it entirely in a few releases.\n \n Link: https://lore.kernel.org/r/20190806101509.GA11280@kroah.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 570aaec7e9436eae5ffe4588d08e5262731e1b75\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Wed Jul 17 07:56:51 2019 -0700\n\n clk: Constify struct clk_bulk_data * where possible\n \n The following functions:\n \n - clk_bulk_enable()\n - clk_bulk_prepare()\n - clk_bulk_disable()\n - clk_bulk_unprepare()\n \n already expect const clk_bulk_data * as a second parameter, however\n their no-op version have mismatching prototypes that don\u0027t. Fix that.\n \n While at it, constify the second argument of clk_bulk_prepare_enable()\n and clk_bulk_disable_unprepare(), since the functions they are\n comprised of already accept const clk_bulk_data *.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Russell King \u003clinux@armlinux.org.uk\u003e\n Cc: Stephen Boyd \u003csboyd@kernel.org\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: linux-clk@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lkml.kernel.org/r/20190717145651.17250-1-andrew.smirnov@gmail.com\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit 56bf8740ff475c118fd6e9e623020cde0d2cc001\nAuthor: Marc Gonzalez \u003cmarc.w.gonzalez@free.fr\u003e\nDate: Thu Jun 13 17:09:34 2019 +0200\n\n clk: qcom: msm8916: Don\u0027t build by default\n \n QCOM_A53PLL and QCOM_CLK_APCS_MSM8916 stand out as the only options\n built by default. Let\u0027s bring them back in line with the rest.\n \n Signed-off-by: Marc Gonzalez \u003cmarc.w.gonzalez@free.fr\u003e\n Link: https://lkml.kernel.org/r/d654907d-a3a2-a00f-d6f5-3a34ae25ebcf@free.fr\n Reviewed-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Acked-by: Georgi Djakov \u003cgeorgi.djakov@linaro.org\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit 6e0b7ca281d764b5ae3d2cb97cf2dfe185e7b91e\nAuthor: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\nDate: Sat Aug 3 09:59:57 2019 -0400\n\n scsi: mpt3sas: Update driver version to 31.100.00.00\n \n Updated driver version from 29.100.00.00 to 31.100.00.00 which is\n equivalent to Phase 12 OOB.\n \n Signed-off-by: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 54d74e6b9d98bc7520e41fea3b3de07a24a766ca\nAuthor: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\nDate: Sat Aug 3 09:59:56 2019 -0400\n\n scsi: mpt3sas: Run SAS DEVICE STATUS CHANGE EVENT from ISR\n \n In some cases, like while performing extensive expander reset or phy reset,\n user may observe that drives are not visible in OS. Driver\u0027s\n firmware-worker thread is blocked for more than 120 seconds resulting in a\n call trace.\n \n 1. Received target add event for Device A and hence driver has registered\n this device to SML by calling sas_rphy_add(). SML has half added this\n device and returned the control to the driver by quitting from\n sas_rphy_add() API, and started some background scanning on this device A.\n \n 2. While background scanning is going on device A, driver has received SAS\n DEVICE STATUS CHANGE EVENT with RC code \"Internal device reset\" event and\n hence driver has set tm_busy flag for this Device A from FW worker thread\n context. When tm_busy flag is set then driver return scsi commands with\n device busy status asking the kernel to retry the command after some time.\n So background scanning for device A will be waiting for this tm_busy to be\n cleared.\n \n 3. Meanwhile driver has received a target add event for Device B and hence\n driver called sas_rphy_add() API to register this device with SML. But\n since background scanning for Device A is still pending and SML is not\n quitting from sas_rphy_add(), the driver’s firmware worker thread got\n blocked.\n \n 4. Now driver has received SAS DEVICE STATUS CHANGE EVENT with RC code\n \"Internal device reset complete\" event. But as driver’s firmware worker\n thread got blocked in Step3, it can’t process this event and it was not\n clearing the tm_busy flag and deadlock occurred (where SML was waiting for\n tm_busy flag to be cleared and our FW worker thread is waiting for SML to\n quit from sas_device_rphy_add() API).\n \n Same deadlock will be observed even if device B is getting removed in\n step3. So to limit these types of deadlocks driver will process the SAS\n DEVICE STATUS CHANGE EVENT events from ISR context instead of processing\n this event from worker thread context. This improvement avoids above\n deadlock.\n \n Signed-off-by: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 1edc677019c23eda6d2e9b6c3c88ef3972e41bd1\nAuthor: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\nDate: Sat Aug 3 09:59:55 2019 -0400\n\n scsi: mpt3sas: Reduce the performance drop\n \n This patch is to reduce the performance drop depth observed on SATA HDD\n when ATA PT command is outstanding.\n \n Driver returns IO commands with status \"SAM_STAT_BUSY\" whenever ATA PT\n command is outstanding. With this, IO commands will be retried until this\n outstanding ATA PT to complete and hence we will observe drop in\n performance.\n \n As the driver is completing the subsequent IOs commands with SAM_STAT_BUSY\n status, these IOs has to go though the block layer. Hence it adds latency\n to the IOs and large performance drop is observed.\n \n So to reduce this performance dropp, added improvement in driver to return\n the subsequent IOs with SCSI_MLQUEUE_DEVICE_BUSY status instead of\n completing the IOs with SAM_STAT_BUSY status when ATA PT command is\n outstanding. Sending command back with SCSI_MLQUEUE_DEVICE_BUSY does not go\n through complete block layer stack (as scsi_done won\u0027t be called) SML will\n immediately retry the command and this method will avoid latency of block\n layer stack and the performance impact will be reduced.\n \n On Local setup, ran 512k sequential read IO operation on HGST SATA drive\n with existing driver \u0026 with this improvement drivers and here is the\n result,\n \n 1. With existing driver: IOs are running at bandwidth of ~230 rMB/s and\n whenever any ATA PT command is outstanding (e.g issued from systemd-udevd\n daemon) then this bandwidth drops to ~150 rMB/s.\n \n 2. With this improvement driver: IOs are running at bandwidth of ~230 rMB/s\n and whenever any ATA PT command is outstanding then this bandwidth drops to\n just ~190 rMB/s.\n \n Signed-off-by: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 9c067c053f94d36006cd0a29cf02b0b6be54c6ca\nAuthor: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\nDate: Sat Aug 3 09:59:54 2019 -0400\n\n scsi: mpt3sas: Handle fault during HBA initialization\n \n During HBA initialization time, if handshake operation fails due to some\n firmware fault then currently driver is terminating the HBA\n initialization. It is possible that HBA may come up properly if diag reset\n is issued.\n \n So improvement is made in driver in such a way that before terminating the\n HBA initialization, driver checks the IOC state and if IOC state is in\n fault state then issue diag reset for once. If diag reset is successful\n then continue with HBA initialization else terminate the HBA\n initialization.\n \n Signed-off-by: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 3ac8e47bbf2dfada890396b2994fe2765afd8cef\nAuthor: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\nDate: Sat Aug 3 09:59:53 2019 -0400\n\n scsi: mpt3sas: Add sysfs to know supported features\n \n Currently with sysfs parameter \"drv_support_bitmap\" driver exposes whether\n driver supports toolbox memory move command or not.\n \n And application should issue the toolbox memory move command only if driver\n tell that memory move tool box command is supported through this sysfs\n parameter.\n \n In future we can utilize this sysfs parameter if any new feature is added\n and need to notify the same to applications.\n \n Signed-off-by: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit ba630ea068d7cb593ae3bb7085846c5a137d10a6\nAuthor: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\nDate: Sat Aug 3 09:59:52 2019 -0400\n\n scsi: mpt3sas: Support MEMORY MOVE Tool box command\n \n Host uses the Memory Move Tool to copy data from any source/destination\n combination of system memory and IOC memory.\n \n Memory Move Tool box request contains two SGE fields, First SGE field must\n contains the source buffer details described by an MPI Simple SGE. The\n second SGE field must contains the destination buffer details described by\n an MPI Simple SGE.\n \n Source -\u003e Destination\n \n 1. IOC -\u003e IOC (Both the SGE\u0027s will be filled by application)\n \n 2. HOST -\u003e HOST (Both the SGE\u0027s will be filled by the host,\n application should give sgl_offset to first SGE offset)\n \n 3. IOC -\u003e HOST (Application will fill the first SGE and set the\n sgl_offset to second SGE and hence driver fills\n the second SGE)\n 4. HOST -\u003e IOC (Application will fill IOC buffer information in the\n first SGE and set the sgl_offset to second SGE.\n Then driver will fill the second SGE with Host buffer\n information and just before posting the command to the\n firmware, driver will swap these two SGEs so that first\n SGE contains the HOST buffer information and second SGE\n contains the IOC information.\n \n Driver has to take care only of the 4th case, other three cases are by\n default supported by the current driver design.\n \n Signed-off-by: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 3c090ce3f0fba1efb44eb9317ebf8e8150197a21\nAuthor: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\nDate: Sat Aug 3 09:59:51 2019 -0400\n\n scsi: mpt3sas: Allow ioctls to blocked access status NVMe\n \n If driver sees the NVMe drive with \"DEVICE_BLOCKED\" AccessStatus in its\n PCIe Device Page0, then driver removes the drive from its internal list and\n does not allow any IOCTL commands to be sent to the drive and will return\n the IOCTLs with \"-ENODEV\" status.\n \n The driver will now allow NVMe Encapsulated IOCTL issued to the NVMe device\n with an access status of DEVICE_BLOCKED. This change allows the user to\n flash new drive firmware online and revive the drive.\n \n Add NVMe device only the driver\u0027s internal list even though the device is\n in the blocked state so that the device will be visible to Apps. This way\n Apps can send NVMe Encapsulated IOCTLs to this drive and bring the drive\n online. This NVMe drive with DEVICE_BLOCKED access status won\u0027t added to\n the SML, it will be added only in the driver\u0027s internal list.\n \n [mkp: clarified desc]\n \n Signed-off-by: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 5bb309dbbbf2157786004201eb70dafd283b54b5\nAuthor: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\nDate: Sat Aug 3 09:59:50 2019 -0400\n\n scsi: mpt3sas: Enumerate SES of a managed PCIe switch\n \n SES device of managed PCIe switch will be enumerated same as NVMe drives.\n \n The device info type for this SES device is\n \n MPI26_PCIE_DEVINFO_SCSI (0x4),\n \n whereas the device info type for NVMe drives is\n \n MPI26_PCIE_DEVINFO_NVME (0x3).\n \n Based on this device info type driver determines whether the device is NVMe\n drive or a SES device of a managed PCIe switch.\n \n This SES device doesn\u0027t have the PCIe device page 2 information like NVMe\n drives, so driver won\u0027t read PCIe device page 2 information for SES device.\n \n This SES device uses only IEEE SGL\u0027s, So driver build\u0027s IEEE SGL\u0027s whenever\n it receives any SCSI commands for this SES device.\n \n Signed-off-by: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 635ee6c73034ea4488a8e26bb0d0291d83146863\nAuthor: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\nDate: Sat Aug 3 09:59:49 2019 -0400\n\n scsi: mpt3sas: Update MPI headers to 2.6.8 spec\n \n Updated MPI to 2.6.8 specification and header files to 2.00.54.\n \n Signed-off-by: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit ffedeae1fa545a1d07e6827180c3923bf67af59f\nAuthor: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\nDate: Sat Aug 3 09:59:48 2019 -0400\n\n scsi: mpt3sas: Gracefully handle online firmware update\n \n Issue:\n \n During online Firmware upgrade operations it is possible that MaxDevHandles\n filled in IOCFacts may change with new FW. With this we may observe kernel\n panics when driver try to access the pd_handles or blocking_handles buffers\n at offset greater than the old firmware\u0027s MaxDevHandle value.\n \n Fix:\n \n _base_check_ioc_facts_changes() looks for increase/decrease in IOCFacts\n attributes during online firmware upgrade and increases the pd_handles,\n blocking_handles, etc buffer sizes to new firmware\u0027s MaxDevHandle value if\n this new firmware\u0027s MaxDevHandle value is greater than the old firmware\u0027s\n MaxDevHandle value.\n \n Signed-off-by: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit e224e03b0c6a2381ed1ea5325c846582d87d6fae\nAuthor: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\nDate: Sat Aug 3 09:59:47 2019 -0400\n\n scsi: mpt3sas: memset request frame before reusing\n \n Driver gets a request frame from the free pool of DMA-able request frames\n and fill in the required information and pass the address of the frame to\n IOC/FW to pull the complete request frame. In certain places the driver\n used the request frame allocated from the free pool without completely\n clearing the previous data stored in it. The request contents were cleared\n only for the size of the new request to be issued and that left out some\n stale data in the unused part of the request. Though the IOC/FW is not\n expected to access the request beyond the specified size, it is good\n practice to clear complete request message frame.\n \n So reinitialize the complete request message frame with 0s before using\n it.\n \n Signed-off-by: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit f23ca2cb2781102b560dbd96fe093b146fd8ec1a\nAuthor: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\nDate: Sat Aug 3 09:59:46 2019 -0400\n\n scsi: mpt3sas: Add support for PCIe Lane margin\n \n PCIe Lane margin tool box request requires IEEE sgl\u0027s and hence driver\n fills the SGL field with IEEE sgl\u0027s while issuing the PCIe Lane margin\n ioctl request to the HBA firmware.\n \n Signed-off-by: Suganath Prabu \u003csuganath-prabu.subramani@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit afcd609e8e7907ccfa04fef0a3adb7d60a298ed6\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Wed Jul 31 23:22:14 2019 +0100\n\n scsi: pm80xx: remove redundant assignments to variable rc\n \n There are several occasions where variable rc is being initialized with a\n value that is never read and error is being re-assigned a little later on.\n Clean up the code by removing rc entirely and just returning the return\n value from the call to pm8001_issue_ssp_tmf\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Acked-by: Jack Wang \u003cjinpu.wang@cloud.ionos.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit a5ac1f5d9a4285adf4e88ffde7ca8ae823d72441\nAuthor: Luo Jiaxing \u003cluojiaxing@huawei.com\u003e\nDate: Mon Aug 5 21:48:12 2019 +0800\n\n scsi: hisi_sas: Consolidate internal abort calls in LU reset operation\n \n In hisi_sas_lu_reset(), we call internal abort for SAS and SATA device\n codepaths -\u003e consolidate into a single call.\n \n Signed-off-by: Luo Jiaxing \u003cluojiaxing@huawei.com\u003e\n Signed-off-by: John Garry \u003cjohn.garry@huawei.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit e7513f666bc73c357805df498df34179c8b4fc9b\nAuthor: Xiang Chen \u003cchenxiang66@hisilicon.com\u003e\nDate: Mon Aug 5 21:48:11 2019 +0800\n\n scsi: hisi_sas: replace \"%p\" with \"%pK\"\n \n The format specifier \"%p\" can leak kernel address, and use \"%pK\" instead.\n \n Signed-off-by: Xiang Chen \u003cchenxiang66@hisilicon.com\u003e\n Signed-off-by: John Garry \u003cjohn.garry@huawei.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit a07b48766c5232b98154f68010512a9269f2841e\nAuthor: Xiang Chen \u003cchenxiang66@hisilicon.com\u003e\nDate: Mon Aug 5 21:48:10 2019 +0800\n\n scsi: hisi_sas: Remove some unnecessary code\n \n Remove some unnecessary code, including:\n \n - Explicit zeroing of memory allocated for dmam_alloc_coherent()\n \n - Some duplicated code\n \n - Some redundant masking\n \n Signed-off-by: Xiang Chen \u003cchenxiang66@hisilicon.com\u003e\n Signed-off-by: John Garry \u003cjohn.garry@huawei.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 7bf18e849d80e1beb118fd6d3bd5ccfcb1267d09\nAuthor: Luo Jiaxing \u003cluojiaxing@huawei.com\u003e\nDate: Mon Aug 5 21:48:09 2019 +0800\n\n scsi: hisi_sas: Modify return type of debugfs functions\n \n For functions which always return 0, which is never checked, make to return\n void.\n \n Signed-off-by: Luo Jiaxing \u003cluojiaxing@huawei.com\u003e\n Signed-off-by: John Garry \u003cjohn.garry@huawei.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit e16963f378faf82d307cad2796953f198497c614\nAuthor: Xiang Chen \u003cchenxiang66@hisilicon.com\u003e\nDate: Mon Aug 5 21:48:08 2019 +0800\n\n scsi: hisi_sas: Drop free_irq() when devm_request_irq() failed\n \n It will free irq automatically if devm_request_irq() failed, so drop\n free_irq() if devm_request_irq() failed.\n \n Signed-off-by: Xiang Chen \u003cchenxiang66@hisilicon.com\u003e\n Signed-off-by: John Garry \u003cjohn.garry@huawei.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 5f6c32d7ce576e9275ab2e9b21192f5cd5f24273\nAuthor: John Garry \u003cjohn.garry@huawei.com\u003e\nDate: Mon Aug 5 21:48:07 2019 +0800\n\n scsi: hisi_sas: Drop SMP resp frame DMA mapping\n \n The SMP frame response is written to the command table and not the SMP\n response pointer from libsas, so don\u0027t bother DMA mapping (and unmapping)\n the SMP response from libsas.\n \n Suggested-by: Xiang Chen \u003cchenxiang66@hisilicon.com\u003e\n Signed-off-by: John Garry \u003cjohn.garry@huawei.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 1c003146c64bb3ae86f1a08d73a7e4551d7cd04a\nAuthor: John Garry \u003cjohn.garry@huawei.com\u003e\nDate: Mon Aug 5 21:48:06 2019 +0800\n\n scsi: hisi_sas: Drop kmap_atomic() in SMP command completion\n \n The call to kmap_atomic() in the SMP command completion code is\n unnecessary, since kmap() is only really concerned with highmem, which is\n not relevant on arm64. The controller only finds itself in arm64 systems.\n \n Signed-off-by: John Garry \u003cjohn.garry@huawei.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 599aefc81effaaf78c02d44461f7157f22ee3b1d\nAuthor: Xiang Chen \u003cchenxiang66@hisilicon.com\u003e\nDate: Mon Aug 5 21:48:05 2019 +0800\n\n scsi: hisi_sas: Make slot buf minimum allocation of PAGE_SIZE\n \n For a system with PAGE_SIZE of 16K or 64K, the size every time we want to\n alloc may be small like 4K, but for function dmam_alloc_coherent(), the\n least size it allocates is PAGE_SIZE, so it will waste much memory for the\n situation.\n \n To solve the issue, limit the minimum allocation size of slot buf to\n PAGE_SIZE.\n \n Signed-off-by: Xiang Chen \u003cchenxiang66@hisilicon.com\u003e\n Signed-off-by: John Garry \u003cjohn.garry@huawei.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit d380f55503ed28461422546900baedc3b512b4b7\nAuthor: Xiang Chen \u003cchenxiang66@hisilicon.com\u003e\nDate: Mon Aug 5 21:48:04 2019 +0800\n\n scsi: hisi_sas: Don\u0027t bother clearing status buffer IU in task prep\n \n For struct hisi_sas_status_buffer, it contains struct hisi_sas_err_record\n and iu[1024]. The struct iu[1024] will be filled fully by the response of\n disks, so it is not need to initialize them to 0, but for the struct\n hisi_sas_err_record, SAS controller only fill some fields of\n hisi_sas_err_record according to hw designer, so it should be initialised\n to 0. After the change, cpu utilization percentage of memset() is changed\n from 1.7% to 0.12%.\n \n Signed-off-by: Xiang Chen \u003cchenxiang66@hisilicon.com\u003e\n Signed-off-by: John Garry \u003cjohn.garry@huawei.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 445ee2de112a18419aeae72fdae4221cd90f2948\nAuthor: Luo Jiaxing \u003cluojiaxing@huawei.com\u003e\nDate: Mon Aug 5 21:48:03 2019 +0800\n\n scsi: hisi_sas: Fix out of bound at debug_I_T_nexus_reset()\n \n Fix a possible out-of-bounds access in hisi_sas_debug_I_T_nexus_reset().\n \n Signed-off-by: Luo Jiaxing \u003cluojiaxing@huawei.com\u003e\n Signed-off-by: John Garry \u003cjohn.garry@huawei.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit b0b3e4290e288bb633c4ff6331b2c0b9530aa9b8\nAuthor: Luo Jiaxing \u003cluojiaxing@huawei.com\u003e\nDate: Mon Aug 5 21:48:02 2019 +0800\n\n scsi: hisi_sas: Snapshot AXI and RAS register at debugfs\n \n The AXI and RAS register values should also should be snapshot at debugfs.\n \n Signed-off-by: Luo Jiaxing \u003cluojiaxing@huawei.com\u003e\n Signed-off-by: John Garry \u003cjohn.garry@huawei.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit bbe0a7b348b336625292092c74fc7817aeb8d30b\nAuthor: Luo Jiaxing \u003cluojiaxing@huawei.com\u003e\nDate: Mon Aug 5 21:48:01 2019 +0800\n\n scsi: hisi_sas: Snapshot HW cache of IOST and ITCT at debugfs\n \n The value of IOST/ITCT is updated to cache first, and then synchronize to\n DDR periodically. So the value in IOST/ITCT cache is the latest data and\n it\u0027s important for debugging.\n \n So, the HW cache of IOST and ITCT should be snapshot at debugfs.\n \n Signed-off-by: Luo Jiaxing \u003cluojiaxing@huawei.com\u003e\n Signed-off-by: John Garry \u003cjohn.garry@huawei.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit bee0cf25c030776a8ecfc3c951d3b73259dc6839\nAuthor: Luo Jiaxing \u003cluojiaxing@huawei.com\u003e\nDate: Mon Aug 5 21:48:00 2019 +0800\n\n scsi: hisi_sas: Fix pointer usage error in show debugfs IOST/ITCT\n \n Fix how the pointer is set in hisi_sas_debugfs_iost_show() and\n hisi_sas_debugfs_itct_show().\n \n Signed-off-by: Luo Jiaxing \u003cluojiaxing@huawei.com\u003e\n Signed-off-by: John Garry \u003cjohn.garry@huawei.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 897cc769bcc09204cb80ce5db7a8d9c404bdb6ec\nAuthor: John Garry \u003cjohn.garry@huawei.com\u003e\nDate: Mon Aug 5 21:47:59 2019 +0800\n\n scsi: hisi_sas: Drop hisi_sas_hw.get_free_slot\n \n In commit 1273d65f29045 (\"scsi: hisi_sas: change queue depth from 512 to\n 4096\"), the depth of each queue is the same as the max IPTT in the system.\n \n As such, as long as we have an IPTT allocated, we will have enough space on\n any delivery queue.\n \n All .get_free_slot functions were checking for space on the queue by\n reading the DQ read pointer. Drop this, and also raise the code into common\n code, as there is nothing hw specific remaining.\n \n Signed-off-by: John Garry \u003cjohn.garry@huawei.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 93352abc81a90314bf032038200ce96989a32c62\nAuthor: John Garry \u003cjohn.garry@huawei.com\u003e\nDate: Mon Aug 5 21:47:58 2019 +0800\n\n scsi: hisi_sas: Make max IPTT count equal for all hw revisions\n \n There is a small optimisation to be had by making the max IPTT the same for\n all hw revisions, that being we can drop the check for read and write\n pointer being the same in the get free slot function.\n \n Change v1 hw to have max IPTT of 4096 - same as v2 and v3 hw - and\n drop hisi_sas_hw.max_command_entries.\n \n Signed-off-by: John Garry \u003cjohn.garry@huawei.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit b7a2758f20251e1d7f884d96e46ea165810431c3\nAuthor: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nDate: Mon Aug 5 14:49:59 2019 +1000\n\n hwrng: n2-drv - fix typo\n \n Fixes: 3e75241be808 (\"hwrng: drivers - Use device-managed registration API\")\n Signed-off-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 69a76d08f2e41fa7f27bd6c4c0dce92cce01744a\nAuthor: Helge Deller \u003cdeller@gmx.de\u003e\nDate: Sat Aug 3 21:34:43 2019 +0200\n\n scsi: ncr53c8xx: Mark expected switch fall-through\n \n Signed-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 7d8948627a7cd84c45069276e5fcddf6ab074baa\nAuthor: Li Zhong \u003clizhongfs@gmail.com\u003e\nDate: Mon Aug 5 08:43:36 2019 +0800\n\n scsi: target: tcmu: clean the nl_cmd of the udev when nl send fails\n \n If the userspace process crashes while we send the nl msg, it is possible\n that the cmd in curr_nl_cmd of tcmu_dev never gets reset to 0, and and\n returns busy for other commands after the userspace process is restartd.\n \n More details below:\n \n /backstores/user:file/file\u003e set attribute dev_size\u003d2048\n Cannot set attribute dev_size: [Errno 3] No such process\n /backstores/user:file/file\u003e set attribute dev_size\u003d2048\n Cannot set attribute dev_size: [Errno 16] Device or resource busy\n \n with following kernel messages:\n [173605.747169] Unable to reconfigure device\n [173616.686674] tcmu daemon: command reply support 1.\n [173623.866978] netlink cmd 3 already executing on file\n [173623.866984] Unable to reconfigure device\n \n Also, it is not safe to leave the nl_cmd in the list, and not get deleted.\n \n This patch removes the nl_cmd from the list, and clear its data if it is\n not sent successfully.\n \n Signed-off-by: Li Zhong \u003clizhongfs@gmail.com\u003e\n Acked-by: Mike Christie \u003cmchristi@redhat.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit a92c7ba982e3950a82fd63b3fd289248bd99e8f4\nAuthor: Valdis Klētnieks \u003cvaldis.kletnieks@vt.edu\u003e\nDate: Wed Aug 7 19:22:34 2019 -0400\n\n fs/handle.c - fix up kerneldoc\n \n When building with W\u003d1, we get some kerneldoc warnings:\n \n CC fs/fhandle.o\n fs/fhandle.c:259: warning: Function parameter or member \u0027flags\u0027 not described in \u0027sys_open_by_handle_at\u0027\n fs/fhandle.c:259: warning: Excess function parameter \u0027flag\u0027 description in \u0027sys_open_by_handle_at\u0027\n \n Fix the typo that caused it.\n \n Signed-off-by: Valdis Kletnieks \u003cvaldis.kletnieks@vt.edu\u003e\n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit cb8b3359eb9c606421a42f3968c12211d6462906\nAuthor: Anil Varughese \u003caniljoy@cadence.com\u003e\nDate: Fri Aug 2 12:21:12 2019 +0100\n\n scsi: ufs: Configure clock in .hce_enable_notify() in Cadence UFS\n \n Configure CDNS_UFS_REG_HCLKDIV in .hce_enable_notify() instead of\n .setup_clock() because if UFSHCD resets the controller ip because of phy or\n device related errors then CDNS_UFS_REG_HCLKDIV is reset to default value\n and .setup_clock() is not called later in the sequence whereas\n .hce_enable_notify will be called everytime controller is reenabled.\n \n Signed-off-by: Anil Varughese \u003caniljoy@cadence.com\u003e\n Reviewed-by: Vignesh Raghavendra \u003cvigneshr@ti.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit dccc96abfb21dc19d69e707c38c8ba439bba7160\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 1 15:38:14 2019 -0700\n\n scsi: core: Reduce memory required for SCSI logging\n \n The data structure used for log messages is so large that it can cause a\n boot failure. Since allocations from that data structure can fail anyway,\n use kmalloc() / kfree() instead of that data structure.\n \n See also https://bugzilla.kernel.org/show_bug.cgi?id\u003d204119.\n See also commit ded85c193a39 (\"scsi: Implement per-cpu logging buffer\") # v4.0.\n \n Reported-by: Jan Palus \u003cjpalus@fastmail.com\u003e\n Cc: Christoph Hellwig \u003chch@lst.de\u003e\n Cc: Hannes Reinecke \u003chare@suse.com\u003e\n Cc: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Cc: Ming Lei \u003cming.lei@redhat.com\u003e\n Cc: Jan Palus \u003cjpalus@fastmail.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 94ef80a5f0d72376a0dd2cfed0bd123123b405e2\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 1 15:38:13 2019 -0700\n\n scsi: core: Complain if scsi_target_block() fails\n \n If scsi_target_block() fails that can break the code that calls this\n function. Hence complain loudly if scsi_target_block() fails.\n \n Cc: Christoph Hellwig \u003chch@lst.de\u003e\n Cc: Hannes Reinecke \u003chare@suse.com\u003e\n Cc: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Cc: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 09addb1d169ed2e67a0314e6275b42e7b8605d79\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 1 15:38:12 2019 -0700\n\n scsi: core: Make scsi_internal_device_unblock_nowait() reject invalid new_state\n \n The only \u0027new_state\u0027 values passed by upstream kernel code to\n scsi_internal_device_unblock_nowait() are SDEV_RUNNING and\n SDEV_TRANSPORT_OFFLINE. These are the only values that should be passed to\n this function. Hence check the value of the \u0027new_state\u0027 argument to avoid\n that scsi_internal_device_unblock_nowait() would be used to trigger an\n illegal SCSI device state transition. In this context \u0027illegal\u0027 means not\n allowed by scsi_device_set_state().\n \n Cc: Christoph Hellwig \u003chch@lst.de\u003e\n Cc: Hannes Reinecke \u003chare@suse.com\u003e\n Cc: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Cc: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 4b828fe156a662a4b6135019bf077040340f299b\nAuthor: Tomas Winkler \u003ctomas.winkler@intel.com\u003e\nDate: Tue Jul 30 08:55:17 2019 +0300\n\n scsi: ufs: revamp string descriptor reading\n \n Define new a type: uc_string_id for easier string handling and less\n casting. Reduce number or string copies in price of a dynamic allocation.\n \n Signed-off-by: Tomas Winkler \u003ctomas.winkler@intel.com\u003e\n Tested-by: Avri Altman \u003cavri.altman@wdc.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit c72a9692923f67a0ba862093848d1834259e01c3\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Jul 29 16:03:13 2019 -0500\n\n scsi: wd33c93: Mark expected switch fall-through\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warning (Building: m68k):\n \n drivers/scsi/wd33c93.c: In function round_4 :\n drivers/scsi/wd33c93.c:1856:11: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n case 2: ++x;\n ^~~\n drivers/scsi/wd33c93.c:1857:3: note: here\n case 3: ++x;\n ^~~~\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit c78a66584d0318e2a01a3f10857c9b517d23bf7a\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Jul 29 09:30:07 2019 -0500\n\n scsi: sun3_scsi: Mark expected switch fall-throughs\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warnings:\n \n drivers/scsi/sun3_scsi.c: warning: this statement may fall through\n [-Wimplicit-fallthrough\u003d]: \u003d\u003e 399:9, 403:9\n \n Reported-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Tested-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 6d44c692c12c33bab92d3e0f524f66c3e97b7bdd\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Jul 29 06:03:45 2019 -0500\n\n scsi: qlogicpti: Mark expected switch fall-throughs\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warnings (Building: sparc defconfig):\n \n drivers/scsi/qlogicpti.c: In function \u0027qlogicpti_mbox_command\u0027:\n drivers/scsi/qlogicpti.c:202:10: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n case 6: sbus_writew(param[5], qpti-\u003eqregs + MBOX5);\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/qlogicpti.c:203:2: note: here\n case 5: sbus_writew(param[4], qpti-\u003eqregs + MBOX4);\n ^~~~\n drivers/scsi/qlogicpti.c:203:10: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n case 5: sbus_writew(param[4], qpti-\u003eqregs + MBOX4);\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/qlogicpti.c:204:2: note: here\n case 4: sbus_writew(param[3], qpti-\u003eqregs + MBOX3);\n ^~~~\n drivers/scsi/qlogicpti.c:204:10: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n case 4: sbus_writew(param[3], qpti-\u003eqregs + MBOX3);\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/qlogicpti.c:205:2: note: here\n case 3: sbus_writew(param[2], qpti-\u003eqregs + MBOX2);\n ^~~~\n drivers/scsi/qlogicpti.c:205:10: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n case 3: sbus_writew(param[2], qpti-\u003eqregs + MBOX2);\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/qlogicpti.c:206:2: note: here\n case 2: sbus_writew(param[1], qpti-\u003eqregs + MBOX1);\n ^~~~\n drivers/scsi/qlogicpti.c:206:10: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n case 2: sbus_writew(param[1], qpti-\u003eqregs + MBOX1);\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/qlogicpti.c:207:2: note: here\n case 1: sbus_writew(param[0], qpti-\u003eqregs + MBOX0);\n ^~~~\n drivers/scsi/qlogicpti.c:256:19: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n case 6: param[5] \u003d sbus_readw(qpti-\u003eqregs + MBOX5);\n ~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/qlogicpti.c:257:2: note: here\n case 5: param[4] \u003d sbus_readw(qpti-\u003eqregs + MBOX4);\n ^~~~\n drivers/scsi/qlogicpti.c:257:19: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n case 5: param[4] \u003d sbus_readw(qpti-\u003eqregs + MBOX4);\n ~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/qlogicpti.c:258:2: note: here\n case 4: param[3] \u003d sbus_readw(qpti-\u003eqregs + MBOX3);\n ^~~~\n drivers/scsi/qlogicpti.c:258:19: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n case 4: param[3] \u003d sbus_readw(qpti-\u003eqregs + MBOX3);\n ~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/qlogicpti.c:259:2: note: here\n case 3: param[2] \u003d sbus_readw(qpti-\u003eqregs + MBOX2);\n ^~~~\n drivers/scsi/qlogicpti.c:259:19: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n case 3: param[2] \u003d sbus_readw(qpti-\u003eqregs + MBOX2);\n ~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/qlogicpti.c:260:2: note: here\n case 2: param[1] \u003d sbus_readw(qpti-\u003eqregs + MBOX1);\n ^~~~\n drivers/scsi/qlogicpti.c:260:19: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n case 2: param[1] \u003d sbus_readw(qpti-\u003eqregs + MBOX1);\n ~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/qlogicpti.c:261:2: note: here\n case 1: param[0] \u003d sbus_readw(qpti-\u003eqregs + MBOX0);\n ^~~~\n \n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 4c73598732f7954867d95e315e5465b6ce7767a0\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Sun Jul 28 19:26:08 2019 -0500\n\n scsi: ibmvfc: Mark expected switch fall-throughs\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warnings:\n \n drivers/scsi/ibmvscsi/ibmvfc.c: In function \u0027ibmvfc_npiv_login_done\u0027:\n drivers/scsi/ibmvscsi/ibmvfc.c:4022:3: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n ibmvfc_retry_host_init(vhost);\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/ibmvscsi/ibmvfc.c:4023:2: note: here\n case IBMVFC_MAD_DRIVER_FAILED:\n ^~~~\n drivers/scsi/ibmvscsi/ibmvfc.c: In function \u0027ibmvfc_bsg_request\u0027:\n drivers/scsi/ibmvscsi/ibmvfc.c:1830:11: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n port_id \u003d (bsg_request-\u003erqst_data.h_els.port_id[0] \u003c\u003c 16) |\n ~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n (bsg_request-\u003erqst_data.h_els.port_id[1] \u003c\u003c 8) |\n ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n bsg_request-\u003erqst_data.h_els.port_id[2];\n ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/ibmvscsi/ibmvfc.c:1833:2: note: here\n case FC_BSG_RPT_ELS:\n ^~~~\n drivers/scsi/ibmvscsi/ibmvfc.c:1838:11: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n port_id \u003d (bsg_request-\u003erqst_data.h_ct.port_id[0] \u003c\u003c 16) |\n ~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n (bsg_request-\u003erqst_data.h_ct.port_id[1] \u003c\u003c 8) |\n ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n bsg_request-\u003erqst_data.h_ct.port_id[2];\n ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/ibmvscsi/ibmvfc.c:1841:2: note: here\n case FC_BSG_RPT_CT:\n ^~~~\n \n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Acked-by: Tyrel Datwyler \u003ctyreld@linux.ibm.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit e5460f084b84f06756e558fbfd2313202fb1dbd0\nAuthor: Qian Cai \u003ccai@lca.pw\u003e\nDate: Fri Jul 26 10:25:43 2019 -0400\n\n scsi: megaraid_sas: Fix a compilation warning\n \n The commit de516379e85f (\"scsi: megaraid_sas: changes to function\n prototypes\") introduced a comilation warning due to it changed the function\n prototype of read_fw_status_reg() to take an instance pointer instead, but\n forgot to remove an unused variable.\n \n drivers/scsi/megaraid/megaraid_sas_fusion.c: In function\n \u0027megasas_fusion_update_can_queue\u0027:\n drivers/scsi/megaraid/megaraid_sas_fusion.c:326:39: warning: variable\n \u0027reg_set\u0027 set but not used [-Wunused-but-set-variable]\n struct megasas_register_set __iomem *reg_set;\n ^~~~~~~\n Fixes: de516379e85f (\"scsi: megaraid_sas: changes to function prototypes\")\n Signed-off-by: Qian Cai \u003ccai@lca.pw\u003e\n Acked-by: Sumit Saxena \u003csumit.saxena@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 88d5c343949ed4dc7547f307bbbb1aff3e9fa1d4\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Jul 26 21:55:40 2019 +0800\n\n scsi: megaraid_sas: Make a bunch of functions static\n \n Fix sparse warnings:\n \n drivers/scsi/megaraid/megaraid_sas_fusion.c:3369:1: warning: symbol \u0027complete_cmd_fusion\u0027 was not declared. Should it be static?\n drivers/scsi/megaraid/megaraid_sas_fusion.c:3535:6: warning: symbol \u0027megasas_sync_irqs\u0027 was not declared. Should it be static?\n drivers/scsi/megaraid/megaraid_sas_fusion.c:3554:1: warning: symbol \u0027megasas_complete_cmd_dpc_fusion\u0027 was not declared. Should it be static?\n drivers/scsi/megaraid/megaraid_sas_fusion.c:3573:13: warning: symbol \u0027megasas_isr_fusion\u0027 was not declared. Should it be static?\n drivers/scsi/megaraid/megaraid_sas_fusion.c:3604:1: warning: symbol \u0027build_mpt_mfi_pass_thru\u0027 was not declared. Should it be static?\n drivers/scsi/megaraid/megaraid_sas_fusion.c:3661:40: warning: symbol \u0027build_mpt_cmd\u0027 was not declared. Should it be static?\n drivers/scsi/megaraid/megaraid_sas_fusion.c:3688:1: warning: symbol \u0027megasas_issue_dcmd_fusion\u0027 was not declared. Should it be static?\n drivers/scsi/megaraid/megaraid_sas_fusion.c:3881:5: warning: symbol \u0027megasas_wait_for_outstanding_fusion\u0027 was not declared. Should it be static?\n drivers/scsi/megaraid/megaraid_sas_fusion.c:4005:6: warning: symbol \u0027megasas_refire_mgmt_cmd\u0027 was not declared. Should it be static?\n drivers/scsi/megaraid/megaraid_sas_fusion.c:4525:25: warning: symbol \u0027megasas_get_peer_instance\u0027 was not declared. Should it be static?\n drivers/scsi/megaraid/megaraid_sas_fusion.c:4825:7: warning: symbol \u0027megasas_fusion_crash_dump\u0027 was not declared. Should it be static?\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Acked-by: Sumit Saxena \u003csumit.saxena@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit b707659213d3c70f2c704ec950df6263b4bffe84\nAuthor: Yonghong Song \u003cyhs@fb.com\u003e\nDate: Wed Aug 7 17:38:56 2019 -0700\n\n tools/bpf: fix core_reloc.c compilation error\n \n On my local machine, I have the following compilation errors:\n \u003d\u003d\u003d\u003d\u003d\n In file included from prog_tests/core_reloc.c:3:0:\n ./progs/core_reloc_types.h:517:46: error: expected ‘\u003d’, ‘,’, ‘;’, ‘asm’ or ‘__attribute__’ before ‘fancy_char_ptr_t’\n typedef const char * const volatile restrict fancy_char_ptr_t;\n ^\n ./progs/core_reloc_types.h:527:2: error: unknown type name ‘fancy_char_ptr_t’\n fancy_char_ptr_t d;\n ^\n \u003d\u003d\u003d\u003d\u003d\n \n I am using gcc 4.8.5. Later compilers may change their behavior not emitting the\n error. Nevertheless, let us fix the issue. \"restrict\" can be tested\n without typedef.\n \n Fixes: 9654e2ae908e (\"selftests/bpf: add CO-RE relocs modifiers/typedef tests\")\n Cc: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Yonghong Song \u003cyhs@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit b684702f6e471d5e7dd80dfbb926489b87f60bd8\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Sun Aug 4 22:03:28 2019 +0530\n\n clk: ti: dm814x: Add of_node_put() to prevent memory leak\n \n In function dm814x_adpll_early_init, variable np takes the value\n returned by of_find_node_by_name, which gets a node but does not put it.\n If np is not put before return, it may cause a memory leak. Hence put np\n before return.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lkml.kernel.org/r/20190804163328.6693-1-nishkadg.linux@gmail.com\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit d432d04560301c6350158dc4a7e8275b331fd480\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Sun Aug 4 22:01:51 2019 +0530\n\n clk: st: clk-flexgen: Add of_node_put() in st_of_flexgen_setup()\n \n In function st_of_flexgen_setup, variable pnode takes the return value\n of of_get_parent, which gets a node but does not put it. If pnode is not\n put before the function returns, it may cause a memory leak. Hence put\n pnode after its last occurrence.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lkml.kernel.org/r/20190804163151.6511-1-nishkadg.linux@gmail.com\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit d55841ce1115ce08614bde66a780211f4ea12ae1\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Sun Aug 4 21:58:24 2019 +0530\n\n clk: davinci: pll: Add of_node_put() in of_davinci_pll_init()\n \n The variable child in the function of_davinci_pll_init takes the value\n of of_get_child_by_name, which gets a node but does not put it. If child\n is not put before the function returns it may cause a memory leak. Hence\n put child before two return statements.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lkml.kernel.org/r/20190804162824.6338-1-nishkadg.linux@gmail.com\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit 5911dba556cf41a14907ed04b5636cf852b17933\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Sun Aug 4 22:04:44 2019 +0530\n\n clk: versatile: Add of_node_put() in cm_osc_setup()\n \n In function cm_osc_setup, variable parent takes the value returned by\n of_get_parent, which gets a node but does not put it. If parent is not\n put before it goes out of scope, it may cause a memory leak.\n Hence put parent before the function terminates.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lkml.kernel.org/r/20190804163445.6862-1-nishkadg.linux@gmail.com\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit cc8f12996e24b102a086a253055ecc58c437c31d\nAuthor: John Keeping \u003cjohn@metanate.com\u003e\nDate: Wed Jul 3 10:51:11 2019 +0100\n\n drm/rockchip: fix VOP_WIN_GET macro\n \n Commit 9a61c54b9bff (\"drm/rockchip: vop: group vop registers\") seems to\n have unintentionally changed the defintion of this macro. Since it is\n unused, this was not spotted but any attempt to use it results in\n compilation errors.\n \n Revert to the previous definition.\n \n Fixes: 9a61c54b9bff (\"drm/rockchip: vop: group vop registers\")\n Signed-off-by: John Keeping \u003cjohn@metanate.com\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703095111.29117-1-john@metanate.com\n\ncommit 2a1d7eb854bb7737b4927f3266ba169dbb4aeae3\nAuthor: Deepak Katragadda \u003cdkatraga@codeaurora.org\u003e\nDate: Mon Jul 22 13:13:48 2019 +0530\n\n clk: qcom: gcc: Add global clock controller driver for SM8150\n \n Add the clocks supported in global clock controller which clock the\n peripherals like BLSPs, SDCC, USB, MDSS etc. Register all the clocks\n to the clock framework for the clients to be able to request for them.\n \n Signed-off-by: Deepak Katragadda \u003cdkatraga@codeaurora.org\u003e\n Signed-off-by: Taniya Das \u003ctdas@codeaurora.org\u003e\n [vkoul: port to upstream and tidy-up\n port to new parent scheme\n Add comments for critical clocks]]\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190722074348.29582-6-vkoul@kernel.org\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit e5ee331ebcf33827d1bd64e984c565b23cf53227\nAuthor: Deepak Katragadda \u003cdkatraga@codeaurora.org\u003e\nDate: Mon Jul 22 13:13:47 2019 +0530\n\n dt-bindings: clock: Document gcc bindings for SM8150\n \n Document the global clock controller found on SM8150.\n \n Signed-off-by: Deepak Katragadda \u003cdkatraga@codeaurora.org\u003e\n Signed-off-by: Taniya Das \u003ctdas@codeaurora.org\u003e\n [vkoul: port to upstream and add external clocks\n split binding to this patch]]\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190722074348.29582-5-vkoul@kernel.org\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit 548a909597d5c176f66e5ffddb25bdf8c151170e\nAuthor: Deepak Katragadda \u003cdkatraga@codeaurora.org\u003e\nDate: Mon Jul 22 13:13:46 2019 +0530\n\n clk: qcom: clk-alpha-pll: Add support for Trion PLLs\n \n Add programming sequence support for managing the Trion\n PLLs.\n \n Signed-off-by: Deepak Katragadda \u003cdkatraga@codeaurora.org\u003e\n Signed-off-by: Taniya Das \u003ctdas@codeaurora.org\u003e\n [vkoul: port to upstream and tidy-up\n use upstream way of specifying PLLs]\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190722074348.29582-4-vkoul@kernel.org\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit 4c45653b0b47a40c013362e17afd884301ad14c1\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Mon Jul 22 13:13:45 2019 +0530\n\n clk: qcom: clk-alpha-pll: Remove post_div_table checks\n \n We want users to code properly and fix the post_div_table missing and\n not rely on core to check. So remove the post_div_table check.\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Link: https://lkml.kernel.org/r/20190722074348.29582-3-vkoul@kernel.org\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit 46e625b3e320e56da830a847f718635c26af9e04\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Mon Jul 22 13:13:44 2019 +0530\n\n clk: qcom: clk-alpha-pll: Remove unnecessary cast\n \n Commit 8f9fab480c7a (\"linux/kernel.h: fix overflow for\n DIV_ROUND_UP_ULL\") fixed the overflow for DIV_ROUND_UP_ULL, so we no\n longer need the cast for DIV_ROUND_UP_ULL, so remove the unnecessary\n u64 casts.\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190722074348.29582-2-vkoul@kernel.org\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit 4d0b32b853f52760bf70288f2751d2ca86581352\nAuthor: Himanshu Madhani \u003chmadhani@marvell.com\u003e\nDate: Fri Jul 26 09:07:40 2019 -0700\n\n scsi: qla2xxx: Update driver version to 10.01.00.18-k\n \n Signed-off-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 03cc44bf682af289d6536eb911e928b415bd0e1f\nAuthor: Quinn Tran \u003cqutran@marvell.com\u003e\nDate: Fri Jul 26 09:07:39 2019 -0700\n\n scsi: qla2xxx: Allow NVMe IO to resume with short cable pull\n \n Current driver report dev_loss_tmo to 0 for NVMe devices with short cable\n pull. This causes NVMe controller to be freed along with NVMe namespace.\n The side affect is IO would stop. By not setting dev_loss_tmo to 0, NVMe\n namespace would stay until cable is plugged back in. This allows IO to\n resume afterward.\n \n [mkp: commit desc]\n \n Signed-off-by: Arun Easi \u003caeasi@marvell.com\u003e\n Signed-off-by: Quinn Tran \u003cqutran@marvell.com\u003e\n Signed-off-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit f00b3428a801758243693e046b34226e92bc56b3\nAuthor: Quinn Tran \u003cqutran@marvell.com\u003e\nDate: Fri Jul 26 09:07:38 2019 -0700\n\n scsi: qla2xxx: Fix hang in fcport delete path\n \n A hang was observed in the fcport delete path when the device was\n responding slow and an issue-lip path (results in session termination) was\n taken.\n \n Fix this by issuing logo requests unconditionally.\n \n PID: 19491 TASK: ffff8e23e67bb150 CPU: 0 COMMAND: \"kworker/0:0\"\n #0 [ffff8e2370297bf8] __schedule at ffffffffb4f7dbb0\n #1 [ffff8e2370297c88] schedule at ffffffffb4f7e199\n #2 [ffff8e2370297c98] schedule_timeout at ffffffffb4f7ba68\n #3 [ffff8e2370297d40] msleep at ffffffffb48ad9ff\n #4 [ffff8e2370297d58] qlt_free_session_done at ffffffffc0c32052 [qla2xxx]\n #5 [ffff8e2370297e20] process_one_work at ffffffffb48bcfdf\n #6 [ffff8e2370297e68] worker_thread at ffffffffb48bdca6\n #7 [ffff8e2370297ec8] kthread at ffffffffb48c4f81\n \n Signed-off-by: Quinn Tran \u003cqutran@marvell.com\u003e\n Signed-off-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit a36f1443e6fc738c1bcfc4be80d6f1609163c614\nAuthor: Andrew Vasquez \u003candrewv@marvell.com\u003e\nDate: Fri Jul 26 09:07:37 2019 -0700\n\n scsi: qla2xxx: Use common update-firmware-options routine for ISP27xx+\n \n Leverage the generic routine, qla24xx_update_fw_options(), for the\n configuration of firmware options for ISP27xx/ISP28xx.\n \n Signed-off-by: Andrew Vasquez \u003candrewv@marvell.com\u003e\n Signed-off-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 9e744591ef1b8df27c25c68dac858dada8688f77\nAuthor: Arun Easi \u003caeasi@marvell.com\u003e\nDate: Fri Jul 26 09:07:36 2019 -0700\n\n scsi: qla2xxx: Fix NVMe port discovery after a short device port loss\n \n The following sequence of event leads to NVME port disappearing:\n \n - device port shut\n - nvme_fc_unregister_remoteport\n - device port online\n - remote port delete completes\n - relogin is scheduled\n - \"post gidpn\" message appears due to rscn generation # mismatch\n \n In short, if a device comes back online sooner than an unregister\n completion, a mismatch in rscn generation number occurs, which is not\n handled correctly during device relogin. Fix this by starting with a redo\n of GNL.\n \n When ql2xextended_error_logging is enabled, the re-plugged device\u0027s\n discovery stops with the following messages printed:\n \n --8\u003c--\n qla2xxx [0000:41:00.0]-480d:3: Relogin scheduled.\n qla2xxx [0000:41:00.0]-4800:3: DPC handler sleeping.\n qla2xxx [0000:41:00.0]-2902:3: qla24xx_handle_relogin_event 21:00:00:24:ff:17:9e:91 DS 0 LS 7 P 0 del 2 cnfl\n (null) rscn 1|2 login 1|2 fl 1\n qla2xxx [0000:41:00.0]-28e9:3: qla24xx_handle_relogin_event 1666 21:00:00:24:ff:17:9e:91 post gidpn\n qla2xxx [0000:41:00.0]-480e:3: Relogin end.\n --8\u003c--\n \n Signed-off-by: Arun Easi \u003caeasi@marvell.com\u003e\n Signed-off-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 26a77799195f4ff105f877042012c7fb355b3da1\nAuthor: Andrew Vasquez \u003candrewv@marvell.com\u003e\nDate: Fri Jul 26 09:07:35 2019 -0700\n\n scsi: qla2xxx: Correct error handling during initialization failures\n \n Current code misses or fails to account for proper recovery during early\n initialization failures:\n \n - Properly unwind allocations during probe() failures.\n \n - Protect against non-initialization memory allocations during\n unwinding.\n \n - Propagate error status during HW initialization.\n \n - Release SCSI host reference when memory allocations fail.\n \n Signed-off-by: Andrew Vasquez \u003candrewv@marvell.com\u003e\n Signed-off-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit f0cecc1eec0c24e790a4daac71858f1f14e6f7b5\nAuthor: Quinn Tran \u003cqutran@marvell.com\u003e\nDate: Fri Jul 26 09:07:34 2019 -0700\n\n scsi: qla2xxx: Retry fabric Scan on IOCB queue full\n \n when fabric scan thread encounters IOCB Q Full, schedule a delayed work to\n retry fabric scan.\n \n Signed-off-by: Quinn Tran \u003cqutran@marvell.com\u003e\n Signed-off-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 3a4b6cc7332130ac5cbf3b505d8cddf0aa2ea745\nAuthor: Quinn Tran \u003cqutran@marvell.com\u003e\nDate: Fri Jul 26 09:07:33 2019 -0700\n\n scsi: qla2xxx: Fix premature timer expiration\n \n For any qla2xxx async command, the SRB buffer is used to send it. In\n setting up the SRB buffer, the timer for this command is started before all\n memory allocation has finished. Under low memory pressure, memory alloc\n can go to sleep and not wake up before the timer expires. Once timer has\n expired, the timer thread will access uninitialize fields resulting into\n NULL pointer crash.\n \n This patch fixes this crash by moving the start of timer after everything\n is setup.\n \n backtrace shows following\n \n PID: 3720 TASK: ffff996928401040 CPU: 0 COMMAND: \"qla2xxx_1_dpc\"\n 0 [ffff99652751b698] __schedule at ffffffff965676c7\n 1 [ffff99652751b728] schedule at ffffffff96567bc9\n 2 [ffff99652751b738] schedule_timeout at ffffffff965655e8\n 3 [ffff99652751b7e0] io_schedule_timeout at ffffffff9656726d\n 4 [ffff99652751b810] congestion_wait at ffffffff95fd8d12\n 5 [ffff99652751b870] isolate_migratepages_range at ffffffff95fddaf3\n 6 [ffff99652751b930] compact_zone at ffffffff95fdde96\n 7 [ffff99652751b980] compact_zone_order at ffffffff95fde0bc\n 8 [ffff99652751ba20] try_to_compact_pages at ffffffff95fde481\n 9 [ffff99652751ba80] __alloc_pages_direct_compact at ffffffff9655cc31\n 10 [ffff99652751bae0] __alloc_pages_slowpath at ffffffff9655d101\n 11 [ffff99652751bbd0] __alloc_pages_nodemask at ffffffff95fc0e95\n 12 [ffff99652751bc80] dma_generic_alloc_coherent at ffffffff95e3217f\n 13 [ffff99652751bcc8] x86_swiotlb_alloc_coherent at ffffffff95e6b7a1\n 14 [ffff99652751bcf8] qla2x00_rft_id at ffffffffc055b5e0 [qla2xxx]\n 15 [ffff99652751bd50] qla2x00_loop_resync at ffffffffc0533e71 [qla2xxx]\n 16 [ffff99652751be68] qla2x00_do_dpc at ffffffffc05210ca [qla2xxx]\n \n PID: 0 TASK: ffffffff96a18480 CPU: 0 COMMAND: \"swapper/0\"\n 0 [ffff99652fc03ae0] machine_kexec at ffffffff95e63674\n 1 [ffff99652fc03b40] __crash_kexec at ffffffff95f1ce12\n 2 [ffff99652fc03c10] crash_kexec at ffffffff95f1cf00\n 3 [ffff99652fc03c28] oops_end at ffffffff9656c758\n 4 [ffff99652fc03c50] no_context at ffffffff9655aa7e\n 5 [ffff99652fc03ca0] __bad_area_nosemaphore at ffffffff9655ab15\n 6 [ffff99652fc03cf0] bad_area_nosemaphore at ffffffff9655ac86\n 7 [ffff99652fc03d00] __do_page_fault at ffffffff9656f6b0\n 8 [ffff99652fc03d70] do_page_fault at ffffffff9656f915\n 9 [ffff99652fc03da0] page_fault at ffffffff9656b758\n [exception RIP: unknown or invalid address]\n RIP: 0000000000000000 RSP: ffff99652fc03e50 RFLAGS: 00010202\n RAX: 0000000000000000 RBX: ffff99652b79a600 RCX: ffff99652b79a760\n RDX: ffff99652b79a600 RSI: ffffffffc0525ad0 RDI: ffff99652b79a600\n RBP: ffff99652fc03e60 R8: ffffffff96a18a18 R9: ffffffff96ee3c00\n R10: 0000000000000002 R11: ffff99652fc03de8 R12: ffff99652b79a760\n R13: 0000000000000100 R14: ffffffffc0525ad0 R15: ffff99652b79a600\n ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0018\n 10 [ffff99652fc03e50] qla2x00_sp_timeout at ffffffffc0525af8 [qla2xxx]\n 11 [ffff99652fc03e68] call_timer_fn at ffffffff95ea7f58\n 12 [ffff99652fc03ea0] run_timer_softirq at ffffffff95eaa3bd\n 13 [ffff99652fc03f18] __do_softirq at ffffffff95ea0f05\n 14 [ffff99652fc03f88] call_softirq at ffffffff9657832c\n 15 [ffff99652fc03fa0] do_softirq at ffffffff95e2e675\n 16 [ffff99652fc03fc0] irq_exit at ffffffff95ea1285\n 17 [ffff99652fc03fd8] smp_apic_timer_interrupt at ffffffff965796c8\n 18 [ffff99652fc03ff0] apic_timer_interrupt at ffffffff96575df2\n \n Signed-off-by: Quinn Tran \u003cqutran@marvell.com\u003e\n Signed-off-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 8b5292bcfcacf15182a77a973a98d310e76fd58b\nAuthor: Quinn Tran \u003cqutran@marvell.com\u003e\nDate: Fri Jul 26 09:07:32 2019 -0700\n\n scsi: qla2xxx: Fix Relogin to prevent modifying scan_state flag\n \n Relogin fails to move forward due to scan_state flag indicating device is\n not there. Before relogin process, Session delete process accidently\n modified the scan_state flag.\n \n [mkp: typos plus corrected Fixes: sha as reported by sfr]\n \n Fixes: 2dee5521028c (\"scsi: qla2xxx: Fix login state machine freeze\")\n Cc: stable@vger.kernel.org\n Signed-off-by: Quinn Tran \u003cqutran@marvell.com\u003e\n Signed-off-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 29e1c66872450adba0ad552ff6019932168676f3\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Wed Aug 7 14:40:01 2019 -0700\n\n selftests/bpf: add CO-RE relocs misc tests\n \n Add tests validating few edge-cases of capturing offset relocations.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit c1f5e7dd19e71cd3607572bb957def618a33519a\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Wed Aug 7 14:40:00 2019 -0700\n\n selftests/bpf: add CO-RE relocs ints tests\n \n Add various tests validating handling compatible/incompatible integer\n types.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit d698f9dbdbed036ef28a96cd34a1b5d7fe58750e\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Wed Aug 7 14:39:59 2019 -0700\n\n selftests/bpf: add CO-RE relocs ptr-as-array tests\n \n Add test validating correct relocation handling for cases where pointer\n to something is used as an array. E.g.:\n \n int *ptr \u003d ...;\n int x \u003d ptr[42];\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 9654e2ae908eb0d51b0b79c7c50df0754ed38edd\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Wed Aug 7 14:39:58 2019 -0700\n\n selftests/bpf: add CO-RE relocs modifiers/typedef tests\n \n Add tests validating correct handling of various combinations of\n typedefs and const/volatile/restrict modifiers.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit d9db3550300f4390e457c79189e2601c107f9fe6\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Wed Aug 7 14:39:57 2019 -0700\n\n selftests/bpf: add CO-RE relocs enum/ptr/func_proto tests\n \n Test CO-RE relocation handling of ints, enums, pointers, func protos, etc.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 20a9ad2e71368da8e831317f0a545e4bfb31cce1\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Wed Aug 7 14:39:56 2019 -0700\n\n selftests/bpf: add CO-RE relocs array tests\n \n Add tests for various array handling/relocation scenarios.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit ec6438a988a43dcb03f0a04f3f51a48aba54764a\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Wed Aug 7 14:39:55 2019 -0700\n\n selftests/bpf: add CO-RE relocs nesting tests\n \n Add a bunch of test validating correct handling of nested\n structs/unions.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 002d3afce65518dc5dfc398a37c2be2a6bf559c4\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Wed Aug 7 14:39:54 2019 -0700\n\n selftests/bpf: add CO-RE relocs struct flavors tests\n \n Add tests verifying that BPF program can use various struct/union\n \"flavors\" to extract data from the same target struct/union.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit df36e621418b0de4d4afec5f8002d45ee636bb5c\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Wed Aug 7 14:39:53 2019 -0700\n\n selftests/bpf: add CO-RE relocs testing setup\n \n Add CO-RE relocation test runner. Add one simple test validating that\n libbpf\u0027s logic for searching for kernel image and loading BTF out of it\n works.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 2dc26d5a4f2e97364542030a4a51e4a50c14bae3\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Wed Aug 7 14:39:52 2019 -0700\n\n selftests/bpf: add BPF_CORE_READ relocatable read macro\n \n Add BPF_CORE_READ macro used in tests to do bpf_core_read(), which\n automatically captures offset relocation.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit ddc7c3042614e273044f698d2beab25cc3842d45\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Wed Aug 7 14:39:51 2019 -0700\n\n libbpf: implement BPF CO-RE offset relocation algorithm\n \n This patch implements the core logic for BPF CO-RE offsets relocations.\n Every instruction that needs to be relocated has corresponding\n bpf_offset_reloc as part of BTF.ext. Relocations are performed by trying\n to match recorded \"local\" relocation spec against potentially many\n compatible \"target\" types, creating corresponding spec. Details of the\n algorithm are noted in corresponding comments in the code.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 4cedc0dad9b5bf55c4180c833be35e27e5d6cdbb\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Wed Aug 7 14:39:50 2019 -0700\n\n libbpf: add .BTF.ext offset relocation section loading\n \n Add support for BPF CO-RE offset relocations. Add section/record\n iteration macros for .BTF.ext. These macro are useful for iterating over\n each .BTF.ext record, either for dumping out contents or later for BPF\n CO-RE relocation handling.\n \n To enable other parts of libbpf to work with .BTF.ext contents, moved\n a bunch of type definitions into libbpf_internal.h.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit b03bc6853c0e0c97da842434e8056f1b9d9a1f4a\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Wed Aug 7 14:39:49 2019 -0700\n\n libbpf: convert libbpf code to use new btf helpers\n \n Simplify code by relying on newly added BTF helper functions.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit ef20a9b27c66278ac2f85006db8ea11d5f61a781\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Wed Aug 7 14:39:48 2019 -0700\n\n libbpf: add helpers for working with BTF types\n \n Add lots of frequently used helpers that simplify working with BTF\n types.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 568b9de48d80bcf1a92e2c4fa67651abbb8ebfe2\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Mon Jul 1 13:36:06 2019 +0200\n\n clk: ingenic/jz4740: Fix \"pll half\" divider not read/written properly\n \n The code was setting the bit 21 of the CPCCR register to use a divider\n of 2 for the \"pll half\" clock, and clearing the bit to use a divider\n of 1.\n \n This is the opposite of how this register field works: a cleared bit\n means that the /2 divider is used, and a set bit means that the divider\n is 1.\n \n Restore the correct behaviour using the newly introduced .div_table\n field.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Link: https://lkml.kernel.org/r/20190701113606.4130-1-paul@crapouillou.net\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit e2f2402f3be89e4407d94d92d326204f70261754\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Mon Jul 1 17:50:20 2019 +0100\n\n clk: Si5341/Si5340: remove redundant assignment to n_den\n \n The variable n_den is initialized however that value is never read\n as n_den is re-assigned a little later in the two paths of a\n following if-statement. Remove the redundant assignment.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Link: https://lkml.kernel.org/r/20190701165020.19840-1-colin.king@canonical.com\n Acked-by: Mike Looijmans \u003cmike.looijmans@topic.nl\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit a95fb581b144b5e73da382eaedb2e32027610597\nAuthor: Nathan Huckleberry \u003cnhuck@google.com\u003e\nDate: Thu Jun 27 15:06:42 2019 -0700\n\n clk: qoriq: Fix -Wunused-const-variable\n \n drivers/clk/clk-qoriq.c:138:38: warning: unused variable\n \u0027p5020_cmux_grp1\u0027 [-Wunused-const-variable] static const struct\n clockgen_muxinfo p5020_cmux_grp1\n \n drivers/clk/clk-qoriq.c:146:38: warning: unused variable\n \u0027p5020_cmux_grp2\u0027 [-Wunused-const-variable] static const struct\n clockgen_muxinfo p5020_cmux_grp2\n \n In the definition of the p5020 chip, the p2041 chip\u0027s info was used\n instead. The p5020 and p2041 chips have different info. This is most\n likely a typo.\n \n Link: https://github.com/ClangBuiltLinux/linux/issues/525\n Cc: clang-built-linux@googlegroups.com\n Signed-off-by: Nathan Huckleberry \u003cnhuck@google.com\u003e\n Link: https://lkml.kernel.org/r/20190627220642.78575-1-nhuck@google.com\n Reviewed-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Acked-by: Scott Wood \u003coss@buserror.net\u003e\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit ebd5f82d32ade6f864917bf868bfa32f4d3c0486\nAuthor: Joel Stanley \u003cjoel@jms.id.au\u003e\nDate: Wed Jul 10 23:40:09 2019 +0930\n\n clk: aspeed: Add SDIO gate\n \n The clock divisor comes with an enable bit (gate). This was not\n implemented as we didn\u0027t have access to SD hardware when writing the\n driver. Now that we can test it, add the gate as a parent to the\n divisor.\n \n There is no reason to expose the gate separately, so users will enable\n it by turning on the ASPEED_CLK_SDIO divisor.\n \n Signed-off-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n [aj: Minor style cleanup]\n Signed-off-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Link: https://lkml.kernel.org/r/20190710141009.20651-1-andrew@aj.id.au\n Signed-off-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n\ncommit 5e0a809af2a2b577934ec54c0e50897b806f0732\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Wed Aug 7 18:37:59 2019 +0000\n\n drm/i915/uc: Hardening firmware fetch\n \n Insert few more failure points into firmware fetch procedure to check\n use of the wrong blob name or use of the mismatched firmware versions.\n \n Also update some messages (remove ptr, duplicated infos) and stop\n treating all fetch errors as missing firmware case.\n \n v2: update log levels (Chris)\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n [ickle: fixup compiler warning for non-debug builds]\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807183759.8588-1-michal.wajdeczko@intel.com\n\ncommit 97299a3035328d7ae2f4fccaf6e549974df6e118\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 15:15:25 2019 -0300\n\n media: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 25a3d6bac6b9ff7d62033185780daa1b4f7bee2f\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Aug 7 05:43:32 2019 -0300\n\n media: adv7511/cobalt: rename driver name to adv7511-v4l2\n \n Commit b2ce5617dad2 (\"media: i2c: fix warning same module names\")\n renamed the adv7511 module in the media tree to adv7511-v4l2.\n \n This patch does the same rename for the driver name and device id to\n keep the naming consistent.\n \n Since the cobalt driver loads this module, it had to be renamed there\n as well.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 6898dd580a045341f844862ceb775144156ec1af\nAuthor: Randy Dunlap \u003crdunlap@infradead.org\u003e\nDate: Tue Aug 6 19:01:36 2019 -0300\n\n media: media/platform: fsl-viu.c: fix build for MICROBLAZE\n \n arch/microblaze/ defines out_be32() and in_be32(), so don\u0027t do that\n again in the driver source.\n \n Fixes these build warnings:\n \n ../drivers/media/platform/fsl-viu.c:36: warning: \"out_be32\" redefined\n ../arch/microblaze/include/asm/io.h:50: note: this is the location of the previous definition\n ../drivers/media/platform/fsl-viu.c:37: warning: \"in_be32\" redefined\n ../arch/microblaze/include/asm/io.h:53: note: this is the location of the previous definition\n \n Fixes: 29d750686331 (\"media: fsl-viu: allow building it with COMPILE_TEST\")\n Signed-off-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 8fb294ca20886a723007c543a65fcdf286ec667f\nAuthor: Fabio Estevam \u003cfestevam@gmail.com\u003e\nDate: Thu Aug 1 14:56:11 2019 -0300\n\n media: imx7.rst: Fix the references to the CSI multiplexer\n \n In imx7s.dtsi the node name for the CSI multiplexer is \"csi-mux\", not\n \"csi_mux\", so fix all the references in the document.\n \n This fixes the following error when the instructions are followed:\n \n Unable to parse link: Invalid argument (22)\n \n While at it, provide the \"media-ctl -p\" output from 5.2 kernel\n version, so that users can see a more updated output.\n \n Fixes: fa88fbdafb4a (\"media: imx7.rst: add documentation for i.MX7 media driver\")\n Signed-off-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Reviewed-by: Rui Miguel Silva \u003crmfrfs@gmail.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 00ab9c1a620dc29f3651d5b61ad12aa327d4dd12\nAuthor: Rui Wang \u003cgtk_ruiwang@mediatek.com\u003e\nDate: Wed Jul 31 01:29:39 2019 -0300\n\n media: mtk-vcodec: Handle H264 error bitstreams\n \n Error h264 bitstreams which picture info are out range of\n decoder hardware specification, and no nal start code at the\n beginning of the buffer, stop decoding and exit.\n \n Signed-off-by: Rui Wang \u003cgtk_ruiwang@mediatek.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit b40dc2bf3581748aa772ad2690bd6092b1fca8da\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Tue Jul 30 10:00:20 2019 -0300\n\n media: vicodec: add support for 4 new RGB32 pixelformats\n \n Add support for V4L2_PIX_FMT_BGRA/X32 and V4L2_PIX_FMT_RGBA/X32 formats.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 6b2265975239ab655069d796b822835a593e1cc7\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Jul 24 23:41:29 2019 -0300\n\n media: staging: tegra-vde: Fix build error\n \n If IOMMU_SUPPORT is not set, and COMPILE_TEST is y,\n IOMMU_IOVA may be set to m. So building will fails:\n \n drivers/staging/media/tegra-vde/iommu.o: In function `tegra_vde_iommu_map\u0027:\n iommu.c:(.text+0x41): undefined reference to `alloc_iova\u0027\n iommu.c:(.text+0x56): undefined reference to `__free_iova\u0027\n \n Select IOMMU_IOVA while COMPILE_TEST is set to fix this.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Suggested-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Fixes: b301f8de1925 (\"media: staging: media: tegra-vde: Add IOMMU support\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Acked-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit a0752c87080ad72e838489a5807b791a117fcb11\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Wed Aug 7 17:00:33 2019 +0000\n\n drm/i915/uc: WOPCM programming errors are not always real\n \n WOPCM programming error might be due to inserted earlier probe\n failure that could affects HuC firmware loading and thus impacts\n result of WOPCM partitioning that would be now incompatible with\n previously programmed values.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807170034.8440-7-michal.wajdeczko@intel.com\n\ncommit d4b2cfb8c29f7447d57d0d836d9e8331fa0513aa\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Wed Aug 7 17:00:32 2019 +0000\n\n drm/i915: Make wopcm_to_i915() private\n \n No need to define it globally as we\u0027re only using it in wopcm.c\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807170034.8440-6-michal.wajdeczko@intel.com\n\ncommit 7dbd03e289043f8d3484810cc694cd9d7a003740\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Wed Aug 7 17:00:31 2019 +0000\n\n drm/i915: Don\u0027t try to partition WOPCM without GuC firmware\n \n For meaningful WOPCM partitioning we need GuC (and optionally HuC)\n firmware size(s) and we shouldn\u0027t just rely on GuC support flag,\n as we might fail to fetch GuC firmware and it\u0027s size will be 0\n and all calculations will be just wrong/useless.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807170034.8440-5-michal.wajdeczko@intel.com\n\ncommit 4157a962f59d0261030355aaa1646f55c2bed8c9\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Wed Aug 7 17:00:30 2019 +0000\n\n drm/i915/uc: Don\u0027t fetch HuC fw if GuC fw fetch already failed\n \n When we failed to fetch GuC firmware there is no point in fetching\n HuC firmware as we will not be able to use it without working GuC.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807170034.8440-4-michal.wajdeczko@intel.com\n\ncommit 2399ab52da5fe1c03c556e08d5e964afdf571e1b\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Wed Aug 7 17:00:29 2019 +0000\n\n drm/i915/uc: HuC firmware can\u0027t be supported without GuC\n \n There is no point in selecting HuC firmware if GuC is unsupported\n or it was already disabled, as we need GuC to authenticate HuC.\n \n While around, make uc_fw_init_early work without direct access\n to whole i915, pass only needed platform/rev info.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807170034.8440-3-michal.wajdeczko@intel.com\n\ncommit 12be4a4c75a2a41ba32b4b1368e9f40e43bfda0a\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Wed Aug 7 17:00:28 2019 +0000\n\n drm/i915/uc: Prefer dev_info for reporting options\n \n While modparams are global for the i915 module, we are reporting\n status of the params applied against specific device instance.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807170034.8440-2-michal.wajdeczko@intel.com\n\ncommit a37f08a882b01a6e86a07512a082b14d20ee0773\nAuthor: Umesh Nerlige Ramappa \u003cumesh.nerlige.ramappa@intel.com\u003e\nDate: Tue Aug 6 16:30:02 2019 -0700\n\n drm/i915/perf: Refactor oa object to better manage resources\n \n The oa object manages the oa buffer and must be allocated when the user\n intends to read performance counter snapshots. This can be achieved by\n making the oa object part of the stream object which is allocated when a\n stream is opened by the user.\n \n Attributes in the oa object that are gen-specific are moved to the perf\n object so that they can be initialized on driver load.\n \n The split provides a better separation of the objects used in perf\n implementation of i915 driver so that resources are allocated and\n initialized only when needed.\n \n v2: Fix checkpatch warnings\n v3: Addressed Lionel\u0027s review comment\n v4: Rebase\n v5: Fix rebase/merge issue with ratelimit_state_init\n \n Signed-off-by: Umesh Nerlige Ramappa \u003cumesh.nerlige.ramappa@intel.com\u003e\n Reviewed-by: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190806233002.984-1-umesh.nerlige.ramappa@intel.com\n\ncommit 653a23ca7e1e1ae907654e91d028bc6dfc7fce0c\nAuthor: Marc Koderer \u003cmarc@koderer.com\u003e\nDate: Tue Aug 6 15:24:12 2019 +0200\n\n Use kvmalloc in cgroups-v1\n \n Instead of using its own logic for k-/vmalloc rely on\n kvmalloc which is actually doing quite the same.\n \n Signed-off-by: Marc Koderer \u003cmarc@koderer.com\u003e\n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n\ncommit 10971e2ab6eb86c490fea92eb67549a8f46f0ae9\nAuthor: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\nDate: Mon Aug 5 18:27:10 2019 +0200\n\n pinctrl: samsung: Fix device node refcount leaks in init code\n \n Several functions use for_each_child_of_node() loop with a break to find\n a matching child node. Although each iteration of\n for_each_child_of_node puts the previous node, but early exit from loop\n misses it. This leads to leak of device node.\n \n Cc: \u003cstable@vger.kernel.org\u003e\n Fixes: 9a2c1c3b91aa (\"pinctrl: samsung: Allow grouping multiple pinmux/pinconf nodes\")\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 295f9b84ea9f6bd9d9ec6f6c1891d5c6897ba101\nAuthor: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\nDate: Mon Aug 5 18:27:09 2019 +0200\n\n pinctrl: samsung: Fix device node refcount leaks in S3C64xx wakeup controller init\n \n In s3c64xx_eint_eint0_init() the for_each_child_of_node() loop is used\n with a break to find a matching child node. Although each iteration of\n for_each_child_of_node puts the previous node, but early exit from loop\n misses it. This leads to leak of device node.\n \n Cc: \u003cstable@vger.kernel.org\u003e\n Fixes: 61dd72613177 (\"pinctrl: Add pinctrl-s3c64xx driver\")\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 68c9106bc0466c1273fc853967d3136f4af2e585\nAuthor: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\nDate: Mon Aug 5 18:27:08 2019 +0200\n\n pinctrl: samsung: Fix device node refcount leaks in S3C24xx wakeup controller init\n \n In s3c24xx_eint_init() the for_each_child_of_node() loop is used with a\n break to find a matching child node. Although each iteration of\n for_each_child_of_node puts the previous node, but early exit from loop\n misses it. This leads to leak of device node.\n \n Cc: \u003cstable@vger.kernel.org\u003e\n Fixes: af99a7507469 (\"pinctrl: Add pinctrl-s3c24xx driver\")\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit a969525f75730d8bd7b7c7d6deb2626610d10518\nAuthor: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\nDate: Mon Aug 5 18:27:07 2019 +0200\n\n pinctrl: samsung: Fix device node refcount leaks in Exynos wakeup controller init\n \n In exynos_eint_wkup_init() the for_each_child_of_node() loop is used\n with a break to find a matching child node. Although each iteration of\n for_each_child_of_node puts the previous node, but early exit from loop\n misses it. This leads to leak of device node.\n \n Cc: \u003cstable@vger.kernel.org\u003e\n Fixes: 43b169db1841 (\"pinctrl: add exynos4210 specific extensions for samsung pinctrl driver\")\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit ca343167b8c5b5dfd5dcdbb72472523d3c0e46be\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Sun Aug 4 21:32:00 2019 +0530\n\n pinctrl: samsung: Add of_node_put() before return in error path\n \n Each iteration of for_each_child_of_node puts the previous node, but in\n the case of a return from the middle of the loop, there is no put, thus\n causing a memory leak. Hence add an of_node_put before the return of\n exynos_eint_wkup_init() error path.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Cc: \u003cstable@vger.kernel.org\u003e\n Fixes: 14c255d35b25 (\"pinctrl: exynos: Add irq_chip instance for Exynos7 wakeup interrupts\")\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 94f3e14e00fd43024b1c4d8e0c1e442db9b4d964\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Aug 6 09:59:50 2019 +0800\n\n mlx5: Use refcount_t for refcount\n \n Reference counters are preferred to use refcount_t instead of\n atomic_t.\n This is because the implementation of refcount_t can prevent\n overflows and detect possible use-after-free.\n So convert atomic_t ref counters to refcount_t.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Acked-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 9c240a7bb33787d4fed56508425ae0b2936b1674\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 6 19:05:52 2019 +0300\n\n mm/hmm: make HMM_MIRROR an implicit option\n \n Make HMM_MIRROR an option that is selected by drivers wanting to use it\n instead of a user visible option as it is just a low-level implementation\n detail.\n \n Link: https://lore.kernel.org/r/20190806160554.14046-15-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit f442c283efe4bda53c8e52b230159d020ea11c8e\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 6 19:05:51 2019 +0300\n\n mm/hmm: allow HMM_MIRROR on all architectures with MMU\n \n There isn\u0027t really any architecture specific code in this page table walk\n implementation, so drop the dependencies.\n \n Link: https://lore.kernel.org/r/20190806160554.14046-14-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 251bbe59b7a6362f4c417aa8af281d44f9edfa0e\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 6 19:05:50 2019 +0300\n\n mm/hmm: cleanup the hmm_vma_walk_hugetlb_entry stub\n \n Stub out the whole function and assign NULL to the .hugetlb_entry method\n if CONFIG_HUGETLB_PAGE is not set, as the method won\u0027t ever be called in\n that case.\n \n Link: https://lore.kernel.org/r/20190806160554.14046-13-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 9d3973d60f0abd3985229bd895f45e2b8974344c\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 6 19:05:49 2019 +0300\n\n mm/hmm: cleanup the hmm_vma_handle_pmd stub\n \n Stub out the whole function when CONFIG_TRANSPARENT_HUGEPAGE is not set to\n make the function easier to read.\n \n Link: https://lore.kernel.org/r/20190806160554.14046-12-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit f0b3c45c8931fd7448a638557752f2743f76f51a\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 6 19:05:48 2019 +0300\n\n mm/hmm: only define hmm_vma_walk_pud if needed\n \n We only need the special pud_entry walker if PUD-sized hugepages and pte\n mappings are supported, else the common pagewalk code will take care of\n the iteration. Not implementing this callback reduced the amount of code\n compiled for non-x86 platforms, and also fixes compile failures with other\n architectures when helpers like pud_pfn are not implemented.\n \n Link: https://lore.kernel.org/r/20190806160554.14046-11-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 309f9a4f5e1a233d5df2101b9394ee689d9e463f\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 6 19:05:47 2019 +0300\n\n mm/hmm: don\u0027t abuse pte_index() in hmm_vma_handle_pmd\n \n pte_index is an internal arch helper in various architectures, without\n consistent semantics. Open code that calculation of a PMD index based on\n the virtual address instead.\n \n Link: https://lore.kernel.org/r/20190806160554.14046-10-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 05c23af4a1b34df5838ebab2da1d6f802cf5ece3\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 6 19:05:46 2019 +0300\n\n mm/hmm: remove the mask variable in hmm_vma_walk_hugetlb_entry\n \n The pagewalk code already passes the value as the hmask parameter.\n \n Link: https://lore.kernel.org/r/20190806160554.14046-9-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 7f08263d9bc6627382da14f9e81d643d0329d5d1\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 6 19:05:45 2019 +0300\n\n mm/hmm: remove the page_shift member from struct hmm_range\n \n All users pass PAGE_SIZE here, and if we wanted to support single entries\n for huge pages we should really just add a HMM_FAULT_HUGEPAGE flag instead\n that uses the huge page size instead of having the caller calculate that\n size once, just for the hmm code to verify it.\n \n Link: https://lore.kernel.org/r/20190806160554.14046-8-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Acked-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit fac555ac93d453a0d2265eef88bf4c249dd63e07\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 6 19:05:44 2019 +0300\n\n mm/hmm: remove superfluous arguments from hmm_range_register\n \n The start, end and page_shift values are all saved in the range structure,\n so we might as well use that for argument passing.\n \n Link: https://lore.kernel.org/r/20190806160554.14046-7-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Reviewed-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 2cbeb41913e639890bf2c6485d7bdc6a25fdfd56\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 6 19:05:43 2019 +0300\n\n mm/hmm: remove the unused vma argument to hmm_range_dma_unmap\n \n Link: https://lore.kernel.org/r/20190806160554.14046-6-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 5aa0acb374edd3c5001328276d34ac86f14a3867\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 6 19:05:41 2019 +0300\n\n nouveau: pass struct nouveau_svmm to nouveau_range_fault\n \n We\u0027ll need the nouveau_svmm structure to improve the function soon. For\n now this allows using the svmm-\u003emm reference to unlock the mmap_sem, and\n thus the same dereference chain that the caller uses to lock and unlock\n it.\n \n Link: https://lore.kernel.org/r/20190806160554.14046-4-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 07d82211b8d277d409836264075c57a21af3c023\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 6 19:05:40 2019 +0300\n\n amdgpu: don\u0027t initialize range-\u003elist in amdgpu_hmm_init_range\n \n The list is used to add the range to another list as an entry in the core\n hmm code, and intended as a private member not exposed to drivers. There\n is no need to initialize it in a driver.\n \n Link: https://lore.kernel.org/r/20190806160554.14046-3-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Reviewed-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 9d0a16658f840335325508e06a797be1883e6c56\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Aug 6 19:05:39 2019 +0300\n\n amdgpu: remove -EAGAIN handling for hmm_range_fault\n \n hmm_range_fault can only return -EAGAIN if called with the\n HMM_FAULT_ALLOW_RETRY flag, which amdgpu never does. Remove the handling\n for the -EAGAIN case with its non-standard locking scheme.\n \n Link: https://lore.kernel.org/r/20190806160554.14046-2-hch@lst.de\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Reviewed-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 91a49470c29cc0abea8e8481a2f2744a59f49caf\nAuthor: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\nDate: Fri Aug 2 12:18:22 2019 +0200\n\n ARM: exynos_defconfig: Enable AHCI-platform SATA driver\n \n Exynos5250-based Arndale board has one eSATA port, so enable AHCI-platform\n driver, which handles it.\n \n Signed-off-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 1e1c735a6ba0698a107b99979aad72f48392a441\nAuthor: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\nDate: Fri Aug 2 12:18:21 2019 +0200\n\n ARM: dts: exynos: Add port map to Exynos5250 AHCI node\n \n Exynos AHCI (SATA) controller has only one port for SATA device. According\n to AHCI driver bindings (ata/ahci-platform.txt), if the bootloader doesn\u0027t\n program the PORTS_IMPL register to proper value, the available port map has\n to be provided by \u0027ports-implemented\u0027 device tree property. This fixes\n SATA operation on Exynos5250-based boards since Linux v4.5.\n \n Signed-off-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 449808a254fd567d3dbeb11595a0af238c687c82\nAuthor: Hou Tao \u003choutao1@huawei.com\u003e\nDate: Sat Jul 27 14:02:58 2019 +0800\n\n raid1: factor out a common routine to handle the completion of sync write\n \n It\u0027s just code clean-up.\n \n Signed-off-by: Hou Tao \u003choutao1@huawei.com\u003e\n Signed-off-by: Song Liu \u003csongliubraving@fb.com\u003e\n\ncommit 0d8ed0e9bf9643f27f4816dca61081784dedb38d\nAuthor: Guoqing Jiang \u003cjgq516@gmail.com\u003e\nDate: Wed Jul 24 11:09:21 2019 +0200\n\n md: don\u0027t call spare_active in md_reap_sync_thread if all member devices can\u0027t work\n \n When add one disk to array, the md_reap_sync_thread is responsible\n to activate the spare and set In_sync flag for the new member in\n spare_active().\n \n But if raid1 has one member disk A, and disk B is added to the array.\n Then we offline A before all the datas are synchronized from A to B,\n obviously B doesn\u0027t have the latest data as A, but B is still marked\n with In_sync flag.\n \n So let\u0027s not call spare_active under the condition, otherwise B is\n still showed with \u0027U\u0027 state which is not correct.\n \n Signed-off-by: Guoqing Jiang \u003cguoqing.jiang@cloud.ionos.com\u003e\n Signed-off-by: Song Liu \u003csongliubraving@fb.com\u003e\n\ncommit 062f5b2ae12a153644c765e7ba3b0f825427be1d\nAuthor: Guoqing Jiang \u003cjgq516@gmail.com\u003e\nDate: Wed Jul 24 11:09:20 2019 +0200\n\n md: don\u0027t set In_sync if array is frozen\n \n When a disk is added to array, the following path is called in mdadm.\n \n Manage_subdevs -\u003e sysfs_freeze_array\n -\u003e Manage_add\n -\u003e sysfs_set_str(\u0026info, NULL, \"sync_action\",\"idle\")\n \n Then from kernel side, Manage_add invokes the path (add_new_disk -\u003e\n validate_super \u003d super_1_validate) to set In_sync flag.\n \n Since In_sync means \"device is in_sync with rest of array\", and the new\n added disk need to resync thread to help the synchronization of data.\n And md_reap_sync_thread would call spare_active to set In_sync for the\n new added disk finally. So don\u0027t set In_sync if array is in frozen.\n \n Signed-off-by: Guoqing Jiang \u003cguoqing.jiang@cloud.ionos.com\u003e\n Signed-off-by: Song Liu \u003csongliubraving@fb.com\u003e\n\ncommit 9a567843f7ce0037bfd4d5fdc58a09d0a527b28b\nAuthor: Guoqing Jiang \u003cjgq516@gmail.com\u003e\nDate: Wed Jul 24 11:09:19 2019 +0200\n\n md: allow last device to be forcibly removed from RAID1/RAID10.\n \n When the \u0027last\u0027 device in a RAID1 or RAID10 reports an error,\n we do not mark it as failed. This would serve little purpose\n as there is no risk of losing data beyond that which is obviously\n lost (as there is with RAID5), and there could be other sectors\n on the device which are readable, and only readable from this device.\n This in general this maximises access to data.\n \n However the current implementation also stops an admin from removing\n the last device by direct action. This is rarely useful, but in many\n case is not harmful and can make automation easier by removing special\n cases.\n \n Also, if an attempt to write metadata fails the device must be marked\n as faulty, else an infinite loop will result, attempting to update\n the metadata on all non-faulty devices.\n \n So add \u0027fail_last_dev\u0027 member to \u0027struct mddev\u0027, then we can bypasses\n the \u0027last disk\u0027 checks for RAID1 and RAID10, and control the behavior\n per array by change sysfs node.\n \n Signed-off-by: NeilBrown \u003cneilb@suse.de\u003e\n [add sysfs node for fail_last_dev by Guoqing]\n Signed-off-by: Guoqing Jiang \u003cguoqing.jiang@cloud.ionos.com\u003e\n Signed-off-by: Song Liu \u003csongliubraving@fb.com\u003e\n\ncommit cf89160793c439dca00e2563d0b7f153c274027b\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 23:41:55 2019 +0300\n\n md: Convert to use int_pow()\n \n Instead of linear approach to calculate power of 10, use generic int_pow()\n which does it better.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Song Liu \u003csongliubraving@fb.com\u003e\n\ncommit 7cee6d4e6035603d42acd56d591e624921aa1b14\nAuthor: Yufen Yu \u003cyuyufen@huawei.com\u003e\nDate: Fri Jul 19 13:48:47 2019 +0800\n\n md/raid10: end bio when the device faulty\n \n Just like raid1, we do not queue write error bio to retry write\n and acknowlege badblocks, when the device is faulty.\n \n Signed-off-by: Yufen Yu \u003cyuyufen@huawei.com\u003e\n Signed-off-by: Song Liu \u003csongliubraving@fb.com\u003e\n\ncommit eeba6809d8d58908b5ed1b5ceb5fcb09a98a7cad\nAuthor: Yufen Yu \u003cyuyufen@huawei.com\u003e\nDate: Fri Jul 19 13:48:46 2019 +0800\n\n md/raid1: end bio when the device faulty\n \n When write bio return error, it would be added to conf-\u003eretry_list\n and wait for raid1d thread to retry write and acknowledge badblocks.\n \n In narrow_write_error(), the error bio will be split in the unit of\n badblock shift (such as one sector) and raid1d thread issues them\n one by one. Until all of the splited bio has finished, raid1d thread\n can go on processing other things, which is time consuming.\n \n But, there is a scene for error handling that is not necessary.\n When the device has been set faulty, flush_bio_list() may end\n bios in pending_bio_list with error status. Since these bios\n has not been issued to the device actually, error handlding to\n retry write and acknowledge badblocks make no sense.\n \n Even without that scene, when the device is faulty, badblocks info\n can not be written out to the device. Thus, we also no need to\n handle the error IO.\n \n Signed-off-by: Yufen Yu \u003cyuyufen@huawei.com\u003e\n Signed-off-by: Song Liu \u003csongliubraving@fb.com\u003e\n\ncommit 143f6e733b73051cd22dcb80951c6c929da413ce\nAuthor: Xiao Ni \u003cxni@redhat.com\u003e\nDate: Mon Jul 8 10:14:32 2019 +0800\n\n md/raid6: Set R5_ReadError when there is read failure on parity disk\n \n 7471fb77ce4d (\"md/raid6: Fix anomily when recovering a single device in\n RAID6.\") avoids rereading P when it can be computed from other members.\n However, this misses the chance to re-write the right data to P. This\n patch sets R5_ReadError if the re-read fails.\n \n Also, when re-read is skipped, we also missed the chance to reset\n rdev-\u003eread_errors to 0. It can fail the disk when there are many read\n errors on P member disk (other disks don\u0027t have read error)\n \n V2: upper layer read request don\u0027t read parity/Q data. So there is no\n need to consider such situation.\n \n This is Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n \n Fixes: 7471fb77ce4d (\"md/raid6: Fix anomily when recovering a single device in RAID6.\")\n Cc: \u003cstable@vger.kernel.org\u003e #4.4+\n Signed-off-by: Xiao Ni \u003cxni@redhat.com\u003e\n Signed-off-by: Song Liu \u003csongliubraving@fb.com\u003e\n\ncommit 4675719d0f47d18bc13db62bd21cffd4e4ec8001\nAuthor: Hou Tao \u003choutao1@huawei.com\u003e\nDate: Tue Jul 2 22:35:48 2019 +0800\n\n raid1: use an int as the return value of raise_barrier()\n \n Using a sector_t as the return value is misleading, because\n raise_barrier() only return 0 or -EINTR.\n \n Also add comments for the return values of raise_barrier().\n \n Signed-off-by: Hou Tao \u003choutao1@huawei.com\u003e\n Signed-off-by: Song Liu \u003csongliubraving@fb.com\u003e\n\ncommit f19b4fe783e753cfc0322969a8bc80c43c979316\nAuthor: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\nDate: Sat Jul 27 16:16:56 2019 +0200\n\n ARM: dts: exynos: Use space after \u0027\u003d\u0027 in exynos4412-itop-scp-core\n \n Replace tab with space after assignment operator.\n \n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n\ncommit 80107e764846a6f9b40fc2a78306329ed9052733\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Wed Jul 31 09:52:08 2019 +0100\n\n mtd: rawnand: remove redundant assignment to variable ret\n \n Variable ret is being initialized with a value that is never read\n and ret is being re-assigned a little later on. The assignment is\n redundant and hence can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n\ncommit 75de0eb28d03ebe3c9448fc04372df5c39b6cacc\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 20:47:27 2019 +0800\n\n mtd: rawnand: r852: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n\ncommit 3194166052401e04ca6ca71446a1ca5dd9af2433\nAuthor: Julia Lawall \u003cjulia.lawall@lip6.fr\u003e\nDate: Mon Jul 22 18:39:34 2019 -0500\n\n mtd: rawnand: ingenic: fix devm_platform_ioremap_resource.cocci warnings\n \n drivers/mtd/nand/raw/ingenic/ingenic_nand_drv.c:330:1-9: WARNING: Use devm_platform_ioremap_resource for cs -\u003e base\n \n Use devm_platform_ioremap_resource helper which wraps\n platform_get_resource() and devm_ioremap_resource() together.\n \n Generated by: scripts/coccinelle/api/devm_platform_ioremap_resource.cocci\n \n Fixes: c403ec33b613 (\"mtd: rawnand: ingenic: Fix ingenic_ecc dependency\")\n CC: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Julia Lawall \u003cjulia.lawall@lip6.fr\u003e\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n\ncommit abd4e745ffc4d762b0ed4eada9ff732d8861174d\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Wed Aug 7 10:52:47 2019 -0400\n\n Revert \"drm/gem: Rename drm_gem_dumb_map_offset() to drm_gem_map_offset()\"\n \n This reverts commit 220df83a5394fbf7c1486ba7848794b7b351d598.\n \n Turns out drm_gem_dumb_map_offset really only worked for the dumb buffer\n case, so revert the name change.\n \n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Sean Paul \u003csean@poorly.run\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807145253.2037-2-sean@poorly.run\n\ncommit e6be0a9913c817689b61a7dc36550b80a3b27f62\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Wed Aug 7 10:52:48 2019 -0400\n\n Revert \"drm/panfrost: Use drm_gem_map_offset()\"\n \n This reverts commit 583bbf46133c726bae277e8f4e32bfba2a528c7f.\n \n Turns out we need mmap to work on imported BOs even if the current code\n is buggy.\n \n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Sean Paul \u003csean@poorly.run\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807145253.2037-3-sean@poorly.run\n\ncommit 0a0b65d8eae3eb25dde57f72596b7a7c81a126fa\nAuthor: Emil Velikov \u003cemil.velikov@collabora.com\u003e\nDate: Mon May 27 09:17:39 2019 +0100\n\n drm/vgem: drop DRM_AUTH usage from the driver\n \n The authentication can be circumvented, by design, by using the render\n node.\n \n From the driver POV there is no distinction between primary and render\n nodes, thus we can drop the token.\n \n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190527081741.14235-11-emil.l.velikov@gmail.com\n\ncommit 34127c7ad86ea144b85a6f405216ef1187debe84\nAuthor: Emil Velikov \u003cemil.velikov@collabora.com\u003e\nDate: Mon May 27 09:17:35 2019 +0100\n\n drm/msm: drop DRM_AUTH usage from the driver\n \n The authentication can be circumvented, by design, by using the render\n node.\n \n From the driver POV there is no distinction between primary and render\n nodes, thus we can drop the token.\n \n Cc: Rob Clark \u003crobdclark@gmail.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: freedreno@lists.freedesktop.org\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190527081741.14235-7-emil.l.velikov@gmail.com\n\ncommit 94533c410f9beafe96d539527acb1156e0f6c7f2\nAuthor: Emil Velikov \u003cemil.velikov@collabora.com\u003e\nDate: Wed May 22 16:02:19 2019 +0100\n\n drm/nouveau: remove open-coded drm_invalid_op()\n \n Cc: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n Cc: nouveau@lists.freedesktop.org\n Signed-off-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190522150219.13913-2-emil.l.velikov@gmail.com\n\ncommit 26f00514d944db8761d62ebf6f94766a81b97480\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Aug 7 15:20:41 2019 +0100\n\n drm/i915: Isolate i915_getparam_ioctl()\n \n This giant switch has tendrils all other the struct and does not fit\n in with the rest of the driver bring up and control in i915_drv.c. Push\n it to one side so that it can grow in peace.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Acked-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807142041.32699-1-chris@chris-wilson.co.uk\n\ncommit a4ba050dbc8825ce52657ba5837e8ef92bd4835c\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Aug 6 12:43:14 2019 +0300\n\n serial: 8250_dw: Use a unified new dev variable in remove\n \n The commit 2cb78eab2376 (\"serial: 8250_dw: Use a unified new dev variable in\n probe\") introduced a local dev variable in -\u003eprobe(). Do the same in -\u003eremove()\n in order to prepare for sequential patches.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Kefeng Wang \u003cwangkefeng.wang@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190806094322.64987-1-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d6026e9befe2e950d6f9e11948df6deb91e1dba8\nAuthor: Rahul Tanwar \u003crahul.tanwar@linux.intel.com\u003e\nDate: Wed Aug 7 17:21:31 2019 +0800\n\n serial: lantiq: Add SMP support\n \n The existing driver can only support single core SoC. But new multicore\n platforms which reuse the same driver/IP need SMP support. This patch adds\n multicore support in the driver.\n \n Signed-off-by: Rahul Tanwar \u003crahul.tanwar@linux.intel.com\u003e\n Reviewed-by: Andy Shevchenko \u003candriy.shevchenko@intel.com\u003e\n Link: https://lore.kernel.org/r/7912786cccad60c72b20ea724af1def505ab22aa.1565160764.git.rahul.tanwar@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 5adf5781019dd21233d6d13ab0d78bf03a13d4f1\nAuthor: Saravana Kannan \u003csaravanak@google.com\u003e\nDate: Tue Aug 6 19:05:58 2019 -0700\n\n of/platform: Fix device_links_supplier_sync_state_resume() warning\n \n In platforms/devices which have CONFIG_OF turned on but don\u0027t have a\n populated DT, the calls to device_links_supplier_sync_state_pause() and\n device_links_supplier_sync_state_resume() can get mismatched. This will\n cause a warning during boot. Fix the warning by making sure the calls are\n matched even in that case.\n \n Reported-by: Qian Cai \u003ccai@lca.pw\u003e\n Signed-off-by: Saravana Kannan \u003csaravanak@google.com\u003e\n Link: https://lore.kernel.org/r/20190807020559.74458-1-saravanak@google.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 0ad0bdd8ba6489d05044aa5ca6c6f7491be7743c\nAuthor: Saravana Kannan \u003csaravanak@google.com\u003e\nDate: Tue Aug 6 12:26:53 2019 -0700\n\n of/platform: Fix fn definitons for of_link_is_valid() and of_link_property()\n \n of_link_is_valid() can be static since it\u0027s not used anywhere else.\n \n of_link_property() return type should have been int instead of bool.\n \n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Saravana Kannan \u003csaravanak@google.com\u003e\n Acked-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190806192654.138605-1-saravanak@google.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 08f103b9a9502974109fab47ea35ca8542c4e57a\nAuthor: Julien Grall \u003cjulien.grall@arm.com\u003e\nDate: Wed Aug 7 11:34:45 2019 +0100\n\n arm64/ptrace: Fix typoes in sve_set() comment\n \n The ptrace trace SVE flags are prefixed with SVE_PT_*. Update the\n comment accordingly.\n \n Reviewed-by: Dave Martin \u003cDave.Martin@arm.com\u003e\n Signed-off-by: Julien Grall \u003cjulien.grall@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 2951d5efaf8b67cc4851373de92a91a60899611c\nAuthor: Miles Chen \u003cmiles.chen@mediatek.com\u003e\nDate: Wed Aug 7 08:33:36 2019 +0800\n\n arm64: mm: print hexadecimal EC value in mem_abort_decode()\n \n This change prints the hexadecimal EC value in mem_abort_decode(),\n which makes it easier to lookup the corresponding EC in\n the ARM Architecture Reference Manual.\n \n The commit 1f9b8936f36f (\"arm64: Decode information from ESR upon mem\n faults\") prints useful information when memory abort occurs. It would\n be easier to lookup \"0x25\" instead of \"DABT\" in the document. Then we\n can check the corresponding ISS.\n \n For example:\n Current info Document\n EC Exception class\n \"CP15 MCR/MRC\" 0x3 \"MCR or MRC access to CP15a...\"\n \"ASIMD\" 0x7 \"Access to SIMD or floating-point...\"\n \"DABT (current EL)\" 0x25 \"Data Abort taken without...\"\n ...\n \n Before:\n Unable to handle kernel paging request at virtual address 000000000000c000\n Mem abort info:\n ESR \u003d 0x96000046\n Exception class \u003d DABT (current EL), IL \u003d 32 bits\n SET \u003d 0, FnV \u003d 0\n EA \u003d 0, S1PTW \u003d 0\n Data abort info:\n ISV \u003d 0, ISS \u003d 0x00000046\n CM \u003d 0, WnR \u003d 1\n \n After:\n Unable to handle kernel paging request at virtual address 000000000000c000\n Mem abort info:\n ESR \u003d 0x96000046\n EC \u003d 0x25: DABT (current EL), IL \u003d 32 bits\n SET \u003d 0, FnV \u003d 0\n EA \u003d 0, S1PTW \u003d 0\n Data abort info:\n ISV \u003d 0, ISS \u003d 0x00000046\n CM \u003d 0, WnR \u003d 1\n \n Cc: Anshuman Khandual \u003canshuman.khandual@arm.com\u003e\n Cc: James Morse \u003cjames.morse@arm.com\u003e\n Acked-by: Mark Rutland \u003cMark.rutland@arm.com\u003e\n Signed-off-by: Miles Chen \u003cmiles.chen@mediatek.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit b99286b088ea843b935dcfb29f187697359fe5cd\nAuthor: Qian Cai \u003ccai@lca.pw\u003e\nDate: Mon Aug 5 23:05:03 2019 -0400\n\n arm64/prefetch: fix a -Wtype-limits warning\n \n The commit d5370f754875 (\"arm64: prefetch: add alternative pattern for\n CPUs without a prefetcher\") introduced MIDR_IS_CPU_MODEL_RANGE() to be\n used in has_no_hw_prefetch() with rv_min\u003d0 which generates a compilation\n warning from GCC,\n \n In file included from ./arch/arm64/include/asm/cache.h:8,\n from ./include/linux/cache.h:6,\n from ./include/linux/printk.h:9,\n from ./include/linux/kernel.h:15,\n from ./include/linux/cpumask.h:10,\n from arch/arm64/kernel/cpufeature.c:11:\n arch/arm64/kernel/cpufeature.c: In function \u0027has_no_hw_prefetch\u0027:\n ./arch/arm64/include/asm/cputype.h:59:26: warning: comparison of\n unsigned expression \u003e\u003d 0 is always true [-Wtype-limits]\n _model \u003d\u003d (model) \u0026\u0026 rv \u003e\u003d (rv_min) \u0026\u0026 rv \u003c\u003d (rv_max); \\\n ^~\n arch/arm64/kernel/cpufeature.c:889:9: note: in expansion of macro\n \u0027MIDR_IS_CPU_MODEL_RANGE\u0027\n return MIDR_IS_CPU_MODEL_RANGE(midr, MIDR_THUNDERX,\n ^~~~~~~~~~~~~~~~~~~~~~~\n \n Fix it by converting MIDR_IS_CPU_MODEL_RANGE to a static inline\n function.\n \n Signed-off-by: Qian Cai \u003ccai@lca.pw\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 71c67a31f09fa8fdd1495dffd96a5f0d4cef2ede\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Wed Aug 7 12:48:33 2019 +0100\n\n init/Kconfig: Fix infinite Kconfig recursion on PPC\n \n Commit 5cf896fb6be3 (\"arm64: Add support for relocating the kernel with\n RELR relocations\") introduced CONFIG_TOOLS_SUPPORT_RELR, which checks\n for RELR support in the toolchain as part of the kernel configuration.\n During this procedure, \"$(NM)\" is invoked to see if it supports the new\n relocation format, however PowerPC conditionally overrides this variable\n in the architecture Makefile in order to pass \u0027--synthetic\u0027 when\n targetting PPC64.\n \n This conditional override causes Kconfig to recurse forever, since\n CONFIG_TOOLS_SUPPORT_RELR cannot be determined without $(NM) being\n defined, but that in turn depends on CONFIG_PPC64:\n \n $ make ARCH\u003dpowerpc CROSS_COMPILE\u003dpowerpc-linux-gnu-\n scripts/kconfig/conf --syncconfig Kconfig\n scripts/kconfig/conf --syncconfig Kconfig\n scripts/kconfig/conf --syncconfig Kconfig\n [...]\n \n In this particular case, it looks like PowerPC may be able to pass\n \u0027--synthetic\u0027 unconditionally to nm or even drop it altogether. While\n that is being resolved, let\u0027s just bodge the RELR check by picking up\n $(NM) directly from the environment in whatever state it happens to be\n in.\n \n Cc: Peter Collingbourne \u003cpcc@google.com\u003e\n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Suggested-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 67028421e4e5a720c6a8e2fa8e98fc2848d3b243\nAuthor: Sean Paul \u003cseanpaul@chromium.org\u003e\nDate: Wed Aug 7 10:20:58 2019 -0400\n\n Revert \"drm/nouveau: remove open-coded drm_invalid_op()\"\n \n This reverts commit ccdae42575695ab442941310bd67c7ed1714e273.\n \n Mandatory review was missing from this patch.\n \n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807142101.251400-6-sean@poorly.run\n\ncommit 7d5000d7d21dd5d58ddbb75c534090d100d463a0\nAuthor: Sean Paul \u003cseanpaul@chromium.org\u003e\nDate: Wed Aug 7 10:20:57 2019 -0400\n\n Revert \"drm/msm: drop DRM_AUTH usage from the driver\"\n \n This reverts commit 88209d2c5035737f96bcfc2fd73c0fd8d80e9bf1.\n \n Mandatory review was missing from this patch.\n \n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807142101.251400-5-sean@poorly.run\n\ncommit 54fc01b775fe35fa0889cb3344ed98c72a52d2c1\nAuthor: Sean Paul \u003cseanpaul@chromium.org\u003e\nDate: Wed Aug 7 10:20:56 2019 -0400\n\n Revert \"drm/vgem: drop DRM_AUTH usage from the driver\"\n \n This reverts commit e4eee93d25776da998ec2dfaabe7d2206598d26d.\n \n Mandatory review was missing from this patch.\n \n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807142101.251400-4-sean@poorly.run\n\ncommit 912e21bf855cc46a7e9d7ac12b01b353fe703087\nAuthor: Sean Paul \u003cseanpaul@chromium.org\u003e\nDate: Wed Aug 7 10:20:55 2019 -0400\n\n Revert \"Revert \"drm/panfrost: Use drm_gem_map_offset()\"\"\n \n This reverts commit be855382bacb5ccfd24f9be6098d87acf4cfbb15.\n \n Mandatory review was missing from this patch.\n \n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807142101.251400-3-sean@poorly.run\n\ncommit 87160d2bfc5ed2eeb61d0d8921a38377a6ea3a36\nAuthor: Sean Paul \u003cseanpaul@chromium.org\u003e\nDate: Wed Aug 7 10:20:54 2019 -0400\n\n Revert \"Revert \"drm/gem: Rename drm_gem_dumb_map_offset() to drm_gem_map_offset()\"\"\n \n This reverts commit 415d2e9e07574d3de63b8df77dc686e0ebf64865.\n \n Mandatory review was missing from this patch.\n \n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807142101.251400-2-sean@poorly.run\n\ncommit 57afe3ea27141720cf8c4e061ba8cec2f9c60d0c\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:34 2019 -0700\n\n pinctrl: cherryview: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Cc: linux-gpio@vger.kernel.org\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n Cc: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 4e73d02f164b528b48d7e008a8f4a456c5a50fc7\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:34 2019 -0700\n\n pinctrl: intel: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Cc: linux-gpio@vger.kernel.org\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n Cc: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 6a33a1d6d66a46157951388ce4c5f2674b251982\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Wed Aug 7 16:41:50 2019 +0300\n\n pinctrl: intel: Use NSEC_PER_USEC for debounce calculus\n \n Replace hard coded constants with self-explanatory names, i.e.\n use NSEC_PER_USEC for debounce calculus.\n \n While here, add a unit suffix to debounce period constant.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n\ncommit 34e656703e85c0485b2903cd54e2c50c0bbe425b\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Fri Jul 26 23:16:36 2019 +0300\n\n pinctrl: baytrail: Re-use data structures from pinctrl-intel.h (part 2)\n \n We have some data structures duplicated across the drivers.\n Let\u0027s deduplicate them by using ones that being provided by\n pinctrl-intel.h.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n\ncommit 179e5a6114cc80d8254f524d7bf7b94523b89faf\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 18:56:33 2019 +0300\n\n pinctrl: intel: Remove default Interrupt Status offset\n \n Since some of the GPIO controllers use different Interrupt Status offset,\n it make sense to provide it explicitly in the drivers.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n\ncommit 3b1e9188df9ca794b089090de22d7cfc23732dc5\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 18:56:32 2019 +0300\n\n pinctrl: lewisburg: Provide Interrupt Status register offset\n \n Since some of the GPIO controllers use different Interrupt Status offset,\n it make sense to provide it explicitly in the driver.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n\ncommit 2a2ed4752108a641651bb103542000810d5d89b4\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 18:56:31 2019 +0300\n\n pinctrl: denverton: Provide Interrupt Status register offset\n \n Since some of the GPIO controllers use different Interrupt Status offset,\n it make sense to provide it explicitly in the driver.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n\ncommit f702e0b93cdb7785d94d6a508e5b1dac99fd9bff\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 18:56:30 2019 +0300\n\n pinctrl: sunrisepoint: Provide Interrupt Status register offset\n \n Since some of the GPIO controllers use different Interrupt Status offset,\n it make sense to provide it explicitly in the driver.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n\ncommit b8d728687ee5c011ba6f3054553abdac21021ed3\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 18:56:29 2019 +0300\n\n pinctrl: icelake: Provide Interrupt Status register offset\n \n Since some of the GPIO controllers use different Interrupt Status offset,\n it make sense to provide it explicitly in the driver.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n\ncommit cd5acb2bce5bc12dd83b3738e47b01aee7693c14\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 18:56:28 2019 +0300\n\n pinctrl: geminilake: Provide Interrupt Status register offset\n \n Since some of the GPIO controllers use different Interrupt Status offset,\n it make sense to provide it explicitly in the driver.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n\ncommit 82af745d2f934b03dd430419b7c33808ebc1d4f7\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 18:56:27 2019 +0300\n\n pinctrl: cannonlake: Provide Interrupt Status register offset\n \n Since some of the GPIO controllers use different Interrupt Status offset,\n it make sense to provide it explicitly in the driver.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n\ncommit 3818d73612caa5a3651300bf73ab7dc9fdbc03e1\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 18:56:26 2019 +0300\n\n pinctrl: broxton: Provide Interrupt Status register offset\n \n Since some of the GPIO controllers use different Interrupt Status offset,\n it make sense to provide it explicitly in the driver.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n\ncommit 7eb7ecddd3bdb9fd616bc62348ed3aaccaa9ef83\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 18:55:14 2019 +0300\n\n pinctrl: intel: Simplify offset validation in intel_get_padcfg()\n \n There is more generic and simpler validation just against the nregs.\n Using it allows to drop customization from the intel_get_padcfg().\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n\ncommit d71c4b034165951f97ec135510686a3c1b93170b\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Aug 7 12:38:27 2019 +0100\n\n drm/i915: Include the DRIVER_DATE in the error state\n \n For forklifted frankenkernels, the reported kernel version has no\n bearing on the actual driver version. Include our own driver date that\n is updated every time we tag drm-intel-next.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807113827.28127-1-chris@chris-wilson.co.uk\n\ncommit ed29da7123a6275f0b2b1ee05f761b25ab0f20d4\nAuthor: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nDate: Wed Aug 7 15:28:32 2019 +0300\n\n drm/i915: Fix some NULL vs IS_ERR() conditions\n \n There were several places which check for NULL when they should have\n been checking for IS_ERR().\n \n Fixes: d8af05ff38ae (\"drm/i915: Allow sharing the idle-barrier from other kernel requests\")\n Signed-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807122832.GA10517@mwanda\n\ncommit 9ae06cad821bbeaca9b34dbb9bdb8591a1d9762a\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Aug 7 09:37:01 2019 +0100\n\n drm/i915: Use intel_engine_lookup_user for probing HAS_BSD etc\n \n Use the same mechanism to determine if a backend engine exists for a\n uabi mapping as used internally.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807083702.16349-2-chris@chris-wilson.co.uk\n\ncommit 2edda80db3d065b875731de2a846f9e700ccb5ce\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Aug 7 12:04:31 2019 +0100\n\n drm/i915: Rename engines to match their user interface\n \n During engine setup, we may find that some engines are fused off causing\n a misalignment between internal names and the instances seen by users,\n e.g. (I915_ENGINE_CLASS_VIDEO_DECODE, 1) may be vcs2 in hardware.\n Normally this is invisible to the user, but a few debug interfaces (and\n our own internal tracing) use the original HW name not the name the user\n would expect as formed from their class:instance tuple. Replace our\n internal name with the uabi name for consistency with, for example, error\n states.\n \n v2: Keep the pretty printing of class name in the selftest\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111311\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807110431.8130-1-chris@chris-wilson.co.uk\n\ncommit a860fac420971c5a90d4f78959b44ead793aee4f\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Tue Aug 6 18:12:13 2019 +0300\n\n ASoC: fsl_sai: Add support for imx7ulp/imx8mq\n \n SAI module on imx7ulp/imx8m features 2 new registers (VERID and PARAM)\n at the beginning of register address space.\n \n On imx7ulp FIFOs can held up to 16 x 32 bit samples.\n On imx8mq FIFOs can held up to 128 x 32 bit samples.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Acked-by: Nicolin Chen \u003cnicoleotsuka@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190806151214.6783-5-daniel.baluta@nxp.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 4f7a0728b5305e2d865f543fbcffd617e03c7674\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Tue Aug 6 18:12:12 2019 +0300\n\n ASoC: fsl_sai: Add support for SAI new version\n \n New IP version introduces Version ID and Parameter registers\n and optionally added Timestamp feature.\n \n VERID and PARAM registers are placed at the top of registers\n address space and some registers are shifted according to\n the following table:\n \n Tx/Rx data registers and Tx/Rx FIFO registers keep their\n addresses, all other registers are shifted by 8.\n \n SAI Memory map is described in chapter 13.10.4.1.1 I2S Memory map\n of the Reference Manual [1].\n \n In order to make as less changes as possible we attach an offset\n to each register offset to each changed register definition. The\n offset is read from each board private data.\n \n [1]https://cache.nxp.com/secured/assets/documents/en/reference-manual/IMX8MDQLQRM.pdf?__gda__\u003d1563728701_38bea7f0f726472cc675cb141b91bec7\u0026fileExt\u003d.pdf\n \n Signed-off-by: Mihai Serban \u003cmihai.serban@nxp.com\u003e\n [initial coding in the NXP internal tree]\n Signed-off-by: Shengjiu Wang \u003cshengjiu.wang@nxp.com\u003e\n [bugfixing and cleanups]\n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n [adapted to linux-next]\n Acked-by: Nicolin Chen \u003cnicoleotsuka@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190806151214.6783-4-daniel.baluta@nxp.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit b84f50b0fcb497a62068926fca793d2d213c7dbd\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Tue Aug 6 18:12:11 2019 +0300\n\n ASoC: fsl_sai: Update Tx/Rx channel enable mask\n \n Tx channel enable (TCE) / Rx channel enable (RCE) bits\n enable corresponding data channel for Tx/Rx operation.\n \n Because SAI supports up the 8 channels TCE/RCE occupy\n up the 8 bits inside TCR3/RCR3 registers we need to extend\n the mask to reflect this.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Acked-by: Nicolin Chen \u003cnicoleotsuka@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190806151214.6783-3-daniel.baluta@nxp.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 5f0ac20ed6db1d6da2eea8b862cf3d54fdfb5830\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Tue Aug 6 18:12:10 2019 +0300\n\n ASoC: fsl_sai: Add registers definition for multiple datalines\n \n SAI IP supports up to 8 data lines. The configuration of\n supported number of data lines is decided at SoC integration\n time.\n \n This patch adds definitions for all related data TX/RX registers:\n * TDR0..7, Transmit data register\n * TFR0..7, Transmit FIFO register\n * RDR0..7, Receive data register\n * RFR0..7, Receive FIFO register\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Acked-by: Nicolin Chen \u003cnicoleotsuka@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190806151214.6783-2-daniel.baluta@nxp.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit abf31feea26c0f412a191c83f408311a0de7435c\nAuthor: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\nDate: Tue Jul 23 16:43:40 2019 +0200\n\n ASoC: Intel: Update request-reply IPC model\n \n struct ipc_message contains fields: header, tx_data and tx_size which\n represent TX i.e. request while RX is represented by rx_data and rx_size\n with reply\u0027s header equivalent missing.\n \n Reply header may contain some vital information including, but not\n limited to, received payload size. Some IPCs have entire payload found\n within RX header instead. Content and value of said header is context\n dependent and may vary between firmware versions and target platform.\n Current model does not allow such IPCs to function at all.\n \n Rather than appending yet another parameter to an already long list of\n such for sst_ipc_tx_message_XXXs, declare message container in form of\n struct sst_ipc_message and add them to parent\u0027s ipc_message declaration.\n \n Align haswell, baytrail and skylake with updated request-reply model and\n modify their reply processing functions to save RX header within message\n container. Despite the range of changes, status quo is achieved.\n \n Signed-off-by: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190723144341.21339-2-cezary.rojewski@intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit b8ba3b572c70c8559ef76c1e0ca02e7b05126e8c\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Wed Aug 7 10:30:53 2019 +0900\n\n ASoC: soc-core: tidyup for snd_soc_dapm_new_controls()\n \n snd_soc_dapm_new_controls() registers controls by using\n for(... i \u003c num; ...). It means if widget was NULL, num should be zero.\n Thus, we don\u0027t need to check about widget pointer.\n This patch also cares missing return value.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87ftmdahow.wl-kuninori.morimoto.gx@renesas.com\n Reviewed-by: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 32d2c172fe88705058310635a732259c0a98e89b\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Wed Aug 7 10:30:47 2019 +0900\n\n ASoC: soc-core: reuse rtdcom at snd_soc_rtdcom_add()\n \n snd_soc_rtdcom_add() is using both \"rtdcom\" and \"new_rtdcom\" as\n variable name, but these are not used at same time.\n Let\u0027s reuse rtdcom.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87h86tahp2.wl-kuninori.morimoto.gx@renesas.com\n Reviewed-by: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 42849064500b24892e2abfe8679507dfffd585f1\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Wed Aug 7 10:30:41 2019 +0900\n\n ASoC: soc-core: don\u0027t use for_each_card_links_safe() at snd_soc_find_dai_link()\n \n It doesn\u0027t removes list during loop at snd_soc_find_dai_link().\n We don\u0027t need to use _safe loop.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87imr9ahp9.wl-kuninori.morimoto.gx@renesas.com\n Reviewed-by: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 5b99a0aad08a3428cc3262ecee29a71c88c981c2\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Wed Aug 7 10:30:36 2019 +0900\n\n ASoC: soc-core: check return value of snd_soc_add_dai_link()\n \n snd_soc_add_dai_link() might return error, we need to check it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87k1bpahpd.wl-kuninori.morimoto.gx@renesas.com\n Reviewed-by: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 6634e3d6ea8cd92fce10d1d2da0c8e42ab84da32\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Wed Aug 7 10:30:31 2019 +0900\n\n ASoC: soc-core: add comment for for_each_xxx\n \n soc-core has many for_each_xxx, but it is a little bit\n difficult to know which list is relead to which for_each_xxx.\n This patch adds missing comment for it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87lfw5ahpj.wl-kuninori.morimoto.gx@renesas.com\n Reviewed-by: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit ad64bfbd09d7b7997a1b1510983350836bcd6ed9\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Wed Aug 7 10:30:13 2019 +0900\n\n ASoC: soc-core: set component-\u003edebugfs_root NULL\n \n To be more safety code, let\u0027s set NULL to component-\u003edebugfs_root\n when it was cleanuped.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87muglahq0.wl-kuninori.morimoto.gx@renesas.com\n Reviewed-by: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit b977fcf477c176e5f41775f0ea139f935b0f25b7\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Wed Jul 31 15:13:19 2019 +0100\n\n irqdomain/debugfs: Use PAs to generate fwnode names\n \n Booting a large arm64 server (HiSi D05) leads to the following\n shouting at boot time:\n \n [ 20.722132] debugfs: File \u0027irqchip@(____ptrval____)-3\u0027 in directory \u0027domains\u0027 already present!\n [ 20.730851] debugfs: File \u0027irqchip@(____ptrval____)-3\u0027 in directory \u0027domains\u0027 already present!\n [ 20.739560] debugfs: File \u0027irqchip@(____ptrval____)-3\u0027 in directory \u0027domains\u0027 already present!\n [ 20.748267] debugfs: File \u0027irqchip@(____ptrval____)-3\u0027 in directory \u0027domains\u0027 already present!\n [ 20.756975] debugfs: File \u0027irqchip@(____ptrval____)-3\u0027 in directory \u0027domains\u0027 already present!\n [ 20.765683] debugfs: File \u0027irqchip@(____ptrval____)-3\u0027 in directory \u0027domains\u0027 already present!\n [ 20.774391] debugfs: File \u0027irqchip@(____ptrval____)-3\u0027 in directory \u0027domains\u0027 already present!\n \n and many more... Evidently, we expect something a bit more informative\n than ____ptrval____, and certainly we want all of our domains, not just\n the first one.\n \n For that, turn the %p used to generate the fwnode name into something\n that won\u0027t be repainted (%pa). Given that we\u0027ve now fixed all users to\n pass a pointer to a PA, it will actually do the right thing.\n \n Acked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 467a3bb974320a23c0c000559cc54b64e06e9787\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Tue Aug 6 15:23:33 2019 +0100\n\n PCI: hv: Allocate a named fwnode instead of an address-based one\n \n To allocate its fwnode that is then used to allocate an irqdomain,\n the driver uses irq_domain_alloc_fwnode(), passing it a VA as an\n identifier. This is a rather bad idea, as this address ends up\n published in debugfs (and we want to move away from VAs there\n anyway).\n \n Instead, let\u0027s allocate a named fwnode by using the device GUID as\n an identifier. It is allegedly unique, and can be traced back to\n the original device.\n \n Acked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 920d947af9cd528338bbfe989449489a9962142f\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Aug 7 10:52:13 2019 +0200\n\n spi: sh-msiof: Use devm_platform_ioremap_resource() helper\n \n Use the devm_platform_ioremap_resource() helper instead of open-coding\n the same operation.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Link: https://lore.kernel.org/r/20190807085213.24666-1-geert+renesas@glider.be\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit daa19fe5b082779962988a5ba9e38509004db3de\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Wed Jul 31 16:13:42 2019 +0100\n\n gpio/ixp4xx: Register the base PA instead of its VA in fwnode\n \n Do not expose the base VA (it appears in debugfs). Instead,\n record the PA, which at least can be used to precisely identify\n the associated irqchip and domain.\n \n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Acked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 9adc54d4e8c56ed5851bb51314ffcbaebdfbf17e\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Wed Jul 31 16:13:42 2019 +0100\n\n irqchip/ixp4xx: Register the base PA instead of its VA in fwnode\n \n Do not expose the base VA (it appears in debugfs). Instead,\n record the PA, which at least can be used to precisely identify\n the associated irqchip and domain.\n \n Acked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 7d5b7695c22f6e22d9a16df17f477832eef8e2b7\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Wed Jul 31 16:13:42 2019 +0100\n\n irqchip/gic-v2m: Register the frame\u0027s PA instead of its VA in fwnode\n \n Do not expose the frame\u0027s VA (it appears in debugfs). Instead,\n record the PA, which at least can be used to precisely identify\n the associated irqchip and domain.\n \n Acked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 188a8471ef03718709bd785d3976f6cd72bb551d\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Wed Jul 31 16:13:42 2019 +0100\n\n irqchip/gic: Register the distributor\u0027s PA instead of its VA in fwnode\n \n Do not expose the distributor\u0027s VA (it appears in debugfs). Instead,\n record the PA, which at least can be used to precisely identify\n the associated irqchip and domain.\n \n Acked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 5778cc77111f09fd2033c66cdf1654f542771ed0\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Wed Jul 31 16:13:42 2019 +0100\n\n irqchip/gic-v3-its: Register the ITS\u0027 PA instead of its VA in fwnode\n \n Do not expose the ITS\u0027 VA (it appears in debugfs). Instead, record\n the PA, which at least can be used to precisely identify the associated\n irqchip and domain.\n \n Acked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 6444b40eeda4f78f57b255dd7ecb8d3e5936eea2\nAuthor: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\nDate: Mon Aug 5 14:21:34 2019 -0700\n\n x86/apic: Annotate global config variables as \"read-only after init\"\n \n Mark the APIC\u0027s global config variables that are constant after boot as\n __ro_after_init to help document that the majority of the APIC config is\n not changed at runtime, and to harden the kernel a smidge.\n \n Signed-off-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190805212134.12001-1-sean.j.christopherson@intel.com\n\ncommit eeee0d093afc2e352e15b23bca38f69aad3d689b\nAuthor: Marc Zyngier \u003cmaz@kernel.org\u003e\nDate: Wed Jul 31 16:13:42 2019 +0100\n\n irqchip/gic-v3: Register the distributor\u0027s PA instead of its VA in fwnode\n \n Do not expose the distributor\u0027s VA (it appears in debugfs). Instead,\n record the PA, which at least can be used to precisely identify\n the associated irqchip and domain.\n \n Acked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Signed-off-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n\ncommit 79629d0f7ce5b38515c1716911a0181f01b91102\nAuthor: Rayagonda Kokatanur \u003crayagonda.kokatanur@broadcom.com\u003e\nDate: Tue Aug 6 15:37:50 2019 +0530\n\n spi: bcm-qspi: Fix BSPI QUAD and DUAL mode support when using flex mode\n \n Fix data transfer width settings based on DT field \u0027spi-rx-bus-width\u0027\n to configure BSPI in single, dual or quad mode by using data width\n and not the command width.\n \n Fixes: 5f195ee7d830c (\"spi: bcm-qspi: Implement the spi_mem interface\")\n \n Signed-off-by: Rayagonda Kokatanur \u003crayagonda.kokatanur@broadcom.com\u003e\n Link: https://lore.kernel.org/r/1565086070-28451-1-git-send-email-rayagonda.kokatanur@broadcom.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit e15fdcf80e2968b01268c4710275389acdc0ce17\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Aug 5 11:57:01 2019 -0700\n\n tty: serial: fsl_lpuart: Ignore TX/RX interrupts if DMA is enabled\n \n In a mixed DMA/IRQ use-case (e.g.: DMA for TX, IRQ for RX), interrupt\n handler might try to handle Rx/Tx condition it shouldn\u0027t. Change the\n code to only handle TX/RX event if corresponding path isn\u0027t being\n handled by DMA.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190805185701.22863-7-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 1fe3896855e62f17d488f0d7501f26d6f62978d1\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Aug 5 11:57:00 2019 -0700\n\n tty: serial: fsl_lpuart: Don\u0027t enable TIE in .startup() or .resume()\n \n Enabling TIE in .startup() callback causes the driver to start (or at\n least try) to transmit data before .start_tx() is called. Which, while\n harmless (since TIE handler will immediately disable it), is a no-op\n and shouldn\u0027t really happen. Drop UARTCR2_TIE from list of bits set in\n lpuart_startup().\n \n This change will also not enable TIE in .resume(), but it seems that,\n similart to .startup(), transmit interrupt shouldn\u0027t be enabled there\n either.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190805185701.22863-6-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit e5b01b657f838b2bb4c657f6d3a376e1ba732457\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Aug 5 11:56:59 2019 -0700\n\n tty: serial: fsl_lpuart: Introduce lpuart*_setup_watermark_enable()\n \n Most users of lpuart*_setup_watermark() enable identical set of flags\n right after the call, so combine those two action into a subroutine\n and make use of it.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190805185701.22863-5-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f2a90ac0243dfc3f747d6a9f9d0092257260fcbf\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Aug 5 11:56:58 2019 -0700\n\n tty: serial: fsl_lpuart: Introduce lpuart32_configure()\n \n Code doing final steps of TX/RX configuration in lpuart32_startup()\n and lpuart_resume() is identical, so move it into a standalone\n subroutine.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190805185701.22863-4-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 87a46c2a5b0e20a85b5d5619f0f8b75b12baf2ae\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Aug 5 11:56:57 2019 -0700\n\n tty: serial: fsl_lpuart: Introduce lpuart_rx_dma_startup()\n \n Code doing initial DMA RX configuration in lpuart_startup() and\n lpuart32_startup() is exactly the same, so move it into a standalone\n subroutine.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190805185701.22863-3-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 3aa263f1ac1a8d5c252d69d06578433f24383045\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Aug 5 11:56:56 2019 -0700\n\n tty: serial: fsl_lpuart: Introduce lpuart_tx_dma_startup()\n \n Code configure DMA TX path in lpuart_startup(), lpuart32_startup() and\n lpuart_resume() is doing exactly the same thing, so move it into a\n standalone subroutine.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190805185701.22863-2-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 371be51a925a619f1fb149b8d7707e353d9c9f86\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Tue Aug 6 18:12:14 2019 +0300\n\n ASoC: dt-bindings: Introduce compatible strings for 7ULP and 8MQ\n \n For i.MX7ULP and i.MX8MQ register map is changed. Add two new compatbile\n strings to differentiate this.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Acked-by: Nicolin Chen \u003cnicoleotsuka@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190806151214.6783-6-daniel.baluta@nxp.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit bb67dd1878de5766c24fe7a4002d1a12c14da518\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Tue Aug 6 12:06:03 2019 -0500\n\n ASoC: SOF: Intel: hda: fix MSI handling\n \n The addition of a kernel module parameter to optionally disable MSI\n had the side effect of permanently disabling it.\n \n The return value of pci_alloc_irq_vectors() is the number of allocated\n vectors or a negative number on error, so testing with the ! operator\n is not quite right. It was one optimization too far.\n \n Restore previous behavior to use MSI by default, unless the user\n selects not to do so or the allocation of irq_vectors fails.\n \n Fixes: 672ff5e3596ee (\u0027ASoC: SOF: Intel: hda: add a parameter to disable MSI\u0027)\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190806170603.10815-1-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 0922c7a51c775b71a297598623debe9cb6ce3a08\nAuthor: Shuming Fan \u003cshumingf@realtek.com\u003e\nDate: Tue Aug 6 17:14:59 2019 +0800\n\n ASoC: rt1011: Add a flag for the R0 calibration test\n \n The factory test needs to know whether the calibration completed.\n This flag helps to confirm the calibration completed or not.\n \n Signed-off-by: Shuming Fan \u003cshumingf@realtek.com\u003e\n Link: https://lore.kernel.org/r/20190806091459.14382-1-shumingf@realtek.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 3403b808a0eb6d25883b72a6c4d2a4a452f8a233\nAuthor: Shuming Fan \u003cshumingf@realtek.com\u003e\nDate: Tue Aug 6 17:14:35 2019 +0800\n\n ASoC: rt1011: Add R0 temperature and TDM1 ADC2DAT Swap control\n \n - The user level application could set the R0 temperature after booting system.\n The degree Celsius of R0 temperature store in the non-volatile space\n when doing R0 calibration.\n - TDM1 ADC2DAT Swap controls use to control TDM slot2/3 data\n \n Signed-off-by: Shuming Fan \u003cshumingf@realtek.com\u003e\n Link: https://lore.kernel.org/r/20190806091435.14329-1-shumingf@realtek.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit cc352735d49c9c25fd711795bee35d3d001ddffa\nAuthor: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\nDate: Tue Aug 6 15:19:58 2019 -0700\n\n ASoC: SOF: Intel: hda: Initialize HDA controller after i915 init\n \n On some platforms, sound card registration fails when a HDMI\n monitor is not connected. This is caused by a recent commit\n that switched the order in which the HDA controller and the\n i915 are initialized. Initializing the i915 before initializing\n the HDA controller fixes the problem.\n \n Fixes: be1b577d01787c (\"ASoC: SOF: Intel: hda: fix the hda init chip\"\n Signed-off-by: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Reviewed-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n Link: https://lore.kernel.org/r/20190806221958.19180-1-ranjani.sridharan@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 06e8f5c842f2dbb232897ba967ea7b422745c271\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Tue Aug 6 12:45:38 2019 +0900\n\n ASoC: rsnd: don\u0027t call clk_get_rate() under atomic context\n \n ADG is using clk_get_rate() under atomic context, thus, we might\n have scheduling issue.\n To avoid this issue, we need to get/keep clk rate under\n non atomic context.\n \n We need to handle ADG as special device at Renesas Sound driver.\n From SW point of view, we want to impletent it as\n rsnd_mod_ops :: prepare, but it makes code just complicate.\n \n To avoid complicated code/patch, this patch adds new clk_rate[] array,\n and keep clk IN rate when rsnd_adg_clk_enable() was called.\n \n Reported-by: Leon Kong \u003cLeon.KONG@cn.bosch.com\u003e\n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Tested-by: Leon Kong \u003cLeon.KONG@cn.bosch.com\u003e\n Link: https://lore.kernel.org/r/87v9vb0xkp.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 42d038c4fb00f1ec1a4c4616784da4561385b628\nAuthor: Leo Yan \u003cleo.yan@linaro.org\u003e\nDate: Tue Aug 6 18:00:14 2019 +0800\n\n arm64: Add support for function error injection\n \n Inspired by the commit 7cd01b08d35f (\"powerpc: Add support for function\n error injection\"), this patch supports function error injection for\n Arm64.\n \n This patch mainly support two functions: one is regs_set_return_value()\n which is used to overwrite the return value; the another function is\n override_function_with_return() which is to override the probed\n function returning and jump to its caller.\n \n Reviewed-by: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\n Signed-off-by: Leo Yan \u003cleo.yan@linaro.org\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 45880f7b7b19e043ce0aaa4cb7d05369425c82fa\nAuthor: Leo Yan \u003cleo.yan@linaro.org\u003e\nDate: Tue Aug 6 18:00:13 2019 +0800\n\n error-injection: Consolidate override function definition\n \n The function override_function_with_return() is defined separately for\n each architecture and every architecture\u0027s definition is almost same\n with each other. E.g. x86 and powerpc both define function in its own\n asm/error-injection.h header and override_function_with_return() has\n the same definition, the only difference is that x86 defines an extra\n function just_return_func() but it is specific for x86 and is only used\n by x86\u0027s override_function_with_return(), so don\u0027t need to export this\n function.\n \n This patch consolidates override_function_with_return() definition into\n asm-generic/error-injection.h header, thus all architectures can use the\n common definition. As result, the architecture specific headers are\n removed; the include/linux/error-injection.h header also changes to\n include asm-generic/error-injection.h header rather than architecture\n header, furthermore, it includes linux/compiler.h for successful\n compilation.\n \n Reviewed-by: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\n Signed-off-by: Leo Yan \u003cleo.yan@linaro.org\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit f2ae04c45b1a9d6c61faa8a946d515f81876e703\nAuthor: Jitendra Sharma \u003cshajit@codeaurora.org\u003e\nDate: Tue Aug 6 11:35:36 2019 +0530\n\n pinctrl: qcom: Add SC7180 pinctrl driver\n \n Add initial pinctrl driver to support pin configuration with\n pinctrl framework for SC7180\n \n Signed-off-by: Jitendra Sharma \u003cshajit@codeaurora.org\u003e\n Signed-off-by: Vivek Gautam \u003cvivek.gautam@codeaurora.org\u003e\n [rnayak: modify to use upstream tile support\n sort and squash some functions]\n Signed-off-by: Rajendra Nayak \u003crnayak@codeaurora.org\u003e\n Reviewed-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Reviewed-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190806060536.18094-2-rnayak@codeaurora.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 6d2db2965ba2079309cb94343c6136b294256c47\nAuthor: Jitendra Sharma \u003cshajit@codeaurora.org\u003e\nDate: Tue Aug 6 11:35:35 2019 +0530\n\n dt-bindings: pinctrl: qcom: Add SC7180 pinctrl binding\n \n Add the binding for the TLMM pinctrl block found in the SC7180 platform\n \n Signed-off-by: Jitendra Sharma \u003cshajit@codeaurora.org\u003e\n Signed-off-by: Vivek Gautam \u003cvivek.gautam@codeaurora.org\u003e\n [rnayak: Fix some copy-paste issues, sort and fix functions]\n Signed-off-by: Rajendra Nayak \u003crnayak@codeaurora.org\u003e\n Reviewed-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Reviewed-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190806060536.18094-1-rnayak@codeaurora.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 40b0bcd3e0e258cff3382717fb287f8be161a398\nAuthor: Song Hui \u003chui.song_1@nxp.com\u003e\nDate: Wed Aug 7 10:12:54 2019 +0800\n\n gpio: mpc8xxx: Add new platforms GPIO DT node description\n \n Update the NXP GPIO node dt-binding file for QorIQ and\n Layerscape platforms, and add one more example with\n ls1028a GPIO node.\n \n Signed-off-by: Song Hui \u003chui.song_1@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190807021254.49092-1-hui.song_1@nxp.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit b19e8f30ec05ef9609edc06d3f7855f10ca3a21a\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Aug 5 17:25:35 2019 +0300\n\n serial: 8250_exar: Replace msleep(1) with usleep_range()\n \n As explained in Documentation/timers/timers-howto.rst\n the small amount of milliseconds sometimes produces\n much longer delays.\n \n Replace msleep(1) with usleep_range(1000, 1100).\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190805142535.21948-2-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d6a59fec33a2a635b52e4763d2c0e2c0ab0f5662\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Aug 5 17:25:34 2019 +0300\n\n serial: 8250_exar: Consolidate callback assignments in default_setup()\n \n For better maintenance consolidate port callbacks in default_setup().\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190805142535.21948-1-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7d9c1d2f7aca2651b3821947bf928ee131df102a\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 6 16:44:52 2019 +0200\n\n USB: add support for dev_groups to struct usb_device_driver\n \n Now that the driver core supports dev_groups for individual drivers,\n expose that pointer to struct usb_device_driver to make it easier for USB\n drivers to also use it.\n \n Yes, users of usb_device_driver are much rare, but there are instances\n already that use custom sysfs files, so adding this support will make\n things easier for those drivers. usbip is one example, hubs might be\n another one.\n \n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Link: https://lore.kernel.org/r/20190806144502.17792-3-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b71b283e3d6df40cef03b5ce882b2385971f58a3\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Tue Aug 6 16:44:51 2019 +0200\n\n USB: add support for dev_groups to struct usb_driver\n \n Now that the driver core supports dev_groups for individual drivers,\n expose that pointer to struct usb_driver to make it easier for USB\n drivers to also use it.\n \n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Link: https://lore.kernel.org/r/20190806144502.17792-2-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit fcfce9f354769b060e82bccc05b8fa2956eb1b04\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Aug 4 22:16:29 2019 +0200\n\n drm/sti: fix opencoded use of drm_panel_*\n \n Use the drm_panel_(enable|disable|get_modes) functions.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Cc: Vincent Abriou \u003cvincent.abriou@st.com\u003e\n Signed-off-by: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190804201637.1240-9-sam@ravnborg.org\n\ncommit c33fe752de9669366e4578ba69fc0911e27a1ec6\nAuthor: Vidya Sagar \u003cvidyas@nvidia.com\u003e\nDate: Fri Aug 2 11:47:27 2019 +0530\n\n firmware: tegra: Move BPMP resume to noirq phase\n \n Modules like PCIe in Tegra194 need BPMP firmware services in noirq phase\n and hence move BPMP resume to noirq phase.\n This patch is verified on Tegra210, Tegra186 and Tegra194.\n \n Signed-off-by: Vidya Sagar \u003cvidyas@nvidia.com\u003e\n Reviewed-by: Timo Alho \u003ctalho@nvidia.com\u003e\n Signed-off-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n\ncommit b8c036dfc66f774624ba7b9e7414fd90a61e0527\nAuthor: Christian König \u003cchristian.koenig@amd.com\u003e\nDate: Mon Aug 5 11:49:20 2019 +0200\n\n dma-buf: simplify reservation_object_get_fences_rcu a bit\n \n We can add the exclusive fence to the list after making sure we got\n a consistent state.\n \n Signed-off-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/322034/?series\u003d64786\u0026rev\u003d1\n\ncommit 4c2488cfaa997e396aeb9d6496db94c25b97c671\nAuthor: Christian König \u003cchristian.koenig@amd.com\u003e\nDate: Mon Aug 5 13:18:43 2019 +0200\n\n drm/i915: stop using seqcount for fence pruning\n \n After waiting for a reservation object use reservation_object_test_signaled_rcu\n to opportunistically prune the fences on the object.\n \n This allows removal of the seqcount handling in the reservation object.\n \n Signed-off-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/322032/?series\u003d64786\u0026rev\u003d1\n\ncommit fdde3097488eebc9c62fe7311388521a2711d940\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Aug 7 10:41:28 2019 +0100\n\n drm/i915: Drop expectations of VM_IO from our GGTT mmappings\n \n We don\u0027t really want to use VM_IO for our GGTT mmaps (it implies that\n the mmap contains memory mapped registers, which we do not expose) yet I\n overzealously added it to an assert just because that\u0027s how we always\n had setup the vma.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Abdiel Janulgue \u003cabdiel.janulgue@linux.intel.com\u003e\n Reviewed-by: Abdiel Janulgue \u003cabdiel.janulgue@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190807094128.9993-1-chris@chris-wilson.co.uk\n\ncommit 96e95496b02dbf1b19a2d4ce238810572e149606\nAuthor: Christian König \u003cchristian.koenig@amd.com\u003e\nDate: Tue Aug 6 13:33:12 2019 +0200\n\n dma-buf: fix shared fence list handling in reservation_object_copy_fences\n \n Add some helpers to correctly allocate/free reservation_object_lists.\n \n Otherwise we might forget to drop dma_fence references on list destruction.\n \n Signed-off-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/322031/?series\u003d64786\u0026rev\u003d1\n\ncommit 93505ee7d05e836fd18894019e93c3875198fcc5\nAuthor: Christian König \u003cchristian.koenig@amd.com\u003e\nDate: Mon Aug 5 11:14:27 2019 +0200\n\n dma-buf: fix busy wait for new shared fences\n \n When reservation_object_add_shared_fence is replacing an old fence with a new\n one we should not drop the old one before the new one is in place.\n \n Otherwise other cores can busy wait for the new one to appear.\n \n Signed-off-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/322030/\n\ncommit 0cccd42e6193e168cbecc271dae464e4a53fd7b3\nAuthor: Jianjun Wang \u003cjianjun.wang@mediatek.com\u003e\nDate: Fri Jun 28 15:34:25 2019 +0800\n\n PCI: mediatek: Add controller support for MT7629\n \n MT7629 is an ARM platform SoC which has the same PCIe IP as MT7622.\n \n The HW default value of its PCI host controller Device ID is invalid,\n fix it to match the hardware implementation.\n \n Signed-off-by: Jianjun Wang \u003cjianjun.wang@mediatek.com\u003e\n [lorenzo.pieralisi@arm.com: commit log/minor spelling update]\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Reviewed-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n Acked-by: Ryder Lee \u003cryder.lee@mediatek.com\u003e\n\ncommit f4c737d6194f509733641017b50548abc01f4b0d\nAuthor: Jianjun Wang \u003cjianjun.wang@mediatek.com\u003e\nDate: Fri Jun 28 15:34:24 2019 +0800\n\n dt-bindings: PCI: Add support for MT7629\n \n MT7629 is an ARM based platform SoC integrating the same PCIe IP as\n MT7622, add a binding for it.\n \n Signed-off-by: Jianjun Wang \u003cjianjun.wang@mediatek.com\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Acked-by: Ryder Lee \u003cryder.lee@mediatek.com\u003e\n\ncommit 071b68cceee7e8232a221f464dd26667f57fbd8f\nAuthor: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\nDate: Tue Aug 6 15:22:08 2019 +0300\n\n drm/i915: abstract display suspend/resume operations\n \n Increase abstraction of display suspend/resume operations by providing\n higher level functions, and hiding the details inside\n intel_display_power.c.\n \n v2: Make checkpatch happy:\n - braces {} are not necessary for single statement blocks\n v3: Also move hsw/bdw PC8 sequences since they are related to\n display PM anyways. (Ville)\n v4: Rebase after a long time, plus Move functions to the new\n intel_display_power so we can stop exporting platform specific\n functions as pointed by Jani.\n v5: Remove unnecessary braces.\n v6 by Jani: make this purely non-functional cleanup, make functions static\n \n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190806122208.16786-2-jani.nikula@intel.com\n\ncommit 1c0023d4f5fd16bc9795c073cb31654d792712c5\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Tue Aug 6 13:07:31 2019 +0300\n\n drm/i915: move property enums to intel_display_types.h\n \n Move the property enums closer to home. Acually make the broadcast\n macros an enum while at it.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/589d5449a504ec9742dc1fa8d4871d0d6d0cddca.1565085692.git.jani.nikula@intel.com\n\ncommit 6da4a2c411e8d9be6f848f9207beb69f13113546\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Tue Aug 6 13:07:30 2019 +0300\n\n drm/i915: remove unnecessary includes of intel_display_types.h header\n \n With its original name intel_drv.h the intel_display_types.h header was\n superfluously cargo-cult included all over the place, while it\u0027s really\n mostly about display internals. Remove the unnecessary includes.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/e3d737f0ab87c55969e62c1e077e15c04c238297.1565085692.git.jani.nikula@intel.com\n\ncommit 1d455f8de8e8a211cc91e19484eeda2e454531a1\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Tue Aug 6 14:39:33 2019 +0300\n\n drm/i915: rename intel_drv.h to display/intel_display_types.h\n \n Everything about the file is about display, and mostly about types\n related to display. Move under display/ as intel_display_types.h to\n reflect the facts.\n \n There\u0027s still plenty to clean up, but start off with moving the file\n where it logically belongs and naming according to contents.\n \n v2: fix the include guard name in the renamed file\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190806113933.11799-1-jani.nikula@intel.com\n\ncommit a09d9a8002368e1b8754c8621ad06d3d7bade6f4\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Tue Aug 6 13:07:28 2019 +0300\n\n drm/i915: avoid including intel_drv.h via i915_drv.h-\u003ei915_trace.h\n \n Disentangle i915_drv.h from intel_drv.h, which gets included via\n i915_trace.h. This necessitates including i915_trace.h wherever it\u0027s\n needed.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/ed82bf259d3b725a1a1a3c3e9d6fb5c08bc4d489.1565085691.git.jani.nikula@intel.com\n\ncommit 683df830e6bd12e9eb7da1fbdfccc6ac444508ae\nAuthor: Fabien Parent \u003cfparent@baylibre.com\u003e\nDate: Tue Aug 6 11:50:29 2019 +0200\n\n cpufreq: mediatek-cpufreq: Add compatible for MT8516\n \n Add the compatible for MT8516 in order to take advantage of the\n MediaTek CPUFreq driver for Mediatek\u0027s MT8516 SoC.\n \n Signed-off-by: Fabien Parent \u003cfparent@baylibre.com\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit d0d392a8deed8442191bdf5b2d6bb17e04c928b6\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Tue Aug 6 14:37:00 2019 +0300\n\n drm/i915/tc: un-inline intel_tc_port_ref_held()\n \n Avoid including the intel_drv.h mega header from other header files to\n make further header cleanup easier.\n \n v2: restore the over-eagerly dropped \u003clinux/types.h\u003e (Imre)\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190806113700.18816-1-jani.nikula@intel.com\n\ncommit e15fd1bee2a998bffce776dd312108a2676bee57\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Tue Aug 6 13:07:26 2019 +0300\n\n drm/i915/mst: un-inline intel_dp_mst_encoder_active_links()\n \n Avoid including the intel_drv.h mega header from other header files to\n make further header cleanup easier.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/ae8f7a9514a9e952e1e0dba411aeb42293d1e183.1565085691.git.jani.nikula@intel.com\n\ncommit 366b6200f76e0f1241be262ea044eee44ded8fc6\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Tue Aug 6 13:07:25 2019 +0300\n\n drm/i915/bw: make intel_atomic_get_bw_state() static\n \n No need for this function to be accessible outside of intel_bw.c. Avoid\n including the i915_drv.h mega header from other header files to make\n further header cleanup easier.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/6c898ec6511af47c1c5b679e516dc757cd207146.1565085691.git.jani.nikula@intel.com\n\ncommit d64575ee8d91b067a2ea77b3fa31d973f7931591\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Tue Aug 6 13:07:24 2019 +0300\n\n drm/i915/irq: un-inline functions to avoid i915_drv.h include\n \n Avoid including the i915_drv.h mega header from other header files to\n make further header cleanup easier.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/45431103f6cbd4322808907154336fdc176ff20d.1565085691.git.jani.nikula@intel.com\n\ncommit 5b51f28fa7ba67aec4515a19f3ea31f14bf18599\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Tue Aug 6 13:07:23 2019 +0300\n\n drm/i915/sprite: un-inline icl_is_hdr_plane()\n \n Avoid including the i915_drv.h mega header from other header files to\n make further header cleanup easier.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/5f998c72018c0f1cc5cdb239327a1281d21f4c0e.1565085691.git.jani.nikula@intel.com\n\ncommit 3e1876251bc34ba0dc94d7a5f8bf9ef312e2f278\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Tue Aug 6 13:07:22 2019 +0300\n\n drm/i915: move intel_display.c function declarations\n \n Move the declarations of functions in intel_display.c to\n intel_display.h. There is still plenty of cleanup to do in\n intel_display.[ch], but one step at a time.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/ebe19d3c508faa82e651ab3c8eb7a0975bfa52b7.1565085691.git.jani.nikula@intel.com\n\ncommit 8faa1cf6ed82f33009f63986c3776cc48af1b7b2\nAuthor: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nDate: Mon Jun 24 16:47:17 2019 +0300\n\n EDAC/altera: Use the proper type for the IRQ status bits\n \n Smatch complains about the cast of a u32 pointer to unsigned long:\n \n drivers/edac/altera_edac.c:1878 altr_edac_a10_irq_handler()\n warn: passing casted pointer \u0027\u0026irq_status\u0027 to \u0027find_first_bit()\u0027\n \n This code wouldn\u0027t work on a 64 bit big endian system because it would\n read past the end of \u0026irq_status.\n \n [ bp: massage. ]\n \n Fixes: 13ab8448d2c9 (\"EDAC, altera: Add ECC Manager IRQ controller support\")\n Signed-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Reviewed-by: Thor Thayer \u003cthor.thayer@linux.intel.com\u003e\n Cc: James Morse \u003cjames.morse@arm.com\u003e\n Cc: kernel-janitors@vger.kernel.org\n Cc: linux-edac \u003clinux-edac@vger.kernel.org\u003e\n Cc: Mauro Carvalho Chehab \u003cmchehab@kernel.org\u003e\n Cc: Tony Luck \u003ctony.luck@intel.com\u003e\n Link: https://lkml.kernel.org/r/20190624134717.GA1754@mwanda\n\ncommit 7bea41c4a25675a32bc3a63cfbd98099fdfb2695\nAuthor: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\nDate: Tue Aug 6 19:23:56 2019 -0700\n\n soc: qcom: socinfo: Annotate switch cases with fall through\n \n Introduce fall through annotations in the switch statements of\n socinfo_debugfs_init() to silence compiler warnings.\n \n Acked-by: Vaishali Thakkar \u003cvaishali.thakkar@linaro.org\u003e\n Fixes: 9c84c1e78634 (\"soc: qcom: socinfo: Expose custom attributes\")\n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 16e910d4467ccdf1cae5035e71035e5f7197e77d\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Tue Aug 6 10:45:29 2019 -0700\n\n selftests/bpf: test_progs: drop extra trailing tab\n \n Small (un)related cleanup.\n \n Cc: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 66bd2ec1e0d9781133eb1a14eddb68facc69d54b\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Tue Aug 6 10:45:28 2019 -0700\n\n selftests/bpf: test_progs: test__printf -\u003e printf\n \n Now that test__printf is a simple wraper around printf, let\u0027s drop it\n (and test__vprintf as well).\n \n Cc: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 946152b3c5d6bab128db8eee226ec2665429b79c\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Tue Aug 6 10:45:27 2019 -0700\n\n selftests/bpf: test_progs: switch to open_memstream\n \n Use open_memstream to override stdout during test execution.\n The copy of the original stdout is held in env.stdout and used\n to print subtest info and dump failed log.\n \n test_{v,}printf are now simple wrappers around stdout and will be\n removed in the next patch.\n \n v5:\n * fix -v crash by always setting env.std{in,err} (Alexei Starovoitov)\n * drop force_log check from stdio_hijack (Andrii Nakryiko)\n \n v4:\n * one field per line for stdout/stderr (Andrii Nakryiko)\n \n v3:\n * don\u0027t do strlen over log_buf, log_cnt has it already (Andrii Nakryiko)\n \n v2:\n * add ifdef __GLIBC__ around open_memstream (maybe pointless since\n we already depend on glibc for argp_parse)\n * hijack stderr as well (Andrii Nakryiko)\n * don\u0027t hijack for every test, do it once (Andrii Nakryiko)\n * log_cap -\u003e log_size (Andrii Nakryiko)\n * do fseeko in a proper place (Andrii Nakryiko)\n * check open_memstream returned value (Andrii Nakryiko)\n \n Cc: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit aa733660dbd8d9192b8c528ae0f4b84f3fef74e4\nAuthor: Yifeng Sun \u003cpkusunyifeng@gmail.com\u003e\nDate: Sun Aug 4 19:56:11 2019 -0700\n\n openvswitch: Print error when ovs_execute_actions() fails\n \n Currently in function ovs_dp_process_packet(), return values of\n ovs_execute_actions() are silently discarded. This patch prints out\n an debug message when error happens so as to provide helpful hints\n for debugging.\n Acked-by: Pravin B Shelar \u003cpshelar@ovn.org\u003e\n \n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6393e60644862478a9da8c6599ffc04f7f94bedf\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Tue Aug 6 13:24:50 2019 +0200\n\n mips: fix vdso32 build, again\n \n The generic vdso support adds the same #if hack in two places,\n asm/vdso/vdso.h and config-n32-o32-env.c, but only the second\n is actually used. The result lacks the BUILD_VDSO32_64 macro,\n and that triggers a build error:\n \n ./include/linux/page-flags-layout.h:95:2: error: #error \"Not enough bits in page flags\"\n \n Move the macro into the other place, and remove the duplicated\n bits.\n \n Reported-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Fixes: ee38d94a0ad8 (\"page flags: prioritize kasan bits over last-cpuid\")\n Fixes: 24640f233b46 (\"mips: Add support for generic vDSO\")\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n Cc: Andrey Konovalov \u003candreyknvl@google.com\u003e\n Cc: Andrey Ryabinin \u003caryabinin@virtuozzo.com\u003e\n Cc: Dmitry Vyukov \u003cdvyukov@google.com\u003e\n Cc: Will Deacon \u003cwill.deacon@arm.com\u003e\n Cc: Christoph Lameter \u003ccl@linux.com\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n\ncommit 905421ee15357c31d5d201de9c1efafc69ea78db\nAuthor: Paul Burton \u003cpaul.burton@mips.com\u003e\nDate: Tue Aug 6 12:52:31 2019 -0700\n\n MIPS: octeon: Remove ARCH_HAS_IRQ_PER_CPU\n \n ARCH_HAS_IRQ_PER_CPU is unused anywhere in the kernel - remove the\n definition.\n \n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit ece51529a470f849ed8652612cc325188c166a9d\nAuthor: Paul Burton \u003cpaul.burton@mips.com\u003e\nDate: Tue Aug 6 12:43:20 2019 -0700\n\n MIPS: octeon: Remove cpu_has_saa\n \n The cpu_has_saa feature macro was added along with Cavium Octeon CPU\n support back in commit 5b3b16880f40 (\"MIPS: Add Cavium OCTEON processor\n support files to arch/mips/cavium-octeon.\") but has never been used.\n \n Remove the dead code.\n \n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit 2e0bc7f3cb5553812f5808ede2cea746aabfbd03\nAuthor: John Hurley \u003cjohn.hurley@netronome.com\u003e\nDate: Sun Aug 4 16:10:49 2019 +0100\n\n nfp: flower: encode mac indexes with pre-tunnel rule check\n \n When a tunnel packet arrives on the NFP card, its destination MAC is\n looked up and MAC index returned for it. This index can help verify the\n tunnel by, for example, ensuring that the packet arrived on the expected\n port. If the packet is destined for a known MAC that is not connected to a\n given physical port then the mac index can have a global value (e.g. when\n a series of bonded ports shared the same MAC).\n \n If the packet is to be detunneled at a bridge device or internal port like\n an Open vSwitch VLAN port, then it should first match a \u0027pre-tunnel\u0027 rule\n to direct it to that internal port.\n \n Use the MAC index to indicate if a packet should match a pre-tunnel rule\n before decap is allowed. Do this by tracking the number of internal ports\n associated with a MAC address and, if the number if \u003e0, set a bit in the\n mac_index to forward the packet to the pre-tunnel table before continuing\n with decap.\n \n Signed-off-by: John Hurley \u003cjohn.hurley@netronome.com\u003e\n Reviewed-by: Simon Horman \u003csimon.horman@netronome.com\u003e\n Acked-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 09aa811bb7def147e230f53ed3c19ff1a54a1575\nAuthor: John Hurley \u003cjohn.hurley@netronome.com\u003e\nDate: Sun Aug 4 16:09:11 2019 +0100\n\n nfp: flower: remove offloaded MACs when reprs are applied to OvS bridges\n \n MAC addresses along with an identifying index are offloaded to firmware to\n allow tunnel decapsulation. If a tunnel packet arrives with a matching\n destination MAC address and a verified index, it can continue on the\n decapsulation process. This replicates the MAC verifications carried out\n in the kernel network stack.\n \n When a netdev is added to a bridge (e.g. OvS) then packets arriving on\n that dev are directed through the bridge datapath instead of passing\n through the network stack. Therefore, tunnelled packets matching the MAC\n of that dev will not be decapped here.\n \n Replicate this behaviour on firmware by removing offloaded MAC addresses\n when a MAC representer is added to an OvS bridge. This can prevent any\n false positive tunnel decaps.\n \n Signed-off-by: John Hurley \u003cjohn.hurley@netronome.com\u003e\n Reviewed-by: Simon Horman \u003csimon.horman@netronome.com\u003e\n Acked-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f12725d98cbe862c7680c63fd1f8b381af965f7b\nAuthor: John Hurley \u003cjohn.hurley@netronome.com\u003e\nDate: Sun Aug 4 16:09:10 2019 +0100\n\n nfp: flower: offload pre-tunnel rules\n \n Pre-tunnel rules are TC flower and OvS rules that forward a packet to the\n tunnel end point where it can then pass through the network stack and be\n decapsulated. These are required if the tunnel end point is, say, an OvS\n internal port.\n \n Currently, firmware determines that a packet is in a tunnel and decaps it\n if it has a known destination IP and MAC address. However, this bypasses\n the flower pre-tunnel rule and so does not update the stats. Further to\n this it ignores VLANs that may exist outside of the tunnel header.\n \n Offload pre-tunnel rules to the NFP. This embeds the pre-tunnel rule into\n the tunnel decap process based on (firmware) mac index and VLAN. This\n means that decap can be carried out correctly with VLANs and that stats\n can be updated for all kernel rules correctly.\n \n Signed-off-by: John Hurley \u003cjohn.hurley@netronome.com\u003e\n Reviewed-by: Simon Horman \u003csimon.horman@netronome.com\u003e\n Acked-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 120ffd84a9ec2430faba83db274a6946374f4631\nAuthor: John Hurley \u003cjohn.hurley@netronome.com\u003e\nDate: Sun Aug 4 16:09:09 2019 +0100\n\n nfp: flower: verify pre-tunnel rules\n \n Pre-tunnel rules must direct packets to an internal port based on L2\n information. Rules that egress to an internal port are already indicated\n by a non-NULL device in its nfp_fl_payload struct. Verfiy the rest of the\n match fields indicate that the rule is a pre-tunnel rule. This requires a\n full match on the destination MAC address, an option VLAN field, and no\n specific matches on other lower layer fields (with the exception of L4\n proto and flags).\n \n If a rule is identified as a pre-tunnel rule then mark it for offload to\n the pre-tunnel table. Similarly, remove it from the pre-tunnel table on\n rule deletion. The actual offloading of these commands is left to a\n following patch.\n \n Signed-off-by: John Hurley \u003cjohn.hurley@netronome.com\u003e\n Reviewed-by: Simon Horman \u003csimon.horman@netronome.com\u003e\n Acked-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f5c977eed725a000dd4efa6d56e86d88085d0b1b\nAuthor: John Hurley \u003cjohn.hurley@netronome.com\u003e\nDate: Sun Aug 4 16:09:08 2019 +0100\n\n nfp: flower: detect potential pre-tunnel rules\n \n Pre-tunnel rules are used when the tunnel end-point is on an \u0027internal\n port\u0027. These rules are used to direct the tunnelled packets (based on outer\n header fields) to the internal port where they can be detunnelled. The\n rule must send the packet to ingress the internal port at the TC layer.\n \n Currently FW does not support an action to send to ingress so cannot\n offload such rules. However, in preparation for populating the pre-tunnel\n table to represent such rules, check for rules that send to the ingress of\n an internal port and mark them as such. Further validation of such rules\n is left to subsequent patches.\n \n Signed-off-by: John Hurley \u003cjohn.hurley@netronome.com\u003e\n Reviewed-by: Simon Horman \u003csimon.horman@netronome.com\u003e\n Acked-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4b10c53d81fd1350a5510f3038f52b5db9c953e0\nAuthor: John Hurley \u003cjohn.hurley@netronome.com\u003e\nDate: Sun Aug 4 16:09:07 2019 +0100\n\n nfp: flower: push vlan after tunnel in merge\n \n NFP allows the merging of 2 flows together into a single offloaded flow.\n In the kernel datapath the packet must match 1 flow, impliment its\n actions, recirculate, match the 2nd flow and also impliment its actions.\n Merging creates a single flow with all actions from the 2 original flows.\n \n Firmware impliments a tunnel header push as the packet is about to egress\n the card. Therefore, if the first merge rule candiate pushes a tunnel,\n then the second rule can only have an egress action for a valid merge to\n occur (or else the action ordering will be incorrect). This prevents the\n pushing of a tunnel header followed by the pushing of a vlan header.\n \n In order to support this behaviour, firmware allows VLAN information to\n be encoded in the tunnel push action. If this is non zero then the fw will\n push a VLAN after the tunnel header push meaning that 2 such flows with\n these actions can be merged (with action order being maintained).\n \n Support tunnel in VLAN pushes by encoding VLAN information in the tunnel\n push action of any merge flow requiring this.\n \n Signed-off-by: John Hurley \u003cjohn.hurley@netronome.com\u003e\n Reviewed-by: Simon Horman \u003csimon.horman@netronome.com\u003e\n Acked-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 48e584ac583b08a923d4d872596cc7b049e99f12\nAuthor: John Hurley \u003cjohn.hurley@netronome.com\u003e\nDate: Sun Aug 4 16:09:06 2019 +0100\n\n net: sched: add ingress mirred action to hardware IR\n \n TC mirred actions (redirect and mirred) can send to egress or ingress of a\n device. Currently only egress is used for hw offload rules.\n \n Modify the intermediate representation for hw offload to include mirred\n actions that go to ingress. This gives drivers access to such rules and\n can decide whether or not to offload them.\n \n Signed-off-by: John Hurley \u003cjohn.hurley@netronome.com\u003e\n Reviewed-by: Simon Horman \u003csimon.horman@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d7609c96c6da0831e196d970a20dc960bcc4a4d6\nAuthor: John Hurley \u003cjohn.hurley@netronome.com\u003e\nDate: Sun Aug 4 16:09:05 2019 +0100\n\n net: tc_act: add helpers to detect ingress mirred actions\n \n TC mirred actions can send to egress or ingress on a given netdev. Helpers\n exist to detect actions that are mirred to egress. Extend the header file\n to include helpers to detect ingress mirred actions.\n \n Signed-off-by: John Hurley \u003cjohn.hurley@netronome.com\u003e\n Reviewed-by: Simon Horman \u003csimon.horman@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit fb1b775a247ee8d846152841f780eba6cb71bcfc\nAuthor: John Hurley \u003cjohn.hurley@netronome.com\u003e\nDate: Sun Aug 4 16:09:04 2019 +0100\n\n net: sched: add skbedit of ptype action to hardware IR\n \n TC rules can impliment skbedit actions. Currently actions that modify the\n skb mark are passed to offloading drivers via the hardware intermediate\n representation in the flow_offload API.\n \n Extend this to include skbedit actions that modify the packet type of the\n skb. Such actions may be used to set the ptype to HOST when redirecting a\n packet to ingress.\n \n Signed-off-by: John Hurley \u003cjohn.hurley@netronome.com\u003e\n Reviewed-by: Simon Horman \u003csimon.horman@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 77feb4eed7560215a724df6e7d4f1beaf98ba49d\nAuthor: John Hurley \u003cjohn.hurley@netronome.com\u003e\nDate: Sun Aug 4 16:09:03 2019 +0100\n\n net: tc_act: add skbedit_ptype helper functions\n \n The tc_act header file contains an inline function that checks if an\n action is changing the skb mark of a packet and a further function to\n extract the mark.\n \n Add similar functions to check for and get skbedit actions that modify\n the packet type of the skb.\n \n Signed-off-by: John Hurley \u003cjohn.hurley@netronome.com\u003e\n Reviewed-by: Simon Horman \u003csimon.horman@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a7a92cf81589ca2e9bef5443e8a99d0855b3722f\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Sun Aug 4 09:52:33 2019 +0200\n\n r8169: sync PCIe PHY init with vendor driver 8.047.01\n \n Synchronize PCIe PHY initialization with vendor driver version 8.047.01.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ef712ede3541dd364131d2fe29603cb362bd5aaa\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Sun Aug 4 09:47:51 2019 +0200\n\n r8169: add helper r8168_mac_ocp_modify\n \n Add a helper for MAC OCP read-modify-write operations.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a14cc4d24a645dd13abe66f50e1439136337fa1a\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Sun Aug 4 09:42:57 2019 +0200\n\n r8169: remove access to legacy register MultiIntr\n \n This code piece was inherited from RTL8139 code, the register at\n address 0x5c however has a different meaning on RTL8169 and is unused.\n So we can remove this.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 77ddaff218fc505f6930a2bf3e4eec2ff74255f5\nAuthor: Dave Taht \u003cdave.taht@gmail.com\u003e\nDate: Sat Aug 3 16:37:29 2019 -0700\n\n fq_codel: Kill useless per-flow dropped statistic\n \n It is almost impossible to get anything other than a 0 out of\n flow-\u003edropped statistic with a tc class dump, as it resets to 0\n on every round.\n \n It also conflates ecn marks with drops.\n \n It would have been useful had it kept a cumulative drop count, but\n it doesn\u0027t. This patch doesn\u0027t change the API, it just stops\n tracking a stat and state that is impossible to measure and nobody\n uses.\n \n Signed-off-by: Dave Taht \u003cdave.taht@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ae697f3bf784898a9be1d68ff7fc38819ca5040f\nAuthor: Dave Taht \u003cdave.taht@gmail.com\u003e\nDate: Sat Aug 3 16:37:28 2019 -0700\n\n Increase fq_codel count in the bulk dropper\n \n In the field fq_codel is often used with a smaller memory or\n packet limit than the default, and when the bulk dropper is hit,\n the drop pattern bifircates into one that more slowly increases\n the codel drop rate and hits the bulk dropper more than it should.\n \n The scan through the 1024 queues happens more often than it needs to.\n \n This patch increases the codel count in the bulk dropper, but\n does not change the drop rate there, relying on the next codel round\n to deliver the next packet at the original drop rate\n (after that burst of loss), then escalate to a higher signaling rate.\n \n Signed-off-by: Dave Taht \u003cdave.taht@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b8fb640643fcdb3bca84137c4cec0c649b25e056\nAuthor: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\nDate: Fri Aug 2 23:01:56 2019 -0700\n\n net: mdio-octeon: Fix Kconfig warnings and build errors\n \n After commit 171a9bae68c7 (\"staging/octeon: Allow test build on\n !MIPS\"), the following combination of configs cause a few Kconfig\n warnings and build errors (distilled from arm allyesconfig and Randy\u0027s\n randconfig builds):\n \n CONFIG_NETDEVICES\u003dy\n CONFIG_STAGING\u003dy\n CONFIG_COMPILE_TEST\u003dy\n \n and CONFIG_OCTEON_ETHERNET as either a module or built-in.\n \n WARNING: unmet direct dependencies detected for MDIO_OCTEON\n Depends on [n]: NETDEVICES [\u003dy] \u0026\u0026 MDIO_DEVICE [\u003dy] \u0026\u0026 MDIO_BUS [\u003dy]\n \u0026\u0026 64BIT [\u003dn] \u0026\u0026 HAS_IOMEM [\u003dy] \u0026\u0026 OF_MDIO [\u003dn]\n Selected by [y]:\n - OCTEON_ETHERNET [\u003dy] \u0026\u0026 STAGING [\u003dy] \u0026\u0026 (CAVIUM_OCTEON_SOC ||\n COMPILE_TEST [\u003dy]) \u0026\u0026 NETDEVICES [\u003dy]\n \n In file included from ../drivers/net/phy/mdio-octeon.c:14:\n ../drivers/net/phy/mdio-cavium.h:111:36: error: implicit declaration of\n function ‘writeq’; did you mean ‘writel’?\n [-Werror\u003dimplicit-function-declaration]\n 111 | #define oct_mdio_writeq(val, addr) writeq(val, (void *)addr)\n | ^~~~~~\n \n CONFIG_64BIT is not strictly necessary if the proper readq/writeq\n definitions are included from io-64-nonatomic-lo-hi.h.\n \n CONFIG_OF_MDIO is not needed when CONFIG_COMPILE_TEST is enabled because\n of commit f9dc9ac51610 (\"of/mdio: Add dummy functions in of_mdio.h.\").\n \n Fixes: 171a9bae68c7 (\"staging/octeon: Allow test build on !MIPS\")\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Reported-by: Mark Brown \u003cbroonie@kernel.org\u003e\n Reported-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\n Signed-off-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Acked-by: Randy Dunlap \u003crdunlap@infradead.org\u003e # build-tested\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 48e2331197b82c425e12ab728bb5332e49c78f63\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Fri Aug 2 15:34:55 2019 -0400\n\n net: dsa: dump CPU port regs through master\n \n Merge the CPU port registers dump into the master interface registers\n dump through ethtool, by nesting the ethtool_drvinfo and ethtool_regs\n structures of the CPU port into the dump.\n \n drvinfo-\u003eregdump_len will contain the full data length, while regs-\u003elen\n will contain only the master interface registers dump length.\n \n This allows for example to dump the CPU port registers on a ZII Dev\n C board like this:\n \n # ethtool -d eth1\n 0x004: 0x00000000\n 0x008: 0x0a8000aa\n 0x010: 0x01000000\n 0x014: 0x00000000\n 0x024: 0xf0000102\n 0x040: 0x6d82c800\n 0x044: 0x00000020\n 0x064: 0x40000000\n 0x084: RCR (Receive Control Register) 0x47c00104\n MAX_FL (Maximum frame length) 1984\n FCE (Flow control enable) 0\n BC_REJ (Broadcast frame reject) 0\n PROM (Promiscuous mode) 0\n DRT (Disable receive on transmit) 0\n LOOP (Internal loopback) 0\n 0x0c4: TCR (Transmit Control Register) 0x00000004\n RFC_PAUSE (Receive frame control pause) 0\n TFC_PAUSE (Transmit frame control pause) 0\n FDEN (Full duplex enable) 1\n HBC (Heartbeat control) 0\n GTS (Graceful transmit stop) 0\n 0x0e4: 0x76735d6d\n 0x0e8: 0x7e9e8808\n 0x0ec: 0x00010000\n .\n .\n .\n 88E6352 Switch Port Registers\n ------------------------------\n 00: Port Status 0x4d04\n Pause Enabled 0\n My Pause 1\n 802.3 PHY Detected 0\n Link Status Up\n Duplex Full\n Speed 100 or 200 Mbps\n EEE Enabled 0\n Transmitter Paused 0\n Flow Control 0\n Config Mode 0x4\n 01: Physical Control 0x003d\n RGMII Receive Timing Control Default\n RGMII Transmit Timing Control Default\n 200 BASE Mode 100\n Flow Control\u0027s Forced value 0\n Force Flow Control 0\n Link\u0027s Forced value Up\n Force Link 1\n Duplex\u0027s Forced value Full\n Force Duplex 1\n Force Speed 100 or 200 Mbps\n .\n .\n .\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 694a296024d4cd5e637fa679c4c7e4c63a7c24f3\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Aug 2 16:13:16 2019 +0100\n\n net/mlx5: remove self-assignment on esw-\u003edev\n \n There is a self assignment of esw-\u003edev to itself, clean this up by\n removing it. Also make dev a const pointer.\n \n Addresses-Coverity: (\"Self assignment\")\n Fixes: 6cedde451399 (\"net/mlx5: E-Switch, Verify support QoS element type\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Reviewed-by: Parav Pandit \u003cparav@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit bbeb6b6c07960b8d33bed5352ef462228110c5ab\nAuthor: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\nDate: Thu Jul 25 15:56:18 2019 +0800\n\n i2c: sprd: Validate the return value of clock initialization\n \n The \u0027enable\u0027 clock of I2C master is required, we should return an error\n if failed to get the \u0027enable\u0027 clock, to make sure the I2C driver can be\n defer probe if the clock resource is not ready.\n \n Signed-off-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit 3c2588fab65fb083873c6c1e74a1be54345615eb\nAuthor: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\nDate: Thu Jul 25 15:56:17 2019 +0800\n\n i2c: sprd: Change to use devm_platform_ioremap_resource()\n \n Use the new helper that wraps the calls to platform_get_resource()\n and devm_ioremap_resource() together.\n \n Signed-off-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit 4d7802aa434a39921c699683c4ca2ad9068b0033\nAuthor: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\nDate: Thu Jul 25 15:56:16 2019 +0800\n\n i2c: sprd: Make I2C driver can be built as a module\n \n Now there is no need to keep our I2C driver to be initialized so early,\n thus changing to module level and let it can be built as a module,\n meanwhile adding some module information.\n \n Signed-off-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit 3e99834cc0c7f0612dc790ad7342f18c375d285b\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Fri Jul 26 16:14:21 2019 +0300\n\n i2c: Drop unneeded check for of_node\n \n of_find_property() will return NULL if of_node is NULL,\n thus of_irq_get_by_name() returns -EINVAL which we ignore,\n so no need to double check.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit f91b2ab0e0c5f7cb4b11012748a184b91a396f0f\nAuthor: Shaokun Zhang \u003czhangshaokun@hisilicon.com\u003e\nDate: Mon Aug 5 17:31:08 2019 +0800\n\n i2c: designware: Fix unused variable warning\n \n drivers/i2c/busses/i2c-designware-master.c: In function ‘i2c_dw_init_recovery_info’:\n drivers/i2c/busses/i2c-designware-master.c:658:6: warning: unused variable ‘r’ [-Wunused-variable]\n int r;\n ^\n Fixes: 33eb09a02e8d (\"i2c: designware: make use of devm_gpiod_get_optional\")\n Signed-off-by: Shaokun Zhang \u003czhangshaokun@hisilicon.com\u003e\n Acked-by: Jarkko Nikula \u003cjarkko.nikula@linux.intel.com\u003e\n Reviewed-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit e611ee0b3b06b4d28216567a251e4de46e595b5d\nAuthor: Wolfram Sang \u003cwsa@the-dreams.de\u003e\nDate: Tue Aug 6 22:19:32 2019 +0200\n\n Revert \"dt-bindings: i2c: riic: Rename bindings documentation file\"\n \n This reverts commit 684ca71259a69c5a3019da72fe718bf983841926.\n I overlooked that the Rev-by tag was given in advance assuming a fix was\n made for the next revision.\n \n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit f29b7f39c0acbc66f18309227719460adab6851d\nAuthor: Wolfram Sang \u003cwsa@the-dreams.de\u003e\nDate: Tue Aug 6 22:17:51 2019 +0200\n\n Revert \"dt-bindings: i2c: rcar: Rename bindings documentation file\"\n \n This reverts commit d13ed84b195cc6e5789b446f07aede357939f7ad.\n I overlooked that the Rev-by tag was given in advance assuming a fix was\n made for the next revision.\n \n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit b19d955055480ac4e03f5afec0ca80f0de7b7013\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Tue Aug 6 16:19:56 2019 +0300\n\n drop_monitor: Use pre_doit / post_doit hooks\n \n Each operation from user space should be protected by the global drop\n monitor mutex. Use the pre_doit / post_doit hooks to take / release the\n lock instead of doing it explicitly in each function.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 965100966efe85e636178166fbf006e9b74f78d4\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Tue Aug 6 16:19:55 2019 +0300\n\n drop_monitor: Add extack support\n \n Add various extack messages to make drop_monitor more user friendly.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ff3818ca39c9f9ce07c4d50db594b9673dfa422c\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Tue Aug 6 16:19:54 2019 +0300\n\n drop_monitor: Avoid multiple blank lines\n \n Remove multiple blank lines which are visually annoying and useless.\n \n This suppresses the \"Please don\u0027t use multiple blank lines\" checkpatch\n messages.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 01921d53f870653f04ebf8d3c375029ee3ca4a31\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Tue Aug 6 16:19:53 2019 +0300\n\n drop_monitor: Document scope of spinlock\n \n While \u0027per_cpu_dm_data\u0027 is a per-CPU variable, its \u0027skb\u0027 and\n \u0027send_timer\u0027 fields can be accessed concurrently by the CPU sending the\n netlink notification to user space from the workqueue and the CPU\n tracing kfree_skb(). This spinlock is meant to protect against that.\n \n Document its scope and suppress the checkpatch message \"spinlock_t\n definition without comment\".\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit dbf896b70d4a4bdbd10aec060af42d7c70e6a88d\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Tue Aug 6 16:19:52 2019 +0300\n\n drop_monitor: Rename and document scope of mutex\n \n The \u0027trace_state_mutex\u0027 does not only protect the global \u0027trace_state\u0027\n variable, but also the global \u0027hw_stats_list\u0027.\n \n Subsequent patches are going add more operations from user space to\n drop_monitor and these all need to be mutually exclusive.\n \n Rename \u0027trace_state_mutex\u0027 to the more fitting \u0027net_dm_mutex\u0027 name and\n document its scope.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 2230a7ef5198632bdbf844fcf0abdd7958a6ac7d\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Tue Aug 6 16:19:51 2019 +0300\n\n drop_monitor: Use correct error code\n \n The error code \u0027ENOTSUPP\u0027 is reserved for use with NFS. Use \u0027EOPNOTSUPP\u0027\n instead.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 267df70fe81b0b89502ca23ca7aeb9409ec45bdf\nAuthor: Marek Vasut \u003cmarex@denx.de\u003e\nDate: Tue Aug 6 15:06:09 2019 +0200\n\n net: dsa: ksz: Drop NET_DSA_TAG_KSZ9477\n \n This Kconfig option is unused, drop it.\n \n Signed-off-by: Marek Vasut \u003cmarex@denx.de\u003e\n Cc: Andrew Lunn \u003candrew@lunn.ch\u003e\n Cc: David S. Miller \u003cdavem@davemloft.net\u003e\n Cc: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Cc: Tristram Ha \u003cTristram.Ha@microchip.com\u003e\n Cc: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Cc: Woojung Huh \u003cwoojung.huh@microchip.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6a7abc610250b28675e98e0bf2fb5634d810c8d5\nAuthor: Marek Vasut \u003cmarex@denx.de\u003e\nDate: Tue Aug 6 15:06:08 2019 +0200\n\n net: dsa: ksz: Merge ksz_priv.h into ksz_common.h\n \n Merge the two headers into one, no functional change.\n \n Signed-off-by: Marek Vasut \u003cmarex@denx.de\u003e\n Cc: Andrew Lunn \u003candrew@lunn.ch\u003e\n Cc: David S. Miller \u003cdavem@davemloft.net\u003e\n Cc: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Cc: Tristram Ha \u003cTristram.Ha@microchip.com\u003e\n Cc: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Cc: Woojung Huh \u003cwoojung.huh@microchip.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ffc60b55d9ccd9950a77686d14da69eb6e437394\nAuthor: Marek Vasut \u003cmarex@denx.de\u003e\nDate: Tue Aug 6 15:06:07 2019 +0200\n\n net: dsa: ksz: Remove dead code and fix warnings\n \n Remove ksz_port_cleanup(), which is unused. Add missing include\n \"ksz_common.h\", which fixes the following warning when built with\n make ... W\u003d1\n \n drivers/net/dsa/microchip/ksz_common.c:23:6: warning: no previous prototype for ‘...’ [-Wmissing-prototypes]\n \n Note that the order of the headers cannot be swapped, as that would\n trigger missing forward declaration errors, which would indicate the\n way forward is to merge the two headers into one.\n \n Signed-off-by: Marek Vasut \u003cmarex@denx.de\u003e\n Cc: Andrew Lunn \u003candrew@lunn.ch\u003e\n Cc: David S. Miller \u003cdavem@davemloft.net\u003e\n Cc: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Cc: Tristram Ha \u003cTristram.Ha@microchip.com\u003e\n Cc: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Cc: Woojung Huh \u003cwoojung.huh@microchip.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 2a1e00c3c0d37f65241236d7731ef6bb92f0d07f\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Mon Aug 5 22:51:38 2019 +0530\n\n gpu: drm: amd: powerplay: Remove logically dead code\n \n Result of pointer airthmentic is never null\n \n fix coverity defect:1451876\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b56a53db4ce10320a27224d4b2f83f9aaec27c89\nAuthor: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\nDate: Sun Aug 4 13:37:13 2019 -0700\n\n drm/amd/powerplay: Zero initialize some variables\n \n Clang warns (only Navi warning shown but Arcturus warns as well):\n \n drivers/gpu/drm/amd/amdgpu/../powerplay/navi10_ppt.c:1534:4: warning:\n variable \u0027asic_default_power_limit\u0027 is used uninitialized whenever \u0027?:\u0027\n condition is false [-Wsometimes-uninitialized]\n smu_read_smc_arg(smu, \u0026asic_default_power_limit);\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/gpu/drm/amd/amdgpu/../powerplay/inc/amdgpu_smu.h:588:3: note:\n expanded from macro \u0027smu_read_smc_arg\u0027\n ((smu)-\u003efuncs-\u003eread_smc_arg? (smu)-\u003efuncs-\u003eread_smc_arg((smu), (arg)) : 0)\n ^~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/gpu/drm/amd/amdgpu/../powerplay/navi10_ppt.c:1550:30: note:\n uninitialized use occurs here\n smu-\u003edefault_power_limit \u003d asic_default_power_limit;\n ^~~~~~~~~~~~~~~~~~~~~~~~\n drivers/gpu/drm/amd/amdgpu/../powerplay/navi10_ppt.c:1534:4: note:\n remove the \u0027?:\u0027 if its condition is always true\n smu_read_smc_arg(smu, \u0026asic_default_power_limit);\n ^\n drivers/gpu/drm/amd/amdgpu/../powerplay/inc/amdgpu_smu.h:588:3: note:\n expanded from macro \u0027smu_read_smc_arg\u0027\n ((smu)-\u003efuncs-\u003eread_smc_arg? (smu)-\u003efuncs-\u003eread_smc_arg((smu), (arg)) : 0)\n ^\n drivers/gpu/drm/amd/amdgpu/../powerplay/navi10_ppt.c:1517:35: note:\n initialize the variable \u0027asic_default_power_limit\u0027 to silence this\n warning\n uint32_t asic_default_power_limit;\n ^\n \u003d 0\n 1 warning generated.\n \n As the code is currently written, if read_smc_arg were ever NULL, arg\n would fail to be initialized but the code would continue executing as\n normal because the return value would just be zero.\n \n There are a few different possible solutions to resolve this class\n of warnings which have appeared in these drivers before:\n \n 1. Assume the function pointer will never be NULL and eliminate the\n wrapper macros.\n \n 2. Have the wrapper macros initialize arg when the function pointer is\n NULL.\n \n 3. Have the wrapper macros return an error code instead of 0 when the\n function pointer is NULL so that the callsites can properly bail out\n before arg can be used.\n \n 4. Initialize arg at the top of its function.\n \n Number four is the path of least resistance right now as every other\n change will be driver wide so do that here. I only make the comment\n now as food for thought.\n \n Fixes: b4af964e75c4 (\"drm/amd/powerplay: make power limit retrieval as asic specific\")\n Link: https://github.com/ClangBuiltLinux/linux/issues/627\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f848be467c98e726b130b1726706f6a84f85a046\nAuthor: Oded Gabbay \u003coded.gabbay@gmail.com\u003e\nDate: Thu Jul 4 09:32:20 2019 +0300\n\n MAINTAINERS: update amdkfd maintainer (v3)\n \n I\u0027m leaving the role of amdkfd maintainer. Therefore, update the relevant\n entry in the MAINTAINERS file with the name of the new maintainer.\n \n Good Luck!\n \n v3: update mailing list, file list (Alex)\n \n Reviewed-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e (v2)\n Signed-off-by: Oded Gabbay \u003coded.gabbay@gmail.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b5507c7e06076d874923aa958ddb8829e3cfc573\nAuthor: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\nDate: Fri Aug 2 16:48:08 2019 -0400\n\n drm/amdgpu: Fix GPU reset crash regression.\n \n amdgpu_ip_block.status.hw for GMC wasn\u0027t set to\n false on suspend during GPU reset and so on resume gmc_v9_0_resume\n wasn\u0027t called.\n Caused by \u0027drm/amdgpu: fix double ucode load by PSP(v3)\u0027\n \n Signed-off-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8dd45504a7d70a23be6e4415edbac0e197935686\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Tue Aug 6 16:14:22 2019 +0800\n\n drm/amd/powerplay: check before issuing messages for max sustainable clocks\n \n Those messages are not supported on Arcturus and should not be\n issued.\n \n Affected ASIC: Arcturus\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b5c73856408b8431eeb29bc1f7d41dbe7549b75b\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Mon Aug 5 16:19:45 2019 +0800\n\n drm/amdgpu/discovery: move common discovery code out of navi1*_reg_base_init()\n \n move amdgpu_discovery_reg_base_init() from navi1*_reg_base_init() to a\n common function nv_reg_base_init().\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 35ef88fa115d0810d28d960b951e7260ed46ad59\nAuthor: tiancyin \u003ctianci.yin@amd.com\u003e\nDate: Mon Aug 5 17:32:45 2019 +0800\n\n drm/amdgpu/soc15: fix external_rev_id for navi14\n \n fix the hard code external_rev_id.\n \n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: tiancyin \u003ctianci.yin@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2a3c7ff6e37c0d0bf8d8e9031922d41c2409d7e5\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Mon Aug 5 15:48:30 2019 +0800\n\n drm/amdgpu: update ras sysfs feature info\n \n remove confused ras error type info\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9234c587a81f360b80aeb7bed643d1405627d6c9\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Mon Aug 5 14:53:12 2019 +0800\n\n drm/amd/powerplay: skip pcie params override on Arcturus V2\n \n This is not supported on Arcturus.\n \n Affected ASIC: Arcturus\n \n V2: minor cosmetic fix\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Le Ma \u003cLe.Ma@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 876923fb92a9e298625067284977917d4741ee2e\nAuthor: xinhui pan \u003cxinhui.pan@amd.com\u003e\nDate: Mon Aug 5 14:53:49 2019 +0800\n\n drm/amdgpu: Fix panic during gpu reset\n \n Clear the flag after hw suspend, otherwise it skips the corresponding hw\n resume.\n \n Signed-off-by: xinhui pan \u003cxinhui.pan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1f288afc2cb74e201bf8962d1c4d2f4dff2add51\nAuthor: Likun Gao \u003cLikun.Gao@amd.com\u003e\nDate: Fri Aug 2 15:18:57 2019 +0800\n\n drm/amdgpu: pin the csb buffer on hw init for gfx v8\n \n Without this pin, the csb buffer will be filled with inconsistent\n data after S3 resume. And that will causes gfx hang on gfxoff\n exit since this csb will be executed then.\n \n Signed-off-by: Likun Gao \u003cLikun.Gao@amd.com\u003e\n Tested-by: Paul Gover \u003cpmw.gover@yahoo.co.uk\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Reviewed-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit caff0e6654377b623e37922d084b1a33aa9d44d7\nAuthor: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\nDate: Fri Aug 2 10:45:11 2019 -0400\n\n drm/amd/display: Block immediate flips for non-fast updates\n \n [Why]\n Underflow can occur in the case where we change buffer pitch, DCC state,\n rotation or mirroring for a plane while also performing an immediate\n flip. It can also generate a p-state warning stack trace on DCN1 which\n is typically observed during the cursor handler pipe locking because of\n how frequent cursor updates can occur.\n \n [How]\n Store the update type on each CRTC - every plane will have access to\n the CRTC state if it\u0027s flipping. If the update type is not\n UPDATE_TYPE_FAST then the immediate flip should be disallowed.\n \n No changes to the target vblank sequencing need to be done, we just\n need to ensure that the surface registers do a double buffered update.\n \n Signed-off-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Reviewed-by: David Francis \u003cdavid.francis@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2cc450ceaae4febc1539975247d235eb06c11979\nAuthor: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\nDate: Fri Aug 2 10:31:29 2019 -0400\n\n drm/amd/display: Validate dc_plane_info and dc_plane_size in atomic check\n \n [Why]\n Pitch, DCC, rotation and mirroring can result in updates that are not\n UPDATE_TYPE_FAST but UPDATE_TYPE_MED instead. DC needs dc_plane_info\n and dc_plane_size to make this determination and we aren\u0027t currently\n passing this into DC during atomic check.\n \n Underflow (visible or non-visible) can occur if we don\u0027t validate this\n correctly. This also will generally trigger p-state warnings, typically\n via the cursor handler when locking.\n \n [How]\n Get the framebuffer tiling flags and generate the required structures\n for DC in dm_determine_update_type_for_commit.\n \n Signed-off-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Reviewed-by: David Francis \u003cdavid.francis@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit aed5a8df3dbbd804409d40f821e0768891cfd5ab\nAuthor: Suren Baghdasaryan \u003csurenb@google.com\u003e\nDate: Fri Jul 26 09:22:26 2019 -0700\n\n tests: add pidfd poll tests\n \n This adds testing for polling on pidfd of a process being killed. Test runs\n 10000 iterations by default to stress test pidfd polling functionality.\n It accepts an optional command-line parameter to override the number or\n iterations to run.\n Specifically, it tests for:\n - pidfd_open on a child process succeeds\n - pidfd_send_signal on a child process succeeds\n - polling on pidfd succeeds and returns exactly one event\n - returned event is POLLIN\n - event is received within 3 secs of the process being killed\n \n 10000 iterations was chosen because of the race condition being tested\n which is not consistently reproducible but usually is revealed after less\n than 2000 iterations.\n Reveals race fixed by commit b191d6491be6 (\"pidfd: fix a poll race when setting exit_state\")\n \n Signed-off-by: Suren Baghdasaryan \u003csurenb@google.com\u003e\n Link: https://lore.kernel.org/r/20190726162226.252750-2-surenb@google.com\n Signed-off-by: Christian Brauner \u003cchristian.brauner@ubuntu.com\u003e\n\ncommit 70433f67ec3a54710744902d782f8954325e25b8\nAuthor: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nDate: Tue Aug 6 12:15:19 2019 +1000\n\n MODSIGN: make new include file self contained\n \n Fixes: c8424e776b09 (\"MODSIGN: Export module signature definitions\")\n Signed-off-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n\ncommit 2ec2f99abd2c75b507410c5def731fe1b483271b\nAuthor: Suren Baghdasaryan \u003csurenb@google.com\u003e\nDate: Fri Jul 26 09:22:25 2019 -0700\n\n tests: move common definitions and functions into pidfd.h\n \n Move definitions and functions used across different pidfd tests into\n pidfd.h header.\n \n Signed-off-by: Suren Baghdasaryan \u003csurenb@google.com\u003e\n Reviewed-by: Christian Brauner \u003cchristian@brauner.io\u003e\n Link: https://lore.kernel.org/r/20190726162226.252750-1-surenb@google.com\n Signed-off-by: Christian Brauner \u003cchristian.brauner@ubuntu.com\u003e\n\ncommit e63f3085709e92bd876e2977a07099d56bbda1c2\nAuthor: Christian Brauner \u003cchristian.brauner@ubuntu.com\u003e\nDate: Sun Jul 28 00:22:30 2019 +0200\n\n pidfd: add pidfd_wait tests\n \n Add tests for pidfd_wait() and CLONE_WAIT_PID:\n - test that waitid(P_PIDFD) fails on /proc/\u003cpid\u003e\n - test that waitid(P_PIDFD) fails on /dev/null\n - test that waitid(P_PIDFD) can wait on a pidfd\n - test that waitid(P_PIDFD) can wait on a pidfd and return siginfo_t\n - test that waitid(P_PIDFD) works with WEXITED\n - test that waitid(P_PIDFD) works with WSTOPPED\n - test that waitid(P_PIDFD) works with WUNTRACED\n - test that waitid(P_PIDFD) works with WCONTINUED\n - test that waitid(P_PIDFD) works with WNOWAIT\n - test that waitid(P_PIDFD)works with WNOHANG\n \n Signed-off-by: Christian Brauner \u003cchristian.brauner@ubuntu.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\n Cc: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: David Howells \u003cdhowells@redhat.com\u003e\n Cc: Jann Horn \u003cjannh@google.com\u003e\n Cc: Andy Lutomirsky \u003cluto@kernel.org\u003e\n Cc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n Cc: Oleg Nesterov \u003coleg@redhat.com\u003e\n Cc: Aleksa Sarai \u003ccyphar@cyphar.com\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n Link: https://lore.kernel.org/r/20190727222229.6516-3-christian@brauner.io\n\ncommit aa48e31b874774bef42fd2b433474202cbe8ac79\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Fri Aug 2 09:21:35 2019 +0800\n\n mailmap: add entry for Jaegeuk Kim\n \n Add entry to connect all Jaegeuk\u0027s email addresses.\n \n Acked-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 758f2512546f70fdb70bd1d752d975a09ad9741b\nAuthor: Todor Tomov \u003ctodor.too@gmail.com\u003e\nDate: Thu Aug 1 20:26:03 2019 +0300\n\n mailmap: Add an entry for my email address\n \n My @linaro.org email address doesn\u0027t exist anymore so add a\n mailmap entry to map it to my @gmail.com address.\n \n Signed-off-by: Todor Tomov \u003ctodor.too@gmail.com\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit a4a03bdccfd02eda6613a892fb418f15adeb2b21\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Thu Aug 1 11:17:50 2019 +0800\n\n mailmap: add entry for Gao Xiang\n \n Add entry to connect all Gao Xiang\u0027s email addresses.\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Acked-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 178e5f3a5bc1d67d1248a74c0abab41040abe7c4\nAuthor: Brian Starkey \u003cbrian.starkey@arm.com\u003e\nDate: Tue Aug 6 13:46:22 2019 +0100\n\n drm/crc-debugfs: Add notes about CRC\u003c-\u003ecommit interactions\n \n CRC generation can be impacted by commits coming from userspace, and\n enabling CRC generation may itself trigger a commit. Add notes about\n this to the kerneldoc.\n \n Changes since v1:\n - Clarified that anything that would disable CRCs counts as a full\n modeset, and so userspace needs to reconfigure after full modesets\n \n Changes since v2:\n - Add these notes\n - Rebase onto drm-misc-next (trivial conflict in comment)\n \n Signed-off-by: Brian Starkey \u003cbrian.starkey@arm.com\u003e\n Signed-off-by: Ayan Kumar Halder \u003cayan.halder@arm.com\u003e\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n \n Link:- https://patchwork.freedesktop.org/patch/321974/\n\ncommit ac862503639ff824e0e1f32fbca220ed75261138\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Mon Aug 5 11:49:51 2019 +0100\n\n Input: docs: fix spelling mistake \"potocol\" -\u003e \"protocol\"\n \n There is a minor spelling mistake in the documentation, fix it.\n \n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 95e760cbf62755bd3d8999bee1e6579f0ff1425e\nAuthor: Randy Dunlap \u003crdunlap@infradead.org\u003e\nDate: Mon Aug 5 09:29:50 2019 -0700\n\n kernel-doc: ignore __printf attribute\n \n Ignore __printf() function attributes just as other __attribute__\n strings are ignored.\n \n Fixes this kernel-doc warning message:\n include/kunit/kunit-stream.h:58: warning: Function parameter or member \u00272\u0027 not described in \u0027__printf\u0027\n \n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\n Cc: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\n Tested-by: Brendan Higgins \u003cbrendanhiggins@google.com\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit c5e5c48c16422521d363c33cfb0dcf58f88c119b\nAuthor: chenzefeng \u003cchenzefeng2@huawei.com\u003e\nDate: Tue Aug 6 15:46:33 2019 +0800\n\n ia64:unwind: fix double free for mod-\u003earch.init_unw_table\n \n The function free_module in file kernel/module.c as follow:\n \n void free_module(struct module *mod) {\n ......\n module_arch_cleanup(mod);\n ......\n module_arch_freeing_init(mod);\n ......\n }\n \n Both module_arch_cleanup and module_arch_freeing_init function\n would free the mod-\u003earch.init_unw_table, which cause double free.\n \n Here, set mod-\u003earch.init_unw_table \u003d NULL after remove the unwind\n table to avoid double free.\n \n Signed-off-by: chenzefeng \u003cchenzefeng2@huawei.com\u003e\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 9ce1263033cd2ad393e2ff0df4a1c4ab4992c9df\nAuthor: Andrey Konovalov \u003candreyknvl@google.com\u003e\nDate: Tue Jul 23 19:58:52 2019 +0200\n\n selftests, arm64: add a selftest for passing tagged pointers to kernel\n \n This patch is a part of a series that extends kernel ABI to allow to pass\n tagged user pointers (with the top byte set to something else other than\n 0x00) as syscall arguments.\n \n This patch adds a simple test, that calls the uname syscall with a\n tagged user pointer as an argument. Without the kernel accepting tagged\n user pointers the test fails with EFAULT.\n \n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Acked-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Signed-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 63f0c60379650d82250f22e4cf4137ef3dc4f43d\nAuthor: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nDate: Tue Jul 23 19:58:39 2019 +0200\n\n arm64: Introduce prctl() options to control the tagged user addresses ABI\n \n It is not desirable to relax the ABI to allow tagged user addresses into\n the kernel indiscriminately. This patch introduces a prctl() interface\n for enabling or disabling the tagged ABI with a global sysctl control\n for preventing applications from enabling the relaxed ABI (meant for\n testing user-space prctl() return error checking without reconfiguring\n the kernel). The ABI properties are inherited by threads of the same\n application and fork()\u0027ed children but cleared on execve(). A Kconfig\n option allows the overall disabling of the relaxed ABI.\n \n The PR_SET_TAGGED_ADDR_CTRL will be expanded in the future to handle\n MTE-specific settings like imprecise vs precise exceptions.\n \n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Signed-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 2b835e24b5c6f9c633ff51973581ee7ca7b3e8ec\nAuthor: Andrey Konovalov \u003candreyknvl@google.com\u003e\nDate: Tue Jul 23 19:58:38 2019 +0200\n\n arm64: untag user pointers in access_ok and __uaccess_mask_ptr\n \n This patch is a part of a series that extends kernel ABI to allow to pass\n tagged user pointers (with the top byte set to something else other than\n 0x00) as syscall arguments.\n \n copy_from_user (and a few other similar functions) are used to copy data\n from user memory into the kernel memory or vice versa. Since a user can\n provided a tagged pointer to one of the syscalls that use copy_from_user,\n we need to correctly handle such pointers.\n \n Do this by untagging user pointers in access_ok and in __uaccess_mask_ptr,\n before performing access validity checks.\n \n Note, that this patch only temporarily untags the pointers to perform the\n checks, but then passes them as is into the kernel internals.\n \n Reviewed-by: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\n [will: Add __force to casting in untagged_addr() to kill sparse warning]\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 033a856c73683924e28e329502572c413beb86c0\nAuthor: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\nDate: Fri Aug 2 11:38:56 2019 -0700\n\n drm/i915/dmc: Load DMC on TGL\n \n Add Support to load DMC v2.03 on TGL.\n \n v2: Use version 2.03 that is already available since that works with\n PSR2\n \n Signed-off-by: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Matt Atwood \u003cmatthew.s.atwood@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190802183856.27280-1-lucas.demarchi@intel.com\n\ncommit 8c30396074c131765b19eb3cb7ff764a4f2f2913\nAuthor: Alexei Starovoitov \u003cast@kernel.org\u003e\nDate: Fri Aug 2 16:23:40 2019 -0700\n\n selftests/bpf: add loop test 5\n \n Add a test with multiple exit conditions.\n It\u0027s not an infinite loop only when the verifier can properly track\n all math on variable \u0027i\u0027 through all possible ways of executing this loop.\n \n barrier()s are needed to disable llvm optimization that combines multiple\n branches into fewer branches.\n \n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n Acked-by: Yonghong Song \u003cyhs@fb.com\u003e\n\ncommit a78d0dbec712999ecd2f800eea0f62dc93867a78\nAuthor: Alexei Starovoitov \u003cast@kernel.org\u003e\nDate: Fri Aug 2 15:54:01 2019 -0700\n\n selftests/bpf: add loop test 4\n \n Add a test that returns a \u0027random\u0027 number between [0, 2^20)\n If state pruning is not working correctly for loop body the number of\n processed insns will be 2^20 * num_of_insns_in_loop_body and the program\n will be rejected.\n \n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n Acked-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Yonghong Song \u003cyhs@fb.com\u003e\n\ncommit 00ec4f3039a9e36cbccd1aea82d06c77c440a51c\nAuthor: Hans Holmberg \u003chans@owltronix.com\u003e\nDate: Wed Jul 31 11:41:36 2019 +0200\n\n block: stop exporting bio_map_kern\n \n Now that there no module users left of bio_map_kern, stop exporting the\n symbol.\n \n Reviewed-by: Javier González \u003cjavier@javigon.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Hans Holmberg \u003chans@owltronix.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit ff8f352070b204ab1721f41339547d897f8120fd\nAuthor: Hans Holmberg \u003chans@owltronix.com\u003e\nDate: Wed Jul 31 11:41:35 2019 +0200\n\n lightnvm: pblk: use kvmalloc for metadata\n \n There is no reason now not to use kvmalloc, so replace the internal\n metadata allocation scheme.\n \n Reviewed-by: Javier González \u003cjavier@javigon.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Hans Holmberg \u003chans@owltronix.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 48e5da725581c1f7444e45cccbafc33e11430b48\nAuthor: Hans Holmberg \u003chans@owltronix.com\u003e\nDate: Wed Jul 31 11:41:34 2019 +0200\n\n lightnvm: move metadata mapping to lower level driver\n \n Now that blk_rq_map_kern can map both kmem and vmem, move internal\n metadata mapping down to the lower level driver.\n \n Reviewed-by: Javier González \u003cjavier@javigon.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Hans Holmberg \u003chans@owltronix.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 98d87f70f4ab84b9e50e16b7848937ae07518cd4\nAuthor: Hans Holmberg \u003chans@owltronix.com\u003e\nDate: Wed Jul 31 11:41:33 2019 +0200\n\n lightnvm: remove nvm_submit_io_sync_fn\n \n Move the redundant sync handling interface and wait for a completion in\n the lightnvm core instead.\n \n Reviewed-by: Javier González \u003cjavier@javigon.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Hans Holmberg \u003chans@owltronix.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 8df92d676c919040a9d0e05d3707f00367c88a22\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Sun Aug 4 21:19:48 2019 +0530\n\n pinctrl: spear: spear: Add of_node_put() before return\n \n Each iteration of for_each_child_of_node puts the previous node, but in\n the case of a return from the middle of the loop, there is no put, thus\n causing a memory leak. Hence add an of_node_put before the return in\n two places.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Acked-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190804154948.4584-1-nishkadg.linux@gmail.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 750e76b4f9f63c95bf4c283ccf8e2af0e258d3bb\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Aug 6 13:43:00 2019 +0100\n\n drm/i915/gt: Move the [class][inst] lookup for engines onto the GT\n \n To maintain a fast lookup from a GT centric irq handler, we want the\n engine lookup tables on the intel_gt. To avoid having multiple copies of\n the same multi-dimension lookup table, move the generic user engine\n lookup into an rbtree (for fast and flexible indexing).\n \n v2: Split uabi_instance cf uabi_class\n v3: Set uabi_class/uabi_instance after collating all engines to provide a\n stable uabi across parallel unordered construction.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e #v2\n Link: https://patchwork.freedesktop.org/patch/msgid/20190806124300.24945-2-chris@chris-wilson.co.uk\n\ncommit c29579d2fabe7448a444681d8229384249d315f9\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Aug 6 13:42:59 2019 +0100\n\n drm/i915/gem: Make caps.scheduler static\n \n We do not notify userspace when the scheduler capabilities are changed\n (due to wedging the driver) and as such userspace will expect the caps\n to be static and unchanging. Make it so, and so we only need to compute\n our caps once during driver registration.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@linux.intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190806124300.24945-1-chris@chris-wilson.co.uk\n\ncommit ea651ffd4f7febd4a0b3fdb98e39a6980e978a2c\nAuthor: Gustavo Pimentel \u003cgustavo.pimentel@synopsys.com\u003e\nDate: Tue Jul 23 17:17:28 2019 +0200\n\n reset: Add DesignWare IP support to simple reset\n \n The reset-simple driver can be now used on DesignWare IPs by\n default by selecting the following compatible strings:\n - snps,dw-high-reset for active high resets inputs\n - snps,dw-low-reset for active low resets inputs\n \n Signed-off-by: Gustavo Pimentel \u003cgustavo.pimentel@synopsys.com\u003e\n Signed-off-by: Luis Oliveira \u003cluis.oliveira@synopsys.com\u003e\n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit e7ef681a8ab8568100eedac1d50fe5973178147b\nAuthor: Luis Oliveira \u003cLuis.Oliveira@synopsys.com\u003e\nDate: Tue Jul 23 17:17:27 2019 +0200\n\n dt-bindings: Document the DesignWare IP reset bindings\n \n This adds documentation of device tree bindings for the\n DesignWare IP reset controller.\n \n Signed-off-by: Gustavo Pimentel \u003cgustavo.pimentel@synopsys.com\u003e\n Signed-off-by: Luis Oliveira \u003cluis.oliveira@synopsys.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit 646d90c3c2c19dcc85b3d1c7c43a26e900d4074f\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Sun Aug 4 21:34:20 2019 +0530\n\n pinctrl: freescale: mxs: Add of_node_put() before return\n \n Each iteration of for_each_child_of_node puts the previous node, but in\n the case of a return from the middle of the loop, there is no put, thus\n causing a memory leak. Hence add an of_node_put before the return in\n three places.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190804160420.5309-1-nishkadg.linux@gmail.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit f452444709c07ea5d126e9f312f602ac9c5e34fd\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Sun Aug 4 21:21:54 2019 +0530\n\n pinctrl: nomadik: abx500: Add of_node_put() before return\n \n Each iteration of for_each_child_of_node puts the previous node, but in\n the case of a return from the middle of the loop, there is no put, thus\n causing a memory leak. Hence add an of_node_put before the return.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190804155154.4916-1-nishkadg.linux@gmail.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit ea8cf5c518502a12cb5bab935fd7bc152822e520\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Sun Aug 4 21:21:17 2019 +0530\n\n pinctrl: nomadik: nomadik: Add of_node_put() before return\n \n Each iteration of for_each_child_of_node puts the previous node, but in\n the case of a return from the middle of the loop, there is no put, thus\n causing a memory leak. Hence add an of_node_put before the return.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190804155117.4753-1-nishkadg.linux@gmail.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit f17d2f54d36dd316bcff4e619a1a349f04214a83\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Sun Aug 4 20:57:45 2019 +0530\n\n pinctrl: falcon: Add of_node_put() before return\n \n Each iteration of for_each_compatible_node puts the previous node, but in\n the case of a return from the middle of the loop, there is no put, thus\n causing a memory leak. Hence add an of_node_put before the return in two\n places.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190804152745.2231-1-nishkadg.linux@gmail.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 6004cf298a4180199dc40bc40466126df8a5a88c\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Fri Aug 2 13:58:40 2019 +0200\n\n b43legacy: Remove pointless cond_resched() wrapper\n \n cond_resched() can be used unconditionally. If CONFIG_PREEMPT is set, it\n becomes a NOP scheduler wise.\n \n Also the B43_BUG_ON() in that wrapper is a homebrewn variant of\n __might_sleep() which is part of cond_resched() already.\n \n Remove the cruft and invoke cond_resched() directly.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\n Tested-by: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 654026df2635863fd695b2ca833e5c62454bd5ee\nAuthor: Brian Norris \u003cbriannorris@chromium.org\u003e\nDate: Mon Aug 5 10:15:04 2019 -0700\n\n Revert \"mwifiex: fix system hang problem after resume\"\n \n This reverts commit 437322ea2a36d112e20aa7282c869bf924b3a836.\n \n This above-mentioned \"fix\" does not actually do anything to prevent a\n race condition. It simply papers over it so that the issue doesn\u0027t\n appear.\n \n If this is a real problem, it should be explained better than the above\n commit does, and an alternative, non-racy solution should be found.\n \n For further reason to revert this: there\u0027s no reason we can\u0027t try\n resetting the card when it\u0027s *actually* stuck in host-sleep mode. So\n instead, this is unnecessarily creating scenarios where we can\u0027t recover\n Wifi (and in fact, I\u0027m fielding reports of Chromebooks that can\u0027t\n recover after the aforementioned commit).\n \n Note that this was proposed in 2017 and Ack\u0027ed then, but due to my\n marking as RFC, it never went anywhere:\n \n https://patchwork.kernel.org/patch/9657277/\n [RFC] Revert \"mwifiex: fix system hang problem after resume\"\n \n Cc: Amitkumar Karwar \u003cakarwar@marvell.com\u003e\n Signed-off-by: Brian Norris \u003cbriannorris@chromium.org\u003e\n Reviewed-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n Acked-by: Amitkumar Karwar \u003camitkarwar@gmail.com\u003e\n Tested-by: Matthias Kaehlcke \u003cmka@chromium.org\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit c85a6376673805a68633e5d7d7d1fccf4b207685\nAuthor: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\nDate: Thu Aug 1 20:21:02 2019 -0500\n\n rtlwifi: rtl8192ce: Convert inline routines to little-endian words\n \n In this step, the read/write routines for the descriptors are converted\n to use __le32 quantities, thus a lot of casts can be removed. Callback\n routines still use the 8-bit arrays, but these are changed within the\n specified routine.\n \n The macro that cleared a descriptor has now been converted into an inline\n routine.\n \n Signed-off-by: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 98fd8db59a007c05f30f911180f745b243d56746\nAuthor: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\nDate: Thu Aug 1 20:21:01 2019 -0500\n\n rtlwifi: rtl8192ce: Convert macros that set descriptor\n \n As a first step in the conversion, the macros that set the RX and TX\n descriptors are converted to static inline routines, and the names are\n changed from upper to lower case. To minimize the changes in a given\n step, the input descriptor information is left as as a byte array\n (u8 *), even though it should be a little-endian word array (__le32 *).\n That will be changed in the next patch.\n \n Several places where checkpatch.pl complains about a space after a cast\n are fixed.\n \n Signed-off-by: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit a246b92914339e539b5fd27b347601fc9f0e8460\nAuthor: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\nDate: Thu Aug 1 20:21:00 2019 -0500\n\n rtlwifi: rtl8192ce: Replace local bit manipulation macros\n \n This driver uses a set of local macros to manipulate the RX and TX\n descriptors, which are all little-endian quantities. These macros\n are replaced by the bitfield macros le32p_replace_bits() and\n le32_get_bits(). In several places, the macros operated on an entire\n 32-bit word. In these cases, a direct read or replacement is used.\n \n Signed-off-by: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 49a52d05a3c7252357ab6b1e49c17ed0c02cf80f\nAuthor: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\nDate: Thu Aug 1 20:20:59 2019 -0500\n\n rtlwifi: rtl8192ce: Remove unused GET_XXX and SET_XXX\n \n As the first step in converting from macros that get/set information\n in the RX and TX descriptors, unused macros are being removed.\n \n Signed-off-by: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 4136214f7c46839c15f0f177fe1d5052302c0205\nAuthor: Yan-Hsuan Chuang \u003cyhchuang@realtek.com\u003e\nDate: Wed Jul 31 20:22:47 2019 +0800\n\n rtw88: add BT co-existence support\n \n Both RTL8822BE/RTL8822CE are WiFi + BT combo chips. Since\n WiFi and BT use 2.4GHz to transmit, it is important to\n make sure they run concurrently without interfering each\n other. To achieve this, WiFi driver requires a mechanism\n to collaborate with BT, whether they share the antenna(s)\n or not.\n \n The final decision made by the co-existence mechanism is\n to choose a proper strategy, or called \"tdma/table\", and\n inform either firmware or hardware of the strategy.\n To choose a strategy, co-existence mechanism needs to\n have enough information from WiFi and BT.\n \n BT information is provided through firmware C2H.\n The contents describe the current status of BT, such as\n if BT is connected or is idle, or the profile that is\n being used.\n \n WiFi information can be provided by WiFi itself. The WiFi\n driver will call various of \"notify\" functions each time\n the state of WiFi changed, such as WiFi is going to switch\n channel or is connected. Also WiFi driver can know if it\n shares antenna with BT by reading efuse content. Antenna\n configuration of the module will finally get a different\n strategy.\n \n Upon receiving any information from WiFi or BT, the WiFi\n driver will run the co-existence mechanism immediately.\n It will set the RF antenna configuration according to the\n strategy through the TDMA H2C to firmware and a hardware\n table. Based on the tdma/table, WiFi + BT should work with\n each other, and having a better user experience.\n \n Signed-off-by: Yan-Hsuan Chuang \u003cyhchuang@realtek.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 713a30de45a2ec8619228280e4832b5d6a34e759\nAuthor: Yan-Hsuan Chuang \u003cyhchuang@realtek.com\u003e\nDate: Wed Jul 31 20:22:46 2019 +0800\n\n rtw88: enclose c2h cmd handle with mutex\n \n C2H commands that cannot be handled in IRQ context should\n be protected by rtwdev-\u003emutex. Because they might have a\n sequece of hardware operations that does not want to be\n interfered.\n \n Signed-off-by: Yan-Hsuan Chuang \u003cyhchuang@realtek.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 0d762f031d702272a17910fbeb45ab15b9673617\nAuthor: Yan-Hsuan Chuang \u003cyhchuang@realtek.com\u003e\nDate: Wed Jul 31 20:22:45 2019 +0800\n\n rtw88: allow c2h operation in irq context\n \n Some of the c2h operations are small and can be done\n under interrupt context. For the rest that requires\n more operations or can go sleep, enqueue onto c2h queue.\n \n Signed-off-by: Yan-Hsuan Chuang \u003cyhchuang@realtek.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit d1d1a96bdb4408d02e2bfcb32b71aba165458a80\nAuthor: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\nDate: Tue Jul 30 19:33:04 2019 -0500\n\n rtlwifi: rtl8188ee: Remove local configuration variable\n \n The configuration variable IS_LITTLE_ENDIAN is replaced by the standard\n __LITTLE_ENDIAN. In addition, an unused struct is removed.\n \n Signed-off-by: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit e53e30ba15da85a21195573db28ac35ba4e6ecf0\nAuthor: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\nDate: Tue Jul 30 19:33:03 2019 -0500\n\n rtlwifi: rtl8188ee: Convert inline routines to little-endian words\n \n In this step, the read/write routines for the descriptors are converted\n to use __le32 quantities, thus a lot of casts can be removed. Callback\n routines still use the 8-bit arrays, but these are changed within the\n specified routine.\n \n The macro that cleared a descriptor has now been converted into an inline\n routine.\n \n Signed-off-by: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 36eda7568f2e4e18ba248974f73767de7778f43a\nAuthor: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\nDate: Tue Jul 30 19:33:02 2019 -0500\n\n rtlwifi: rtl8188ee: Convert macros that set descriptor\n \n As a first step in the conversion, the macros that set the RX and TX\n descriptors are converted to static inline routines, and the names are\n changed from upper to lower case. To minimize the changes in a given\n step, the input descriptor information is left as as a byte array\n (u8 *), even though it should be a little-endian word array (__le32 *).\n That will be changed in the next patch.\n \n Signed-off-by: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 66b2b064ee54fea02a68589b0513a2586c7a482b\nAuthor: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\nDate: Tue Jul 30 19:33:01 2019 -0500\n\n rtlwifi: rtl88188ee: Replace local bit manipulation macros\n \n This driver uses a set of local macros to manipulate the RX and TX\n descriptors, which are all little-endian quantities. These macros\n are replaced by the bitfield macros le32p_replace_bits() and\n le32_get_bits(). In several places, the macros operated on an entire\n 32-bit word. In these cases, a direct read or replacement is used.\n \n Signed-off-by: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit b97494f48d5a0cf0b5a8bfec861c7311db987ad2\nAuthor: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\nDate: Tue Jul 30 19:33:00 2019 -0500\n\n rtlwifi: rtl8188ee: Remove unused GET_XXX and SET_XXX descriptor macros\n \n As the first step in converting from macros that get/set information\n in the RX and TX descriptors, unused macros are being removed.\n \n Signed-off-by: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 706f0182b1add0fc41a8c40662f659b7426f0629\nAuthor: Masanari Iida \u003cstandby24x7@gmail.com\u003e\nDate: Sun Jul 28 23:07:42 2019 +0900\n\n rt2800usb: Add new rt2800usb device PLANEX GW-USMicroN\n \n This patch add a device ID for PLANEX GW-USMicroN.\n Without this patch, I had to echo the device IDs in order to\n recognize the device.\n \n # lsusb |grep PLANEX\n Bus 002 Device 005: ID 2019:ed14 PLANEX GW-USMicroN\n \n Signed-off-by: Masanari Iida \u003cstandby24x7@gmail.com\u003e\n Acked-by: Stanislaw Gruszka \u003csgruszka@redhat.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit d1b68c1182380e50cad4b7bd76ee68f64951a64b\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Jul 26 22:20:18 2019 +0800\n\n rtw88: pci: remove set but not used variable \u0027ip_sel\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/net/wireless/realtek/rtw88/pci.c: In function \u0027rtw_pci_phy_cfg\u0027:\n drivers/net/wireless/realtek/rtw88/pci.c:993:6: warning:\n variable \u0027ip_sel\u0027 set but not used [-Wunused-but-set-variable]\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit de019a3bdd6e68552e061f8c66b7cf7a18109c54\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Jul 26 22:15:35 2019 +0800\n\n brcmsmac: remove three set but not used variables\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c: In function \u0027brcms_c_set_gmode\u0027:\n drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c:5257:7: warning: variable \u0027preamble_restrict\u0027 set but not used [-Wunused-but-set-variable]\n drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c:5256:6: warning: variable \u0027preamble\u0027 set but not used [-Wunused-but-set-variable]\n drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c:5251:7: warning: variable \u0027shortslot_restrict\u0027 set but not used [-Wunused-but-set-variable]\n \n They are never used so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 937a194ae8651b434c4b4c62fe94099f41a93e14\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Jul 26 11:06:14 2019 +0100\n\n ipw2x00: remove redundant assignment to err\n \n Variable err is initialized to a value that is never read and it\n is re-assigned later. The initialization is redundant and can\n be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit dfc7a8f7c857369c3cee4aa4c12d6c8da7aa1597\nAuthor: Ganapathi Kondraju \u003cganapathirajukondraju@gmail.com\u003e\nDate: Thu Jul 25 16:21:50 2019 +0530\n\n rsi: fix for sdio reset card issue\n \n Issue: While removing and inserting the driver module, observed driver\n loading is not successful.\n \n Root cause: Card is not resetted completely without issuing cmd5.\n \n Fix: Issued cmd5 properly.\n \n Signed-off-by: Ganapathi Kondraju \u003cganapathirajukondraju@gmail.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 3915a252ce71b836a830bcc537220389a3acf3a5\nAuthor: Dan Williams \u003cdcbw@redhat.com\u003e\nDate: Wed Jul 24 09:38:15 2019 -0500\n\n libertas: Fix a double free in if_spi_c2h_data()\n \n The lbs_process_rxed_packet() frees the skb. It didn\u0027t originally, but\n we fixed it in commit f54930f36311 (\"libertas: don\u0027t leak skb on receive\n error\").\n \n Reported-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Signed-off-by: Dan Williams \u003cdcbw@redhat.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 04f7745300edbb3fbb9447586b0aafeef28abd51\nAuthor: J. Bruce Fields \u003cbfields@fieldses.org\u003e\nDate: Tue Aug 6 06:57:50 2019 -0400\n\n thunderbolt: Show key using %*pE not %*pEp\n \n %*pEp (without \"h\" or \"o\") is a no-op. This string could contain\n arbitrary (non-NULL) characters, so we do want escaping. Use %*pE like\n every other caller.\n \n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n Signed-off-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n\ncommit b95c732234fa40319c6aae2db6b35e9a7f41d79c\nAuthor: Navid Emamdoost \u003cnavid.emamdoost@gmail.com\u003e\nDate: Wed Jul 24 09:17:36 2019 -0500\n\n mt7601u: null check the allocation\n \n devm_kzalloc may fail and return NULL. So the null check is needed.\n \n Signed-off-by: Navid Emamdoost \u003cnavid.emamdoost@gmail.com\u003e\n Acked-by: Jakub Kicinski \u003ckubakici@wp.pl\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit cddecd92d1ec2fd05ed1123455e7c6cf6906b5a5\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Jul 24 22:12:01 2019 +0800\n\n brcmfmac: remove set but not used variable \u0027dtim_period\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c: In function brcmf_update_bss_info:\n drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c:2962:5: warning: variable dtim_period set but not used [-Wunused-but-set-variable]\n drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c: In function brcmf_update_bss_info:\n drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c:2961:6: warning: variable beacon_interval set but not used [-Wunused-but-set-variable]\n \n They are never used so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit a4a68f727fb8be4cd738a45e01b3781d3fed79f9\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Jul 24 22:10:20 2019 +0800\n\n rtlwifi: remove unneeded function _rtl_dump_channel_map()\n \n Now _rtl_dump_channel_map() does not do any actual\n thing using the channel. So remove it.\n \n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Acked-by: Ping-Ke Shih \u003cpkshih@realtek.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit e7338e0319854e4bab7a6767c95d258f32cd18ea\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Wed Jul 24 19:28:02 2019 +0800\n\n rtlwifi: rtl_pci: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 1f5f5ea72fc993a1ffdc8f7c16430c921cc16e16\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Wed Jul 24 19:27:53 2019 +0800\n\n qtnfmac_pcie: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit ffa4d78cbc2644b4867b8129b3fbb5ddcdfcdba2\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Wed Jul 24 19:27:45 2019 +0800\n\n mwifiex: pcie: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit a40c28700d98d955f0d8fc96b5feb00b23fd8a6a\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Wed Jul 24 19:27:30 2019 +0800\n\n iwlegacy: Use dev_get_drvdata where possible\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 8f9af6309b84a33db2c790c9df77017d865d3983\nAuthor: Ganapathi Kondraju \u003cganapathirajukondraju@gmail.com\u003e\nDate: Wed Jul 24 16:07:21 2019 +0530\n\n rsi: fix for sdio interface setup in 9116\n \n Issue: RS-9116 Card is not responding after firmware got loaded.\n \n Root cause: After firmware got loaded, we need to reset the program\n counter and few device specific registers. Those registers were not\n resetted properly.\n \n Fix: Properly resetting those registers.\n \n Signed-off-by: Ganapathi Kondraju \u003cganapathirajukondraju@gmail.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 6334dea8880a4add8e5e0e4d3c311e7886c26f1c\nAuthor: Pavel Machek \u003cpavel@ucw.cz\u003e\nDate: Wed Jul 24 11:50:15 2019 +0200\n\n mwifiex: make error values consistent in mwifiex_update_bss_desc_with_ie()\n \n Surrounding code uses -ERRNO as a result, so don\u0027t pass plain -1.\n \n Signed-off-by: Pavel Machek \u003cpavel@denx.de\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 9236c1250cb7c6c645a658485efc2f99d57e96c2\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 16:29:19 2019 +0800\n\n bcma: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 3457f86da60de73705bce8fe32a36651441e639e\nAuthor: Brian Norris \u003cbriannorris@chromium.org\u003e\nDate: Fri Jul 12 18:32:32 2019 -0700\n\n rtw88: use txpwr_lmt_cfg_pair struct, not arrays\n \n We\u0027re just trusting that these tables are of the right dimensions, when\n we could do better by just using the struct directly. Let\u0027s expose the\n struct txpwr_lmt_cfg_pair instead.\n \n The table changes were made by using some Vim macros, so that should\n help prevent any translation mistakes along the way.\n \n Remaining work: get the \u0027void *data\u0027 out of the generic struct\n rtw_table; all of these tables really deserve to be their own data\n structure, with proper type fields.\n \n Signed-off-by: Brian Norris \u003cbriannorris@chromium.org\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 100c12f20dfa463d52f9327b71c1806dcde82852\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 22:21:54 2019 +0300\n\n thunderbolt: Switch to use device_property_count_uXX()\n \n Use device_property_count_uXX() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Yehezkel Bernat \u003cYehezkelShB@gmail.com\u003e\n Signed-off-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n\ncommit 118b2806a07605d6e3c880e35a5c472bfda7582f\nAuthor: Ard van Breemen \u003card@kwaak.net\u003e\nDate: Fri Aug 2 13:52:15 2019 +0200\n\n ALSA: usb-audio: Add Pioneer DDJ-SX3 PCM quirck\n \n The Pioneer DDJ-SX3 is a plain 12 32bit channel out and 10 channel in\n PCM/midi controller. The PCM part is \"vendor specific\".\n It needs the \"ignore invalid bsynchaddress\" patch as it uses 0 for that.\n \n Signed-off-by: Ard van Breemen \u003card@kwaak.net\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 1b34121d9f26d272b0b2334209af6b6fc82d4bf1\nAuthor: Ard van Breemen \u003card@kwaak.net\u003e\nDate: Fri Aug 2 13:52:14 2019 +0200\n\n ALSA: usb-audio: Skip bSynchAddress endpoint check if it is invalid\n \n The Linux kernel assumes that get_endpoint(alts,0) and\n get_endpoint(alts,1) are eachothers feedback endpoints.\n To reassure that validity it will test bsynchaddress to comply with that\n assumption. But if the bsyncaddress is 0 (invalid), it will flag that as\n a wrong assumption and return an error.\n Fix: Skip the test if bSynchAddress is 0.\n Note: those with a valid bSynchAddress should have a code quirck added.\n \n Signed-off-by: Ard van Breemen \u003card@kwaak.net\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit a037d269221c0ae15f47046757afcbd1a7177bbf\nAuthor: Mukesh Ojha \u003cmojha@codeaurora.org\u003e\nDate: Wed Jul 31 20:35:04 2019 +0530\n\n locking/mutex: Use mutex flags macro instead of hard code\n \n Use the mutex flag macro instead of hard code value inside\n __mutex_owner().\n \n Signed-off-by: Mukesh Ojha \u003cmojha@codeaurora.org\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: mingo@redhat.com\n Cc: will@kernel.org\n Link: https://lkml.kernel.org/r/1564585504-3543-2-git-send-email-mojha@codeaurora.org\n\ncommit 5f35d5a66b3ec62cb5ec4ec2ad9aebe2ac325673\nAuthor: Mukesh Ojha \u003cmojha@codeaurora.org\u003e\nDate: Wed Jul 31 20:35:03 2019 +0530\n\n locking/mutex: Make __mutex_owner static to mutex.c\n \n __mutex_owner() should only be used by the mutex api\u0027s.\n So, to put this restiction let\u0027s move the __mutex_owner()\n function definition from linux/mutex.h to mutex.c file.\n \n There exist functions that uses __mutex_owner() like\n mutex_is_locked() and mutex_trylock_recursive(), So\n to keep legacy thing intact move them as well and\n export them.\n \n Move mutex_waiter structure also to keep it private to the\n file.\n \n Signed-off-by: Mukesh Ojha \u003cmojha@codeaurora.org\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: mingo@redhat.com\n Cc: will@kernel.org\n Link: https://lkml.kernel.org/r/1564585504-3543-1-git-send-email-mojha@codeaurora.org\n\ncommit 24a376d65177009a4dd8d846543c5dc69f5c4ced\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Thu Aug 1 15:30:28 2019 +0200\n\n locking/qspinlock,x86: Clarify virt_spin_lock_key\n \n Add a few comments to clarify how this is supposed to work.\n \n Reported-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Juergen Gross \u003cjgross@suse.com\u003e\n\ncommit fce45cd41101f1a9620267146b21f09b3454d8db\nAuthor: Davidlohr Bueso \u003cdave@stgolabs.net\u003e\nDate: Sun Jul 28 21:47:35 2019 -0700\n\n locking/rwsem: Check for operations on an uninitialized rwsem\n \n Currently rwsems is the only locking primitive that lacks this\n debug feature. Add it under CONFIG_DEBUG_RWSEMS and do the magic\n checking in the locking fastpath (trylock) operation such that\n we cover all cases. The unlocking part is pretty straightforward.\n \n Signed-off-by: Davidlohr Bueso \u003cdbueso@suse.de\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Acked-by: Waiman Long \u003clongman@redhat.com\u003e\n Cc: mingo@kernel.org\n Cc: Davidlohr Bueso \u003cdave@stgolabs.net\u003e\n Link: https://lkml.kernel.org/r/20190729044735.9632-1-dave@stgolabs.net\n\ncommit 91d2a812dfb98b3b4dad661529c33bc38d303461\nAuthor: Waiman Long \u003clongman@redhat.com\u003e\nDate: Tue Jun 25 10:39:13 2019 -0400\n\n locking/rwsem: Make handoff writer optimistically spin on owner\n \n When the handoff bit is set by a writer, no other tasks other than\n the setting writer itself is allowed to acquire the lock. If the\n to-be-handoff\u0027ed writer goes to sleep, there will be a wakeup latency\n period where the lock is free, but no one can acquire it. That is less\n than ideal.\n \n To reduce that latency, the handoff writer will now optimistically spin\n on the owner if it happens to be a on-cpu writer. It will spin until\n it releases the lock and the to-be-handoff\u0027ed writer can then acquire\n the lock immediately without any delay. Of course, if the owner is not\n a on-cpu writer, the to-be-handoff\u0027ed writer will have to sleep anyway.\n \n The optimistic spinning code is also modified to not stop spinning\n when the handoff bit is set. This will prevent an occasional setting of\n handoff bit from causing a bunch of optimistic spinners from entering\n into the wait queue causing significant reduction in throughput.\n \n On a 1-socket 22-core 44-thread Skylake system, the AIM7 shared_memory\n workload was run with 7000 users. The throughput (jobs/min) of the\n following kernels were as follows:\n \n 1) 5.2-rc6\n - 8,092,486\n 2) 5.2-rc6 + tip\u0027s rwsem patches\n - 7,567,568\n 3) 5.2-rc6 + tip\u0027s rwsem patches + this patch\n - 7,954,545\n \n Using perf-record(1), the %cpu time used by rwsem_down_write_slowpath(),\n rwsem_down_write_failed() and their callees for the 3 kernels were 1.70%,\n 5.46% and 2.08% respectively.\n \n Signed-off-by: Waiman Long \u003clongman@redhat.com\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: x86@kernel.org\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: Will Deacon \u003cwill.deacon@arm.com\u003e\n Cc: huang ying \u003chuang.ying.caritas@gmail.com\u003e\n Cc: Tim Chen \u003ctim.c.chen@linux.intel.com\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Borislav Petkov \u003cbp@alien8.de\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Davidlohr Bueso \u003cdave@stgolabs.net\u003e\n Cc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\n Link: https://lkml.kernel.org/r/20190625143913.24154-1-longman@redhat.com\n\ncommit f7f530181461aaf6afa0b3f0be4d239dea410896\nAuthor: Alexander Tsoy \u003calexander@tsoy.me\u003e\nDate: Fri Aug 11 02:36:14 2017 +0300\n\n ALSA: usb-audio: fix PCM device order\n \n Some cards have alternate setting with non-PCM format as the first\n altsetting in the interface descriptors. This confuses userspace, since\n alsa-lib uses device 0 by default. So lets parse interfaces in two steps:\n 1. Parse altsettings with PCM formats.\n 2. Parse altsettings with non-PCM formats.\n \n This fixes at least following cards:\n - Audinst HUD-mx2\n - Audinst HUD-mini\n \n [ Adapted to 5.3 kernel by tiwai ]\n \n Signed-off-by: Alexander Tsoy \u003calexander@tsoy.me\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit c1ae5e7f057beb1daa72831ab22ec06e71669af7\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Sat Jul 27 10:55:49 2019 +0200\n\n ALSA: usb-audio: Unify audioformat release code\n \n There are many open code for releasing audioformat object.\n Provide a unified helper and call it from the all places.\n \n Only a cleanup, no functional changes.\n \n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit e018ed6ecebf71b7c539d1a2f49d092a75844148\nAuthor: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\nDate: Sat Jul 27 21:34:14 2019 +0200\n\n nvmem: meson-mx-efuse: allow reading data smaller than word_size\n \n Some Amlogic boards store the Ethernet MAC address inside the eFuse. The\n Ethernet MAC address uses 6 bytes. The existing logic in\n meson_mx_efuse_read() would write beyond the end of the data buffer when\n trying to read data with a size that is not aligned to word_size (4\n bytes on Meson8, Meson8b and Meson8m2).\n \n Calculate the remaining data to copy inside meson_mx_efuse_read() so\n reading 6 bytes doesn\u0027t write beyond the end of the data buffer.\n \n Signed-off-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n\ncommit d0d3f4defa90c1c6c0a1e0aa927b2370835c4bac\nAuthor: Stefan Mavrodiev \u003cstefan@olimex.com\u003e\nDate: Wed Jul 31 10:14:47 2019 +0300\n\n nvmem: sunxi_sid: fix A64 SID controller support\n \n Like in H3, A64 SID controller doesn\u0027t return correct data\n when using direct access. It appears that on A64, SID needs\n 8 bytes of word_size.\n \n Workaround is to enable read by registers.\n \n Signed-off-by: Stefan Mavrodiev \u003cstefan@olimex.com\u003e\n Acked-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\n Tested-by: Vasily Khoruzhick \u003canarsoul@gmail.com\u003e\n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n\ncommit e55d8079472b97bc9ede3f84f35524f4f71eadf9\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Thu Jul 11 10:37:14 2019 +0800\n\n nvmem: imx-ocotp: Add i.MX8MN support\n \n i.MX8MN is a new SoC of i.MX8M series, it is similar to i.MX8MM\n in terms of addressing and clock setup, add support for its fuse\n read/write.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n\ncommit dfa1a3b67468d5bddf525dc8afec89d5f190bbd7\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Thu Jul 11 10:37:13 2019 +0800\n\n dt-bindings: imx-ocotp: Add i.MX8MN compatible\n \n Add compatible for i.MX8MN and add i.MX8MM/i.MX8MN to the description.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n\ncommit 0a43bffb7e21345444cd09302808b75cd8b6f9d5\nAuthor: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\nDate: Tue Aug 6 09:51:45 2019 +0100\n\n slimbus: fix slim_tid_txn()\n \n fix below issue reported by coccicheck\n ./drivers/slimbus/slimbus.h:440:3-46: duplicated argument to \u0026\u0026 or ||\n \n Looks like this was a typo, SLIM_MSG_MC_REQUEST_CHANGE_VALUE is command\n which requires transaction ID, so fix it, this also fix the warning.\n \n Reported-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n\ncommit a373775ca3b1aaba7e0ace11925c05a096ff5870\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Sun Aug 4 21:52:01 2019 +0530\n\n slimbus: qcom-ngd-ctrl: Add of_node_put() before return\n \n Each iteration of for_each_available_child_of_node puts the previous\n node, but in the case of a return from the middle of the loop, there is\n no put, thus causing a memory leak. Hence add an of_node_put before the\n return in two places.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Reviewed-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Signed-off-by: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n\ncommit 361f9dc2436812cbcf727fe73d0a4c11ca895fbb\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Aug 6 08:42:19 2019 +0100\n\n drm/i915: Use drm_i915_private directly from drv_get_drvdata()\n \n As we store a pointer to i915 in the drvdata field (as the pointer is both\n an alias to the drm_device and drm_i915_private), we can use the stored\n pointer directly as the i915 device.\n \n v2: Store and use i915 inside drv_get_drvdata()\n v3: Only expect i915 inside drv_get_drvdata() so drop the assumed\n i915/drm equivalence.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Andi Shyti \u003candi.shyti@intel.com\u003e\n Reviewed-by: Andi Shyti \u003candi.shyti@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190806074219.11043-1-chris@chris-wilson.co.uk\n\ncommit bc0380b9e5587d0c3fd207f16c53a101a9ea81dd\nAuthor: Ramalingam C \u003cramalingam.c@intel.com\u003e\nDate: Thu Aug 1 17:11:19 2019 +0530\n\n drm/hdcp: reference for srm file format\n \n In the kernel documentation, HDCP specifications links are shared as a\n reference for SRM table format.\n \n v2:\n Fixed small nits. [Shashank]\n \n Signed-off-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Reviewed-by: Shashank Sharma \u003cshashank.sharma@intel.com\u003e\n Acked-by: Pekka Paalanen \u003cpekka.paalanen@collabora.com\u003e\n Acked-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/320968/?series\u003d57232\u0026rev\u003d14\n\ncommit a41e71f4e78819403859950c34a638655cfd7da3\nAuthor: Ramalingam C \u003cramalingam.c@intel.com\u003e\nDate: Thu Aug 1 17:11:18 2019 +0530\n\n drm/i915: update the hdcp state with uevent\n \n drm function to update the content protection property state and to\n generate a uevent is invoked from the intel hdcp property work.\n \n Hence whenever kernel changes the property state, userspace will be\n updated with a uevent.\n \n v2:\n state update is moved into drm function [daniel]\n \n Signed-off-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Acked-by: Pekka Paalanen \u003cpekka.paalanen@collabora.com\u003e\n Acked-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/320965/?series\u003d57232\u0026rev\u003d14\n\ncommit bb5a45d40d50163df275c5a9578e19acdac361a2\nAuthor: Ramalingam C \u003cramalingam.c@intel.com\u003e\nDate: Thu Aug 1 17:11:17 2019 +0530\n\n drm/hdcp: update content protection property with uevent\n \n drm function is defined and exported to update a connector\u0027s\n content protection property state and to generate a uevent along\n with it.\n \n Pekka have completed the Weston DRM-backend review in\n https://gitlab.freedesktop.org/wayland/weston/merge_requests/48\n and the UAPI for HDCP 2.2 looks good.\n \n The userspace is accepted in Weston.\n \n v2:\n Update only when state is different from old one.\n v3:\n KDoc is added [Daniel]\n v4:\n KDoc is extended bit more [pekka]\n v5:\n Uevent usage is documented at kdoc of \"Content Protection\" also\n [pekka]\n \n Signed-off-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Acked-by: Pekka Paalanen \u003cpekka.paalanen@collabora.com\u003e\n Acked-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/320963/?series\u003d57232\u0026rev\u003d14\n\ncommit 6fe2ce064936efe5fbf2ce62c0b9be790d75189f\nAuthor: Ramalingam C \u003cramalingam.c@intel.com\u003e\nDate: Thu Aug 1 17:11:16 2019 +0530\n\n drm: uevent for connector status change\n \n DRM API for generating uevent for a status changes of connector\u0027s\n property.\n \n This uevent will have following details related to the status change:\n \n HOTPLUG\u003d1, CONNECTOR\u003d\u003cconnector_id\u003e and PROPERTY\u003d\u003cproperty_id\u003e\n \n Pekka have completed the Weston DRM-backend review in\n https://gitlab.freedesktop.org/wayland/weston/merge_requests/48\n and the UAPI for HDCP 2.2 looks good.\n \n The userspace is accepted in Weston.\n \n v2:\n Minor fixes at KDoc comments [Daniel]\n v3:\n Check the property is really attached with connector [Daniel]\n v4:\n Typos and string length suggestions are addressed [Sean]\n \n Signed-off-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Reviewed-by: Sean Paul \u003csean@poorly.run\u003e\n Acked-by: Pekka Paalanen \u003cpekka.paalanen@collabora.com\u003e\n Acked-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/320961/?series\u003d57232\u0026rev\u003d14\n\ncommit d456512c39ccc2d575135fe13156b61eb17c0b2a\nAuthor: Ramalingam C \u003cramalingam.c@intel.com\u003e\nDate: Thu Aug 1 17:11:15 2019 +0530\n\n drm/i915: Attach content type property\n \n Attaches the content type property for HDCP2.2 capable connectors.\n \n Implements the update of content type from property and apply the\n restriction on HDCP version selection.\n \n Need ACK for content type property from userspace consumer.\n \n v2:\n s/cp_content_type/content_protection_type [daniel]\n disable at hdcp_atomic_check to avoid check at atomic_set_property\n [Maarten]\n v3:\n s/content_protection_type/hdcp_content_type [Pekka]\n v4:\n hdcp disable incase of type change is moved into commit [daniel].\n v5:\n Simplified the Type change procedure. [Daniel]\n v6:\n Type change with UNDESIRED state is ignored.\n \n Signed-off-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Acked-by: Pekka Paalanen \u003cpekka.paalanen@collabora.com\u003e\n Acked-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/320959/?series\u003d57232\u0026rev\u003d14\n\ncommit 7672dbba85d30db07c3588f8cf6c6708bbdc0baa\nAuthor: Ramalingam C \u003cramalingam.c@intel.com\u003e\nDate: Thu Aug 1 17:11:14 2019 +0530\n\n drm: Add Content protection type property\n \n This patch adds a DRM ENUM property to the selected connectors.\n This property is used for mentioning the protected content\u0027s type\n from userspace to kernel HDCP authentication.\n \n Type of the stream is decided by the protected content providers.\n Type 0 content can be rendered on any HDCP protected display wires.\n But Type 1 content can be rendered only on HDCP2.2 protected paths.\n \n So when a userspace sets this property to Type 1 and starts the HDCP\n enable, kernel will honour it only if HDCP2.2 authentication is through\n for type 1. Else HDCP enable will be failed.\n \n Pekka have completed the Weston DRM-backend review in\n https://gitlab.freedesktop.org/wayland/weston/merge_requests/48\n and the UAPI for HDCP 2.2 looks good.\n \n The userspace is accepted in Weston.\n \n v2:\n cp_content_type is replaced with content_protection_type [daniel]\n check at atomic_set_property is removed [Maarten]\n v3:\n %s/content_protection_type/hdcp_content_type [Pekka]\n v4:\n property is created for the first requested connector and then reused.\n [Danvet]\n v5:\n kernel doc nits addressed [Daniel]\n Rebased as part of patch reordering.\n v6:\n Kernel docs are modified [pekka]\n v7:\n More details in Kernel docs. [pekka]\n v8:\n Few more clarification into kernel doc of content type [pekka]\n v9:\n Small fixes in coding style.\n v10:\n Moving DRM_MODE_HDCP_CONTENT_TYPEx definition to drm_hdcp.h [pekka]\n \n Signed-off-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Acked-by: Pekka Paalanen \u003cpekka.paalanen@collabora.com\u003e\n Acked-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/320957/?series\u003d57232\u0026rev\u003d14\n\ncommit 91148dbad80bb6024389287be4aefcbf20b2d0e4\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Mon Aug 5 21:36:35 2019 +0200\n\n USB: usbip: convert platform driver to use dev_groups\n \n Platform drivers now have the option to have the platform core create\n and remove any needed sysfs attribute files. So take advantage of that\n and do not register \"by hand\" any sysfs files.\n \n Cc: Valentina Manea \u003cvalentina.manea.m@gmail.com\u003e\n Acked-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n Link: https://lore.kernel.org/r/20190805193636.25560-5-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit c1233578a2b9db920ecd9b647797967ac68dfe46\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 18:11:04 2019 +0800\n\n drm/bochs: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Link: http://patchwork.freedesktop.org/patch/msgid/20190723101103.30250-1-hslester96@gmail.com\n Signed-off-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n\ncommit f6d2b63adb0a55e41ddbc0baa890c0c02858000d\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 18:40:00 2019 +0800\n\n drm/qxl: Use dev_get_drvdata where possible\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Link: http://patchwork.freedesktop.org/patch/msgid/20190723103959.4078-1-hslester96@gmail.com\n Signed-off-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n\ncommit ae76f635d4e1cffa6870cc5472567ca9d6940a22\nAuthor: Guo Ren \u003cren_guo@c-sky.com\u003e\nDate: Tue Jul 30 17:16:28 2019 +0800\n\n csky: Optimize arch_sync_dma_for_cpu/device with dma_inv_range\n \n DMA_FROM_DEVICE only need to read dma data of memory into CPU cache,\n so there is no need to clear cache before. Also clear + inv for\n DMA_FROM_DEVICE won\u0027t cause problem, because the memory range for dma\n won\u0027t be touched by software during dma working.\n \n Changes for V2:\n - Remove clr cache and ignore the DMA_TO_DEVICE in _for_cpu.\n - Change inv to wbinv cache with DMA_FROM_DEVICE in _for_device.\n \n Signed-off-by: Guo Ren \u003cren_guo@c-sky.com\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 4af9027d3f4061992c0b065102a0a666b72f073b\nAuthor: Guo Ren \u003cren_guo@c-sky.com\u003e\nDate: Tue Jul 30 17:02:26 2019 +0800\n\n csky/dma: Fixup cache_op failed when cross memory ZONEs\n \n If the paddr and size are cross between NORMAL_ZONE and HIGHMEM_ZONE\n memory range, cache_op will panic in do_page_fault with bad_area.\n \n Optimize the code to support the range which cross memory ZONEs.\n \n Changes for V2:\n - Revert back to postcore_initcall\n \n Signed-off-by: Guo Ren \u003cren_guo@c-sky.com\u003e\n Cc: Christoph Hellwig \u003chch@infradead.org\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit e7f0141a217fa28049d7a3bbc09bee9642c47687\nAuthor: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\nDate: Mon Aug 5 16:01:19 2019 +0200\n\n drm/ttm: drop ttm_buffer_object-\u003eresv\n \n All users moved to ttm_buffer_object-\u003ebase.resv\n \n Signed-off-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: http://patchwork.freedesktop.org/patch/msgid/20190805140119.7337-18-kraxel@redhat.com\n\ncommit 7a4db29660a9d16024fd843b720fb7449ebc2538\nAuthor: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\nDate: Mon Aug 5 16:01:18 2019 +0200\n\n drm/virtio: switch driver from bo-\u003eresv to bo-\u003ebase.resv\n \n Signed-off-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Acked-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: http://patchwork.freedesktop.org/patch/msgid/20190805140119.7337-17-kraxel@redhat.com\n\ncommit 5c69f132a2660435ec30d8531d77515b7ba4148e\nAuthor: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\nDate: Mon Aug 5 16:01:17 2019 +0200\n\n drm/qxl: switch driver from bo-\u003eresv to bo-\u003ebase.resv\n \n Signed-off-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Acked-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: http://patchwork.freedesktop.org/patch/msgid/20190805140119.7337-16-kraxel@redhat.com\n\ncommit 27c44acebd3fab5448aa3cffdc1996c897965a4a\nAuthor: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\nDate: Mon Aug 5 16:01:16 2019 +0200\n\n drm/nouveau: switch driver from bo-\u003eresv to bo-\u003ebase.resv\n \n Signed-off-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: http://patchwork.freedesktop.org/patch/msgid/20190805140119.7337-15-kraxel@redhat.com\n\ncommit 5a5011a72489545343a1599362e9ec126d7bd297\nAuthor: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\nDate: Mon Aug 5 16:01:15 2019 +0200\n\n drm/amdgpu: switch driver from bo-\u003eresv to bo-\u003ebase.resv\n \n Signed-off-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: http://patchwork.freedesktop.org/patch/msgid/20190805140119.7337-14-kraxel@redhat.com\n\ncommit 4922f55294bbc48d670bb57c025904b4d4878d1b\nAuthor: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\nDate: Mon Aug 5 16:01:14 2019 +0200\n\n drm/vmwgfx: switch driver from bo-\u003eresv to bo-\u003ebase.resv\n \n Signed-off-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Acked-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Link: http://patchwork.freedesktop.org/patch/msgid/20190805140119.7337-13-kraxel@redhat.com\n\ncommit 336ac942f115dd076bd7287c7cf03f37c710895c\nAuthor: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\nDate: Mon Aug 5 16:01:13 2019 +0200\n\n drm/radeon: switch driver from bo-\u003eresv to bo-\u003ebase.resv\n \n Signed-off-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: http://patchwork.freedesktop.org/patch/msgid/20190805140119.7337-12-kraxel@redhat.com\n\ncommit e532a135d7044b5477c1c56169fa131d77c57f75\nAuthor: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\nDate: Mon Aug 5 16:01:12 2019 +0200\n\n drm/ttm: switch ttm core from bo-\u003eresv to bo-\u003ebase.resv\n \n Signed-off-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: http://patchwork.freedesktop.org/patch/msgid/20190805140119.7337-11-kraxel@redhat.com\n\ncommit 2e3c9ec4d151c04d75546dfdc2f85a84ad546eb0\nAuthor: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\nDate: Mon Aug 5 16:01:11 2019 +0200\n\n drm/ttm: set both resv and base.resv pointers\n \n Initialize both ttm_buffer_object-\u003eresv and ttm_buffer_object-\u003ebase.resv\n pointers. This allows to move users from the former to the latter. When\n all users are moved we can drop ttm_buffer_object-\u003eresv.\n \n Signed-off-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: http://patchwork.freedesktop.org/patch/msgid/20190805140119.7337-10-kraxel@redhat.com\n\ncommit b96f3e7c8069b749a40ca3a33c97835d57dd45d2\nAuthor: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\nDate: Mon Aug 5 16:01:10 2019 +0200\n\n drm/ttm: use gem vma_node\n \n Drop vma_node from ttm_buffer_object, use the gem struct\n (base.vma_node) instead.\n \n Signed-off-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: http://patchwork.freedesktop.org/patch/msgid/20190805140119.7337-9-kraxel@redhat.com\n\ncommit 1e053b10ba60eae6a3f9de64cbc74bdf6cb0e715\nAuthor: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\nDate: Mon Aug 5 16:01:09 2019 +0200\n\n drm/ttm: use gem reservation object\n \n Drop ttm_resv from ttm_buffer_object, use the gem reservation object\n (base._resv) instead.\n \n Signed-off-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: http://patchwork.freedesktop.org/patch/msgid/20190805140119.7337-8-kraxel@redhat.com\n\ncommit f8659be8addd731480037eb44ecb521084ae1d11\nAuthor: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\nDate: Mon Aug 5 16:01:08 2019 +0200\n\n drm/nouveau: use embedded gem object\n \n Drop drm_gem_object from nouveau_bo, use the\n ttm_buffer_object.base instead.\n \n Build tested only.\n \n Signed-off-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Acked-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: http://patchwork.freedesktop.org/patch/msgid/20190805140119.7337-7-kraxel@redhat.com\n\ncommit c105de2828e13931cd218e8b801cd6762b618992\nAuthor: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\nDate: Mon Aug 5 16:01:07 2019 +0200\n\n drm/amdgpu: use embedded gem object\n \n Drop drm_gem_object from amdgpu_bo, use the\n ttm_buffer_object.base instead.\n \n Build tested only.\n \n Signed-off-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: http://patchwork.freedesktop.org/patch/msgid/20190805140119.7337-6-kraxel@redhat.com\n\ncommit ce77038fdae385f947757a37573d90f2e83f0271\nAuthor: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\nDate: Mon Aug 5 16:01:06 2019 +0200\n\n drm/radeon: use embedded gem object\n \n Drop drm_gem_object from radeon_bo, use the\n ttm_buffer_object.base instead.\n \n Build tested only.\n \n Signed-off-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: http://patchwork.freedesktop.org/patch/msgid/20190805140119.7337-5-kraxel@redhat.com\n\ncommit e0828d54c81cb111ead1a7c47a5ef1b319610a1d\nAuthor: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\nDate: Mon Aug 5 16:01:05 2019 +0200\n\n drm/qxl: use embedded gem object\n \n Drop drm_gem_object from qxl_bo, use the\n ttm_buffer_object.base instead.\n \n Signed-off-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Acked-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: http://patchwork.freedesktop.org/patch/msgid/20190805140119.7337-4-kraxel@redhat.com\n\ncommit 0e580c6d7d2f2d490add0fcf70c5b7ec2300e636\nAuthor: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\nDate: Mon Aug 5 16:01:04 2019 +0200\n\n drm/vram: use embedded gem object\n \n Drop drm_gem_object from drm_gem_vram_object, use the\n ttm_buffer_object.base instead.\n \n Signed-off-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Acked-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Link: http://patchwork.freedesktop.org/patch/msgid/20190805140119.7337-3-kraxel@redhat.com\n\ncommit 8eb8833e7ed362977c021116d2f34451a7009ca3\nAuthor: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\nDate: Mon Aug 5 16:01:03 2019 +0200\n\n drm/ttm: add gem base object\n \n Add drm_gem_object struct to ttm_buffer_object, so ttm objects are a gdm\n object superclass. Add a function to check whenever a given bo actually\n uses the embedded drm_gem_object.\n \n Signed-off-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: http://patchwork.freedesktop.org/patch/msgid/20190805140119.7337-2-kraxel@redhat.com\n\ncommit 9d4db2f5f0e6dc4c0dfb94110fe488e9eb191748\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:02 2019 -0700\n\n bus: sunxi-rsb: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Chen-Yu Tsai \u003cwens@csie.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\n\ncommit 053951dda71ecb4b554a2cdbe26f5f6f9bee9dd2\nAuthor: Zhou Yanjie \u003czhouyanjie@zoho.com\u003e\nDate: Fri Aug 2 16:27:37 2019 +0800\n\n MIPS: Ingenic: Disable broken BTB lookup optimization.\n \n In order to further reduce power consumption, the XBurst core\n by default attempts to avoid branch target buffer lookups by\n detecting \u0026 special casing loops. This feature will cause\n BogoMIPS and lpj calculate in error. Set cp0 config7 bit 4 to\n disable this feature.\n \n Signed-off-by: Zhou Yanjie \u003czhouyanjie@zoho.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: ralf@linux-mips.org\n Cc: paul@crapouillou.net\n Cc: jhogan@kernel.org\n Cc: malat@debian.org\n Cc: gregkh@linuxfoundation.org\n Cc: tglx@linutronix.de\n Cc: allison@lohutok.net\n Cc: syq@debian.org\n Cc: chenhc@lemote.com\n Cc: jiaxun.yang@flygoat.com\n\ncommit 579de8f86b1cdc3f9ef105fca9f4539d99647ed1\nAuthor: Zhou Yanjie \u003czhouyanjie@zoho.com\u003e\nDate: Fri Aug 2 16:27:36 2019 +0800\n\n MIPS: Ingenic: Fix bugs when detecting X1000\u0027s L2 cache.\n \n 1.fix bugs when detecting L2 cache sets value.\n 2.fix bugs when detecting L2 cache ways value.\n \n Signed-off-by: Zhou Yanjie \u003czhouyanjie@zoho.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: ralf@linux-mips.org\n Cc: paul@crapouillou.net\n Cc: jhogan@kernel.org\n Cc: malat@debian.org\n Cc: gregkh@linuxfoundation.org\n Cc: tglx@linutronix.de\n Cc: allison@lohutok.net\n Cc: syq@debian.org\n Cc: chenhc@lemote.com\n Cc: jiaxun.yang@flygoat.com\n\ncommit 05589b30b21ac0273970b61edd50c07d2ba156af\nAuthor: Thara Gopinath \u003cthara.gopinath@linaro.org\u003e\nDate: Tue Jul 30 11:24:42 2019 -0400\n\n soc: qcom: Extend AOSS QMP driver to support resources that are used to wake up the SoC.\n \n The AOSS QMP driver is extended to communicate with the additional\n resources. These resources are then registered as cooling devices\n with the thermal framework.\n \n Signed-off-by: Thara Gopinath \u003cthara.gopinath@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 4ece3125f21b1d42b84896c5646dbf0e878464e1\nAuthor: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\nDate: Tue Jul 2 10:00:41 2019 +0200\n\n ima: fix freeing ongoing ahash_request\n \n integrity_kernel_read() can fail in which case we forward to call\n ahash_request_free() on a currently running request. We have to wait\n for its completion before we can free the request.\n \n This was observed by interrupting a \"find / -type f -xdev -print0 | xargs -0\n cat 1\u003e/dev/null\" with ctrl-c on an IMA enabled filesystem.\n \n Signed-off-by: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\n Signed-off-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n\ncommit f5e1040196dbfe14c77ce3dfe3b7b08d2d961e88\nAuthor: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\nDate: Tue Jul 2 10:00:40 2019 +0200\n\n ima: always return negative code for error\n \n integrity_kernel_read() returns the number of bytes read. If this is\n a short read then this positive value is returned from\n ima_calc_file_hash_atfm(). Currently this is only indirectly called from\n ima_calc_file_hash() and this function only tests for the return value\n being zero or nonzero and also doesn\u0027t forward the return value.\n Nevertheless there\u0027s no point in returning a positive value as an error,\n so translate a short read into -EINVAL.\n \n Signed-off-by: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\n Signed-off-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n\ncommit e5092255bb3967bcc473dc86492dbbd5f7714023\nAuthor: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\nDate: Thu Jun 27 23:19:33 2019 -0300\n\n ima: Store the measurement again when appraising a modsig\n \n If the IMA template contains the \"modsig\" or \"d-modsig\" field, then the\n modsig should be added to the measurement list when the file is appraised.\n \n And that is what normally happens, but if a measurement rule caused a file\n containing a modsig to be measured before a different rule causes it to be\n appraised, the resulting measurement entry will not contain the modsig\n because it is only fetched during appraisal. When the appraisal rule\n triggers, it won\u0027t store a new measurement containing the modsig because\n the file was already measured.\n \n We need to detect that situation and store an additional measurement with\n the modsig. This is done by adding an IMA_MEASURE action flag if we read a\n modsig and the IMA template contains a modsig field.\n \n Suggested-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n\ncommit 3878d505aa718bcc7b1eb4089ab9b9fb27dee957\nAuthor: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\nDate: Thu Jun 27 23:19:32 2019 -0300\n\n ima: Define ima-modsig template\n \n Define new \"d-modsig\" template field which holds the digest that is\n expected to match the one contained in the modsig, and also new \"modsig\"\n template field which holds the appended file signature.\n \n Add a new \"ima-modsig\" defined template descriptor with the new fields as\n well as the ones from the \"ima-sig\" descriptor.\n \n Change ima_store_measurement() to accept a struct modsig * argument so that\n it can be passed along to the templates via struct ima_event_data.\n \n Suggested-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n\ncommit 15588227e086ec662d59df144e48af82e3e592f1\nAuthor: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\nDate: Thu Jun 27 23:19:31 2019 -0300\n\n ima: Collect modsig\n \n Obtain the modsig and calculate its corresponding hash in\n ima_collect_measurement().\n \n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n\ncommit 39b07096364a42c516415d5f841069e885234e61\nAuthor: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\nDate: Thu Jun 27 23:19:30 2019 -0300\n\n ima: Implement support for module-style appended signatures\n \n Implement the appraise_type\u003dimasig|modsig option, allowing IMA to read and\n verify modsig signatures.\n \n In case a file has both an xattr signature and an appended modsig, IMA will\n only use the appended signature if the key used by the xattr signature\n isn\u0027t present in the IMA or platform keyring.\n \n Because modsig verification needs to convert from an integrity keyring id\n to the keyring itself, add an integrity_keyring_from_id() function in\n digsig.c so that integrity_modsig_verify() can use it.\n \n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n\ncommit a5fbeb615ca42f913ace3291d636e96feabcc545\nAuthor: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\nDate: Thu Jun 27 23:19:29 2019 -0300\n\n ima: Factor xattr_verify() out of ima_appraise_measurement()\n \n Verify xattr signature in a separate function so that the logic in\n ima_appraise_measurement() remains clear when it gains the ability to also\n verify an appended module signature.\n \n The code in the switch statement is unchanged except for having to\n dereference the status and cause variables (since they\u0027re now pointers),\n and fixing the style of a block comment to appease checkpatch.\n \n Suggested-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n\ncommit 9044d627fd18f9fca49b62d4619ee14914b91464\nAuthor: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\nDate: Thu Jun 27 23:19:28 2019 -0300\n\n ima: Add modsig appraise_type option for module-style appended signatures\n \n Introduce the modsig keyword to the IMA policy syntax to specify that\n a given hook should expect the file to have the IMA signature appended\n to it. Here is how it can be used in a rule:\n \n appraise func\u003dKEXEC_KERNEL_CHECK appraise_type\u003dimasig|modsig\n \n With this rule, IMA will accept either a signature stored in the extended\n attribute or an appended signature.\n \n For now, the rule above will behave exactly the same as if\n appraise_type\u003dimasig was specified. The actual modsig implementation\n will be introduced separately.\n \n Suggested-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n\ncommit cf38fed1e183dd2410f62d49ae635fe593082f0c\nAuthor: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\nDate: Thu Jun 27 23:19:27 2019 -0300\n\n integrity: Select CONFIG_KEYS instead of depending on it\n \n This avoids a dependency cycle in soon-to-be-introduced\n CONFIG_IMA_APPRAISE_MODSIG: it will select CONFIG_MODULE_SIG_FORMAT\n which in turn selects CONFIG_KEYS. Kconfig then complains that\n CONFIG_INTEGRITY_SIGNATURE depends on CONFIG_KEYS.\n \n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Signed-off-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n\ncommit e201af16d1ec76ccd19b90484d767984ff451f18\nAuthor: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\nDate: Thu Jun 27 23:19:26 2019 -0300\n\n PKCS#7: Introduce pkcs7_get_digest()\n \n IMA will need to access the digest of the PKCS7 message (as calculated by\n the kernel) before the signature is verified, so introduce\n pkcs7_get_digest() for that purpose.\n \n Also, modify pkcs7_digest() to detect when the digest was already\n calculated so that it doesn\u0027t have to do redundant work. Verifying that\n sinfo-\u003esig-\u003edigest isn\u0027t NULL is sufficient because both places which\n allocate sinfo-\u003esig (pkcs7_parse_message() and pkcs7_note_signed_info())\n use kzalloc() so sig-\u003edigest is always initialized to zero.\n \n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Reviewed-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n Cc: David Howells \u003cdhowells@redhat.com\u003e\n Cc: David Woodhouse \u003cdwmw2@infradead.org\u003e\n Cc: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Signed-off-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n\ncommit 2a7bf671186eb5d1a47ef192aefbdf788f5b38fe\nAuthor: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\nDate: Thu Jun 27 23:19:25 2019 -0300\n\n PKCS#7: Refactor verify_pkcs7_signature()\n \n IMA will need to verify a PKCS#7 signature which has already been parsed.\n For this reason, factor out the code which does that from\n verify_pkcs7_signature() into a new function which takes a struct\n pkcs7_message instead of a data buffer.\n \n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Reviewed-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n Cc: David Howells \u003cdhowells@redhat.com\u003e\n Cc: David Woodhouse \u003cdwmw2@infradead.org\u003e\n Cc: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Signed-off-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n\ncommit c8424e776b093280d3fdd104d850706b3b229ac8\nAuthor: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\nDate: Thu Jul 4 15:57:34 2019 -0300\n\n MODSIGN: Export module signature definitions\n \n IMA will use the module_signature format for append signatures, so export\n the relevant definitions and factor out the code which verifies that the\n appended signature trailer is valid.\n \n Also, create a CONFIG_MODULE_SIG_FORMAT option so that IMA can select it\n and be able to use mod_check_sig() without having to depend on either\n CONFIG_MODULE_SIG or CONFIG_MODULES.\n \n s390 duplicated the definition of struct module_signature so now they can\n use the new \u003clinux/module_signature.h\u003e header instead.\n \n Signed-off-by: Thiago Jung Bauermann \u003cbauerman@linux.ibm.com\u003e\n Acked-by: Jessica Yu \u003cjeyu@kernel.org\u003e\n Reviewed-by: Philipp Rudo \u003cprudo@linux.ibm.com\u003e\n Cc: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\n Signed-off-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n\ncommit 16e8e8072108426029f0c16dff7fbe77fae3df8f\nAuthor: Stephan Gerhold \u003cstephan@gerhold.net\u003e\nDate: Mon Jul 22 11:22:11 2019 +0200\n\n arm64: dts: qcom: Add device tree for Longcheer L8150\n \n Longcheer L8150 is a smartphone based on MSM8916 which is\n used in several rebrands like the Snapdragon 410\n Android One devices or the Wileyfox Swift.\n \n Add a device tree for L8150 with initial support for:\n - SDHCI (internal and external storage)\n - USB Device Mode\n - UART\n - Regulators\n \n Co-developed-by: Nikita Travkin \u003cnikitos.tr@gmail.com\u003e\n Signed-off-by: Nikita Travkin \u003cnikitos.tr@gmail.com\u003e\n Signed-off-by: Stephan Gerhold \u003cstephan@gerhold.net\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 1329c1ab0730b521e6cd3051c56a2ff3d55f21e6\nAuthor: Stephan Gerhold \u003cstephan@gerhold.net\u003e\nDate: Mon Jul 22 11:22:10 2019 +0200\n\n arm64: dts: qcom: Add device tree for Samsung Galaxy A3U/A5U\n \n Samsung Galaxy A3 (SM-A300FU) and Samsung Galaxy A5 (SM-A500FU)\n are smartphones using the MSM8916 SoC released in 2015.\n \n Add a device tree for A3U and A5U with initial support for:\n - SDHCI (internal and external storage)\n - USB Device Mode\n - UART (on USB connector via the SM5502 MUIC)\n - Regulators\n \n The two devices (and all other variants of A3/A5 released in 2015)\n are very similar, with some differences in display, touchscreen\n and sensors. The common parts are shared in\n msm8916-samsung-a2015-common.dtsi to reduce duplication.\n \n The device tree is loosely based on apq8016-sbc.dtsi and the\n downstream kernel provided by Samsung, mixed with a lot of own\n research.\n \n Co-developed-by: Michael Srba \u003cMichael.Srba@seznam.cz\u003e\n Signed-off-by: Michael Srba \u003cMichael.Srba@seznam.cz\u003e\n Signed-off-by: Stephan Gerhold \u003cstephan@gerhold.net\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit b72160fa886d757611b1c7f1127d5585fce43af2\nAuthor: Stephan Gerhold \u003cstephan@gerhold.net\u003e\nDate: Mon Jul 22 11:22:09 2019 +0200\n\n dt-bindings: qcom: Document bindings for new MSM8916 devices\n \n Document the new samsung,a3u/a5u-eur and longcheer,l8150\n device tree bindings used in their device trees.\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Stephan Gerhold \u003cstephan@gerhold.net\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 3af1e2aa1855d976337752c95ac0c4971d8fc9e6\nAuthor: Stephan Gerhold \u003cstephan@gerhold.net\u003e\nDate: Mon Jul 22 11:22:08 2019 +0200\n\n dt-bindings: vendor-prefixes: Add Longcheer Technology Co., Ltd.\n \n Add the \"longcheer\" vendor prefix for Longcheer Technology Co., Ltd.,\n an \"industry-leading service provider of mobile phone design\n and product delivery\". (http://www.longcheer.com)\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Stephan Gerhold \u003cstephan@gerhold.net\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 4e300e439af31ae39669f970121022743980bbce\nAuthor: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\nDate: Thu Jul 25 15:39:49 2019 +0300\n\n arm64: dts: qcom: msm8996: Add Venus video codec DT node\n \n This adds Qualcomm Venus video codec DT node for the video\n codec hardware found in MSM8996 platforms.\n \n Signed-off-by: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 7e4b5f241a91e2e15cd1e19024c3b6a0664beadc\nAuthor: Thara Gopinath \u003cthara.gopinath@linaro.org\u003e\nDate: Tue Jul 30 11:24:43 2019 -0400\n\n arm64: dts: qcom: Extend AOSS QMP node\n \n AOSS hosts resources that can be used to warm up the SoC.\n Add nodes for these resources.\n \n Signed-off-by: Thara Gopinath \u003cthara.gopinath@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit d98de8efa19fb753ca39e6c1c26d8f2bedb6e2d4\nAuthor: Vivek Gautam \u003cvivek.gautam@codeaurora.org\u003e\nDate: Wed Jul 31 11:28:02 2019 +0530\n\n arm64: dts: qcom: msm8996: Add Coresight support\n \n Enable coresight support by adding device nodes for the\n available source, sinks and channel blocks on msm8996.\n \n This also adds coresight cpu debug nodes.\n \n Signed-off-by: Vivek Gautam \u003cvivek.gautam@codeaurora.org\u003e\n Signed-off-by: Sai Prakash Ranjan \u003csaiprakash.ranjan@codeaurora.org\u003e\n Reviewed-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Acked-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 783abfa2249a0a354fdba5d232714c448a5312ca\nAuthor: Sai Prakash Ranjan \u003csaiprakash.ranjan@codeaurora.org\u003e\nDate: Wed Jul 31 11:28:01 2019 +0530\n\n arm64: dts: qcom: msm8998: Add Coresight support\n \n Enable coresight support by adding device nodes for the\n available source, sinks and channel blocks on MSM8998.\n \n Signed-off-by: Sai Prakash Ranjan \u003csaiprakash.ranjan@codeaurora.org\u003e\n Reviewed-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Acked-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit ed7d6110c110730cac33a0c9e1cf5b83092fc2dc\nAuthor: Sai Prakash Ranjan \u003csaiprakash.ranjan@codeaurora.org\u003e\nDate: Wed Jul 31 11:28:00 2019 +0530\n\n arm64: dts: qcom: sdm845: Add Coresight support\n \n Add coresight components found on Qualcomm SDM845 SoC.\n \n Signed-off-by: Sai Prakash Ranjan \u003csaiprakash.ranjan@codeaurora.org\u003e\n Reviewed-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Acked-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 32a5da21dd1cea8e76cf354dee8858da24820292\nAuthor: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\nDate: Mon Jul 22 09:58:23 2019 -0700\n\n arm64: dts: qcom: msm8998: Node ordering, address cleanups\n \n DT nodes should be ordered by address, then node name, and finally label.\n The msm8998 dtsi does not follow this, so clean it up by reordering the\n nodes. While we are at it, extend the addresses to be fully 32-bits wide\n so that ordering is easy to determine when adding new nodes. Also, two\n or so nodes had the wrong address value in their node name (did not match\n the reg property), so fix those up as well.\n \n Hopefully going forward, things can be maintained so that a cleanup like\n this is not needed.\n \n Reviewed-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Marc Gonzalez \u003cmarc.w.gonzalez@free.fr\u003e\n Signed-off-by: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit cd23d1405be666a7c43045abe339bc1acaa3400b\nAuthor: Vaishali Thakkar \u003cvaishali.thakkar@linaro.org\u003e\nDate: Wed Jul 24 04:05:15 2019 +0530\n\n soc: qcom: socinfo: Expose image information\n \n The socinfo driver provides information about version of the various\n images loaded in the system. Expose this to user space for debugging\n purpose.\n \n Tested-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Vaishali Thakkar \u003cvaishali.thakkar@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 9c84c1e78634bd7f93541ad9ad88e99b21495da7\nAuthor: Vaishali Thakkar \u003cvaishali.thakkar@linaro.org\u003e\nDate: Wed Jul 24 04:05:14 2019 +0530\n\n soc: qcom: socinfo: Expose custom attributes\n \n The Qualcomm socinfo provides a number of additional attributes,\n add these to the socinfo driver and expose them via debugfs\n functionality.\n \n Tested-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Vaishali Thakkar \u003cvaishali.thakkar@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit efb448d0a3fca01bb987dd70963da6185b81751e\nAuthor: Imran Khan \u003ckimran@codeaurora.org\u003e\nDate: Wed Jul 24 04:05:13 2019 +0530\n\n soc: qcom: Add socinfo driver\n \n The Qualcomm socinfo driver exposes information about the SoC, its\n version and its serial number to user space.\n \n Tested-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Imran Khan \u003ckimran@codeaurora.org\u003e\n [Bjorn: Extract code to platform_driver, split patch in multiple]\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n [Vaishali: Simplify declarations, introduce qcom_socinfo struct, Fix\n memory leak, Remove extra code and Misc code refactoring]\n Signed-off-by: Vaishali Thakkar \u003cvaishali.thakkar@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit f7ccc7a397cf2ef64aebb2f726970b93203858d2\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Wed Jul 24 04:05:12 2019 +0530\n\n base: soc: Export soc_device_register/unregister APIs\n \n Qcom Socinfo driver can be built as a module, so\n export these two APIs.\n \n Tested-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Vaishali Thakkar \u003cvaishali.thakkar@linaro.org\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Reviewed-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 9aebf4de220344e2f03ae6386272bf98f80fd295\nAuthor: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\nDate: Wed Jul 24 04:05:11 2019 +0530\n\n base: soc: Add serial_number attribute to soc\n \n Add new attribute named \"serial_number\" as a standard interface for\n user space to acquire the serial number of the device.\n \n For ST-Ericsson SoCs this is exposed by the cryptically named \"soc_id\"\n attribute, but this provides a human readable standardized name for this\n property.\n \n Tested-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Signed-off-by: Vaishali Thakkar \u003cvaishali.thakkar@linaro.org\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Reviewed-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit e3c5bc562a5f38b76affc89bdeb2dd83a0014b3e\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Jul 31 15:40:20 2019 +0200\n\n arm64: dts: qcom: sdm845-cheza: Spelling s/conenctors/connectors/\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit d32bb89c9a45a95b8fa0a731352062814805cc3c\nAuthor: Govind Singh \u003cgovinds@codeaurora.org\u003e\nDate: Wed Jul 31 17:37:04 2019 +0530\n\n arm64: dts: qcom: qcs404: Add wifi rails in QCS404 dt node for proxy votes\n \n Add wifi regulators in qcs404 dt node for proxy vote. Proxy votes are\n required for handling driver recovery scenarios to prevent\n un-clocked register access during driver recovery.\n \n Signed-off-by: Govind Singh \u003cgovinds@codeaurora.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit e8c48eb08ab1446b8bc2b99c2254b4505945adb1\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Thu Aug 1 15:20:49 2019 +0530\n\n arm64: dts: qcom: qcs404: remove unit name for thermal trip points\n \n The thermal trip points have unit name but no reg property, so we can\n remove them\n \n arch/arm64/boot/dts/qcom/qcs404.dtsi:1080.31-1084.7: Warning (unit_address_vs_reg): /thermal-zones/aoss-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/qcs404.dtsi:1095.33-1099.7: Warning (unit_address_vs_reg): /thermal-zones/q6-hvx-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/qcs404.dtsi:1110.32-1114.7: Warning (unit_address_vs_reg): /thermal-zones/lpass-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/qcs404.dtsi:1125.31-1129.7: Warning (unit_address_vs_reg): /thermal-zones/wlan-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/qcs404.dtsi:1140.34-1144.7: Warning (unit_address_vs_reg): /thermal-zones/cluster-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/qcs404.dtsi:1145.34-1149.7: Warning (unit_address_vs_reg): /thermal-zones/cluster-thermal/trips/trip-point@1: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/qcs404.dtsi:1174.31-1178.7: Warning (unit_address_vs_reg): /thermal-zones/cpu0-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/qcs404.dtsi:1179.31-1183.7: Warning (unit_address_vs_reg): /thermal-zones/cpu0-thermal/trips/trip-point@1: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/qcs404.dtsi:1208.31-1212.7: Warning (unit_address_vs_reg): /thermal-zones/cpu1-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/qcs404.dtsi:1213.31-1217.7: Warning (unit_address_vs_reg): /thermal-zones/cpu1-thermal/trips/trip-point@1: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/qcs404.dtsi:1242.31-1246.7: Warning (unit_address_vs_reg): /thermal-zones/cpu2-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/qcs404.dtsi:1247.31-1251.7: Warning (unit_address_vs_reg): /thermal-zones/cpu2-thermal/trips/trip-point@1: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/qcs404.dtsi:1276.31-1280.7: Warning (unit_address_vs_reg): /thermal-zones/cpu3-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/qcs404.dtsi:1281.31-1285.7: Warning (unit_address_vs_reg): /thermal-zones/cpu3-thermal/trips/trip-point@1: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/qcs404.dtsi:1310.30-1314.7: Warning (unit_address_vs_reg): /thermal-zones/gpu-thermal/trips/trip-point@0: node has a unit name, but no reg property\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit cbf7f748f27684fee169252070e7b2d76fe81f22\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Thu Aug 1 15:20:48 2019 +0530\n\n arm64: dts: qcom: pms405: remove reduandant properties\n \n pms405@1 nodes specified unnecessary #address-cells/#size-cells but the\n subnodes dont have \"ranges\" or \"reg\" so remove it\n \n arch/arm64/boot/dts/qcom/pms405.dtsi:141.21-150.4: Warning (avoid_unnecessary_addr_size): /soc@0/spmi@200f000/pms405@1: unnecessary #address-cells/#size-cells without \"ranges\" or child \"reg\" property\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit f65018264cb1ae2777b39a3c36cea012c5684bac\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Thu Aug 1 15:20:47 2019 +0530\n\n arm64: dts: qcom: pms405: add unit name adc nodes\n \n The adc nodes have reg property but were missing the unit name, so add\n that to fix these warnings:\n \n arch/arm64/boot/dts/qcom/pms405.dtsi:91.12-94.6: Warning (unit_address_vs_reg): /soc@0/spmi@200f000/pms405@0/adc@3100/ref_gnd: node has a reg or ranges property, but no unit name\n arch/arm64/boot/dts/qcom/pms405.dtsi:96.14-99.6: Warning (unit_address_vs_reg): /soc@0/spmi@200f000/pms405@0/adc@3100/vref_1p25: node has a reg or ranges property, but no unit name\n arch/arm64/boot/dts/qcom/pms405.dtsi:101.19-104.6: Warning (unit_address_vs_reg): /soc@0/spmi@200f000/pms405@0/adc@3100/vph_pwr: node has a reg or ranges property, but no unit name\n arch/arm64/boot/dts/qcom/pms405.dtsi:106.13-109.6: Warning (unit_address_vs_reg): /soc@0/spmi@200f000/pms405@0/adc@3100/die_temp: node has a reg or ranges property, but no unit name\n arch/arm64/boot/dts/qcom/pms405.dtsi:111.27-116.6: Warning (unit_address_vs_reg): /soc@0/spmi@200f000/pms405@0/adc@3100/thermistor1: node has a reg or ranges property, but no unit name\n arch/arm64/boot/dts/qcom/pms405.dtsi:118.27-123.6: Warning (unit_address_vs_reg): /soc@0/spmi@200f000/pms405@0/adc@3100/thermistor3: node has a reg or ranges property, but no unit name\n arch/arm64/boot/dts/qcom/pms405.dtsi:125.22-130.6: Warning (unit_address_vs_reg): /soc@0/spmi@200f000/pms405@0/adc@3100/xo_temp: node has a reg or ranges property, but no unit name\n \n Reviewed-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit ba0411ddd133b34fd715f17b9c0c29027bfa92a8\nAuthor: Sai Prakash Ranjan \u003csaiprakash.ranjan@codeaurora.org\u003e\nDate: Wed Jul 10 16:59:24 2019 +0530\n\n arm64: dts: sdm845: Add device node for Last level cache controller\n \n Last level cache (aka. system cache) controller provides control\n over the last level cache present on SDM845. This cache lies after\n the memory noc, right before the DDR.\n \n Signed-off-by: Sai Prakash Ranjan \u003csaiprakash.ranjan@codeaurora.org\u003e\n Signed-off-by: Vivek Gautam \u003cvivek.gautam@codeaurora.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 0baf212eab4daf8e4882afd624e1403846418610\nAuthor: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\nDate: Wed Jul 31 14:39:55 2019 +0200\n\n soc: amlogic: meson-gx-socinfo: add A311D id\n \n Add the SoC ID for the A311D Amlogic SoC.\n \n Signed-off-by: Christian Hewitt \u003cchristianshewitt@gmail.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 887e54218183a2665c4dc59aa10477185145fad9\nAuthor: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\nDate: Tue Jun 18 17:16:02 2019 -0700\n\n arm64: dts: qcom: msm8996: Rename smmu nodes\n \n Node names shouldn\u0027t include a vendor prefix and should whenever\n possible use a generic identifier. Resolve this by renaming the smmu\n nodes \"iommu\".\n \n Reviewed-by: Vivek Gautam \u003cvivek.gautam@codeaurora.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit dee51cd0d2e8af08c74e5bbb10673af1cf0ee4a3\nAuthor: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\nDate: Sat Jul 27 14:12:57 2019 +0200\n\n ARM: dts: meson8b: mxq: add the VDDEE regulator\n \n The VDDEE regulator is basically a copy of the VCCK regulator. VDDEE\n supplies for example the Mali GPU and is controlled by PWM_D instead of\n PWM_C.\n \n Add the VDDEE PWM regulator and make it the supply of the Mali GPU.\n \n Signed-off-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 8bdf38be712d49093e51e3cf754fd427d4685ba0\nAuthor: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\nDate: Sat Jul 27 14:12:56 2019 +0200\n\n ARM: dts: meson8b: odroidc1: add the VDDEE regulator\n \n The VDDEE regulator is basically a copy of the VCCK regulator. VDDEE\n supplies for example the Mali GPU and is controlled by PWM_D instead of\n PWM_C.\n \n Add the VDDEE PWM regulator and make it the supply of the Mali GPU.\n \n Signed-off-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 087a1d8b4e4cd0f84bde82e21bad83ea3d00a216\nAuthor: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\nDate: Sat Jul 27 14:12:55 2019 +0200\n\n ARM: dts: meson8b: ec100: add the VDDEE regulator\n \n The VDDEE regulator is basically a copy of the VCCK regulator. VDDEE\n supplies for example the Mali GPU and is controlled by PWM_D instead of\n PWM_C.\n \n Add the VDDEE PWM regulator and make it the supply of the Mali GPU.\n \n Signed-off-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit ea241bdfa0ba632416a828ddaa2fde33f30de530\nAuthor: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\nDate: Sat Jul 27 14:12:54 2019 +0200\n\n ARM: dts: meson8b: add the PWM_D output pin\n \n The PWM_D output is used for the VDDEE PWM regulator which supplies for\n example the Mali GPU on the EC-100 and Odroid-C1 boards. Add the output\n pin the VDDEE regulators can be added.\n \n Signed-off-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 4f0303d43973a4086d6355db0c653a5a369b5dbe\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Thu Jul 18 11:36:23 2019 +0200\n\n ARM: dts: meson8b: add ethernet fifo sizes\n \n If unspecified in DT, the fifo sizes are not automatically detected by\n the dwmac1000 dma driver and the reported fifo sizes default to 0.\n Because of this, flow control will be turned off on the device.\n \n Add the fifo sizes provided by the datasheet in the SoC in DT so\n flow control may be enabled if necessary.\n \n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 340ff31ab00bca5c15915e70ad9ada3030c98cf8\nAuthor: Corey Minyard \u003ccminyard@mvista.com\u003e\nDate: Fri Aug 2 07:31:36 2019 -0500\n\n ipmi_si: Only schedule continuously in the thread in maintenance mode\n \n ipmi_thread() uses back-to-back schedule() to poll for command\n completion which, on some machines, can push up CPU consumption and\n heavily tax the scheduler locks leading to noticeable overall\n performance degradation.\n \n This was originally added so firmware updates through IPMI would\n complete in a timely manner. But we can\u0027t kill the scheduler\n locks for that one use case.\n \n Instead, only run schedule() continuously in maintenance mode,\n where firmware updates should run.\n \n Signed-off-by: Corey Minyard \u003ccminyard@mvista.com\u003e\n\ncommit 9b80c36353ed4cce324af21244a65984db21991b\nAuthor: Paul Moore \u003cpaul@paul-moore.com\u003e\nDate: Thu Aug 1 17:55:06 2019 -0400\n\n selinux: always return a secid from the network caches if we find one\n \n Previously if we couldn\u0027t find an entry in the cache and we failed to\n allocate memory for a new cache entry we would fail the network object\n label lookup; this is obviously not ideal. This patch fixes this so\n that we return the object label even if we can\u0027t cache the object at\n this point in time due to memory pressure.\n \n The GitHub issue tracker is below:\n * https://github.com/SELinuxProject/selinux-kernel/issues/3\n \n Signed-off-by: Paul Moore \u003cpaul@paul-moore.com\u003e\n\ncommit c33b2777d01eb0039a53f14f8c0e4cca8df501c7\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Mon Jul 29 15:02:18 2019 +0200\n\n soc: amlogic: meson-clk-measure: add G12B second cluster cpu clk\n \n Add the G12B second CPU cluster CPU and SYS_PLL measure IDs.\n \n These IDs returns 0Hz on G12A.\n \n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 3a760d986568b67d1f8411dab64608075817b90d\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Mon Jul 29 15:02:17 2019 +0200\n\n soc: amlogic: meson-clk-measure: protect measure with a mutex\n \n In order to protect clock measuring when multiple process asks for\n a measure, protect the main measure function with mutexes.\n \n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n\ncommit 0dfa0bed9662db680db2cc06457b8d152a2f22af\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Fri Aug 2 02:17:51 2019 -0400\n\n cnic: Explicitly initialize all reference counts to 0.\n \n The driver is relying on zero\u0027ed allocated memory and does not\n explicitly call atomic_set() to initialize the ref counts to 0. Add\n these atomic_set() calls so that it will be more straight forward\n to convert atomic ref counts to refcount_t.\n \n Reported-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Cc: Rasesh Mody \u003crmody@marvell.com\u003e\n Cc: \u003cGR-Linux-NIC-Dev@marvell.com\u003e\n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f07ea1d4eda2574c6b0f99576db61c86ec27ff5b\nAuthor: Ondrej Mosnacek \u003comosnace@redhat.com\u003e\nDate: Mon Jul 29 10:41:17 2019 +0200\n\n selinux: policydb - rename type_val_to_struct_array\n \n The name is overly long and inconsistent with the other *_val_to_struct\n members. Dropping the \"_array\" prefix makes the code easier to read and\n gets rid of one line over 80 characters warning.\n \n Signed-off-by: Ondrej Mosnacek \u003comosnace@redhat.com\u003e\n Signed-off-by: Paul Moore \u003cpaul@paul-moore.com\u003e\n\ncommit 2492acaf1e53a13a528e31769aa3df15b4d9f6ef\nAuthor: Ondrej Mosnacek \u003comosnace@redhat.com\u003e\nDate: Mon Jul 29 10:41:16 2019 +0200\n\n selinux: policydb - fix some checkpatch.pl warnings\n \n Fix most of the code style warnings discovered when moving code around.\n \n Signed-off-by: Ondrej Mosnacek \u003comosnace@redhat.com\u003e\n Signed-off-by: Paul Moore \u003cpaul@paul-moore.com\u003e\n\ncommit 43a4b60d04362185cd5475fd77a02bf6c56c07e4\nAuthor: David Ahern \u003cdsahern@gmail.com\u003e\nDate: Thu Aug 1 15:18:08 2019 -0700\n\n ipv6: have a single rcu unlock point in __ip6_rt_update_pmtu\n \n Simplify the unlock path in __ip6_rt_update_pmtu by using a\n single point where rcu_read_unlock is called.\n \n Signed-off-by: David Ahern \u003cdsahern@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 0eb2f2962489edb6943fe8a853de024c53aa91b2\nAuthor: Paul Moore \u003cpaul@paul-moore.com\u003e\nDate: Wed Jul 31 19:29:14 2019 -0400\n\n selinux: shuffle around policydb.c to get rid of forward declarations\n \n No code changes, but move a lot of the policydb destructors higher up\n so we can get rid of a forward declaration.\n \n This patch does expose a few old checkpatch.pl errors, but those will\n be dealt with in a separate (set of) patches.\n \n Signed-off-by: Paul Moore \u003cpaul@paul-moore.com\u003e\n\ncommit e7642f35851da14b78be7a38b0bb0d4068254859\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Thu Jul 25 16:32:24 2019 +0200\n\n backlight: drop EARLY_EVENT_BLANK support\n \n There was no users left - so drop the code to support EARLY_EVENT_BLANK.\n This patch removes the support in backlight,\n and drop the notifier in fbmem.\n \n That EARLY_EVENT_BLANK is not used can be verified that no driver set any of:\n \n lcd_ops.early_set_power()\n lcd_ops.r_early_set_power()\n \n Noticed while browsing backlight code for other reasons.\n \n v2:\n - Fix changelog to say \"EARLY_EVENT_BLANK\" (Daniel)\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Lee Jones \u003clee.jones@linaro.org\u003e\n Cc: Daniel Thompson \u003cdaniel.thompson@linaro.org\u003e\n Cc: Jingoo Han \u003cjingoohan1@gmail.com\u003e\n Cc: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: \"Michał Mirosław\" \u003cmirq-linux@rere.qmqm.pl\u003e\n Cc: Peter Rosin \u003cpeda@axentia.se\u003e\n Cc: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-fbdev@vger.kernel.org\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Acked-by: Daniel Thompson \u003cdaniel.thompson@linaro.org\u003e\n Acked-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Acked-by: Lee Jones \u003clee.jones@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725143224.GB31803@ravnborg.org\n\ncommit 558e93c93f7843297752db6c491517b311bfb19a\nAuthor: Czeslaw Zagorski \u003cczeslawx.zagorski@intel.com\u003e\nDate: Tue Jul 2 08:22:59 2019 -0400\n\n i40e: Remove unicast log when VF is leaving multicast mode.\n \n This patch removes unicast log when VF is leaving multicast mode.\n Added check of vf-\u003evf_states \u0026\n I40E_VF_STATE_MC_PROMISC/I40E_VF_STATE_UC_PROMISC.\n Without this commit, leaving multicast mode logs \"unset unicast\"\n in dmsg.\n \n Signed-off-by: Czeslaw Zagorski \u003cczeslawx.zagorski@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit b27223591606f59c1f7c042b8e3dc74affadf492\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Tue Jul 2 08:22:58 2019 -0400\n\n i40e: verify string count matches even on early return\n \n Similar to i40e_get_ethtool_stats, add a goto to verify that the data\n pointer for the strings lines up with the expected stats count. This\n helps ensure that bugs are not introduced when adding stats.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit b603f9dc20afed5e4666642c8713cafb94a23058\nAuthor: Czeslaw Zagorski \u003cczeslawx.zagorski@intel.com\u003e\nDate: Tue Jul 2 08:22:57 2019 -0400\n\n i40e: Log info when PF is entering and leaving Allmulti mode.\n \n Add log when PF is entering and leaving allmulti mode. The\n change of PF state is visible in dmesg now. Without this commit,\n entering and leaving allmulti mode is not logged in dmesg.\n \n Signed-off-by: Czeslaw Zagorski \u003cczeslawx.zagorski@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 0969402fd5dd57268bb7547d7e5ece8fcd81157d\nAuthor: Czeslaw Zagorski \u003cczeslawx.zagorski@intel.com\u003e\nDate: Tue Jul 2 08:22:56 2019 -0400\n\n i40e: Update visual effect for advertised FEC mode.\n \n Updates visual effect for advertised mode after setting desired mode.\n The mode appears in advertised FEC mode correctly, when ethtool\n interface command is called. Without this commit advertised FEC\n is displayed regardless of the settings as \"None BaseR RS\".\n \n Signed-off-by: Czeslaw Zagorski \u003cczeslawx.zagorski@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 6db6032298d2d7c919742e24a4bc59744bd72d62\nAuthor: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\nDate: Tue Jul 2 08:22:55 2019 -0400\n\n i40e: fix code comments\n \n Found a code comment that needed TLC to correct their formatting.\n \n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n\ncommit 2ad1274fa35ace5c6360762ba48d33b63da2396c\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Tue Jul 2 08:22:54 2019 -0400\n\n i40e: don\u0027t report link up for a VF who hasn\u0027t enabled queues\n \n Commit d3d657a90850 (\"i40e: update VFs of link state after\n GET_VF_RESOURCES\") modified the PF driver to notify a VF of\n its link status immediately after it requests resources.\n \n This was intended to fix reporting on VF drivers, so that they would\n properly report link status.\n \n However, some older VF drivers do not respond well to receiving a link\n up notification before queues are enabled. This can cause their state\n machine to think that it is safe to send traffic. This results in a Tx\n hang on the VF.\n \n More recent versions of the old i40evf and all versions of iavf are\n resilient to these early link status messages. However, if a VM happens\n to run an older version of the VF driver, this can be problematic.\n \n Record whether the PF has actually enabled queues for the VF. When\n reporting link status, always report link down if the queues aren\u0027t\n enabled. In this way, the VF driver will never receive a link up\n notification until after its queues are enabled.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit d9f78ceb8ffacb825b1660a510ceb04258653615\nAuthor: Slawomir Laba \u003cslawomirx.laba@intel.com\u003e\nDate: Tue Jul 2 08:22:53 2019 -0400\n\n i40e: Log disable-fw-lldp flag change by ethtool\n \n Add logging for disable-fw-lldp flag by ethtool. Added check\n for I40E_FLAG_DISABLE_FW_LLDP and logging state in dmesg.\n Without this commit there was no clear statement in dmesg\n about FW LLDP state in dmesg.\n \n Signed-off-by: Slawomir Laba \u003cslawomirx.laba@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit f21fa0606c383e8c7289ad89b4a610b441707d08\nAuthor: Dmitrii Golovanov \u003cdmitrii.golovanov@intel.com\u003e\nDate: Tue Jul 2 08:22:51 2019 -0400\n\n i40e: fix incorrect ethtool statistics veb and veb.tc_\n \n This patch fixes missing call of i40e_update_veb_stats() in function\n i40e_get_ethtool_stats() to update stats data of VEB and VEB TC\n counters before they are written into ethtool buffer.\n Before the patch ethtool counters may fell behind interface counters.\n \n Signed-off-by: Dmitrii Golovanov \u003cdmitrii.golovanov@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 74f94e41825549638a5b24410c1f62d5ea57134e\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Sun Aug 4 15:21:38 2019 +0900\n\n ALSA: firewire-lib: localize kernel APIs to start/stop each AMDTP stream\n \n As a result to support AMDTP domain, no drivers call kernel APIs to\n start/stop each AMDTP stream. This commit localize these APIs.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit c9a9ce899cd1798f054f264043bddc021b47cb1e\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Sun Aug 4 15:21:37 2019 +0900\n\n ALSA: fireface: support AMDTP domain\n \n This commit adds AMDTP domain support for ALSA fireface driver.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit ccc6c1b0e604144d4d068eb676b444397cecf1b3\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Sun Aug 4 15:21:36 2019 +0900\n\n ALSA: firewire-motu: support AMDTP domain\n \n This commit adds AMDTP domain support for ALSA firewire-motu driver.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit c281d46a51e31703183aa66ed9315446a0a10953\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Sun Aug 4 15:21:35 2019 +0900\n\n ALSA: firewire-tascam: support AMDTP domain\n \n This commit adds AMDTP domain support for ALSA firewire-tascam driver.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 9a08067ec318cbeaf0caa2d104cf677e723e02a3\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Sun Aug 4 15:21:34 2019 +0900\n\n ALSA: firewire-digi00x: support AMDTP domain\n \n This commit adds AMDTP domain support for ALSA firewire-digi00x driver.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit e9f21129b8d84debf53e06f73d340d5af136270f\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Sun Aug 4 15:21:33 2019 +0900\n\n ALSA: dice: support AMDTP domain\n \n This commit adds AMDTP domain support for ALSA dice driver.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit ac5d77864cfcb0f034221c2b3dac828ae24c722e\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Sun Aug 4 15:21:32 2019 +0900\n\n ALSA: oxfw: support AMDTP domain\n \n This commit adds AMDTP domain support for ALSA oxfw driver.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit db40eeb2469fef4efe26789ea2596372ee43ae2f\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Sun Aug 4 15:21:31 2019 +0900\n\n ALSA: fireworks: support AMDTP domain\n \n This commit adds AMDTP domain support for ALSA fireworks driver.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit b0db4d512941455894e7a13f71594b7d4638b94b\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Sun Aug 4 15:21:30 2019 +0900\n\n ALSA: bebob: support AMDTP domain\n \n This commit adds AMDTP domain support for ALSA bebob driver.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 42355abb9c33503f7b5889d71e1432fabdf2fd7f\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Sun Aug 4 15:21:29 2019 +0900\n\n ALSA: fireface: code refactoring for initialization/destruction of AMDTP stream\n \n This commit is a preparation to support AMDTP domain.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 39e522a5a45fe805a2925aaddc30a4b2c6ded642\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Sun Aug 4 15:21:28 2019 +0900\n\n ALSA: firewire-motu: code refactoring for initialization/destruction of AMDTP stream\n \n This commit is a preparation to support AMDTP domain.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 5f9625a5ba5dcf1815f123fced65f289b558aca4\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Sun Aug 4 15:21:27 2019 +0900\n\n ALSA: firewire-tascam: code refactoring for initialization/destruction of AMDTP stream\n \n This commit is a preparation to support AMDTP domain.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit d79360ebe9c8668caf0e3932216f236fffbfa1b6\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Sun Aug 4 15:21:26 2019 +0900\n\n ALSA: firewire-digi00x: code refactoring for initialization/destruction of AMDTP stream\n \n This commit is a preparation to support AMDTP domain.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 7eb7b18e9fc7c8dc3d3c186ec6b0cfa9147299f8\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Sun Aug 4 15:21:25 2019 +0900\n\n ALSA: fireworks: code refactoring for bus reset handler\n \n This commit is a preparation to support AMDTP domain.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 94491c175d6bf4b838043a3fb5adabd731d8fbab\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Sun Aug 4 15:21:24 2019 +0900\n\n ALSA: fireworks: code refactoring for initialization/destruction of AMDTP streams\n \n This commit is a preparation to support AMDTP domain.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 9b4702b06c0e25abc612e6f02f3e25a51c684a01\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Sun Aug 4 15:21:23 2019 +0900\n\n ALSA: firewire-lib: add a kernel API to start AMDTP streams in AMDTP domain\n \n This commit adds a kernel API to start a couple of isochronous contexts\n for some AMDTP streams.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 157a53eef6a016e6938d9fd266b85221fcd4eaa3\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Sun Aug 4 15:21:22 2019 +0900\n\n ALSA: firewire-lib: add a kernel API to add AMDTP stream into AMDTP domain\n \n This commit adds a kernel API to insert AMDTP stream to list in AMDTP\n domain.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 6261f90bdb0b605ffb158717ec7a03b1753aded5\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Sun Aug 4 15:21:21 2019 +0900\n\n ALSA: firewire-lib: add a kernel API to stop a couple of AMDTP streams in AMDTP domain\n \n This commit adds a kernel API to stop a couple of isochronous contexts\n for AMDTP streams. The API is not protected with any lock primitive.\n Callers should use this with enough lock against concurrent access.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 3ec3d7a3ff10692d4f8a2baa0ff18fe10a9b6ad4\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Sun Aug 4 15:21:20 2019 +0900\n\n ALSA: firewire-lib: add AMDTP domain structure to handle several isoc contexts\n \n This commit adds \u0027struct amdtp_domain\u0027 structure. This structure\n has list of instance of AMDTP stream to handle a couple of\n isochronous contexts.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit a9e21bea1f815c2ec917ecc0efe0a7049c825d5b\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Mon Aug 5 11:52:11 2019 +0100\n\n ][next] selftests: nettest: fix spelling mistake: \"potocol\" -\u003e \"protocol\"\n \n There is a spelling mistake in an error messgae. Fix it.\n \n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Acked-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6c86e5ff2617efaca3db745132de0f004a7494e7\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Sat Aug 3 11:22:21 2019 +0100\n\n drm/i915/gt: Remove stale kerneldoc for internal MOCS functions\n \n The kerneldoc were stale, generating mismatching parameters warning, but\n furthermore they were for internal routines, not part of the MOCS\n interface so the instructions were superfluous.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190803102221.21344-1-chris@chris-wilson.co.uk\n\ncommit d9a910f5ed72c11773a11cff84e0b957d3634d5f\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Sun Aug 4 19:50:52 2019 +0000\n\n drm/i915/uc: Don\u0027t fail on HuC early init errors\n \n Since commit 301efe96f777 (\"drm/i915/uc: Don\u0027t fail on HuC\n firmware failure\") we can continue driver load after error\n during HuC firmware load or authentication, but we could\n still fail on any error during early HuC initialization.\n Change that by ignoring HuC related errors until hardware\n initialization phase where we can decide about next steps.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190804195052.31140-5-michal.wajdeczko@intel.com\n\ncommit 34f0a20fa4ea8ad08b2d1fd43fa6e2febb9a4eff\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Sun Aug 4 19:50:51 2019 +0000\n\n drm/i915/uc: Remove redundant GuC support checks\n \n Since commit db81bc6eb9c0 (\"drm/i915/uc: Consider enable_guc\n modparam during fw selection\") we have started using firmware\n status as main indicator of the GuC support. No need to use\n same checks twice.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190804195052.31140-4-michal.wajdeczko@intel.com\n\ncommit 2afc794c3ea026edcf4134b683e5c478073e8781\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Sun Aug 4 19:50:50 2019 +0000\n\n drm/i915/huc: Prefer intel_huc_is_supported\n \n No need to dance with intel_uc_supports_huc(uc) as we\n can directly use intel_huc_is_supported(huc)\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190804195052.31140-3-michal.wajdeczko@intel.com\n\ncommit c8be1a5fc5c63b78f86f3c15fd700163ecfed57f\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Sun Aug 4 19:50:49 2019 +0000\n\n drm/i915/guc: Prefer intel_guc_is_submission_supported\n \n No need to use intel_uc_supports_guc_submission(uc) as we\n can directly use intel_guc_is_submission_supported(guc)\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190804195052.31140-2-michal.wajdeczko@intel.com\n\ncommit 5156eb8cd680a87123c448429129f95c3dac3c31\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Mon Jul 8 13:31:54 2019 +0200\n\n drm/bridge/parade: Drop legacy GPIO header\n \n This driver uses the new GPIO API from \u003clinux/gpio/consumer.h\u003e\n so drop the inclusion of the legacy header.\n \n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190708113154.12985-1-linus.walleij@linaro.org\n\ncommit f15543ac6f44da5973085152d6017657e5864361\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Mon Jul 8 13:30:09 2019 +0200\n\n drm/bridge/nxp-ptn3460: Drop legacy GPIO headers\n \n This driver uses exclusively the new GPIO API from\n \u003clinux/gpio/consumer.h\u003e so just drop the old API headers.\n \n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190708113009.12723-1-linus.walleij@linaro.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit b316c92e9a0ea156f91b79840f39a1d964aad55f\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Mon Jul 8 13:28:03 2019 +0200\n\n drm/bridge/megachips: Drop GPIO header\n \n This file isn\u0027t using any interfaces from \u003clinux/gpio.h\u003e so\n just drop the include.\n \n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190708112803.12432-1-linus.walleij@linaro.org\n\ncommit 2e1bc0e86ec77d5568e2bb5767529f86e61cd93a\nAuthor: Shaokun Zhang \u003czhangshaokun@hisilicon.com\u003e\nDate: Mon Aug 5 17:14:16 2019 +0800\n\n drm/pl111: Fix unused variable warning\n \n drivers/gpu/drm/pl111/pl111_display.c: In function ‘pl111_display_init’:\n drivers/gpu/drm/pl111/pl111_display.c:551:17: warning: unused variable\n ‘dev’ [-Wunused-variable]\n struct device *dev \u003d drm-\u003edev;\n ^\n Fixes: d6781e490179 (\"drm/pl111: Drop special pads config check\")\n Cc: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Shaokun Zhang \u003czhangshaokun@hisilicon.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/1564996456-55677-1-git-send-email-zhangshaokun@hisilicon.com\n\ncommit 854da7511a05c9a05a4f28fb2225349380520b96\nAuthor: Jean-Baptiste Maneyrol \u003cJManeyrol@invensense.com\u003e\nDate: Mon Aug 5 12:26:50 2019 +0000\n\n iio: imu: inv_mpu6050: be more explicit on supported chips\n \n Since every chip has a different whoami, we are not supporting all\n existing variant of all chips. Add an explicit supported chips\n list in Kconfig description.\n \n Signed-off-by: Jean-Baptiste Maneyrol \u003cjmaneyrol@invensense.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 741172d18e8a9cffb5c9c9c236d1311b6fcf6111\nAuthor: Martyn Welch \u003cmartyn.welch@collabora.com\u003e\nDate: Fri Aug 2 12:42:28 2019 +0100\n\n iio: light: noa1305: Add support for NOA1305\n \n This driver adds the initial support for the ON Semiconductor\n NOA1305 Ambient Light Sensor.\n \n Originally written by Sergei Miroshnichenko. Found here:\n https://github.com/EmcraftSystems/linux-upstream/commit/196d6cf897e632d2cb82d45484bd7a1bfdd5b6d9\n \n Signed-off-by: Sergei M \u003cfizik1@yandex.com\u003e\n Signed-off-by: Martyn Welch \u003cmartyn.welch@collabora.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit cfa1f5f27c79997926f89df5dd1b4f8c920b9623\nAuthor: Gal Pressman \u003cgalpress@amazon.com\u003e\nDate: Thu Aug 1 20:14:47 2019 +0300\n\n RDMA/efa: Rate limit admin queue error prints\n \n Admin queue error prints should never happen unless something wrong\n happened to the device. However, in the unfortunate case that it does,\n we should take extra care not to flood the log with error messages.\n \n Reviewed-by: Firas JahJah \u003cfirasj@amazon.com\u003e\n Reviewed-by: Yossi Leybovich \u003csleybo@amazon.com\u003e\n Signed-off-by: Gal Pressman \u003cgalpress@amazon.com\u003e\n Link: https://lore.kernel.org/r/20190801171447.54440-3-galpress@amazon.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 05bb411ada9508b48044ef5d84dd8bc46cece607\nAuthor: Gal Pressman \u003cgalpress@amazon.com\u003e\nDate: Thu Aug 1 20:14:46 2019 +0300\n\n RDMA/core: Introduce ratelimited ibdev printk functions\n \n Add ratelimited helpers to the ibdev_* printk functions.\n Implementation inspired by counterpart dev_*_ratelimited functions.\n \n Signed-off-by: Gal Pressman \u003cgalpress@amazon.com\u003e\n Reviewed-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Link: https://lore.kernel.org/r/20190801171447.54440-2-galpress@amazon.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 7a63b31efbb22391a3f508a2a3d97d3de7c8dca0\nAuthor: Leon Romanovsky \u003cleonro@mellanox.com\u003e\nDate: Thu Aug 1 14:48:27 2019 +0300\n\n RDMA/hns: Remove not used UAR assignment\n \n UAR in CQ is not used and generates the following compilation\n warning, clean the code by removing uar assignment.\n \n drivers/infiniband/hw/hns/hns_roce_cq.c: In function _create_user_cq_:\n drivers/infiniband/hw/hns/hns_roce_cq.c:305:27: warning: parameter _uar_ set but not used [-Wunused-but-set-parameter]\n 305 | struct hns_roce_uar *uar,\n | ~~~~~~~~~~~~~~~~~~~~~^~~\n \n Fixes: 4f8f0d5e33dd (\"RDMA/hns: Package the flow of creating cq\")\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Link: https://lore.kernel.org/r/20190801114827.24263-1-leon@kernel.org\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit f51cf9e23b70aa01fff2473d7bc8535570a8afc8\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Wed Jul 31 07:16:22 2019 -0700\n\n misc: Remove spear13xx pcie gadget driver\n \n This driver has been marked broken since 2013, see commit 98097858ccf3\n (\"misc: mark spear13xx-pcie-gadget as broken\"). Let\u0027s remove this file\n now that it\u0027s been more than 5 years of existing in a broken state.\n \n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Link: https://lore.kernel.org/r/5d41a2b7.1c69fb81.c8d56.edb6@mx.google.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 52eb6d31a1c2e85a3eaf08ab599c9ad58c890c28\nAuthor: Wu Hao \u003chao.wu@intel.com\u003e\nDate: Sun Aug 4 18:20:20 2019 +0800\n\n fpga: dfl: fme: add capability sysfs interfaces\n \n This patch adds 3 read-only sysfs interfaces for FPGA Management Engine\n (FME) block for capabilities including cache_size, fabric_version and\n socket_id.\n \n Signed-off-by: Luwei Kang \u003cluwei.kang@intel.com\u003e\n Signed-off-by: Xu Yilun \u003cyilun.xu@intel.com\u003e\n Signed-off-by: Wu Hao \u003chao.wu@intel.com\u003e\n Acked-by: Alan Tull \u003catull@kernel.org\u003e\n Signed-off-by: Moritz Fischer \u003cmdf@kernel.org\u003e\n Link: https://lore.kernel.org/r/1564914022-3710-11-git-send-email-hao.wu@intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 3c51ff772278d291117dae9cad09ddef07e0d504\nAuthor: Wu Hao \u003chao.wu@intel.com\u003e\nDate: Sun Aug 4 18:20:18 2019 +0800\n\n fpga: dfl: make uinit callback optional\n \n This patch makes uinit callback of sub features optional. With\n this change, people don\u0027t need to prepare any empty uinit callback.\n \n Signed-off-by: Wu Hao \u003chao.wu@intel.com\u003e\n Link: https://lore.kernel.org/r/1564914022-3710-9-git-send-email-hao.wu@intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 15bbb300fcef4e62e4f6063cc29e698796027b98\nAuthor: Wu Hao \u003chao.wu@intel.com\u003e\nDate: Sun Aug 4 18:20:15 2019 +0800\n\n fpga: dfl: add id_table for dfl private feature driver\n \n This patch adds id_table for each dfl private feature driver,\n it allows to reuse same private feature driver to match and support\n multiple dfl private features.\n \n Signed-off-by: Xu Yilun \u003cyilun.xu@intel.com\u003e\n Signed-off-by: Wu Hao \u003chao.wu@intel.com\u003e\n Acked-by: Moritz Fischer \u003cmdf@kernel.org\u003e\n Acked-by: Alan Tull \u003catull@kernel.org\u003e\n Signed-off-by: Moritz Fischer \u003cmdf@kernel.org\u003e\n Link: https://lore.kernel.org/r/1564914022-3710-6-git-send-email-hao.wu@intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d2ad5ac1cda7c30c9ded04d0e21aba528f1f96ec\nAuthor: Wu Hao \u003chao.wu@intel.com\u003e\nDate: Sun Aug 4 18:20:13 2019 +0800\n\n fpga: dfl: afu: add AFU state related sysfs interfaces\n \n This patch introduces more sysfs interfaces for Accelerated\n Function Unit (AFU). These interfaces allow users to read\n current AFU Power State (APx), read / clear AFU Power (APx)\n events which are sticky to identify transient APx state,\n and manage AFU\u0027s LTR (latency tolerance reporting).\n \n Signed-off-by: Ananda Ravuri \u003cananda.ravuri@intel.com\u003e\n Signed-off-by: Xu Yilun \u003cyilun.xu@intel.com\u003e\n Signed-off-by: Wu Hao \u003chao.wu@intel.com\u003e\n Acked-by: Alan Tull \u003catull@kernel.org\u003e\n Signed-off-by: Moritz Fischer \u003cmdf@kernel.org\u003e\n Link: https://lore.kernel.org/r/1564914022-3710-4-git-send-email-hao.wu@intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit bdd4f307956ae7c80a831aed67b0ddd131537481\nAuthor: Wu Hao \u003chao.wu@intel.com\u003e\nDate: Sun Aug 4 18:20:12 2019 +0800\n\n fpga: dfl: pci: enable SRIOV support.\n \n This patch enables the standard sriov support. It allows user to\n enable SRIOV (and VFs), then user could pass through accelerators\n (VFs) into virtual machine or use VFs directly in host.\n \n Signed-off-by: Zhang Yi Z \u003cyi.z.zhang@intel.com\u003e\n Signed-off-by: Xu Yilun \u003cyilun.xu@intel.com\u003e\n Signed-off-by: Wu Hao \u003chao.wu@intel.com\u003e\n Acked-by: Alan Tull \u003catull@kernel.org\u003e\n Acked-by: Moritz Fischer \u003cmdf@kernel.org\u003e\n Signed-off-by: Moritz Fischer \u003cmdf@kernel.org\u003e\n Link: https://lore.kernel.org/r/1564914022-3710-3-git-send-email-hao.wu@intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 69bb18ddfc4331ba1dea9db811caf93e95726408\nAuthor: Wu Hao \u003chao.wu@intel.com\u003e\nDate: Sun Aug 4 18:20:11 2019 +0800\n\n fpga: dfl: fme: add DFL_FPGA_FME_PORT_RELEASE/ASSIGN ioctl support.\n \n In order to support virtualization usage via PCIe SRIOV, this patch\n adds two ioctls under FPGA Management Engine (FME) to release and\n assign back the port device. In order to safely turn Port from PF\n into VF and enable PCIe SRIOV, it requires user to invoke this\n PORT_RELEASE ioctl to release port firstly to remove userspace\n interfaces, and then configure the PF/VF access register in FME.\n After disable SRIOV, it requires user to invoke this PORT_ASSIGN\n ioctl to attach the port back to PF.\n \n Ioctl interfaces:\n * DFL_FPGA_FME_PORT_RELEASE\n Release platform device of given port, it deletes port platform\n device to remove related userspace interfaces on PF. After this\n function, then it\u0027s safe to configure PF/VF access mode to VF,\n and enable VFs via SRIOV.\n \n * DFL_FPGA_FME_PORT_ASSIGN\n Assign platform device of given port back to PF. After configure\n PF/VF access mode to PF, this ioctl adds port platform device\n back to re-enable related userspace interfaces on PF.\n \n Signed-off-by: Zhang Yi Z \u003cyi.z.zhang@intel.com\u003e\n Signed-off-by: Xu Yilun \u003cyilun.xu@intel.com\u003e\n Signed-off-by: Wu Hao \u003chao.wu@intel.com\u003e\n Acked-by: Alan Tull \u003catull@kernel.org\u003e\n Acked-by: Moritz Fischer \u003cmdf@kernel.org\u003e\n Signed-off-by: Moritz Fischer \u003cmdf@kernel.org\u003e\n Link: https://lore.kernel.org/r/1564914022-3710-2-git-send-email-hao.wu@intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7008aff290e1dba6618b18219967fd0d58783127\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 2 21:50:10 2019 +0800\n\n w1: omap-hdq: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190802135010.24052-1-yuehaibing@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b0a523fa9e9d61febb9e817f07b28927190c3c2f\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 2 21:48:19 2019 +0800\n\n w1: mxc_w1: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190802134819.9088-1-yuehaibing@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 9e18a80c2e06c154c9c745091f7caec487423519\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Thu Aug 1 20:14:08 2019 +0100\n\n parport: parport_serial: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Sudip Mukherjee \u003csudipm.mukherjee@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190801191408.10977-3-sudipm.mukherjee@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 5b25380d906af3be2489066295ea8c832efc71e8\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Thu Aug 1 20:14:06 2019 +0100\n\n parport: Add missing newline at end of file\n \n \"git diff\" says:\n \n \\ No newline at end of file\n \n after modifying the file.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Sudip Mukherjee \u003csudipm.mukherjee@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190801191408.10977-1-sudipm.mukherjee@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 1441beda92e7c0ab0759d28fa0472456e929803e\nAuthor: Martyn Welch \u003cmartyn.welch@collabora.com\u003e\nDate: Fri Aug 2 12:42:27 2019 +0100\n\n dt-bindings: Add binding document for NOA1305\n \n Document the ON Semiconductor NOA1305 ambient light sensor devicetree\n bindings.\n \n Signed-off-by: Martyn Welch \u003cmartyn.welch@collabora.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 20cf4e026730104892fa1268de0371a631cee294\nAuthor: Chuck Lever \u003cchuck.lever@oracle.com\u003e\nDate: Mon Jul 29 13:22:09 2019 -0400\n\n rdma: Enable ib_alloc_cq to spread work over a device\u0027s comp_vectors\n \n Send and Receive completion is handled on a single CPU selected at\n the time each Completion Queue is allocated. Typically this is when\n an initiator instantiates an RDMA transport, or when a target\n accepts an RDMA connection.\n \n Some ULPs cannot open a connection per CPU to spread completion\n workload across available CPUs and MSI vectors. For such ULPs,\n provide an API that allows the RDMA core to select a completion\n vector based on the device\u0027s complement of available comp_vecs.\n \n ULPs that invoke ib_alloc_cq() with only comp_vector 0 are converted\n to use the new API so that their completion workloads interfere less\n with each other.\n \n Suggested-by: Håkon Bugge \u003chaakon.bugge@oracle.com\u003e\n Signed-off-by: Chuck Lever \u003cchuck.lever@oracle.com\u003e\n Reviewed-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Cc: \u003clinux-cifs@vger.kernel.org\u003e\n Cc: \u003cv9fs-developer@lists.sourceforge.net\u003e\n Link: https://lore.kernel.org/r/20190729171923.13428.52555.stgit@manet.1015granger.net\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 9cd15d521a3adcb687a0f9a312e32caaa94f44c2\nAuthor: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\nDate: Fri Aug 2 10:59:13 2019 -0700\n\n iio: remove get_irq_data_ready() function pointer and use IRQ number directly\n \n Not even sure why it was there since the beginning. Just use IRQ\n number in the sensor_data struct.\n \n Signed-off-by: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 7fc009cbd7d11d6b27ba5443dbfa80a0513829e4\nAuthor: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\nDate: Fri Aug 2 08:34:08 2019 +0100\n\n arm64: dts: renesas: r8a774c0: cat874: Add definition for 12V regulator\n \n Power rail \"D12.0V\" comes straight from the power barrel connector,\n and it\u0027s used in both main board and sub board.\n \n Signed-off-by: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit a2fe2cd26285061fd9ef2649bac529f1db33d50b\nAuthor: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\nDate: Fri Aug 2 08:34:07 2019 +0100\n\n arm64: dts: renesas: r8a774c0: Point LVDS0 to its companion LVDS1\n \n Add the new renesas,companion property to the LVDS0 node to point to the\n companion LVDS encoder LVDS1.\n Based on similar work from Laurent Pinchart for the r8a7799[05].\n \n Signed-off-by: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit c070bf341093652fd82ebd1de5f1dcf5e10afbb6\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Thu Aug 1 23:25:13 2019 +0900\n\n arm64: dts: renesas: r8a77995: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 1ab0a43afdfb6683cfd76216d3361a3813bcb75e\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Thu Aug 1 23:24:55 2019 +0900\n\n arm64: dts: renesas: r8a77995: draak: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit ddd56410dac00cb9cf36ead6b795242952a4ed6d\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Thu Aug 1 23:24:36 2019 +0900\n\n arm64: dts: renesas: r8a77990: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 880925e84dbea9786c23cebd51dfa7861129da4f\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Thu Aug 1 23:24:12 2019 +0900\n\n arm64: dts: renesas: r8a77990: ebisu: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n [geert: Sort sound child nodes]\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 52f95a0935c1a0c8073ffe3b44f0c30d62b05b37\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Thu Aug 1 23:23:52 2019 +0900\n\n arm64: dts: renesas: r8a77980: v3hsk: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 8ccb4c9788814b95c059d9663949aaea03957ac9\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Thu Aug 1 23:23:34 2019 +0900\n\n arm64: dts: renesas: r8a77980: condor: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 43afe2062267332fbb253924de0f3522b02d1a66\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Thu Aug 1 23:23:17 2019 +0900\n\n arm64: dts: renesas: r8a77970: v3msk: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 83c5cf19a3f2d249e180a7b45a2559dd6b514f54\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Thu Aug 1 23:21:53 2019 +0900\n\n arm64: dts: renesas: r8a77970: eagle: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 48cceecfa7625e700ed9c57392bd17923c4832c0\nAuthor: Martin Kepplinger \u003cmartin.kepplinger@puri.sm\u003e\nDate: Thu Aug 1 16:39:08 2019 +0200\n\n iio: imu: st_lsm6dsx: make IIO_CHAN_INFO_SCALE shared by type\n \n in_accel_x_scale, in_accel_y_scale and in_accel_z_scale are always\n the same. The scale is still defined to be in \"info_mask_separate\".\n \n Userspace (iio-sensor-proxy and others) is not used to that and only\n looks for \"in_accel_scale\" for the scaling factor to apply.\n \n Change IIO_CHAN_INFO_SCALE from being separate in all channel to be\n shared by type.\n \n This removes in_accel_x_scale, in_accel_y_scale and in_accel_z_scale and\n makes available in_accel_scale.\n \n This is an ABI change, but hopefully in the \u0027no one will notice\u0027\n category. The cleanup in interface is worth the small risk that\n there is a custom script out there somewhere that this will break.\n \n Signed-off-by: Martin Kepplinger \u003cmartin.kepplinger@puri.sm\u003e\n Acked-by: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit a6513a878e2115911f8aff656ca3beac7bb8242c\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Thu Aug 1 15:36:19 2019 +0800\n\n iio: tsl2772: Use regulator_bulk_() APIs\n \n Use regulator_bulk_() APIs to shrink driver size.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Reviewed-by: Brian Masney \u003cmasneyb@onstation.org\u003e\n Tested-by: Brian Masney \u003cmasneyb@onstation.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 92cb3e5980638a37c56091e605aa837d0af05a9d\nAuthor: Christian König \u003cchristian.koenig@amd.com\u003e\nDate: Thu Aug 1 15:11:14 2019 +0200\n\n dma-buf: fix stack corruption in dma_fence_chain_release\n \n We can\u0027t free up the chain using recursion or we run into a stack overflow.\n \n Manually free up the dangling chain nodes to avoid recursion.\n \n Signed-off-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Acked-by: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Fixes: 7bf60c52e093 (\"dma-buf: add new dma_fence_chain container v7\")\n Link: https://patchwork.freedesktop.org/patch/321612/\n\ncommit 480a98c7380799de1f1836526b4894d649e2e619\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 2 21:08:17 2019 +0800\n\n tty: serial: qcom_geni_serial: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190802130817.16220-1-yuehaibing@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7e7f631c95467bcf01e7bc94ec69df9feaed7751\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Thu Aug 1 15:36:12 2019 +0800\n\n iio: tsl2772: Use devm_iio_device_register\n \n Use devm_iio_device_register, which removes the tsl2772_remove\n function.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Reviewed-by: Brian Masney \u003cmasneyb@onstation.org\u003e\n Tested-by: Brian Masney \u003cmasneyb@onstation.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 338084135aeddb103624a6841972fb8588295cc6\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Thu Aug 1 15:36:05 2019 +0800\n\n iio: tsl2772: Use devm_add_action_or_reset for tsl2772_chip_off\n \n Use devm_add_action_or_reset to call tsl2772_chip_off\n when the device is removed.\n This also fixes the issue that the chip is turned off\n before the device is unregistered.\n \n Not marked for stable as fairly hard to hit the bug and\n this is in the middle of a set making other cleanups\n to the driver. Hence will probably need explicit backporting.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Fixes: c06c4d793584 (\"staging: iio: tsl2x7x/tsl2772: move out of staging\")\n Reviewed-by: Brian Masney \u003cmasneyb@onstation.org\u003e\n Tested-by: Brian Masney \u003cmasneyb@onstation.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit db2a17320a25a63b46ddb081a306af9ded1b906e\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Sun Aug 4 21:50:23 2019 +0530\n\n regulator: core: Add of_node_put() before return\n \n Each iteration of for_each_child_of_node puts the previous node, but in\n the case of a return from the middle of the loop, there is no put, thus\n causing a memory leak. Hence add an of_node_put before the return in\n two places.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190804162023.5673-1-nishkadg.linux@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 79776da0989733a5bac0a1e635e3a284c3f5c745\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:52:12 2019 +0900\n\n ASoC: soc-component: add snd_soc_pcm_component_pcm_free()\n \n Current ALSA SoC is directly using component-\u003edriver-\u003exxx,\n thus, the code nested deeply, and it makes code difficult to read,\n and is not good for encapsulation.\n \n We want to implement component related function at soc-component.c,\n but, some of them need to care whole snd_soc_pcm_runtime (\u003d rtd)\n connected component.\n \n Let\u0027s call component related function which need to care with\n for_each_rtdcom() loop as snd_soc_pcm_component_xxx().\n This patch adds new snd_soc_pcm_component_pcm_free() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87k1c54czu.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 7484291e9b7564af65b2581dcdebeeaf98bc86d0\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:52:08 2019 +0900\n\n ASoC: soc-component: add snd_soc_pcm_component_pcm_new()\n \n Current ALSA SoC is directly using component-\u003edriver-\u003exxx,\n thus, the code nested deeply, and it makes code difficult to read,\n and is not good for encapsulation.\n \n We want to implement component related function at soc-component.c,\n but, some of them need to care whole snd_soc_pcm_runtime (\u003d rtd)\n connected component.\n \n Let\u0027s call component related function which need to care with\n for_each_rtdcom() loop as snd_soc_pcm_component_xxx().\n This patch adds new snd_soc_pcm_component_pcm() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87lfwl4czy.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 205875e1a12ef9c61e939db9ded90fe3f6352e75\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:52:04 2019 +0900\n\n ASoC: soc-component: add snd_soc_pcm_component_mmap()\n \n Current ALSA SoC is directly using component-\u003edriver-\u003eops-\u003exxx,\n thus, the code nested deeply, and it makes code difficult to read,\n and is not good for encapsulation.\n \n We want to implement component related function at soc-component.c,\n but, some of them need to care whole snd_soc_pcm_runtime (\u003d rtd)\n connected component.\n \n Let\u0027s call component related function which need to care with\n for_each_rtdcom() loop as snd_soc_pcm_component_xxx().\n This patch adds new snd_soc_pcm_component_mmap() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87muh14d02.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9c712e4f57229081e837d593fc1e4183b068a41c\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:52:00 2019 +0900\n\n ASoC: soc-component: add snd_soc_pcm_component_page()\n \n Current ALSA SoC is directly using component-\u003edriver-\u003eops-\u003exxx,\n thus, the code nested deeply, and it makes code difficult to read,\n and is not good for encapsulation.\n \n We want to implement component related function at soc-component.c,\n but, some of them need to care whole snd_soc_pcm_runtime (\u003d rtd)\n connected component.\n \n Let\u0027s call component related function which need to care with\n for_each_rtdcom() loop as snd_soc_pcm_component_xxx().\n This patch adds new snd_soc_pcm_component_page() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87o91h4d06.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 82d81f5cced36e480b581ae51c595e2deb9f2d56\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:51:56 2019 +0900\n\n ASoC: soc-component: add snd_soc_pcm_component_copy_user()\n \n Current ALSA SoC is directly using component-\u003edriver-\u003eops-\u003exxx,\n thus, the code nested deeply, and it makes code difficult to read,\n and is not good for encapsulation.\n \n We want to implement component related function at soc-component.c,\n but, some of them need to care whole snd_soc_pcm_runtime (\u003d rtd)\n connected component.\n \n Let\u0027s call component related function which need to care with\n for_each_rtdcom() loop as snd_soc_pcm_component_xxx().\n This patch adds new snd_soc_pcm_component_copy_user() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87pnlx4d0a.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 96a47908d8769479f5217bf3f432ccdc06a29747\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:51:51 2019 +0900\n\n ASoC: soc-component: add snd_soc_pcm_component_ioctrl()\n \n Current ALSA SoC is directly using component-\u003edriver-\u003eops-\u003exxx,\n thus, the code nested deeply, and it makes code difficult to read,\n and is not good for encapsulation.\n \n We want to implement component related function at soc-component.c,\n but, some of them need to care whole snd_soc_pcm_runtime (\u003d rtd)\n connected component.\n \n Let\u0027s call component related function which need to care with\n for_each_rtdcom() loop as snd_soc_pcm_component_xxx().\n This patch adds new snd_soc_pcm_component_ioctrl() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87r26d4d0f.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 0035e2565b93e0902a06320ba1716bc1ddd753b3\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:51:47 2019 +0900\n\n ASoC: soc-component: add snd_soc_pcm_component_pointer()\n \n Current ALSA SoC is directly using component-\u003edriver-\u003eops-\u003exxx,\n thus, the code nested deeply, and it makes code difficult to read,\n and is not good for encapsulation.\n \n We want to implement component related function at soc-component.c,\n but, some of them need to care whole snd_soc_pcm_runtime (\u003d rtd)\n connected component.\n \n Let\u0027s call component related function which need to care with\n for_each_rtdcom() loop as snd_soc_pcm_component_xxx().\n This patch adds new snd_soc_pcm_component_pointer() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87sgqt4d0j.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 7951b14611851bdae18e9bca18015b1d84731d0d\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:51:43 2019 +0900\n\n ASoC: soc-component: move snd_soc_component_set_bias_level()\n \n Current soc-dapm / soc-core are using a long way round to call\n .set_bias_level.\n \n if (driver-\u003eset_bias_level)\n dapm-\u003eset_bias_level \u003d ...;\n ...\n if (dapm-\u003eset_bias_level)\n ret \u003d dapm-\u003eset_bias_level(...);\n \n We can directly call it via driver-\u003eset_bias_level.\n One note here is that both Card and Component have dapm,\n but, Card\u0027s dapm doesn\u0027t have dapm-\u003ecomponent.\n We need to check it.\n \n This patch moves snd_soc_component_set_bias_level() to soc-component.c\n and updates parameters.\n dapm-\u003eset_bias_level is no longer needed\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87tvb94d0n.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 8e2a990d76aced95c6f01c2d67d8835c86f0ca67\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:51:39 2019 +0900\n\n ASoC: soc-component: move snd_soc_component_stream_event()\n \n Current soc-dapm / soc-core are using a long way round to call\n .stream_event.\n \n if (driver-\u003estream_event)\n dapm-\u003estream_event \u003d ...;\n ...\n if (dapm-\u003estream_event)\n ret \u003d dapm-\u003estream_event(...);\n \n We can directly call it via driver-\u003estream_event.\n One note here is that both Card and Component have dapm,\n but, Card\u0027s dapm doesn\u0027t have dapm-\u003ecomponent.\n We need to check it.\n \n This patch moves snd_soc_component_stream_event() to soc-component.c\n and updates parameters.\n dapm-\u003estream_event is no longer needed\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87v9vp4d0r.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 6b9ab1cf4b21b92c4b3bbebe1b5dc4f6d44f35b6\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Thu Aug 1 15:35:57 2019 +0800\n\n iio: tsl2772: Use devm_add_action_or_reset\n \n Use devm_add_action_or_reset to remove the call to\n tsl2772_disable_regulators_action to simplify the error path.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Reviewed-by: Brian Masney \u003cmasneyb@onstation.org\u003e\n Tested-by: Brian Masney \u003cmasneyb@onstation.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 9d415fbf773f162a5c274e671741c6fa94b74287\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:51:35 2019 +0900\n\n ASoC: soc-component: move snd_soc_component_seq_notifier()\n \n Current soc-dapm / soc-core are using a long way round to call\n .seq_notifier.\n \n if (driver-\u003eseq_notifier)\n dapm-\u003eseq_notifier \u003d ...;\n ...\n if (dapm-\u003eseq_notifier)\n ret \u003d dapm-\u003eseq_notifier(...);\n \n We can directly call it via driver-\u003eseq_notifier.\n One note here is that both Card and Component have dapm,\n but, Card\u0027s dapm doesn\u0027t have dapm-\u003ecomponent.\n We need to check it.\n \n This patch moves snd_soc_component_seq_notifier() to soc-component.c,\n and updates parameters.\n dapm-\u003eseq_notifier is no longer needed\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87wog54d0v.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit a2a341752558cc67d6fe5c8ada7c16f9c3690f89\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:51:31 2019 +0900\n\n ASoC: soc-component: add snd_soc_component_of_xlate_dai_name()\n \n Current ALSA SoC is directly using component-\u003edriver-\u003exxx,\n thus, it is deep nested, and makes code difficult to read,\n and is not good for encapsulation.\n This patch adds new snd_soc_component_of_xlate_dai_name() and use it\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87y30l4d0z.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 2c7b1704819435d188c7697c6815f788bf9e6200\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:51:26 2019 +0900\n\n ASoC: soc-component: add snd_soc_component_of_xlate_dai_id()\n \n Current ALSA SoC is directly using component-\u003edriver-\u003exxx,\n thus, it is deep nested, and makes code difficult to read,\n and is not good for encapsulation.\n This patch adds new snd_soc_component_of_xlate_dai_id() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87zhl14d14.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 03b34dd7d87ce3493cb1837c9e59c3b3aac4724f\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:51:22 2019 +0900\n\n ASoC: soc-component: add snd_soc_component_remove()\n \n Current ALSA SoC is directly using component-\u003edriver-\u003exxx,\n thus, it is deep nested, and makes code difficult to read,\n and is not good for encapsulation.\n This patch adds new snd_soc_component_remove() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/871ryd5rlo.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 08e837dd9e39bd3e25b1fa1a13f6ba44040e3f0d\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:51:17 2019 +0900\n\n ASoC: soc-component: add snd_soc_component_probe()\n \n Current ALSA SoC is directly using component-\u003edriver-\u003exxx,\n thus, it is deep nested, and makes code difficult to read,\n and is not good for encapsulation.\n This patch adds new snd_soc_component_probe() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/8736it5rlt.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit e40fadbcef583808c11d2e86b8ac1c652731468e\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:51:13 2019 +0900\n\n ASoC: soc-component: add snd_soc_component_is_suspended()\n \n Current ALSA SoC is directly using component-\u003exxx,\n But, it is not good for encapsulation.\n This patch adds new snd_soc_component_is_suspended() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/874l395rlx.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9a840cbac77a90e8406296aaa132ebf2c84ed9e3\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:51:08 2019 +0900\n\n ASoC: soc-component: add snd_soc_component_resume()\n \n Current ALSA SoC is directly using component-\u003edriver-\u003exxx,\n thus, it is deep nested, and makes code difficult to read,\n and is not good for encapsulation.\n This patch adds new snd_soc_component_resume() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/875znp5rm2.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 66c51573b89d0a5c1089139a2f0dd029a755c37d\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:50:34 2019 +0900\n\n ASoC: soc-component: add snd_soc_component_suspend()\n \n Current ALSA SoC is directly using component-\u003edriver-\u003exxx,\n thus, it is deep nested, and makes code difficult to read,\n and is not good for encapsulation.\n This patch adds new snd_soc_component_suspend() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/877e855rn0.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 5693d50c830272cb3c4a04d2ce4db502debd1259\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:50:29 2019 +0900\n\n ASoC: soc-component: add snd_soc_component_trigger()\n \n Current ALSA SoC is directly using component-\u003edriver-\u003eops-\u003exxx,\n thus, it is deep nested, and makes code difficult to read,\n and is not good for encapsulation.\n This patch adds new snd_soc_component_trigger() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/878ssl5rn5.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 582ee9c5ad2661b359309f379f5c411fd33e511a\nAuthor: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\nDate: Mon Aug 5 22:24:12 2019 +0800\n\n USB: Disable USB2 LPM at shutdown\n \n The QCA Rome USB Bluetooth controller has several issues once LPM gets\n enabled:\n - Fails to get enumerated in coldboot. [1]\n - Drains more power (~ 0.2W) when the system is in S5. [2]\n - Disappears after a warmboot. [2]\n \n The issue happens because the device lingers at LPM L1 in S5, so device\n can\u0027t get enumerated even after a reboot.\n \n Disable LPM at shutdown to solve the issue.\n \n [1] https://bugs.launchpad.net/bugs/1757218\n [2] https://patchwork.kernel.org/patch/10607097/\n \n Signed-off-by: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190805142412.23965-1-kai.heng.feng@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d3523b631403897bf560927321ca8c2ea0aed873\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Aug 2 21:04:08 2019 +0800\n\n usb: dwc3: meson-g12a: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Acked-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://lore.kernel.org/r/20190802130408.20336-1-yuehaibing@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit eae7136aa2083699c69de5890fd6c32c501952b5\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:50:24 2019 +0900\n\n ASoC: soc-component: add snd_soc_component_hw_free()\n \n Current ALSA SoC is directly using component-\u003edriver-\u003eops-\u003exxx,\n thus, it is deep nested, and makes code difficult to read,\n and is not good for encapsulation.\n This patch adds new snd_soc_component_hw_free() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87a7d15rna.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 245c539a1206d74e8508a07550fb7c99d0817b8c\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:50:19 2019 +0900\n\n ASoC: soc-component: add snd_soc_component_hw_params()\n \n Current ALSA SoC is directly using component-\u003edriver-\u003eops-\u003exxx,\n thus, it is deep nested, and makes code difficult to read,\n and is not good for encapsulation.\n This patch adds new snd_soc_component_hw_params() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87blxh5rnf.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 6d53723380ed73a2ced648d6e92774b39e5af1bd\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:50:13 2019 +0900\n\n ASoC: soc-component: add snd_soc_component_prepare()\n \n Current ALSA SoC is directly using component-\u003edriver-\u003eops-\u003exxx,\n thus, it is deep nested, and makes code difficult to read,\n and is not good for encapsulation.\n This patch adds new snd_soc_component_prepare() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87d0hx5rnm.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 3672beb8cad6beb7b8c017514aef0f4f507debcf\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:50:07 2019 +0900\n\n ASoC: soc-component: add snd_soc_component_close()\n \n Current ALSA SoC is directly using component-\u003edriver-\u003eops-\u003exxx,\n thus, it is deep nested, and makes code difficult to read,\n and is not good for encapsulation.\n This patch adds new snd_soc_component_close() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87ef2d5rnr.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit ae2f4849286eed48a3aa79a7b73bb5bcd0c9213b\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:50:01 2019 +0900\n\n ASoC: soc-component: add snd_soc_component_open()\n \n Current ALSA SoC is directly using component-\u003edriver-\u003eops-\u003exxx,\n thus, it is deep nested, and makes code difficult to read,\n and is not good for encapsulation.\n This patch adds new snd_soc_component_open() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87ftmt5rnx.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 4a81e8f30d0b422b7f10562952124d719f73b071\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:49:54 2019 +0900\n\n ASoC: soc-component: add snd_soc_component_get/put()\n \n ALSA SoC is calling try_module_get()/module_put() based on\n component-\u003edriver-\u003emodule_get_upon_open.\n To keep simple and readable code, we should create its function.\n This patch adds new snd_soc_component_get/put().\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87h8795ro4.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 4ff1fef10f353b928bcc9d56d31fda53f2c43191\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:49:48 2019 +0900\n\n ASoC: add soc-component.c\n \n ALSA SoC has many snd_soc_component_xxx(), but these are randomly\n located in many files. Because of it, code is difficult to read.\n This patch creates new soc-component.c, and moves existing\n snd_soc_component_xxx() into it.\n But not yet fully. We need more cleanup it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87imrp5roa.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 57cb28856ed8db11faa7cc1c751eb6ad00e7c3f0\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:49:43 2019 +0900\n\n ASoC: soc-pcm: remove soc_fill_silence()\n \n No ALSA SoC driver has .fill_silence at component-\u003edriver-\u003eops.\n We can revive it if some-driver want to use it, but let\u0027s remove it\n so far to avoid maintaining complex code\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87k1c55rof.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 4efbb20971af527990f3c51056d142e3f7dc6b4a\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:49:29 2019 +0900\n\n ASoC: soc-pcm: remove soc_rtdcom_copy_kernel()\n \n No ALSA SoC driver has .copy_kernel at component-\u003edriver-\u003eops.\n We can revive it if some-driver want to use it, but let\u0027s remove it\n so far to avoid maintaining complex code\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87lfwl5rot.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 18bd7b5ef3a313a9af1314508bca1ed982c05f75\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Fri Jul 26 13:49:23 2019 +0900\n\n ASoC: soc-pcm: remove soc_rtdcom_ack()\n \n No ALSA SoC driver has .ack at component-\u003edriver-\u003eops.\n We can revive it if some-driver want to use it, but let\u0027s remove it\n so far to avoid maintaining complex code\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87muh15roz.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 36d042bd6865ada4fc5b14e2cb5fd325eeb248db\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Sun Aug 4 08:18:32 2019 +0530\n\n staging: rtl8192e: Make use kmemdup\n \n As kmemdup API does kmalloc + memcpy . We can make use of it instead of\n calling kmalloc and memcpy independetly.\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190804024832.GA14352@hari-Inspiron-1545\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d8ba499d53d92fbee87638f3313b2281d04300bd\nAuthor: Saurav Girepunje \u003csaurav.girepunje@gmail.com\u003e\nDate: Sun Aug 4 00:55:55 2019 +0530\n\n staging: rtl8723bs: core: add spaces around \u0027-\u0027, \u0027+\u0027, \u0027\u0026\u0027,\u0027|\u0027 and \u0027?\u0027\n \n Add spaces around \u0027-\u0027, \u0027+\u0027, \u0027\u0026\u0027,\u0027|\u0027 and \u0027?\u0027 to improve readability\n and follow kernel coding style in rtw_ap.c\n \n Signed-off-by: Saurav Girepunje \u003csaurav.girepunje@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190803192551.GA24041@saurav\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 67ca96bbd0315cc804668d42568934b88fe1d968\nAuthor: Saurav Girepunje \u003csaurav.girepunje@gmail.com\u003e\nDate: Sun Aug 4 00:17:54 2019 +0530\n\n staging: rtl8723bs: core: Remove Unnecessary parentheses\n \n Remove Unnecessary parentheses around padapter-\u003emlmepriv,\n pmlmeext-\u003emlmext_info and pmlmeinfo-\u003enetwork in rtw_ap.c\n \n Signed-off-by: Saurav Girepunje \u003csaurav.girepunje@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190803184750.GA23507@saurav\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 783f9a6a2d2c2b1e5c5bc51f6407e66473385b3a\nAuthor: Saurav Girepunje \u003csaurav.girepunje@gmail.com\u003e\nDate: Sun Aug 4 00:00:54 2019 +0530\n\n staging: rtl8723bs: core: correct spelling mistake in rtw_ap.c\n \n Correct spelling mistake in rtw_ap.c\n \n Signed-off-by: Saurav Girepunje \u003csaurav.girepunje@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190803183051.GA23151@saurav\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit fe4e6e0368885f77fef81adff486f4f7116e9e8f\nAuthor: Saurav Girepunje \u003csaurav.girepunje@gmail.com\u003e\nDate: Sat Aug 3 18:48:42 2019 +0530\n\n staging: rtl8188eu: core: add spaces around \u0027-\u0027, \u0027+\u0027, \u0027\u003e\u003e\u0027,\u0027\u003c\u003c\u0027 and \u0027*\u0027 in rtw_efuse.c\n \n Add spaces around \u0027-\u0027, \u0027+\u0027, \u0027\u003e\u003e\u0027,\u0027\u003c\u003c\u0027 and \u0027*\u0027 to improve readability\n and follow kernel coding style.\n \n Signed-off-by: Saurav Girepunje \u003csaurav.girepunje@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190803131839.GA14077@saurav\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7c279229f980e5423c946fd692860b4071e0c708\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:19 2019 -0700\n\n iio: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Jonathan Cameron \u003cjic23@kernel.org\u003e\n Cc: Hartmut Knaack \u003cknaack.h@gmx.de\u003e\n Cc: Lars-Peter Clausen \u003clars@metafoo.de\u003e\n Cc: Peter Meerwald-Stadler \u003cpmeerw@pmeerw.net\u003e\n Cc: linux-iio@vger.kernel.org\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit d3017f5fbb2dc597b8cc9b59144c63d9a1396392\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Mon Jul 29 09:43:01 2019 +0800\n\n iio: light: si1145: Use device-managed APIs\n \n Use device-managed APIs to simplify the code.\n The remove functions are redundant now and can\n be deleted.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit eb1d8ebf2d0a5648a4e13e44d2906f9bb4e28352\nAuthor: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\nDate: Wed Jul 31 14:49:00 2019 -0700\n\n iio:pressure: preenable/postenable/predisable fixup for ST press buffer\n \n This patch is trying to cleanup for good the buffers operation functions.\n There is no need of using preenable, all can be done into\n postenable. Let\u0027s also use logical sequence of operations as\n already done in accel driver.\n Finally also rename the goto label using operation to perform and not\n where it fails.\n \n Not stable material as not fixing a \u0027bug\u0027 but rather bringing the\n driver in line with general \u0027patterns\u0027 to allow a subsystem wide\n cleanup.\n \n Signed-off-by: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\n Reviewed-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 6ee19af415c55f8fee565613a2fa7892030ca9ac\nAuthor: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\nDate: Wed Jul 31 14:48:59 2019 -0700\n\n iio:magn: preenable/postenable/predisable fixup for ST magn buffer\n \n This patch is trying to cleanup for good the buffers operation functions.\n Let\u0027s rename the goto label using operation to perform and not\n where it fails.\n \n Not stable material as not fixing a \u0027bug\u0027 but rather bringing the\n driver in line with general \u0027patterns\u0027 to allow a subsystem wide\n cleanup.\n \n Signed-off-by: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\n Reviewed-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 21b1ee3190340007378448fa1cae2f46046c2d35\nAuthor: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\nDate: Wed Jul 31 14:48:58 2019 -0700\n\n iio:gyro: preenable/postenable/predisable fixup for ST gyro buffer\n \n This patch is trying to cleanup for good the buffers operation functions.\n There is no need of using preenable, all can be done into\n postenable. Let\u0027s also use logical sequence of operations as\n already done in accel driver.\n Finally also rename the goto label using operation to perform and not\n where it fails.\n \n Not stable material as not fixing a \u0027bug\u0027 but rather bringing the\n driver in line with general \u0027patterns\u0027 to allow a subsystem wide\n cleanup.\n \n Signed-off-by: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\n Reviewed-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit d4e851983beec3c2c9f20f8be8d819ba4fcf3489\nAuthor: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\nDate: Wed Jul 31 14:48:57 2019 -0700\n\n iio:accel: preenable/postenable/predisable fixup for ST accel buffer\n \n This patch is trying to cleanup for good the buffers operation functions.\n There is no need of using preenable, all can be done into\n postenable.\n Let\u0027s also rename the goto label using operation to perform and not\n where it fails.\n \n Note, not stable material. This is cleaning up the logic flow rather\n than fixing a bug.\n \n Signed-off-by: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\n Reviewed-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 5a57a9bf52b0ab3fca3e14265b98c8051b288e1b\nAuthor: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\nDate: Mon Aug 5 12:18:44 2019 +0200\n\n dt-bindings: iio: imu: st_lsm6dsx: add ism330dhcx device bindings\n \n Signed-off-by: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit db947a79fc3aa78e995d0d4879f4aa7d9927e171\nAuthor: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\nDate: Mon Aug 5 12:18:43 2019 +0200\n\n iio: imu: st_lsm6dsx: add support to ISM330DHCX\n \n Add support to STM ISM330DHCX 6-axis (acc + gyro) Mems sensor\n https://www.st.com/resource/en/datasheet/ism330dhcx.pdf\n \n Signed-off-by: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit f48bc49b97470bbd27fcdaab159973e0c26c3d4f\nAuthor: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\nDate: Mon Jul 29 21:22:31 2019 +0200\n\n iio: imu: st_lsm6dsx: move iio chan definitions in st_lsm6dsx_sensor_settings\n \n Move IIO channel definitions in st_lsm6dsx_sensor_settings in order to\n support sensors with different channels maps.\n This is a preliminary patch to add support for LSM9DS1 sensor to\n st_lsm6dsx driver.\n \n Signed-off-by: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\n Tested-by: Martin Kepplinger \u003cmartin.kepplinger@puri.sm\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit ec19a6f4999344fcd5e3a3bf8037f8c9e8c24433\nAuthor: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\nDate: Mon Jul 29 21:22:30 2019 +0200\n\n iio: imu: st_lsm6dsx: inline output acc/gyro output regs\n \n Remove output register definition and inline register value since\n they are used only for iio channel definition. This is a preliminary\n patch to add support for LSM9DS1 sensor to st_lsm6dsx driver\n \n Signed-off-by: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\n Tested-by: Martin Kepplinger \u003cmartin.kepplinger@puri.sm\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 640aca3fb897866297402f1387c12317d084b492\nAuthor: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\nDate: Mon Jul 29 21:22:29 2019 +0200\n\n iio: imu: st_lsm6dsx: move fs_table in st_lsm6dsx_sensor_settings\n \n Move fs_table in st_lsm6dsx_sensor_settings in order to support\n sensors with different gain maps. This is a preliminary patch to add\n support for LSM9DS1 sensor to st_lsm6dsx driver\n \n Tested-by: Martin Kepplinger \u003cmartin.kepplinger@puri.sm\u003e\n Signed-off-by: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 40dd7343897760c4b617faa78d213e25652de9a6\nAuthor: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\nDate: Mon Jul 29 21:22:28 2019 +0200\n\n iio: imu: st_lsm6dsx: move odr_table in st_lsm6dsx_sensor_settings\n \n Move sensor odr table in st_lsm6dsx_sensor_settings in order to support\n sensors with different odr maps. This is a preliminary patch to add\n support for LSM9DS1 sensor to st_lsm6dsx driver\n \n Tested-by: Martin Kepplinger \u003cmartin.kepplinger@puri.sm\u003e\n Signed-off-by: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit d42974e438feebda372fe4a39d7761cc596abc57\nAuthor: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\nDate: Thu Jun 27 12:59:11 2019 -0300\n\n media: venus: dec: populate properly timestamps and flags for capture buffers\n \n Cache flags, timestamps and timecode structure of OUTPUT buffers\n in per-instance structure array and fill correctly the same when\n the CAPTURE buffers are done.\n \n This will make v4l2-compliance decoder streaming test happy.\n \n Signed-off-by: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 059790467bf4a75441615c69e5efd2b4bac3cba4\nAuthor: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\nDate: Thu Jun 27 09:09:34 2019 -0300\n\n media: venus: helpers: handle correctly vbuf field\n \n Correct handling of OUTPUT buffers field and make v4l2-compliance\n happy.\n \n Signed-off-by: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit beac82904a870b220d45b883e493b869d8d0dce5\nAuthor: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\nDate: Thu Jan 17 07:43:31 2019 -0200\n\n media: venus: make decoder compliant with stateful codec API\n \n This refactors code for start/stop streaming vb2 operations and\n adds a state machine handling similar to the one in stateful codec\n API documentation. One major change is that now the HFI session is\n started on STREAMON(OUTPUT) and stopped on REQBUF(OUTPUT,count\u003d0),\n during that time STREAMOFF(CAP,OUT) just flush buffers but doesn\u0027t\n stop the session. The other major change is that now the capture\n and output queues are completely separated.\n \n Signed-off-by: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit be76f150bcadec1707e3718b0750a8b2261b3c20\nAuthor: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\nDate: Wed Jan 16 14:48:47 2019 -0200\n\n media: venus: vdec: allow bigger sizeimage set by clients\n \n In most of the cases the client will know better what could be\n the maximum size for compressed data buffers. Change the driver\n to permit the user to set bigger size for the compressed buffer\n but make reasonable sanitation.\n \n Signed-off-by: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit dff38165a5a873741c67e728f80d5e923852cc67\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Sun Jul 28 22:36:30 2019 +0800\n\n iio: light: cm3323: Use device-managed APIs\n \n Use device-managed APIs to simplify the code.\n The remove functions are redundant now and can\n be deleted.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit cef697d8a0632f649c9b0450e14a370b85d2d10d\nAuthor: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\nDate: Wed Jan 16 14:40:58 2019 -0200\n\n media: venus: vdec_ctrls: get real minimum buffers for capture\n \n Until now we returned num_output_bufs set during reqbuf but\n that could be wrong when we implement stateful Codec API. So\n get the minimum buffers for capture from HFI. This is supposed\n to be called after stream header parsing, i.e. after dequeue\n v4l2 event for change resolution.\n \n Signed-off-by: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 14ea00d65c657707b425d5cd8035c6c94ad9fe56\nAuthor: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\nDate: Wed Jan 16 11:48:04 2019 -0200\n\n media: venus: helpers: add three more helper functions\n \n This adds three more helper functions:\n * for internal buffers reallocation, applicable when we are doing\n dynamic resolution change\n * for initial buffer processing of capture and output queue buffer\n types\n \n All of them will be needed for stateful Codec API support.\n \n Signed-off-by: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit a367b9ab3eb5de3836af1ce2a6cae4d56e7de4bd\nAuthor: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\nDate: Wed Jan 16 11:41:14 2019 -0200\n\n media: venus: hfi: return an error if session_init is already called\n \n This makes hfi_session_init to return an error when it is\n already called without a call to hfi_session_deinit.\n \n Signed-off-by: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit e42dbe8e86e89f77921bd117873604b9e26bbb19\nAuthor: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\nDate: Wed Jan 16 11:39:08 2019 -0200\n\n media: venus: hfi: export few HFI functions\n \n Export few HFI functions to use them from decoder to implement\n more granular control needed for stateful Codec API compliance.\n \n Signed-off-by: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 05dea7980d0b9eba9f7fc5b269985d94a108e133\nAuthor: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\nDate: Wed Jan 16 11:08:36 2019 -0200\n\n media: venus: hfi: add type argument to hfi flush function\n \n Make hfi_flush function to receive an argument for the type\n of flush.\n \n Signed-off-by: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 1e485ee5a7240b1c230ef0b74899253a512deb60\nAuthor: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\nDate: Wed Jan 16 10:56:33 2019 -0200\n\n media: venus: helpers: export few helper functions\n \n Here we export few helper function to use them from decoder to\n implement more granular control needed for stateful Codec API\n compliance.\n \n Signed-off-by: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit d162aea8f633f885ebdd1d235764afe601d97458\nAuthor: Malathi Gottam \u003cmgottam@codeaurora.org\u003e\nDate: Mon Nov 26 08:12:53 2018 -0200\n\n media: venus: venc: amend buffer size for bitstream plane\n \n Accept the buffer size requested by client and compare it\n against driver calculated size and set the maximum to\n bitstream plane.\n \n Signed-off-by: Malathi Gottam \u003cmgottam@codeaurora.org\u003e\n Signed-off-by: Stanimir Varbanov \u003cstanimir.varbanov@linaro.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 5e8447d1b0a79c318a8a71cf5e6533a5a5e1b02c\nAuthor: Martin Kaiser \u003cmartin@kaiser.cx\u003e\nDate: Wed Jul 31 16:07:06 2019 +0200\n\n dt-bindings: iio: potentiometer: add max5432.yaml binding\n \n Add a binding for the Maxim Integrated MAX5432-MAX5435 family of digital\n potentiometers.\n \n Signed-off-by: Martin Kaiser \u003cmartin@kaiser.cx\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 425c6f1063955fd2335b44b6ba628de3e3a81b4d\nAuthor: Martin Kaiser \u003cmartin@kaiser.cx\u003e\nDate: Wed Jul 31 16:07:05 2019 +0200\n\n iio: potentiometer: add a driver for Maxim 5432-5435\n \n Add a driver for the Maxim Integrated MAX5432-MAX5435 family of digital\n potentiometers.\n \n These potentiometers are connected via I2C and have 32 wiper\n positions.\n \n Supported functionality\n - set the volatile wiper position\n - read the potentiometer scale\n \n Datasheet:\n https://datasheets.maximintegrated.com/en/ds/MAX5432-MAX5435.pdf\n \n Signed-off-by: Martin Kaiser \u003cmartin@kaiser.cx\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 5b6cc1277a0536dae20f37d531f29b3416cee670\nAuthor: Olivier Gay \u003cogay@logitech.com\u003e\nDate: Mon Jul 29 19:21:52 2019 +0200\n\n HID: logitech-dj: extend consumer usages range\n \n Extend the range of usage codes in the consumer page descriptor of\n the driver. Some Logitech HID devices send usages in that upper range.\n \n Signed-off-by: Olivier Gay \u003cogay@logitech.com\u003e\n Tested-by: Benson Leung \u003cbleung@chromium.org\u003e\n Signed-off-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n\ncommit c801aff1a57697d79618ea3e84dc8ff79617669d\nAuthor: Fabian Henneke \u003cfabian.henneke@gmail.com\u003e\nDate: Thu Jul 18 22:50:58 2019 +0200\n\n hiddev: Return EPOLLOUT from hiddev_poll\n \n Always return EPOLLOUT from hiddev_poll when a device is connected.\n This is safe since hiddev_write always fails and improves compatibility\n with tools like socat.\n \n Signed-off-by: Fabian Henneke \u003cfabian@henneke.me\u003e\n In-reply-to: \u003cnycvar.YFH.7.76.1907171333160.5899@cbobk.fhfr.pm\u003e\n Signed-off-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n\ncommit 378b80370aa1fe50f9c48a3ac8af3e416e73b89f\nAuthor: Fabian Henneke \u003cfabian.henneke@gmail.com\u003e\nDate: Tue Jul 9 13:03:37 2019 +0200\n\n hidraw: Return EPOLLOUT from hidraw_poll\n \n Always return EPOLLOUT from hidraw_poll when a device is connected.\n This is safe since writes are always possible (but will always block).\n \n hidraw does not support non-blocking writes and instead always calls\n blocking backend functions on write requests. Hence, so far, a call to\n poll never returned EPOLLOUT, which confuses tools like socat.\n \n Signed-off-by: Fabian Henneke \u003cfabian.henneke@gmail.com\u003e\n In-reply-to: \u003cCA+hv5qkyis03CgYTWeWX9cr0my-d2Oe+aZo+mjmWRXgjrGqyrw@mail.gmail.com\u003e\n Signed-off-by: Jiri Kosina \u003cjkosina@suse.cz\u003e\n\ncommit 5cf896fb6be3effd9aea455b22213e27be8bdb1d\nAuthor: Peter Collingbourne \u003cpcc@google.com\u003e\nDate: Wed Jul 31 18:18:42 2019 -0700\n\n arm64: Add support for relocating the kernel with RELR relocations\n \n RELR is a relocation packing format for relative relocations.\n The format is described in a generic-abi proposal:\n https://groups.google.com/d/topic/generic-abi/bX460iggiKg/discussion\n \n The LLD linker can be instructed to pack relocations in the RELR\n format by passing the flag --pack-dyn-relocs\u003drelr.\n \n This patch adds a new config option, CONFIG_RELR. Enabling this option\n instructs the linker to pack vmlinux\u0027s relative relocations in the RELR\n format, and causes the kernel to apply the relocations at startup along\n with the RELA relocations. RELA relocations still need to be applied\n because the linker will emit RELA relative relocations if they are\n unrepresentable in the RELR format (i.e. address not a multiple of 2).\n \n Enabling CONFIG_RELR reduces the size of a defconfig kernel image\n with CONFIG_RANDOMIZE_BASE by 3.5MB/16% uncompressed, or 550KB/5%\n compressed (lz4).\n \n Signed-off-by: Peter Collingbourne \u003cpcc@google.com\u003e\n Tested-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Reviewed-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 66cbdf5d0c96f5fe570b548e764583ea9d793077\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Jul 31 15:35:20 2019 +0200\n\n arm64: Move TIF_* documentation to individual definitions\n \n Some TIF_* flags are documented in the comment block at the top, some\n next to their definitions, some in both places.\n \n Move all documentation to the individual definitions for consistency,\n and for easy lookup.\n \n Acked-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 13776f9d40a028a245bb766269e360f5b7a62721\nAuthor: Junhua Huang \u003chuang.junhua@zte.com.cn\u003e\nDate: Sat Jul 6 14:41:15 2019 +0800\n\n arm64: mm: free the initrd reserved memblock in a aligned manner\n \n We should free the initrd reserved memblock in an aligned manner,\n because the initrd reserves the memblock in an aligned manner\n in arm64_memblock_init().\n Otherwise there are some fragments in memblock_reserved regions\n after free_initrd_mem(). e.g.:\n /sys/kernel/debug/memblock # cat reserved\n 0: 0x0000000080080000..0x00000000817fafff\n 1: 0x0000000083400000..0x0000000083ffffff\n 2: 0x0000000090000000..0x000000009000407f\n 3: 0x00000000b0000000..0x00000000b000003f\n 4: 0x00000000b26184ea..0x00000000b2618fff\n The fragments like the ranges from b0000000 to b000003f and\n from b26184ea to b2618fff should be freed.\n \n And we can do free_reserved_area() after memblock_free(),\n as free_reserved_area() calls __free_pages(), once we\u0027ve done\n that it could be allocated somewhere else,\n but memblock and iomem still say this is reserved memory.\n \n Fixes: 05c58752f9dc (\"arm64: To remove initrd reserved area entry from memblock\")\n Signed-off-by: Junhua Huang \u003chuang.junhua@zte.com.cn\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 22ec71615d824f4f11d38d0e55a88d8956b7e45f\nAuthor: Will Deacon \u003cwill.deacon@arm.com\u003e\nDate: Fri Jun 7 15:48:58 2019 +0100\n\n arm64: io: Relax implicit barriers in default I/O accessors\n \n The arm64 implementation of the default I/O accessors requires barrier\n instructions to satisfy the memory ordering requirements documented in\n memory-barriers.txt [1], which are largely derived from the behaviour of\n I/O accesses on x86.\n \n Of particular interest are the requirements that a write to a device\n must be ordered against prior writes to memory, and a read from a device\n must be ordered against subsequent reads from memory. We satisfy these\n requirements using various flavours of DSB: the most expensive barrier\n we have, since it implies completion of prior accesses. This was deemed\n necessary when we first implemented the accessors, since accesses to\n different endpoints could propagate independently and therefore the only\n way to enforce order is to rely on completion guarantees [2].\n \n Since then, the Armv8 memory model has been retrospectively strengthened\n to require \"other-multi-copy atomicity\", a property that requires memory\n accesses from an observer to become visible to all other observers\n simultaneously [3]. In other words, propagation of accesses is limited\n to transitioning from locally observed to globally observed. It recently\n became apparent that this change also has a subtle impact on our I/O\n accessors for shared peripherals, allowing us to use the cheaper DMB\n instruction instead.\n \n As a concrete example, consider the following:\n \n memcpy(dma_buffer, data, bufsz);\n writel(DMA_START, dev-\u003ectrl_reg);\n \n A DMB ST instruction between the final write to the DMA buffer and the\n write to the control register will ensure that the writes to the DMA\n buffer are observed before the write to the control register by all\n observers. Put another way, if an observer can see the write to the\n control register, it can also see the writes to memory. This has always\n been the case and is not sufficient to provide the ordering required by\n Linux, since there is no guarantee that the master interface of the\n DMA-capable device has observed either of the accesses. However, in an\n other-multi-copy atomic world, we can infer two things:\n \n 1. A write arriving at an endpoint shared between multiple CPUs is\n visible to all CPUs\n \n 2. A write that is visible to all CPUs is also visible to all other\n observers in the shareability domain\n \n Pieced together, this allows us to use DMB OSHST for our default I/O\n write accessors and DMB OSHLD for our default I/O read accessors (the\n outer-shareability is for handling non-cacheable mappings) for shared\n devices. Memory-mapped, DMA-capable peripherals that are private to a\n CPU (i.e. inaccessible to other CPUs) still require the DSB, however\n these are few and far between and typically require special treatment\n anyway which is outside of the scope of the portable driver API (e.g.\n GIC, page-table walker, SPE profiler).\n \n Note that our mandatory barriers remain as DSBs, since there are cases\n where they are used to flush the store buffer of the CPU, e.g. when\n publishing page table updates to the SMMU.\n \n [1] https://git.kernel.org/linus/4614bbdee357\n [2] https://www.youtube.com/watch?v\u003di6DayghhA8Q\n [3] https://www.cl.cam.ac.uk/~pes20/armv8-mca/\n \n Reviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\n\ncommit d55b7fdd58ac12e76ef65979af4a13b9c15fc00d\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Jul 31 15:29:17 2019 +0200\n\n pinctrl: xway: Use devm_kasprintf() instead of fixed buffer formatting\n \n Improve readability and maintainability by replacing a hardcoded string\n allocation and formatting by the use of the devm_kasprintf() helper.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Link: https://lore.kernel.org/r/20190731132917.17607-4-geert+renesas@glider.be\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 811604d01e1a98a5cfdebde211f20368d151fcd4\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Jul 31 15:29:16 2019 +0200\n\n pinctrl: lantiq: Use kasprintf() instead of fixed buffer formatting\n \n Improve readability and maintainability by replacing a hardcoded string\n allocation and formatting by the use of the kasprintf() helper.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Link: https://lore.kernel.org/r/20190731132917.17607-3-geert+renesas@glider.be\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit f0b0e923e07aa3efd3d5db0bd6ff421bd9a9a0d4\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Jul 31 15:29:15 2019 +0200\n\n pinctrl: devicetree: Use strlen() instead of hardcoded number\n \n Improve readability by replacing a hardcoded number requiring a comment\n by strlen().\n \n Gcc is smart enough to evaluate the length of a constant string at\n compile-time.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Link: https://lore.kernel.org/r/20190731132917.17607-2-geert+renesas@glider.be\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 15bddb7d787aebc837912cecf107b7eaccf03913\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:15 2019 -0700\n\n gpio: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: linux-gpio@vger.kernel.org\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190730181557.90391-16-swboyd@chromium.org\n Acked-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 64c4dcbfcc4ea3ae8f898e6dac836e57acb16b6b\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:33 2019 -0700\n\n pinctrl: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Cc: linux-gpio@vger.kernel.org\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190730181557.90391-34-swboyd@chromium.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 31b4c4b12435911bad5960e88e0675713ae404fb\nAuthor: Kunihiko Hayashi \u003chayashi.kunihiko@socionext.com\u003e\nDate: Tue Jul 30 14:43:30 2019 +0900\n\n pinctrl: uniphier: Fix Pro5 SD pin-mux setting\n \n SD uses the following pins starting from 247:\n SDCD, SDWP, SDVOLC, SDCLK, SDCMD, SDDAT{0,1,2,3}\n \n Signed-off-by: Kunihiko Hayashi \u003chayashi.kunihiko@socionext.com\u003e\n Acked-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Link: https://lore.kernel.org/r/1564465410-9165-6-git-send-email-hayashi.kunihiko@socionext.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 490bbdcfc4d253410e50d1f83fd91f642402c3e1\nAuthor: Kunihiko Hayashi \u003chayashi.kunihiko@socionext.com\u003e\nDate: Tue Jul 30 14:43:29 2019 +0900\n\n pinctrl: uniphier: Add Pro5 PCIe pin-mux settings\n \n Pro5 PCIe interface uses the following pins:\n XPERST, XPEWAKE, XPECLKRQ\n \n Signed-off-by: Kunihiko Hayashi \u003chayashi.kunihiko@socionext.com\u003e\n Link: https://lore.kernel.org/r/1564465410-9165-5-git-send-email-hayashi.kunihiko@socionext.com\n Acked-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit fc1d910ae9dff82a92156c1897646976e785e5f6\nAuthor: Kunihiko Hayashi \u003chayashi.kunihiko@socionext.com\u003e\nDate: Tue Jul 30 14:43:28 2019 +0900\n\n pinctrl: uniphier: Add 5th LD20 MPEG2-TS input pin-mux setting\n \n The 5th serial TS interface uses the following pins:\n hscin4_s: PCA[11-14]\n \n Signed-off-by: Kunihiko Hayashi \u003chayashi.kunihiko@socionext.com\u003e\n Acked-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Link: https://lore.kernel.org/r/1564465410-9165-4-git-send-email-hayashi.kunihiko@socionext.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit f686867266c4d65b49036dbd85110a1300e2d42e\nAuthor: Kunihiko Hayashi \u003chayashi.kunihiko@socionext.com\u003e\nDate: Tue Jul 30 14:43:27 2019 +0900\n\n pinctrl: uniphier: Add another audio I/O pin-mux settings for LD20\n \n This adds support for pinmux settings of aout1b group. This group includes\n audio I/O signals derived from xirq pins, and it is equivalent to \"aout1\"\n in functionality.\n \n Signed-off-by: Kunihiko Hayashi \u003chayashi.kunihiko@socionext.com\u003e\n Link: https://lore.kernel.org/r/1564465410-9165-3-git-send-email-hayashi.kunihiko@socionext.com\n Acked-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 111a8fcb2c2da8256b17a589221c0ad0b5b3cd50\nAuthor: Kunihiko Hayashi \u003chayashi.kunihiko@socionext.com\u003e\nDate: Tue Jul 30 14:43:26 2019 +0900\n\n pinctrl: uniphier: Separate modem group from UART ctsrts group\n \n It depends on the board implementation whether to have each pins of\n CTS/RTS, and others for modem. So it is necessary to divide current\n uart_ctsrts group into uart_ctsrts and uart_modem groups.\n \n Since the number of implemented pins for modem differs depending\n on SoC, each uart_modem group also has a different number of pins.\n \n Signed-off-by: Kunihiko Hayashi \u003chayashi.kunihiko@socionext.com\u003e\n Link: https://lore.kernel.org/r/1564465410-9165-2-git-send-email-hayashi.kunihiko@socionext.com\n Acked-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 726e8d813771ed9c714d75c9ce58a97b9ddf343d\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Mon Jul 29 14:58:38 2019 +0200\n\n pinctrl: meson-g12a: add pwm_a on GPIOE_2 pinmux\n \n Add the missing pinmux for the pwm_a function on the GPIOE_2 pin.\n \n Reviewed-by: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://lore.kernel.org/r/20190729125838.6498-1-narmstrong@baylibre.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 87e1ee238ef24bb1e43f9ac6e9226087004c0f96\nAuthor: Andrew Jeffery \u003candrew@aj.id.au\u003e\nDate: Wed Jul 24 17:43:13 2019 +0930\n\n dt-bindings: aspeed: Remove mention of deprecated compatibles\n \n Guide readers away from using the aspeed,g[45].* compatible patterns.\n \n Signed-off-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Link: https://lore.kernel.org/r/20190724081313.12934-4-andrew@aj.id.au\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 73c732c5714955514bb237f01b14e9e5aa7db47e\nAuthor: Andrew Jeffery \u003candrew@aj.id.au\u003e\nDate: Wed Jul 24 17:43:12 2019 +0930\n\n pinctrl: aspeed: Document existence of deprecated compatibles\n \n Otherwise they look odd in the face of not being listed in the bindings\n documents.\n \n Signed-off-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Link: https://lore.kernel.org/r/20190724081313.12934-3-andrew@aj.id.au\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit a24810673638d5da0336ccae5407c3fd59da14ac\nAuthor: Kevin Brodsky \u003ckevin.brodsky@arm.com\u003e\nDate: Thu Jul 25 18:12:14 2019 +0100\n\n arm64: dts: fast models: Remove clcd\u0027s max-memory-bandwidth\n \n It is unclear why max-memory-bandwidth should be set for CLCD on the\n fast model. Removing that property allows allocating and using 32bpp\n buffers, which may be desirable on certain platforms such as\n Android.\n \n Reported-by: Ruben Ayrapetyan \u003cruben.ayrapetyan@arm.com\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Kevin Brodsky \u003ckevin.brodsky@arm.com\u003e\n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit 2eda1cdec49f8ae7878e60d1b06bd8157a95424f\nAuthor: Andrew Jeffery \u003candrew@aj.id.au\u003e\nDate: Thu Jul 11 13:49:42 2019 +0930\n\n pinctrl: aspeed: Add AST2600 pinmux support\n \n The AST2600 pinmux is fairly similar to the previous generations of\n ASPEED BMC SoCs in terms of architecture, though differ in some of the\n design details. The complexity of the pin expressions is largely reduced\n (e.g. there are no-longer signals with multiple expressions muxing them\n to the associated pin), and there are now signals and buses with\n multiple pin groups.\n \n The driver implements pinmux support for all 244 GPIO-capable pins plus\n a further four pins that are not GPIO capable but which expose multiple\n signals. pinconf will be implemented in a follow-up patch.\n \n The implementation has been smoke-tested under qemu, and run on hardware\n by ASPEED.\n \n Debugged-by: Johnny Huang \u003cjohnny_huang@aspeedtech.com\u003e\n Signed-off-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Link: https://lore.kernel.org/r/20190711041942.23202-7-andrew@aj.id.au\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 86392fac9a9c92f36c0a422a3075865a0fe959f9\nAuthor: Andrew Jeffery \u003candrew@aj.id.au\u003e\nDate: Mon Jul 29 15:26:03 2019 +0930\n\n pinctrl: aspeed: Add SIG_DESC_CLEAR() helper\n \n The complement of SIG_DESC_SET().\n \n Signed-off-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Link: https://lore.kernel.org/r/20190729055604.13239-6-andrew@aj.id.au\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit e7a96b0b7d1669e95aa77c007e71e3c88c531cc4\nAuthor: Andrew Jeffery \u003candrew@aj.id.au\u003e\nDate: Mon Jul 29 15:26:02 2019 +0930\n\n pinctrl: aspeed: Add multiple pin group support for functions\n \n The AST2400 and AST2500 SoCs only exposed one pin group per function.\n Lone pin groups drove some implementation simplifications in the ASPEED\n pinmux infrastructure that is now invalid for the AST2600, which\n supports multiple groups per function for some functions on the chip\n (SMBus Alert pins and UARTs among others).\n \n This patch reworks the macro jungle to enable support for multiple pin\n groups. In the process we inflict some collateral damage on the existing\n AST2400 and AST2500 drivers, but the rework is mostly a relatively\n straight-forward, automated transform of adding the pin name as an\n argument to some macro calls and implementing wrappers to paper over\n groups in the cases where there aren\u0027t multiple.\n \n As previously documented, the macro infrastructure exposes mux\n configuration as symbols in the source file which are used to detect\n accidental duplication. Previously these symbols were named in terms of\n the signal for a given expression. As the AST2600 supports multiple pin\n groups for a function, the signal name on its own is no-longer unique,\n and we must switch to the (signal, group) tuple. However, this means\n that we can no-longer derive the signal expression symbol name from the\n signal name alone, which among other cases, impacts the operation of the\n PIN_DECL_x() macros.\n \n To fix that and avoid requiring we awkwardly provide the associated\n group name for every signal for every PIN_DECL_x() invocation, instead\n opportunistically alias the name of the signal expression symbol from\n the unique (signal, group) tuple to the also unique (pin, signal) tuple,\n then reference the alias symbol in the tables generated by PIN_DECL_x().\n This way we do not require extra group parameters for PIN_DECL_x() as\n the pin name was already provided as an argument, and instead simply\n require that the pin name be provided to the expression declaration\n macros in order to generate the alias symbol.\n \n The patch implements the alias strategy and fixes up all the expression\n definition macro calls in the AST2400 and AST2500 drivers to account for\n pin groups. Given the implementation strategy has the property that\n compilation either fails or loudly warns for bad pin descriptions, this\n patch is theoretically tested by successfully compiling both affected\n drivers. For a more practical test I\u0027ve inspected the diff of the\n content of the pinctrl debugfs entries before and after the patch under\n qemu; all pins, functions and groups match.\n \n Signed-off-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Link: https://lore.kernel.org/r/20190729055604.13239-5-andrew@aj.id.au\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 27d1f73670774e3ea3e304f989f0e1176b57cd7c\nAuthor: Andrew Jeffery \u003candrew@aj.id.au\u003e\nDate: Mon Jul 29 15:26:01 2019 +0930\n\n pinctrl: aspeed: Add PIN_DECL_3() helper\n \n This case is common in the AST2600, so add to the collection.\n \n Signed-off-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Link: https://lore.kernel.org/r/20190729055604.13239-4-andrew@aj.id.au\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 7b38897081666532b434e067bdfa85789f42fa4e\nAuthor: Andrew Jeffery \u003candrew@aj.id.au\u003e\nDate: Mon Jul 29 15:26:00 2019 +0930\n\n pinctrl: aspeed: Rename pin declaration macros\n \n Rename macros as follows:\n \n * s/SS_PIN_DECL()/PIN_DECL_1()/\n * s/MS_PIN_DECL()/PIN_DECL_2()/\n * s/MS_PIN_DECL_()/PIN_DECL_()/\n \n This is in preparation for adding PIN_DECL_3(). We could clean this up\n with e.g. CPPMAGIC_MAP() from ccan, but that might be a bridge too far\n given how much of a macro jungle we already have.\n \n Signed-off-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Link: https://lore.kernel.org/r/20190729055604.13239-3-andrew@aj.id.au\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 7488838f23157092038347dd67b1047657a5c4f0\nAuthor: Andrew Jeffery \u003candrew@aj.id.au\u003e\nDate: Mon Jul 29 15:25:59 2019 +0930\n\n dt-bindings: pinctrl: aspeed: Document AST2600 pinmux\n \n The AST260 differs from the 2400 and 2500 in that it supports multiple\n groups for a subset of functions.\n \n Signed-off-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Link: https://lore.kernel.org/r/20190729055604.13239-2-andrew@aj.id.au\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit fb18f1887fddd55e9d9003456136f44d0e2d3fac\nAuthor: Icenowy Zheng \u003cicenowy@aosc.io\u003e\nDate: Sun Jul 28 11:12:22 2019 +0800\n\n pinctrl: sunxi: v3s: introduce support for V3\n \n Introduce the GPIO pins that is only available on V3 (not on V3s) to the\n V3s pinctrl driver.\n \n Signed-off-by: Icenowy Zheng \u003cicenowy@aosc.io\u003e\n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n \n Link: https://lore.kernel.org/r/20190728031227.49140-2-icenowy@aosc.io\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 6161dc03587b8057b97c61aa3a2f8b16562d8991\nAuthor: Anders Roxell \u003canders.roxell@linaro.org\u003e\nDate: Fri Jul 26 13:28:16 2019 +0200\n\n pinctrl: qcom: spmi-gpio: Mark expected switch fall-through\n \n When fall-through warnings was enabled by default the following warnings\n was starting to show up:\n \n ../drivers/pinctrl/qcom/pinctrl-spmi-gpio.c: In function ‘pmic_gpio_populate’:\n ../drivers/pinctrl/qcom/pinctrl-spmi-gpio.c:815:20: warning: this statement may fall\n through [-Wimplicit-fallthrough\u003d]\n pad-\u003ehave_buffer \u003d true;\n ~~~~~~~~~~~~~~~~~^~~~~~\n ../drivers/pinctrl/qcom/pinctrl-spmi-gpio.c:816:2: note: here\n case PMIC_GPIO_SUBTYPE_GPIOC_4CH:\n ^~~~\n ../drivers/pinctrl/qcom/pinctrl-spmi-gpio.c:820:20: warning: this statement may fall\n through [-Wimplicit-fallthrough\u003d]\n pad-\u003ehave_buffer \u003d true;\n ~~~~~~~~~~~~~~~~~^~~~~~\n ../drivers/pinctrl/qcom/pinctrl-spmi-gpio.c:821:2: note: here\n case PMIC_GPIO_SUBTYPE_GPIOC_8CH:\n ^~~~\n \n Rework so that the compiler doesn\u0027t warn about fall-through.\n \n Fixes: d93512ef0f0e (\"Makefile: Globally enable fall-through warning\")\n Signed-off-by: Anders Roxell \u003canders.roxell@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190726112816.19723-1-anders.roxell@linaro.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit cd927f140d5199c04270241fb675732275c2d20f\nAuthor: Anders Roxell \u003canders.roxell@linaro.org\u003e\nDate: Fri Jul 26 13:28:12 2019 +0200\n\n pinctrl: rockchip: Mark expected switch fall-through\n \n When fall-through warnings was enabled by default the following warning\n was starting to show up:\n \n ../drivers/pinctrl/pinctrl-rockchip.c: In function ‘rockchip_gpio_set_config’:\n ../drivers/pinctrl/pinctrl-rockchip.c:2783:3: warning: this statement may fall\n through [-Wimplicit-fallthrough\u003d]\n rockchip_gpio_set_debounce(gc, offset, true);\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n ../drivers/pinctrl/pinctrl-rockchip.c:2795:2: note: here\n default:\n ^~~~~~~\n \n Rework so that the compiler doesn\u0027t warn about fall-through. Add\n \u0027return -ENOTSUPP;\u0027 to match the comment.\n \n Fixes: d93512ef0f0e (\"Makefile: Globally enable fall-through warning\")\n Signed-off-by: Anders Roxell \u003canders.roxell@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190726112812.19665-1-anders.roxell@linaro.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit be65c8bb0c64c552b159d509329df4e55bd18e44\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Jul 25 22:24:19 2019 +0800\n\n pinctrl: oxnas: remove set but not used variable \u0027arg\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/pinctrl/pinctrl-oxnas.c: In function oxnas_ox810se_pinconf_set:\n drivers/pinctrl/pinctrl-oxnas.c:905:6: warning: variable arg set but not used [-Wunused-but-set-variable]\n drivers/pinctrl/pinctrl-oxnas.c: In function oxnas_ox820_pinconf_set:\n drivers/pinctrl/pinctrl-oxnas.c:944:6: warning: variable arg set but not used [-Wunused-but-set-variable]\n \n It is never used since commit 4b0c0c25fa79 (\"pinctrl:\n oxnas: Add support for OX820\"), so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190725142419.29892-1-yuehaibing@huawei.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit a502b343ebd0eab38f3cb33fbb84011847cf5aac\nAuthor: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\nDate: Thu Jul 25 13:16:56 2019 +0200\n\n pinctrl: stmfx: update pinconf settings\n \n According to the following tab (coming from STMFX datasheet), updates\n have to done in stmfx_pinconf_set function:\n \n -\"type\" has to be set when \"bias\" is configured as \"pull-up or pull-down\"\n -PIN_CONFIG_DRIVE_PUSH_PULL should only be used when gpio is configured as\n output. There is so no need to check direction.\n \n DIR | TYPE | PUPD | MFX GPIO configuration\n ----|------|------|---------------------------------------------------\n 1 | 1 | 1 | OUTPUT open drain with internal pull-up resistor\n ----|------|------|---------------------------------------------------\n 1 | 1 | 0 | OUTPUT open drain with internal pull-down resistor\n ----|------|------|---------------------------------------------------\n 1 | 0 | 0/1 | OUTPUT push pull no pull\n ----|------|------|---------------------------------------------------\n 0 | 1 | 1 | INPUT with internal pull-up resistor\n ----|------|------|---------------------------------------------------\n 0 | 1 | 0 | INPUT with internal pull-down resistor\n ----|------|------|---------------------------------------------------\n 0 | 0 | 1 | INPUT floating\n ----|------|------|---------------------------------------------------\n 0 | 0 | 0 | analog (GPIO not used, default setting)\n \n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n Signed-off-by: Amelie Delaunay \u003camelie.delaunay@st.com\u003e\n Link: https://lore.kernel.org/r/1564053416-32192-1-git-send-email-amelie.delaunay@st.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 2f22e2028c03d95c47b077887297dfa068da2c8a\nAuthor: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\nDate: Thu Jul 25 17:56:31 2019 +0800\n\n pinctrl: sprd: Combine the condition of MISC_PIN and COMMON_PIN\n \n Since the follow-up pin design on Spreadtrum platform has some changes,\n some configuration of MISC_PIN moved to COMMON_PIN. To support current\n pin design and keep backward compatibility, we should combine the\n condition of MISC_PIN and COMMON_PIN to configure an individual pin.\n \n Signed-off-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Link: https://lore.kernel.org/r/17af5e761e0515d288a7ea4078ac9aa4a82a7a4e.1564048446.git.baolin.wang@linaro.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit e89febc75df909fadd43b76754778109a05383c0\nAuthor: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\nDate: Thu Jul 25 17:56:30 2019 +0800\n\n pinctrl: sprd: Change to use devm_platform_ioremap_resource()\n \n The devm_platform_ioremap_resource() function wraps platform_get_resource()\n and devm_ioremap_resource() in a single helper, thus use it to simplify\n the code.\n \n Signed-off-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Link: https://lore.kernel.org/r/ff410d312ed0047b5a36e5113daf7df78bcf1aa8.1564048446.git.baolin.wang@linaro.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 2f8f180b3ceed7a16a92cc3c164368c26e1f9320\nAuthor: Mark Brown \u003cbroonie@kernel.org\u003e\nDate: Tue Jul 30 11:26:31 2019 +0100\n\n arm64: Remove unused cpucap_multi_entry_cap_cpu_enable()\n \n The function cpucap_multi_entry_cap_cpu_enable() is unused, remove it to\n avoid any confusion reading the code and potential for bit rot.\n \n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 73961dc1182ed7e1eb80a90ee28871ab3547af7e\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Mon Jul 15 14:28:17 2019 +0100\n\n arm64: sysreg: Remove unused and rotting SCTLR_ELx field definitions\n \n Our SCTLR_ELx field definitions are somewhat over-engineered in that\n they carefully define masks describing the RES0/RES1 bits and then use\n these to construct further masks representing bits to be set/cleared for\n the _EL1 and _EL2 registers.\n \n However, most of the resulting definitions aren\u0027t actually used by\n anybody and have subsequently started to bit-rot when new fields have\n been added by the architecture, resulting in fields being part of the\n RES0 mask despite being defined and used elsewhere.\n \n Rather than fix up these masks, simply remove the unused parts entirely\n so that we can drop the maintenance burden. We can always add things\n back if we need them in the future.\n \n Acked-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 332e5281a4e8269b96233a7babc98b03596b7e6d\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Jul 16 08:14:19 2019 +0100\n\n arm64: esr: Add ESR exception class encoding for trapped ERET\n \n The ESR.EC encoding of 0b011010 (0x1a) describes an exception generated\n by an ERET, ERETAA or ERETAB instruction as a result of a nested\n virtualisation trap to EL2.\n \n Add an encoding for this EC and a string description so that we identify\n it correctly if we take one unexpectedly.\n \n Acked-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Acked-by: Marc Zyngier \u003cmaz@kernel.org\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit b3e089cd446b26bb1e12860e1afb9da314453fd6\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 30 10:44:15 2019 +0800\n\n arm64: Replace strncmp with str_has_prefix\n \n In commit b6b2735514bc\n (\"tracing: Use str_has_prefix() instead of using fixed sizes\")\n the newly introduced str_has_prefix() was used\n to replace error-prone strncmp(str, const, len).\n Here fix codes with the same pattern.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 3e77eeb7a27fc3dcf6b65e7ee01ac00bf5d2b4fb\nAuthor: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\nDate: Mon Jul 22 17:14:33 2019 +0100\n\n ACPI/IORT: Rename arm_smmu_v3_set_proximity() \u0027node\u0027 local variable\n \n Commit 36a2ba07757d (\"ACPI/IORT: Reject platform device creation on NUMA\n node mapping failure\") introduced a local variable \u0027node\u0027 in\n arm_smmu_v3_set_proximity() that shadows the struct acpi_iort_node\n pointer function parameter.\n \n Execution was unaffected but it is prone to errors and can lead\n to subtle bugs.\n \n Rename the local variable to prevent any issue.\n \n Reviewed-by: Hanjun Guo \u003cguohanjun@huawei.com\u003e\n Reported-by: Will Deacon \u003cwill@kernel.org\u003e\n Signed-off-by: Lorenzo Pieralisi \u003clorenzo.pieralisi@arm.com\u003e\n Cc: Will Deacon \u003cwill@kernel.org\u003e\n Cc: Hanjun Guo \u003cguohanjun@huawei.com\u003e\n Cc: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Cc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Cc: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Cc: Kefeng Wang \u003cwangkefeng.wang@huawei.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit b717480f5415201286cc9b5e3244513f54a327c4\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Tue Jul 23 20:29:22 2019 +0900\n\n arm64: remove unneeded uapi/asm/stat.h\n \n stat.h is listed in include/uapi/asm-generic/Kbuild, so Kbuild will\n automatically generate it.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit c19d050f80881296aab3ba90fe5b2c107a238dcb\nAuthor: Bhupesh Sharma \u003cbhsharma@redhat.com\u003e\nDate: Thu Jul 11 17:27:32 2019 +0530\n\n arm64/kexec: Use consistent convention of initializing \u0027kxec_buf.mem\u0027 with KEXEC_BUF_MEM_UNKNOWN\n \n With commit b6664ba42f14 (\"s390, kexec_file: drop arch_kexec_mem_walk()\"),\n we introduced the KEXEC_BUF_MEM_UNKNOWN macro. If kexec_buf.mem is set\n to this value, kexec_locate_mem_hole() will try to allocate free memory.\n \n While other arch(s) like s390 and x86_64 already use this macro to\n initialize kexec_buf.mem with, arm64 uses an equivalent value of 0.\n Replace it with KEXEC_BUF_MEM_UNKNOWN, to keep the convention of\n initializing \u0027kxec_buf.mem\u0027 consistent across various archs.\n \n Cc: takahiro.akashi@linaro.org\n Cc: james.morse@arm.com\n Reviewed-by: Matthias Brugger \u003cmbrugger@suse.com\u003e\n Signed-off-by: Bhupesh Sharma \u003cbhsharma@redhat.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit b907b80d7ae7b2b65ef9f534f3e9a32ce6a4b539\nAuthor: Mark Rutland \u003cmark.rutland@arm.com\u003e\nDate: Mon Jul 8 17:36:40 2019 +0100\n\n arm64: remove pointless __KERNEL__ guards\n \n For a number of years, UAPI headers have been split from kernel-internal\n headers. The latter are never exposed to userspace, and always built\n with __KERNEL__ defined.\n \n Most headers under arch/arm64 don\u0027t have __KERNEL__ guards, but there\n are a few stragglers lying around. To make things more consistent, and\n to set a good example going forward, let\u0027s remove these redundant\n __KERNEL__ guards.\n \n In a couple of cases, a trailing #endif lacked a comment describing its\n corresponding #if or #ifdef, so these are fixes up at the same time.\n \n Guards in auto-generated crypto code are left as-is, as these guards are\n generated by scripting imported from the upstream openssl project\n scripts. Guards in UAPI headers are left as-is, as these can be included\n by userspace or the kernel.\n \n There should be no functional change as a result of this patch.\n \n Signed-off-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Cc: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Cc: Will Deacon \u003cwill.deacon@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit c87857945b0e61c46222798b56fb1b0f4868088b\nAuthor: Julien Thierry \u003cjulien.thierry@arm.com\u003e\nDate: Thu Jul 4 14:44:04 2019 +0100\n\n arm64: Remove unused assembly macro\n \n As of commit 4141c857fd09dbed480f021b3eece4f46c653161 (\"arm64: convert\n raw syscall invocation to C\"), moving syscall handling from assembly to\n C, the macro mask_nospec64 is no longer referenced.\n \n Acked-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Julien Thierry \u003cjulien.thierry@arm.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 2c8db5bef9fb442a5a0d2fee28aed20100362ce3\nAuthor: Kalesh Singh \u003ckaleshsingh@google.com\u003e\nDate: Wed Jul 31 14:29:33 2019 -0700\n\n PM/sleep: Expose suspend stats in sysfs\n \n Userspace can get suspend stats from the suspend stats debugfs node.\n Since debugfs doesn\u0027t have stable ABI, expose suspend stats in\n sysfs under /sys/power/suspend_stats.\n \n Signed-off-by: Kalesh Singh \u003ckaleshsingh@google.com\u003e\n Reviewed-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit db5b44f82a6d29049e38853c40dcc128213f16c5\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 22:27:38 2019 +0300\n\n pinctrl: qdf2xxx: Switch to use device_property_count_uXX()\n \n Use use device_property_count_uXX() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190723192738.68486-2-andriy.shevchenko@linux.intel.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 720b8ec676cd2b1d30289c977b5fde08840cb6dd\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 22:27:37 2019 +0300\n\n pinctrl: msm: Switch to use device_property_count_uXX()\n \n Use use device_property_count_uXX() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190723192738.68486-1-andriy.shevchenko@linux.intel.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit bb100b64763c5f93a4d5e97d3a1abedadfaaf384\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Wed Jul 17 13:55:43 2019 +0300\n\n ACPI / APEI: Get rid of NULL_UUID_LE constant\n \n This is a missed part of the commit 5b53696a30d5\n (\"ACPI / APEI: Switch to use new generic UUID API\"), i.e.\n replacing old definition with a global constant variable.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: James Morse \u003cjames.morse@arm.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit a1db8da7c28131ceb3200732e429eb9d30ce5929\nAuthor: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\nDate: Mon Jul 22 10:10:15 2019 +0100\n\n pinctrl: madera: Add configuration for Cirrus Logic CS47L92\n \n This adds the pinctrl configuration for the CS47L92 codec to\n the madera pinctrl driver.\n \n Signed-off-by: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n Link: https://lore.kernel.org/r/20190722091015.20884-3-ckeepax@opensource.cirrus.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit b0bca3e4ee934a3f06e6ddb4337c679f54575009\nAuthor: Richard Fitzgerald \u003crf@opensource.cirrus.com\u003e\nDate: Mon Jul 22 10:10:14 2019 +0100\n\n pinctrl: madera: Add configuration for Cirrus Logic CS47L15\n \n This adds the pinctrl configuration for the CS47L15 codec to\n the madera pinctrl driver.\n \n Signed-off-by: Richard Fitzgerald \u003crf@opensource.cirrus.com\u003e\n Signed-off-by: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n Link: https://lore.kernel.org/r/20190722091015.20884-2-ckeepax@opensource.cirrus.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit f134b8517736bddc52955fef08e70bde7d408c0b\nAuthor: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\nDate: Mon Jul 22 10:10:13 2019 +0100\n\n pinctrl: madera: Use local copy of pdata\n \n A local copy of the pdata exists and it should be used rather than\n pulling a fresh copy.\n \n Signed-off-by: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n Link: https://lore.kernel.org/r/20190722091015.20884-1-ckeepax@opensource.cirrus.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 74d2d0e68701bcd53d2cf771dd3b3cb9f84bed5c\nAuthor: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\nDate: Mon Jul 22 10:07:48 2019 +0100\n\n gpio: madera: Add support for Cirrus Logic CS47L92\n \n As the gpio is common to all madera codecs all that is needed\n is to setup the correct number of GPIO pins for the CS47L92.\n \n Signed-off-by: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n Link: https://lore.kernel.org/r/20190722090748.20807-4-ckeepax@opensource.cirrus.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit d06be8bc290aa255b9fd8602e60fb9f487aa0f48\nAuthor: Richard Fitzgerald \u003crf@opensource.cirrus.com\u003e\nDate: Mon Jul 22 10:07:47 2019 +0100\n\n gpio: madera: Add support for Cirrus Logic CS47L15\n \n As the gpio is common to all madera codecs all that is needed\n is to setup the correct number of GPIO pins for the CS47L15.\n \n Signed-off-by: Richard Fitzgerald \u003crf@opensource.cirrus.com\u003e\n Signed-off-by: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n Link: https://lore.kernel.org/r/20190722090748.20807-3-ckeepax@opensource.cirrus.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit fe2afa0d7b5ce05f8d2c4508e4041f3a216a5539\nAuthor: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\nDate: Mon Jul 22 10:07:46 2019 +0100\n\n gpio: madera: Use local copy of pdata\n \n A local copy of the pdata exists and it should be used rather than\n pulling a fresh copy.\n \n Signed-off-by: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n Link: https://lore.kernel.org/r/20190722090748.20807-2-ckeepax@opensource.cirrus.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 72ce665f170ee22069726840cd13ccf5af4ef34d\nAuthor: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\nDate: Mon Jul 22 10:07:45 2019 +0100\n\n gpio: arizona: Use local copy of pdata\n \n A local copy of the pdata exists and it should be used rather than\n pulling a fresh copy.\n \n Signed-off-by: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n Link: https://lore.kernel.org/r/20190722090748.20807-1-ckeepax@opensource.cirrus.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit e38a9a437fb93ddafab5030165e4c6a3a5021669\nAuthor: Stefan Wahren \u003cwahrenst@gmx.net\u003e\nDate: Mon Jul 22 08:23:25 2019 +0200\n\n pinctrl: bcm2835: Add support for BCM2711 pull-up functionality\n \n The BCM2711 has a new way of selecting the pull-up/pull-down setting\n for a GPIO pin. The registers used for the BCM2835, GP_PUD and\n GP_PUDCLKn0, are no longer connected. A new set of registers,\n GP_GPIO_PUP_PDN_CNTRL_REGx must be used. This commit will add\n a new compatible string \"brcm,bcm2711-gpio\" and the kernel\n driver will use it to select which method is used to select\n pull-up/pull-down.\n \n This patch based on a patch by Al Cooper which was intended for the\n BCM7211. This is a bugfixed and improved version.\n \n Signed-off-by: Stefan Wahren \u003cwahrenst@gmx.net\u003e\n Link: https://lore.kernel.org/r/1563776607-8368-3-git-send-email-wahrenst@gmx.net\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 9cd6237d660fa70f6592e28b7c3c8ecb8e76807a\nAuthor: Stefan Wahren \u003cwahrenst@gmx.net\u003e\nDate: Mon Jul 22 08:23:24 2019 +0200\n\n pinctrl: bcm2835: Add brcm,bcm2711 compatible\n \n Add a new compatible for the BCM2711.\n \n Signed-off-by: Stefan Wahren \u003cwahrenst@gmx.net\u003e\n Link: https://lore.kernel.org/r/1563776607-8368-2-git-send-email-wahrenst@gmx.net\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit d111e7a387d71b444b9a785d754f348d9a623204\nAuthor: Hennie Muller \u003chm@bitlabs.co.za\u003e\nDate: Sun Jul 21 14:52:58 2019 +0200\n\n gpio: viperboard: Replace \u0027unsigned\u0027 with \u0027unsigned int\u0027\n \n Fixes a couple of warnings by checkpatch and sparse.\n \n Signed-off-by: Hennie Muller \u003chm@bitlabs.co.za\u003e\n Link: https://lore.kernel.org/r/20190721125259.13990-1-hm@bitlabs.co.za\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 6c172af579cffd0968ef94b9ffa4f9b12cc21dcb\nAuthor: Sowjanya Komatineni \u003cskomatineni@nvidia.com\u003e\nDate: Wed Jul 31 14:10:45 2019 -0700\n\n pinctrl: tegra210: Add Tegra210 pinctrl pm ops\n \n This patch adds suspend and resume functionality to Tegra210 pinctrl.\n \n Signed-off-by: Sowjanya Komatineni \u003cskomatineni@nvidia.com\u003e\n Link: https://lore.kernel.org/r/1564607463-28802-3-git-send-email-skomatineni@nvidia.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 9870acd3747a3535a1d45e915265be85cd92f72d\nAuthor: Sowjanya Komatineni \u003cskomatineni@nvidia.com\u003e\nDate: Wed Jul 31 14:10:44 2019 -0700\n\n pinctrl: tegra: Add suspend and resume support\n \n This patch adds support for Tegra pinctrl driver suspend and resume.\n \n During suspend, context of all pinctrl registers are stored and\n on resume they are all restored to have all the pinmux and pad\n configuration for normal operation.\n \n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Reviewed-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Signed-off-by: Sowjanya Komatineni \u003cskomatineni@nvidia.com\u003e\n Link: https://lore.kernel.org/r/1564607463-28802-2-git-send-email-skomatineni@nvidia.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 3d6ade0abfe1ac833b525349ade156a269979cf0\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sat Jul 20 20:58:58 2019 +0900\n\n pinctrl: pinctrl-single: add header include guard\n \n Add a header include guard just in case.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Link: https://lore.kernel.org/r/20190720115858.7015-1-yamada.masahiro@socionext.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit cab09f3d2d2a0a6cb3dfb678660d67a2c3764f50\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Fri Jul 19 12:12:42 2019 +0200\n\n cpuidle: teo: Allow tick to be stopped if PM QoS is used\n \n The TEO goveror prevents the scheduler tick from being stopped (unless\n stopped already) if there is a PM QoS latency constraint for the given\n CPU and the target residency of the deepest idle state matching that\n constraint is below the tick boundary.\n \n However, that is problematic if CPUs with PM QoS latency constraints\n are idle for long times, because it effectively causes the tick to\n run on them all the time which is wasteful. [It is also confusing\n and questionable if they are full dynticks CPUs.]\n \n To address that issue, modify the TEO governor to carry out the\n entire search for the most suitable idle state (from the target\n residency perspective) even if a latency constraint is present,\n to allow it to determine the expected idle duration in all cases.\n \n Also, when using the last several measured idle duration values\n to refine the idle state selection, make it compare those values\n with the current expected idle duration value (instead of\n comparing them with the target residency of the idle state\n selected so far) which should prevent the tick from being\n retained when it makes sense to stop it sometimes (especially\n in the presence of PM QoS latency constraints).\n \n Fixes: b26bf6ab716f (\"cpuidle: New timer events oriented governor for tickless systems\")\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 32b91ca15353b2803d27cfc747156e72dd2cd5d8\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Thu Jul 18 10:53:21 2019 +0200\n\n cpuidle: menu: Allow tick to be stopped if PM QoS is used\n \n After commit 554c8aa8ecad (\"sched: idle: Select idle state before\n stopping the tick\") the menu governor prevents the scheduler tick from\n being stopped (unless stopped already) if there is a PM QoS latency\n constraint for the given CPU and the target residency of the deepest\n idle state matching that constraint is below the tick boundary.\n \n However, that is problematic if CPUs with PM QoS latency constraints\n are idle for long times, because it effectively causes the tick to\n run on them all the time which is wasteful. [It is also confusing\n and questionable if they are full dynticks CPUs.]\n \n To address that issue, make the menu governor allow the tick to be\n stopped only if the idle duration predicted by it is beyond the tick\n boundary, except when the shallowest idle state is selected upfront\n and it is not a \"polling\" one.\n \n Fixes: 554c8aa8ecad (\"sched: idle: Select idle state before stopping the tick\")\n Link: https://lore.kernel.org/lkml/79b247b3-e056-610e-9a07-e685dfdaa6c9@gmail.com/\n Reported-by: Thomas Lindroth \u003cthomas.lindroth@gmail.com\u003e\n Tested-by: Thomas Lindroth \u003cthomas.lindroth@gmail.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 0d1e16c640ee67dcae65ae8e3c6f58e6d8101047\nAuthor: Ulf Hansson \u003culf.hansson@linaro.org\u003e\nDate: Tue Jul 16 11:43:20 2019 +0200\n\n PM / Domains: Align in-parameter names for some genpd functions\n \n Align in-parameter names for the declarations of pm_genpd_add|\n remove_subdomain() and of_genpd_add_subdomain() according to their\n implementations, as to improve consistency.\n \n Signed-off-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 9616dda1aaddb2080122aaeab96ad7fc064e36b4\nAuthor: Leonardo Bras \u003cleonardo@linux.ibm.com\u003e\nDate: Thu Aug 1 19:52:51 2019 -0300\n\n powerpc/pseries/hotplug-memory.c: Replace nested ifs by switch-case\n \n I noticed these nested ifs can be easily replaced by switch-cases,\n which can improve readability.\n \n Signed-off-by: Leonardo Bras \u003cleonardo@linux.ibm.com\u003e\n Reviewed-by: David Hildenbrand \u003cdavid@redhat.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190801225251.17864-1-leonardo@linux.ibm.com\n\ncommit 1ebe0dcce1750109181d666394b7dfd9af9ff645\nAuthor: Jordan Niethe \u003cjniethe5@gmail.com\u003e\nDate: Fri Aug 2 10:08:35 2019 +1000\n\n powerpc/xive: Update comment referencing magic loads from an ESB\n \n The comment above xive_esb_read() references magic loads from an ESB as\n described xive.h. This has been inaccurate since commit 12c1f339cd49\n (\"powerpc/xive: Move definition of ESB bits\") which moved the\n description. Update the comment to reference the new location of the\n description in xive-regs.h\n \n Signed-off-by: Jordan Niethe \u003cjniethe5@gmail.com\u003e\n Acked-by: Stewart Smith \u003cstewart@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190802000835.26191-1-jniethe5@gmail.com\n\ncommit 461cef2a676e7c578d0ef62969dbcb8237c8631d\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Wed Jul 31 06:31:41 2019 +0000\n\n powerpc/32: activate ARCH_HAS_PMEM_API and ARCH_HAS_UACCESS_FLUSHCACHE\n \n PPC32 also have flush_dcache_range() so it can also support\n ARCH_HAS_PMEM_API and ARCH_HAS_UACCESS_FLUSHCACHE without changes.\n \n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/a682a2f9db308c5cfe77e45aa3352e41bc9f4e33.1564554634.git.christophe.leroy@c-s.fr\n\ncommit e7de4f7b64c23e503a8c42af98d56f2a7462bd6d\nAuthor: Andrew Donnellan \u003cajd@linux.ibm.com\u003e\nDate: Fri May 3 17:52:53 2019 +1000\n\n powerpc/powernv: Restrict OPAL symbol map to only be readable by root\n \n Currently the OPAL symbol map is globally readable, which seems bad as\n it contains physical addresses.\n \n Restrict it to root.\n \n Fixes: c8742f85125d (\"powerpc/powernv: Expose OPAL firmware symbol map\")\n Cc: stable@vger.kernel.org # v3.19+\n Suggested-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Signed-off-by: Andrew Donnellan \u003cajd@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190503075253.22798-1-ajd@linux.ibm.com\n\ncommit 8b856a0942a1b4d832966985fcdf1a455eb6ab8c\nAuthor: Andrew Donnellan \u003cajd@linux.ibm.com\u003e\nDate: Thu May 9 15:11:19 2019 +1000\n\n powerpc/configs: Disable SCOM_DEBUGFS in powernv_defconfig\n \n SCOM_DEBUGFS is really not needed for anything other than low-level\n hardware debugging.\n \n mpe: It also introduces a large and poorly documented/understood\n attack surface. Although the interface is only available to root, the\n kernel still aspires to restrict root to accessing hardware through\n well defined interfaces, which this is not.\n \n opal-prd uses its own interface (/dev/prd) for SCOM access, so it\n doesn\u0027t need SCOM_DEBUGFS.\n \n At some point in the future we\u0027ll introduce a debug config fragment\n where this can go instead.\n \n Signed-off-by: Andrew Donnellan \u003cajd@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190509051119.7694-5-ajd@linux.ibm.com\n\ncommit 8c98db8d5851db45c50024a29c3dd1779dc4da0b\nAuthor: Andrew Donnellan \u003cajd@linux.ibm.com\u003e\nDate: Thu May 9 15:11:18 2019 +1000\n\n powerpc/powernv: Fix checkpatch warnings in opal-xscom.c\n \n Signed-off-by: Andrew Donnellan \u003cajd@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190509051119.7694-4-ajd@linux.ibm.com\n\ncommit bfd2f0d49aef8abfe6bf58f12719f39912993cc6\nAuthor: Andrew Donnellan \u003cajd@linux.ibm.com\u003e\nDate: Thu May 9 15:11:17 2019 +1000\n\n powerpc/powernv: Get rid of old scom_controller abstraction\n \n Once upon a time, the SCOM access code was used by the WSP platform as\n well as powernv. Thus it made sense to have a generic SCOM access\n interface to abstract between different platforms.\n \n Now that it\u0027s just powernv, with no other platforms currently on the\n horizon, let\u0027s rip out scom_controller and make everything much\n simpler and more direct.\n \n While we\u0027re here, fix up the comment block at the top.\n \n Signed-off-by: Andrew Donnellan \u003cajd@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190509051119.7694-3-ajd@linux.ibm.com\n\ncommit 9edc6b919d9fbec694dc37ea14f9587be1b354ec\nAuthor: Andrew Donnellan \u003cajd@linux.ibm.com\u003e\nDate: Thu May 9 15:11:16 2019 +1000\n\n powerpc/powernv: Remove dead SCOM access code\n \n Nothing is using scom_map_device() or scom_find_parent(). Remove them.\n \n Also don\u0027t export scom_controller, there are no other users of it.\n \n Signed-off-by: Andrew Donnellan \u003cajd@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190509051119.7694-2-ajd@linux.ibm.com\n\ncommit 08a456aa643776757e07adfdebe7f7681117d144\nAuthor: Andrew Donnellan \u003cajd@linux.ibm.com\u003e\nDate: Thu May 9 15:11:15 2019 +1000\n\n powerpc/powernv: Move SCOM access code into powernv platform\n \n The powernv platform is the only one that directly accesses SCOMs.\n Move the support code to platforms/powernv, and get rid of the\n PPC_SCOM Kconfig option, as SCOM support is always selected when\n compiling for powernv.\n \n This also means that the Kconfig item for CONFIG_SCOM_DEBUGFS will\n show up in menuconfig in the platform menu, rather than at the root,\n which is a much better location.\n \n Signed-off-by: Andrew Donnellan \u003cajd@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190509051119.7694-1-ajd@linux.ibm.com\n\ncommit 0df3e42167caaf9f8c7b64de3da40a459979afe8\nAuthor: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\nDate: Mon Jun 3 15:11:58 2019 -0700\n\n PCI: rpaphp: Avoid a sometimes-uninitialized warning\n \n When building with -Wsometimes-uninitialized, clang warns:\n \n drivers/pci/hotplug/rpaphp_core.c:243:14: warning: variable \u0027fndit\u0027 is\n used uninitialized whenever \u0027for\u0027 loop exits because its condition is\n false [-Wsometimes-uninitialized]\n for (j \u003d 0; j \u003c entries; j++) {\n ^~~~~~~~~~~\n drivers/pci/hotplug/rpaphp_core.c:256:6: note: uninitialized use occurs\n here\n if (fndit)\n ^~~~~\n drivers/pci/hotplug/rpaphp_core.c:243:14: note: remove the condition if\n it is always true\n for (j \u003d 0; j \u003c entries; j++) {\n ^~~~~~~~~~~\n drivers/pci/hotplug/rpaphp_core.c:233:14: note: initialize the variable\n \u0027fndit\u0027 to silence this warning\n int j, fndit;\n ^\n \u003d 0\n \n fndit is only used to gate a sprintf call, which can be moved into the\n loop to simplify the code and eliminate the local variable, which will\n fix this warning.\n \n Fixes: 2fcf3ae508c2 (\"hotplug/drc-info: Add code to search ibm,drc-info property\")\n Suggested-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Acked-by: Tyrel Datwyler \u003ctyreld@linux.ibm.com\u003e\n Acked-by: Joel Savitz \u003cjsavitz@redhat.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://github.com/ClangBuiltLinux/linux/issues/504\n Link: https://lore.kernel.org/r/20190603221157.58502-1-natechancellor@gmail.com\n\ncommit ae2e953fdca791270e80c08d6a830d9aa472a111\nAuthor: Nathan Lynch \u003cnathanl@linux.ibm.com\u003e\nDate: Thu Jul 18 11:22:14 2019 -0500\n\n powerpc/rtas: Unexport rtas_online_cpus_mask, rtas_offline_cpus_mask\n \n These aren\u0027t used by modular code, nor should they be.\n \n Fixes: 120496ac2d2d (\"powerpc: Bring all threads online prior to migration/hibernation\")\n Signed-off-by: Nathan Lynch \u003cnathanl@linux.ibm.com\u003e\n Signed-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Link: https://lore.kernel.org/r/20190718162214.5694-1-nathanl@linux.ibm.com\n\ncommit ce52ad5dd52cfaf3398058384e0ff94134bbd89c\nAuthor: Stanislav Lisovskiy \u003cstanislav.lisovskiy@intel.com\u003e\nDate: Fri Jul 12 11:19:38 2019 +0300\n\n drm/i915: Fix wrong escape clock divisor init for GLK\n \n According to Bspec clock divisor registers in GeminiLake\n should be initialized by shifting 1(\u003c\u003c) to amount of correspondent\n divisor. While i915 was writing all this time that value as is.\n \n Surprisingly that it by accident worked, until we met some issues\n with Microtech Etab.\n \n v2: Added Fixes tag and cc\n v3: Added stable to cc as well.\n \n Signed-off-by: Stanislav Lisovskiy \u003cstanislav.lisovskiy@intel.com\u003e\n Reviewed-by: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d108826\n Fixes: bcc657004841 (\"drm/i915/glk: Program txesc clock divider for GLK\")\n Cc: Deepak M \u003cm.deepak@intel.com\u003e\n Cc: Madhav Chauhan \u003cmadhav.chauhan@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@linux.intel.com\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: intel-gfx@lists.freedesktop.org\n Cc: stable@vger.kernel.org\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712081938.14185-1-stanislav.lisovskiy@intel.com\n\ncommit aaefca8e30d9df7a4ca13c9c8e135dd227b8ff19\nAuthor: Tony Luck \u003ctony.luck@intel.com\u003e\nDate: Thu Jul 18 11:29:20 2019 -0700\n\n x86/mce: Don\u0027t check for the overflow bit on action optional machine checks\n \n We currently do not process SRAO (Software Recoverable Action Optional)\n machine checks if they are logged with the overflow bit set to 1 in the\n machine check bank status register. This is overly conservative.\n \n There are two cases where we could end up with an SRAO+OVER log based\n on the SDM volume 3 overwrite rules in \"Table 15-8. Overwrite Rules for\n UC, CE, and UCR Errors\"\n \n 1) First a corrected error is logged, then the SRAO error overwrites.\n The second error overwrites the first because uncorrected errors\n have a higher severity than corrected errors.\n 2) The SRAO error was logged first, followed by a correcetd error.\n In this case the first error is retained in the bank.\n \n So in either case the machine check bank will contain the address\n of the SRAO error. So we can process that even if the overflow bit\n was set.\n \n Reported-by: Yongkai Wu \u003cyongkaiwu@tencent.com\u003e\n Signed-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Cc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: linux-edac \u003clinux-edac@vger.kernel.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: x86-ml \u003cx86@kernel.org\u003e\n Link: https://lkml.kernel.org/r/20190718182920.32621-1-tony.luck@intel.com\n\ncommit 0dbd555a011c2d096a7b7e40c83c5776a7df367c\nAuthor: Christian König \u003cchristian.koenig@amd.com\u003e\nDate: Wed Jul 31 09:41:50 2019 +0200\n\n dma-buf: add more reservation object locking wrappers\n \n Complete the abstraction of the ww_mutex inside the reservation object.\n \n This allows us to add more handling and debugging to the reservation\n object in the future.\n \n Signed-off-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/320761/\n\ncommit 2f64dc70baba18dc0b2d91d92f13bda7f2f28112\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Sun Aug 4 21:10:29 2019 +0530\n\n pinctrl: rzn1: Add of_node_put() before return\n \n Each iteration of for_each_child_of_node puts the previous node, but in\n the case of a return from the middle of the loop, there is no put, thus\n causing a memory leak. Hence add an of_node_put before the return in\n three places.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 556f36e90dbe7dded81f4fac084d2bc8a2458330\nAuthor: Ming Lei \u003cming.lei@redhat.com\u003e\nDate: Thu Jul 25 17:41:46 2019 +0800\n\n blk-mq: balance mapping between present CPUs and queues\n \n Spread queues among present CPUs first, then building mapping on other\n non-present CPUs.\n \n So we can minimize count of dead queues which are mapped by un-present\n CPUs only. Then bad IO performance can be avoided by unbalanced mapping\n between present CPUs and queues.\n \n The similar policy has been applied on Managed IRQ affinity.\n \n Cc: Yi Zhang \u003cyi.zhang@redhat.com\u003e\n Reported-by: Yi Zhang \u003cyi.zhang@redhat.com\u003e\n Reviewed-by: Bob Liu \u003cbob.liu@oracle.com\u003e\n Signed-off-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit b7e9e1fb7a9227be34ad4a5e778022c3164494cf\nAuthor: Ming Lei \u003cming.lei@redhat.com\u003e\nDate: Thu Jul 25 10:05:00 2019 +0800\n\n scsi: implement .cleanup_rq callback\n \n Implement .cleanup_rq() callback for freeing driver private part\n of the request. Then we can avoid to leak this part if the request isn\u0027t\n completed by SCSI, and freed by blk-mq or upper layer(such as dm-rq) finally.\n \n Cc: Ewan D. Milne \u003cemilne@redhat.com\u003e\n Cc: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Cc: Hannes Reinecke \u003chare@suse.com\u003e\n Cc: Christoph Hellwig \u003chch@lst.de\u003e\n Cc: Mike Snitzer \u003csnitzer@redhat.com\u003e\n Cc: dm-devel@redhat.com\n Cc: \u003cstable@vger.kernel.org\u003e\n Fixes: 396eaf21ee17 (\"blk-mq: improve DM\u0027s blk-mq IO merging via blk_insert_cloned_request feedback\")\n Signed-off-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 226b4fc75c78f9c497c5182d939101b260cfb9f3\nAuthor: Ming Lei \u003cming.lei@redhat.com\u003e\nDate: Thu Jul 25 10:04:59 2019 +0800\n\n blk-mq: add callback of .cleanup_rq\n \n SCSI maintains its own driver private data hooked off of each SCSI\n request, and the pridate data won\u0027t be freed after scsi_queue_rq()\n returns BLK_STS_RESOURCE or BLK_STS_DEV_RESOURCE. An upper layer driver\n (e.g. dm-rq) may need to retry these SCSI requests, before SCSI has\n fully dispatched them, due to a lower level SCSI driver\u0027s resource\n limitation identified in scsi_queue_rq(). Currently SCSI\u0027s per-request\n private data is leaked when the upper layer driver (dm-rq) frees and\n then retries these requests in response to BLK_STS_RESOURCE or\n BLK_STS_DEV_RESOURCE returns from scsi_queue_rq().\n \n This usecase is so specialized that it doesn\u0027t warrant training an\n existing blk-mq interface (e.g. blk_mq_free_request) to allow SCSI to\n account for freeing its driver private data -- doing so would add an\n extra branch for handling a special case that all other consumers of\n SCSI (and blk-mq) won\u0027t ever need to worry about.\n \n So the most pragmatic way forward is to delegate freeing SCSI driver\n private data to the upper layer driver (dm-rq). Do so by adding\n new .cleanup_rq callback and calling a new blk_mq_cleanup_rq() method\n from dm-rq. A following commit will implement the .cleanup_rq() hook\n in scsi_mq_ops.\n \n Cc: Ewan D. Milne \u003cemilne@redhat.com\u003e\n Cc: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Cc: Hannes Reinecke \u003chare@suse.com\u003e\n Cc: Christoph Hellwig \u003chch@lst.de\u003e\n Cc: Mike Snitzer \u003csnitzer@redhat.com\u003e\n Cc: dm-devel@redhat.com\n Cc: \u003cstable@vger.kernel.org\u003e\n Fixes: 396eaf21ee17 (\"blk-mq: improve DM\u0027s blk-mq IO merging via blk_insert_cloned_request feedback\")\n Signed-off-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit a61dbfb12bc6dadce91600fe3409a3675751c716\nAuthor: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\nDate: Thu Aug 1 10:26:38 2019 -0700\n\n null_blk: implement REQ_OP_ZONE_RESET_ALL\n \n This patch implements newly introduced zone reset all operation for\n null_blk driver.\n \n Reviewed-by: Damien Le Moal \u003cdamien.lemoal@wdc.com\u003e\n Reviewed-by: Hannes Reinecke \u003chare@suse.com\u003e\n Signed-off-by: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit d81e9d494354064cd0a99d08401097dd16f335c0\nAuthor: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\nDate: Thu Aug 1 10:26:37 2019 -0700\n\n scsi: implement REQ_OP_ZONE_RESET_ALL\n \n This patch implements the zone reset all operation for sd_zbc.c. We add\n a new boolean parameter for the sd_zbc_setup_reset_cmd() to indicate\n REQ_OP_ZONE_RESET_ALL command setup. Along with that we add support in\n the completion path for the zone reset all.\n \n Reviewed-by: Hannes Reinecke \u003chare@suse.com\u003e\n Reviewed-by: Damien Le Moal \u003cdamien.lemoal@wdc.com\u003e\n Signed-off-by: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\n Acked-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 6e33dbf280d60db8c1c11dbf99c0bc475946f9c8\nAuthor: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\nDate: Thu Aug 1 10:26:36 2019 -0700\n\n blk-zoned: implement REQ_OP_ZONE_RESET_ALL\n \n This implements REQ_OP_ZONE_RESET_ALL as a special case of the block\n device zone reset operations where we just simply issue bio with the\n newly introduced req op.\n \n We issue this req op when the number of sectors is equal to the device\u0027s\n partition\u0027s number of sectors and device has no partitions.\n \n We also add support so that blk_op_str() can print the new reset-all\n zone operation.\n \n This patch also adds a generic make request check for newly\n introduced REQ_OP_ZONE_RESET_ALL req_opf. We simply return error\n when queue is zoned and reset-all flag is not set for\n REQ_OP_ZONE_RESET_ALL.\n \n Reviewed-by: Hannes Reinecke \u003chare@suse.com\u003e\n Reviewed-by: Damien Le Moal \u003cdamien.lemoal@wdc.com\u003e\n Signed-off-by: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit e84e8f0663956f45c747df5629046794cff93893\nAuthor: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\nDate: Thu Aug 1 10:26:35 2019 -0700\n\n block: add req op to reset all zones and flag\n \n This patch introduces a new request operation REQ_OP_ZONE_RESET_ALL.\n This is useful for the applications like mkfs where it needs to reset\n all the zones present on the underlying block device. As part for this\n patch we also introduce new QUEUE_FLAG_ZONE_RESETALL which indicates the\n queue zone reset all capability and corresponding helper macro.\n \n Reviewed-by: Damien Le Moal \u003cdamien.lemoal@wdc.com\u003e\n Reviewed-by: Hannes Reinecke \u003chare@suse.com\u003e\n Signed-off-by: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 67ed8b738633f8c309cfdbfdf501e09d3759ce0c\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 1 15:39:55 2019 -0700\n\n block: Fix a comment in blk_cleanup_queue()\n \n Change a reference to the legacy block layer into a reference to blk-mq.\n \n Reviewed-by: Chaitanya Kulkarni \u003cchaitanya.kulkarni@wdc.com\u003e\n Cc: Christoph Hellwig \u003chch@infradead.org\u003e\n Cc: Hannes Reinecke \u003chare@suse.com\u003e\n Cc: James Smart \u003cjames.smart@broadcom.com\u003e\n Cc: Ming Lei \u003cming.lei@redhat.com\u003e\n Cc: Jianchao Wang \u003cjianchao.w.wang@oracle.com\u003e\n Cc: Dongli Zhang \u003cdongli.zhang@oracle.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 012d4a652ca172d93315cb69f2adf7df37ea77e6\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 1 15:39:07 2019 -0700\n\n block: Fix spelling in the header above blkg_lookup()\n \n See also commit 8f4236d9008b (\"block: remove QUEUE_FLAG_BYPASS and -\u003ebypass\") # v5.0.\n \n Cc: Christoph Hellwig \u003chch@lst.de\u003e\n Cc: Hannes Reinecke \u003chare@suse.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 9cc5169cd478bc596902a57580804f1da3dfd74e\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 1 15:50:44 2019 -0700\n\n block: Improve physical block alignment of split bios\n \n Consider the following example:\n * The logical block size is 4 KB.\n * The physical block size is 8 KB.\n * max_sectors equals (16 KB \u003e\u003e 9) sectors.\n * A non-aligned 4 KB and an aligned 64 KB bio are merged into a single\n non-aligned 68 KB bio.\n \n The current behavior is to split such a bio into (16 KB + 16 KB + 16 KB\n + 16 KB + 4 KB). The start of none of these five bio\u0027s is aligned to a\n physical block boundary.\n \n This patch ensures that such a bio is split into four aligned and\n one non-aligned bio instead of being split into five non-aligned bios.\n This improves performance because most block devices can handle aligned\n requests faster than non-aligned requests.\n \n Since the physical block size is larger than or equal to the logical\n block size, this patch preserves the guarantee that the returned\n value is a multiple of the logical block size.\n \n Cc: Christoph Hellwig \u003chch@infradead.org\u003e\n Cc: Ming Lei \u003cming.lei@redhat.com\u003e\n Cc: Hannes Reinecke \u003chare@suse.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 708b25b344fd9bedd02ccc0f8eee71f7006d7d07\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 1 15:50:43 2019 -0700\n\n block: Simplify blk_bio_segment_split()\n \n Move the max_sectors check into bvec_split_segs() such that a single\n call to that function can do all the necessary checks. This patch\n optimizes the fast path further, namely if a bvec fits in a page.\n \n Cc: Christoph Hellwig \u003chch@lst.de\u003e\n Cc: Ming Lei \u003cming.lei@redhat.com\u003e\n Cc: Hannes Reinecke \u003chare@suse.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit ff9811b3cf2092fe6c39cf694e5e7f949f3b2c16\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 1 15:50:42 2019 -0700\n\n block: Simplify bvec_split_segs()\n \n Simplify this function by by removing two if-tests. Other than requiring\n that the @sectors pointer is not NULL, this patch does not change the\n behavior of bvec_split_segs().\n \n Reviewed-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Cc: Christoph Hellwig \u003chch@lst.de\u003e\n Cc: Ming Lei \u003cming.lei@redhat.com\u003e\n Cc: Hannes Reinecke \u003chare@suse.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit dad7758459bc6097115f5e783eda232f36b1ad99\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 1 15:50:41 2019 -0700\n\n block: Document the bio splitting functions\n \n Since what the bio splitting functions do is nontrivial, document these\n functions.\n \n Reviewed-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Cc: Christoph Hellwig \u003chch@infradead.org\u003e\n Cc: Ming Lei \u003cming.lei@redhat.com\u003e\n Cc: Hannes Reinecke \u003chare@suse.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit af2c68fe94e8c0a628519b60ba070c5cf6526a99\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Thu Aug 1 15:50:40 2019 -0700\n\n block: Declare several function pointer arguments \u0027const\u0027\n \n Make it clear to the compiler and also to humans that the functions\n that query request queue properties do not modify any member of the\n request_queue data structure.\n \n Reviewed-by: Johannes Thumshirn \u003cjthumshirn@suse.de\u003e\n Cc: Christoph Hellwig \u003chch@infradead.org\u003e\n Cc: Ming Lei \u003cming.lei@redhat.com\u003e\n Cc: Hannes Reinecke \u003chare@suse.com\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit a87ccce0b5a06ee546931859fa62e10f1bce54f9\nAuthor: Ming Lei \u003cming.lei@redhat.com\u003e\nDate: Wed Jul 24 11:48:43 2019 +0800\n\n blk-mq: remove blk_mq_complete_request_sync\n \n blk_mq_tagset_wait_completed_request() has been applied for waiting\n for completed request\u0027s fn, so not necessary to use\n blk_mq_complete_request_sync() any more.\n \n Cc: Max Gurtovoy \u003cmaxg@mellanox.com\u003e\n Cc: Sagi Grimberg \u003csagi@grimberg.me\u003e\n Cc: Keith Busch \u003ckeith.busch@intel.com\u003e\n Cc: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n Signed-off-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 622b8b6893ff3096e130250c1298adf57a0cab03\nAuthor: Ming Lei \u003cming.lei@redhat.com\u003e\nDate: Wed Jul 24 11:48:42 2019 +0800\n\n nvme: wait until all completed request\u0027s complete fn is called\n \n When aborting in-flight request for recovering controller, we have\n to make sure that queue\u0027s complete function is called on completed\n request before moving on. Otherwise, for example, the warning of\n WARN_ON_ONCE(qp-\u003emrs_used \u003e 0) in ib_destroy_qp_user() may be\n triggered on nvme-rdma.\n \n Fix this issue by using blk_mq_tagset_wait_completed_request.\n \n Cc: Max Gurtovoy \u003cmaxg@mellanox.com\u003e\n Cc: Sagi Grimberg \u003csagi@grimberg.me\u003e\n Cc: Keith Busch \u003ckeith.busch@intel.com\u003e\n Cc: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n Signed-off-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 78ca40724713bd422873cb4ebee86f9f499650f7\nAuthor: Ming Lei \u003cming.lei@redhat.com\u003e\nDate: Wed Jul 24 11:48:41 2019 +0800\n\n nvme: don\u0027t abort completed request in nvme_cancel_request\n \n Before aborting in-flight requests, all IO queues and their interrupts\n have been shutdown. However, request\u0027s completion function may not be\n done yet because it can be scheduled to run via IPI.\n \n So don\u0027t abort one request if it is marked as completed, otherwise\n we may abort one normal completed request.\n \n Cc: Max Gurtovoy \u003cmaxg@mellanox.com\u003e\n Cc: Sagi Grimberg \u003csagi@grimberg.me\u003e\n Cc: Keith Busch \u003ckeith.busch@intel.com\u003e\n Cc: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n Signed-off-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit f9934a80f91dba8c7029ba7601459e41ea7770aa\nAuthor: Ming Lei \u003cming.lei@redhat.com\u003e\nDate: Wed Jul 24 11:48:40 2019 +0800\n\n blk-mq: introduce blk_mq_tagset_wait_completed_request()\n \n blk-mq may schedule to call queue\u0027s complete function on remote CPU via\n IPI, but doesn\u0027t provide any way to synchronize the request\u0027s complete\n fn. The current queue freeze interface can\u0027t provide the synchonization\n because aborted requests stay at blk-mq queues during EH.\n \n In some driver\u0027s EH(such as NVMe), hardware queue\u0027s resource may be freed \u0026\n re-allocated. If the completed request\u0027s complete fn is run finally after the\n hardware queue\u0027s resource is released, kernel crash will be triggered.\n \n Prepare for fixing this kind of issue by introducing\n blk_mq_tagset_wait_completed_request().\n \n Cc: Max Gurtovoy \u003cmaxg@mellanox.com\u003e\n Cc: Sagi Grimberg \u003csagi@grimberg.me\u003e\n Cc: Keith Busch \u003ckeith.busch@intel.com\u003e\n Cc: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n Signed-off-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit aa306ab703e9452b1e25cc8e8f04b8df523d0bb8\nAuthor: Ming Lei \u003cming.lei@redhat.com\u003e\nDate: Wed Jul 24 11:48:39 2019 +0800\n\n blk-mq: introduce blk_mq_request_completed()\n \n NVMe needs this function to decide if one request to be aborted has\n been completed in normal IO path already.\n \n So introduce it.\n \n Cc: Max Gurtovoy \u003cmaxg@mellanox.com\u003e\n Cc: Sagi Grimberg \u003csagi@grimberg.me\u003e\n Cc: Keith Busch \u003ckeith.busch@intel.com\u003e\n Cc: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Sagi Grimberg \u003csagi@grimberg.me\u003e\n Signed-off-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\n\ncommit 9cb9a17813bf0de1f8ad6deb9538296d5148b5a8\nAuthor: Linus Lüssing \u003clinus.luessing@c0d3.blue\u003e\nDate: Sun Aug 4 20:54:54 2019 +0200\n\n batman-adv: BATMAN_V: aggregate OGMv2 packets\n \n Instead of transmitting individual OGMv2 packets from the aggregation\n queue merge those OGMv2 packets into a single one and transmit this\n aggregate instead.\n \n This reduces overhead as it saves an ethernet header and a transmission\n per aggregated OGMv2 packet.\n \n Signed-off-by: Linus Lüssing \u003clinus.luessing@c0d3.blue\u003e\n Signed-off-by: Sven Eckelmann \u003csven@narfation.org\u003e\n Signed-off-by: Simon Wunderlich \u003csw@simonwunderlich.de\u003e\n\ncommit f89255a02f1d75d8e1b9d1c31435fcb64840cb2a\nAuthor: Linus Lüssing \u003clinus.luessing@c0d3.blue\u003e\nDate: Sun Aug 4 20:54:53 2019 +0200\n\n batman-adv: BATMAN_V: introduce per hard-iface OGMv2 queues\n \n In preparation for the OGMv2 packet aggregation, hold OGMv2 packets for\n up to BATADV_MAX_AGGREGATION_MS milliseconds (100ms) on per\n hard-interface queues, before transmitting.\n \n This allows us to later squash multiple OGMs into a single frame\n and transmission for reduced overhead.\n \n Signed-off-by: Linus Lüssing \u003clinus.luessing@c0d3.blue\u003e\n Signed-off-by: Sven Eckelmann \u003csven@narfation.org\u003e\n Signed-off-by: Simon Wunderlich \u003csw@simonwunderlich.de\u003e\n\ncommit b40d73784ffc33f3c6431e7ceec3b20fffcd95c3\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Sun Aug 4 13:48:26 2019 +0100\n\n drm/i915: Replace struct_mutex for batch pool serialisation\n \n Switch to tracking activity via i915_active on individual nodes, only\n keeping a list of retired objects in the cache, and reaping the cache\n when the engine itself idles.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190804124826.30272-2-chris@chris-wilson.co.uk\n\ncommit a4e57f9031ccd543c549a34524af16c0c246e628\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Sun Aug 4 13:48:25 2019 +0100\n\n drm/i915: Teach execbuffer to take the engine wakeref not GT\n \n In the next patch, we would like to couple into the engine wakeref to\n free the batch pool on idling. The caveat here is that we therefore want\n to track the engine wakeref more precisely and to hold it instead of the\n broader GT wakeref as we process the ioctl.\n \n v2: Avoid introducing odd semantics for a shortlived timeline-\u003emutex\n acquisition interface.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190804124826.30272-1-chris@chris-wilson.co.uk\n\ncommit 1fa475fee4909777a83f1c46832dde1141b09364\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Mon Jul 8 16:12:36 2019 -0700\n\n fm10k: fix fm10k_get_fault_pf to read correct address\n \n Fix assignment of the FM10K_FAULT_ADDR_LO register into fault-\u003eaddress\n by using a bit-wise |\u003d operation. Without this, the low address is\n completely overwriting the high potion of the address. This caused the\n fault to incorrectly return only the lower 32 bits of the fault address.\n \n This issue was detected by cppcheck and resolves the following warnings\n produced by that tool:\n \n [fm10k_pf.c:1668] -\u003e [fm10k_pf.c:1670]: (style) Variable\n \u0027fault-\u003eaddress\u0027 is reassigned a value before the old one has been used.\n \n [fm10k_pf.c:1669] -\u003e [fm10k_pf.c:1670]: (style) Variable\n \u0027fault-\u003eaddress\u0027 is reassigned a value before the old one has been used.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit a3ffeaf7c2bedb5b8658f06e4ca09dc8a352ead6\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Mon Jul 8 16:12:35 2019 -0700\n\n fm10k: convert NON_Q_VECTORS(hw) into NON_Q_VECTORS\n \n The driver currently uses a macro to decide whether we should use\n NON_Q_VECTORS_PF or NON_Q_VECTORS_VF.\n \n However, we also define NON_Q_VECTORS_VF to the same value as\n NON_Q_VECTORS_PF. This means that the macro NON_Q_VECTORS(hw) will\n always return the same value.\n \n Let\u0027s just remove this macro, and replace it directly with an enum value\n on the enum non_q_vectors.\n \n This was detected by cppcheck and fixes the following warnings when\n building with BUILD\u003dKERNEL\n \n [fm10k_ethtool.c:1123]: (style) Same value in both branches of ternary\n operator.\n \n [fm10k_ethtool.c:1142]: (style) Same value in both branches of ternary\n operator.\n \n [fm10k_main.c:1826]: (style) Same value in both branches of ternary\n operator.\n \n [fm10k_main.c:1849]: (style) Same value in both branches of ternary\n operator.\n \n [fm10k_main.c:1858]: (style) Same value in both branches of ternary\n operator.\n \n [fm10k_pci.c:901]: (style) Same value in both branches of ternary\n operator.\n \n [fm10k_pci.c:1040]: (style) Same value in both branches of ternary\n operator.\n \n [fm10k_pci.c:1726]: (style) Same value in both branches of ternary\n operator.\n \n [fm10k_pci.c:1763]: (style) Same value in both branches of ternary\n operator.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit d5c2f39500549a7e2cd397f3ec713bd8d85af3e1\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Mon Jul 8 16:12:34 2019 -0700\n\n fm10k: mark unused parameters with __always_unused\n \n Several functions in the fm10k driver have specific function templates,\n as they are used as function pointers. The parameters in these functions\n are not always used. Explicitly mark unused parameters with the\n __always_unused macro, so that the compiler will not warn about them\n when building with the -Wunused-parameter warning enabled.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 27429be75e0d291a74b57f9753537e727b7635b7\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Mon Jul 8 16:12:33 2019 -0700\n\n fm10k: cast page_addr to u8 * when incrementing it\n \n The page_addr variable is a void pointer. Incrementing it before calling\n prefetch is technically undefined. Fix this by casting it to a u8*\n pointer before incrementing it. This ensures that we increment the\n pointer value in byte units, instead of relying on this undefined\n behavior.\n \n This was detected by cppcheck, and resolves the following warning\n produced by that tool:\n \n [fm10k_main.c:328]: (portability) \u0027page_addr\u0027 is of type \u0027void *\u0027. When\n using void pointers in calculations, the behaviour is undefined.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 9aac0fbd471bf09b614b2038faddc91f53b788bb\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Mon Jul 8 16:12:32 2019 -0700\n\n fm10k: explicitly return 0 on success path in function\n \n In the fm10k_handle_resume function, return 0 explicitly at the end of\n the function instead of returning the err value.\n \n This was detected by cppcheck and resolves the following style warning\n produced by that tool:\n \n [fm10k_pci.c:2768] -\u003e [fm10k_pci.c:2787]: (warning) Identical condition\n \u0027err\u0027, second condition is always false\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit cb1b5226cbdf0bc01dcdd8d422d4051fec1b77ef\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Mon Jul 8 16:12:31 2019 -0700\n\n fm10k: remove needless initialization of size local variable\n \n The local variable \u0027size\u0027 in fm10k_dfwd_add_station is initialized, but\n is always re-assigned immediately before use. Remove this unnecessary\n initialization.\n \n This was detected by cppcheck and resolves the following warning\n produced by that tool:\n \n [fm10k_netdev.c:1466]: (style) Variable \u0027size\u0027 is assigned a value that is never used.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 4d12002fd29ad8d8c3da0232153217bee4f10326\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Mon Jul 8 16:12:30 2019 -0700\n\n fm10k: remove needless assignment of err local variable\n \n The local variable err in several functions in the fm10k_netdev.c file\n is initialized with a value that is never used. The err value is\n immediately re-assigned in all cases where it will be checked. Remove\n the unnecessary initializers.\n \n This was detected by cppcheck and resolves the following warnings\n produced by that tool:\n \n [fm10k_netdev.c:999] -\u003e [fm10k_netdev.c:1004]: (style) Variable \u0027err\u0027 is\n reassigned a value before the old one has been used.\n \n [fm10k_netdev.c:1019] -\u003e [fm10k_netdev.c:1024]: (style) Variable \u0027err\u0027\n is reassigned a value before the old one has been used.\n \n [fm10k_netdev.c:64]: (style) Variable \u0027err\u0027 is assigned a value that is\n never used.\n \n [fm10k_netdev.c:131]: (style) Variable \u0027err\u0027 is assigned a value that\n is never used.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit d9ecd1f748f2ddb7cc15d56acadbc90b3ce235a9\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Mon Jul 8 16:12:29 2019 -0700\n\n fm10k: remove unnecessary variable initializer\n \n The err variable in the fm10k_tlv_attr_parse function is initialized\n with zero. However, the function never reads err without first assigning\n it from a function call. Remove this unnecessary initialization.\n \n This was detected by cppcheck and resolves the following warning\n produced by that tool:\n \n [fm10k_tlv.c:498]: (style) Variable \u0027err\u0027 is assigned a value that is\n never used.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit d3dd552da3eb16e369d2459e3fa927542f11d17c\nAuthor: Clément Péron \u003cperon.clem@gmail.com\u003e\nDate: Fri Jun 7 20:10:57 2019 -0300\n\n media: dt-bindings: media: sunxi-ir: Add H6 compatible\n \n There are some minor differences between A31 or A64 with H6 IR peripheral.\n \n But A31 IR driver is compatible with H6.\n \n Signed-off-by: Clément Péron \u003cperon.clem@gmail.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 020826f4ec57eb957d3f1a01902371d83dfce174\nAuthor: Clément Péron \u003cperon.clem@gmail.com\u003e\nDate: Fri Jun 7 20:10:54 2019 -0300\n\n media: dt-bindings: media: sunxi-ir: Add A64 compatible\n \n There are some minor differences between A31 and A64 driver.\n \n But A31 IR driver is compatible with A64.\n \n Signed-off-by: Clément Péron \u003cperon.clem@gmail.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit b136d72cb89dc2bd11ba001c90cdc65b5f5a1034\nAuthor: Clément Péron \u003cperon.clem@gmail.com\u003e\nDate: Fri Jun 7 20:10:51 2019 -0300\n\n media: rc: sunxi: Add RXSTA bits definition\n \n We are using RXINT bits definition when looking at RXSTA register.\n \n These bits are equal but it\u0027s not really proper.\n \n Introduce the RXSTA bits and use them to have coherency.\n \n Signed-off-by: Clément Péron \u003cperon.clem@gmail.com\u003e\n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 87d0609801ebcdf18639bb30ec5ec9a380f15be8\nAuthor: Clément Péron \u003cperon.clem@gmail.com\u003e\nDate: Fri Jun 7 20:10:50 2019 -0300\n\n media: rc: sunxi: Add A31 compatible\n \n Allwiner A31 has a different memory mapping so add the compatible\n we will need it later.\n \n Signed-off-by: Clément Péron \u003cperon.clem@gmail.com\u003e\n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 6b197cb5b4dc7be463599daeb28dfb8d24674746\nAuthor: Clément Péron \u003cperon.clem@gmail.com\u003e\nDate: Fri Jun 7 20:10:49 2019 -0300\n\n media: rc: Introduce sunxi_ir_quirks\n \n This driver is used in various Allwinner SoC with different configuration.\n \n Introduce a quirks struct to know the fifo size and if a reset is required.\n \n Signed-off-by: Clément Péron \u003cperon.clem@gmail.com\u003e\n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 47fabc9cbc05bd5ebc70e1953d118d2a424df841\nAuthor: Clément Péron \u003cperon.clem@gmail.com\u003e\nDate: Fri Jun 7 20:10:48 2019 -0300\n\n media: dt-bindings: media: sunxi-ir: Add A31 compatible\n \n Allwinner A31 has introduced a new memory mapping and a\n reset line.\n \n The difference in memory mapping are :\n \n - In the configure register there is a new sample bit\n and Allwinner has introduced the active threshold feature.\n \n - In the status register a new STAT bit is present.\n \n Note: CGPO and DRQ_EN bits are removed on A31 but present on A13\n and on new SoCs like A64/H6.\n This is actually not an issue as these bits are togglable and new\n SoCs have a dedicated bindings.\n \n Introduce this bindings to make a difference since this generation.\n And declare the reset line required since A31.\n \n Signed-off-by: Clément Péron \u003cperon.clem@gmail.com\u003e\n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit b3185ab502749539872402b6e7198c5229ff20d6\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Tue Jul 30 14:55:55 2019 -0300\n\n media: ir-kbd-i2c: remove outdated comments\n \n The \"free memory\" comment is obsolete since 2013 and the other ones\n explain the obvious. Just remove the comments.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit ab1cbdf159beba7395a13ab70bc71180929ca064\nAuthor: Oliver Neukum \u003coneukum@suse.com\u003e\nDate: Tue Jul 30 05:50:44 2019 -0300\n\n media: iguanair: add sanity checks\n \n The driver needs to check the endpoint types, too, as opposed\n to the number of endpoints. This also requires moving the check earlier.\n \n Reported-by: syzbot+01a77b82edaa374068e1@syzkaller.appspotmail.com\n Signed-off-by: Oliver Neukum \u003coneukum@suse.com\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 7328d682371722433c96416c33fa5e2b7a6d97bc\nAuthor: Sean Young \u003csean@mess.org\u003e\nDate: Thu Jul 25 16:32:31 2019 -0300\n\n media: rc: describe rc protocols and their scancodes\n \n This lists the rc protocols the kernel knows about and how they are\n converted to and from scancodes.\n \n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 16407a6af464301f1d5c32599ab07e6f89a98e62\nAuthor: Sean Young \u003csean@mess.org\u003e\nDate: Thu Jul 25 09:49:45 2019 -0300\n\n media: lirc: document BPF IR decoding\n \n This is just a start.\n \n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 14e3cdbb00a885eedc95c0cf8eda8fe28d26d6b4\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Wed Jun 19 10:24:17 2019 -0300\n\n media: don\u0027t drop front-end reference count for -\u003edetach\n \n A bugfix introduce a link failure in configurations without CONFIG_MODULES:\n \n In file included from drivers/media/usb/dvb-usb/pctv452e.c:20:0:\n drivers/media/usb/dvb-usb/pctv452e.c: In function \u0027pctv452e_frontend_attach\u0027:\n drivers/media/dvb-frontends/stb0899_drv.h:151:36: error: weak declaration of \u0027stb0899_attach\u0027 being applied to a already existing, static definition\n \n The problem is that the !IS_REACHABLE() declaration of stb0899_attach()\n is a \u0027static inline\u0027 definition that clashes with the weak definition.\n \n I further observed that the bugfix was only done for one of the five users\n of stb0899_attach(), the other four still have the problem. This reverts\n the bugfix and instead addresses the problem by not dropping the reference\n count when calling \u0027-\u003edetach()\u0027, instead we call this function directly\n in dvb_frontend_put() before dropping the kref on the front-end.\n \n I first submitted this in early 2018, and after some discussion it\n was apparently discarded. While there is a long-term plan in place,\n that plan is obviously not nearing completion yet, and the current\n kernel is still broken unless this patch is applied.\n \n Link: https://patchwork.kernel.org/patch/10140175/\n Link: https://patchwork.linuxtv.org/patch/54831/\n \n Cc: Max Kellermann \u003cmax.kellermann@gmail.com\u003e\n Cc: Wolfgang Rohdewald \u003cwolfgang@rohdewald.de\u003e\n Cc: stable@vger.kernel.org\n Fixes: f686c14364ad (\"[media] stb0899: move code to \"detach\" callback\")\n Fixes: 6cdeaed3b142 (\"media: dvb_usb_pctv452e: module refcount changes were unbalanced\")\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 7084ed30ae2aba4efb4d3ef72c0e7042644e7637\nAuthor: Mark Zhang \u003cmarkz@mellanox.com\u003e\nDate: Wed Jul 31 14:40:14 2019 +0300\n\n IB/mlx5: Support MLX5_CMD_OP_QUERY_LAG as a DEVX general command\n \n The \"MLX5_CMD_OP_QUERY_LAG\" is one of the DEVX general commands, add it.\n \n Fixes: 8aa8c95ce4cc (\"IB/mlx5: Add support for DEVX general command\")\n Signed-off-by: Mark Zhang \u003cmarkz@mellanox.com\u003e\n Reviewed-by: Yishai Hadas \u003cyishaih@mellanox.com\u003e\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n\ncommit ea77388b02270b0af8dc57f668f311235ea068f0\nAuthor: Mark Zhang \u003cmarkz@mellanox.com\u003e\nDate: Wed Jul 31 14:40:13 2019 +0300\n\n net/mlx5: Fix mlx5_ifc_query_lag_out_bits\n \n Remove the \"reserved_at_40\" field to match the device specification.\n \n Fixes: 84df61ebc69b (\"net/mlx5: Add HW interfaces used by LAG\")\n Signed-off-by: Mark Zhang \u003cmarkz@mellanox.com\u003e\n Reviewed-by: Yishai Hadas \u003cyishaih@mellanox.com\u003e\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n\ncommit 56eba15d1c601d7e8a40b2997c9aff72bdae9b0f\nAuthor: David Ahern \u003cdsahern@gmail.com\u003e\nDate: Thu Aug 1 11:56:48 2019 -0700\n\n selftests: Add use case section to fcnal-test\n \n Add use case section to fcnal-test.\n \n Initial test is VRF based with a bridge and vlans. The commands\n stem from bug reports fixed by:\n \n a173f066c7cf (\"netfilter: bridge: Don\u0027t sabotage nf_hook calls from an l3mdev\")\n cd6428988bf4 (\"netfilter: bridge: Don\u0027t sabotage nf_hook calls for an l3mdev slave\")\n \n Signed-off-by: David Ahern \u003cdsahern@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit db6641ee6e9e728efb71bda93740559dc55f696c\nAuthor: David Ahern \u003cdsahern@gmail.com\u003e\nDate: Thu Aug 1 11:56:47 2019 -0700\n\n selftests: Add ipv6 netfilter tests to fcnal-test\n \n Add IPv6 netfilter tests to send tcp reset or icmp unreachable for a\n port. Initial tests are VRF only.\n \n Signed-off-by: David Ahern \u003cdsahern@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 88f2b36053b97d3299976dd3af1c768a7f5d9c55\nAuthor: David Ahern \u003cdsahern@gmail.com\u003e\nDate: Thu Aug 1 11:56:46 2019 -0700\n\n selftests: Add ipv4 netfilter tests to fcnal-test\n \n Add netfilter tests to send tcp reset or icmp unreachable for a port.\n Initial tests are VRF only.\n \n Signed-off-by: David Ahern \u003cdsahern@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4cd12f61b55bc6a670900d75806a2f0122fc6658\nAuthor: David Ahern \u003cdsahern@gmail.com\u003e\nDate: Thu Aug 1 11:56:45 2019 -0700\n\n selftests: Add ipv6 runtime tests to fcnal-test\n \n Add IPv6 runtime tests where passive (no traffic flowing) and active\n (with traffic) sockets are expected to be reset on device deletes.\n \n Signed-off-by: David Ahern \u003cdsahern@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 0113f726856e389461fb0fc8d519fc2e8fe52d46\nAuthor: David Ahern \u003cdsahern@gmail.com\u003e\nDate: Thu Aug 1 11:56:44 2019 -0700\n\n selftests: Add ipv4 runtime tests to fcnal-test\n \n Add runtime tests where passive (no traffic flowing) and active (with\n traffic) sockets are expected to be reset on device deletes.\n \n Signed-off-by: David Ahern \u003cdsahern@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 34d0302ab86172b373e38ef02ea10519ab1c2388\nAuthor: David Ahern \u003cdsahern@gmail.com\u003e\nDate: Thu Aug 1 11:56:43 2019 -0700\n\n selftests: Add ipv6 address bind tests to fcnal-test\n \n Add IPv6 address bind tests to fcnal-test.sh. Verifies socket binding to\n local addresses for raw, tcp and udp including device and VRF cases.\n \n Signed-off-by: David Ahern \u003cdsahern@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 75b2b2b3db4ce660541709c52e826e2ac5c602ad\nAuthor: David Ahern \u003cdsahern@gmail.com\u003e\nDate: Thu Aug 1 11:56:42 2019 -0700\n\n selftests: Add ipv4 address bind tests to fcnal-test\n \n Add address bind tests to fcnal-test.sh. Verifies socket binding to\n local addresses for raw, tcp and udp including device and VRF cases.\n \n Signed-off-by: David Ahern \u003cdsahern@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6abdb651255784f1907d8c8fcbf7e4ba4196b1da\nAuthor: David Ahern \u003cdsahern@gmail.com\u003e\nDate: Thu Aug 1 11:56:41 2019 -0700\n\n selftests: Add ipv6 udp tests to fcnal-test\n \n Add IPv6 udp tests to fcnal-test.sh. Covers the permutations of directly\n connected addresses, routed destinations, VRF and non-VRF, and expected\n failures for both clients and servers. Includes permutations with\n net.ipv4.udp_l3mdev_accept set to 0 and 1.\n \n Signed-off-by: David Ahern \u003cdsahern@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a4368be9ad23d2a8e13c2b98409f9df166a4c9c5\nAuthor: David Ahern \u003cdsahern@gmail.com\u003e\nDate: Thu Aug 1 11:56:40 2019 -0700\n\n selftests: Add ipv4 udp tests to fcnal-test\n \n Add udp tests to fcnal-test.sh. Covers the permutations of directly\n connected addresses, routed destinations, VRF and non-VRF, and expected\n failures for both clients and servers. Includes permutations with\n net.ipv4.udp_l3mdev_accept set to 0 and 1.\n \n Signed-off-by: David Ahern \u003cdsahern@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a071bbf20539fa9166820c34c71cf2167dc23833\nAuthor: David Ahern \u003cdsahern@gmail.com\u003e\nDate: Thu Aug 1 11:56:39 2019 -0700\n\n selftests: Add ipv6 tcp tests to fcnal-test\n \n Add IPv6 tcp tests to fcnal-test.sh. Covers the permutations of directly\n connected addresses, routed destinations, VRF and non-VRF, and expected\n failures for both clients and servers. Includes permutations with\n net.ipv4.tcp_l3mdev_accept set to 0 and 1.\n \n Signed-off-by: David Ahern \u003cdsahern@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit bbd7c764086b4e6bea8c37e94914f6e5fee34f7d\nAuthor: David Ahern \u003cdsahern@gmail.com\u003e\nDate: Thu Aug 1 11:56:38 2019 -0700\n\n selftests: Add ipv4 tcp tests to fcnal-test\n \n Add tcp tests to fcnal-test.sh. Covers the permutations of directly\n connected addresses, routed destinations, VRF and non-VRF, and expected\n failures for both clients and servers. Includes permutations with\n net.ipv4.tcp_l3mdev_accept set to 0 and 1.\n \n Signed-off-by: David Ahern \u003cdsahern@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c0644e71df3302ce44ef4572d3df9fb4758bcb8b\nAuthor: David Ahern \u003cdsahern@gmail.com\u003e\nDate: Thu Aug 1 11:56:37 2019 -0700\n\n selftests: Add ipv6 ping tests to fcnal-test\n \n Add IPv6 ping tests to fcnal-test.sh. Covers the permutations of directly\n connected addresses, routed destinations, VRF and non-VRF, and expected\n failures.\n \n Setup includes unreachable routes and fib rules blocking traffic.\n \n Signed-off-by: David Ahern \u003cdsahern@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c032dd8cc7e23bbcc53e00b262594657ff1b7532\nAuthor: David Ahern \u003cdsahern@gmail.com\u003e\nDate: Thu Aug 1 11:56:36 2019 -0700\n\n selftests: Add ipv4 ping tests to fcnal-test\n \n Add ping tests to fcnal-test.sh. Covers the permutations of directly\n connected addresses, routed destinations, VRF and non-VRF, and expected\n failures.\n \n Setup includes unreachable routes and fib rules blocking traffic.\n \n Signed-off-by: David Ahern \u003cdsahern@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6f9d5cacfe07308fd4007ebdcb76861752d0a1ad\nAuthor: David Ahern \u003cdsahern@gmail.com\u003e\nDate: Thu Aug 1 11:56:35 2019 -0700\n\n selftests: Setup for functional tests for fib and socket lookups\n \n Initial commit for functional test suite for fib and socket lookups.\n This commit contains the namespace setup, networking config, test options\n and other basic infrastructure.\n \n Signed-off-by: David Ahern \u003cdsahern@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit acda655fefae352a48eec87c8f8487de1608a48b\nAuthor: David Ahern \u003cdsahern@gmail.com\u003e\nDate: Thu Aug 1 11:56:34 2019 -0700\n\n selftests: Add nettest\n \n Add nettest - a simple program with an implementation for various networking\n APIs. nettest is used for tcp, udp and raw functional tests for both IPv4\n and IPv6.\n \n Point of this command versus existing utilities:\n - controlled implementation of the APIs and the order in which they\n are called,\n - ability to verify ingress device, local and remote addresses,\n - timeout for controlled test length,\n - ability to discriminate a timeout from a system call failure, and\n - simplicity with test scripts.\n \n The command returns:\n 0 on success,\n 1 for any system call failure, and\n 2 on timeout.\n \n Signed-off-by: David Ahern \u003cdsahern@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1b90af292e71b20d03b837d39406acfbdc5d4b2a\nAuthor: Junwei Hu \u003chujunwei4@huawei.com\u003e\nDate: Thu Aug 1 00:03:30 2019 +0800\n\n ipvs: Improve robustness to the ipvs sysctl\n \n The ipvs module parse the user buffer and save it to sysctl,\n then check if the value is valid. invalid value occurs\n over a period of time.\n Here, I add a variable, struct ctl_table tmp, used to read\n the value from the user buffer, and save only when it is valid.\n I delete proc_do_sync_mode and use extra1/2 in table for the\n proc_dointvec_minmax call.\n \n Fixes: f73181c8288f (\"ipvs: add support for sync threads\")\n Signed-off-by: Junwei Hu \u003chujunwei4@huawei.com\u003e\n Acked-by: Julian Anastasov \u003cja@ssi.bg\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit e84fb4b3666dabd3917952fb33588daa891a6ad3\nAuthor: Matteo Croce \u003cmcroce@redhat.com\u003e\nDate: Tue Jul 23 03:23:03 2019 +0200\n\n netfilter: conntrack: use shared sysctl constants\n \n Use shared sysctl variables for zero and one constants, as in commit\n eec4844fae7c (\"proc/sysctl: add shared variables for range check\")\n \n Fixes: 8f14c99c7eda (\"netfilter: conntrack: limit sysctl setting for boolean options\")\n Signed-off-by: Matteo Croce \u003cmcroce@redhat.com\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit 8c0bb7873815bf8c3c4dfb24e8ebf4fefb4c35d2\nAuthor: Fernando Fernandez Mancera \u003cffmancera@riseup.net\u003e\nDate: Wed Jul 10 12:05:59 2019 +0200\n\n netfilter: synproxy: rename mss synproxy_options field\n \n After introduce \"mss_encode\" field in the synproxy_options struct the field\n \"mss\" is a little confusing. It has been renamed to \"mss_option\".\n \n Signed-off-by: Fernando Fernandez Mancera \u003cffmancera@riseup.net\u003e\n Signed-off-by: Pablo Neira Ayuso \u003cpablo@netfilter.org\u003e\n\ncommit d81765d693db67ee9afab40b4476e205723defe3\nAuthor: Marek Vasut \u003cmarex@denx.de\u003e\nDate: Sat Aug 3 15:49:40 2019 +0200\n\n ARM: dts: imx53: Update LCD panel node on M53Menlo\n \n Update the panel node with latest version of the panel used on the system.\n Add missing pincontrol phandle to the panel node.\n \n Signed-off-by: Marek Vasut \u003cmarex@denx.de\u003e\n Cc: Shawn Guo \u003cshawnguo@kernel.org\u003e\n Cc: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Cc: NXP Linux Team \u003clinux-imx@nxp.com\u003e\n To: linux-arm-kernel@lists.infradead.org\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit cd1b1264632a3fe739016843005e49e42c60be55\nAuthor: Christian Gmeiner \u003cchristian.gmeiner@gmail.com\u003e\nDate: Fri Aug 2 14:20:28 2019 +0200\n\n arm64: defconfig: CONFIG_DRM_ETNAVIV\u003dm\n \n For imx8 we want to enable etnaviv, let\u0027s enable it\n in defconfig, it will be useful to have it enabled for KernelCI\n boot and runtime testing.\n \n Signed-off-by: Christian Gmeiner \u003cchristian.gmeiner@gmail.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 2c1a3f4dc4b927d182f725c83e31bb055c9eb881\nAuthor: Martin Kaiser \u003cmartin@kaiser.cx\u003e\nDate: Mon Jul 29 16:23:16 2019 +0200\n\n ARM: dts: imx25-pdk: native-mode is part of display-timings\n \n Move the native-mode property inside the display-timings node.\n \n According to\n Documentation/devicetree/bindings/display/panel/display-timing.txt.\n native-mode is a property of the display-timings node.\n \n If it\u0027s located outside of display-timings, the native-mode setting is\n ignored and the first display timing is used (which is a problem only if\n someone adds another display timing).\n \n Signed-off-by: Martin Kaiser \u003cmartin@kaiser.cx\u003e\n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 9b9c60bed562c3718ae324a86f3f30a4ff983cf8\nAuthor: Abel Vesa \u003cabel.vesa@nxp.com\u003e\nDate: Tue Jul 30 10:22:55 2019 +0300\n\n clk: imx8mq: Mark AHB clock as critical\n \n Initially, the TMU_ROOT clock was marked as critical, which automatically\n made the AHB clock to stay always on. Since the TMU_ROOT clock is not\n marked as critical anymore, following commit:\n \n \"clk: imx8mq: Remove CLK_IS_CRITICAL flag for IMX8MQ_CLK_TMU_ROOT\"\n \n all the clocks that derive from ipg_root clock (and implicitly ahb clock)\n would also have to enable, along with their own gate, the AHB clock.\n \n But considering that AHB is actually a bus that has to be always on, we mark\n it as critical in the clock provider driver and then all the clocks that\n derive from it can be controlled through the dedicated per IP gate which\n follows after the ipg_root clock.\n \n Signed-off-by: Abel Vesa \u003cabel.vesa@nxp.com\u003e\n Tested-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit ba7372fbba5e3c592d4e646c22b2d4e63f60a2a2\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Sun Jul 28 17:08:17 2019 +0300\n\n arm64: dts: imx8mq-evk: Unbypass audio_pll1\n \n Making audio_pll1 parent of audio_pll1_bypass, will allow\n setting rates multiple of 8000 for children.\n \n After unbypass clk hierarchy looks like this:\n * osc_25m\n * audio_pll1\n * audio_pll1_bypass\n * audio_pll1_out\n * sai2\n * sai2_root_clk\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 2a44db130351877733938f0283b0dda9828a71b7\nAuthor: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\nDate: Sat Jul 27 16:26:40 2019 +0200\n\n ARM: dts: imx: Cleanup style around assignment operator\n \n Use a space before and after assignment operator to have consistent\n style.\n \n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 05103ea9a3159b9bb7004e9c0693948d4d6124f9\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Fri Jul 26 23:06:58 2019 +0200\n\n drm/vblank: drop use of DRM_WAIT_ON()\n \n DRM_WAIT_ON() is from the deprecated drm_os_linux header and\n the modern replacement is the wait_event_*.\n \n The return values differ, so a conversion is needed to\n keep the original interface towards userspace.\n Introduced a switch/case to make code obvious.\n \n Analysis from Michel Dänzer:\n \n The waiting condition rely on all relevant places where vblank_count\n is modified calls wake_up(\u0026vblank-\u003equeue).\n \n drm_handle_vblank():\n - Calls wake_up(\u0026vblank-\u003equeue)\n \n drm_vblank_enable():\n - There is no need here because there can be no sleeping waiters\n in the queue, because vblank-\u003eenabled \u003d\u003d false immediately\n terminates any waits.\n \n drm_crtc_accurate_vblank_count():\n - This is called from interrupt handlers, at least from\n amdgpu_dm.c:dm_pflip_high_irq(). Not sure it needs to wake up\n the queue though, the driver should call\n drm_(crtc_)_handle_vblank anyway.\n \n drm_vblank_disable_and_save():\n - It can be called from an interrupt, via drm_handle_vblank -\u003e\n vblank_disable_fn. However, the only place where\n drm_vblank_disable_and_save can be called with sleeping waiters\n in the queue is in drm_crtc_vblank_off, which wakes up the queue\n afterwards (which terminates all waits, because\n vblank-\u003eenabled \u003d\u003d false at this point).\n \n v3:\n - Added analysis to changelog from Michel Dänzer\n - Moved return result handling inside if (req_seq !\u003d seq) (Daniel V)\n - Reused more of the former logic - resulting in simpler code\n - Dropped Reviewed-by from Sean Paul as this is a new implementation\n \n v2:\n - Fix so the case where req_seq equals seq was handled properly\n - quick hack to check if IGT became happy\n - Only sent to igt, not to dri-devel\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Michel Dänzer \u003cmichel.daenzer@amd.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190726210658.GA6299@ravnborg.org\n\ncommit 13f138d3fcbcb7d22b341522512844bf50cffd82\nAuthor: Fabio Estevam \u003cfestevam@gmail.com\u003e\nDate: Thu Jul 25 10:44:12 2019 -0300\n\n ARM: dts: imx6qdl-nit6xlite: Remove invalid properties\n \n The \"vqmmc-1-8-v\" and \"ocr-limit\" properties are not documented\n anywhere, so just remove them.\n \n Signed-off-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 0d9df5815ccc9c5e4822b4e52e0a2da63e9c1adb\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Tue Jul 9 16:00:15 2019 +0800\n\n arm64: dts: imx8mm: Add opp-suspend property to OPP table\n \n Add opp-suspend property to each OPP, the of opp core will\n select the OPP HW supported and with highest rate to be\n suspend opp, it will speed up the suspend/resume process.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit db4cfe2fef509cf46c4de4869104eb47347436e6\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Tue Jul 9 16:00:14 2019 +0800\n\n arm64: dts: imx8mq: Add opp-suspend property to OPP table\n \n Add opp-suspend property to each OPP, the of opp core will\n select the OPP HW supported and with highest rate to be\n suspend opp, it will speed up the suspend/resume process.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 66f1f58055a4d3a25b9fcb93c2a3c894092094aa\nAuthor: Chuanhua Han \u003cchuanhua.han@nxp.com\u003e\nDate: Wed May 29 16:32:54 2019 +0800\n\n arm64: dts: ls1088a: Revise gpio registers to little-endian\n \n Since fsl-ls1088a Soc GPIO registers are used as little endian,\n the patch adds the little-endian attribute to each gpio node.\n \n Signed-off-by: Chuanhua Han \u003cchuanhua.han@nxp.com\u003e\n Acked-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 546d92d34f3c60e0165f405ce2156d23f29aeca1\nAuthor: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\nDate: Mon Jul 22 13:30:43 2019 +0300\n\n arm64: dts: add the console node for DPAA2 platforms\n \n Add the console device tree node for the following\n DPAA2 based platforms: LS1088A, LS2080A, LS2088A and LX2160A.\n \n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 356c27227b3b6aa824dcf11ffe632095e3cffe8a\nAuthor: Richard Hu \u003crichard.hu@technexion.com\u003e\nDate: Mon Jul 22 13:27:29 2019 +0300\n\n arm64: dts: fsl: pico-pi: Add a device tree for the PICO-PI-IMX8M\n \n TechNexion PICO-PI-IMX8M-DEV evaluation and development kit based on\n NXP i.MX8M Quad applications processor. Datasheet can be found at:\n \n https://s3.us-east-2.amazonaws.com/technexion/datasheets/picopiimx8m.pdf\n \n The current level of support yields a working console and is able to boot\n userspace from NFS or init ramdisk.\n \n Additional subsystems that are active :\n - Ethernet\n - USB\n \n Cc: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Richard Hu \u003crichard.hu@technexion.com\u003e\n Signed-off-by: Andra Danciu \u003candradanciu1997@gmail.com\u003e\n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit adc898e3d11ab2c28c760a05b40b8f2372dc44f5\nAuthor: Fugang Duan \u003cfugang.duan@nxp.com\u003e\nDate: Mon Jul 22 16:28:24 2019 +0800\n\n arm64: dts: imx8qxp: add serial alias\n \n Add i.MX8QXP serial alias for lpuart ports.\n \n Signed-off-by: Fugang Duan \u003cfugang.duan@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit d04fd0b423f40643209e7ee9261692a32964326f\nAuthor: Fugang Duan \u003cfugang.duan@nxp.com\u003e\nDate: Mon Jul 22 16:28:23 2019 +0800\n\n arm64: dts: imx8qxp: add lpuart baud clock\n \n Add imx8qxp lpuart baud clock.\n \n Signed-off-by: Fugang Duan \u003cfugang.duan@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 3a3f06085fac791f3f949063c1da9fd244168ba1\nAuthor: Wen He \u003cwen.he_1@nxp.com\u003e\nDate: Fri Jul 19 17:59:56 2019 +0800\n\n arm64: dts: ls1028a: Add optional property node for Mali DP500\n \n This patch use the optional property node \"arm,malidp-arqos-value\" to\n can be dynamic configure QoS signaling.\n \n Signed-off-by: Wen He \u003cwen.he_1@nxp.com\u003e\n Reviewed-by: Liviu Dudau \u003cliviu.dudau@arm.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 56838644f9ddbaf8100051a7a7d5c5b0a028c3d0\nAuthor: Manivannan Sadhasivam \u003cmanivannan.sadhasivam@linaro.org\u003e\nDate: Fri Jul 19 12:39:26 2019 +0530\n\n arm64: dts: freescale: Add support for i.MX8QXP AI_ML board\n \n Add support for i.MX8QXP AI_ML board from Einfochips. This board is one\n of the Consumer Edition boards of the 96Boards family based on i.MX8QXP\n SoC from NXP/Freescale.\n \n The initial support includes following peripherals which are tested and\n known to be working:\n \n 1. Debug serial via UART2\n 2. uSD\n 3. WiFi\n 4. Ethernet\n \n More information about this board can be found in Arrow website:\n https://www.arrow.com/en/products/imx8-ai-ml/arrow-development-tools\n \n Signed-off-by: Manivannan Sadhasivam \u003cmanivannan.sadhasivam@linaro.org\u003e\n Reviewed-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit b5547e22be7298bbc8685c5becc7195be275e5d2\nAuthor: Fabio Estevam \u003cfestevam@gmail.com\u003e\nDate: Thu Jul 18 21:17:17 2019 -0300\n\n arm64: dts: imx8mm-evk: Remove invalid properties\n \n All these at803x properties are not documented anywhere, so\n just remove them.\n \n Signed-off-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit bf95c394f5a32ffc5aa1184de3aa398e1d9d574a\nAuthor: Pramod Kumar \u003cpramod.kumar_1@nxp.com\u003e\nDate: Tue Jul 16 13:43:31 2019 +0000\n\n arm64: dts: nxp: add ls1046a-frwy board support\n \n ls1046afrwy board is based on nxp ls1046a SoC.\n Board support\u0027s 4GB ddr memory, i2c, microSD card,\n serial console,qspi nor flash,ifc nand flash,qsgmii network interface,\n usb 3.0 and serdes interface to support two x1gen3 pcie interface.\n \n Signed-off-by: Vabhav Sharma \u003cvabhav.sharma@nxp.com\u003e\n Signed-off-by: Pramod Kumar \u003cpramod.kumar_1@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 9ae303f9edde61ebc9870b5a486dfbd46f1037b1\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Fri Jul 12 15:59:48 2019 +0200\n\n arm64: dts: rdu3: add UCS1002 charge controller node\n \n Add the charge controller node. With the controller driver loaded\n the VBUS of the user USB socket is controlled exclusively via i2c\n with the GPIO controls ignored, so vbus-supply for the user USB\n port must be linked to the charge controller.\n \n Hog the previously used GPIO control to unconditionally enable\n VBUS until the driver is loaded.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 74bd5951dd357c8cce6e74dccb182afb68bb5720\nAuthor: Li Jun \u003cjun.li@nxp.com\u003e\nDate: Wed Jul 10 19:19:17 2019 +0800\n\n arm64: dts: imx8mq: correct usb controller clocks\n \n The correct clock for \"bus_early\", \"ref\", \"suspend\" should be:\n IMX8MQ_CLK_USB1_CTRL_ROOT, IMX8MQ_CLK_USB_CORE_REF, IMX8MQ_CLK_32K,\n especially we may need the right suspend clock rate to set register\n in controller driver.\n \n Signed-off-by: Li Jun \u003cjun.li@nxp.com\u003e\n Reviewed-by: Abel Vesa \u003cabel.vesa@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 150736b88a7eab51b6950f179c0df4dedfe20bbf\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Fri Jul 5 12:56:12 2019 +0800\n\n arm64: dts: imx8mq: Add clock for TMU node\n \n i.MX8MQ has clock gate for TMU module, add clock info to TMU\n node for clock management.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 9eced3a2f224a62a233761e8af18c907c532e192\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Sat Jun 29 18:21:57 2019 +0800\n\n arm64: dts: imx8mq: Correct OPP table according to latest datasheet\n \n According to latest datasheet (Rev.1, 10/2018) from below links,\n in the consumer datasheet, 1.5GHz is mentioned as highest opp but\n depends on speed grading fuse, and in the industrial datasheet,\n 1.3GHz is mentioned as highest opp but depends on speed grading\n fuse. 1.5GHz and 1.3GHz opp use same voltage, so no need for\n consumer part to support 1.3GHz opp, with same voltage, CPU should\n run at highest frequency in order to go into idle as quick as\n possible, this can save power.\n \n That means for consumer part, 1GHz/1.5GHz are supported, for\n industrial part, 800MHz/1.3GHz are supported, and then check the\n speed grading fuse to limit the highest CPU frequency further.\n Correct the market segment bits in opp table to make them work\n according to datasheets.\n \n https://www.nxp.com/docs/en/data-sheet/IMX8MDQLQIEC.pdf\n https://www.nxp.com/docs/en/data-sheet/IMX8MDQLQCEC.pdf\n \n Fixes: 12629c5c3749 (\"arm64: dts: imx8mq: Add cpu speed grading and all OPPs\")\n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit cd7c2ddfda3629a18fcf6d20779bdcc73eda6bec\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Sat Jun 29 18:21:56 2019 +0800\n\n arm64: dts: imx8mm: Correct OPP table according to latest datasheet\n \n According to latest datasheet (Rev.0.2, 04/2019) from below links,\n 1.8GHz is ONLY available for consumer part, so the market segment\n bits for 1.8GHz opp should ONLY available for consumer part accordingly.\n \n https://www.nxp.com/docs/en/data-sheet/IMX8MMIEC.pdf\n https://www.nxp.com/docs/en/data-sheet/IMX8MMCEC.pdf\n \n Fixes: f403a26c865b (arm64: dts: imx8mm: Add cpu speed grading and all OPPs)\n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 46b29f4bd67e1007a2372abcfc48e41ab1d5bf7b\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Fri Jul 5 16:54:06 2019 +0800\n\n arm64: dts: imx8mm: Add \"fsl,imx8mq-src\" as src\u0027s fallback compatible\n \n i.MX8MM can reuse i.MX8MQ\u0027s src driver, add \"fsl,imx8mq-src\" as\n src\u0027s fallback compatible to enable it.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 1efe85c9056af0f20d8440a841bd89f4ec7ba735\nAuthor: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\nDate: Thu Jul 4 11:53:21 2019 +0300\n\n arm64: dts: imx8m: Add ddr-pmu nodes\n \n The same ddr perfomance counter IP from 8qxp is also available on imx8m\n series so add it to dts.\n \n Tested with `perf stat` and `memtester` on imx8mm-evk and obtained\n plausible results.\n \n Signed-off-by: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Acked-by: Frank Li \u003cfrank.li@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit ef4c47ab888af4da0a3b52562a3e60ab93e7bded\nAuthor: Li Jun \u003cjun.li@nxp.com\u003e\nDate: Wed Jul 3 15:23:27 2019 +0800\n\n arm64: dts: imx8mm-evk: enable usb1 and typec support\n \n USB1 port has typec connector with power delivery support:\n - Dual data role: host and device.\n - Dual power role: source and sink, prefer power sink.\n \n Signed-off-by: Li Jun \u003cjun.li@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 8b01840edcfcbc5d099996693f488875d0f6aba0\nAuthor: Li Jun \u003cjun.li@nxp.com\u003e\nDate: Wed Jul 3 15:23:26 2019 +0800\n\n arm64: dts: imx8mm: Remove setting for IMX8MM_CLK_USB_CORE_REF\n \n Since IMX8MM_CLK_USB_CORE_REF is not used at all, so remove the setting\n for it.\n \n Signed-off-by: Li Jun \u003cjun.li@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 156263599e9e8956af86ebeba35a73f367d40af7\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Tue Jul 2 09:44:00 2019 +0800\n\n arm64: dts: imx8mm: Add gpio-ranges property\n \n Add \"gpio-ranges\" property to establish connections between GPIOs\n and PINs on i.MX8MM pinctrl driver.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Abel Vesa \u003cabel.vesa@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 26c2f55a6a15153eed977e94dd42f8565b159526\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Tue Jul 2 09:43:59 2019 +0800\n\n arm64: dts: imx8mq: Add gpio-ranges property\n \n Add \"gpio-ranges\" property to establish connections between GPIOs\n and PINs on i.MX8MQ pinctrl driver.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Abel Vesa \u003cabel.vesa@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 6b392e16307f967c6922522b04a77c76174a4296\nAuthor: Abel Vesa \u003cabel.vesa@nxp.com\u003e\nDate: Wed Jul 3 15:08:22 2019 +0300\n\n arm64: dts: imx8mm: Init rates and parents configs for clocks\n \n Add the initial configuration for clocks that need default parent and rate\n setting. This is based on the vendor tree clock provider parents and rates\n configuration except this is doing the setup in dts rather than using clock\n consumer API in a clock provider driver.\n \n Signed-off-by: Abel Vesa \u003cabel.vesa@nxp.com\u003e\n Acked-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 3724ace582d9f675134985727fd5e9811f23c059\nAuthor: Robert Richter \u003crrichter@marvell.com\u003e\nDate: Mon Jun 24 15:08:55 2019 +0000\n\n EDAC/mc: Fix grain_bits calculation\n \n The grain in EDAC is defined as \"minimum granularity for an error\n report, in bytes\". The following calculation of the grain_bits in\n edac_mc is wrong:\n \n grain_bits \u003d fls_long(e-\u003egrain) + 1;\n \n Where grain_bits is defined as:\n \n grain \u003d 1 \u003c\u003c grain_bits\n \n Example:\n \n grain \u003d 8 # 64 bit (8 bytes)\n grain_bits \u003d fls_long(8) + 1\n grain_bits \u003d 4 + 1 \u003d 5\n \n grain \u003d 1 \u003c\u003c grain_bits\n grain \u003d 1 \u003c\u003c 5 \u003d 32\n \n Replace it with the correct calculation:\n \n grain_bits \u003d fls_long(e-\u003egrain - 1);\n \n The example gives now:\n \n grain_bits \u003d fls_long(8 - 1)\n grain_bits \u003d fls_long(7)\n grain_bits \u003d 3\n \n grain \u003d 1 \u003c\u003c 3 \u003d 8\n \n Also, check if the hardware reports a reasonable grain !\u003d 0 and fallback\n with a warning to 1 byte granularity otherwise.\n \n [ bp: massage a bit. ]\n \n Signed-off-by: Robert Richter \u003crrichter@marvell.com\u003e\n Signed-off-by: Borislav Petkov \u003cbp@suse.de\u003e\n Cc: \"linux-edac@vger.kernel.org\" \u003clinux-edac@vger.kernel.org\u003e\n Cc: James Morse \u003cjames.morse@arm.com\u003e\n Cc: Mauro Carvalho Chehab \u003cmchehab@kernel.org\u003e\n Cc: Tony Luck \u003ctony.luck@intel.com\u003e\n Link: https://lkml.kernel.org/r/20190624150758.6695-2-rrichter@marvell.com\n\ncommit d6781e490179f7ba710dd924187109da70c185b0\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Wed Jul 24 15:49:59 2019 +0200\n\n drm/pl111: Drop special pads config check\n \n This drops the check of the surplus \"pads\" configuration\n from the device tree that is completely unused in the DRM\n driver.\n \n This was only used to work around limitations in the earlier\n fbdev driver.\n \n Cc: Pawel Moll \u003cpawel.moll@arm.com\u003e\n Cc: Liviu Dudau \u003cLiviu.Dudau@arm.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190724134959.2365-2-linus.walleij@linaro.org\n\ncommit 1af824649b44382af2361eefc5bd50a92f6b3d1b\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Wed Jul 24 15:49:58 2019 +0200\n\n drm/pl111: Deprecate the pads from the DT binding\n \n The pads were an earlier workaround for the internal image\n pipeline in the Linux fbdev subsystem. As we move to generic\n definition of display properties and drivers that no longer\n need this to work, deprecate this property.\n \n Cc: Pawel Moll \u003cpawel.moll@arm.com\u003e\n Cc: Liviu Dudau \u003cLiviu.Dudau@arm.com\u003e\n Cc: devicetree@vger.kernel.org\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190724134959.2365-1-linus.walleij@linaro.org\n\ncommit f7988ba3b2ed3ec8e6c3623e281c278fea845bd8\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jul 24 15:50:17 2019 +0800\n\n clk: imx8mn: Keep uart clocks on for early console\n \n Call imx_register_uart_clocks() API to keep uart clocks enabled\n when earlyprintk or earlycon is active.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Abel Vesa \u003cabel.vesa@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 5133f1f3cc41bb3702dfd259bfc253dcb1bc915a\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jul 24 14:24:35 2019 +0800\n\n clk: imx: Remove unused function statement\n \n imx_register_uart_clocks_hws() function is NOT implemented\n at all, remove it.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Acked-by: Uwe Kleine-König \u003cu.kleine-koenig@pengutronix.de\u003e\n Reviewed-by: Abel Vesa \u003cabel.vesa@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit b87340998a87f338a0e1501c25ab650ce2c6ce95\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jul 24 11:06:00 2019 +0800\n\n clk: imx7ulp: Make sure earlycon\u0027s clock is enabled\n \n Earlycon\u0027s clock could be disabled during kernel boot up,\n if earlycon is enabled and its clock is gated, then kernel\n boot up will fail. Make sure earlycon\u0027s clock is enabled\n during kernel boot up.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit f27c3a359e6ed0f220fa767f6c965d581e7a7c6d\nAuthor: Fabio Estevam \u003cfestevam@gmail.com\u003e\nDate: Mon Jul 22 10:45:46 2019 -0300\n\n ARM: imx_v6_v7_defconfig: Select the OV5645 camera driver\n \n OV5645 camera sensor can be used on several i.MX boards, such as\n imx6qdl-wandboard, imx7d-pico, imx6ul-pico, etc.\n \n Select the OV5645 driver by default.\n \n Signed-off-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 8ad2d1dcce5493d422de9f38a83ad2e879f490c8\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Mon Jul 22 10:45:45 2019 -0300\n\n ARM: dts: imx6qdl-wandboard: Add OV5645 camera support\n \n imx6qdl-wandboard can be connected to a OV5645 camera via\n MIPI CSI port. Add support for it.\n \n PAD_GPIO_6 has been originally used for the Ethernet FEC\n ERR006687 workaround, but it needs to be used to provide the\n camera sensor clock, so adjust it accordingly.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 77266e722feabb6eefc8a7e84ac2415837d91c5f\nAuthor: Sven Van Asbroeck \u003cthesven73@gmail.com\u003e\nDate: Fri Jul 12 16:43:15 2019 -0400\n\n bus: imx-weim: optionally enable burst clock mode\n \n To enable burst clock mode, add the fsl,burst-clk-enable\n property to the weim bus\u0027s devicetree node.\n \n Example:\n weim: weim@21b8000 {\n compatible \u003d \"fsl,imx6q-weim\";\n reg \u003d \u003c0x021b8000 0x4000\u003e;\n clocks \u003d \u003c\u0026clks 196\u003e;\n #address-cells \u003d \u003c2\u003e;\n #size-cells \u003d \u003c1\u003e;\n ranges \u003d \u003c0 0 0x08000000 0x08000000\u003e;\n fsl,weim-cs-gpr \u003d \u003c\u0026gpr\u003e;\n fsl,burst-clk-enable;\n \n client-device@0,0 {\n compatible \u003d \"something\";\n reg \u003d \u003c0 0 0x02000000\u003e;\n #address-cells \u003d \u003c1\u003e;\n #size-cells \u003d \u003c1\u003e;\n bank-width \u003d \u003c2\u003e;\n fsl,weim-cs-timing \u003d \u003c0x00620081 0x00000001 0x1c022000\n 0x0000c000 0x1404a38e 0x00000000\u003e;\n };\n };\n \n Signed-off-by: Sven Van Asbroeck \u003cTheSven73@gmail.com\u003e\n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 0008c1f3bc274fcea3b9654a07718d6012773e25\nAuthor: Sven Van Asbroeck \u003cthesven73@gmail.com\u003e\nDate: Fri Jul 12 16:43:16 2019 -0400\n\n dt-bindings: bus: imx-weim: document optional burst clock mode\n \n An optional devicetree property was added to the imx-weim driver,\n which if present instructs it to operate in burst clock mode.\n Update the dt-bindings to reflect this.\n \n Signed-off-by: Sven Van Asbroeck \u003cTheSven73@gmail.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit af7e7ee0e4280c29c41b6ec64b892bb53987a997\nAuthor: Abel Vesa \u003cabel.vesa@nxp.com\u003e\nDate: Tue Jul 9 17:20:03 2019 +0300\n\n clk: imx8mm: Switch to platform driver\n \n There is no strong reason for this to use CLK_OF_DECLARE instead\n of being a platform driver. Plus, this will now be aligned with the\n other i.MX8M clock drivers which are platform drivers.\n \n In order to make the clock provider a platform driver\n all the data and code needs to be outside of .init section.\n \n Signed-off-by: Abel Vesa \u003cabel.vesa@nxp.com\u003e\n Acked-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 053a4ffe298836bb973d2cba59f82fff60c7db5b\nAuthor: Peng Fan \u003cpeng.fan@nxp.com\u003e\nDate: Mon Jul 15 02:55:43 2019 +0000\n\n clk: imx: imx8mm: fix audio pll setting\n \n The AUDIO PLL max support 650M, so the original clk settings violate\n spec. This patch makes the output 786432000 -\u003e 393216000,\n and 722534400 -\u003e 361267200 to aligned with NXP vendor kernel without any\n impact on audio functionality and go within 650MHz PLL limit.\n \n Cc: \u003cstable@vger.kernel.org\u003e\n Fixes: ba5625c3e272 (\"clk: imx: Add clock driver support for imx8mm\")\n Signed-off-by: Peng Fan \u003cpeng.fan@nxp.com\u003e\n Acked-by: Abel Vesa \u003cabel.vesa@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit d4c5792e2a4afbde62c7e8b38025351f173d30cd\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jun 26 09:28:03 2019 +0800\n\n clk: imx8mm: GPT1 clock mux option #5 should be sys_pll1_80m\n \n i.MX8MM\u0027s GPT1 clock mux option #5 should be sys_pll1_80m,\n NOT sys_pll1_800m, correct it.\n \n Fixes: ba5625c3e272 (\"clk: imx: Add clock driver support for imx8mm\")\n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit d52fb0193ffb78a4a21a095bd9dbf6ec4058bd71\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jun 26 09:28:02 2019 +0800\n\n clk: imx8mm: Fix typo of pwm3 clock\u0027s mux option #4\n \n i.MX8MM has no sys3_pll2_out clock, PWM3 clock\u0027s mux option #4\n should be sys_pll3_out, sys3_pll2_out is a typo, fix it.\n \n Fixes: ba5625c3e272 (\"clk: imx: Add clock driver support for imx8mm\")\n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit e50bf7a67cc66b245ae6a8d062982e64e9dbebda\nAuthor: Abel Vesa \u003cabel.vesa@nxp.com\u003e\nDate: Thu Jul 11 18:11:50 2019 +0300\n\n clk: imx: Remove unused clk based API\n \n Now that the i.MX6 and i.MX7 clock drivers have been switched to clk_hw based,\n we can remove the clk based API that is not used by any i.MX clock driver.\n \n The following APIs are going away now:\n - imx_clk_busy_divider\n - imx_clk_busy_mux\n - imx_clk_fixup_divider\n - imx_clk_fixup_mux\n - imx_clk_mux_ldb\n - imx_clk_gate_dis_flags\n - imx_clk_gate_flags\n \n Signed-off-by: Abel Vesa \u003cabel.vesa@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 0d381f4c80ee1d885cba14f1004bdc6f17507b63\nAuthor: Li Jun \u003cjun.li@nxp.com\u003e\nDate: Wed Jul 10 19:19:16 2019 +0800\n\n clk: imx8mq: set correct parent for usb ctrl clocks\n \n Per latest imx8mq datasheet of CCM, the parent of usb1_ctrl_root_clk\n and usb2_ctrl_root_clk is usb_bus.\n \n Signed-off-by: Li Jun \u003cjun.li@nxp.com\u003e\n Reviewed-by: Abel Vesa \u003cabel.vesa@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 6ad7cb7122cee4d3e672e58bf542b070b9363c15\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Tue Jul 2 18:20:07 2019 +0300\n\n clk: imx8: Add DSP related clocks\n \n i.MX8QXP contains Hifi4 DSP. There are four clocks\n associated with DSP:\n * dsp_lpcg_core_clk\n * dsp_lpcg_ipg_clk\n * dsp_lpcg_adb_aclk\n * ocram_lpcg_ipg_clk\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Reviewed-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 951c1aef9691491ddf4dd5aab76f2665d56bd5d3\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Fri Jul 5 12:56:11 2019 +0800\n\n clk: imx8mq: Remove CLK_IS_CRITICAL flag for IMX8MQ_CLK_TMU_ROOT\n \n IMX8MQ_CLK_TMU_ROOT is ONLY used for thermal module, the driver\n should manage this clock, so no need to have CLK_IS_CRITICAL flag\n set.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Abel Vesa \u003cabel.vesa@nxp.com\u003e\n Acked-by: Stephen Boyd \u003csboyd@kernel.org\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 96d6392b54dbb1ff2b39448a2516fa6abb33114b\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jun 19 13:52:46 2019 +0800\n\n clk: imx: Add support for i.MX8MN clock driver\n \n This patch adds i.MX8MN clock driver support.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 0209001674de31177aa10dce47843cc6bc339bb9\nAuthor: Fancy Fang \u003cchen.fang@nxp.com\u003e\nDate: Tue Jul 9 07:18:01 2019 +0000\n\n clk: imx8mm: rename \u0027share_count_dcss\u0027 to \u0027share_count_disp\u0027\n \n Rename \u0027share_count_dcss\u0027 to \u0027share_count_disp\u0027, since the\n DCSS module does not exist on imx8mm platform. So rename it\n to avoid any unnecessary confusion.\n \n Signed-off-by: Fancy Fang \u003cchen.fang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit d1b395b36e907f7ff0b06deefc7ba163ea7bc398\nAuthor: Li Jun \u003cjun.li@nxp.com\u003e\nDate: Wed Jul 3 15:23:25 2019 +0800\n\n clk: imx8mm: correct the usb1_ctrl parent to be usb_bus\n \n Per latest imx8mm datasheet of CCM, the parent of usb1_ctrl_root_clk\n should be usb_bus.\n \n Signed-off-by: Li Jun \u003cjun.li@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit fd6ef2857b7484ab947beb2f2a1bbb3662553298\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jun 19 13:52:45 2019 +0800\n\n clk: imx: Add API for clk unregister when driver probe fail\n \n For i.MX clock drivers probe fail case, clks should be unregistered\n in the return path, this patch adds a common API for i.MX clock\n drivers to unregister clocks when fail.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 34af5179d239e42698e2aff0be94de2aeac9a733\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jun 19 13:52:44 2019 +0800\n\n clk: imx8mm: Make 1416X/1443X PLL macro definitions common for usage\n \n 1416X/1443X PLL are used on i.MX8MM and i.MX8MN and maybe\n other i.MX8M series SoC later, the macro definitions of\n these PLLs\u0027 initialization should be common for usage.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 1e80936a42e16b7a7c4265de666c472256b7cf36\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jun 19 13:52:43 2019 +0800\n\n dt-bindings: imx: Add clock binding doc for i.MX8MN\n \n Add the clock binding doc for i.MX8MN.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 9fd827c699aaaac8521a3246638850913d140854\nAuthor: Marco Villegas \u003cgit@marvil07.net\u003e\nDate: Fri Aug 2 16:47:37 2019 -0300\n\n staging: rtl8712: Add spaces around \u003c\u003c\n \n Fix checkpatch error \"CHECK: spaces preferred around that \u0027\u003c\u003c\u0027\".\n \n Signed-off-by: Marco Villegas \u003cgit@marvil07.net\u003e\n Link: https://lore.kernel.org/r/20190802194737.12252-1-git@marvil07.net\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7cbf0108247da2001a591583d2fc6433918b8a47\nAuthor: Thiago Bonotto \u003cthbonotto@gmail.com\u003e\nDate: Fri Aug 2 20:23:23 2019 +0000\n\n staging: isdn: remove unnecessary parentheses\n \n Fix the following checkpatch error:\n \n ERROR: return is not a function, parentheses are not required\n FILE: drivers/staging/isdn/hysdn/hysdn_net.c:289:\n + return (0); /* and return success */\n \n Signed-off-by: Thiago Bonotto \u003cthbonotto@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190802202323.27117-1-thbonotto@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 8ca9d8ad30f39651d6ff9428aaeb1940489a7754\nAuthor: Saurav Girepunje \u003csaurav.girepunje@gmail.com\u003e\nDate: Sat Aug 3 12:14:12 2019 +0530\n\n staging: rtl8188eu: core: rtw_recv.c: Remove Unnecessary parentheses\n \n Remove Unnecessary parentheses around precvframe-\u003elist,\n pmlmeext-\u003emlmext_info and precvpriv-\u003efree_recv_queue\n \n Signed-off-by: Saurav Girepunje \u003csaurav.girepunje@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190803064408.GA30111@saurav\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 83af58f8068ea3f7b3c537c37a30887bfa585069\nAuthor: Helge Deller \u003cdeller@gmx.de\u003e\nDate: Wed Feb 6 23:21:10 2019 +0100\n\n parisc: Add assembly implementations for memset, strlen, strcpy, strncpy and strcat\n \n Add performance-optimized versions of some string functions.\n \n Signed-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n Tested-by: Sven Schnelle \u003csvens@stackframe.org\u003e\n\ncommit ec4d396b635070e0caf5888d58cb9eedc8dd73d9\nAuthor: Sven Schnelle \u003csvens@stackframe.org\u003e\nDate: Thu Aug 1 22:08:42 2019 +0200\n\n parisc: trigger die notifier chain in parisc_terminate()\n \n This will trigger kgdb/kdb when they are enabled.\n \n Signed-off-by: Sven Schnelle \u003csvens@stackframe.org\u003e\n Signed-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n\ncommit c8bbbc67cb314981e7ba22af0bd703225f0add80\nAuthor: Sven Schnelle \u003csvens@stackframe.org\u003e\nDate: Tue Jul 23 22:37:53 2019 +0200\n\n parisc: Update feature list\n \n Add jump labels to the list of supported features.\n \n Signed-off-by: Sven Schnelle \u003csvens@stackframe.org\u003e\n Signed-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n\ncommit 52a22e6c27be9089fdd9c4e1857939f0dec8f57c\nAuthor: Sven Schnelle \u003csvens@stackframe.org\u003e\nDate: Tue Jul 23 22:37:52 2019 +0200\n\n parisc/ftrace: Add KPROBES_ON_FTRACE\n \n Allow KPROBES to use the ftrace infrastructure on PA-RISC.\n \n Signed-off-by: Sven Schnelle \u003csvens@stackframe.org\u003e\n Signed-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n\ncommit d562aca37a543cc3de85be6a6946771a27d50c6a\nAuthor: Sven Schnelle \u003csvens@stackframe.org\u003e\nDate: Tue Jul 23 22:37:51 2019 +0200\n\n parisc/ftrace: Add ARCH_SUPPORTS_FTRACE_OPS support\n \n Pass ftrace_ops to ftrace functions to ftrace_trace_function().\n \n Signed-off-by: Sven Schnelle \u003csvens@stackframe.org\u003e\n Signed-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n\ncommit 8488d8e90c03983912318deaae41546f950667d6\nAuthor: Claudiu Manoil \u003cclaudiu.manoil@nxp.com\u003e\nDate: Thu Aug 1 14:52:53 2019 +0300\n\n arm64: dts: fsl: ls1028a: Enable eth port1 on the ls1028a QDS board\n \n LS1028a has one Ethernet management interface. On the QDS board, the\n MDIO signals are multiplexed to either on-board AR8035 PHY device or\n to 4 PCIe slots allowing for SGMII cards.\n To enable the Ethernet ENETC Port 1, which can only be connected to a\n RGMII PHY, the multiplexer needs to be configured to route the MDIO to\n the AR8035 PHY. The MDIO/MDC routing is controlled by bits 7:4 of FPGA\n board config register 0x54, and value 0 selects the on-board RGMII PHY.\n The FPGA board config registers are accessible on the i2c bus, at address\n 0x66.\n \n The PF3 MDIO PCIe integrated endpoint device allows for centralized access\n to the MDIO bus. Add the corresponding devicetree node and set it to be\n the MDIO bus parent.\n \n Signed-off-by: Alex Marginean \u003calexandru.marginean@nxp.com\u003e\n Signed-off-by: Claudiu Manoil \u003cclaudiu.manoil@nxp.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 288a91d5cd0c947cfd0cf8430f588974ea697c5e\nAuthor: Claudiu Manoil \u003cclaudiu.manoil@nxp.com\u003e\nDate: Thu Aug 1 14:52:52 2019 +0300\n\n dt-bindings: net: fsl: enetc: Add bindings for the central MDIO PCIe endpoint\n \n The on-chip PCIe root complex that integrates the ENETC ethernet\n controllers also integrates a PCIe endpoint for the MDIO controller\n providing for centralized control of the ENETC mdio bus.\n Add bindings for this \"central\" MDIO Integrated PCIe Endpoint.\n \n Signed-off-by: Claudiu Manoil \u003cclaudiu.manoil@nxp.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 231ece36f50df5d0d648011c58d4255d112a8bbf\nAuthor: Claudiu Manoil \u003cclaudiu.manoil@nxp.com\u003e\nDate: Thu Aug 1 14:52:51 2019 +0300\n\n enetc: Add mdio bus driver for the PCIe MDIO endpoint\n \n ENETC ports can manage the MDIO bus via local register\n interface. However there\u0027s also a centralized way\n to manage the MDIO bus, via the MDIO PCIe endpoint\n device integrated by the same root complex that also\n integrates the ENETC ports (eth controllers).\n \n Depending on board design and use case, centralized\n access to MDIO may be better than using local ENETC\n port registers. For instance, on the LS1028A QDS board\n where MDIO muxing is required. Also, the LS1028A on-chip\n switch doesn\u0027t have a local MDIO register interface.\n \n The current patch registers the above PCIe endpoint as a\n separate MDIO bus and provides a driver for it by re-using\n the code used for local MDIO access. It also allows the\n ENETC port PHYs to be managed by this driver if the local\n \"mdio\" node is missing from the ENETC port node.\n \n Signed-off-by: Claudiu Manoil \u003cclaudiu.manoil@nxp.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 0c010a9deb33179169fa867d3c76833ce80165b7\nAuthor: Claudiu Manoil \u003cclaudiu.manoil@nxp.com\u003e\nDate: Thu Aug 1 14:52:50 2019 +0300\n\n enetc: Clean up makefile\n \n Clean up overcomplicated makefile to make it more maintainable.\n Basically, there\u0027s a set of common objects shared between\n the PF and VF driver modules. This can be implemented in a\n simpler way, without conditionals, less repetition, allowing\n also for easier updates in the future.\n \n Signed-off-by: Claudiu Manoil \u003cclaudiu.manoil@nxp.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 2152e7a2d9a8c29e8850adabd2175fa042ad5bb8\nAuthor: Claudiu Manoil \u003cclaudiu.manoil@nxp.com\u003e\nDate: Thu Aug 1 14:52:49 2019 +0300\n\n enetc: Clean up local mdio bus allocation\n \n What\u0027s needed is basically a pointer to the mdio registers.\n This is one way to store it inside bus-\u003epriv allocated space,\n without upsetting sparse.\n Reworked accessors to avoid __iomem casting.\n Used devm_* variant to further clean up the init error /\n remove paths.\n \n Fixes following sparse warning:\n warning: incorrect type in assignment (different address spaces)\n expected void *priv\n got struct enetc_mdio_regs [noderef] \u003casn:2\u003e*[assigned] regs\n \n Fixes: ebfcb23d62ab (\"enetc: Add ENETC PF level external MDIO support\")\n \n Signed-off-by: Claudiu Manoil \u003cclaudiu.manoil@nxp.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7150961487c5b4521ef5b6557373546e370709d2\nAuthor: Hubert Feurstein \u003ch.feurstein@gmail.com\u003e\nDate: Wed Jul 31 10:23:51 2019 +0200\n\n net: dsa: mv88e6xxx: add PTP support for MV88E6250 family\n \n This adds PTP support for the MV88E6250 family.\n \n Signed-off-by: Hubert Feurstein \u003ch.feurstein@gmail.com\u003e\n Acked-by: Richard Cochran \u003crichardcochran@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 8858ccc837e6e89c917f4b4bb1d7335d62e1baab\nAuthor: Hubert Feurstein \u003ch.feurstein@gmail.com\u003e\nDate: Wed Jul 31 10:23:50 2019 +0200\n\n net: dsa: mv88e6xxx: order ptp structs numerically ascending\n \n As it is done for all the other structs within this driver.\n \n Signed-off-by: Hubert Feurstein \u003ch.feurstein@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 121b8fe2fdc931a60f7437a94db3b8af8c62ee54\nAuthor: Hubert Feurstein \u003ch.feurstein@gmail.com\u003e\nDate: Wed Jul 31 10:23:49 2019 +0200\n\n net: dsa: mv88e6xxx: setup message port is not supported in the 6250 familiy\n \n The MV88E6250 family doesn\u0027t support the MV88E6XXX_PORT_CTL1_MESSAGE_PORT\n bit.\n \n Signed-off-by: Hubert Feurstein \u003ch.feurstein@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c857486a4b6d5c9250832315341e5889fdcb03c3\nAuthor: Hubert Feurstein \u003ch.feurstein@gmail.com\u003e\nDate: Wed Jul 31 10:23:48 2019 +0200\n\n net: dsa: mv88e6xxx: introduce invalid_port_mask in mv88e6xxx_info\n \n With this it is possible to mark certain chip ports as invalid. This is\n required for example for the MV88E6220 (which is in general a MV88E6250\n with 7 ports) but the ports 2-4 are not routed to pins.\n \n If a user configures an invalid port, an error is returned.\n \n Signed-off-by: Hubert Feurstein \u003ch.feurstein@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 83c5ee315f73048d0c06c5736eb8ffe4ded2b7cc\nAuthor: Hubert Feurstein \u003ch.feurstein@gmail.com\u003e\nDate: Wed Jul 31 10:23:47 2019 +0200\n\n dt-bindings: net: dsa: marvell: add 6220 model to the 6250 family\n \n The MV88E6220 is part of the MV88E6250 family.\n \n Signed-off-by: Hubert Feurstein \u003ch.feurstein@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4902264745b524ae61d5d2df379d5c74577af823\nAuthor: Hubert Feurstein \u003ch.feurstein@gmail.com\u003e\nDate: Wed Jul 31 10:23:46 2019 +0200\n\n net: dsa: mv88e6xxx: add support for MV88E6220\n \n The MV88E6220 is almost the same as MV88E6250 except that the ports 2-4 are\n not routed to pins. So the usable ports are 0, 1, 5 and 6.\n \n Signed-off-by: Hubert Feurstein \u003ch.feurstein@gmail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 82f151def2153f34a1f6f58499f22ceb2bc94042\nAuthor: Andrew Jeffery \u003candrew@aj.id.au\u003e\nDate: Wed Jul 31 15:09:59 2019 +0930\n\n net: ftgmac100: Select ASPEED MDIO driver for the AST2600\n \n Ensures we can talk to a PHY via MDIO on the AST2600, as the MDIO\n controller is now separate from the MAC.\n \n Signed-off-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 39bfab8844a0fabea812f99dc6aa88734323a920\nAuthor: Andrew Jeffery \u003candrew@aj.id.au\u003e\nDate: Wed Jul 31 15:09:58 2019 +0930\n\n net: ftgmac100: Add support for DT phy-handle property\n \n phy-handle is necessary for the AST2600 which separates the MDIO\n controllers from the MAC.\n \n I\u0027ve tried to minimise the intrusion of supporting the AST2600 to the\n FTGMAC100 by leaving in place the existing MDIO support for the embedded\n MDIO interface. The AST2400 and AST2500 continue to be supported this\n way, as it avoids breaking/reworking existing devicetrees.\n \n The AST2600 support by contrast requires the presence of the phy-handle\n property in the MAC devicetree node to specify the appropriate PHY to\n associate with the MAC. In the event that someone wants to specify the\n MDIO bus topology under the MAC node on an AST2400 or AST2500, the\n current auto-probe approach is done conditional on the absence of an\n \"mdio\" child node of the MAC.\n \n Signed-off-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f160e99462c68ab5b9e2b9097a4867459730b49a\nAuthor: Andrew Jeffery \u003candrew@aj.id.au\u003e\nDate: Wed Jul 31 15:09:57 2019 +0930\n\n net: phy: Add mdio-aspeed\n \n The AST2600 design separates the MDIO controllers from the MAC, which is\n where they were placed in the AST2400 and AST2500. Further, the register\n interface is reworked again, so now we have three possible different\n interface implementations, however this driver only supports the\n interface provided by the AST2600. The AST2400 and AST2500 will continue\n to be supported by the MDIO support embedded in the FTGMAC100 driver.\n \n The hardware supports both C22 and C45 mode, but for the moment only C22\n support is implemented.\n \n Signed-off-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 94166fd21af318be13aa9b66c4de381a4817ed3d\nAuthor: Andrew Jeffery \u003candrew@aj.id.au\u003e\nDate: Wed Jul 31 15:09:56 2019 +0930\n\n dt-bindings: net: Add aspeed, ast2600-mdio binding\n \n The AST2600 splits out the MDIO bus controller from the MAC into its own\n IP block and rearranges the register layout. Add a new binding to\n describe the new hardware.\n \n Signed-off-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 515b8b7e935ef3f59c4efda04a3b05353ed6fbb7\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 2 22:21:37 2019 +0100\n\n drm/i915: Flush the freed object list on file close\n \n As we increase the number of RCU objects, it becomes easier for us to\n have several hundred thousand objects in the deferred RCU free queues.\n An example is gem_ctx_create/files which continually creates active\n contexts, which are not immediately freed upon close as they are kept\n alive by outstanding requests. This lack of backpressure allows the\n context objects to persist until they overwhelm and starve the system.\n We can increase our backpressure by flushing the freed object queue upon\n closing the device fd which should then not impact other clients.\n \n Testcase: igt/gem_ctx_create/*files\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190802212137.22207-2-chris@chris-wilson.co.uk\n\ncommit 1aff1903d0ff53f055088a77948ac8d8224d42db\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 2 22:21:36 2019 +0100\n\n drm/i915: Hide unshrinkable context objects from the shrinker\n \n The shrinker cannot touch objects used by the contexts (logical state\n and ring). Currently we mark those as \"pin_global\" to let the shrinker\n skip over them, however, if we remove them from the shrinker lists\n entirely, we don\u0027t event have to include them in our shrink accounting.\n \n By keeping the unshrinkable objects in our shrinker tracking, we report\n a large number of objects available to be shrunk, and leave the shrinker\n deeply unsatisfied when we fail to reclaim those. The shrinker will\n persist in trying to reclaim the unavailable objects, forcing the system\n into a livelock (not even hitting the dread oomkiller).\n \n v2: Extend unshrinkable protection for perma-pinned scratch and guc\n allocations (Tvrtko)\n v3: Notice that we should be pinned when marking unshrinkable and so the\n link cannot be empty; merge duplicate paths.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190802212137.22207-1-chris@chris-wilson.co.uk\n\ncommit 6f79fcc6c9e311d489989e0909d86c0762d3ac61\nAuthor: Matthias Kaehlcke \u003cmka@chromium.org\u003e\nDate: Thu Aug 1 15:03:54 2019 -0700\n\n ARM: dts: rockchip: Add pin names for rk3288-veyron fievel\n \n This is like commit 0ca87bd5baa6 (\"ARM: dts: rockchip: Add pin names\n for rk3288-veyron-jerry\") and other similar commits, but for the\n veyron fievel board (and tiger, which includes the fievel .dtsi).\n \n Signed-off-by: Matthias Kaehlcke \u003cmka@chromium.org\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit 94562040da08611ca15eecb39c68be8caac4bf06\nAuthor: Matthias Kaehlcke \u003cmka@chromium.org\u003e\nDate: Wed Jul 31 08:15:27 2019 -0700\n\n ARM: dts: rockchip: A few fixes for veyron-{fievel,tiger}\n \n Fix/improve a few things for veyron fievel/tiger:\n \n - move \u0027vccsys\u0027 regulator from tiger to fievel, both boards\n have it (and tiger includes the fievel .dtsi)\n - move \u0027ext_gmac\u0027 node below regulators\n - fix GPIO ids of vcc5_host1 and vcc5_host2 regulators\n - remove reset configuration from \u0027gmac\u0027 node, this is already done\n in rk3288.dtsi\n - fixed style issues of some multi-line comments\n - switch \u0027vcc18_lcdt\u0027, \u0027vdd10_lcd\u0027 and \u0027vcc33_ccd\u0027 regulators off\n during suspend\n - no pull-up on the Bluetooth wake-up pin, there is an external\n pull-up. The signal is active low, add the \u0027bt_host_wake_l\u0027\n pinctrl config\n - move BC 1.2 pins up in the pinctrl config to keep \u0027wake only\u0027 pins\n separate\n - add BC 1.2 pins to sleep config\n \n Fixes: 0067692b662e (\"ARM: dts: rockchip: add veyron-fievel board\")\n Reviewed-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n Signed-off-by: Matthias Kaehlcke \u003cmka@chromium.org\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit 1d1def89bd2ed573d5978f39545596ab3f3d2c8b\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Jul 19 03:24:14 2019 +0000\n\n pinctrl: sprd: Fix platform_no_drv_owner.cocci warnings\n \n Remove .owner field if calls are used which set it automatically\n Generated by: scripts/coccinelle/api/platform_no_drv_owner.cocci\n \n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190719032414.85369-1-yuehaibing@huawei.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 7bdbd1eccc9fb3875f33af1a2257001128824e81\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Thu Jul 18 15:51:01 2019 +0900\n\n gpio: refactor gpiochip_allocate_mask() with bitmap_alloc()\n \n Refactor gpiochip_allocate_mask() slightly by using bitmap_alloc().\n \n I used bitmap_free() for the corresponding free parts. Actually,\n bitmap_free() is a wrapper of kfree(), but I did this for consistency.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Link: https://lore.kernel.org/r/20190718065101.26994-1-yamada.masahiro@socionext.com\n Reviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit bd4bd337039df9d5c6a3c11c439511fd99e11f52\nAuthor: Song Hui \u003chui.song_1@nxp.com\u003e\nDate: Thu Jul 18 17:49:02 2019 +0800\n\n gpio: mpc8xxx: Add ls1028a device specify function.\n \n There is a device specify register(named GPIO_IBE)\n on ls1028a need to enable in initial stage.\n \n Signed-off-by: Song Hui \u003chui.song_1@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190718094902.15562-2-hui.song_1@nxp.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 6bd0fbe156f1cc5dc97590ce994d1848f593569e\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Fri Aug 2 18:40:55 2019 +0000\n\n drm/i915/wopcm: Don\u0027t fail on WOPCM partitioning failure\n \n We don\u0027t have to immediately fail on WOPCM partitioning, we can wait\n until we will start programming WOPCM registers. This should give us\n more options if we decide to restore fallback in case of GuC failures.\n \n v3: rebased\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190802184055.31988-7-michal.wajdeczko@intel.com\n\ncommit 5d1ef2b4270de45e1b1b40a00838e3b6196eefd7\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Fri Aug 2 18:40:54 2019 +0000\n\n drm/i915/uc: Inject probe errors into intel_uc_init_hw\n \n Inject probe errors into intel_uc_init_hw to make sure we\n correctly handle any uC initialization failure.\n \n To avoid complains from CI about injected errors use\n i915_probe_error to lower message level.\n \n v4: rebased after moving hot fixes moved to separate patches\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e #v1\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190802184055.31988-6-michal.wajdeczko@intel.com\n\ncommit 32ff76e80c2400c511b37a498eead9caba5656b8\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Fri Aug 2 18:40:53 2019 +0000\n\n drm/i915/uc: Move GuC error log to uc and release it on fini\n \n When we fail to load GuC and want to abort probe, we hit:\n \n \u003c7\u003e [229.915779] i915 0000:00:02.0: [drm:intel_uc_init_hw [i915]] GuC initialization failed -6\n \u003c7\u003e [229.915813] i915 0000:00:02.0: [drm:i915_gem_init_hw [i915]] Enabling uc failed (-6)\n \u003c4\u003e [229.953354] ------------[ cut here ]------------\n \u003c4\u003e [229.953355] WARN_ON(dev_priv-\u003emm.shrink_count)\n \u003c4\u003e [229.953406] WARNING: CPU: 9 PID: 3287 at drivers/gpu/drm/i915/i915_gem.c:1684 i915_gem_cleanup_early+0xfc/0x110 [i915]\n \u003c4\u003e [229.953464] Call Trace:\n \u003c4\u003e [229.953489] i915_driver_late_release+0x19/0x60 [i915]\n \u003c4\u003e [229.953514] i915_driver_probe+0xb82/0x18a0 [i915]\n \u003c4\u003e [229.953519] ? __pm_runtime_resume+0x4f/0x80\n \u003c4\u003e [229.953545] i915_pci_probe+0x43/0x1b0 [i915]\n ...\n \u003c4\u003e [229.962951] ------------[ cut here ]------------\n \u003c4\u003e [229.962956] DEBUG_LOCKS_WARN_ON(lock-\u003emagic !\u003d lock)\n \u003c4\u003e [229.962959] WARNING: CPU: 8 PID: 2395 at kernel/locking/mutex.c:912 __mutex_lock+0x750/0x9b0\n \u003c4\u003e [229.963091] Call Trace:\n \u003c4\u003e [229.963129] ? i915_vma_destroy+0x86/0x350 [i915]\n \u003c4\u003e [229.963166] ? i915_vma_destroy+0x86/0x350 [i915]\n \u003c4\u003e [229.963201] i915_vma_destroy+0x86/0x350 [i915]\n \u003c4\u003e [229.963236] __i915_gem_free_objects+0xb8/0x510 [i915]\n \u003c4\u003e [229.963270] __i915_gem_free_work+0x5a/0x90 [i915]\n \u003c4\u003e [229.963275] process_one_work+0x245/0x610\n \n as since commit 6f76098fe0f3 (\"drm/i915/uc: Move uC early functions\n inside the GT ones\") we cleanup uc after gem.\n \n Move captured GuC load error log to uc struct and release it\n in intel_uc_fini() instead of intel_uc_driver_late_release()\n \n Note that intel_uc_driver_late_release() is now empty, but\n we can leave it as a placeholder for future code.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190802184055.31988-5-michal.wajdeczko@intel.com\n\ncommit 3243bd096bb1656dfab7a1f3699d4d94692fd14c\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Fri Aug 2 18:40:52 2019 +0000\n\n drm/i915/uc: Reorder firmware status codes\n \n On Gen9 when we try to reload HuC due to GuC upload error, we hit:\n \n \u003c7\u003e [232.025927] [drm:intel_uc_init_hw [i915]] GuC fw load failed: -8; will reset and retry 2 more time(s)\n \u003c7\u003e [232.026004] [drm:intel_uc_fw_upload [i915]] HuC fw load i915/kbl_huc_ver02_00_1810.bin\n \u003c7\u003e [232.026686] [drm:intel_uc_fw_upload [i915]] HuC fw xfer completed\n \u003c6\u003e [232.026688] [drm] HuC: Loaded firmware i915/kbl_huc_ver02_00_1810.bin (version 2.0)\n \u003c3\u003e [232.026703] intel_uc_fw_copy_rsa:541 GEM_BUG_ON(!intel_uc_fw_is_available(uc_fw))\n \n as firmware that previously failed to load was wrongly treated as\n unavailable since its status code was not matching status check logic.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190802184055.31988-4-michal.wajdeczko@intel.com\n\ncommit 771051eaa74661f63d893589137f5d70515266c0\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Fri Aug 2 18:40:51 2019 +0000\n\n drm/i915/uc: Do full sanitize instead of pure reset\n \n On Gen9 when we try to reload HuC due to GuC upload error, we hit:\n \n \u003c7\u003e [229.656688] [drm:intel_uc_init_hw [i915]] GuC fw load failed: -8; will reset and retry 2 more time(s)\n \u003c7\u003e [229.656739] [drm:intel_uc_fw_upload [i915]] HuC fw load i915/kbl_huc_ver02_00_1810.bin\n \u003c3\u003e [229.656740] intel_uc_fw_upload:425 GEM_BUG_ON(intel_uc_fw_is_loaded(uc_fw))\n \n as we performed only pure reset and didn\u0027t sanitized HuC fw status.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190802184055.31988-3-michal.wajdeczko@intel.com\n\ncommit 50d84418f58699272a9220f9447bf89f0440c671\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Fri Aug 2 18:40:50 2019 +0000\n\n drm/i915: Add i915 to i915_inject_probe_failure\n \n With i915 added to i915_inject_probe_failure we can use dedicated\n printk when injecting artificial load failure.\n \n Also make this function look like other i915 functions that return\n error code and make it more flexible to return any provided error\n code instead of previously assumed -ENODEV.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190802184055.31988-2-michal.wajdeczko@intel.com\n\ncommit 31d0e6c149b8c9a9bddc6d68f8600918bb771cb9\nAuthor: Joe Perches \u003cjoe@perches.com\u003e\nDate: Thu Nov 1 00:24:08 2018 -0700\n\n mlx5: Fix formats with line continuation whitespace\n \n The line continuations unintentionally add whitespace so\n instead use coalesced formats to remove the whitespace.\n \n Signed-off-by: Joe Perches \u003cjoe@perches.com\u003e\n Reviewed-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Link: https://lore.kernel.org/r/f14db3287b23ed8af9bdbf8001e2e2fe7ae9e43a.camel@perches.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 2e737e520548d4456fc8dfa800ab189e0e4b2939\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Thu Jul 4 12:43:37 2019 +0200\n\n drm/etnaviv: clean up includes\n \n Drop unused includes, move more includes from the generic etnaviv_drv.h to\n the units where they are actually used, sort includes.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n\ncommit facb180d3a519040e877eb3a6810c4fc53320ed7\nAuthor: Fabio Estevam \u003cfestevam@gmail.com\u003e\nDate: Wed Jun 5 14:57:02 2019 -0300\n\n drm/etnaviv: Use devm_platform_ioremap_resource()\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n \n Signed-off-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n\ncommit 6eae41fea75039136707c02cf99431462d590c5f\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 30 07:21:03 2019 +0200\n\n drm/etnaviv: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n Fix fallout in all .c files.\n \n The etnaviv_drv.h header file was made self-contained,\n and missing includes was then added to the .c files that needed them.\n In a few cases the list of include files was sorted.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Russell King \u003clinux+etnaviv@armlinux.org.uk\u003e\n Cc: Christian Gmeiner \u003cchristian.gmeiner@gmail.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: etnaviv@lists.freedesktop.org\n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n\ncommit 15ed3ea2280e3afaea82afa664a6d1adc65b0009\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Thu Jul 25 18:40:10 2019 -0500\n\n soundwire: bus: split handling of Device0 events\n \n Assigning a device number to a Slave will result in additional events\n when it reports its status in a PING frame. There is no point in\n dealing with all the other devices in a loop, this can be done when a\n non-device0 event happens.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190725234032.21152-19-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 8965ad8433eadadc5a41df2915eec4a144e8aed6\nAuthor: james qian wang (Arm Technology China) \u003cjames.qian.wang@arm.com\u003e\nDate: Tue Jun 18 09:10:49 2019 +0100\n\n drm/komeda: Enable dual-link support\n \n Komeda HW can support dual-link which splits display frame to two halves\n (left/link0, right/link1) and output them by two output links.\n Due to the halved pixel rate of each link, the pxlclk of dual-link can be\n reduced two times compare with single-link.\n \n For enabling dual-link:\n - The DT need to configure two output-links for the pipeline node.\n - Komeda enable dual-link when both link0 and link1 have been connected.\n \n Example of how the pipeline node will look like for dual-link setup\n \n pipe0: pipeline@0 {\n clocks \u003d \u003c\u0026fpgaosc2\u003e;\n clock-names \u003d \"pxclk\";\n reg \u003d \u003c0\u003e;\n \n #address-cells \u003d \u003c1\u003e;\n #size-cells \u003d \u003c0\u003e;\n \n port@0 {\n reg \u003d \u003c0\u003e;\n \n #address-cells \u003d \u003c1\u003e;\n #size-cells \u003d \u003c0\u003e;\n dp0_pipe0_link0: endpoint@0 {\n reg \u003d \u003c0\u003e;\n remote-endpoint \u003d \u003c\u0026dlink_connector_in0\u003e;\n \n };\n dp0_pipe0_link1: endpoint@1 {\n reg \u003d \u003c1\u003e;\n remote-endpoint \u003d \u003c\u0026dlink_connector_in1\u003e;\n };\n };\n };\n \n Signed-off-by: James Qian Wang (Arm Technology China) \u003cjames.qian.wang@arm.com\u003e\n Signed-off-by: Liviu Dudau \u003cLiviu.Dudau@arm.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190618081013.13638-3-james.qian.wang@arm.com\n\ncommit ed22c6d9304d35fbdfe4818016ae57a9da5df9bc\nAuthor: james qian wang (Arm Technology China) \u003cjames.qian.wang@arm.com\u003e\nDate: Tue Jun 18 09:10:40 2019 +0100\n\n drm/komeda: Use drm_display_mode \"crtc_\" prefixed hardware timings\n \n struct drm_display_mode contains two copies of timings.\n - plain timings.\n - hardware timings, the ones with \"crtc_\" prefix.\n According to the definition, update komeda to use the hardware timing.\n \n Signed-off-by: James Qian Wang (Arm Technology China) \u003cjames.qian.wang@arm.com\u003e\n Signed-off-by: Liviu Dudau \u003cLiviu.Dudau@arm.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190618081013.13638-2-james.qian.wang@arm.com\n\ncommit 84a68104193efd0e11dd9284e5be9070dde3fdd1\nAuthor: Liviu Dudau \u003cLiviu.Dudau@arm.com\u003e\nDate: Wed Jul 3 15:56:03 2019 +0100\n\n drm/drm_debugfs_crc.c: Document that .verify_crc_source vfunc is required for enabling CRC support.\n \n drm_debugfs_crtc_crc_add() function checks that both .set_crc_source and\n .verify_crc_source hooks are provided before enabling debugfs support for\n reading per-frame CRC data. Make that explicit in the documentation.\n \n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Signed-off-by: Liviu Dudau \u003cliviu.dudau@arm.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703150330.21992-1-Liviu.Dudau@arm.com\n\ncommit 28ba1b1da49a20ba8fb767d6ddd7c521ec79a119\nAuthor: Anders Roxell \u003canders.roxell@linaro.org\u003e\nDate: Tue Jul 30 17:30:56 2019 +0200\n\n drm: mali-dp: Mark expected switch fall-through\n \n Now that -Wimplicit-fallthrough is passed to GCC by default, the\n following warnings shows up:\n \n ../drivers/gpu/drm/arm/malidp_hw.c: In function ‘malidp_format_get_bpp’:\n ../drivers/gpu/drm/arm/malidp_hw.c:387:8: warning: this statement may fall\n through [-Wimplicit-fallthrough\u003d]\n bpp \u003d 30;\n ~~~~^~~~\n ../drivers/gpu/drm/arm/malidp_hw.c:388:3: note: here\n case DRM_FORMAT_YUV420_10BIT:\n ^~~~\n ../drivers/gpu/drm/arm/malidp_hw.c: In function ‘malidp_se_irq’:\n ../drivers/gpu/drm/arm/malidp_hw.c:1311:4: warning: this statement may fall\n through [-Wimplicit-fallthrough\u003d]\n drm_writeback_signal_completion(\u0026malidp-\u003emw_connector, 0);\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n ../drivers/gpu/drm/arm/malidp_hw.c:1313:3: note: here\n case MW_START:\n ^~~~\n \n Rework to add a \u0027break;\u0027 in a case that didn\u0027t have it so that\n the compiler doesn\u0027t warn about fall-through.\n \n Cc: stable@vger.kernel.org # v5.2+\n Fixes: b8207562abdd (\"drm/arm/malidp: Specified the rotation memory requirements for AFBC YUV formats\")\n Acked-by: Liviu Dudau \u003cliviu.dudau@arm.com\u003e\n Signed-off-by: Anders Roxell \u003canders.roxell@linaro.org\u003e\n Signed-off-by: Liviu Dudau \u003cLiviu.Dudau@arm.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730153056.3606-1-anders.roxell@linaro.org\n\ncommit 3cf7bf2e4837fa1836f09b751e6130473a69fd11\nAuthor: shaoyunl \u003cshaoyun.liu@amd.com\u003e\nDate: Fri Jul 26 14:19:02 2019 -0500\n\n drm/amdgpu: enable Navi12 kfd support for amdgpu\n \n Navi12 has the same interface as Navi10\n \n Signed-off-by: shaoyunl \u003cshaoyun.liu@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit fbd2afe5366cd78997153b615232482fb97011c5\nAuthor: Roman Li \u003cRoman.Li@amd.com\u003e\nDate: Tue Jul 16 17:26:14 2019 -0400\n\n drm/amd/display: Add missing NV12 asic IDs\n \n Add missing navi12 asic ids.\n \n Signed-off-by: Roman Li \u003cRoman.Li@amd.com\u003e\n Reviewed-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 078655d982c5fcf03a3120b52383482b3dc04ab1\nAuthor: Leo Li \u003csunpeng.li@amd.com\u003e\nDate: Tue Jul 16 18:12:13 2019 -0400\n\n drm/amdgpu: Add nv12 DC ip block\n \n Load DC and amdgpu display manager\n \n Signed-off-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Reviewed-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 36bbb634b75399b98e09be41842fdf8429453e31\nAuthor: Leo Li \u003csunpeng.li@amd.com\u003e\nDate: Tue Jul 16 11:50:06 2019 -0400\n\n drm/amd/display: Add ASICREV_IS_NAVI macros\n \n They are used by DC to determine ASIC revs.\n \n Signed-off-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Reviewed-by: Roman Li \u003cRoman.Li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 400e9c5ea66ae2ec101992b6f3af8441c5b58497\nAuthor: Boyuan Zhang \u003cboyuan.zhang@amd.com\u003e\nDate: Thu Jul 18 17:39:07 2019 -0400\n\n drm/amdgpu: enable DPG mode for Navi12\n \n Enable Dynamic Power Gating VCN for Navi12.\n \n Signed-off-by: Boyuan Zhang \u003cboyuan.zhang@amd.com\u003e\n Reviewed-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1fbed280a2445aba3700eae72b31a623311b2c2e\nAuthor: Boyuan Zhang \u003cboyuan.zhang@amd.com\u003e\nDate: Thu Jul 18 10:13:23 2019 -0400\n\n drm/amdgpu: add VCN ip block for Navi12\n \n Add VCN2 ip block for Navi12\n \n Signed-off-by: Boyuan Zhang \u003cboyuan.zhang@amd.com\u003e\n Reviewed-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a3219816c407282a0f1b40d1dc5efc320b051767\nAuthor: Boyuan Zhang \u003cboyuan.zhang@amd.com\u003e\nDate: Thu Jul 18 09:17:24 2019 -0400\n\n drm/amdgpu: add Navi12 VCN firmware support\n \n Add Navi12 to VCN family\n \n Signed-off-by: Boyuan Zhang \u003cboyuan.zhang@amd.com\u003e\n Reviewed-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6b66ae2e556c78e26fc75fb8ff01753857fbd457\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu Jul 18 02:54:29 2019 +0800\n\n drm/amdgpu: add psp ip block for navi12\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7f47efeb9e212dce2abfb5510b9a19741e51820b\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Jul 16 03:26:49 2019 +0800\n\n drm/amdgpu: add smu ip block for navi12\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e60cc94b268a65cc46f8fd430e98e764a4f46370\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu Jul 18 05:00:00 2019 +0800\n\n drm/amdgpu: start autoload till RLCG fw for navi12\n \n rlc save restore list is not ready yet for navi12\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 739cdbd6a21b61677890b648fe49f40a9dbcc909\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Fri Aug 2 08:59:36 2019 -0500\n\n drm/amdgpu/psp11: add psp support for navi12\n \n Same as other navi asics.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9ea8da75ba43a344a06b7c811a61277e9075a96b\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu Aug 1 14:54:59 2019 -0500\n\n drm/amdgpu/smu11: add smu support for navi12\n \n Same as other Navi asics.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 02938eed74a3d75c743c3b8a80c80ba130dd6674\nAuthor: Jack Xiao \u003cJack.Xiao@amd.com\u003e\nDate: Mon Jul 15 05:12:21 2019 +0800\n\n drm/amdgpu: correct smu rlc handshake enablement bit\n \n Correct the enablement bit of SMU RLC handshake.\n \n Signed-off-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 87190edcf3a965cb1ef050bf2c407f6cef94e676\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Wed Jul 10 18:50:20 2019 +0800\n\n drm/amdgpu: add CGTT_GS_NGG_CLK_CTRL register to gc header\n \n gc 10.1.2 introduced this new register\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c726fbf0fb64ef6142deeff2d211fa4996556094\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Jul 9 14:17:08 2019 +0800\n\n drm/amdgpu/sdma5: add golden settings for navi12 (v2)\n \n common golden settings are put in golden_settings_sdma_5 array\n \n v2: update settings (Alex)\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f8984cb9e3e984611d9dbd3cb77e93686e525e48\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Jul 9 14:16:22 2019 +0800\n\n drm/amdgpu/gfx10: add golden settings for navi12 (v2)\n \n Add initial golden settings for navi12 gfx.\n \n v2: update settings\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7990202903135173893beb211b14d3a37e97a5e7\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Wed Jun 26 19:19:57 2019 +0800\n\n drm/amdgpu: enable virtual display for navi12\n \n Virtual display is a sw display interface for\n bring up and virtualization or for cards without\n display hardware.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 71745cf4740945e3143f1cabb51fab4136fd4278\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Jan 29 22:36:15 2019 +0800\n\n drm/amdgpu/gfx10: set tcp harvest for navi12\n \n Same as navi10.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 44e9e7c96c5e987354ac7a976c1c17a88c9bf901\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu May 16 19:58:19 2019 +0800\n\n drm/amdgpu: add ip blocks for navi12\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4a0e815fb3180875f8763321e5d2e2b432d35f1a\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu May 16 18:05:37 2019 +0800\n\n drm/amdgpu/gmc10: set gart size and vm size for navi12\n \n Same as other navi asics.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f2d6731d77cfbcdac9724e249cebdce0a75a2d00\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Mon Dec 17 18:07:22 2018 +0800\n\n drm/amdgpu/sdma5: add placeholder for navi12 golden settings\n \n None yet.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6f523fd7b3d4e42d7d0e7591c8a121048d0142f8\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Mon Dec 17 18:05:32 2018 +0800\n\n drm/amdgpu/sdma5: declare sdma firmwares for navi12\n \n Declare the firmwares and load the proper ones for navi12.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4cdfc4a2be144ee84406e8e6dce5474a67f00c2c\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu May 16 20:02:14 2019 +0800\n\n drm/amdgpu/gfx10: set rlc funcs for navi12\n \n Same as other navi asics.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9ff3dba6d68db11f096d89b07c5a1db39ffdd6ff\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu May 16 20:01:03 2019 +0800\n\n drm/amdgpu/gfx10: set number of me(c)/pipe/queue for navi12\n \n Same as other navi asics.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 716e9bb099c52efe5290c0b1f329e855b453663a\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Jun 11 11:16:54 2019 +0800\n\n drm/amdgpu/gfx10: add placeholder for navi12 golden settings\n \n Not used yet.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 92c123aec1a28439c31a38500f35fef391b5953b\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Jun 11 11:16:38 2019 +0800\n\n drm/amdgpu/gfx10: declare cp/rlc firmwares for navi12\n \n Set the name properly to load the right ucode.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6983469c1a654571529705339e097ca44ce3d57e\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu May 16 19:01:19 2019 +0800\n\n drm/amdgpu/gfx10: add gfx config for navi12\n \n got from mmCP_MAX_CONTEXT and mmPA_SC_FIFO_SIZE\n \n v2: squash all navi asics together because the\n settings are the same.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 59ab8c292b9ada03c05b691252a5a6f1d625e66d\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu May 16 19:03:01 2019 +0800\n\n drm/amdgpu/gfx10: set gfx cg for navi12\n \n Same as other navi asics.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7e17e58bdde26d1bb3588a8c1c546b9beed33405\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu May 16 19:51:12 2019 +0800\n\n drm/amdgpu: set nbio/hdp cg for navi12\n \n Same as navi10.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 74b5e509a082c70fa20798343cd0bbfa532ceda1\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu May 16 19:47:33 2019 +0800\n\n drm/amdgpu: initialize cg/pg flags and external rev id for navi12\n \n don\u0027t enable any cg/pg features yet.\n \n v2: calculate external revision id from revision id so that we can\n differentiate navi12 A0 from A1 directly.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d4d838ba4e9d1ca0d46743141bba84de1f04a007\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Mon Dec 17 18:04:19 2018 +0800\n\n drm/amdgpu: use front door firmware loading for navi12\n \n Same as other navi asics.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4808cf9c2a82427148a0c66f8833d1f743e1c960\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu May 16 17:36:41 2019 +0800\n\n drm/amdgpu: set asic family and ip blocks for navi12\n \n same with navi10\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 42b325e5ec39655d9f0bf0aef73f3f636cc972aa\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Mon Dec 17 18:01:38 2018 +0800\n\n drm/amdgpu: add gpu_info firmware for navi12\n \n gpu_info firmare store asic configuration details.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9802f5d78b3747467644edc609393f95d0b10e61\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Mon Dec 17 18:00:26 2018 +0800\n\n drm/amdgpu: add navi12 asic type\n \n Add asic type.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 03d0a073cf3f8eced0e27774f15b700a85c9f976\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue May 14 15:22:53 2019 +0800\n\n drm/amdgpu: initialize reg base for navi12\n \n Set up the register offset map for navi12.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6d622903285a878e1177981fb952e662e0f416a0\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue May 14 15:18:19 2019 +0800\n\n drm/amdgpu: add ip offset header for navi12 (v2)\n \n This adds the absolute offsets of each IP regiser block.\n \n v2: Squash in MP1 update\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8c2ef8ca0e0ea6a5b10b52a5eac8f3e7db0b663e\nAuthor: John Clements \u003cJohn.Clements@amd.com\u003e\nDate: Thu Aug 1 15:16:12 2019 +0800\n\n drm/amdgpu: update SDMA V4 microcode init\n \n Removed loading duplicate instances of SDMA FW for Arcturus.\n We use a single image for all instances.\n \n Signed-off-by: John Clements \u003cjohn.clements@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b86f8d8b2bc07b0f6802e9c6b481049a63f4a637\nAuthor: John Clements \u003cJohn.Clements@amd.com\u003e\nDate: Thu Aug 1 17:59:55 2019 +0800\n\n drm/amdgpu: extend PSP FW loading support to 8 SDMA instances\n \n Arcturus has 8 instances of SDMA. Update host to PSP interface\n to handle it.\n \n Signed-off-by: John Clements \u003cjohn.clements@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8fda90e8214413cc7f9e3c64ad0df7eddffcff62\nAuthor: John Clements \u003cJohn.Clements@amd.com\u003e\nDate: Wed Jul 31 16:11:08 2019 +0800\n\n drm/amdgpu: disable MEC2 JT context init for Arcturus\n \n We don\u0027t need to handle it like other asics.\n \n Signed-off-by: John Clements \u003cjohn.clements@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6c37bde9c63b4116913fd79abbb2e5abaeb52e23\nAuthor: John Clements \u003cJohn.Clements@amd.com\u003e\nDate: Thu Jul 25 15:29:11 2019 +0800\n\n drm/amdgpu: update PSP CMD fail response status print\n \n Print the response in hex with the apprpriate mask.\n \n Signed-off-by: John Clements \u003cjohn.clements@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit dc0d962297af444fa45ecea077e4ae53d7518719\nAuthor: John Clements \u003cJohn.Clements@amd.com\u003e\nDate: Mon Jul 22 18:06:58 2019 +0800\n\n drm/amdgpu: add PSP KDB loading support for Arcturus\n \n Add support for the arcturus specific psp metadata to the\n amdgpu firmware and properly parse it when loading it.\n \n Signed-off-by: John Clements \u003cjohn.clements@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f36d9ab95f0f775fd21d3b8dc4e15fd34eda8004\nAuthor: John Clements \u003cJohn.Clements@amd.com\u003e\nDate: Thu Aug 1 18:05:50 2019 +0800\n\n drm/amdgpu: add PSP SW init support for Arcturus\n \n Add arcturus cases to psp init sewquence.\n \n Signed-off-by: John Clements \u003cjohn.clements@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c0dac3c9f5458c623fab7856f24c258303d15cfb\nAuthor: John Clements \u003cJohn.Clements@amd.com\u003e\nDate: Wed Jul 31 16:11:38 2019 +0800\n\n drm/amdgpu: removed duplicate line\n \n Remove duplicate break.\n \n Signed-off-by: John Clements \u003cjohn.clements@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 706e50829eb7f667f259c83629bdf9e6c992772b\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Fri Aug 2 16:38:32 2019 +0800\n\n drm/amd/powerplay: correct navi10 vcn powergate\n \n vcn dpm on is a prerequisite for vcn power gate control.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4abc1765d2f74f706fdae067f3aec5c0778154d6\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Wed Jul 31 10:34:36 2019 +0800\n\n drm/amd/powerplay: enable SW SMU power profile switch support in KFD\n \n Hook up the SW SMU power profile switch in KFD routine.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7aa3f675d1439bb0ca68505b1eab06f75891e81b\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Mon Jul 29 15:17:27 2019 +0800\n\n drm/amd/powerplay: support power profile retrieval and setting on arcturus\n \n Enable arcturus power profile retrieval and setting.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 839f9117e1c27593245fb2af952a81d11a13b7c9\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Tue Jul 30 16:39:45 2019 +0800\n\n drm/amd/powerplay: guard consistency between CPU copy and local VRAM\n \n This can prevent CPU to use the out-dated copy.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit bd2280da46dbdbc3d70dc538f0613afb6fcc4efa\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Thu Aug 1 17:30:35 2019 +0800\n\n drm/amdgpu: replace AMDGPU_RAS_UE with AMDGPU_RAS_SUCCESS\n \n ce can also trigger interrupt, and even both ce and ue error can be\n found in one ras query, distinguishing between ce and ue in interrupt\n handler is uncessary.\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Suggested-by: Guchun Chen \u003cguchun.chen@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 91ba68f8d5727adb1c17d039af34af0f8297dff3\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Thu Aug 1 12:52:54 2019 +0800\n\n drm/amdgpu: only uncorrectable error needs gpu reset\n \n we only read error information for correctable error in interrupt\n handler, gpu reset is unnecessary since there is no data lost\n in correctable error\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b1a5895352dc1a154f1605702745ef2f63a5d797\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Mon Jul 29 17:19:57 2019 +0800\n\n drm/amdgpu: update the calc algorithm of umc ecc error count\n \n the initial value of ecc error count can be adjusted\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b7f92097f5bc5129cb386340ec54e6f40639d6e3\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Mon Jul 29 17:01:39 2019 +0800\n\n drm/amdgpu: implement umc ras init function\n \n enable umc ce interrupt and initialize ecc error count\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 51437623a0a14a5af74aee144a0f2d5790ab1420\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Mon Jul 29 16:04:33 2019 +0800\n\n drm/amdgpu: support ce interrupt in ras module\n \n correctable error can also trigger interrupt in some ras blocks\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 13b7c46c18e26182450d01efd0988e5c9fa8260f\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Thu Aug 1 11:41:39 2019 +0800\n\n drm/amdgpu: add error address query for umc ras\n \n umc error address query can get ce/ue error address and clear error\n status\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2b671b6049efafc7ae6de9f67acb964b9c534f3a\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Thu Aug 1 11:37:25 2019 +0800\n\n drm/amdgpu: apply umc_for_each_channel macro to umc_6_1\n \n use umc_for_each_channel to make code simpler\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit fee858ba5f96e6cbcbe8167444aeb6532519bb8c\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Mon Jul 29 14:50:35 2019 +0800\n\n drm/amdgpu: add macro of umc for each channel\n \n common function for all umc versions, loop for each umc channel is\n a frequent used operation in umc block, define it as a macro to\n simplify code\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 3aacf4ea1102f24c8dc63eb6f3d734cbc8bad86e\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Mon Jul 29 14:28:35 2019 +0800\n\n drm/amdgpu: initialize new parameters and functions for amdgpu_umc structure\n \n add initialization for new members of amdgpu_umc structure\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 33b97cf896d4ec6c05e8febdf73ee30c508a0481\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Mon Jul 29 14:10:54 2019 +0800\n\n drm/amdgpu: add more parameters and functions to amdgpu_umc structure\n \n expose more parameters and functions of specific umc version to common\n umc layer, so amdgpu_umc layer and other blocks could access them\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a55c8d7bda4f83e86e2b7ed7b1704e762ed50db3\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Mon Jul 29 10:28:57 2019 +0800\n\n drm/amdgpu: remove the clear of MCA_ADDR\n \n clearing MCA_STATUS is enough to reset the whole MCA, writing zero to\n MCA_ADDR is unnecessary\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b94afb61cdaeaa86f77846fdec419eb5e9d9f537\nAuthor: Kevin Wang \u003ckevin1.wang@amd.com\u003e\nDate: Fri Aug 2 12:01:00 2019 +0800\n\n drm/amd/powerplay: honor hw limit on fetching metrics data for navi10\n \n too frequently to update mertrics table will cause smu internal error.\n \n Signed-off-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit bd200d190f45b62c006d1ad0a63eeffd87db7a47\nAuthor: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\nDate: Wed Jul 31 10:33:54 2019 -0400\n\n drm/amd/display: Don\u0027t replace the dc_state for fast updates\n \n [Why]\n DRM private objects have no hw_done/flip_done fencing mechanism on their\n own and cannot be used to sequence commits accordingly.\n \n When issuing commits that don\u0027t touch the same set of hardware resources\n like page-flips on different CRTCs we can run into the issue below\n because of this:\n \n 1. Client requests non-blocking Commit #1, has a new dc_state #1,\n state is swapped, commit tail is deferred to work queue\n \n 2. Client requests non-blocking Commit #2, has a new dc_state #2,\n state is swapped, commit tail is deferred to work queue\n \n 3. Commit #2 work starts, commit tail finishes,\n atomic state is cleared, dc_state #1 is freed\n \n 4. Commit #1 work starts,\n commit tail encounters null pointer deref on dc_state #1\n \n In order to change the DC state as in the private object we need to\n ensure that we wait for all outstanding commits to finish and that\n any other pending commits must wait for the current one to finish as\n well.\n \n We do this for MEDIUM and FULL updates. But not for FAST updates, nor\n would we want to since it would cause stuttering from the delays.\n \n FAST updates that go through dm_determine_update_type_for_commit always\n create a new dc_state and lock the DRM private object if there are\n any changed planes.\n \n We need the old state to validate, but we don\u0027t actually need the new\n state here.\n \n [How]\n If the commit isn\u0027t a full update then the use after free can be\n resolved by simply discarding the new state entirely and retaining\n the existing one instead.\n \n With this change the sequence above can be reexamined. Commit #2 will\n still free Commit #1\u0027s reference, but before this happens we actually\n added an additional reference as part of Commit #2.\n \n If an update comes in during this that needs to change the dc_state\n it will need to wait on Commit #1 and Commit #2 to finish. Then it\u0027ll\n swap the state, finish the work in commit tail and drop the last\n reference on Commit #2\u0027s dc_state.\n \n Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id\u003d204181\n Fixes: 004b3938e637 (\"drm/amd/display: Check scaling info when determing update type\")\n \n Signed-off-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: David Francis \u003cdavid.francis@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 43d10d30df156f7834fa91aecb69614fefc8bb0a\nAuthor: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\nDate: Wed Jul 31 09:45:16 2019 -0400\n\n drm/amd/display: Skip determining update type for async updates\n \n [Why]\n By passing through the dm_determine_update_type_for_commit for atomic\n commits that can be done asynchronously we are incurring a\n performance penalty by locking access to the global private object\n and holding that access until the end of the programming sequence.\n \n This is also allocating a new large dc_state on every access in addition\n to retaining all the references on each stream and plane until the end\n of the programming sequence.\n \n [How]\n Shift the determination for async update before validation. Return early\n if it\u0027s going to be an async update.\n \n Signed-off-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: David Francis \u003cdavid.francis@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e16e37efb4c9eb7bcb9dab756c975040c5257e98\nAuthor: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\nDate: Mon Jun 10 08:47:57 2019 -0400\n\n drm/amd/display: Allow cursor async updates for framebuffer swaps\n \n [Why]\n We previously allowed framebuffer swaps as async updates for cursor\n planes but had to disable them due to a bug in DRM with async update\n handling and incorrect ref counting. The check to block framebuffer\n swaps has been added to DRM for a while now, so this check is redundant.\n \n The real fix that allows this to properly in DRM has also finally been\n merged and is getting backported into stable branches, so dropping\n this now seems to be the right time to do so.\n \n [How]\n Drop the redundant check for old_fb !\u003d new_fb.\n \n With the proper fix in DRM, this should also fix some cursor stuttering\n issues with xf86-video-amdgpu since it double buffers the cursor.\n \n IGT tests that swap framebuffers (-varying-size for example) should\n also pass again.\n \n Signed-off-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: David Francis \u003cdavid.francis@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ac4bf4a1ebce7554628db89498ec49054c7bcdba\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Thu Aug 1 12:01:45 2019 +0100\n\n drm/amdgpu: fix unsigned variable instance compared to less than zero\n \n Currenly the error check on variable instance is always false because\n it is a uint32_t type and this is never less than zero. Fix this by\n making it an int type.\n \n Addresses-Coverity: (\"Unsigned compared against 0\")\n Fixes: 7d0e6329dfdc (\"drm/amdgpu: update more sdma instances irq support\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f0ced3f61b4d2a21a3e0f0aa79fb5ad6c6717c31\nAuthor: Matt Coffin \u003cmcoffin13@gmail.com\u003e\nDate: Wed Jul 31 14:14:35 2019 -0600\n\n drm/amd/powerplay: Allow changing of fan_control in smu_v11_0\n \n [Why]\n Before this change, the fan control state on smu_v11 was not able to be\n changed because the capability check for checking if the fan control\n capability existed was inverted.\n \n [How]\n The capability check for fan control in smu_v11_0_auto_fan_control was\n inverted, to correctly check for the absence, instead of presence of fan\n control capabilities.\n \n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Matt Coffin \u003cmcoffin13@gmail.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ab63131155b0d86156b892a5d380dc337271aed9\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Thu Aug 1 09:39:41 2019 +0100\n\n drm/amd/powerplay: fix a few spelling mistakes\n \n There are a few spelling mistakes \"unknow\" -\u003e \"unknown\" and\n \"enabeld\" -\u003e \"enabled\". Fix these.\n \n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f3eb9b8f67bc28783eddc142ad805ebdc53d6339\nAuthor: Jia-Ju Bai \u003cbaijiaju1990@gmail.com\u003e\nDate: Mon Jul 29 16:36:44 2019 +0800\n\n gpu: drm: radeon: Fix a possible null-pointer dereference in radeon_connector_set_property()\n \n In radeon_connector_set_property(), there is an if statement on line 743\n to check whether connector-\u003eencoder is NULL:\n if (connector-\u003eencoder)\n \n When connector-\u003eencoder is NULL, it is used on line 755:\n if (connector-\u003eencoder-\u003ecrtc)\n \n Thus, a possible null-pointer dereference may occur.\n \n To fix this bug, connector-\u003eencoder is checked before being used.\n \n This bug is found by a static analysis tool STCheck written by us.\n \n Signed-off-by: Jia-Ju Bai \u003cbaijiaju1990@gmail.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e3bf125bdbb6efed2d650effbc3c13320602d64a\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Thu Aug 1 12:15:41 2019 +0100\n\n drm/amd/powerplay: fix off-by-one upper bounds limit checks\n \n There are two occurrances of off-by-one upper bound checking of indexes\n causing potential out-of-bounds array reads. Fix these.\n \n Addresses-Coverity: (\"Out-of-bounds read\")\n Fixes: cb33363d0e85 (\"drm/amd/powerplay: add smu feature name support\")\n Fixes: 6b294793e384 (\"drm/amd/powerplay: add smu message name support\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6f7fe9a93e6c09bf988c5059403f5f88e17e21e6\nAuthor: KyleMahlkuch \u003ckmahlkuc@linux.vnet.ibm.com\u003e\nDate: Wed Jul 31 17:10:14 2019 -0500\n\n drm/radeon: Fix EEH during kexec\n \n During kexec some adapters hit an EEH since they are not properly\n shut down in the radeon_pci_shutdown() function. Adding\n radeon_suspend_kms() fixes this issue.\n \n Signed-off-by: KyleMahlkuch \u003ckmahlkuc@linux.vnet.ibm.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5145d57ec5f5cf7dadaa6ccd9c9f1e4dae82570b\nAuthor: Jay Cornwall \u003cJay.Cornwall@amd.com\u003e\nDate: Thu Jul 18 16:57:22 2019 -0500\n\n drm/amdkfd: Extend CU mask to 8 SEs (v3)\n \n Following bitmap layout logic introduced by:\n \"drm/amdgpu: support get_cu_info for Arcturus\".\n \n v2: squash in fixup for gfx_v9_0.c (Alex)\n v3: squash in debug print output fix\n \n Signed-off-by: Jay Cornwall \u003cJay.Cornwall@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 857b82d0df4c85d3d1c043b325dcd8354fd09c6e\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Mon Jul 8 20:17:48 2019 +0800\n\n drm/amdgpu: support get_cu_info for Arcturus\n \n This change is because SE/SH layout on Arcturus is 8*1, different from\n 4*2(or 4*1) on Vega ASICs.\n \n Currently the cu bitmap array is 4x4 size, and besides the bitmap is used widely\n across SW stack. To mostly reduce the scale of impact, we make the cu bitmap\n array compatible with SE/SH layout on Arcturus. Then the store of cu bits of\n each shader array for Arcturus will be like below:\n SE0,SH0 --\u003e bitmap[0][0]\n SE1,SH0 --\u003e bitmap[1][0]\n SE2,SH0 --\u003e bitmap[2][0]\n SE3,SH0 --\u003e bitmap[3][0]\n SE4,SH0 --\u003e bitmap[0][1]\n SE5,SH0 --\u003e bitmap[1][1]\n SE6,SH0 --\u003e bitmap[2][1]\n SE7,SH0 --\u003e bitmap[3][1]\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 612e4ed99bdd7d4aae207a69dea16e609d304e3b\nAuthor: Kent Russell \u003ckent.russell@amd.com\u003e\nDate: Wed Jul 31 09:24:32 2019 -0400\n\n drm/amdgpu: Fix pcie_bw on Vega20\n \n The registers used for VG20 are different in that certain performance\n counters were split off to TXCLK3/4. Vega10/12 doesn\u0027t have this, so add\n a new vg20_get_pcie_usage to reflect this change.\n \n Signed-off-by: Kent Russell \u003ckent.russell@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 57d352f7699efedeebcf809df5e01e6fd51e2191\nAuthor: Kent Russell \u003ckent.russell@amd.com\u003e\nDate: Wed Jul 31 09:23:45 2019 -0400\n\n drm/amdgpu: Update NBIO headers to add TXCLK3/4\n \n These are added for VG20, and are needed for PCIe bandwidth.\n \n Signed-off-by: Kent Russell \u003ckent.russell@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 19ed70ff5d2458798387274186e738dabc2e05ad\nAuthor: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\nDate: Thu Aug 1 11:44:17 2019 -0400\n\n drm/amdgpu: Add amdgpu_asic_funcs.reset_method for Vega20\n \n Fixes GPU reset crash.\n \n Signed-off-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6856e4b65f64eeb3f17148f79b36c1d60c627529\nAuthor: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\nDate: Mon Jul 8 20:01:22 2019 -0400\n\n drm/amdgpu: Mark KFD VRAM allocations for wipe on release\n \n Memory used by KFD applications can contain sensitive information that\n should not be leaked to other processes. The current approach to prevent\n leaks is to clear VRAM at allocation time. This is not effective because\n memory can be reused in other ways without being cleared. Synchronously\n clearing memory on the allocation path also carries a significant\n performance penalty.\n \n Stop clearing memory at allocation time. Instead mark the memory for\n wipe on release.\n \n Signed-off-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ab2f7a5c18b5c17cc94aaab7ae2e7d1fa08993d6\nAuthor: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\nDate: Tue Jul 9 19:12:44 2019 -0400\n\n drm/amdgpu: Implement VRAM wipe on release\n \n Wipe VRAM memory containing sensitive data when moving or releasing\n BOs. Clearing the memory is pipelined to minimize any impact on\n subsequent memory allocation latency. Use of a poison value should\n help debug future use-after-free bugs.\n \n When moving BOs, the existing ttm_bo_pipelined_move ensures that the\n memory won\u0027t be reused before being wiped.\n \n When releasing BOs, the BO is fenced with the memory fill operation,\n which results in queuing the BO for a delayed delete.\n \n v2: Move amdgpu_amdkfd_unreserve_memory_limit into\n amdgpu_bo_release_notify so that KFD can use memory that\u0027s still\n being cleared in the background\n \n Signed-off-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d8f4981e2e8a968411105db568f3d48256b2ebbc\nAuthor: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\nDate: Mon Jul 8 20:09:21 2019 -0400\n\n drm/amdgpu: Add flag to wipe VRAM on release\n \n This memory allocation flag will be used to indicate BOs containing\n sensitive data that should not be leaked to other processes.\n \n Signed-off-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 274840e544225657fbca4f12efa1ee55474bb800\nAuthor: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\nDate: Tue Jul 9 19:09:42 2019 -0400\n\n drm/ttm: Add release_notify callback to ttm_bo_driver\n \n This notifies the driver that a BO is about to be released.\n \n Releasing a BO also invokes the move_notify callback from\n ttm_bo_cleanup_memtype_use, but that happens too late for anything\n that would add fences to the BO and require a delayed delete.\n \n Signed-off-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d9ec5cfd5a2eca87b7c5f697ae07b5c9ae9b771f\nAuthor: Leo Li \u003csunpeng.li@amd.com\u003e\nDate: Thu Jul 25 13:12:24 2019 -0400\n\n drm/amd/display: Use switch table for dc_to_smu_clock_type\n \n Using a static int array will cause errors if the given dm_pp_clk_type\n is out-of-bounds. For robustness, use a switch table, with a default\n case to handle all invalid values.\n \n v2: 0 is a valid clock type for smu_clk_type. Return SMU_CLK_COUNT\n instead on invalid mapping.\n \n Signed-off-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Reviewed-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d196bbbc28fab82624f7686f8b0da8e8644b6e6a\nAuthor: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\nDate: Wed Jul 3 22:52:16 2019 -0700\n\n drm/amd/display: Use proper enum conversion functions\n \n clang warns:\n \n drivers/gpu/drm/amd/amdgpu/../display/amdgpu_dm/amdgpu_dm_pp_smu.c:336:8:\n warning: implicit conversion from enumeration type \u0027enum smu_clk_type\u0027\n to different enumeration type \u0027enum amd_pp_clock_type\u0027\n [-Wenum-conversion]\n dc_to_smu_clock_type(clk_type),\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/gpu/drm/amd/amdgpu/../display/amdgpu_dm/amdgpu_dm_pp_smu.c:421:14:\n warning: implicit conversion from enumeration type \u0027enum\n amd_pp_clock_type\u0027 to different enumeration type \u0027enum smu_clk_type\u0027\n [-Wenum-conversion]\n dc_to_pp_clock_type(clk_type),\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n \n There are functions to properly convert between all of these types, use\n them so there are no longer any warnings.\n \n Fixes: a43913ea50a5 (\"drm/amd/powerplay: add function get_clock_by_type_with_latency for navi10\")\n Fixes: e5e4e22391c2 (\"drm/amd/powerplay: add interface to get clock by type with latency for display (v2)\")\n Link: https://github.com/ClangBuiltLinux/linux/issues/586\n Signed-off-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Reviewed-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 482f0e5385805ebbca25a1967da3538773bcdbea\nAuthor: Monk Liu \u003cMonk.Liu@amd.com\u003e\nDate: Wed Jul 31 16:47:56 2019 +0800\n\n drm/amdgpu: fix double ucode load by PSP(v3)\n \n previously the ucode loading of PSP was repreated, one executed in\n phase_1 init/re-init/resume and the other in fw_loading routine\n \n Avoid this double loading by clearing ip_blocks.status.hw in suspend or reset\n prior to the FW loading and any block\u0027s hw_init/resume\n \n v2:\n still do the smu fw loading since it is needed by bare-metal\n \n v3:\n drop the change in reinit_early_sriov, just clear all block\u0027s status.hw\n in the head place and set the status.hw after hw_init done is enough\n \n Signed-off-by: Monk Liu \u003cMonk.Liu@amd.com\u003e\n Reviewed-by: Emily Deng \u003cEmily.Deng@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9244d3a6eb45248b07ec2ec084b05b9ef102c588\nAuthor: Monk Liu \u003cMonk.Liu@amd.com\u003e\nDate: Tue Jul 30 17:32:27 2019 +0800\n\n drm/amdgpu: fix incorrect judge on sos fw version\n \n for SRIOV the SOS fw of PSP is loaded in hypervisor thus\n guest won\u0027t tell the version of it, and judging feature by\n reading the sos fw version in guest side is completely wrong\n \n Signed-off-by: Monk Liu \u003cMonk.Liu@amd.com\u003e\n Reviewed-by: Emily Deng \u003cEmily.Deng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4cd4c5c064bd5cb9045bab3ed79bce292504d5b8\nAuthor: Monk Liu \u003cMonk.Liu@amd.com\u003e\nDate: Tue Jul 30 17:21:19 2019 +0800\n\n drm/amdgpu: cleanup vega10 SRIOV code path\n \n we can simplify all those unnecessary function under\n SRIOV for vega10 since:\n 1) PSP L1 policy is by force enabled in SRIOV\n 2) original logic always set all flags which make itself\n a dummy step\n \n besides,\n 1) the ih_doorbell_range set should also be skipped\n for VEGA10 SRIOV.\n 2) the gfx_common registers should also be skipped\n for VEGA10 SRIOV.\n \n Signed-off-by: Monk Liu \u003cMonk.Liu@amd.com\u003e\n Reviewed-by: Emily Deng \u003cEmily.Deng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 67194518cb1979d775e10089d00c015d6737c194\nAuthor: Kevin Wang \u003ckevin1.wang@amd.com\u003e\nDate: Wed Jul 31 15:37:07 2019 +0800\n\n drm/amd/powerplay: sort feature status index by asic feature id for smu\n \n before this change, the pp_feature sysfs show feature enable state by\n logic feature id, it is not easy to read.\n this change will sort pp_features show index by asic feature id.\n \n before:\n features high: 0x00000623 low: 0xb3cdaffb\n 00. DPM_PREFETCHER ( 0) : enabeld\n 01. DPM_GFXCLK ( 1) : enabeld\n 02. DPM_UCLK ( 3) : enabeld\n 03. DPM_SOCCLK ( 4) : enabeld\n 04. DPM_MP0CLK ( 5) : enabeld\n 05. DPM_LINK ( 6) : enabeld\n 06. DPM_DCEFCLK ( 7) : enabeld\n 07. DS_GFXCLK (10) : enabeld\n 08. DS_SOCCLK (11) : enabeld\n 09. DS_LCLK (12) : disabled\n 10. PPT (23) : enabeld\n 11. TDC (24) : enabeld\n 12. THERMAL (33) : enabeld\n 13. RM (35) : disabled\n ......\n \n after:\n features high: 0x00000623 low: 0xb3cdaffb\n 00. DPM_PREFETCHER ( 0) : enabeld\n 01. DPM_GFXCLK ( 1) : enabeld\n 02. DPM_GFX_PACE ( 2) : disabled\n 03. DPM_UCLK ( 3) : enabeld\n 04. DPM_SOCCLK ( 4) : enabeld\n 05. DPM_MP0CLK ( 5) : enabeld\n 06. DPM_LINK ( 6) : enabeld\n 07. DPM_DCEFCLK ( 7) : enabeld\n 08. MEM_VDDCI_SCALING ( 8) : enabeld\n 09. MEM_MVDD_SCALING ( 9) : enabeld\n 10. DS_GFXCLK (10) : enabeld\n 11. DS_SOCCLK (11) : enabeld\n 12. DS_LCLK (12) : disabled\n 13. DS_DCEFCLK (13) : enabeld\n ......\n \n Signed-off-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit cd2a4eaf8c79daa41bdeb7251dbf66413291fd70\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jul 30 21:58:05 2019 +0100\n\n drm/i915: Report resv_obj allocation failure\n \n Since commit 64d6c500a384 (\"drm/i915: Generalise GPU activity\n tracking\"), we have been prepared for i915_vma_move_to_active() to fail.\n We can take advantage of this to report the failure for allocating the\n shared-fence slot in the reservation_object.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730205805.3733-1-chris@chris-wilson.co.uk\n\ncommit b5deabd439ed644c7e22e3217883550d86791851\nAuthor: Patrice Chotard \u003cpatrice.chotard@st.com\u003e\nDate: Wed Jul 31 09:22:04 2019 +0200\n\n ARM: multi_v7_defconfig: Enable SPI_STM32_QSPI support\n \n Enable support for QSPI block on STM32 SoCs.\n \n Signed-off-by: Patrice Chotard \u003cpatrice.chotard@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit b343037de48f1da690364609bfdfba81005b49a4\nAuthor: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\nDate: Mon Jul 1 10:14:24 2019 +0200\n\n ARM: multi_v7_defconfig: enable STM32 booster regulator\n \n This enables the driver for STM32 booster regulator found on stm32mp1.\n \n Signed-off-by: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit 320b96e41bb7579ebdc1e852e39a7cb18627bec3\nAuthor: Christophe Kerello \u003cchristophe.kerello@st.com\u003e\nDate: Fri Jun 21 16:49:50 2019 +0200\n\n ARM: multi_v7_defconfig: add FMC2 NAND controller support\n \n This patch adds FMC2 NAND controller support used by STM32MP SOCs.\n \n Signed-off-by: Christophe Kerello \u003cchristophe.kerello@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit 49490d51743cedad368f3d76f76406bc4b75fd51\nAuthor: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\nDate: Wed Jul 31 17:26:09 2019 +0200\n\n ARM: dts: stm32: remove useless pinctrl entries in stm32mp157-pinctrl\n \n This patch removes \"ngpios\" and \"gpio-ranges\" information from\n stm32mp157-pinctrl.dtsi file as it is now filled in stm32mp157 pinctrl\n package files.\n \n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit 98414811d773e86187d3ece2c47c79df928d8a9d\nAuthor: Yannick Fertré \u003cyannick.fertre@st.com\u003e\nDate: Mon Jul 29 16:29:08 2019 +0200\n\n ARM: dts: stm32: add phy-dsi-supply property on stm32mp157c-ev1\n \n The dsi physical layer is powered by the 1v8 power controller supply.\n \n Signed-off-by: Yannick Fertré \u003cyannick.fertre@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit 08b9092ff3b6ef4c53fba46f5844c054f25c2c15\nAuthor: Olivier Moysan \u003colivier.moysan@st.com\u003e\nDate: Fri Jul 5 13:53:00 2019 +0200\n\n ARM: dts: stm32: add audio codec support on stm32mp157a-dk1 board\n \n Add support of Cirrus cs42l51 audio codec on stm32mp157a-dk1 board.\n Configuration overview:\n - SAI2A is the CPU interface used for the codec audio playback\n - SAI2B is the CPU interface used for the codec audio record\n - SAI2A is configured as a clock provider for the audio codec\n - SAI2A\u0026B are configured as slave of the audio codec\n - SAI2A\u0026B share the same interface of the audio codec\n \n Note:\n In master mode, cs42l51 audio codec provides a bitclock\n at 64 x FS, regardless of data width. This means that\n slot width is always 32 bits.\n Set slot width to 32 bits and slot number to 2\n in SAI2A\u0026B endpoint nodes, to match this constraint.\n dai-tdm-slot-num and dai-tdm-slot-width properties are used here,\n assuming that i2s is a special case of tdm, where slot number is 2.\n \n Signed-off-by: Olivier Moysan \u003colivier.moysan@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit 92edf0a2851d44d0f456a4f177736bd45f059c26\nAuthor: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\nDate: Wed Jul 3 12:08:16 2019 +0200\n\n ARM: dts: stm32: add syscfg to ADC on stm32mp157c\n \n On stm32mp157c, the ADC inputs are multiplexed with analog switches which\n have reduced performances when their supply is below 2.7V (vdda by\n default).\n Add syscfg registers that can be used on stm32mp157c, to get full ADC\n analog performances.\n \n Signed-off-by: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit 4f0f89dd9060cc04de48f3efc4358b687f406158\nAuthor: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\nDate: Wed Jun 19 11:52:05 2019 +0200\n\n ARM: dts: stm32: add pwm cells to stm32f746\n \n STM32 Timers support generic 3 cells PWM to encode PWM number, period and\n polarity.\n \n Fixes: 9bd7b77af8e4 (\"ARM: dts: stm32: add Timers driver for stm32f746\n MCU\")\n \n Signed-off-by: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit 289459737869b55f8a6b4618184cb827c88e097c\nAuthor: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\nDate: Wed Jun 19 11:52:04 2019 +0200\n\n ARM: dts: stm32: add pwm cells to stm32f429\n \n STM32 Timers support generic 3 cells PWM to encode PWM number, period and\n polarity.\n \n Fixes: c0e14fc712d9 (\"ARM: dts: stm32: add Timers driver for stm32f429\n MCU\")\n \n Signed-off-by: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit eb5f46e54910e76c95f4251e767b5ac15532763b\nAuthor: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\nDate: Wed Jun 19 11:52:03 2019 +0200\n\n ARM: dts: stm32: add pwm cells to stm32mp157c\n \n STM32 Timers support generic 3 cells PWM to encode PWM number, period and\n polarity.\n \n Fixes: 61fc211c484d (\"ARM: dts: stm32: add timers support to stm32mp157c\")\n \n Signed-off-by: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit 1f9510c2ff74645dd0779cdc3fba095458443d9e\nAuthor: Erwan Le Ray \u003cerwan.leray@st.com\u003e\nDate: Mon Jul 1 11:18:06 2019 +0200\n\n ARM: dts: stm32: fix -Wall W\u003d1 compilation in stm32mp157 pinctrl for mcan\n \n Fix compilations warnings detected by -Wall W\u003d1 compilation option:\n - node has a unit name, but no reg property\n \n Signed-off-by: Erwan Le Ray \u003cerwan.leray@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit e7c3a5beb31e88acfedb07244b51fbe345bedc25\nAuthor: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\nDate: Mon Jul 1 10:14:25 2019 +0200\n\n ARM: dts: stm32: add booster for ADC analog switches on stm32mp157c\n \n Booster for ADC analog input switches can be used when Vdda is below 2.7V\n to get maximum ADC analog performances.\n Add booster for ADC analog switches on stm32mp157c.\n \n Signed-off-by: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit ae5dda2d1d05f7c6263c87bc461bf6a6e12220e2\nAuthor: Fabien Dessenne \u003cfabien.dessenne@st.com\u003e\nDate: Fri Jul 26 18:09:53 2019 +0200\n\n ARM: dts: stm32: enable m4 coprocessor support on STM32MP157a-dk1\n \n Enable m4 coprocessor for STM32MP157a-dk1 board.\n \n Signed-off-by: Fabien Dessenne \u003cfabien.dessenne@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit c39ca797fff9c843def5f66081ea60646e579702\nAuthor: Fabien Dessenne \u003cfabien.dessenne@st.com\u003e\nDate: Fri Jul 26 18:08:35 2019 +0200\n\n ARM: dts: stm32: declare copro reserved memories on STM32MP157a-dk1\n \n Declare reserved memories shared by the processors for STM32MP157a-dk1\n \n Signed-off-by: Fabien Dessenne \u003cfabien.dessenne@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit f2d0ea33d251e6aa2f29c4a84c7a74aea0f02da1\nAuthor: Fabien Dessenne \u003cfabien.dessenne@st.com\u003e\nDate: Tue May 14 10:27:01 2019 +0200\n\n ARM: dts: stm32: enable m4 coprocessor support on STM32MP157c-ed1\n \n Enable m4 coprocessor for STM32MP157c-ed1 board.\n \n Signed-off-by: Fabien Dessenne \u003cfabien.dessenne@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit 037dc38a25b4d98210d9b5a433da7302a8e474c7\nAuthor: Fabien Dessenne \u003cfabien.dessenne@st.com\u003e\nDate: Fri Jul 26 18:06:57 2019 +0200\n\n ARM: dts: stm32: declare copro reserved memories on STM32MP157c-ed1\n \n Declare reserved memories shared by the processors for STM32MP157c-ed1\n board.\n \n Signed-off-by: Fabien Dessenne \u003cfabien.dessenne@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit c19335e2af91dabdb5b0c8b130bad6f8fb7af5c5\nAuthor: Fabien Dessenne \u003cfabien.dessenne@st.com\u003e\nDate: Tue May 14 10:26:59 2019 +0200\n\n ARM: dts: stm32: add m4 remoteproc support on STM32MP157c\n \n Declare the M4 remote processor in a sub-node of the mlahb simple bus.\n \n Signed-off-by: Fabien Dessenne \u003cfabien.dessenne@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit 493e84c5dc4d703d976b5875f5db22dae08a0782\nAuthor: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\nDate: Wed Jun 19 14:29:58 2019 +0200\n\n ARM: dts: stm32: add missing vdda-supply to adc on stm32h743i-eval\n \n Add missing vdda-supply required by STM32 ADC.\n \n Fixes: 090992a9ca54 (\"ARM: dts: stm32: enable ADC on stm32h743i-eval\n board\")\n \n Signed-off-by: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit 1425d00aff015ca1b0ecd6d6c112b7802be1383c\nAuthor: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\nDate: Wed Jun 19 14:29:57 2019 +0200\n\n ARM: dts: stm32: add missing vdda-supply to adc on stm32429i-eval\n \n Add missing vdda-supply required by STM32 ADC.\n \n Fixes: 7465d81191a1 (\"ARM: dts: stm32: enable ADC on stm32f429i-eval\n board\")\n \n Signed-off-by: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit bb06b54721fb58f51f051dc925263f480500b506\nAuthor: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\nDate: Wed Jun 19 14:29:56 2019 +0200\n\n ARM: dts: stm32: remove fixed regulator unit address on stm32429i-eval\n \n vref fixed regulator shouldn\u0027t have unit address and reg properties.\n Rename the label and phandle to \"vref\" according to the schematics.\n Also remove it from simple-bus.\n \n Fixes: 7465d81191a1 (\"ARM: dts: stm32: enable ADC on stm32f429i-eval\n board\")\n \n Signed-off-by: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit a049a377164c3eca7d5323cd9896b64d2120c415\nAuthor: Thomas Huth \u003cthuth@redhat.com\u003e\nDate: Wed Jul 31 17:15:25 2019 +0200\n\n KVM: selftests: Enable dirty_log_test on s390x\n \n To run the dirty_log_test on s390x, we have to make sure that we\n access the dirty log bitmap with little endian byte ordering and\n we have to properly align the memslot of the guest.\n Also all dirty bits of a segment are set once on s390x when one\n of the pages of a segment are written to for the first time, so\n we have to make sure that we touch all pages during the first\n iteration to keep the test in sync here.\n DEFAULT_GUEST_TEST_MEM needs an adjustment, too. On some s390x\n distributions, the ELF binary is linked to address 0x80000000,\n so we have to avoid that our test region overlaps into this area.\n 0xc0000000 seems to be a good alternative that should work on x86\n and aarch64, too.\n \n Acked-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n Reviewed-by: Andrew Jones \u003cdrjones@redhat.com\u003e\n Signed-off-by: Thomas Huth \u003cthuth@redhat.com\u003e\n Link: https://lore.kernel.org/r/20190731151525.17156-4-thuth@redhat.com\n Signed-off-by: Christian Borntraeger \u003cborntraeger@de.ibm.com\u003e\n\ncommit f90f57b3971a3370be5a3cf02ae783b4bf728b7f\nAuthor: Thomas Huth \u003cthuth@redhat.com\u003e\nDate: Wed Jul 31 17:15:24 2019 +0200\n\n KVM: selftests: Implement ucall() for s390x\n \n On s390x, we can neither exit via PIO nor MMIO, but have to use an\n instruction like DIAGNOSE. Now that ucall() is implemented, we can\n use it in the sync_reg_test on s390x, too.\n \n Reviewed-by: Andrew Jones \u003cdrjones@redhat.com\u003e\n Signed-off-by: Thomas Huth \u003cthuth@redhat.com\u003e\n Link: https://lore.kernel.org/r/20190731151525.17156-3-thuth@redhat.com\n Signed-off-by: Christian Borntraeger \u003cborntraeger@de.ibm.com\u003e\n\ncommit 2040f414d12f31be3e73eb3f5048d2b1cdec48f6\nAuthor: Thomas Huth \u003cthuth@redhat.com\u003e\nDate: Wed Jul 31 17:15:23 2019 +0200\n\n KVM: selftests: Split ucall.c into architecture specific files\n \n The way we exit from a guest to userspace is very specific to the\n architecture: On x86, we use PIO, on aarch64 we are using MMIO and on\n s390x we\u0027re going to use an instruction instead. The possibility to\n select a type via the ucall_type_t enum is currently also completely\n unused, so the code in ucall.c currently looks more complex than\n required. Let\u0027s split this up into architecture specific ucall.c\n files instead, so we can get rid of the #ifdefs and the unnecessary\n ucall_type_t handling.\n \n Reviewed-by: Andrew Jones \u003cdrjones@redhat.com\u003e\n Signed-off-by: Thomas Huth \u003cthuth@redhat.com\u003e\n Acked-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n Link: https://lore.kernel.org/r/20190731151525.17156-2-thuth@redhat.com\n Signed-off-by: Christian Borntraeger \u003cborntraeger@de.ibm.com\u003e\n\ncommit e4661f144497f588b6baa6f00d7f3e4d58b29077\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Fri Aug 2 12:47:39 2019 +0000\n\n drm/i915: Fix documentation for __intel_wait_for_register_fw*\n \n Use section name \"Return\" and proper error code -ETIMEDOUT\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190802124739.12548-1-michal.wajdeczko@intel.com\n\ncommit 57d8396504b3a93f284e51b866740a3e7419a3d9\nAuthor: Jyri Sarha \u003cjsarha@ti.com\u003e\nDate: Fri Aug 2 15:55:22 2019 +0300\n\n drm/tilcdc: Remove obsolete crtc_mode_valid() hack\n \n Earlier there were no mode_valid() helper for crtc and tilcdc had a\n hack to over come this limitation. But now the mode_valid() helper is\n there (has been since v4.13), so it is about time to get rid of that\n hack.\n \n Signed-off-by: Jyri Sarha \u003cjsarha@ti.com\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/ \u003c5c4dcb5b1e7975bd2b7ca86f7addf219cd0f9a06.1564750248.git.jsarha@ti.com\n\ncommit 8f0df898b27926e443d13770adfd828cc0f50148\nAuthor: Frieder Schrempf \u003cfrieder.schrempf@kontron.de\u003e\nDate: Fri Aug 2 10:04:11 2019 +0000\n\n serial: 8250: Don\u0027t check for mctrl_gpio_init() returning -ENOSYS\n \n Now that the mctrl_gpio code returns NULL instead of ERR_PTR(-ENOSYS)\n if CONFIG_GPIOLIB is disabled, we can safely remove this check.\n \n Signed-off-by: Frieder Schrempf \u003cfrieder.schrempf@kontron.de\u003e\n Acked-by: Uwe Kleine-Knig \u003cu.kleine-koenig@pengutronix.de\u003e\n Link: https://lore.kernel.org/r/20190802100349.8659-4-frieder.schrempf@kontron.de\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit eea8b4fc48c719d83c6c3c764fda21d43c45f0d2\nAuthor: Frieder Schrempf \u003cfrieder.schrempf@kontron.de\u003e\nDate: Fri Aug 2 10:04:10 2019 +0000\n\n serial: sh-sci: Don\u0027t check for mctrl_gpio_init() returning -ENOSYS\n \n Now that the mctrl_gpio code returns NULL instead of ERR_PTR(-ENOSYS)\n if CONFIG_GPIOLIB is disabled, we can safely remove this check.\n \n Signed-off-by: Frieder Schrempf \u003cfrieder.schrempf@kontron.de\u003e\n Acked-by: Uwe Kleine-Knig \u003cu.kleine-koenig@pengutronix.de\u003e\n Link: https://lore.kernel.org/r/20190802100349.8659-3-frieder.schrempf@kontron.de\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit c3595221945938153f3c7bb6a1c0ab5d989b1c0b\nAuthor: Frieder Schrempf \u003cfrieder.schrempf@kontron.de\u003e\nDate: Fri Aug 2 10:04:09 2019 +0000\n\n serial: mctrl_gpio: Avoid probe failures in case of missing gpiolib\n \n If CONFIG_GPIOLIB is not enabled, mctrl_gpio_init() and\n mctrl_gpio_init_noauto() will currently return an error pointer with\n -ENOSYS. As the mctrl GPIOs are usually optional, drivers need to\n check for this condition to allow continue probing.\n \n To avoid the need for this check in each driver, we return NULL\n instead, as all the mctrl_gpio_*() functions are skipped anyway.\n We also adapt mctrl_gpio_to_gpiod() to be in line with this change.\n \n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Frieder Schrempf \u003cfrieder.schrempf@kontron.de\u003e\n Reviewed-by: Uwe Kleine-Knig \u003cu.kleine-koenig@pengutronix.de\u003e\n Link: https://lore.kernel.org/r/20190802100349.8659-1-frieder.schrempf@kontron.de\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 93b6984b31182cfc340495af17691c0b9d53f6b2\nAuthor: Corey Minyard \u003ccminyard@mvista.com\u003e\nDate: Wed Jul 31 19:46:53 2019 -0500\n\n ipmi_si: Remove ipmi_ from the device attr names\n \n Better conform with kernel style.\n \n Signed-off-by: Corey Minyard \u003ccminyard@mvista.com\u003e\n\ncommit a6f4c33187d038dc2ca4ab885eb5e9c44940760f\nAuthor: Corey Minyard \u003ccminyard@mvista.com\u003e\nDate: Wed Jul 31 19:37:43 2019 -0500\n\n ipmi_si: Convert device attr permissions to octal\n \n Kernel preferences are for octal values instead of symbols.\n \n Signed-off-by: Corey Minyard \u003ccminyard@mvista.com\u003e\n\ncommit 104fb25f60077e4696145bcea51ca56f0959d7e3\nAuthor: Corey Minyard \u003ccminyard@mvista.com\u003e\nDate: Wed Jul 31 19:18:25 2019 -0500\n\n ipmi_si: Rework some include files\n \n ipmi_si_sm.h was getting included in lots of places it didn\u0027t\n belong. Rework things a bit to remove all the dependencies,\n mostly just moving things between include files that were in\n the wrong place and removing bogus includes.\n \n Signed-off-by: Corey Minyard \u003ccminyard@mvista.com\u003e\n\ncommit 2205d920f4ada49013d63ca58dfd74dcf53df38b\nAuthor: Aaron Sierra \u003casierra@xes-inc.com\u003e\nDate: Thu Aug 1 13:59:56 2019 -0500\n\n serial: 8250_exar: Absorb remaining 8250_port INT0 support\n \n Move INT0 clearing out of common, per-port serial8250_do_startup()\n into PCI device probe/resume.\n \n As described in commit 2c0ac5b48a35 (\"serial: exar: Fix stuck MSIs\"),\n the purpose of clearing INT0 is to prevent the PCI interrupt line from\n becoming stuck asserted, \"which is fatal with edge-triggered MSIs\".\n \n Like the clearing via interrupt handler that moved from common code in\n commit c7e1b4059075 (\"tty: serial: exar: Relocate sleep wake-up\n handling\"), this clearing at startup can be better handled at the PCI\n device level.\n \n Cc: Jan Kiszka \u003cjan.kiszka@siemens.com\u003e\n Cc: Sudip Mukherjee \u003csudip.mukherjee@codethink.co.uk\u003e\n Signed-off-by: Aaron Sierra \u003casierra@xes-inc.com\u003e\n Reviewed-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190801185956.3222-1-asierra@xes-inc.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit ae0f8f0f9a4b7d2b37adcec54e540f8a03238f3d\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Tue Jul 30 14:34:32 2019 +0300\n\n drm/i915/oa: update the generated files\n \n Update the generated files to make the headers self-contained, switch to\n the kernel preferred SPDX comment format, and update the copyright\n year. Also add the Makefile stanza to run header tests on the files.\n \n Other changes produced by gputop i915-perf-kernelgen.py were manually\n stripped out, and left to the folks who actually know something about\n the OA stuff.\n \n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730113432.22146-3-jani.nikula@intel.com\n\ncommit 68acd85996f6456ad95b458f19bc7309ed530a4b\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Thu Jul 25 18:40:03 2019 -0500\n\n soundwire: cadence_master: simplify bus clash interrupt clear\n \n The bus clash interrupts are generated when the status is one, and\n also cleared by writing a one. It\u0027s overkill/useless to use an OR when\n the bit is already set.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190725234032.21152-12-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit c23ef285fc67f5fd2d7faac4d14d8b8c5a2b23ab\nAuthor: Guido Günther \u003cagx@sigxcpu.org\u003e\nDate: Fri Aug 2 12:55:18 2019 +0200\n\n drm/imx: Drop unused imx-ipuv3-crtc.o build\n \n Since\n \n commit 3d1df96ad468 (\"drm/imx: merge imx-drm-core and ipuv3-crtc in one module\")\n \n the former contents of imx-ipuv3-crtc.o are built via imxdrm-objs. So\n there\u0027s no need to keep an extra entry with a non existing config value\n (CONFIG_DRM_IMX_IPUV3).\n \n Fixes: 3d1df96ad468 (\"drm/imx: merge imx-drm-core and ipuv3-crtc in one module\")\n Signed-off-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit b35105a3a32343ecdd777497a300a7680f15bf08\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 2 12:12:12 2019 +0530\n\n staging: rtl8712: r8712_aes_decrypt(): Change return type\n \n Change return type of r8712_aes_decrypt from u8 to void as its return\n value is never checked. Modify return statements accordingly.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190802064212.30476-9-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit eb43590f50a393ec4156be59eec9287fb711ffd8\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 2 12:12:11 2019 +0530\n\n staging: rtl8712: aes_decipher(): Change return type\n \n Change return type of aes_decipher from sint to void as it always\n returns _SUCCESS and this value is never used.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190802064212.30476-8-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit e48a3add05d1b3e0c2ead1e809edd4d02f2f5f10\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 2 12:12:10 2019 +0530\n\n staging: rtl8712: aes_cipher(): Change return type\n \n Change return type of aes_cipher from sint to void as it always returns\n _SUCCESS and its return value is never used.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190802064212.30476-7-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f8dbe3f0acaff86c8c12540bb1b76c34d32a563f\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 2 12:12:09 2019 +0530\n\n staging: rtl8712: r8712_tkip_decrypt(): Change return type\n \n Change return type of r8712_tkip_decrypt from u8 to void as the\n return value is never used. Modify or remove return statements\n accordingly.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190802064212.30476-6-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit dcca75a495f9a8b7a6a0fd66bdbe0ed08ffe4884\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 2 12:12:08 2019 +0530\n\n staging: rtl8712: r8712_os_recvbuf_resource_free(): Change return type\n \n Change return type of r8712_os_recvbuf_resource_free from int to void as\n it always returns _SUCCESS and this return value is never stored,\n checked or otherwise used. Remove return statement accordingly.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190802064212.30476-5-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 14b4302ff028afc5fc09f243a21ee6cad6118c40\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 2 12:12:07 2019 +0530\n\n staging: rtl8712: r8712_os_recvbuf_resource_alloc(): Change return values\n \n Change return values of r8712_os_recvbuf_resource_alloc from\n _SUCCESS/_FAIL to 0/-ENOMEM respectively.\n Modify check at call site to check for non-zero return value instead of\n _FAIL. Thereafter remove variable at call site that stored the return\n value and perform the check directly.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190802064212.30476-4-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 0bf36e605286eaa53ec90c09a2f8eaac5d4d1cc4\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 2 12:12:06 2019 +0530\n\n staging: rtl8712: r8712_init_recv_priv(): Change return type\n \n Change return type of r8712_init_recv_priv from int to void as its\n return value is never used. Remove return statements accordingly.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190802064212.30476-3-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 1ce0212ab8c103f36aa8d7b43ff8d738bc98699b\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 2 12:12:05 2019 +0530\n\n staging: rtl8712: _r8712_init_recv_priv(): Change return type\n \n Change return type of _r8712_init_recv_priv as its return value is never\n checked or used. Modify return statements accordingly.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190802064212.30476-2-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a0afad481b83690449a119a9384a3b0f54cacf11\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 2 12:12:04 2019 +0530\n\n staging: rtl8712: r8712_os_recv_resource_alloc(): Change return type\n \n Change return type of function r8712_os_recv_resource_alloc from int to\n void as its return value is never used.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190802064212.30476-1-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 6250a57e6cf45440e2230bda226162532b0130e4\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 2 11:54:44 2019 +0530\n\n staging: rtl8723bs: Replace hal_btcoex_Initialize()\n \n Remove hal_btcoex_Initialize as all it does is perform a memset and call\n EXhalbtcoutsrc_InitlizeVariables.\n Rename EXhalbtcoutsrc_InitlizeVariables to hal_btcoex_Initialize and add\n the memset of hal_btcoex_Initialize in order to maintain compatibility\n with call sites of the latter (EXhalbtcoutsrc is not called anywhere\n else except in now-removed old hal_btcoex_Initialize).\n Change return type of new hal_btcoex_Initialize from u8 to void and\n remove its return statement as the return value of hal_btcoex_Initialize\n is never used.\n Change the type of function argument at call site to match the function\n parameter of new hal_btcoex_Initialize.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190802062444.30384-4-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 365231fcd67c97ea464c17d8d413f0860400547b\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 2 11:54:43 2019 +0530\n\n staging: rtl8723bs: Change return type of hal_btcoex_IsLpsOn()\n \n Change return type of hal_btcoex_IsLpsOn from u8 to bool as the\n function only returns boolean values.\n Modify a comparison of a boolean value to false to instead check that\n boolean value directly.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190802062444.30384-3-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 84881dd37beb0626d2622ccbfbdcfdc4bda2546d\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 2 11:54:42 2019 +0530\n\n staging: rtl8723bs: Change return type of hal_btcoex_IsBtControlLps()\n \n Change return type of hal_btcoex_IsBtControlLps from u8 to bool as it\n only returns boolean values.\n Change comparison to false within this function; check the boolean\n variable directly instead.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190802062444.30384-2-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 42755adcce8897a60a766bc102fcb23f1dd28f43\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 2 11:54:41 2019 +0530\n\n staging: rtl8723bs: Change return type of hal_btcoex_IsBtExist()\n \n Change return type of hal_btcoex_IsBtExist from u8 to bool as its only\n possible return values are true and false.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190802062444.30384-1-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 0fc6041d7abc449198f895906f7ffac7647b603a\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Thu Jul 25 18:40:00 2019 -0500\n\n soundwire: intel: remove BIOS work-arounds\n \n the values passed by all existing BIOS are fine, let\u0027s use them as is.\n The existing code must have been needed only on early prototypes.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190725234032.21152-9-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 9104688c321a7e82ec95fddff6c98729f4e4fefd\nAuthor: Adham Abozaeid \u003cadham.abozaeid@microchip.com\u003e\nDate: Thu Aug 1 20:15:12 2019 +0000\n\n staging: wilc1000: remove unused function\n \n function wilc_resolve_disconnect_aberration isn\u0027t referenced, so\n removing it\n \n Signed-off-by: Adham Abozaeid \u003cadham.abozaeid@microchip.com\u003e\n Link: https://lore.kernel.org/r/20190801200909.12605-3-adham.abozaeid@microchip.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 4b1b8b08ab9f4db41e30e3a1bbdba1d58aa624a3\nAuthor: Adham Abozaeid \u003cadham.abozaeid@microchip.com\u003e\nDate: Thu Aug 1 20:15:11 2019 +0000\n\n staging: wilc1000: merge drv_handle and operation_mode wids\n \n wilc_set_wfi_drv_handler and wilc_set_operation_mode sends the same\n parameters to the FW, so it\u0027s better to combine them together.\n \n Kept wilc_set_wfi_drv_handler implementation since it sends all the\n required parameters, and renamed it to wilc_set_operation_mode to be\n more descriptive.\n \n Signed-off-by: Adham Abozaeid \u003cadham.abozaeid@microchip.com\u003e\n Link: https://lore.kernel.org/r/20190801200909.12605-2-adham.abozaeid@microchip.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b7a40c32f18a88b2a1e894a4406cf9d97423c078\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Aug 2 11:39:13 2019 +0530\n\n staging: wlan-ng: Remove function prism2mib_excludeunencrypted()\n \n Remove function prism2mib_excludeunencrypted as all it does is call\n prism2mib_flag.\n Modify call sites accordingly.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190802060913.30241-1-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 129d0561a3e81836aeeba2bbdb89249c0012af24\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Jul 31 15:42:07 2019 +0200\n\n staging: bcm2835-camera: Spelling s/conencted/connected/\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Link: https://lore.kernel.org/r/20190731134207.18549-1-geert+renesas@glider.be\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a5589269e660a14ab8580235909100f245d73cb9\nAuthor: Wei Yongjun \u003cweiyongjun1@huawei.com\u003e\nDate: Fri Aug 2 01:31:49 2019 +0000\n\n staging: fsl-dpaa2/ethsw: Remove useless set memory to zero use memset()\n \n The memory return by kzalloc() has already be set to zero, so remove\n useless memset(0).\n \n Signed-off-by: Wei Yongjun \u003cweiyongjun1@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190802013149.80952-1-weiyongjun1@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 0b800f3dbae710ed72ad544d17fd186ea5b5e504\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:52 2019 +0800\n\n staging: erofs: update Kconfig\n \n Keep in line with erofs-outofstaging patchset:\n - turn on CONFIG_EROFS_FS_ZIP by default;\n - turn on CONFIG_EROFS_FS_SECURITY by default suggested by David;\n - update Kconfig description.\n \n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-23-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit bc33d9f35da87cccd73fe5d3e007acb084411309\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:51 2019 +0800\n\n staging: erofs: update super.c\n \n Keep in line with erofs-outofstaging patchset:\n - \"Chao Yu\" is most commonly used in Linux community;\n - quoted string split across lines.\n \n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-22-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 2bb90cc275e242eb3965a9091c359c2c8327a1e7\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:50 2019 +0800\n\n staging: erofs: tidy up utils.c\n \n Keep in line with erofs-outofstaging patchset:\n - Update comments in erofs_try_to_release_workgroup;\n - code style cleanup.\n \n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-21-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 4279f3f9889f26bcdc4d35c0bf868a08e09470c7\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:49 2019 +0800\n\n staging: erofs: turn cache strategies into mount options\n \n Kill all kconfig cache strategies and turn them into mount options\n \"cache_strategy\u003d{disable|readahead|readaround}\".\n \n As the first step, cached pages can still be usable after cache\n is disabled by remounting, and these pages will be fallen out\n over time, which can be refined in the later version if some\n requirement is needed. Update related document as well.\n \n Suggested-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-20-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit eace994a116c9704738be3388f5777c740d9558d\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:48 2019 +0800\n\n staging: erofs: remove clusterbits in sbi\n \n clustersize can now be set on per-file basis\n rather than per-filesystem basis.\n \n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-19-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 97e86a858bc360c9cded74ce9a06595d40f7ad81\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:47 2019 +0800\n\n staging: erofs: tidy up decompression frontend\n \n Although this patch has an amount of changes, it is hard to\n separate into smaller patches.\n \n Most changes are due to structure renaming for better understand\n and straightforward,\n z_erofs_vle_workgroup to z_erofs_pcluster\n since it represents a physical cluster;\n z_erofs_vle_work to z_erofs_collection\n since it represents a collection of logical pages;\n z_erofs_vle_work_builder to z_erofs_collector\n since it\u0027s used to fill z_erofs_{pcluster,collection}.\n \n struct z_erofs_vle_work_finder has no extra use compared with\n struct z_erofs_collector, delete it.\n \n FULL_LENGTH bit is integrated into .length of pcluster so that it\n can be updated with the corresponding length change in atomic.\n \n Minor, add comments for better description.\n \n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-18-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit febde7b3ac4de7d12f03a555e400d08ae3237a01\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:46 2019 +0800\n\n staging: erofs: remove redundant braces in inode.c\n \n Remove redundant braces in inode.c\n since these are all single statements.\n \n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-17-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 046d64e1189fea79eb8705015ccc621a2f25e1da\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:45 2019 +0800\n\n staging: erofs: tidy up zpvec.h\n \n - use shorter function names:\n z_erofs_pagevec_enqueue and z_erofs_pagevec_dequeue;\n - minor code cleanup.\n \n In order to keep in line with erofs-outofstaging patchset.\n \n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-16-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 46c2d1494332ece4f136d167b62afe91a63796fb\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:44 2019 +0800\n\n staging: erofs: kill CONFIG_EROFS_FS_USE_VM_MAP_RAM\n \n Turn into a module parameter (\"use_vmap\") as it\n can be set at runtime.\n \n Suggested-by: David Sterba \u003cdsterba@suse.cz\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-15-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b25a15191c5cf090f79fd220238b93f30b022b0d\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:43 2019 +0800\n\n staging: erofs: refine erofs_allocpage()\n \n remove duplicated code in decompressor by introducing\n failable erofs_allocpage().\n \n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-14-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 8494c29ffe22dcd166a86b490dce10ecf1c20065\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:42 2019 +0800\n\n staging: erofs: drop __GFP_NOFAIL for managed inode\n \n For historical reasons, __GFP_NOFAIL was set for managed inode.\n It\u0027s no need using that since EROFS can handle it properly.\n \n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-13-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 8f7acdae2cd4b843d1fcb49e449cc7860556e3ef\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:41 2019 +0800\n\n staging: erofs: kill all failure handling in fill_super()\n \n .kill_sb() will do that instead in order to remove duplicated code.\n \n Note that the initialzation of managed_cache is now moved\n after s_root is assigned since it\u0027s more preferred to iput()\n in .put_super() and all inodes should be evicted before\n the end of generic_shutdown_super(sb).\n \n Suggested-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-12-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 9e794de573d7d54e3742628227e585bd90acc38a\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:40 2019 +0800\n\n staging: erofs: kill sbi-\u003edev_name\n \n As Al said, \"the only use of sbi-\u003edev_name is debugging\n printks and all of those have sb-\u003es_id available, with\n device name stored in there. Which makes the whole\n thing bloody weird\".\n \n sbi-\u003edev_name was used for our debugging use and it\u0027s\n better to just use s_id in community and delete\n the whole erofs_mount_private stuff.\n \n Suggested-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-11-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 22fe04a77d104996d838a487a1d031eda8d439b7\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:39 2019 +0800\n\n staging: erofs: clean up shrinker stuffs\n \n - rename erofs_register_super / erofs_unregister_super\n to erofs_shrinker_register / erofs_shrinker_unregister;\n - fold the only erofs_shrink_workstation external call\n to erofs_shrinker_unregister;\n - localize erofs_shrink_workstation;\n - localize erofs_shrinker_info by introducing\n erofs_init_shrinker and erofs_exit_shrinker.\n \n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-10-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit beb5f3c4a1f8daa9175f0044d2fa4185c7813a30\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:38 2019 +0800\n\n staging: erofs: kill CONFIG_EROFS_FS_IO_MAX_RETRIES\n \n CONFIG_EROFS_FS_IO_MAX_RETRIES seems a runtime setting\n and users have no idea about the change in behaviour.\n \n Let\u0027s remove the setting currently and could turn it\n into a module parameter if it\u0027s really needed.\n \n Suggested-by: David Sterba \u003cdsterba@suse.cz\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-9-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 22ee1347a8fb6d3eba91aacad7c691dee274cc1f\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:37 2019 +0800\n\n staging: erofs: remove redundant #include \"internal.h\"\n \n Because #include \"internal.h\" is included in xattr.h\n \n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-8-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 14f362b4f405804e219ac016b9e771d8f5dff5db\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:36 2019 +0800\n\n staging: erofs: clean up internal.h\n \n Tidy up relative order of variables / declarations in internal.h,\n move some local static functions out into other files and\n add tags at the end of #endif acrossing several lines.\n \n No logic change.\n \n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-7-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b1897c60b14dc565e9a308c09bbb696faf9d1944\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:35 2019 +0800\n\n staging: erofs: sunset erofs_workstn_{lock,unlock}\n \n It was used for Linux backward compatibility, and\n no use for upstream kernel.\n \n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-6-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit cead56f85baa783a1ed56bea235a4c1c2556c3fa\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:34 2019 +0800\n\n staging: erofs: keep up erofs_fs.h with erofs-outofstaging patchset\n \n The main change is to reserve all checksums except for superblock,\n since it\u0027s more useful to do block-based verity for read-only fs.\n \n Some comments change as well, which is minor.\n \n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-5-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 5734fa217f27bc08667930e862dbe536ed6a77ea\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:33 2019 +0800\n\n staging: erofs: fix dummy functions erofs_{get, list}xattr\n \n dummy functions erofs_{get,list}xattr should be inlined\n without xattr enabled.\n \n Signed-off-by: Yue Hu \u003czbestahu@gmail.com\u003e\n [ Gao Xiang : this patch was \"staging: erofs: remove needless\n dummy functions of erofs_{get,list}xattr. \"]\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-4-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 57b78c9fd9ce4df7281d36b0e642bee9b06a4071\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:32 2019 +0800\n\n staging: erofs: rename source files for better understanding\n \n Keep in line with erofs-outofstaging patchset as well, see\n https://lore.kernel.org/linux-fsdevel/20190725095658.155779-1-gaoxiang25@huawei.com/\n \n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-3-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 29b24f6ca112d168ae17c66343464a2581c573e0\nAuthor: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\nDate: Wed Jul 31 23:57:31 2019 +0800\n\n staging: erofs: update source file headers\n \n - Use the correct style for all SPDX License Identifiers;\n - Get rid of the unnecessary license boilerplate;\n - Use \"GPL-2.0-only\" instead of \"GPL-2.0\" suggested-by Stephen.\n \n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731155752.210602-2-gaoxiang25@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit be46cfba5452f7d0ccb1e74c32d44457bc34833e\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 22:42:18 2019 +0300\n\n soundwire: mipi_disco: Switch to use fwnode_property_count_uXX()\n \n Use use fwnode_property_count_uXX() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190723194218.69168-1-andriy.shevchenko@linux.intel.com\n Reviewed-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 0fbb93fce79b938643889862e74ee30213e1d5fe\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Wed Jul 31 14:43:45 2019 +0200\n\n olpc: x01: convert platform driver to use dev_groups\n \n Platform drivers now have the option to have the platform core create\n and remove any needed sysfs attribute files. So take advantage of that\n and do not register \"by hand\" a lid sysfs file.\n \n Cc: Darren Hart \u003cdvhart@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: Borislav Petkov \u003cbp@alien8.de\u003e\n Cc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\n Cc: x86@kernel.org\n Cc: platform-driver-x86@vger.kernel.org\n Acked-by: Andy Shevchenko \u003candy.shevchenko@gmail.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Link: https://lore.kernel.org/r/20190731124349.4474-7-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 5bd08a4ae3d06cf598b91444c2d727aa25045bf0\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Wed Jul 31 14:43:46 2019 +0200\n\n platform: x86: hp-wmi: convert platform driver to use dev_groups\n \n Platform drivers now have the option to have the platform core create\n and remove any needed sysfs attribute files. So take advantage of that\n and do not register \"by hand\" a bunch of sysfs files.\n \n Cc: Darren Hart \u003cdvhart@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: Borislav Petkov \u003cbp@alien8.de\u003e\n Cc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\n Cc: x86@kernel.org\n Cc: platform-driver-x86@vger.kernel.org\n Acked-by: Andy Shevchenko \u003candy.shevchenko@gmail.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Link: https://lore.kernel.org/r/20190731124349.4474-8-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit de3dacf0347e3dc6f63cd5abe944bc60077c5abf\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Wed Jul 31 14:43:49 2019 +0200\n\n video: fbdev: sm501fb: convert platform driver to use dev_groups\n \n Platform drivers now have the option to have the platform core create\n and remove any needed sysfs attribute files. So take advantage of that\n and do not register \"by hand\" a bunch of sysfs files.\n \n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-fbdev@vger.kernel.org\n Acked-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Link: https://lore.kernel.org/r/20190731124349.4474-11-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit e14018cc34d633a55e8fabd2085f1da51d6c60fb\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Wed Jul 31 14:43:48 2019 +0200\n\n video: fbdev: w100fb: convert platform driver to use dev_groups\n \n Platform drivers now have the option to have the platform core create\n and remove any needed sysfs attribute files. So take advantage of that\n and do not register \"by hand\" a bunch of sysfs files.\n \n Cc: Tony Prisk \u003clinux@prisktech.co.nz\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-fbdev@vger.kernel.org\n Acked-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Link: https://lore.kernel.org/r/20190731124349.4474-10-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 0abd02ede7dfd0f3ea7a6ac68e00ca6946827889\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Wed Jul 31 14:43:47 2019 +0200\n\n video: fbdev: wm8505fb: convert platform driver to use dev_groups\n \n Platform drivers now have the option to have the platform core create\n and remove any needed sysfs attribute files. So take advantage of that\n and do not register \"by hand\" a sysfs file.\n \n Cc: Tony Prisk \u003clinux@prisktech.co.nz\u003e\n Cc: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-fbdev@vger.kernel.org\n Acked-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Link: https://lore.kernel.org/r/20190731124349.4474-9-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 43b9ac937be6f2afb2f0cf060d40cea365c09d07\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Wed Jul 31 14:43:44 2019 +0200\n\n firmware: arm_scpi: convert platform driver to use dev_groups\n \n Platform drivers now have the option to have the platform core create\n and remove any needed sysfs attribute files. So take advantage of that\n and do not register \"by hand\" a sysfs group of attributes.\n \n Acked-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Link: https://lore.kernel.org/r/20190731124349.4474-6-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 3c0448d507d42605ab6811bcd61c1cd94390d2d8\nAuthor: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\nDate: Thu Aug 1 22:47:10 2019 +0200\n\n spi: atmel: add tracing to custom .transfer_one_message callback\n \n Driver specific implementations for .transfer_one_message need to call\n the tracing stuff themself. This is necessary to make spi tracing\n actually useful.\n \n Signed-off-by: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\n Link: https://lore.kernel.org/r/20190801204710.27309-1-uwe@kleine-koenig.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 6b8ac10e0dd4b49eda513c1aa5045b3b1660d350\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:41 2019 -0700\n\n spi: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Mark Brown \u003cbroonie@kernel.org\u003e\n Cc: linux-spi@vger.kernel.org\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190730181557.90391-42-swboyd@chromium.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 4ff13d00ebd5533947b083c66d44da3243a2d559\nAuthor: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nDate: Thu Aug 1 13:13:53 2019 +0200\n\n spi: Reduce kthread priority\n \n The SPI thingies request FIFO-99 by default, reduce this to FIFO-50.\n \n FIFO-99 is the very highest priority available to SCHED_FIFO and\n it not a suitable default; it would indicate the SPI work is the\n most important work on the machine.\n \n Cc: Benson Leung \u003cbleung@chromium.org\u003e\n Cc: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Cc: Guenter Roeck \u003cgroeck@chromium.org\u003e\n Cc: Mark Brown \u003cbroonie@kernel.org\u003e\n Cc: linux-spi@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Reviewed-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190801111541.917256884@infradead.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 47241933b43dec5b10d067e6e094f4d31a1cf3e7\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:37 2019 -0700\n\n regulator: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Liam Girdwood \u003clgirdwood@gmail.com\u003e\n Cc: Mark Brown \u003cbroonie@kernel.org\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190730181557.90391-38-swboyd@chromium.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit cf9441adb1a35506d7606866c382b9d8614169b5\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:49 2019 -0700\n\n ASoC: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Liam Girdwood \u003clgirdwood@gmail.com\u003e\n Cc: Mark Brown \u003cbroonie@kernel.org\u003e\n Cc: Jaroslav Kysela \u003cperex@perex.cz\u003e\n Cc: Takashi Iwai \u003ctiwai@suse.com\u003e\n Cc: alsa-devel@alsa-project.org\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190730181557.90391-50-swboyd@chromium.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 0682e005d1705a740a36860cc75199a411f46025\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Wed Jul 31 14:43:41 2019 +0200\n\n uio: uio_fsl_elbc_gpcm: convert platform driver to use dev_groups\n \n Platform drivers now have the option to have the platform core create\n and remove any needed sysfs attribute files. So take advantage of that\n and do not register \"by hand\" a sysfs group of attributes.\n \n Link: https://lore.kernel.org/r/20190731124349.4474-3-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d8af05ff38ae7a42819b285ffef314942414ef8b\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 2 11:00:15 2019 +0100\n\n drm/i915: Allow sharing the idle-barrier from other kernel requests\n \n By placing our idle-barriers in the i915_active fence tree, we expose\n those for reuse by other components that are issuing requests along the\n kernel_context. Reusing the proto-barrier active_node is perfectly fine\n as the new request implies a context-switch, and so an opportune point\n to run the idle-barrier. However, the proto-barrier is not equivalent\n to a normal active_node and care must be taken to avoid dereferencing the\n ERR_PTR used as its request marker.\n \n v2: Comment the more egregious cheek\n v3: A glossary!\n \n Reported-by: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Fixes: ce476c80b8bf (\"drm/i915: Keep contexts pinned until after the next kernel context switch\")\n Fixes: a9877da2d629 (\"drm/i915/oa: Reconfigure contexts on the fly\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190802100015.1281-1-chris@chris-wilson.co.uk\n\ncommit 23b6904442d08b7dbed7622ed33b236d41a3aa8b\nAuthor: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nDate: Wed Jul 31 14:43:40 2019 +0200\n\n driver core: add dev_groups to all drivers\n \n Add the ability for the driver core to create and remove a list of\n attribute groups automatically when the device is bound/unbound from a\n specific driver.\n \n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n Tested-by: Richard Gong \u003crichard.gong@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190731124349.4474-2-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f4160faac98c36556231621b03b44b646508b08e\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Thu Aug 1 20:39:38 2019 +0300\n\n gpio: pca953x: Drop %s for constant string literals\n \n There is no need to use %s for constant string literals\n w/o special characters inside.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n\ncommit 6dd6a2d27e5a0d192b542389f4c9a035d1b77b8e\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Thu Aug 1 20:39:37 2019 +0300\n\n gpio: pca953x: Remove explicit comparison with 0\n \n There is no need to explicitly compare return code with 0.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n\ncommit 63b484c2a69a4a98a22e072ea2ad1317a85e3abf\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Thu Aug 1 20:39:36 2019 +0300\n\n gpio: pca953x: Use GENMASK() consistently\n \n Use GENMASK() macro for all definitions where it\u0027s appropriate.\n No functional change intended.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n\ncommit 8eeb467d3b4f8a5f05e49dd88e7674a2d6c0f346\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Thu Aug 1 20:39:35 2019 +0300\n\n gpio: pca953x: Switch to use device_get_match_data()\n \n Instead of open coded variants, switch to direct use of\n device_get_match_data().\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n\ncommit 576f05865581f82ac988ffec70e4e2ebd31165db\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jul 30 12:21:51 2019 +0100\n\n drm/i915: Flush extra hard after writing relocations through the GTT\n \n Recently discovered in commit bdae33b8b82b (\"drm/i915: Use maximum write\n flush for pwrite_gtt\") was that we needed to our full write barrier\n before changing the GGTT PTE to ensure that our indirect writes through\n the GTT landed before the PTE changed (and the writes end up in a\n different page). That also applies to our GGTT relocation path.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: stable@vger.kernel.org\n Reviewed-by: Prathap Kumar Valsan \u003cprathap.kumar.valsan@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730112151.5633-4-chris@chris-wilson.co.uk\n\ncommit 51fbd8de87dcf09f3929e5438e4344bea4338990\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Aug 2 00:36:16 2019 +0100\n\n drm/i915/pmu: Atomically acquire the gt_pm wakeref\n \n Currently, we only sample if the intel_gt is awake, but we acquire our\n own runtime_pm wakeref. Since intel_gt has transitioned to tracking its\n own wakeref, we can atomically test and acquire that wakeref instead.\n \n v2: Take engine-\u003ewakeref for engine sampling\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190801233616.23007-1-chris@chris-wilson.co.uk\n\ncommit 7beb620f805dcb6757b5e7870bebd75e3e9b37e0\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Thu Aug 1 16:44:39 2019 +0800\n\n gpio: mxc: Use devm_clk_get_optional instead of devm_clk_get\n \n i.MX SoC\u0027s GPIO clock is optional, so it is better to use\n devm_clk_get_optional instead of devm_clk_get.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n\ncommit 4c349c7dafad613366574b89a9245f07504822ad\nAuthor: Vijay Khemka \u003cvijaykhemka@fb.com\u003e\nDate: Tue Jul 23 10:58:20 2019 -0700\n\n ARM: dts: aspeed: tiogapass: Add Riser card\n \n Added i2c mux for riser card and multiple ava card and its sensor\n components for Facebook tiogapass platform\n \n Signed-off-by: Vijay Khemka \u003cvijaykhemka@fb.com\u003e\n Signed-off-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n\ncommit 46a9f947b3f65e488d03056765523291dc8bf4f6\nAuthor: Vijay Khemka \u003cvijaykhemka@fb.com\u003e\nDate: Thu Jul 25 10:45:11 2019 -0700\n\n ARM: dts: aspeed: tiogapass: Move battery sensor\n \n Moved adc7 hwmon battery sensor to correct label to be read\n by single applications for all adc sensors.\n \n Signed-off-by: Vijay Khemka \u003cvijaykhemka@fb.com\u003e\n Signed-off-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n\ncommit b6fcab14ff027c4d54505d9cdff41535c3daf077\nAuthor: Tony W Wang-oc \u003cTonyWWang-oc@zhaoxin.com\u003e\nDate: Fri Aug 2 03:04:08 2019 +0000\n\n ALSA: hda: Add support of Zhaoxin controller\n \n Add the new PCI ID 0x1d17 0x3288 Zhaoxin controller support\n \n Signed-off-by: Tony W Wang-oc \u003cTonyWWang-oc@zhaoxin.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 3bb4aefd2aed628452b8ce544b76abda0ce531f0\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Thu Aug 1 17:28:24 2019 +0100\n\n ALSA: isa/wavefront: remove redundant assignment to pointer bptr\n \n The pointer bptr is being assigned a value that is never read\n and it is being updated in the next statement with a new value.\n The initialization is redundant and can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 155e4db324874d22aa8d3892da2a37fb39c89578\nAuthor: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nDate: Thu Aug 1 12:11:33 2019 +1000\n\n asm-generic: Remove redundant arch-specific rules for simd.h\n \n Now that simd.h is in include/asm-generic/Kbuild we don\u0027t need\n the arch-specific Kbuild rules for them.\n \n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Fixes: 82cb54856874 (\"asm-generic: make simd.h a mandatory...\")\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n Acked-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Acked-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 176435ad2ac7733014d6429369130ddb78f14ee0\nAuthor: Horia Geantă \u003choria.geanta@nxp.com\u003e\nDate: Sun Jul 28 22:26:38 2019 +0300\n\n crypto: caam - defer probing until QMan is available\n \n When QI (Queue Interface) support is enabled on DPAA 1.x platforms,\n defer probing if dependencies (QMan drivers) are not available yet.\n \n Signed-off-by: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 2ef540476e0077ca2f9992fde9a014891000c4b5\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Mon Jul 29 00:32:36 2019 +0900\n\n hwrng: timeriomem - add include guard to timeriomem-rng.h\n \n Add a header include guard just in case.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit f9981bc53825de2c001659d35c870786771f1e90\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Fri Jul 26 22:19:07 2019 +0200\n\n crypto: api - Remove redundant #ifdef in crypto_yield()\n \n While looking at CONFIG_PREEMPT dependencies treewide the #ifdef in\n crypto_yield() matched.\n \n CONFIG_PREEMPT and CONFIG_PREEMPT_VOLUNTARY are mutually exclusive so the\n extra !CONFIG_PREEMPT conditional is redundant.\n \n cond_resched() has only an effect when CONFIG_PREEMPT_VOLUNTARY is set,\n otherwise it\u0027s a stub which the compiler optimizes out.\n \n Remove the whole conditional.\n \n No functional change.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: linux-crypto@vger.kernel.org\n Cc: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 3e75241be8081f22f7382f4041496b10a4d9aed0\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Thu Jul 25 16:01:55 2019 +0800\n\n hwrng: drivers - Use device-managed registration API\n \n Use devm_hwrng_register to simplify the implementation.\n Manual unregistration and some remove functions can be\n removed now.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Acked-by: Łukasz Stelmach \u003cl.stelmach@samsung.com\u003e\n Acked-by: Ludovic Desroches \u003cludovic.desroches@microchip.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit dec0fb3946c478e7bb6165a1b7a03092ceefd419\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Wed Jul 24 20:51:55 2019 +0200\n\n crypto: jitterentropy - build without sanitizer\n \n Recent clang-9 snapshots double the kernel stack usage when building\n this file with -O0 -fsanitize\u003dkernel-hwaddress, compared to clang-8\n and older snapshots, this changed between commits svn364966 and\n svn366056:\n \n crypto/jitterentropy.c:516:5: error: stack frame size of 2640 bytes in function \u0027jent_entropy_init\u0027 [-Werror,-Wframe-larger-than\u003d]\n int jent_entropy_init(void)\n ^\n crypto/jitterentropy.c:185:14: error: stack frame size of 2224 bytes in function \u0027jent_lfsr_time\u0027 [-Werror,-Wframe-larger-than\u003d]\n static __u64 jent_lfsr_time(struct rand_data *ec, __u64 time, __u64 loop_cnt)\n ^\n \n I prepared a reduced test case in case any clang developers want to\n take a closer look, but from looking at the earlier output it seems\n that even with clang-8, something was very wrong here.\n \n Turn off any KASAN and UBSAN sanitizing for this file, as that likely\n clashes with -O0 anyway. Turning off just KASAN avoids the warning\n already, but I suspect both of these have undesired side-effects\n for jitterentropy.\n \n Link: https://godbolt.org/z/fDcwZ5\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 3bd4952853d69659e58d1d4a4964383cb536b5be\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Wed Jul 24 13:24:33 2019 +0530\n\n crypto: nx - Add of_node_put() before return in 842\n \n Each iteration of for_each_compatible_node puts the previous node, but\n in the case of a return from the middle of the loop, there is no put,\n thus causing a memory leak. Add an of_node_put before the return.\n Issue found with Coccinelle.\n \n Acked-by: Stewart Smith \u003cstewart@linux.ibm.com\u003e\n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 939f146b89c22c387d51f69b27ef031664f64cbe\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Tue Jul 23 20:43:44 2019 +0900\n\n crypto: user - fix potential warnings in cryptouser.h\n \n Function definitions in headers are usually marked as \u0027static inline\u0027.\n \n Since \u0027inline\u0027 is missing for crypto_reportstat(), if it were not\n referenced from a .c file that includes this header, it would produce\n a warning.\n \n Also, \u0027struct crypto_user_alg\u0027 is not declared in this header.\n \n I included \u003clinux/crytouser.h\u003e instead of adding the forward declaration\n as suggested [1].\n \n Detected by compile-testing this header as a standalone unit:\n \n ./include/crypto/internal/cryptouser.h:6:44: warning: ‘struct crypto_user_alg’ declared inside parameter list will not be visible outside of this definition or declaration\n struct crypto_alg *crypto_alg_match(struct crypto_user_alg *p, int exact);\n ^~~~~~~~~~~~~~~\n ./include/crypto/internal/cryptouser.h:11:12: warning: ‘crypto_reportstat’ defined but not used [-Wunused-function]\n static int crypto_reportstat(struct sk_buff *in_skb, struct nlmsghdr *in_nlh, struct nlattr **attrs)\n ^~~~~~~~~~~~~~~~~\n \n [1] https://lkml.org/lkml/2019/6/13/1121\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 68289c634344431d6f97480eb5384516f37e39b8\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Tue Jul 23 20:43:43 2019 +0900\n\n crypto: add header include guards\n \n Add header include guards in case they are included multiple times.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 1c0ab408bb6e16285fcddc9b4ce74507081d053f\nAuthor: Vakul Garg \u003cvakul.garg@nxp.com\u003e\nDate: Tue Jul 23 09:14:24 2019 +0000\n\n crypto: caam/qi2 - Add printing dpseci fq stats using debugfs\n \n Add support of printing the dpseci frame queue statistics using debugfs.\n \n Signed-off-by: Vakul Garg \u003cvakul.garg@nxp.com\u003e\n Reviewed-by: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 1b82feb6c5e1996513d0fb0bbb475417088b4954\nAuthor: Alexander Sverdlin \u003calexander.sverdlin@nokia.com\u003e\nDate: Tue Jul 23 07:24:01 2019 +0000\n\n crypto: qat - Silence smp_processor_id() warning\n \n It seems that smp_processor_id() is only used for a best-effort\n load-balancing, refer to qat_crypto_get_instance_node(). It\u0027s not feasible\n to disable preemption for the duration of the crypto requests. Therefore,\n just silence the warning. This commit is similar to e7a9b05ca4\n (\"crypto: cavium - Fix smp_processor_id() warnings\").\n \n Silences the following splat:\n BUG: using smp_processor_id() in preemptible [00000000] code: cryptomgr_test/2904\n caller is qat_alg_ablkcipher_setkey+0x300/0x4a0 [intel_qat]\n CPU: 1 PID: 2904 Comm: cryptomgr_test Tainted: P O 4.14.69 #1\n ...\n Call Trace:\n dump_stack+0x5f/0x86\n check_preemption_disabled+0xd3/0xe0\n qat_alg_ablkcipher_setkey+0x300/0x4a0 [intel_qat]\n skcipher_setkey_ablkcipher+0x2b/0x40\n __test_skcipher+0x1f3/0xb20\n ? cpumask_next_and+0x26/0x40\n ? find_busiest_group+0x10e/0x9d0\n ? preempt_count_add+0x49/0xa0\n ? try_module_get+0x61/0xf0\n ? crypto_mod_get+0x15/0x30\n ? __kmalloc+0x1df/0x1f0\n ? __crypto_alloc_tfm+0x116/0x180\n ? crypto_skcipher_init_tfm+0xa6/0x180\n ? crypto_create_tfm+0x4b/0xf0\n test_skcipher+0x21/0xa0\n alg_test_skcipher+0x3f/0xa0\n alg_test.part.6+0x126/0x2a0\n ? finish_task_switch+0x21b/0x260\n ? __schedule+0x1e9/0x800\n ? __wake_up_common+0x8d/0x140\n cryptomgr_test+0x40/0x50\n kthread+0xff/0x130\n ? cryptomgr_notify+0x540/0x540\n ? kthread_create_on_node+0x70/0x70\n ret_from_fork+0x24/0x50\n \n Fixes: ed8ccaef52 (\"crypto: qat - Add support for SRIOV\")\n Cc: stable@vger.kernel.org\n Signed-off-by: Alexander Sverdlin \u003calexander.sverdlin@nokia.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 01970282a4ce0a049eddeb85bbe677e765111ee5\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 15:19:36 2019 +0800\n\n crypto: atmel-sha204a - Use device-managed registration API\n \n Use devm_hwrng_register to get rid of manual\n unregistration.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Acked-by: Ludovic Desroches \u003cludovic.desroches@microchip.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit fdc0417be5260d5f2aa796ed5e1ac98d327888fc\nAuthor: Tao Ren \u003ctaoren@fb.com\u003e\nDate: Thu Aug 1 21:10:10 2019 -0700\n\n ARM: dts: aspeed: Add Facebook Wedge100 BMC\n \n Add initial version of device tree for Facebook Wedge100 AST2400 BMC\n platform.\n \n Signed-off-by: Tao Ren \u003ctaoren@fb.com\u003e\n Reviewed-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Signed-off-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n\ncommit 5cb98b41abb38249eef04fa2884660cd772a0520\nAuthor: Tao Ren \u003ctaoren@fb.com\u003e\nDate: Thu Aug 1 21:06:50 2019 -0700\n\n ARM: dts: aspeed: Add Facebook Wedge40 BMC\n \n Add initial version of device tree for Facebook Wedge40 AST2400 BMC\n platform.\n \n Signed-off-by: Tao Ren \u003ctaoren@fb.com\u003e\n Reviewed-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Signed-off-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n\ncommit c9f1fd4f2f74f322d5bdc4ec2c6a38ab7462967b\nAuthor: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nDate: Fri Aug 2 13:31:35 2019 +1000\n\n Revert \"crypto: aegis128 - add support for SIMD acceleration\"\n \n This reverts commit ecc8bc81f2fb3976737ef312f824ba6053aa3590\n (\"crypto: aegis128 - provide a SIMD implementation based on NEON\n intrinsics\") and commit 7cdc0ddbf74a19cecb2f0e9efa2cae9d3c665189\n (\"crypto: aegis128 - add support for SIMD acceleration\").\n \n They cause compile errors on platforms other than ARM because\n the mechanism to selectively compile the SIMD code is broken.\n \n Repoted-by: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 518ea582cb6e74af54c3b9c38b3183fbf1937d11\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Thu Aug 1 17:23:30 2019 +0100\n\n drm/i915/pmu: Make get_rc6 take intel_gt\n \n RC6 is a GT state so make the function parameter reflect that.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190801162330.2729-4-tvrtko.ursulin@linux.intel.com\n\ncommit 08ce5c64b25d0d434ef55fed0e0a3d7b526d793b\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Thu Aug 1 17:23:29 2019 +0100\n\n drm/i915/pmu: Convert sampling to gt\n \n Engines and frequencies are a GT thing so adjust sampling routines to\n match.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190801162330.2729-3-tvrtko.ursulin@linux.intel.com\n\ncommit 28fba0961de40d401070608c668de76fba848ea1\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Thu Aug 1 17:23:28 2019 +0100\n\n drm/i915/pmu: Convert engine sampling to uncore mmio\n \n Drops one macro using implicit dev_priv.\n \n v2:\n * Use ENGINE_READ_FW. (Chris)\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190801162330.2729-2-tvrtko.ursulin@linux.intel.com\n\ncommit 908091c85003d00742b215c996646e37c750f88b\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Thu Aug 1 17:23:27 2019 +0100\n\n drm/i915/pmu: Make more struct i915_pmu centric\n \n Just tidy the code a bit by removing a sea of overly verbose i915-\u003epmu.*.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190801162330.2729-1-tvrtko.ursulin@linux.intel.com\n\ncommit 7c5b42055964f587e55bd87ef334c3a27e95d144\nAuthor: Jon Maloy \u003cjon.maloy@ericsson.com\u003e\nDate: Tue Jul 30 16:23:18 2019 +0200\n\n tipc: reduce risk of wakeup queue starvation\n \n In commit 365ad353c256 (\"tipc: reduce risk of user starvation during\n link congestion\") we allowed senders to add exactly one list of extra\n buffers to the link backlog queues during link congestion (aka\n \"oversubscription\"). However, the criteria for when to stop adding\n wakeup messages to the input queue when the overload abates is\n inaccurate, and may cause starvation problems during very high load.\n \n Currently, we stop adding wakeup messages after 10 total failed attempts\n where we find that there is no space left in the backlog queue for a\n certain importance level. The counter for this is accumulated across all\n levels, which may lead the algorithm to leave the loop prematurely,\n although there may still be plenty of space available at some levels.\n The result is sometimes that messages near the wakeup queue tail are not\n added to the input queue as they should be.\n \n We now introduce a more exact algorithm, where we keep adding wakeup\n messages to a level as long as the backlog queue has free slots for\n the corresponding level, and stop at the moment there are no more such\n slots or when there are no more wakeup messages to dequeue.\n \n Fixes: 365ad35 (\"tipc: reduce risk of user starvation during link congestion\")\n Reported-by: Tung Nguyen \u003ctung.q.nguyen@dektech.com.au\u003e\n Acked-by: Ying Xue \u003cying.xue@windriver.com\u003e\n Signed-off-by: Jon Maloy \u003cjon.maloy@ericsson.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit df87b8fcf8fe658518fa8e75c9379e5bdfaf908c\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Mon Jul 8 16:12:28 2019 -0700\n\n fm10k: reduce scope of the ring variable\n \n Reduce the scope of the ring local variable in the fm10k_assign_l2_accel\n function.\n \n This was detected by cppcheck and resolves the following warning\n produced by that tool:\n \n [fm10k_netdev.c:1447]: (style) The scope of the variable \u0027ring\u0027 can be\n reduced.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 8e03f26b716f87c3bd1da212ec0fb3869ea47e64\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Mon Jul 8 16:12:27 2019 -0700\n\n fm10k: reduce the scope of the result local variable\n \n Reduce the scope of the result local variable in the\n fm10k_iov_msg_lport_state_pf function.\n \n This was detected by cppcheck and resolves the following warning\n produced by that tool:\n \n [fm10k_pf.c:1435]: (style) The scope of the variable \u0027result\u0027 can be\n reduced.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 71974d7e8556860c5ebe9bb37d598975f1e0c4e3\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Mon Jul 8 16:12:26 2019 -0700\n\n fm10k: reduce the scope of the local msg variable\n \n The msg variable in the fm10k_mbx_validate_msg_size and\n fm10k_sm_mbx_transmit functions is only used within the do {} loop\n scope. Reduce its scope only to where it is used.\n \n This was detected by cppcheck, and resolves the following warnings\n produced by that tool:\n \n [fm10k_mbx.c:299]: (style) The scope of the variable \u0027msg\u0027 can be reduced.\n [fm10k_mbx.c:2004]: (style) The scope of the variable \u0027msg\u0027 can be reduced.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit d56b47791d3464178be2aa2d61db694331d50d5a\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Mon Jul 8 16:12:25 2019 -0700\n\n fm10k: reduce the scope of the local i variable\n \n Reduce the scope of the local loop variable in the\n fm10k_check_hang_subtask function.\n \n This was detected by cppcheck and resolves the following warning\n produced by that tool:\n \n [driver/fm10k_pci.c:852]: (style) The scope of the variable \u0027i\u0027 can be\n reduced.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit b731d079e1baa18137fff3daed59c7f0d5ab2424\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Mon Jul 8 16:12:24 2019 -0700\n\n fm10k: reduce the scope of the err variable\n \n Reduce the scope of the local variable err in the fm10k_detach_subtask\n function.\n \n This was detected by cppcheck and resolves the following warning\n produced by that tool:\n \n [fm10k_pci.c:403]: (style) The scope of the variable \u0027err\u0027 can be reduced.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit fb381e60b8eb1eea61db07bdca436409229a8743\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Mon Jul 8 16:12:23 2019 -0700\n\n fm10k: reduce the scope of the tx_buffer variable\n \n The tx_buffer local variable in the function fm10k_clean_tx_ring is not\n used except inside a smaller block scope. Reduce the scope to its point\n of use.\n \n This was detected by cppcheck and resolves the following style warning\n produced by that tool:\n \n [fm10k_netdev.c:179]: (style) The scope of the variable \u0027tx_buffer\u0027 can\n be reduced.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 2b92a82fe03bd426a90b82d69fe1c340ac6b9d74\nAuthor: Jordan Justen \u003cjordan.l.justen@intel.com\u003e\nDate: Thu Jul 25 17:24:11 2019 -0700\n\n drm/i915/tgl: allow the reg_read ioctl to read the RCS TIMESTAMP register\n \n This enables the Mesa driver to advertise support for ARB_timer_query,\n and thus an OpenGL version higher than 3.2.\n \n Based on the ICL patch by Paulo Zanoni and CNL patch by Nanley Chery.\n \n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Jordan Justen \u003cjordan.l.justen@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190726002412.5827-3-lucas.demarchi@intel.com\n\ncommit 5d571068f71ebf41f434fa3f9b4ebcc9849d5efe\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Thu Jul 25 17:24:10 2019 -0700\n\n drm/i915/tgl: Add and use new DC5 and DC6 residency counter registers\n \n Tiger Lake has a new register offset for DC5 and DC6 residency counters.\n \n v2:\n - Rename registers since they are not in the CSR memory range\n (requested by Anshuman)\n - Fix type (requested by Matthew)\n \n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Anshuman Gupta \u003canshuman.gupta@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190726002412.5827-2-lucas.demarchi@intel.com\n\ncommit 60013d5d2b4031e6027005e5e2dcb6ee6da6b186\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Wed Jul 10 08:30:00 2019 -0700\n\n rcutorture: Aggressive forward-progress tests shouldn\u0027t block shutdown\n \n The more aggressive forward-progress tests can interfere with rcutorture\n shutdown, resulting in false-positive diagnostics. This commit therefore\n ends any such tests 30 seconds prior to shutdown.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 77e9752ce69f36f1be4e366373727fb7921f5909\nAuthor: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\nDate: Thu Jul 4 00:34:30 2019 -0400\n\n rcuperf: Make rcuperf kernel test more robust for !expedited mode\n \n It is possible that the rcuperf kernel test runs concurrently with init\n starting up. During this time, the system is running all grace periods\n as expedited. However, rcuperf can also be run for normal GP tests.\n Right now, it depends on a holdoff time before starting the test to\n ensure grace periods start later. This works fine with the default\n holdoff time however it is not robust in situations where init takes\n greater than the holdoff time to finish running. Or, as in my case:\n \n I modified the rcuperf test locally to also run a thread that did\n preempt disable/enable in a loop. This had the effect of slowing down\n init. The end result was that the \"batches:\" counter in rcuperf was 0\n causing a division by 0 error in the results. This counter was 0 because\n only expedited GPs seem to happen, not normal ones which led to the\n rcu_state.gp_seq counter remaining constant across grace periods which\n unexpectedly happen to be expedited. The system was running expedited\n RCU all the time because rcu_unexpedited_gp() would not have run yet\n from init. In other words, the test would concurrently with init\n booting in expedited GP mode.\n \n To fix this properly, this commit waits until system_state is set to\n SYSTEM_RUNNING before starting the test. This change is made just\n before kernel_init() invokes rcu_end_inkernel_boot(), and this latter\n is what turns off boot-time expediting of RCU grace periods.\n \n Signed-off-by: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 21f57546ceaf4c5537a617f55b809a843b109210\nAuthor: Denis Efremov \u003cefremov@linux.com\u003e\nDate: Thu Jul 4 15:57:19 2019 +0300\n\n torture: Remove exporting of internal functions\n \n The functions torture_onoff_cleanup() and torture_shuffle_cleanup()\n are declared static and marked EXPORT_SYMBOL_GPL(), which is at best an\n odd combination. Because these functions are not used outside of the\n kernel/torture.c file they are defined in, this commit removes their\n EXPORT_SYMBOL_GPL() marking.\n \n Fixes: cc47ae083026 (\"rcutorture: Abstract torture-test cleanup\")\n Signed-off-by: Denis Efremov \u003cefremov@linux.com\u003e\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit bd1bfc51a36f334270b886db6d8467e55fe294ca\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Sat Jun 22 14:35:59 2019 -0700\n\n rcutorture: Emulate userspace sojourn during call_rcu() floods\n \n During an actual call_rcu() flood, there would be frequent trips to\n userspace (in-kernel call_rcu() floods must be otherwise housebroken).\n Userspace execution allows a great many things to interrupt execution,\n and rcutorture needs to also allow such interruptions. This commit\n therefore causes call_rcu() floods to occasionally invoke schedule(),\n thus preventing spurious rcutorture failures due to other parts of the\n kernel becoming irate at the call_rcu() flood events.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit f4e8352928587ef8772df3d269a328efa609daaa\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Thu Jun 27 14:05:54 2019 -0700\n\n rcutorture: Test TREE03 with the threadirqs kernel boot parameter\n \n Since commit 05f415715ce45 (\"rcu: Speed up expedited GPs when interrupting\n RCU reader\") in v5.0 and through v5.1, booting with the threadirqs kernel\n boot parameter caused self-deadlocks, which can be reproduced using the\n following command on an 8-CPU system:\n \n tools/testing/selftests/rcutorture/bin/kvm.sh --duration 5 --configs \"TREE03\" --bootargs \"threadirqs\"\n \n This commit therefore adds the threadirqs kernel boot parameter to\n the TREE03 rcutorture scenario in order to more quickly detect future\n similar bugs.\n \n Link: http://lkml.kernel.org/r/20190626135447.y24mvfuid5fifwjc@linutronix.de\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n Cc: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Cc: Joel Fernandes \u003cjoel@joelfernandes.org\u003e\n\ncommit 2c667e5eae232f7f4a4fc30f58e51abdb0dc43c5\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Fri Jun 21 10:32:57 2019 -0700\n\n torture: Expand last_ts variable in kvm-test-1-run.sh\n \n The kvm-test-1-run.sh script says \u0027test -z \"last_ts\"\u0027 which always\n evaluates to true (AKA zero) regardless of the value of the last_ts shell\n variable. This commit therefore inserts the needed dollar sign (\"$\").\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit b3f3886c59f649ace424d132bd8c06e3611c71a8\nAuthor: Xiao Yang \u003cice_yangxiao@163.com\u003e\nDate: Fri May 31 23:15:45 2019 +0800\n\n rcuperf: Fix perf_type module-parameter description\n \n The rcu_bh rcuperf type was removed by commit 620d246065cd(\"rcuperf:\n Remove the \"rcu_bh\" and \"sched\" torture types\"), but it lives on in the\n MODULE_PARM_DESC() of perf_type. This commit therefore changes that\n module-parameter description to substitute srcu for rcu_bh.\n \n Signed-off-by: Xiao Yang \u003cice_yangxiao@163.com\u003e\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 9147089bee3a6b504821dd8462e2be229e6dbfae\nAuthor: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\nDate: Tue Jul 16 18:12:21 2019 -0400\n\n rcu: Remove redundant debug_locks check in rcu_read_lock_sched_held()\n \n The debug_locks flag can never be true at the end of\n rcu_read_lock_sched_held() because it is already checked by the earlier\n call todebug_lockdep_rcu_enabled(). This commit therefore removes this\n redundant check.\n \n Signed-off-by: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 0a5b99f57873e233ad42ef71e23c629f6ea1fcfe\nAuthor: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\nDate: Thu Jul 11 16:45:41 2019 -0400\n\n treewide: Rename rcu_dereference_raw_notrace() to _check()\n \n The rcu_dereference_raw_notrace() API name is confusing. It is equivalent\n to rcu_dereference_raw() except that it also does sparse pointer checking.\n \n There are only a few users of rcu_dereference_raw_notrace(). This patches\n renames all of them to be rcu_dereference_raw_check() with the \"_check()\"\n indicating sparse checking.\n \n Signed-off-by: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\n [ paulmck: Fix checkpatch warnings about parentheses. ]\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 7a432d57e080ac16645a08e93dbf1159d31f405f\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Mon Jul 8 16:12:22 2019 -0700\n\n fm10k: reduce the scope of the q_idx local variable\n \n Reduce the scope of the q_idx local variable in the fm10k_cache_ring_qos\n function.\n \n This was detected by cppcheck and resolves the following style warning\n produced by that tool:\n \n [fm10k_main.c:2016]: (style) The scope of the variable \u0027q_idx\u0027 can be\n reduced.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 57928c583ded1077c68168b0c7d34943612458da\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Mon Jul 8 16:12:21 2019 -0700\n\n fm10k: reduce the scope of local err variable\n \n Reduce the scope of the local err variable in the fm10k_iov_alloc_data\n function.\n \n This was detected by cppcheck and resolves the following style warning\n produced by that tool:\n \n [fm10k_iov.c:426]: (style) The scope of the variable \u0027err\u0027 can be reduced.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 4f9e05fb44f7fca787586f99a8b7f0da6c15eadf\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Mon Jul 8 16:12:20 2019 -0700\n\n fm10k: reduce the scope of qv local variable\n \n Reduce the scope of the qv vector pointer local variable in the\n fm10k_set_coalesce function.\n \n This was detected by cppcheck and resolves the following style warning\n produced by that tool:\n \n [fm10k_ethtool.c:658]: (style) The scope of the variable \u0027qv\u0027 can be\n reduced.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 0500873de968df6fdef5752d7bbdca317ddc220b\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Mon Jul 8 08:01:50 2019 -0700\n\n doc: Add rcutree.kthread_prio pointer to stallwarn.txt\n \n This commit adds mention of the rcutree.kthread_prio kernel boot parameter\n to the discussion of how high-priority real-time tasks can result in\n RCU CPU stall warnings. (However, this does not necessarily help when\n the high-priority real-time tasks are using dubious deadlines.)\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 3545832fc22e2316d9c289f6ba825710a268bfa6\nAuthor: Byungchul Park \u003cbyungchul.park@lge.com\u003e\nDate: Mon Jul 1 09:40:39 2019 +0900\n\n rcu: Change return type of rcu_spawn_one_boost_kthread()\n \n The return value of rcu_spawn_one_boost_kthread() is not used any longer.\n This commit therefore changes its return type from int to void, and\n removes the cast to void from its callers.\n \n Signed-off-by: Byungchul Park \u003cbyungchul.park@lge.com\u003e\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 7e210a653ec9445512534cd235cac29e7301af2a\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Fri Jun 28 17:11:10 2019 -0700\n\n srcu: Avoid srcutorture security-based pointer obfuscation\n \n Because pointer output is now obfuscated, and because what you really\n want to know is whether or not the callback lists are empty, this commit\n replaces the srcu_data structure\u0027s head callback pointer printout with\n a single character that is \".\" is the callback list is empty or \"C\"\n otherwise.\n \n This is the only remaining user of rcu_segcblist_head(), so this\n commit also removes this function\u0027s definition. It also turns out that\n rcu_segcblist_tail() no longer has any callers, so this commit removes\n that function\u0027s definition while in the area. They were both marked\n \"Interim\", and their end has come.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit fbad01af8c3bb9618848abde8054ab7e0c2330fe\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Wed Jun 19 15:42:51 2019 -0700\n\n rcu: Add destroy_work_on_stack() to match INIT_WORK_ONSTACK()\n \n The synchronize_rcu_expedited() function has an INIT_WORK_ONSTACK(),\n but lacks the corresponding destroy_work_on_stack(). This commit\n therefore adds destroy_work_on_stack().\n \n Reported-by: Andrea Arcangeli \u003caarcange@redhat.com\u003e\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n Acked-by: Andrea Arcangeli \u003caarcange@redhat.com\u003e\n\ncommit cdc694b2359d52cd6d0465d5a6263d97c786fb0c\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Thu Jun 13 15:30:49 2019 -0700\n\n rcu: Add kernel parameter to dump trace after RCU CPU stall warning\n \n This commit adds a rcu_cpu_stall_ftrace_dump kernel boot parameter, that,\n when set, causes the trace buffer to be dumped after an RCU CPU stall\n warning is printed. This kernel boot parameter is disabled by default,\n maintaining compatibility with previous behavior.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 1f3ebc8253ee56bfaa883c5114fb5569c56f6197\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue Jun 4 14:05:52 2019 -0700\n\n rcu: Restore barrier() to rcu_read_lock() and rcu_read_unlock()\n \n Commit bb73c52bad36 (\"rcu: Don\u0027t disable preemption for Tiny and Tree\n RCU readers\") removed the barrier() calls from rcu_read_lock() and\n rcu_write_lock() in CONFIG_PREEMPT\u003dn\u0026\u0026CONFIG_PREEMPT_COUNT\u003dn kernels.\n Within RCU, this commit was OK, but it failed to account for things like\n get_user() that can pagefault and that can be reordered by the compiler.\n Lack of the barrier() calls in rcu_read_lock() and rcu_read_unlock()\n can cause these page faults to migrate into RCU read-side critical\n sections, which in CONFIG_PREEMPT\u003dn kernels could result in too-short\n grace periods and arbitrary misbehavior. Please see commit 386afc91144b\n (\"spinlocks and preemption points need to be at least compiler barriers\")\n and Linus\u0027s commit 66be4e66a7f4 (\"rcu: locking and unlocking need to\n always be at least barriers\"), this last of which restores the barrier()\n call to both rcu_read_lock() and rcu_read_unlock().\n \n This commit removes barrier() calls that are no longer needed given that\n the addition of them in Linus\u0027s commit noted above. The combination of\n this commit and Linus\u0027s commit effectively reverts commit bb73c52bad36\n (\"rcu: Don\u0027t disable preemption for Tiny and Tree RCU readers\").\n \n Reported-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n Reported-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n [ paulmck: Fix embarrassing typo located by Alan Stern. ]\n\ncommit b55bd585551ed2220eefdab96b31e6f935310eec\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Thu May 30 05:39:25 2019 -0700\n\n time/tick-broadcast: Fix tick_broadcast_offline() lockdep complaint\n \n The TASKS03 and TREE04 rcutorture scenarios produce the following\n lockdep complaint:\n \n ------------------------------------------------------------------------\n \n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n WARNING: inconsistent lock state\n 5.2.0-rc1+ #513 Not tainted\n --------------------------------\n inconsistent {IN-HARDIRQ-W} -\u003e {HARDIRQ-ON-W} usage.\n migration/1/14 [HC0[0]:SC0[0]:HE1:SE1] takes:\n (____ptrval____) (tick_broadcast_lock){?...}, at: tick_broadcast_offline+0xf/0x70\n {IN-HARDIRQ-W} state was registered at:\n lock_acquire+0xb0/0x1c0\n _raw_spin_lock_irqsave+0x3c/0x50\n tick_broadcast_switch_to_oneshot+0xd/0x40\n tick_switch_to_oneshot+0x4f/0xd0\n hrtimer_run_queues+0xf3/0x130\n run_local_timers+0x1c/0x50\n update_process_times+0x1c/0x50\n tick_periodic+0x26/0xc0\n tick_handle_periodic+0x1a/0x60\n smp_apic_timer_interrupt+0x80/0x2a0\n apic_timer_interrupt+0xf/0x20\n _raw_spin_unlock_irqrestore+0x4e/0x60\n rcu_nocb_gp_kthread+0x15d/0x590\n kthread+0xf3/0x130\n ret_from_fork+0x3a/0x50\n irq event stamp: 171\n hardirqs last enabled at (171): [\u003cffffffff8a201a37\u003e] trace_hardirqs_on_thunk+0x1a/0x1c\n hardirqs last disabled at (170): [\u003cffffffff8a201a53\u003e] trace_hardirqs_off_thunk+0x1a/0x1c\n softirqs last enabled at (0): [\u003cffffffff8a264ee0\u003e] copy_process.part.56+0x650/0x1cb0\n softirqs last disabled at (0): [\u003c0000000000000000\u003e] 0x0\n \n other info that might help us debug this:\n Possible unsafe locking scenario:\n \n CPU0\n ----\n lock(tick_broadcast_lock);\n \u003cInterrupt\u003e\n lock(tick_broadcast_lock);\n \n *** DEADLOCK ***\n \n 1 lock held by migration/1/14:\n #0: (____ptrval____) (clockevents_lock){+.+.}, at: tick_offline_cpu+0xf/0x30\n \n stack backtrace:\n CPU: 1 PID: 14 Comm: migration/1 Not tainted 5.2.0-rc1+ #513\n Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS Bochs 01/01/2011\n Call Trace:\n dump_stack+0x5e/0x8b\n print_usage_bug+0x1fc/0x216\n ? print_shortest_lock_dependencies+0x1b0/0x1b0\n mark_lock+0x1f2/0x280\n __lock_acquire+0x1e0/0x18f0\n ? __lock_acquire+0x21b/0x18f0\n ? _raw_spin_unlock_irqrestore+0x4e/0x60\n lock_acquire+0xb0/0x1c0\n ? tick_broadcast_offline+0xf/0x70\n _raw_spin_lock+0x33/0x40\n ? tick_broadcast_offline+0xf/0x70\n tick_broadcast_offline+0xf/0x70\n tick_offline_cpu+0x16/0x30\n take_cpu_down+0x7d/0xa0\n multi_cpu_stop+0xa2/0xe0\n ? cpu_stop_queue_work+0xc0/0xc0\n cpu_stopper_thread+0x6d/0x100\n smpboot_thread_fn+0x169/0x240\n kthread+0xf3/0x130\n ? sort_range+0x20/0x20\n ? kthread_cancel_delayed_work_sync+0x10/0x10\n ret_from_fork+0x3a/0x50\n \n ------------------------------------------------------------------------\n \n To reproduce, run the following rcutorture test:\n \n tools/testing/selftests/rcutorture/bin/kvm.sh --duration 5 --kconfig \"CONFIG_DEBUG_LOCK_ALLOC\u003dy CONFIG_PROVE_LOCKING\u003dy\" --configs \"TASKS03 TREE04\"\n \n It turns out that tick_broadcast_offline() was an innocent bystander.\n After all, interrupts are supposed to be disabled throughout\n take_cpu_down(), and therefore should have been disabled upon entry to\n tick_offline_cpu() and thus to tick_broadcast_offline(). This suggests\n that one of the CPU-hotplug notifiers was incorrectly enabling interrupts,\n and leaving them enabled on return.\n \n Some debugging code showed that the culprit was sched_cpu_dying().\n It had irqs enabled after return from sched_tick_stop(). Which in turn\n had irqs enabled after return from cancel_delayed_work_sync(). Which is a\n wrapper around __cancel_work_timer(). Which can sleep in the case where\n something else is concurrently trying to cancel the same delayed work,\n and as Thomas Gleixner pointed out on IRC, sleeping is a decidedly bad\n idea when you are invoked from take_cpu_down(), regardless of the state\n you leave interrupts in upon return.\n \n Code inspection located no reason why the delayed work absolutely\n needed to be canceled from sched_tick_stop(): The work is not\n bound to the outgoing CPU by design, given that the whole point is\n to collect statistics without disturbing the outgoing CPU.\n \n This commit therefore simply drops the cancel_delayed_work_sync() from\n sched_tick_stop(). Instead, a new -\u003estate field is added to the tick_work\n structure so that the delayed-work handler function sched_tick_remote()\n can avoid reposting itself. A cpu_is_offline() check is also added to\n sched_tick_remote() to avoid mucking with the state of an offlined CPU\n (though it does appear safe to do so). The sched_tick_start() and\n sched_tick_stop() functions also update -\u003estate, and sched_tick_start()\n also schedules the delayed work if -\u003estate indicates that it is not\n already in flight.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n [ paulmck: Apply Peter Zijlstra and Frederic Weisbecker atomics feedback. ]\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n\ncommit 519248f36d6f3c80e176f6fa844c10d94f1f5990\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Thu May 30 05:39:25 2019 -0700\n\n lockdep: Make print_lock() address visible\n \n Security is a wonderful thing, but so is the ability to debug based on\n lockdep warnings. This commit therefore makes lockdep lock addresses\n visible in the clear.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit cb4dbbfaa1f5a190f041b174177699a009ab2ecd\nAuthor: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\nDate: Mon Jul 1 00:04:14 2019 -0400\n\n rcu: Simplify rcu_note_context_switch exit from critical section\n \n Because __rcu_read_unlock() can be preempted just before the call to\n rcu_read_unlock_special(), it is possible for a task to be preempted just\n before it would have fully exited its RCU read-side critical section.\n This would result in a needless extension of that critical section until\n that task was resumed, which might in turn result in a needlessly\n long grace period, needless RCU priority boosting, and needless\n force-quiescent-state actions. Therefore, rcu_note_context_switch()\n invokes __rcu_read_unlock() followed by rcu_preempt_deferred_qs() when\n it detects this situation. This action by rcu_note_context_switch()\n ends the RCU read-side critical section immediately.\n \n Of course, once the task resumes, it will invoke rcu_read_unlock_special()\n redundantly. This is harmless because the fact that a preemption\n happened means that interrupts, preemption, and softirqs cannot\n have been disabled, so there would be no deferred quiescent state.\n While -\u003ercu_read_lock_nesting remains less than zero, none of the\n -\u003ercu_read_unlock_special.b bits can be set, and they were all zeroed by\n the call to rcu_note_context_switch() at task-preemption time. Therefore,\n setting -\u003ercu_read_unlock_special.b.exp_hint to false has no effect.\n \n Therefore, the extra call to rcu_preempt_deferred_qs_irqrestore()\n would return immediately. With one possible exception, which is\n if an expedited grace period started just as the task was being\n resumed, which could leave -\u003eexp_deferred_qs set. This will cause\n rcu_preempt_deferred_qs_irqrestore() to invoke rcu_report_exp_rdp(),\n reporting the quiescent state, just as it should. (Such an expedited\n grace period won\u0027t affect the preemption code path due to interrupts\n having already been disabled.)\n \n But when rcu_note_context_switch() invokes __rcu_read_unlock(), it\n is doing so with preemption disabled, hence __rcu_read_unlock() will\n unconditionally defer the quiescent state, only to immediately invoke\n rcu_preempt_deferred_qs(), thus immediately reporting the deferred\n quiescent state. It turns out to be safe (and faster) to instead\n just invoke rcu_preempt_deferred_qs() without the __rcu_read_unlock()\n middleman.\n \n Because this is the invocation during the preemption (as opposed to\n the invocation just after the resume), at least one of the bits in\n -\u003ercu_read_unlock_special.b must be set and -\u003ercu_read_lock_nesting\n must be negative. This means that rcu_preempt_need_deferred_qs() must\n return true, avoiding the early exit from rcu_preempt_deferred_qs().\n Thus, rcu_preempt_deferred_qs_irqrestore() will be invoked immediately,\n as required.\n \n This commit therefore simplifies the CONFIG_PREEMPT\u003dy version of\n rcu_note_context_switch() by removing the \"else if\" branch of its\n \"if\" statement. This change means that all callers that would have\n invoked rcu_read_unlock_special() followed by rcu_preempt_deferred_qs()\n will now simply invoke rcu_preempt_deferred_qs(), thus avoiding the\n rcu_read_unlock_special() middleman when __rcu_read_unlock() is preempted.\n \n Cc: rcu@vger.kernel.org\n Cc: kernel-team@android.com\n Signed-off-by: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 87446b48748b49dd34900904649a5ec95a591699\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Fri Jun 28 11:25:26 2019 -0700\n\n rcu: Make rcu_read_unlock_special() checks match raise_softirq_irqoff()\n \n Threaded interrupts provide additional interesting interactions between\n RCU and raise_softirq() that can result in self-deadlocks in v5.0-2 of\n the Linux kernel. These self-deadlocks can be provoked in susceptible\n kernels within a few minutes using the following rcutorture command on\n an 8-CPU system:\n \n tools/testing/selftests/rcutorture/bin/kvm.sh --duration 5 --configs \"TREE03\" --bootargs \"threadirqs\"\n \n Although post-v5.2 RCU commits have at least greatly reduced the\n probability of these self-deadlocks, this was entirely by accident.\n Although this sort of accident should be rowdily celebrated on those\n rare occasions when it does occur, such celebrations should be quickly\n followed by a principled patch, which is what this patch purports to be.\n \n The key point behind this patch is that when in_interrupt() returns\n true, __raise_softirq_irqoff() will never attempt a wakeup. Therefore,\n if in_interrupt(), calls to raise_softirq*() are both safe and\n extremely cheap.\n \n This commit therefore replaces the in_irq() calls in the \"if\" statement\n in rcu_read_unlock_special() with in_interrupt() and simplifies the\n \"if\" condition to the following:\n \n if (irqs_were_disabled \u0026\u0026 use_softirq \u0026\u0026\n (in_interrupt() ||\n (exp \u0026\u0026 !t-\u003ercu_read_unlock_special.b.deferred_qs))) {\n raise_softirq_irqoff(RCU_SOFTIRQ);\n } else {\n /* Appeal to the scheduler. */\n }\n \n The rationale behind the \"if\" condition is as follows:\n \n 1. irqs_were_disabled: If interrupts are enabled, we should\n instead appeal to the scheduler so as to let the upcoming\n irq_enable()/local_bh_enable() do the rescheduling for us.\n 2. use_softirq: If this kernel isn\u0027t using softirq, then\n raise_softirq_irqoff() will be unhelpful.\n 3. a. in_interrupt(): If this returns true, the subsequent\n call to raise_softirq_irqoff() is guaranteed not to\n do a wakeup, so that call will be both very cheap and\n quite safe.\n b. Otherwise, if !in_interrupt() the raise_softirq_irqoff()\n might do a wakeup, which is expensive and, in some\n contexts, unsafe.\n i. The \"exp\" (an expedited RCU grace period is being\n blocked) says that the wakeup is worthwhile, and:\n ii. The !.deferred_qs says that scheduler locks\n cannot be held, so the wakeup will be safe.\n \n Backporting this requires considerable care, so no auto-backport, please!\n \n Fixes: 05f415715ce45 (\"rcu: Speed up expedited GPs when interrupting RCU reader\")\n Reported-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit d143b3d1cd89f6bcab67dc88160914aa3536c663\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Sat Jun 22 12:05:54 2019 -0700\n\n rcu: Simplify rcu_read_unlock_special() deferred wakeups\n \n In !use_softirq runs, we clearly cannot rely on raise_softirq() and\n its lightweight bit setting, so we must instead do some form of wakeup.\n In the absence of a self-IPI when interrupts are disabled, these wakeups\n can be delayed until the next interrupt occurs. This means that calling\n invoke_rcu_core() doesn\u0027t actually do any expediting.\n \n In this case, it is better to take the \"else\" clause, which sets the\n current CPU\u0027s resched bits and, if there is an expedited grace period\n in flight, uses IRQ-work to force the needed self-IPI. This commit\n therefore removes the \"else if\" clause that calls invoke_rcu_core().\n \n Reported-by: Scott Wood \u003cswood@redhat.com\u003e\n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit a5c0d861280dc7d596d875213e691a046b430597\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Mon Jul 8 16:12:19 2019 -0700\n\n fm10k: reduce scope of *p local variable\n \n Reduce the scope of the char *p local variable to only the block where\n it is used.\n \n This was detected by cppcheck and resolves the following style warning\n produced by that tool:\n \n [fm10k_ethtool.c:229]: (style) The scope of the variable \u0027p\u0027 can be\n reduced.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 30b1b498d7568ea8db387308087d30292fb32b8b\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Mon Jul 8 16:12:18 2019 -0700\n\n fm10k: reduce scope of the err variable\n \n Reduce the scope of the err local variable in the fm10k_dcbnl_ieee_setets\n function.\n \n This was detected using cppcheck, and resolves the following style\n warning:\n \n [fm10k_dcbnl.c:37]: (style) The scope of the variable \u0027err\u0027 can be reduced.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit fd5ef31f370a8b7000794cd8a428b349dbfbbb80\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Mon Jul 29 14:51:11 2019 -0700\n\n selftests/bpf: extend sockopt_sk selftest with TCP_CONGESTION use case\n \n Ignore SOL_TCP:TCP_CONGESTION in getsockopt and always override\n SOL_TCP:TCP_CONGESTION with \"cubic\" in setsockopt hook.\n \n Call setsockopt(SOL_TCP, TCP_CONGESTION) with short optval (\"nv\")\n to make sure BPF program has enough buffer space to replace it\n with \"cubic\".\n \n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 9babe825da769cfbd7080f95e6bddbe98ce6b95d\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Mon Jul 29 14:51:10 2019 -0700\n\n bpf: always allocate at least 16 bytes for setsockopt hook\n \n Since we always allocate memory, allocate just a little bit more\n for the BPF program in case it need to override user input with\n bigger value. The canonical example is TCP_CONGESTION where\n input string might be too small to override (nv -\u003e bbr or cubic).\n \n 16 bytes are chosen to match the size of TCP_CA_NAME_MAX and can\n be extended in the future if needed.\n \n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit b1ac6fb440d6abf77ae82667b57126c6059c7bd6\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Thu Aug 1 14:36:37 2019 -0400\n\n net: dsa: mv88e6xxx: call vtu_getnext directly in vlan_add\n \n Wrapping mv88e6xxx_vtu_getnext makes the code less easy to read and\n _mv88e6xxx_port_vlan_add is the only function requiring the preparation\n of a new VLAN entry.\n \n To simplify things up, remove the mv88e6xxx_vtu_get wrapper and\n explicit the VLAN lookup in _mv88e6xxx_port_vlan_add. This rework\n also avoids programming the broadcast entries again when changing a\n port\u0027s membership, e.g. from tagged to untagged.\n \n At the same time, rename the helper using an old underscore convention.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 5210989283c28b909e13efb61c0e298c178bb76f\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Thu Aug 1 14:36:36 2019 -0400\n\n net: dsa: mv88e6xxx: call vtu_getnext directly in vlan_del\n \n Wrapping mv88e6xxx_vtu_getnext makes the code less easy to read.\n Explicit the call to mv88e6xxx_vtu_getnext in _mv88e6xxx_port_vlan_del\n and the return value expected by switchdev in case of software VLANs.\n \n At the same time, rename the helper using an old underscore convention.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 5ef8d249f8743325832a99fdaf997a28cbd0f78b\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Thu Aug 1 14:36:35 2019 -0400\n\n net: dsa: mv88e6xxx: call vtu_getnext directly in db load/purge\n \n mv88e6xxx_vtu_getnext is simple enough to call it directly in the\n mv88e6xxx_port_db_load_purge function and explicit the return code\n expected by switchdev for software VLANs when an hardware VLAN does\n not exist.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 425d2d37aba6710c6fe3ad9f8c6755c49986f3e5\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Thu Aug 1 14:36:34 2019 -0400\n\n net: dsa: mv88e6xxx: explicit entry passed to vtu_getnext\n \n mv88e6xxx_vtu_getnext interprets two members from the input\n mv88e6xxx_vtu_entry structure: the (excluded) vid member to start\n the iteration from, and the valid argument specifying whether the VID\n must be written or not (only required once at the start of a loop).\n \n Explicit the assignation of these two fields right before calling\n mv88e6xxx_vtu_getnext, as it is done in the mv88e6xxx_vtu_get wrapper.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7095a4c497adb5aca30141f43c7ba00ce18bea74\nAuthor: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\nDate: Thu Aug 1 14:36:33 2019 -0400\n\n net: dsa: mv88e6xxx: lock mutex in vlan_prepare\n \n Lock the mutex in the mv88e6xxx_port_vlan_prepare function\n called by the DSA stack, instead of doing it in the internal\n mv88e6xxx_port_check_hw_vlan helper.\n \n Signed-off-by: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 01158da721c5fd5f321cfd7e3e955fbd83ba3124\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Thu Aug 1 13:28:40 2019 +0000\n\n drm/i915/uc: Stop sanitizing enable_guc modparam\n \n As we already track GuC/HuC uses by other means than modparam\n there is no point in sanitizing it. Just scan modparam for\n major discrepancies between what was requested vs actual.\n \n v2: rebased, reworded info messages\n v3: oops\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190801132840.33176-1-michal.wajdeczko@intel.com\n\ncommit 724df646c80374f0d6e76c160a248547a042082b\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Wed Jul 31 22:33:20 2019 +0000\n\n drm/i915/guc: Use dedicated flag to track submission mode\n \n Instead of relying on enable_guc modparam to represent actual\n GuC submission mode, use dedicated flag and look at modparam\n only to check if submission was explicitly disabled by the user.\n \n v2: rebased, simplified condition (Chris)\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190731223321.36436-4-michal.wajdeczko@intel.com\n\ncommit db81bc6eb9c01598782599ecf6797adfb167f801\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Wed Jul 31 22:33:19 2019 +0000\n\n drm/i915/uc: Consider enable_guc modparam during fw selection\n \n We can use value of enable_guc modparam during firmware path selection\n and start using firmware status to see if GuC/HuC is being used.\n This is first step to make enable_guc modparam read-only.\n \n v2: rebased, don\u0027t care about \u003c0 (Chris)\n v3: oops\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190731223321.36436-3-michal.wajdeczko@intel.com\n\ncommit 57a68c3507161bed25c97df8941fb45166acec99\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Wed Jul 31 22:33:18 2019 +0000\n\n drm/i915/uc: Rename intel_uc_is_using* into intel_uc_supports*\n \n Rename intel_uc_is_using* into intel_uc_supports* to make clear\n distinction from actual state (compare intel_uc_fw_is_running)\n \n Suggested-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190731223321.36436-2-michal.wajdeczko@intel.com\n\ncommit 3695eae5fee0605f316fbaad0b9e3de791d7dfaf\nAuthor: Christian Brauner \u003cchristian.brauner@ubuntu.com\u003e\nDate: Sun Jul 28 00:22:29 2019 +0200\n\n pidfd: add P_PIDFD to waitid()\n \n This adds the P_PIDFD type to waitid().\n One of the last remaining bits for the pidfd api is to make it possible\n to wait on pidfds. With P_PIDFD added to waitid() the parts of userspace\n that want to use the pidfd api to exclusively manage processes can do so\n now.\n \n One of the things this will unblock in the future is the ability to make\n it possible to retrieve the exit status via waitid(P_PIDFD) for\n non-parent processes if handed a _suitable_ pidfd that has this feature\n set. This is similar to what you can do on FreeBSD with kqueue(). It\n might even end up being possible to wait on a process as a non-parent if\n an appropriate property is enabled on the pidfd.\n \n With P_PIDFD no scoping of the process identified by the pidfd is\n possible, i.e. it explicitly blocks things such as wait4(-1), wait4(0),\n waitid(P_ALL), waitid(P_PGID) etc. It only allows for semantics\n equivalent to wait4(pid), waitid(P_PID). Users that need scoping should\n rely on pid-based wait*() syscalls for now.\n \n Signed-off-by: Christian Brauner \u003cchristian.brauner@ubuntu.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Reviewed-by: Oleg Nesterov \u003coleg@redhat.com\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\n Cc: Joel Fernandes (Google) \u003cjoel@joelfernandes.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: David Howells \u003cdhowells@redhat.com\u003e\n Cc: Jann Horn \u003cjannh@google.com\u003e\n Cc: Andy Lutomirsky \u003cluto@kernel.org\u003e\n Cc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n Cc: Aleksa Sarai \u003ccyphar@cyphar.com\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n Link: https://lore.kernel.org/r/20190727222229.6516-2-christian@brauner.io\n\ncommit 6830b468259b45e3b73070474b8cec9388aa8c11\nAuthor: Tonghao Zhang \u003cxiangxia.m.yue@gmail.com\u003e\nDate: Thu Aug 1 16:40:59 2019 +0800\n\n net/mlx5e: Allow dropping specific tunnel packets\n \n In some case, we don\u0027t want to allow specific tunnel packets\n to host that can avoid to take up high CPU (e.g network attacks).\n But other tunnel packets which not matched in hardware will be\n sent to host too.\n \n $ tc filter add dev vxlan_sys_4789 \\\n protocol ip chain 0 parent ffff: prio 1 handle 1 \\\n flower dst_ip 1.1.1.100 ip_proto tcp dst_port 80 \\\n enc_dst_ip 2.2.2.100 enc_key_id 100 enc_dst_port 4789 \\\n action tunnel_key unset pipe action drop\n \n Signed-off-by: Tonghao Zhang \u003cxiangxia.m.yue@gmail.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit c9e6c7209a9a26a0281b311c6880b9e2382ad635\nAuthor: Aya Levin \u003cayal@mellanox.com\u003e\nDate: Mon Jun 24 20:33:52 2019 +0300\n\n net/mlx5e: TX reporter cleanup\n \n Remove redundant include files.\n \n Signed-off-by: Aya Levin \u003cayal@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit baf6dfdb10e9695637d72429159fd26fc36d30c3\nAuthor: Aya Levin \u003cayal@mellanox.com\u003e\nDate: Mon Jun 24 19:34:42 2019 +0300\n\n net/mlx5e: Set tx reporter only on successful creation\n \n When failing to create tx reporter, don\u0027t set the reporter\u0027s pointer.\n Creating a reporter is not mandatory for driver load, avoid\n garbage/error pointer.\n \n Signed-off-by: Aya Levin \u003cayal@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 7f7cc235c2dfc2a9208a743492b80d6cdfee50a6\nAuthor: Aya Levin \u003cayal@mellanox.com\u003e\nDate: Wed Jul 3 09:16:52 2019 +0300\n\n net/mlx5e: Fix mlx5e_tx_reporter_create return value\n \n Return error when failing to create a reporter in devlink. Since\n NET_DEVLINK mandatory to MLX5_CORE in Kconfig, returned pointer\n can\u0027t be NULL and can only hold an error in bad path.\n \n Signed-off-by: Aya Levin \u003cayal@mellanox.com\u003e\n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 8c7698d5caa7852bebae0cf7402b7d3a1f30423b\nAuthor: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\nDate: Fri May 3 15:12:46 2019 -0700\n\n net/mlx5e: Rx, checksum handling refactoring\n \n Move vlan checksum fixup flow into mlx5e_skb_padding_csum(), which is\n supposed to fixup SKB checksum if needed. And rename\n mlx5e_skb_padding_csum() to mlx5e_skb_csum_fixup().\n \n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit b431302e92f00b7acd5617a4d289f8006394bfc2\nAuthor: Tariq Toukan \u003ctariqt@mellanox.com\u003e\nDate: Mon Jul 1 12:08:08 2019 +0300\n\n net/mlx5e: Tx, Soften inline mode VLAN dependencies\n \n If capable, use zero inline mode in TX WQE for non-VLAN packets.\n For VLAN ones, keep the enforcement of at least L2 inline mode,\n unless the WQE VLAN insertion offload cap is on.\n \n Performance:\n Tested single core packet rate of 64Bytes.\n \n NIC: ConnectX-5\n CPU: Intel(R) Xeon(R) Gold 6154 CPU @ 3.00GHz\n \n pktgen:\n Before: 12.46 Mpps\n After: 14.65 Mpps (+17.5%)\n \n XDP_TX:\n The MPWQE flow is not affected, as it already has this optimization.\n So we test with priv-flag xdp_tx_mpwqe: off.\n \n Before: 9.90 Mpps\n After: 10.20 Mpps (+3%)\n \n Signed-off-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Tested-by: Noam Stolero \u003cnoams@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 7cf6f811b72aced0c48e1065fe059d604ef6363d\nAuthor: Tariq Toukan \u003ctariqt@mellanox.com\u003e\nDate: Sun Jul 14 17:50:51 2019 +0300\n\n net/mlx5e: XDP, Slight enhancement for WQE fetch function\n \n Instead of passing an output param, let function return the\n WQE pointer.\n In addition, pass \u0026pi so it gets its value in the function,\n and save the redundant assignment that comes after it.\n \n Signed-off-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 6c085a8aab5183d8658c9a692bcfda3e24195b7a\nAuthor: Shay Agroskin \u003cshayag@mellanox.com\u003e\nDate: Sun May 12 18:28:27 2019 +0300\n\n net/mlx5e: XDP, Close TX MPWQE session when no room for inline packet left\n \n In MPWQE mode, when transmitting packets with XDP, a packet that is smaller\n than a certain size (set to 256 bytes) would be sent inline within its WQE\n TX descriptor (mem-copied), in case the hardware tx queue is congested\n beyond a pre-defined water-mark.\n \n If a MPWQE cannot contain an additional inline packet, we close this\n MPWQE session, and send the packet inlined within the next MPWQE.\n To save some MPWQE session close+open operations, we don\u0027t open MPWQE\n sessions that are contiguously smaller than certain size (set to the\n HW MPWQE maximum size). If there isn\u0027t enough contiguous room in the\n send queue, we fill it with NOPs and wrap the send queue index around.\n \n This way, qualified packets are always sent inline.\n \n Perf tests:\n Tested packet rate for UDP 64Byte multi-stream\n over two dual port ConnectX-5 100Gbps NICs.\n CPU: Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz\n \n XDP_TX:\n \n With 24 channels:\n | ------ | bounced packets | inlined packets | inline ratio |\n | before | 113.6Mpps | 96.3Mpps | 84% |\n | after | 115Mpps | 99.5Mpps | 86% |\n \n With one channel:\n \n | ------ | bounced packets | inlined packets | inline ratio |\n | before | 6.7Mpps | 0pps | 0% |\n | after | 6.8Mpps | 0pps | 0% |\n \n As we can see, there is improvement in both inline ratio and overall\n packet rate for 24 channels. Also, we see no degradation for the\n one-channel case.\n \n Signed-off-by: Shay Agroskin \u003cshayag@mellanox.com\u003e\n Signed-off-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 68865419ba1bf502a5bd279a500deda64000249d\nAuthor: Tariq Toukan \u003ctariqt@mellanox.com\u003e\nDate: Thu Jul 11 11:20:22 2019 +0300\n\n net/mlx5e: Tx, Strict the room needed for SQ edge NOPs\n \n We use NOPs to populate the WQ fragment edge if the WQE does not fit\n in frag, to avoid WQEs crossing a page boundary (or wrap-around the WQ).\n \n The upper bound on the needed number of NOPs is one WQEBB less than\n the largest possible WQE, for otherwise the WQE would certainly fit.\n \n Signed-off-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 558101f1b9807b34d8eeefb352d11e642b7e98dd\nAuthor: Gavi Teitz \u003cgavi@mellanox.com\u003e\nDate: Thu Jun 27 20:53:03 2019 +0300\n\n net/mlx5: Add flow counter pool\n \n Add a pool of flow counters, based on flow counter bulks, removing the\n need to allocate a new counter via a costly FW command during the flow\n creation process. The time it takes to acquire/release a flow counter\n is cut from ~50 [us] to ~50 [ns].\n \n The pool is part of the mlx5 driver instance, and provides flow\n counters for aging flows. mlx5_fc_create() was modified to provide\n counters for aging flows from the pool by default, and\n mlx5_destroy_fc() was modified to release counters back to the pool\n for later reuse. If bulk allocation is not supported or fails, and for\n non-aging flows, the fallback behavior is to allocate and free\n individual counters.\n \n The pool is comprised of three lists of flow counter bulks, one of\n fully used bulks, one of partially used bulks, and one of unused\n bulks. Counters are provided from the partially used bulks first, to\n help limit bulk fragmentation.\n \n The pool maintains a threshold, and strives to maintain the amount of\n available counters below it. The pool is increased in size when a\n counter acquisition request is made and there are no available\n counters, and it is decreased in size when the last counter in a bulk\n is released and there are more available counters than the threshold.\n All pool size changes are done in the context of the\n acquiring/releasing process.\n \n The value of the threshold is directly correlated to the amount of\n used counters the pool is providing, while constrained by a hard\n maximum, and is recalculated every time a bulk is allocated/freed.\n This ensures that the pool only consumes large amounts of memory for\n available counters if the pool is being used heavily. When fully\n populated and at the hard maximum, the buffer of available counters\n consumes ~40 [MB].\n \n Signed-off-by: Gavi Teitz \u003cgavi@mellanox.com\u003e\n Reviewed-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 5d8a02536545080a555fe37064d24a402fd00d6a\nAuthor: Gavi Teitz \u003cgavi@mellanox.com\u003e\nDate: Thu Jun 27 13:58:56 2019 +0300\n\n net/mlx5: Add flow counter bulk infrastructure\n \n Add infrastructure to track bulks of flow counters, providing\n the means to allocate and deallocate bulks, and to acquire and\n release individual counters from the bulks.\n \n Signed-off-by: Gavi Teitz \u003cgavi@mellanox.com\u003e\n Reviewed-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit fcb64c0f5640e629bd77c2cb088f9fd70ff5bde7\nAuthor: Eli Cohen \u003celi@mellanox.com\u003e\nDate: Wed May 8 11:44:56 2019 +0300\n\n net/mlx5: E-Switch, add ingress rate support\n \n Use the scheduling elements to implement ingress rate limiter on an\n eswitch ports ingress traffic. Since the ingress of eswitch port is the\n egress of VF port, we control eswitch ingress by controlling VF egress.\n \n Configuration is done using the ports\u0027 representor net devices.\n \n Please note that burst size configuration is not supported by devices\n ConnectX-5 and earlier generations.\n \n Configuration examples:\n tc:\n tc filter add dev enp59s0f0_0 root protocol ip matchall action police rate 1mbit burst 20k\n \n ovs:\n ovs-vsctl set interface eth0 ingress_policing_rate\u003d1000\n \n Signed-off-by: Eli Cohen \u003celi@mellanox.com\u003e\n Reviewed-by: Paul Blakey \u003cpaulb@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 5d99b32a009e900a561f6a42ea7afe5b21288b8a\nAuthor: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\nDate: Wed Jul 31 00:33:54 2019 +0200\n\n posix-timers: Move rcu_head out of it union\n \n Timer deletion on PREEMPT_RT is prone to priority inversion and live\n locks. The hrtimer code has a synchronization mechanism for this. Posix CPU\n timers will grow one.\n \n But that mechanism cannot be invoked while holding the k_itimer lock\n because that can deadlock against the running timer callback. So the lock\n must be dropped which allows the timer to be freed.\n \n The timer free can be prevented by taking RCU readlock before dropping the\n lock, but because the rcu_head is part of the \u0027it\u0027 union a concurrent free\n will overwrite the hrtimer on which the task is trying to synchronize.\n \n Move the rcu_head out of the union to prevent this.\n \n [ tglx: Fixed up kernel-doc. Rewrote changelog ]\n \n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190730223828.965541887@linutronix.de\n\ncommit 08a3c192c93f4359a94bf47971e55b0324b72b8b\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Jul 31 00:33:55 2019 +0200\n\n posix-timers: Prepare for PREEMPT_RT\n \n Posix timer delete retry loops are affected by the same priority inversion\n and live lock issues as the other timers.\n \n Provide a RT specific synchronization function which keeps a reference to\n the timer by holding rcu read lock to prevent the timer from being freed,\n dropping the timer lock and invoking the timer specific wait function.\n \n This does not yet cover posix CPU timers because they need more special\n treatment on PREEMPT_RT.\n \n Originally-by: Anna-Maria Gleixenr \u003canna-maria@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190730223829.058247862@linutronix.de\n\ncommit 6945e5c2abe008302b20266248d6de95575311a8\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Jul 31 00:33:53 2019 +0200\n\n posix-timers: Rework cancel retry loops\n \n As a preparatory step for adding the PREEMPT RT specific synchronization\n mechanism to wait for a running timer callback, rework the timer cancel\n retry loops so they call a common function. This allows trivial\n substitution in one place.\n \n Originally-by: Anna-Maria Gleixner \u003canna-maria@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190730223828.874901027@linutronix.de\n\ncommit 21670ee44f1e3565030bcabc62178b8e5eb2fce7\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Jul 31 00:33:52 2019 +0200\n\n posix-timers: Cleanup the flag/flags confusion\n \n do_timer_settime() has a \u0027flags\u0027 argument and uses \u0027flag\u0027 for the interrupt\n flags, which is confusing at best.\n \n Rename the argument so \u0027flags\u0027 can be used for interrupt flags as usual.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190730223828.782664411@linutronix.de\n\ncommit c7e6d704a0097e59667495cf52dcc4e1085e620b\nAuthor: Anna-Maria Gleixner \u003canna-maria@linutronix.de\u003e\nDate: Wed Jul 31 00:33:51 2019 +0200\n\n itimers: Prepare for PREEMPT_RT\n \n Use the hrtimer_cancel_wait_running() synchronization mechanism to prevent\n priority inversion and live locks on PREEMPT_RT.\n \n As a benefit the retry loop gains the missing cpu_relax() on !RT.\n \n [ tglx: Split out of combo patch ]\n \n Signed-off-by: Anna-Maria Gleixner \u003canna-maria@linutronix.de\u003e\n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190730223828.690771827@linutronix.de\n\ncommit a125ecc16453a4fe0ba865c7df87b9c722991fdf\nAuthor: Anna-Maria Gleixner \u003canna-maria@linutronix.de\u003e\nDate: Wed Jul 31 00:33:50 2019 +0200\n\n timerfd: Prepare for PREEMPT_RT\n \n Use the hrtimer_cancel_wait_running() synchronization mechanism to prevent\n priority inversion and live locks on PREEMPT_RT.\n \n [ tglx: Split out of combo patch ]\n \n Signed-off-by: Anna-Maria Gleixner \u003canna-maria@linutronix.de\u003e\n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190730223828.600085866@linutronix.de\n\ncommit 51ae33092bb8320497ec75ddc5ab383d8fafd55c\nAuthor: Anna-Maria Gleixner \u003canna-maria@linutronix.de\u003e\nDate: Wed Jul 31 00:33:49 2019 +0200\n\n alarmtimer: Prepare for PREEMPT_RT\n \n Use the hrtimer_cancel_wait_running() synchronization mechanism to prevent\n priority inversion and live locks on PREEMPT_RT.\n \n [ tglx: Split out of combo patch ]\n \n Signed-off-by: Anna-Maria Gleixner \u003canna-maria@linutronix.de\u003e\n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190730223828.508744705@linutronix.de\n\ncommit 850377a875a481c393ce59111b0c9725005e0eb4\nAuthor: Juri Lelli \u003cjuri.lelli@redhat.com\u003e\nDate: Wed Jul 31 12:37:15 2019 +0200\n\n sched/deadline: Ensure inactive_timer runs in hardirq context\n \n SCHED_DEADLINE inactive timer needs to run in hardirq context (as\n dl_task_timer already does) on PREEMPT_RT\n \n Change the mode to HRTIMER_MODE_REL_HARD.\n \n [ tglx: Fixed up the start site, so mode debugging works ]\n \n Signed-off-by: Juri Lelli \u003cjuri.lelli@redhat.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190731103715.4047-1-juri.lelli@redhat.com\n\ncommit 030dcdd197d77374879bb5603d091eee7d8aba80\nAuthor: Anna-Maria Gleixner \u003canna-maria@linutronix.de\u003e\nDate: Fri Jul 26 20:31:00 2019 +0200\n\n timers: Prepare support for PREEMPT_RT\n \n When PREEMPT_RT is enabled, the soft interrupt thread can be preempted. If\n the soft interrupt thread is preempted in the middle of a timer callback,\n then calling del_timer_sync() can lead to two issues:\n \n - If the caller is on a remote CPU then it has to spin wait for the timer\n handler to complete. This can result in unbound priority inversion.\n \n - If the caller originates from the task which preempted the timer\n handler on the same CPU, then spin waiting for the timer handler to\n complete is never going to end.\n \n To avoid these issues, add a new lock to the timer base which is held\n around the execution of the timer callbacks. If del_timer_sync() detects\n that the timer callback is currently running, it blocks on the expiry\n lock. When the callback is finished, the expiry lock is dropped by the\n softirq thread which wakes up the waiter and the system makes progress.\n \n This addresses both the priority inversion and the life lock issues.\n \n This mechanism is not used for timers which are marked IRQSAFE as for those\n preemption is disabled accross the callback and therefore this situation\n cannot happen. The callbacks for such timers need to be individually\n audited for RT compliance.\n \n The same issue can happen in virtual machines when the vCPU which runs a\n timer callback is scheduled out. If a second vCPU of the same guest calls\n del_timer_sync() it will spin wait for the other vCPU to be scheduled back\n in. The expiry lock mechanism would avoid that. It\u0027d be trivial to enable\n this when paravirt spinlocks are enabled in a guest, but it\u0027s not clear\n whether this is an actual problem in the wild, so for now it\u0027s an RT only\n mechanism.\n \n As the softirq thread can be preempted with PREEMPT_RT\u003dy, the SMP variant\n of del_timer_sync() needs to be used on UP as well.\n \n [ tglx: Refactored it for mainline ]\n \n Signed-off-by: Anna-Maria Gleixner \u003canna-maria@linutronix.de\u003e\n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190726185753.832418500@linutronix.de\n\ncommit f61eff83cec9cfab31fd30a2ca8856be379cdcd5\nAuthor: Anna-Maria Gleixner \u003canna-maria@linutronix.de\u003e\nDate: Fri Jul 26 20:30:59 2019 +0200\n\n hrtimer: Prepare support for PREEMPT_RT\n \n When PREEMPT_RT is enabled, the soft interrupt thread can be preempted. If\n the soft interrupt thread is preempted in the middle of a timer callback,\n then calling hrtimer_cancel() can lead to two issues:\n \n - If the caller is on a remote CPU then it has to spin wait for the timer\n handler to complete. This can result in unbound priority inversion.\n \n - If the caller originates from the task which preempted the timer\n handler on the same CPU, then spin waiting for the timer handler to\n complete is never going to end.\n \n To avoid these issues, add a new lock to the timer base which is held\n around the execution of the timer callbacks. If hrtimer_cancel() detects\n that the timer callback is currently running, it blocks on the expiry\n lock. When the callback is finished, the expiry lock is dropped by the\n softirq thread which wakes up the waiter and the system makes progress.\n \n This addresses both the priority inversion and the life lock issues.\n \n The same issue can happen in virtual machines when the vCPU which runs a\n timer callback is scheduled out. If a second vCPU of the same guest calls\n hrtimer_cancel() it will spin wait for the other vCPU to be scheduled back\n in. The expiry lock mechanism would avoid that. It\u0027d be trivial to enable\n this when paravirt spinlocks are enabled in a guest, but it\u0027s not clear\n whether this is an actual problem in the wild, so for now it\u0027s an RT only\n mechanism.\n \n [ tglx: Refactored it for mainline ]\n \n Signed-off-by: Anna-Maria Gleixner \u003canna-maria@linutronix.de\u003e\n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190726185753.737767218@linutronix.de\n\ncommit 1842f5a427f5323f5c19ab99b55d09b3ab5172a5\nAuthor: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\nDate: Fri Jul 26 20:30:58 2019 +0200\n\n hrtimer: Determine hard/soft expiry mode for hrtimer sleepers on RT\n \n On PREEMPT_RT enabled kernels hrtimers which are not explicitely marked for\n hard interrupt expiry mode are moved into soft interrupt context either for\n latency reasons or because the hrtimer callback takes regular spinlocks or\n invokes other functions which are not suitable for hard interrupt context\n on PREEMPT_RT.\n \n The hrtimer_sleeper callback is RT compatible in hard interrupt context,\n but there is a latency concern: Untrusted userspace can spawn many threads\n which arm timers for the same expiry time on the same CPU. On expiry that\n causes a latency spike due to the wakeup of a gazillion threads.\n \n OTOH, priviledged real-time user space applications rely on the low latency\n of hard interrupt wakeups. These syscall related wakeups are all based on\n hrtimer sleepers.\n \n If the current task is in a real-time scheduling class, mark the mode for\n hard interrupt expiry.\n \n [ tglx: Split out of a larger combo patch. Added changelog ]\n \n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190726185753.645792403@linutronix.de\n\ncommit f5c2f0215e36d76fbb9605283dd7535af09f5770\nAuthor: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\nDate: Fri Jul 26 20:30:57 2019 +0200\n\n hrtimer: Move unmarked hrtimers to soft interrupt expiry on RT\n \n On PREEMPT_RT not all hrtimers can be expired in hard interrupt context\n even if that is perfectly fine on a PREEMPT_RT\u003dn kernel, e.g. because they\n take regular spinlocks. Also for latency reasons PREEMPT_RT tries to defer\n most hrtimers\u0027 expiry into softirq context.\n \n hrtimers marked with HRTIMER_MODE_HARD must be kept in hard interrupt\n context expiry mode. Add the required logic.\n \n No functional change for PREEMPT_RT\u003dn kernels.\n \n [ tglx: Split out of a larger combo patch. Added changelog ]\n \n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190726185753.551967692@linutronix.de\n\ncommit 902a9f9c509053161e987778dc5836d2628f53b7\nAuthor: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\nDate: Fri Jul 26 20:30:56 2019 +0200\n\n tick: Mark tick related hrtimers to expiry in hard interrupt context\n \n The tick related hrtimers, which drive the scheduler tick and hrtimer based\n broadcasting are required to expire in hard interrupt context for obvious\n reasons.\n \n Mark them so PREEMPT_RT kernels wont move them to soft interrupt expiry.\n \n Make the horribly formatted RCU_NONIDLE bracket maze readable while at it.\n \n No functional change,\n \n [ tglx: Split out from larger combo patch. Add changelog ]\n \n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190726185753.459144407@linutronix.de\n\ncommit 2c0d278f3293fc59da0d183075415ca1e8c93b40\nAuthor: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\nDate: Fri Jul 26 20:30:55 2019 +0200\n\n KVM: LAPIC: Mark hrtimer to expire in hard interrupt context\n \n On PREEMPT_RT enabled kernels unmarked hrtimers are moved into soft\n interrupt expiry mode by default.\n \n While that\u0027s not a functional requirement for the KVM local APIC timer\n emulation, it\u0027s a latency issue which can be avoided by marking the timer\n so hard interrupt context expiry is enforced.\n \n No functional change.\n \n [ tglx: Split out from larger combo patch. Add changelog. ]\n \n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190726185753.363363474@linutronix.de\n\ncommit d2ab4cf4943576fb060b8a69341d9e0c2a952ba7\nAuthor: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\nDate: Fri Jul 26 20:30:54 2019 +0200\n\n watchdog: Mark watchdog_hrtimer to expire in hard interrupt context\n \n The watchdog hrtimer must expire in hard interrupt context even on\n PREEMPT_RT\u003dy kernels as otherwise the hard/softlockup detection logic would\n not work.\n \n No functional change.\n \n [ tglx: Split out from larger combo patch. Added changelog ]\n \n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190726185753.262895510@linutronix.de\n\ncommit 30f9028b6c43fd17c006550594ea3dbb87afbf80\nAuthor: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\nDate: Fri Jul 26 20:30:53 2019 +0200\n\n perf/core: Mark hrtimers to expire in hard interrupt context\n \n To guarantee that the multiplexing mechanism and the hrtimer driven events\n work on PREEMPT_RT enabled kernels it\u0027s required that the related hrtimers\n expire in hard interrupt context. Mark them so PREEMPT_RT kernels wont\n defer them to soft interrupt context.\n \n No functional change.\n \n [ tglx: Split out of larger combo patch. Added changelog ]\n \n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190726185753.169509224@linutronix.de\n\ncommit d5096aa65acd0ef2d18ac8247260ab4481ade399\nAuthor: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\nDate: Fri Jul 26 20:30:52 2019 +0200\n\n sched: Mark hrtimers to expire in hard interrupt context\n \n The scheduler related hrtimers need to expire in hard interrupt context\n even on PREEMPT_RT enabled kernels. Mark then as such.\n \n No functional change.\n \n [ tglx: Split out from larger combo patch. Add changelog. ]\n \n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190726185753.077004842@linutronix.de\n\ncommit 0ab6a3ddbad40ef5b6b8c2353fd53fa4ecf9c479\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Tue Jul 30 20:15:25 2019 +0200\n\n hrtimer: Make enqueue mode check work on RT\n \n hrtimer_start_range_ns() has a WARN_ONCE() which verifies that a timer\n which is marker for softirq expiry is not queued in the hard interrupt base\n and vice versa.\n \n When PREEMPT_RT is enabled, timers which are not explicitely marked to\n expire in hard interrupt context are deferrred to the soft interrupt. So\n the regular check would trigger.\n \n Change the check, so when PREEMPT_RT is enabled, it is verified that the\n timers marked for hard interrupt expiry are not tried to be queued for soft\n interrupt expiry or any of the unmarked and softirq marked is tried to be\n expired in hard interrupt context.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n\ncommit 5896b97296a7928035590ff3f477629774dce250\nAuthor: Parav Pandit \u003cparav@mellanox.com\u003e\nDate: Mon Jul 29 21:13:12 2019 +0000\n\n net/mlx5: E-switch, Tide up eswitch config sequence\n \n Currently for PF and ECPF vports, representors are created before\n their eswitch hardware ports are initialized in below flow.\n \n mlx5_eswitch_enable()\n esw_offloads_init()\n esw_offloads_load_all_reps()\n [..]\n esw_enable_vport()\n \n However for VFs, vports are initialized before creating their\n respective netdev represnetors in event handling context.\n \n Similarly while disabling eswitch, first hardware vports are disabled,\n followed by destroying their representors.\n Here while underlying vports gets destroyed but its respective user\n facing netdevice can still exist on which user can continue to perform\n more offload operations.\n \n Instead, its more accurate to do\n enable_eswitch switchdev mode:\n 1. perform FDB tables initialization\n 2. initialize hw vport\n 3. create and publish representor for this vport\n \n disable_eswitch switchdev mode:\n 1. destroy user facing representor for the vport\n 2. disable hw vport\n 3. perform FDB tables cleanup\n \n Signed-off-by: Parav Pandit \u003cparav@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 131ce7014043087fbeddbcb3b8dac8891cd0e0fe\nAuthor: Parav Pandit \u003cparav@mellanox.com\u003e\nDate: Mon Jul 29 21:13:10 2019 +0000\n\n net/mlx5: E-Switch, Remove redundant mc_promisc NULL check\n \n mc_promisc pointer points to an instance of struct esw_mc_addr allocated\n as part of the esw structure.\n Hence it cannot be NULL.\n Removed such redundant check and assign where it is actually used.\n \n While at it, add comment around legacy mode fields and move mc_promisc\n close to other legacy mode structures to improve code redability.\n \n Signed-off-by: Parav Pandit \u003cparav@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 9ddb830a14dbd88308354d27cd17009fc97d3a6f\nAuthor: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\nDate: Mon Jul 29 21:13:08 2019 +0000\n\n net/mlx5: E-Switch, remove redundant error handling\n \n We don\u0027t need to handle error flow of esw_create_legacy_table() in the\n same branch, it is already being handled directly after the if statement,\n for both legacy and switchdev modes in one place.\n \n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 5019833d661f5edb6bd63abd3da064d2517966b4\nAuthor: Parav Pandit \u003cparav@mellanox.com\u003e\nDate: Mon Jul 29 21:13:06 2019 +0000\n\n net/mlx5: E-switch, Introduce helper function to enable/disable vports\n \n vports needs to be enabled in switchdev and legacy mode.\n \n In switchdev mode, vports should be enabled after initializing\n the FDB tables and before creating their represntors so that\n representor works on an initialized vport object.\n \n Prepare a helper function which can be called when enabling either of\n the eswitch modes.\n \n Similarly, have disable vports helper function.\n \n Signed-off-by: Parav Pandit \u003cparav@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 610090ebce92ab6a3e1e623344be5a9dd72a8b6d\nAuthor: Parav Pandit \u003cparav@mellanox.com\u003e\nDate: Mon Jul 29 21:13:04 2019 +0000\n\n net/mlx5: E-switch, Initialize TSAR Qos hardware block before its user vports\n \n First enable TSAR Qos hardware block in device before enabling its\n user vports.\n \n This refactor is needed so that vports can be enabled before their\n representor netdevice can be created.\n \n While at it, esw_create_tsar() returns error code which was used only to\n print error. However esw_create_tsar() already prints warning if it hits\n an error.\n Hence, remove the redundant warning.\n \n Signed-off-by: Parav Pandit \u003cparav@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 332bd3a5b931eb67deb370db62d59f9cc7f76dac\nAuthor: Parav Pandit \u003cparav@mellanox.com\u003e\nDate: Mon Jul 29 21:13:02 2019 +0000\n\n net/mlx5: E-switch, Combine metadata enable/disable functionality\n \n Except bit toggling code, rest of the code is same to enable/disable\n metadata passing functionality.\n Hence, combine them to single function and control using enable flag.\n \n Also instead of checking metadata supported at multiple places,\n fold into the helper function.\n \n Signed-off-by: Parav Pandit \u003cparav@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 6cedde4513990af7191afa43528533f80e92c989\nAuthor: Eli Cohen \u003celi@mellanox.com\u003e\nDate: Mon Jul 29 21:13:00 2019 +0000\n\n net/mlx5: E-Switch, Verify support QoS element type\n \n Check if firmware supports the requested element type before\n attempting to create the element type.\n In addition, explicitly specify the request element type and tsar type.\n \n Signed-off-by: Eli Cohen \u003celi@mellanox.com\u003e\n Reviewed-by: Paul Blakey \u003cpaulb@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 0000a5f2507deef8668d23d6406e9d19ba371def\nAuthor: Parav Pandit \u003cparav@mellanox.com\u003e\nDate: Mon Jul 29 21:12:58 2019 +0000\n\n net/mlx5: Make load_one() and unload_one() symmetric\n \n Currently mlx5_load_one() perform device registration using\n mlx5_register_device(). But mlx5_unload_one() doesn\u0027t unregister.\n \n Make them symmetric by doing device unregistration in\n mlx5_unload_one().\n \n Signed-off-by: Parav Pandit \u003cparav@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 7761f9eef3f09f2f4c579313e0c536770b5ecff4\nAuthor: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\nDate: Mon Jul 29 21:12:56 2019 +0000\n\n net/mlx5: Fix offset of tisc bits reserved field\n \n First reserved field is off by one instead of reserved_at_1 it should be\n reserved_at_2, fix that.\n \n Fixes: a12ff35e0fb7 (\"net/mlx5: Introduce TLS TX offload hardware bits and structures\")\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n Reviewed-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 8536a6bf2ea1f3bf4e3159b590fbecce4d854796\nAuthor: Gavi Teitz \u003cgavi@mellanox.com\u003e\nDate: Mon Jul 29 21:12:54 2019 +0000\n\n net/mlx5: Add flow counter bulk allocation hardware bits and command\n \n Add a handle to invoke the new FW capability of allocating a bulk of\n flow counters.\n \n Signed-off-by: Gavi Teitz \u003cgavi@mellanox.com\u003e\n Reviewed-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 6f06e04b67baa1c9da61c8b15b1335a1dbb98bcb\nAuthor: Gavi Teitz \u003cgavi@mellanox.com\u003e\nDate: Mon Jul 29 21:12:52 2019 +0000\n\n net/mlx5: Refactor and optimize flow counter bulk query\n \n Towards introducing the ability to allocate bulks of flow counters,\n refactor the flow counter bulk query process, removing functions and\n structs whose names indicated being used for flow counter bulk\n allocation FW commands, despite them actually only being used to\n support bulk querying, and migrate their functionality to correctly\n named functions in their natural location, fs_counters.c.\n \n Additionally, optimize the bulk query process by:\n * Extracting the memory used for the query to mlx5_fc_stats so\n that it is only allocated once, and not for each bulk query.\n * Querying all the counters in one function call.\n \n Signed-off-by: Gavi Teitz \u003cgavi@mellanox.com\u003e\n Reviewed-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 012fcb52f67cbba95c3e85010a4a9c40ea43866f\nAuthor: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\nDate: Thu Aug 1 11:55:45 2019 +0800\n\n net: hns3: activate reset timer when calling reset_event\n \n When calling hclge_reset_event() within HCLGE_RESET_INTERVAL,\n it returns directly now. If no one call it again, then the\n error which needs a reset to fix it can not be fixed.\n \n So this patch activates the reset timer for this case, and\n adds checking in the end of the reset procedure to make this\n error fixed earlier.\n \n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 72e2fb07997c5ca2c73114624fd64c41be013bdc\nAuthor: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\nDate: Thu Aug 1 11:55:44 2019 +0800\n\n net: hns3: clear reset interrupt status in hclge_irq_handle()\n \n Currently, the reset interrupt is cleared in the reset task, which\n is too late. Since, when the hardware finish the previous reset,\n it can begin to do a new global/IMP reset, if this new coming reset\n type is same as the previous one, the driver will clear them together,\n then driver can not get that there is another reset, but the hardware\n still wait for the driver to deal with the second one.\n \n So this patch clears PF\u0027s reset interrupt status in the\n hclge_irq_handle(), the hardware waits for handshaking from\n driver before doing reset, so the driver and hardware deal with reset\n one by one.\n \n BTW, when VF doing global/IMP reset, it reads PF\u0027s reset interrupt\n register to get that whether PF driver\u0027s re-initialization is done,\n since VF\u0027s re-initialization should be done after PF\u0027s. So we add\n a new command and a register bit to do that. When VF receive reset\n interrupt, it sets up this bit, and PF finishes re-initialization\n send command to clear this bit, then VF do re-initialization.\n \n Fixes: 4ed340ab8f49 (\"net: hns3: Add reset process in hclge_main\")\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Reviewed-by: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6b428b4fbf986077d218287dd6f805f371cdf48f\nAuthor: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\nDate: Thu Aug 1 11:55:43 2019 +0800\n\n net: hns3: fix some reset handshake issue\n \n Currently, the driver sets handshake status to tell the hardware\n that the driver have downed the netdev and it can continue with\n reset process. The driver will clear the handshake status when\n re-initializing the CMDQ, and does not recover this status\n when reset fail, which may cause the hardware to wait for\n the handshake status to be set and not being able to continue\n with reset process.\n \n So this patch delays clearing handshake status just before UP,\n and recovers this status when reset fail.\n \n BTW, this patch adds a new function hclge(vf)_reset_handshake() to\n deal with the reset handshake issue, and renames\n HCLGE(VF)_NIC_CMQ_ENABLE to HCLGE(VF)_NIC_SW_RST_RDY which\n represents this register bit more accurately.\n \n Fixes: ada13ee3db7b (\"net: hns3: add handshake with hardware while doing reset\")\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6e6e7680630c1b80042a79f62509aab5f49e3e88\nAuthor: Guojia Liao \u003cliaoguojia@huawei.com\u003e\nDate: Thu Aug 1 11:55:42 2019 +0800\n\n net: hns3: rename a member in struct hclge_mac_ethertype_idx_rd_cmd\n \n The member \u0027mac_add\u0027 defined in hclge_mac_ethertype_idx_rd_cmd\n means MAC address, so \u0027mac_addr\u0027 is a better name for it.\n \n Signed-off-by: Guojia Liao \u003cliaoguojia@huawei.com\u003e\n Signed-off-by: Guangbin Huang \u003chuangguangbin2@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit dbae56a33f5881bad631b0121f9b4a5e7b2a454d\nAuthor: Weihang Li \u003cliweihang@hisilicon.com\u003e\nDate: Thu Aug 1 11:55:41 2019 +0800\n\n net: hns3: simplify hclge_cmd_query_error()\n \n The 4th and 5th parameter of hclge_cmd_query_error is useless, so this\n patch removes them.\n \n Signed-off-by: Weihang Li \u003cliweihang@hisilicon.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b6872fd361056f3bce3e9fcfe9535c4b6b737531\nAuthor: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\nDate: Thu Aug 1 11:55:40 2019 +0800\n\n net: hns3: minior error handling change for hclge_tm_schd_info_init\n \n When hclge_tm_schd_info_update calls hclge_tm_schd_info_init to\n initialize the schedule info, hdev-\u003etm_info.num_pg and\n hdev-\u003etx_sch_mode is not changed, which makes the checking in\n hclge_tm_schd_info_init unnecessary.\n \n So this patch moves the hdev-\u003etm_info.num_pg and hdev-\u003etx_sch_mode\n checking into hclge_tm_schd_init and changes the return type of\n hclge_tm_schd_info_init from int to void.\n \n Signed-off-by: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a4ee7624c003cc73a2eac9d9d5317b0c9fed89c4\nAuthor: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\nDate: Thu Aug 1 11:55:39 2019 +0800\n\n net: hns3: minor cleanup in hns3_clean_rx_ring\n \n The unused_count variable is used to indicate how many\n RX BD need attaching new buffer in hns3_clean_rx_ring,\n and the clean_count variable has the similar meaning.\n \n This patch removes the clean_count variable and use\n unused_count to uniformly indicate the RX BD that need\n attaching new buffer.\n \n This patch also clean up some coding style related to\n variable assignment in hns3_clean_rx_ring.\n \n Signed-off-by: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6e4139f69163fb5769c92a97b969ea0d9a114ced\nAuthor: Jian Shen \u003cshenjian15@huawei.com\u003e\nDate: Thu Aug 1 11:55:38 2019 +0800\n\n net: hns3: remove unnecessary variable in hclge_get_mac_vlan_cmd_status()\n \n The local variable return_status in hclge_get_mac_val_cmd_status()\n is useless. So this patch returns the error code directly, instead of\n using this variable. Also, replace some \u0027%d\u0027 with \u0027%u\u0027 in\n hclge_get_mac_val_cmd_status().\n \n Signed-off-by: Jian Shen \u003cshenjian15@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a723fb8efe29d3912d2be41edcf38b76f8d6ed36\nAuthor: Jian Shen \u003cshenjian15@huawei.com\u003e\nDate: Thu Aug 1 11:55:37 2019 +0800\n\n net: hns3: refine for set ring parameters\n \n Previously, when changing the ring parameters, we free the old\n ring resources firstly, and then setup the new ring resources.\n In some case of an memory allocation fail, there will be no\n resources to use. This patch refines it by setup new ring\n resources and free the old ring resources in order.\n \n Also reduce the max ring BD number to 32760 according to UM.\n \n Signed-off-by: Jian Shen \u003cshenjian15@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 3f0f325309aca23cb241c1aded0056d78d7366a1\nAuthor: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\nDate: Thu Aug 1 11:55:36 2019 +0800\n\n net: hns3: do not query unsupported commands in debugfs\n \n Some commands are not supported on DCB-unsupported ports.\n This patch distinguishes these commands and does not query\n unsupported commands in debugfs.\n \n This patch also fix an error in the dump \"qos buf cfg\"\n command in debugfs.\n \n Fixes: 2849d4e7a1be (\"net: hns3: Add \"tc config\" info query function\")\n Fixes: 7d9d7f8864ba (\"net: hns3: Add \"qos buffer\" config info query function\")\n Signed-off-by: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b18bf305c46c20e289c6d030b21b13b632e8a459\nAuthor: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\nDate: Thu Aug 1 11:55:35 2019 +0800\n\n net: hns3: add handler for NCSI error mailbox\n \n When NCSI has HW error, the IMP will report this error to the driver\n by sending a mailbox. After received this message, the driver should\n assert a global reset to fix this kind of HW error.\n \n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ed8fb4b262aed95974ccd624402a482c4d59c888\nAuthor: Jian Shen \u003cshenjian15@huawei.com\u003e\nDate: Thu Aug 1 11:55:34 2019 +0800\n\n net: hns3: add link change event report\n \n Previously, PF updates link status per second. For some scenario,\n it requires link down event being reported more quickly.\n To solve it, firmware pushes the link change event to PF with\n CMDQ message, and driver updates the link status directly.\n \n Signed-off-by: Jian Shen \u003cshenjian15@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 0ae9fce32c5e5dafcbc7a5ed97a962d021423d31\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 1 20:46:30 2019 +0800\n\n net: phy: xgene: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9d26cfa5b0d221e8655592e93b86f2e5e5cb4570\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 1 20:39:08 2019 +0800\n\n bcm63xx_enet: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c792c0081db67252c52de4d4b82356edf0dc71c5\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 1 20:34:30 2019 +0800\n\n net: qcom/emac: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 566495de16580a3cb730b7975f1ae75fa1308fd5\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 1 20:33:08 2019 +0800\n\n net: mediatek: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 42376788469bb3d543f9e287796465166d2b40fe\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 1 20:29:11 2019 +0800\n\n net: dsa: bcm_sf2: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 291f4b6de48aa8a4140a189ab974d7c398d72bea\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 1 20:27:32 2019 +0800\n\n net: dsa: b53: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6551c8c807b8d35b404f5a4bfeb02b20bf8c1fad\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 1 20:25:46 2019 +0800\n\n net: dsa: lantiq: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 3230a55b3633aa09c91bd83b794cbdfff5f7c680\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 1 20:22:02 2019 +0800\n\n mvpp2: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9dfe3459ef4152ca4cf66c78cdd1ae5f58aaf2ae\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Wed Jul 31 17:57:09 2019 -0700\n\n drm/i915/gt: Introduce intel_gt_runtime_suspend/resume\n \n To be called from the top level runtime functions, to hide the\n gt-specific bits (mainly related to intel_uc).\n \n v2: rebased\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@linux.intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190801005709.34092-3-daniele.ceraolospurio@intel.com\n\ncommit 6f76098fe0f3f0b519b2ad528b4319195d6d0f73\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Wed Jul 31 17:57:08 2019 -0700\n\n drm/i915/uc: Move uC early functions inside the GT ones\n \n uC is a subcomponent of GT, so initialize/clean it as part of it. The\n wopcm_init_early doesn\u0027t have to be happen before the uC one, but since\n in other parts of the code we consider WOPCM first do the same for\n consistency.\n \n v2: s/cleanup_early/late_release to match the caller\n v3: s/late_release/driver_late_release/ (Chris)\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@linux.intel.com\u003e\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e #v1\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190801005709.34092-2-daniele.ceraolospurio@intel.com\n\ncommit 6cf72db68d20230380a7a37dae31943c4d1d6e07\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Wed Jul 31 17:57:07 2019 -0700\n\n drm/i915/gt: Move gt_cleanup_early out of gem_cleanup_early\n \n We don\u0027t call the init_early function from within the gem code, so we\n shouldn\u0027t do it for the cleanup either.\n \n v2: while at it, s/gt_cleanup_early/gt_late_release (Chris)\n v3: s/late_release/driver_late_release/ (Chris)\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@linux.intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e #v1\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190801005709.34092-1-daniele.ceraolospurio@intel.com\n\ncommit b0d08afda84dde941226f67cb12dd9cbde40ede4\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Jul 31 22:09:03 2019 +0800\n\n staging: rtl8723bs: remove set but not used variable \u0027FirstConnect\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/staging/rtl8723bs/hal/odm.c: In function \u0027odm_RSSIMonitorCheckCE\u0027:\n drivers/staging/rtl8723bs/hal/odm.c:1258:7: warning:\n variable \u0027FirstConnect\u0027 set but not used [-Wunused-but-set-variable]\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731140903.304-1-yuehaibing@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d698b0a09656818beafd8ea7295d7c20bc74f9bc\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Jul 31 21:59:53 2019 +0800\n\n staging: rtl8723bs: remove set but not used variables \u0027prspbuf\u0027 and \u0027auth\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/staging/rtl8723bs/core/rtw_cmd.c: In function rtw_cmd_thread:\n drivers/staging/rtl8723bs/core/rtw_cmd.c:405:16: warning: variable prspbuf set but not used [-Wunused-but-set-variable]\n drivers/staging/rtl8723bs/core/rtw_cmd.c: In function rtw_joinbss_cmd:\n drivers/staging/rtl8723bs/core/rtw_cmd.c:771:6: warning: variable auth set but not used [-Wunused-but-set-variable]\n \n They are never used, so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731135953.16784-1-yuehaibing@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 09acf29c8246a20c6e255969df7e1131317181da\nAuthor: Navid Emamdoost \u003cnavid.emamdoost@gmail.com\u003e\nDate: Wed Jul 31 09:19:21 2019 -0500\n\n staging: rtl8192u: null check the kzalloc\n \n In rtl8192_init_priv_variable allocation for priv-\u003epFirmware may fail,\n so a null check is necessary.priv-\u003epFirmware is accessed later in\n rtl8192_adapter_start. I added the check and made appropriate changes\n to propagate the errno to the caller.\n \n Signed-off-by: Navid Emamdoost \u003cnavid.emamdoost@gmail.com\u003e\n \n Link: https://lore.kernel.org/r/20190731141925.29268-1-navid.emamdoost@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a1c9ca223c3df1b8993abedde777f5462165387c\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jul 30 14:30:26 2019 +0100\n\n drm/i915: Remove lrc default desc from GEM context\n \n We only compute the lrc_descriptor() on pinning the context, i.e.\n infrequently, so we do not benefit from storing the template as the\n addressing mode is also fixed for the lifetime of the intel_context.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Prathap Kumar Valsan \u003cprathap.kumar.valsan@intel.com\u003e\n Acked-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730133035.1977-9-chris@chris-wilson.co.uk\n\ncommit 572d8be0d4dd2251396dca38d176354180be7683\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Aug 1 09:53:07 2019 +0800\n\n staging: rtl8723bs: remove set but not used variable \u0027bEEPROMCheck\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/staging/rtl8723bs//hal/odm_CfoTracking.c: In function \u0027odm_SetCrystalCap\u0027:\n drivers/staging/rtl8723bs//hal/odm_CfoTracking.c:14:7: warning:\n variable \u0027bEEPROMCheck\u0027 set but not used [-Wunused-but-set-variable]\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190801015307.44572-1-yuehaibing@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 986409f74a5afc1d99eeb1b20994063b2d7f1b0e\nAuthor: Andrew Murray \u003candrew.murray@arm.com\u003e\nDate: Tue Jul 30 13:51:53 2019 +0100\n\n coresight: etm4x: Use explicit barriers on enable/disable\n \n Synchronization is recommended before disabling the trace registers\n to prevent any start or stop points being speculative at the point\n of disabling the unit (section 7.3.77 of ARM IHI 0064D).\n \n Synchronization is also recommended after programming the trace\n registers to ensure all updates are committed prior to normal code\n resuming (section 4.3.7 of ARM IHI 0064D).\n \n Let\u0027s ensure these syncronization points are present in the code\n and clearly commented.\n \n Note that we could rely on the barriers in CS_LOCK and\n coresight_disclaim_device_unlocked or the context switch to user\n space - however coresight may be of use in the kernel.\n \n On armv8 the mb macro is defined as dsb(sy) - Given that the etm4x is\n only used on armv8 let\u0027s directly use dsb(sy) instead of mb(). This\n removes some ambiguity and makes it easier to correlate the code with\n the TRM.\n \n Signed-off-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n Reviewed-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n [Fixed capital letter for \"use\" in title]\n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n\ncommit 01e3927f6e0246e7885b7bd3ef647154c60002e0\nAuthor: Andrew Murray \u003candrew.murray@arm.com\u003e\nDate: Tue Jul 30 13:51:52 2019 +0100\n\n coresight: etm4x: Remove superfluous setting of os_unlock\n \n In addition to unlocking the OS lock, etm4_os_unlock will also\n set the os_unlock flag. Therefore let\u0027s avoid unnecessarily\n setting os_unlock flag outside of this function.\n \n Signed-off-by: Andrew Murray \u003candrew.murray@arm.com\u003e\n Reviewed-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n [Fixed capital letter for \"remove\" in the title]\n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n\ncommit 991de72831b313acb9ba7a4fa9c7d3422f880a3d\nAuthor: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\nDate: Fri Jul 26 17:08:39 2019 +0100\n\n coresight: acpi: Static funnel support\n \n The ACPI bindings for CoreSight has been updated to add the device\n id for non-programmable CoreSight funnels (aka static funnels) as of\n v1.1 [0]. Add the ACPI id for static funnels in the driver.\n \n [0] https://static.docs.arm.com/den0067/a/DEN0067_CoreSight_ACPI_1.1.pdf\n \n Signed-off-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n\ncommit 336e86a5232ade6797051d01cfd98a70fdbb27f5\nAuthor: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\nDate: Tue Jul 30 10:37:33 2019 +0100\n\n coresight: Convert pr_warn to dev_warn for obsolete bindings\n \n We warn the users of obsolete bindings in the DT for coresight replicator\n and funnel drivers. However we use pr_warn_once() which doesn\u0027t give a clue\n about which device it is bound to. Let us use dev_warn_once() to give the\n context.\n \n Cc: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Signed-off-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n\ncommit 7e5a36e8a4e55e98bc0a67fb00e33931181af867\nAuthor: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\nDate: Wed Jul 24 12:43:12 2019 +0100\n\n coresight: tmc-etr: Check if non-secure access is enabled\n \n CoreSight TMC-ETR must have the non-secure invasive debug access\n enabled for use by self-hosted tracing. Without it, there is no\n point in enabling the ETR. So, let us check it in the TMC_AUTHSTATUS\n register and fail the probe if it is disabled.\n \n Cc: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Signed-off-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n\ncommit c4770f7e5a988b8e2365df830e72b055af7d59c9\nAuthor: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\nDate: Wed Jul 24 12:43:11 2019 +0100\n\n coresight: tmc-etr: Handle memory errors\n \n We have so far ignored the memory errors, assuming that we have perfect\n hardware and driver. Let us handle the memory errors reported by the\n TMC ETR in status and truncate the buffer.\n \n Cc: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Signed-off-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n [Removed ASCII smiley face from changelog]\n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n\ncommit 3be3f07bfd2166417bbd650c23605fe40a88cb60\nAuthor: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\nDate: Wed Jul 24 12:43:10 2019 +0100\n\n coresight: etr_buf: Consolidate refcount initialization\n \n We now use refcounts for the etr_buf users. Let us initialize it\n while we allocate it.\n \n Cc: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Signed-off-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n\ncommit ac1b0ca667ec3ffb1c57e1307872e7ccfa0ee51c\nAuthor: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\nDate: Wed Jul 24 12:43:08 2019 +0100\n\n coresight: Fix DEBUG_LOCKS_WARN_ON for uninitialized attribute\n \n While running the linux-next with CONFIG_DEBUG_LOCKS_ALLOC enabled,\n I get the following splat.\n \n BUG: key ffffcb5636929298 has not been registered!\n ------------[ cut here ]------------\n DEBUG_LOCKS_WARN_ON(1)\n WARNING: CPU: 1 PID: 53 at kernel/locking/lockdep.c:3669 lockdep_init_map+0x164/0x1f0\n CPU: 1 PID: 53 Comm: kworker/1:1 Tainted: G W 5.2.0-next-20190712-00015-g00ad4634222e-dirty #603\n Workqueue: events amba_deferred_retry_func\n pstate: 60c00005 (nZCv daif +PAN +UAO)\n pc : lockdep_init_map+0x164/0x1f0\n lr : lockdep_init_map+0x164/0x1f0\n \n [ trimmed ]\n \n Call trace:\n lockdep_init_map+0x164/0x1f0\n __kernfs_create_file+0x9c/0x158\n sysfs_add_file_mode_ns+0xa8/0x1d0\n sysfs_add_file_to_group+0x88/0xd8\n etm_perf_add_symlink_sink+0xcc/0x138\n coresight_register+0x110/0x280\n tmc_probe+0x160/0x420\n \n [ trimmed ]\n \n ---[ end trace ab4cc669615ba1b0 ]---\n \n Fix this by initialising the dynamically allocated attribute properly.\n \n Cc: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Fixes: bb8e370bdc141ddf (\"coresight: perf: Add \"sinks\" group to PMU directory\")\n Signed-off-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n [Fixed a typograhic error in the changelog]\n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n\ncommit eb6c2eb6c7fbd24b51fba2fa4b270c839b8b8349\nAuthor: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\nDate: Wed Jul 31 14:11:50 2019 +0200\n\n usb: host: xhci-plat: Prevent an abnormally restrictive PHY init skipping\n \n In the past, USB PHY handling has been moved in the HCD core. Some\n host controller drivers needing more control of the PHYs, they have\n been granted the freedom to handle themselves the PHY states and to\n prevent the HCD core to do so in commit 4e88d4c08301 (\"usb: add a flag\n to skip PHY initialization to struct usb_hcd\"). With this change, any\n USB host controller could set the hcd-\u003eskip_phy_initialization flag so\n that the HCD core would just skip the PHY initialization sequence.\n \n However, in the USB subsystem, there are currently two entirely\n different forms of PHY: one is called \u0027usb_phy\u0027 and is\n USB-subsystem-wide, while there is also the generic and kernel-wide\n \u0027phy\u0027 from the (recent) generic PHY framework.\n \n When the commit above was introduced, both type of PHYs where handled\n by the HCD core.\n \n Later, commit bc40f5341741 (\"USB: core: hcd: drop support for legacy\n phys\") removed the support for the former type of PHYs in the HCD\n core. These \u0027usb_phy\u0027 are still present though, but managed from the\n controller drivers only. Hence, setting the\n hcd-\u003eskip_phy_initialization flag just because a \u0027usb_phy\u0027 is\n initialized by a controller driver is a non-sense.\n \n For instance on Armada CP110, a \u0027usb_phy\u0027 is there to enable the power\n supply to the USB host, while there is also a COMPHY block providing\n SERDES lanes configuration that is referenced as a PHY from the common\n PHY framework.\n \n Right now, users of the xhci-plat.c driver either use a \u0027usb_phy\u0027 only\n and do not care about the attempt of generic PHY initialization within\n the HCD core (as there is none); or they use a single \u0027phy\u0027 and the\n code flow does not pass through the block setting\n hcd-\u003eskip_phy_initialization anyway.\n \n While there is not users of both PHY types at the same time, drop this\n limitation from the xhci-plat.c driver. Note that the tegra driver\n probably has the same limitation and could definitely benefit from a\n similar change.\n \n Cc: Johan Hovold \u003cjohan@kernel.org\u003e\n Cc: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Acked-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Link: https://lore.kernel.org/r/20190731121150.2253-1-miquel.raynal@bootlin.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d1f54df4fbb5e525bea4ecc60ceb1e5cf2387689\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Jul 31 15:42:41 2019 +0200\n\n usb: wusbcore: Spelling s/disconenct/disconnect/\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Link: https://lore.kernel.org/r/20190731134241.18647-1-geert+renesas@glider.be\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 82e17a09c412ad981dd29c0243b73b363835899e\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Wed Jul 31 23:39:17 2019 +0100\n\n usb: musb: remove redundant assignment to variable ret\n \n Variable ret is being initialized with a value that is never read\n and ret is being re-assigned a little later on. The assignment is\n redundant and hence can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190731223917.16532-1-colin.king@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b2567ebb78bdde9a49010957a0ea5a2514f3fed2\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Jul 31 15:37:48 2019 +0800\n\n RDMA/hns: remove set but not used variable \u0027irq_num\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/infiniband/hw/hns/hns_roce_hw_v2.c: In function hns_roce_v2_cleanup_eq_table:\n drivers/infiniband/hw/hns/hns_roce_hw_v2.c:5920:6:\n warning: variable irq_num set but not used [-Wunused-but-set-variable]\n \n It is not used since\n commit 33db6f94847c (\"RDMA/hns: Refactor eq table init for hip08\")\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190731073748.17664-1-yuehaibing@huawei.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit ae6683d815895c2be1e60e1942630fa99488055b\nAuthor: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\nDate: Fri Jul 26 20:30:51 2019 +0200\n\n hrtimer: Introduce HARD expiry mode\n \n On PREEMPT_RT not all hrtimers can be expired in hard interrupt context\n even if that is perfectly fine on a PREEMPT_RT\u003dn kernel, e.g. because they\n take regular spinlocks. Also for latency reasons PREEMPT_RT tries to defer\n most hrtimers\u0027 expiry into soft interrupt context.\n \n But there are hrtimers which must be expired in hard interrupt context even\n when PREEMPT_RT is enabled:\n \n - hrtimers which must expiry in hard interrupt context, e.g. scheduler,\n perf, watchdog related hrtimers\n \n - latency critical hrtimers, e.g. nanosleep, ..., kvm lapic timer\n \n Add a new mode flag HRTIMER_MODE_HARD which allows to mark these timers so\n PREEMPT_RT will not move them into softirq expiry mode.\n \n [ tglx: Split out of a larger combo patch. Added changelog ]\n \n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190726185752.981398465@linutronix.de\n\ncommit 9dd8813ed9f6e2bba75434abc6c8bb06c3d87fdc\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Tue Jul 30 21:16:55 2019 +0200\n\n hrtimer/treewide: Use hrtimer_sleeper_start_expires()\n \n hrtimer_sleepers will gain a scheduling class dependent treatment on\n PREEMPT_RT. Use the new hrtimer_sleeper_start_expires() function to make\n that possible.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n\ncommit 01656464fce946f70b02a84ab218e562ceb1662e\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Tue Jul 30 21:03:53 2019 +0200\n\n hrtimer: Provide hrtimer_sleeper_start_expires()\n \n hrtimer_sleepers will gain a scheduling class dependent treatment on\n PREEMPT_RT. Create a wrapper around hrtimer_start_expires() to make that\n possible.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n\ncommit dbc1625fc9deefb352f6ff26a575ae4b3ddef23a\nAuthor: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\nDate: Fri Jul 26 20:30:50 2019 +0200\n\n hrtimer: Consolidate hrtimer_init() + hrtimer_init_sleeper() calls\n \n hrtimer_init_sleeper() calls require prior initialisation of the hrtimer\n object which is embedded into the hrtimer_sleeper.\n \n Combine the initialization and spare a function call. Fixup all call sites.\n \n This is also a preparatory change for PREEMPT_RT to do hrtimer sleeper\n specific initializations of the embedded hrtimer without modifying any of\n the call sites.\n \n No functional change.\n \n [ anna-maria: Minor cleanups ]\n [ tglx: Adopted to the removal of the task argument of\n hrtimer_init_sleeper() and trivial polishing.\n Folded a fix from Stephen Rothwell for the vsoc code ]\n \n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Anna-Maria Gleixner \u003canna-maria@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190726185752.887468908@linutronix.de\n\ncommit 7455cdd1a0fe9a1367ee99596ea2564031daec00\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Mon Feb 11 12:13:57 2019 -0800\n\n tools/memory-model: Make scripts be executable\n \n This commit simplifies life a bit by making all of the scripts in\n tools/memory-model/scripts be executable.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n\ncommit 21176ebcd7d07b412bbed46620fd223d583c3c62\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Jul 31 14:45:55 2019 +0200\n\n serial: sh-sci: Use DEVICE_ATTR_RW() for rx_fifo_trigger\n \n While commit b6b996b6cdeecf7e (\"treewide: Use DEVICE_ATTR_RW\") converted\n the rx_fifo_timeout attribute, it forgot to convert rx_fifo_trigger due\n to a slightly different function naming.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Ulrich Hecht \u003culi+renesas@fpond.eu\u003e\n Link: https://lore.kernel.org/r/20190731124555.14349-1-geert+renesas@glider.be\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 340eda040af3a8bc4d0a3326f924b7e20244e42f\nAuthor: Vivek Gautam \u003cvivek.gautam@codeaurora.org\u003e\nDate: Thu Aug 1 17:41:53 2019 +0530\n\n tty: serial: qcom_geni_serial: Update the oversampling rate\n \n For QUP IP versions 2.5 and above the oversampling rate is halved\n from 32 to 16. Update this rate after reading hardware version\n register, so that the clock divider value is correctly set to\n achieve required baud rate.\n \n Signed-off-by: Vivek Gautam \u003cvivek.gautam@codeaurora.org\u003e\n Link: https://lore.kernel.org/r/20190801121153.10613-1-vivek.gautam@codeaurora.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 99710869174e97fe85962f4575fd2aa401d34248\nAuthor: Robert Middleton \u003crobert.middleton@rm5248.com\u003e\nDate: Thu Aug 1 10:56:40 2019 -0400\n\n serial: 8250_exar: Clear buffer before shutdown\n \n When closing and shutting down the exar serial port, if the chip\n has not finished sending all of the data in its buffer, the\n remaining bytes will be lost. Hold off on the shutdown until the\n bytes have all been sent.\n \n Signed-off-by: Robert Middleton \u003crobert.middleton@rm5248.com\u003e\n Link: https://lore.kernel.org/r/20190801145640.26080-1-robert.middleton@rm5248.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f9dd5b4321ec4e693d41f33cf402ac21672e84e1\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Wed Jul 31 20:05:58 2019 +0300\n\n serial: 8250_exar: Move custom divisor support out from 8250_port\n \n There are Exar custom divisor support in 8250_port which belongs to\n 8250_exar module. Move it out to the correct module and do not contaminate\n generic code with it.\n \n Cc: Aaron Sierra \u003casierra@xes-inc.com\u003e\n Cc: Jan Kiszka \u003cjan.kiszka@siemens.com\u003e\n Cc: Sudip Mukherjee \u003csudip.mukherjee@codethink.co.uk\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190731170558.52897-3-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7a7b7b73dbca770ca5a987f2b07586d844625b73\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Wed Jul 31 20:05:57 2019 +0300\n\n serial: 8250_exar: Extract PM routine from 8250_port\n \n There are Exar quirks in 8250_port which belong to 8250_exar module.\n Extract PM routine to the correct module and do not contaminate generic code\n with it.\n \n Cc: Aaron Sierra \u003casierra@xes-inc.com\u003e\n Cc: Jan Kiszka \u003cjan.kiszka@siemens.com\u003e\n Cc: Sudip Mukherjee \u003csudip.mukherjee@codethink.co.uk\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190731170558.52897-2-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 1f06cb7f4fabdc1e0c25130864f56422a9367bb4\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Wed Jul 31 20:05:56 2019 +0300\n\n serial: 8250_exar: No need to autoconfigure Exar ports\n \n Since we have a separate driver there is no need to autoconfigure ports,\n we already know what they are.\n \n Drop autoconfiguration in 8250_port and move type detection to 8250_exar.\n \n Cc: Aaron Sierra \u003casierra@xes-inc.com\u003e\n Cc: Jan Kiszka \u003cjan.kiszka@siemens.com\u003e\n Cc: Sudip Mukherjee \u003csudip.mukherjee@codethink.co.uk\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190731170558.52897-1-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f839b4b549a65e1c611f911654012ac65e6d1610\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:55 2019 -0700\n\n platform/x86: intel_int0002_vgpio: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Cc: Hans de Goede \u003chdegoede@redhat.com\u003e\n Cc: \"Darren Hart (VMware)\" \u003cdvhart@infradead.org\u003e\n Cc: Roman Kiryanov \u003crkir@google.com\u003e\n Cc: Vadim Pasternak \u003cvadimp@mellanox.com\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit b2dd2d9a49e0a9251badde3363efc9e7be2d0bde\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:54 2019 -0700\n\n platform/x86: intel_bxtwc_tmu: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Cc: Hans de Goede \u003chdegoede@redhat.com\u003e\n Cc: \"Darren Hart (VMware)\" \u003cdvhart@infradead.org\u003e\n Cc: Roman Kiryanov \u003crkir@google.com\u003e\n Cc: Vadim Pasternak \u003cvadimp@mellanox.com\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit eaae882c4a77c8aa2ca20d4447aeb81d9d5eec97\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:53 2019 -0700\n\n platform/mellanox: mlxreg-hotplug: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Cc: Hans de Goede \u003chdegoede@redhat.com\u003e\n Cc: \"Darren Hart (VMware)\" \u003cdvhart@infradead.org\u003e\n Cc: Roman Kiryanov \u003crkir@google.com\u003e\n Cc: Vadim Pasternak \u003cvadimp@mellanox.com\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 50944213062f7605ef79604418c7e3468d97b109\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:52 2019 -0700\n\n platform/x86: intel_pmc_ipc: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Cc: Hans de Goede \u003chdegoede@redhat.com\u003e\n Cc: \"Darren Hart (VMware)\" \u003cdvhart@infradead.org\u003e\n Cc: Roman Kiryanov \u003crkir@google.com\u003e\n Cc: Vadim Pasternak \u003cvadimp@mellanox.com\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 8ef7e3409ceccec713ca8de6608817fb57954024\nAuthor: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\nDate: Tue Jul 30 15:05:53 2019 -0700\n\n drm/i915/ehl: Don\u0027t forget to handle port C\u0027s hotplug interrupts\n \n We\u0027re mostly re-using ICL\u0027s interrupt handling on EHL, but we still need\n to remember to account for the extra combo port that EHL has. Use TGP\u0027s\n mask (which includes combo port C) rather than ICP\u0027s mask when\n appropriate. Let\u0027s also skip reading TC-specific registers on this\n platform since EHL doesn\u0027t have any TC ports.\n \n v2: Base setup of SHOTPLUG_CTL_TC on whether the tc pin mask is non-zero\n rather than performing another PCH type check. (Jose)\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Vivek Kasireddy \u003cvivek.kasireddy@intel.com\u003e\n Cc: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Signed-off-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730220553.15300-1-matthew.d.roper@intel.com\n\ncommit b3173c2292fbaf24ff7062d366830b012ed04269\nAuthor: Saravana Kannan \u003csaravanak@google.com\u003e\nDate: Wed Jul 31 15:17:20 2019 -0700\n\n of/platform: Don\u0027t create device links for default busses\n \n Default busses also have devices created for them. But there\u0027s no point\n in creating device links for them. It\u0027s especially wasteful as it\u0027ll\n cause the traversal of the entire device tree and also spend a lot of\n time checking and figuring out that creating those links isn\u0027t allowed.\n So check for default busses and skip trying to create device links for\n them.\n \n Signed-off-by: Saravana Kannan \u003csaravanak@google.com\u003e\n Link: https://lore.kernel.org/r/20190731221721.187713-8-saravanak@google.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 709fb8297358f08a125b770d1518a95d03b541db\nAuthor: Saravana Kannan \u003csaravanak@google.com\u003e\nDate: Wed Jul 31 15:17:19 2019 -0700\n\n of/platform: Create device links for all child-supplier depencencies\n \n A parent device can have child devices that it adds when it probes. But\n this probing of the parent device can happen way after kernel init is done\n -- for example, when the parent device\u0027s driver is loaded as a module.\n \n In such cases, if the child devices depend on a supplier in the system, we\n need to make sure the supplier gets the sync_state() callback only after\n these child devices are added and probed.\n \n To achieve this, when creating device links for a device by looking at its\n DT node, don\u0027t just look at DT references at the top node level. Look at DT\n references in all the descendant nodes too and create device links from the\n ancestor device to all these supplier devices.\n \n This way, when the parent device probes and adds child devices, the child\n devices can then create their own device links to the suppliers and further\n delay the supplier\u0027s sync_state() callback to after the child devices are\n probed.\n \n Example:\n In this illustration, -\u003e denotes DT references and indentation\n represents child status.\n \n Device node A\n Device node B -\u003e D\n Device node C -\u003e B, D\n \n Device node D\n \n Assume all these devices have their drivers loaded as modules.\n \n Without this patch, this is the sequence of events:\n 1. D is added.\n 2. A is added.\n 3. Device D probes.\n 4. Device D gets its sync_state() callback.\n 5. Device B and C might malfunction because their resources got\n altered/turned off before they can make active requests for them.\n \n With this patch, this is the sequence of events:\n 1. D is added.\n 2. A is added and creates device links to D.\n 3. Device link from A to B is not added because A is a parent of B.\n 4. Device D probes.\n 5. Device D does not get it\u0027s sync_state() callback because consumer A\n hasn\u0027t probed yet.\n 5. Device A probes.\n 5. a. Devices B and C are added.\n 5. b. Device links from B and C to D are added.\n 5. c. Device A\u0027s probe completes.\n 6. Device D does not get it\u0027s sync_state() callback because consumer A\n has probed but consumers B and C haven\u0027t probed yet.\n 7. Device B and C probe.\n 8. Device D gets it\u0027s sync_state() callback because all its consumers\n have probed.\n 9. None of the devices malfunction.\n \n Signed-off-by: Saravana Kannan \u003csaravanak@google.com\u003e\n Link: https://lore.kernel.org/r/20190731221721.187713-7-saravanak@google.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 21871a99b34c65c56a24193c277a4981529c306f\nAuthor: Saravana Kannan \u003csaravanak@google.com\u003e\nDate: Wed Jul 31 15:17:18 2019 -0700\n\n of/platform: Pause/resume sync state during init and of_platform_populate()\n \n When all the top level devices are populated from DT during kernel\n init, the supplier devices could be added and probed before the\n consumer devices are added and linked to the suppliers. To avoid the\n sync_state() callback from being called prematurely, pause the\n sync_state() callbacks before populating the devices and resume them\n at late_initcall_sync().\n \n Similarly, when children devices are populated after kernel init using\n of_platform_populate(), there could be supplier-consumer dependencies\n between the children devices that are populated. To avoid the same\n problem with sync_state() being called prematurely, pause and resume\n sync_state() callbacks across of_platform_populate().\n \n Signed-off-by: Saravana Kannan \u003csaravanak@google.com\u003e\n Link: https://lore.kernel.org/r/20190731221721.187713-6-saravanak@google.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 8f8184d6bf676a8680d6f441e40317d166b46f73\nAuthor: Saravana Kannan \u003csaravanak@google.com\u003e\nDate: Wed Jul 31 15:17:17 2019 -0700\n\n driver core: Add sync_state driver/bus callback\n \n This sync_state driver/bus callback is called once all the consumers\n of a supplier have probed successfully.\n \n This allows the supplier device\u0027s driver/bus to sync the supplier\n device\u0027s state to the software state with the guarantee that all the\n consumers are actively managing the resources provided by the supplier\n device.\n \n To maintain backwards compatibility and ease transition from existing\n frameworks and resource cleanup schemes, late_initcall_sync is the\n earliest when the sync_state callback might be called.\n \n There is no upper bound on the time by which the sync_state callback\n has to be called. This is because if a consumer device never probes,\n the supplier has to maintain its resources in the state left by the\n bootloader. For example, if the bootloader leaves the display\n backlight at a fixed voltage and the backlight driver is never probed,\n you don\u0027t want the backlight to ever be turned off after boot up.\n \n Also, when multiple devices are added after kernel init, some\n suppliers could be added before their consumer devices get added. In\n these instances, the supplier devices could get their sync_state\n callback called right after they probe because the consumers devices\n haven\u0027t had a chance to create device links to the suppliers.\n \n To handle this correctly, this change also provides APIs to\n pause/resume sync state callbacks so that when multiple devices are\n added, their sync_state callback evaluation can be postponed to happen\n after all of them are added.\n \n kbuild test robot reported missing documentation for device.state_synced\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Saravana Kannan \u003csaravanak@google.com\u003e\n Link: https://lore.kernel.org/r/20190731221721.187713-5-saravanak@google.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 690ff7881b2681afca1c3c063f4a5cb7c71d9d8b\nAuthor: Saravana Kannan \u003csaravanak@google.com\u003e\nDate: Wed Jul 31 15:17:16 2019 -0700\n\n of/platform: Add functional dependency link from DT bindings\n \n Add device-links after the devices are created (but before they are\n probed) by looking at common DT bindings like clocks and\n interconnects.\n \n Automatically adding device-links for functional dependencies at the\n framework level provides the following benefits:\n \n - Optimizes device probe order and avoids the useless work of\n attempting probes of devices that will not probe successfully\n (because their suppliers aren\u0027t present or haven\u0027t probed yet).\n \n For example, in a commonly available mobile SoC, registering just\n one consumer device\u0027s driver at an initcall level earlier than the\n supplier device\u0027s driver causes 11 failed probe attempts before the\n consumer device probes successfully. This was with a kernel with all\n the drivers statically compiled in. This problem gets a lot worse if\n all the drivers are loaded as modules without direct symbol\n dependencies.\n \n - Supplier devices like clock providers, interconnect providers, etc\n need to keep the resources they provide active and at a particular\n state(s) during boot up even if their current set of consumers don\u0027t\n request the resource to be active. This is because the rest of the\n consumers might not have probed yet and turning off the resource\n before all the consumers have probed could lead to a hang or\n undesired user experience.\n \n Some frameworks (Eg: regulator) handle this today by turning off\n \"unused\" resources at late_initcall_sync and hoping all the devices\n have probed by then. This is not a valid assumption for systems with\n loadable modules. Other frameworks (Eg: clock) just don\u0027t handle\n this due to the lack of a clear signal for when they can turn off\n resources. This leads to downstream hacks to handle cases like this\n that can easily be solved in the upstream kernel.\n \n By linking devices before they are probed, we give suppliers a clear\n count of the number of dependent consumers. Once all of the\n consumers are active, the suppliers can turn off the unused\n resources without making assumptions about the number of consumers.\n \n By default we just add device-links to track \"driver presence\" (probe\n succeeded) of the supplier device. If any other functionality provided\n by device-links are needed, it is left to the consumer/supplier\n devices to change the link when they probe.\n \n kbuild test robot reported clang error about missing const\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Saravana Kannan \u003csaravanak@google.com\u003e\n Link: https://lore.kernel.org/r/20190731221721.187713-4-saravanak@google.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 134b23eec9e3a3c795a6ceb0efe2fa63e87983b2\nAuthor: Saravana Kannan \u003csaravanak@google.com\u003e\nDate: Wed Jul 31 15:17:15 2019 -0700\n\n driver core: Add edit_links() callback for drivers\n \n The driver core/bus adding supplier-consumer dependencies by default\n enables functional dependencies to be tracked correctly even when the\n consumer devices haven\u0027t had their drivers registered or loaded (if they\n are modules).\n \n However, when the bus incorrectly adds dependencies that it shouldn\u0027t\n have added, the devices might never probe.\n \n For example, if device-C is a consumer of device-S and they have\n phandles to each other in DT, the following could happen:\n \n 1. Device-S get added first.\n 2. The bus add_links() callback will (incorrectly) try to link it as\n a consumer of device-C.\n 3. Since device-C isn\u0027t present, device-S will be put in\n \"waiting-for-supplier\" list.\n 4. Device-C gets added next.\n 5. All devices in \"waiting-for-supplier\" list are retried for linking.\n 6. Device-S gets linked as consumer to Device-C.\n 7. The bus add_links() callback will (correctly) try to link it as\n a consumer of device-S.\n 8. This isn\u0027t allowed because it would create a cyclic device links.\n \n Neither devices will get probed since the supplier is marked as\n dependent on the consumer. And the consumer will never probe because the\n consumer can\u0027t get resources from the supplier.\n \n Without this patch, things stay in this broken state. However, with this\n patch, the execution will continue like this:\n \n 9. Device-C\u0027s driver is loaded.\n 10. Device-C\u0027s driver removes Device-S as a consumer of Device-C.\n 11. Device-C\u0027s driver adds Device-C as a consumer of Device-S.\n 12. Device-S probes.\n 14. Device-C probes.\n \n kbuild test robot reported missing documentation for device.has_edit_links\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Saravana Kannan \u003csaravanak@google.com\u003e\n Link: https://lore.kernel.org/r/20190731221721.187713-3-saravanak@google.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 5302dd7dd0b6d04c63cdce51d1e9fda9ef0be886\nAuthor: Saravana Kannan \u003csaravanak@google.com\u003e\nDate: Wed Jul 31 15:17:14 2019 -0700\n\n driver core: Add support for linking devices during device addition\n \n When devices are added, the bus might want to create device links to track\n functional dependencies between supplier and consumer devices. This\n tracking of supplier-consumer relationship allows optimizing device probe\n order and tracking whether all consumers of a supplier are active. The\n add_links bus callback is added to support this.\n \n However, when consumer devices are added, they might not have a supplier\n device to link to despite needing mandatory resources/functionality from\n one or more suppliers. A waiting_for_suppliers list is created to track\n such consumers and retry linking them when new devices get added.\n \n Signed-off-by: Saravana Kannan \u003csaravanak@google.com\u003e\n Link: https://lore.kernel.org/r/20190731221721.187713-2-saravanak@google.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 313b46d831189f593840c625d7972092cb0088fc\nAuthor: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\nDate: Thu Aug 1 11:20:26 2019 +0100\n\n drivers: Fix htmldocs warnings with bus_find_next_device()\n \n Document the parameters for bus_find_next_device() to avoid\n htmldocs build warnings as reported below :\n \n include/linux/device.h:236: warning: Function parameter or member \u0027bus\u0027 not described in \u0027bus_find_next_device\u0027\n include/linux/device.h:236: warning: Function parameter or member \u0027cur\u0027 not described in \u0027bus_find_next_device\u0027\n \n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Link: https://lore.kernel.org/r/20190801102026.27312-3-suzuki.poulose@arm.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b9515ecbf6caef2ea911ca59801eff84d034fa48\nAuthor: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\nDate: Thu Aug 1 11:20:25 2019 +0100\n\n drivers: Fix typo in parameter description for driver_find_device_by_acpi_dev\n \n Fix a typo in the comment describing the parameters for the new API, which\n triggers the following warning for htmldocs:\n \n include/linux/device.h:479: warning: Function parameter or member \u0027drv\u0027 not described in \u0027driver_find_device_by_acpi_dev\u0027\n \n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Link: https://lore.kernel.org/r/20190801102026.27312-2-suzuki.poulose@arm.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 644bf60088955421051e716ab9c8fe7fb7997fd7\nAuthor: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\nDate: Thu Aug 1 11:20:24 2019 +0100\n\n i2c: Revert incorrect conversion to use generic helper\n \n The patch \"drivers: Introduce device lookup variants by ACPI_COMPANION device\"\n converted an incorrect instance in i2c driver to a new helper. Revert this\n change.\n \n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Cc: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n Cc: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n Fixes: 00500147cbd3 (\"drivers: Introduce device lookup variants by ACPI_COMPANION device\")\n Signed-off-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Link: https://lore.kernel.org/r/20190801102026.27312-1-suzuki.poulose@arm.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit cc8b4659491d23577eb8992449d82c7f4f4fc170\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Jul 31 14:47:38 2019 +0200\n\n spi: core: Use DEVICE_ATTR_RW() for SPI slave control sysfs attribute\n \n Convert the SPI slave control sysfs attribute from DEVICE_ATTR() to\n DEVICE_ATTR_RW(), to reduce boilerplate.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Link: https://lore.kernel.org/r/20190731124738.14519-1-geert+renesas@glider.be\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9d52a35ebd4bb6e495ab06e4ce1f568daa0f70e1\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Wed Jul 31 15:29:23 2019 +0200\n\n regmap: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n The debugfs core will warn if a file or directory can not be created, so\n there\u0027s no need to duplicate the warning, nor really do anything else.\n \n Cc: Mark Brown \u003cbroonie@kernel.org\u003e\n Cc: \"Rafael J. Wysocki\" \u003crafael@kernel.org\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Link: https://lore.kernel.org/r/20190731132923.GA13829@kroah.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit be9035630d80ac0dd318d07c802bc3c1d4cbe036\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Sun Jul 28 18:51:38 2019 -0500\n\n i2c: s3c2410: Mark expected switch fall-through\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warning:\n \n drivers/i2c/busses/i2c-s3c2410.c: In function \u0027i2c_s3c_irq_nextbyte\u0027:\n drivers/i2c/busses/i2c-s3c2410.c:431:6: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n if (i2c-\u003estate \u003d\u003d STATE_READ)\n ^\n drivers/i2c/busses/i2c-s3c2410.c:439:2: note: here\n case STATE_WRITE:\n ^~~~\n \n Notice that, in this particular case, the code comment is\n modified in accordance with what GCC is expecting to find.\n \n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit 33eb09a02e8d8fad976019694196eec2cc210d62\nAuthor: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\nDate: Thu Jul 25 22:21:36 2019 +0200\n\n i2c: designware: make use of devm_gpiod_get_optional\n \n There is a semantical change: if devm_gpiod_get_optional returns -ENOSYS\n this is passed as error to the caller. This effectively reverts commit\n d1fa74520dcd (\"i2c: designware: Consider SCL GPIO optional\") which\n shouldn\u0027t be necessary any more since gpiod_get_optional doesn\u0027t return\n -ENOSYS any more with GPIOLIB\u003dn.\n \n Signed-off-by: Uwe Kleine-König \u003cuwe@kleine-koenig.org\u003e\n Reviewed-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Jarkko Nikula \u003cjarkko.nikula@linux.intel.com\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit 684ca71259a69c5a3019da72fe718bf983841926\nAuthor: Simon Horman \u003chorms+renesas@verge.net.au\u003e\nDate: Wed Jul 24 14:15:59 2019 +0200\n\n dt-bindings: i2c: riic: Rename bindings documentation file\n \n Rename the bindings documentation file for Renesas EMEV2 IIC controller\n from i2c-emev2.txt to renesas,iic-emev2.txt.\n \n This is part of an ongoing effort to name bindings documentation files for\n Renesas IP blocks consistently, in line with the compat strings they\n document.\n \n Signed-off-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit 1d583590514a4f1c9aa582d1234fca7aa3d627e9\nAuthor: Simon Horman \u003chorms+renesas@verge.net.au\u003e\nDate: Wed Jul 24 14:15:58 2019 +0200\n\n dt-bindings: i2c: riic: Rename bindings documentation file\n \n Rename the bindings documentation file for RIIC controller\n from i2c-riic.txt to renesas,riic.txt.\n \n This is part of an ongoing effort to name bindings documentation files for\n Renesas IP blocks consistently, in line with the compat strings they\n document.\n \n Signed-off-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit d13ed84b195cc6e5789b446f07aede357939f7ad\nAuthor: Simon Horman \u003chorms+renesas@verge.net.au\u003e\nDate: Wed Jul 24 14:15:57 2019 +0200\n\n dt-bindings: i2c: rcar: Rename bindings documentation file\n \n Rename the bindings documentation file for R-Car I2C controller\n from i2c-rcar.txt to renesas,rcar.txt.\n \n This is part of an ongoing effort to name bindings documentation files for\n Renesas IP blocks consistently, in line with the compat strings they\n document.\n \n Signed-off-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit 23c2556d8fbec57e72e2c8fd8b7be2715c192b25\nAuthor: Simon Horman \u003chorms+renesas@verge.net.au\u003e\nDate: Wed Jul 24 14:15:56 2019 +0200\n\n dt-bindings: i2c: sh_mobile: Rename bindings documentation file\n \n Rename the bindings documentation file for sh_mobile I2C controller\n from i2c-sh_mobile.txt to renesas,iic.txt.\n \n This is part of an ongoing effort to name bindings documentation files for\n Renesas IP blocks consistently, in line with the compat strings they\n document.\n \n Signed-off-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit f24e41d3d04f326613d8a7ebecf72c3019826f71\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Wed Jul 31 23:32:34 2019 +0100\n\n ASoC: zx-tdm: remove redundant assignment to ts_width on error return path\n \n The value assigned to ts_width is never read on the error return path\n so the assignment is redundant and can be removed. Remove it.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190731223234.16153-1-colin.king@canonical.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 3ff3a4f657b3fab2d56247983c2ebed180ef2fbb\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Wed Jul 31 15:17:16 2019 +0200\n\n ASoC: SOF: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Also, if a debugfs call fails, userspace is notified with an error in\n the log, so no need to log the error again.\n \n Because we no longer need to check the return value, there\u0027s no need to\n save the dentry returned by debugfs. Just use the dentry in the file\n pointer if we really need to figure out the \"name\" of the file being\n opened.\n \n Cc: Liam Girdwood \u003clgirdwood@gmail.com\u003e\n Cc: Mark Brown \u003cbroonie@kernel.org\u003e\n Cc: Jaroslav Kysela \u003cperex@perex.cz\u003e\n Cc: Takashi Iwai \u003ctiwai@suse.com\u003e\n Cc: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Cc: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Cc: alsa-devel@alsa-project.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190731131716.9764-3-gregkh@linuxfoundation.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit fee531d6fc49b9a616525e30955c4cf3b403f632\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Wed Jul 31 15:17:15 2019 +0200\n\n ASoC: core: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Also, there is no need to store the individual debugfs file name, just\n remove the whole directory all at once, saving a local variable.\n \n Note, the soc-pcm \"state\" file has now moved to a subdirectory, as it is\n only a good idea to save the dentries for debugfs directories, not\n individual files, as the individual file debugfs functions are changing\n to not return a dentry.\n \n Cc: Liam Girdwood \u003clgirdwood@gmail.com\u003e\n Cc: Mark Brown \u003cbroonie@kernel.org\u003e\n Cc: Jaroslav Kysela \u003cperex@perex.cz\u003e\n Cc: Takashi Iwai \u003ctiwai@suse.com\u003e\n Cc: alsa-devel@alsa-project.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Link: https://lore.kernel.org/r/20190731131716.9764-2-gregkh@linuxfoundation.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 7735eeebd2be26d6d3fa151c85d839d62d710d93\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 19:11:10 2019 +0800\n\n i2c: busses: Use dev_get_drvdata where possible\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Reviewed-by: Jean Delvare \u003cjdelvare@suse.de\u003e\n Acked-by: Jarkko Nikula \u003cjarkko.nikula@linux.intel.com\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit d8701fae95c0e6cb587b293cb8e01acc96dd1095\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Wed Jul 31 15:17:14 2019 +0200\n\n ASoC: Intel: SoC: skylake: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Also, if a debugfs call fails, userspace is notified with an error in\n the log, so no need to log the error again.\n \n Cc: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Cc: Liam Girdwood \u003cliam.r.girdwood@linux.intel.com\u003e\n Cc: Jie Yang \u003cyang.jie@linux.intel.com\u003e\n Cc: Mark Brown \u003cbroonie@kernel.org\u003e\n Cc: Jaroslav Kysela \u003cperex@perex.cz\u003e\n Cc: Takashi Iwai \u003ctiwai@suse.com\u003e\n Cc: alsa-devel@alsa-project.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Link: https://lore.kernel.org/r/20190731131716.9764-1-gregkh@linuxfoundation.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9de98628c895d15427138073986eab1e3ce39cb4\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Thu Jul 25 18:59:49 2019 +0200\n\n ASoC: codec2codec: fill some of the runtime stream parameters\n \n Set the information provided struct snd_soc_pcm_stream in the\n struct snd_pcm_runtime of the codec to codec link.\n \n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Link: https://lore.kernel.org/r/20190725165949.29699-7-jbrunet@baylibre.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit a72706ed8208ac3f72d1c3ebbc6509e368b0dcb0\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Thu Jul 25 18:59:48 2019 +0200\n\n ASoC: codec2codec: remove ephemeral variables\n \n Now that codec to codec links struct snd_soc_pcm_runtime have lasting pcm\n and substreams, let\u0027s use them. Alsa allocate and keep the\n struct snd_pcm_runtime as long as the link is powered.\n \n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Link: https://lore.kernel.org/r/20190725165949.29699-6-jbrunet@baylibre.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit a342031cdd0818cb0fbcb44798211c7a02c7ca27\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Thu Jul 25 18:59:47 2019 +0200\n\n ASoC: create pcm for codec2codec links as well\n \n At the moment, codec to codec links uses an ephemeral variable for\n the struct snd_pcm_substream. Also the struct snd_soc_pcm_runtime\n does not have real struct snd_pcm.\n \n This might a problem if the functions used by a codec on codec to\n codec link expect these structures to exist, and keep on existing\n during the life of the codec.\n \n For example, it is the case of the hdmi-codec, which uses\n snd_pcm_add_chmap_ctls(). For the controls to works, the pcm and\n substream must to exist.\n \n This change is first step, it create pcm (and substreams) for codec\n to codec links, in the same way as dpcm backend links.\n \n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Link: https://lore.kernel.org/r/20190725165949.29699-5-jbrunet@baylibre.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 03929cce52e285c9e6da9237fae168264e7c5785\nAuthor: Michał Mirosław \u003cmirq-linux@rere.qmqm.pl\u003e\nDate: Mon Jul 22 21:05:56 2019 +0200\n\n i2c: at91: fix clk_offset for sama5d2\n \n In SAMA5D2 datasheet, TWIHS_CWGR register rescription mentions clock\n offset of 3 cycles (compared to 4 in eg. SAMA5D3).\n \n Cc: stable@vger.kernel.org # 5.2.x\n [needs applying to i2c-at91.c instead for earlier kernels]\n Fixes: 0ef6f3213dac (\"i2c: at91: add support for new alternative command mode\")\n Signed-off-by: Michał Mirosław \u003cmirq-linux@rere.qmqm.pl\u003e\n Acked-by: Ludovic Desroches \u003cludovic.desroches@microchip.com\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit 9c5718e14b811190e847451715939c230ab598c6\nAuthor: Michał Mirosław \u003cmirq-linux@rere.qmqm.pl\u003e\nDate: Mon Jul 22 20:55:27 2019 +0200\n\n i2c: at91: disable TXRDY interrupt after sending data\n \n Driver was not disabling TXRDY interrupt after last TX byte.\n This caused interrupt storm until transfer timeouts for slow\n or broken device on the bus. The patch fixes the interrupt storm\n on my SAMA5D2-based board.\n \n Cc: stable@vger.kernel.org # 5.2.x\n [v5.2 introduced file split; the patch should apply to i2c-at91.c before the split]\n Fixes: fac368a0404 (\"i2c: at91: add new driver\")\n Signed-off-by: Michał Mirosław \u003cmirq-linux@rere.qmqm.pl\u003e\n Acked-by: Ludovic Desroches \u003cludovic.desroches@microchip.com\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit 5667b5b59f45bd0bebadfae9ed55745b4cff2198\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jul 17 16:40:17 2019 +0800\n\n i2c: imx-lpi2c: use devm_platform_ioremap_resource() to simplify code\n \n Use the new helper devm_platform_ioremap_resource() which wraps the\n platform_get_resource() and devm_ioremap_resource() together, to\n simplify the code.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Acked-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit b17e6d19dcd3fcb2b93166034e739e82b49d2a51\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jul 17 16:40:16 2019 +0800\n\n i2c: mxs: use devm_platform_ioremap_resource() to simplify code\n \n Use the new helper devm_platform_ioremap_resource() which wraps the\n platform_get_resource() and devm_ioremap_resource() together, to\n simplify the code.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit 34de3513e668d5ad46bd32ebee63f9e3dc57e1db\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Mon Jul 15 11:17:39 2019 +0800\n\n i2c: ismt: Remove call to memset after dmam_alloc_coherent\n \n In commit 518a2f1925c3 (\"dma-mapping: zero memory returned from\n dma_alloc_*\"), dma_alloc_coherent has already zeroed the memory. So\n memset is not needed.\n \n Acked-by: Neil Horman \u003cnhorman@tuxdriver.com\u003e\n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit e5738bc46d49642dc4804bad3656b23016f8e1f1\nAuthor: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nDate: Mon Jul 8 02:12:34 2019 +0300\n\n i2c: tegra: Compile PM functions unconditionally\n \n The I2C driver fails to probe if CONFIG_PM_SLEEP\u003dn because runtime PM\n doesn\u0027t depend on the PM sleep and in this case the runtime PM ops are\n not included in the driver, resulting in I2C clock not being enabled.\n It\u0027s much cleaner to simply allow compiler to remove the dead code\n instead of messing with the #ifdefs.\n \n This patch fixes such errors when CONFIG_PM_SLEEP\u003dn:\n \n tegra-i2c 7000c400.i2c: timeout waiting for fifo flush\n tegra-i2c 7000c400.i2c: Failed to initialize i2c controller\n \n Signed-off-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Acked-by: Jon Hunter \u003cjonathanh@nvidia.com\u003e\n Signed-off-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit b36f281f4a314de4be0a51d6511b794691f8a244\nAuthor: Mimi Zohar \u003czohar@linux.ibm.com\u003e\nDate: Fri Jul 19 07:16:57 2019 -0400\n\n ima: initialize the \"template\" field with the default template\n \n IMA policy rules are walked sequentially. Depending on the ordering of\n the policy rules, the \"template\" field might be defined in one rule, but\n will be replaced by subsequent, applicable rules, even if the rule does\n not explicitly define the \"template\" field.\n \n This patch initializes the \"template\" once and only replaces the\n \"template\", when explicitly defined.\n \n Fixes: 19453ce0bcfb (\"IMA: support for per policy rule template formats\")\n Signed-off-by: Mimi Zohar \u003czohar@linux.ibm.com\u003e\n\ncommit 8ad4ca6e9a8166495b5ba843375f199c7f0c218e\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Tue Jul 30 14:34:30 2019 +0300\n\n drm/i915/oa: add content to Makefile\n \n Apparently the empty Makefile has caused some confusion. Add the\n subdir-cc-flags-y as in 7fcc7ca549d4 (\"drm/i915: add header search path\n to subdir Makefiles\") which should be useful.\n \n The generated headers still aren\u0027t self-contained, so can\u0027t add that.\n \n References: http://marc.info/?i\u003d80bf2204-558a-6d3f-c493-bf17b891fc8a@infradead.org\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730113432.22146-1-jani.nikula@intel.com\n\ncommit 93386368a138e2ccc40459af5b05dd2e39bcd3db\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Thu Aug 1 13:53:13 2019 +0200\n\n Revert \"drm/radeon: Provide ddc symlink in connector sysfs directory\"\n \n This reverts commit bed7a2182de6833f5d0f990a656bffb1c6000c70.\n \n It causes the following build error:\n ../drm/radeon/radeon_connectors.c: In function \u0027radeon_add_legacy_connector\u0027:\n ../drm/radeon/radeon_connectors.c:2433:5: error: \u0027ddc\u0027 undeclared (first use in this function)\n ddc \u003d \u0026radeon_connector-\u003eddc_bus-\u003eadapter;\n ^~~\n \n Suggested-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Acked-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190801115313.22562-1-narmstrong@baylibre.com\n\ncommit de70fdd7d24cd07e51fbec420f8704d956a47949\nAuthor: Wei Yongjun \u003cweiyongjun1@huawei.com\u003e\nDate: Thu Jul 4 10:45:34 2019 +0000\n\n drm/i915: fix possible memory leak in intel_hdcp_auth_downstream()\n \n \u0027ksv_fifo\u0027 is malloced in intel_hdcp_auth_downstream() and should be\n freed before leaving from the error handling cases, otherwise it will\n cause memory leak.\n \n Fixes: f26ae6a652f2 (\"drm/i915: SRM revocation check for HDCP1.4 and 2.2\")\n Signed-off-by: Wei Yongjun \u003cweiyongjun1@huawei.com\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190704104534.12508-1-weiyongjun1@huawei.com\n\ncommit fb0d72c7ac5d1a3578e15cb56b4f453730790ba6\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Thu Aug 1 09:54:54 2019 +0200\n\n dt-bindings: reset: amlogic,meson8b-reset: update with SPDX Licence identifier\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit b16a0063654516bde54a95262d02207d79f1bf53\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Thu Aug 1 09:54:53 2019 +0200\n\n dt-bindings: reset: amlogic,meson-gxbb-reset: update with SPDX Licence identifier\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit 7c61ccf643a9aa308c96508dceb93afe2083afa5\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Thu Aug 1 09:54:52 2019 +0200\n\n reset: reset-meson: update with SPDX Licence identifier\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit 942b4c10b11bced65c0e553d8355b58bb3d9cdb2\nAuthor: Guido Günther \u003cagx@sigxcpu.org\u003e\nDate: Thu Aug 1 10:20:59 2019 +0200\n\n dt-bindings: reset: Fix typo in imx8mq resets\n \n Some of the mipi dsi resets were called\n \n IMX8MQ_RESET_MIPI_DIS__\n \n instead of\n \n IMX8MQ_RESET_MIPI_DSI__\n \n Since they\u0027re DSI related this looks like a typo. This fixes the\n only in tree user as well to not break bisecting.\n \n Signed-off-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n Reviewed-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit e2557157a9dd47b0335cac325b4eb438db2320d2\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Fri Jul 5 16:54:05 2019 +0800\n\n dt-bindings: reset: imx7: Add support for i.MX8MM\n \n i.MX8MM can reuse i.MX8MQ\u0027s reset driver, update the compatible\n property and related info to support i.MX8MM.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n\ncommit 10e36489abda787e10b21a1e006121dee89c8f1f\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jul 30 14:30:19 2019 +0100\n\n drm/i915/execlists: Always clear pending\u0026inflight requests on reset\n \n If we skip the reset as we found the engine inactive at the time of the\n reset, we still need to clear the residual inflight \u0026 pending request\n bookkeeping to reflect the current state of HW.\n \n Otherwise, we may end up stuck in a loop like:\n \n \u003c7\u003e [416.490346] hangcheck rcs0\n \u003c7\u003e [416.490371] hangcheck Awake? 1\n \u003c7\u003e [416.490376] hangcheck Hangcheck: 8003 ms ago\n \u003c7\u003e [416.490380] hangcheck Reset count: 0 (global 0)\n \u003c7\u003e [416.490383] hangcheck Requests:\n \u003c7\u003e [416.491210] hangcheck RING_START: 0x0017b000\n \u003c7\u003e [416.491983] hangcheck RING_HEAD: 0x00000048\n \u003c7\u003e [416.491992] hangcheck RING_TAIL: 0x00000048\n \u003c7\u003e [416.492006] hangcheck RING_CTL: 0x00000000\n \u003c7\u003e [416.492037] hangcheck RING_MODE: 0x00000200 [idle]\n \u003c7\u003e [416.492044] hangcheck RING_IMR: 00000000\n \u003c7\u003e [416.492809] hangcheck ACTHD: 0x00000000_9ca00048\n \u003c7\u003e [416.492824] hangcheck BBADDR: 0x00000000_00001004\n \u003c7\u003e [416.492838] hangcheck DMA_FADDR: 0x00000000_00000000\n \u003c7\u003e [416.492845] hangcheck IPEIR: 0x00000000\n \u003c7\u003e [416.492852] hangcheck IPEHR: 0x00000000\n \u003c7\u003e [416.492863] hangcheck Execlist status: 0x00018001 00000000, entries 12\n \u003c7\u003e [416.492869] hangcheck Execlist CSB read 1, write 1, tasklet queued? no (enabled)\n \u003c7\u003e [416.492938] hangcheck Pending[0] ring:{start:0017b000, hwsp:fedf9000, seqno:00016fd6}, rq: 20ffa:16fd6!+ prio\u003d-4094 @ 8307ms: signaled\n \u003c7\u003e [416.492972] hangcheck Queue priority hint: -4093\n \u003c7\u003e [416.492979] hangcheck Q 20ffa:16fd8- prio\u003d-4093 @ 8307ms: [i915]\n \u003c7\u003e [416.492985] hangcheck Q 20ffa:16fda prio\u003d-4094 @ 8307ms: [i915]\n \u003c7\u003e [416.492990] hangcheck Q 20ffa:16fdc prio\u003d-4094 @ 8307ms: [i915]\n \u003c7\u003e [416.492996] hangcheck Q 20ffa:16fde prio\u003d-4094 @ 8307ms: [i915]\n \u003c7\u003e [416.493001] hangcheck Q 20ffa:16fe0 prio\u003d-4094 @ 8307ms: [i915]\n \u003c7\u003e [416.493007] hangcheck Q 20ffa:16fe2 prio\u003d-4094 @ 8307ms: [i915]\n \u003c7\u003e [416.493013] hangcheck Q 20ffa:16fe4 prio\u003d-4094 @ 8307ms: [i915]\n \u003c7\u003e [416.493021] hangcheck ...skipping 21 queued requests...\n \u003c7\u003e [416.493027] hangcheck Q 20ffa:17010 prio\u003d-4094 @ 8307ms: [i915]\n \u003c7\u003e [416.493081] hangcheck HWSP:\n \u003c7\u003e [416.493089] hangcheck [0000] 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000\n \u003c7\u003e [416.493094] hangcheck *\n \u003c7\u003e [416.493100] hangcheck [0040] 10008002 00000000 10000018 00000000 10000018 00000000 10000001 00000000\n \u003c7\u003e [416.493106] hangcheck [0060] 10000018 00000000 10000001 00000000 10000018 00000000 10000001 00000000\n \u003c7\u003e [416.493111] hangcheck *\n \u003c7\u003e [416.493117] hangcheck [00a0] 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000001\n \u003c7\u003e [416.493123] hangcheck [00c0] 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000\n \u003c7\u003e [416.493127] hangcheck *\n \u003c7\u003e [416.493132] hangcheck Idle? no\n \u003c6\u003e [416.512124] i915 0000:00:02.0: GPU HANG: ecode 11:0:0x00000000, hang on rcs0\n \u003c6\u003e [416.512205] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.\n \u003c6\u003e [416.512207] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -\u003e DRM/Intel\n \u003c6\u003e [416.512208] [drm] drm/i915 developers can then reassign to the right component if it\u0027s not a kernel issue.\n \u003c6\u003e [416.512210] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it.\n \u003c6\u003e [416.512212] [drm] GPU crash dump saved to /sys/class/drm/card0/error\n \u003c5\u003e [416.513602] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0\n \u003c7\u003e [424.489258] hangcheck rcs0\n \u003c7\u003e [424.489263] hangcheck Awake? 1\n \u003c7\u003e [424.489267] hangcheck Hangcheck: 5954 ms ago\n \u003c7\u003e [424.489271] hangcheck Reset count: 1 (global 0)\n \u003c7\u003e [424.489274] hangcheck Requests:\n \u003c7\u003e [424.490128] hangcheck RING_START: 0x00000000\n \u003c7\u003e [424.490870] hangcheck RING_HEAD: 0x00000000\n \u003c7\u003e [424.490877] hangcheck RING_TAIL: 0x00000000\n \u003c7\u003e [424.490887] hangcheck RING_CTL: 0x00000000\n \u003c7\u003e [424.490897] hangcheck RING_MODE: 0x00000200 [idle]\n \u003c7\u003e [424.490904] hangcheck RING_IMR: 00000000\n \u003c7\u003e [424.490917] hangcheck ACTHD: 0x00000000_00000000\n \u003c7\u003e [424.490930] hangcheck BBADDR: 0x00000000_00000000\n \u003c7\u003e [424.490943] hangcheck DMA_FADDR: 0x00000000_00000000\n \u003c7\u003e [424.490950] hangcheck IPEIR: 0x00000000\n \u003c7\u003e [424.490956] hangcheck IPEHR: 0x00000000\n \u003c7\u003e [424.490968] hangcheck Execlist status: 0x00000001 00000000, entries 12\n \u003c7\u003e [424.490972] hangcheck Execlist CSB read 11, write 11, tasklet queued? no (enabled)\n \u003c7\u003e [424.490983] hangcheck Pending[0] ring:{start:0017b000, hwsp:fedf9000, seqno:00016fd6}, rq: 20ffa:16fd6!+ prio\u003d-4094 @ 16305ms: signaled\n \u003c7\u003e [424.490989] hangcheck Queue priority hint: -4093\n \u003c7\u003e [424.490996] hangcheck Q 20ffa:16fd8- prio\u003d-4093 @ 16305ms: [i915]\n \u003c7\u003e [424.491001] hangcheck Q 20ffa:16fda prio\u003d-4094 @ 16305ms: [i915]\n \u003c7\u003e [424.491006] hangcheck Q 20ffa:16fdc prio\u003d-4094 @ 16305ms: [i915]\n \u003c7\u003e [424.491011] hangcheck Q 20ffa:16fde prio\u003d-4094 @ 16305ms: [i915]\n \u003c7\u003e [424.491016] hangcheck Q 20ffa:16fe0 prio\u003d-4094 @ 16305ms: [i915]\n \u003c7\u003e [424.491022] hangcheck Q 20ffa:16fe2 prio\u003d-4094 @ 16305ms: [i915]\n \u003c7\u003e [424.491048] hangcheck Q 20ffa:16fe4 prio\u003d-4094 @ 16305ms: [i915]\n \u003c7\u003e [424.491057] hangcheck ...skipping 21 queued requests...\n \u003c7\u003e [424.491063] hangcheck Q 20ffa:17010 prio\u003d-4094 @ 16305ms: [i915]\n \u003c7\u003e [424.491095] hangcheck HWSP:\n \u003c7\u003e [424.491102] hangcheck [0000] 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000\n \u003c7\u003e [424.491106] hangcheck *\n \u003c7\u003e [424.491113] hangcheck [0040] 10008002 00000000 10000018 00000000 10000018 00000000 10000001 00000000\n \u003c7\u003e [424.491118] hangcheck [0060] 10000018 00000000 10000001 00000000 10000018 00000000 10000001 00000000\n \u003c7\u003e [424.491122] hangcheck *\n \u003c7\u003e [424.491127] hangcheck [00a0] 00000000 00000000 00000000 00000000 00000000 00000000 00000000 0000000b\n \u003c7\u003e [424.491133] hangcheck [00c0] 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000\n \u003c7\u003e [424.491136] hangcheck *\n \u003c7\u003e [424.491141] hangcheck Idle? no\n \u003c5\u003e [424.491834] i915 0000:00:02.0: Resetting rcs0 for hang on rcs0\n \n Where not having cleared the pending array on reset, it persists\n indefinitely.\n \n Fixes: fff8102aaed5 (\"drm/i915/execlists: Process interrupted context on reset\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Andi Shyti \u003candi.shyti@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730133035.1977-2-chris@chris-wilson.co.uk\n\ncommit 49281a222ac42724b94f6c874f43a284ec58d37e\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Thu Aug 1 00:28:26 2019 +0200\n\n gpio: of: Fix hard-assigned valid_mask for OF case\n \n The recent refactoring to break out OF code to its own file\n contained a bug letting the need_valid_mask\n be overridden by the need of the device tree range check,\n and if there were no ranges, but device tree was active\n and the reserved GPIO used in another way, things likely\n crash.\n \n Cc: Mark Brown \u003cbroonie@kernel.org\u003e\n Cc: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Cc: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Fixes: f626d6dfb709 (\"gpio: of: Break out OF-only code\")\n Reported-by: Mark Brown \u003cbroonie@kernel.org\u003e\n Suggested-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit dc607f6bbafd2d8bf1d518c346560ba9577e402f\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Jul 6 18:47:22 2019 +0200\n\n mfd: aat2870: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Cc: Lee Jones \u003clee.jones@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190706164722.18766-3-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b11f75d032d8706bd39390a63107828985a61de3\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Jul 6 18:47:21 2019 +0200\n\n mfd: ab8500: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Cc: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Cc: Lee Jones \u003clee.jones@linaro.org\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Link: https://lore.kernel.org/r/20190706164722.18766-2-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 45640a383adc83331250f6fc7bf7e2c04f3e7879\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Sat Jul 6 18:47:20 2019 +0200\n\n mfd: ab3100: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Cc: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Cc: Lee Jones \u003clee.jones@linaro.org\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Link: https://lore.kernel.org/r/20190706164722.18766-1-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit cbb19cb1eef050b193c823502bf920097c0f0459\nAuthor: Corey Minyard \u003ccminyard@mvista.com\u003e\nDate: Wed Jul 31 15:33:36 2019 -0500\n\n ipmi_si: Convert timespec64 to timespec\n \n There is no need for timespec64, and it will cause issues in the\n future with i386 and 64-bit division not being available.\n \n Signed-off-by: Corey Minyard \u003ccminyard@mvista.com\u003e\n\ncommit 3247b272048ffefc12c7dcfa3169bd03047a49bc\nAuthor: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\nDate: Tue Jul 30 15:20:41 2019 +0300\n\n net: bridge: mcast: add delete due to fast-leave mdb flag\n \n In user-space there\u0027s no way to distinguish why an mdb entry was deleted\n and that is a problem for daemons which would like to keep the mdb in\n sync with remote ends (e.g. mlag) but would also like to converge faster.\n In almost all cases we\u0027d like to age-out the remote entry for performance\n and convergence reasons except when fast-leave is enabled. In that case we\n want explicit immediate remote delete, thus add mdb flag which is set only\n when the entry is being deleted due to fast-leave.\n \n Signed-off-by: Nikolay Aleksandrov \u003cnikolay@cumulusnetworks.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 0eba31ef5c8913adfd103c45c32d4856f1aa85cc\nAuthor: Lucas Bates \u003clucasb@mojatatu.com\u003e\nDate: Mon Jul 29 19:18:12 2019 -0400\n\n tc-testing: Clarify the use of tdc\u0027s -d option\n \n The -d command line argument to tdc requires the name of a physical device\n on the system where the tests will be run. If -d has not been used, tdc\n will skip tests that require a physical device.\n \n This patch is intended to better document what the -d option does and how\n it is used.\n \n Signed-off-by: Lucas Bates \u003clucasb@mojatatu.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 3015b8fcb60d448a9521b49359e068d021aaec97\nAuthor: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\nDate: Wed Jun 26 02:20:27 2019 -0700\n\n ice: Bump version number\n \n Update driver version to 0.7.5\n \n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit b67f25d76e9f5e73024b8782acb47f7e7bd339b8\nAuthor: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\nDate: Wed Jun 26 02:20:26 2019 -0700\n\n ice: Remove flag to track VF interrupt status\n \n As a result of refactoring of VF VSIs interrupts code, there is no\n need to track its configuration status again with ICE_VF_STATE_CFG_INTR\n flag - In fact, it is not being checked anywhere in the code right now, so\n this patch removes the dead code as applicable to the flag.\n \n Signed-off-by: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit ba880734ba9c24eca589b3f734d38309568301b2\nAuthor: Brett Creeley \u003cbrett.creeley@intel.com\u003e\nDate: Wed Jun 26 02:20:25 2019 -0700\n\n ice: Remove unnecessary flag ICE_FLAG_MSIX_ENA\n \n This flag is not needed and is called every time we re-enable interrupts\n in the hotpath so remove it. Also remove ice_vsi_req_irq() because it\n was a wrapper function for ice_vsi_req_irq_msix() whose sole purpose was\n checking the ICE_FLAG_MSIX_ENA flag.\n \n Signed-off-by: Brett Creeley \u003cbrett.creeley@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 992149446353bc22dff8edf231f2a5d8c5a3bd50\nAuthor: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\nDate: Wed Jun 26 02:20:24 2019 -0700\n\n ice: Don\u0027t return error for disabling LAN Tx queue that does exist\n \n Since Tx rings are being managed by FW/NVM, Tx rings might have not been\n set up or driver had already wiped them off - In that case, call to\n disable LAN Tx queue is being returned as not in existence. This patch\n makes sure we don\u0027t return unnecessary error for such scenario.\n \n Signed-off-by: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit a1e9968593234c179fcb3e71875897ae585c8362\nAuthor: Brett Creeley \u003cbrett.creeley@intel.com\u003e\nDate: Wed Jun 26 02:20:23 2019 -0700\n\n ice: Remove duplicate code in ice_alloc_rx_bufs\n \n Currently if the call to ice_alloc_mapped_page() fails we jump to the\n no_buf label, possibly call ice_release_rx_desc(), and return true\n indicating that there is more work to do. In the success case we just\n fall out of the while loop, possibly call ice_alloc_mapped_page(), and\n return false saying we exhausted cleaned_count. This flow can be\n improved by breaking if ice_alloc_mapped_page() fails and then the flow\n outside of the while loop is the same for the failure and success case.\n \n Signed-off-by: Brett Creeley \u003cbrett.creeley@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 56923ab66467c2edbd1ff97240ff14805e0b1cdc\nAuthor: Brett Creeley \u003cbrett.creeley@intel.com\u003e\nDate: Wed Jun 26 02:20:22 2019 -0700\n\n ice: Add stats for Rx drops at the port level\n \n Currently we are not reporting dropped counts at the port level to\n ethtool or netlink. This was found when debugging Rx dropped issues\n and the total packets sent did not equal the total packets received\n minus the rx_dropped, which was very confusing. To determine dropped\n counts at the port level we need to read the PRTRPB_RDPC register.\n To fix reporting we will store the dropped counts in the PF\u0027s\n rx_discards. This will be reported to netlink by storing it in the\n PF VSI\u0027s rx_missed_errors signaling that the receiver missed the\n packet. Also, we will report this to ethtool in the rx_dropped.nic\n field.\n \n Signed-off-by: Brett Creeley \u003cbrett.creeley@intel.com\u003e\n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 66b29e7a884e53e7e32ffb893d9897b299cd5cdf\nAuthor: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\nDate: Wed Jun 26 02:20:21 2019 -0700\n\n ice: Update number of VF queue before setting VSI resources\n \n In case there is a request from a VF to change its number of queues, and\n the request was successful, we need to update number of queues\n configured on the VF before updating corresponding VSI for that VF,\n especially LAN Tx queue tree and TC update, otherwise, we would continued\n to use old value of vf-\u003enum_vf_qs for allocated Tx/Rx queues...\n \n Signed-off-by: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit d5a46359171a00539179aefa04115c9c30002617\nAuthor: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\nDate: Wed Jun 26 02:20:20 2019 -0700\n\n ice: Set up Tx scheduling tree based on alloc VSI Tx queues\n \n This patch uses allocated number of Tx queues per VSI to set up its\n scheduling tree instead of using total number of available Tx queues.\n Only PF VSIs have total number of allocated Tx queues equal to number\n of available Tx queues, other VSIs have different number of queues\n configured.\n \n Signed-off-by: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit cb7db35641c9a508247bdcd73831c855c8758cd3\nAuthor: Brett Creeley \u003cbrett.creeley@intel.com\u003e\nDate: Wed Jun 26 02:20:19 2019 -0700\n\n ice: Only bump Rx tail and release buffers once per napi_poll\n \n Currently we bump the Rx tail and release/give buffers to hardware every\n 16 descriptors. This causes us to bump Rx tail up to 4 times per\n napi_poll call. Also we are always bumping tail on an odd index and this\n is a problem because hardware ignores the lower 3 bits in the QRX_TAIL\n register. This is making it so hardware sees tail bumps only every 8\n descriptors. Instead lets only bump Rx tail once per napi_poll if\n the value aligns with hardware\u0027s expectations of the lower 3 bits being\n cleared. Also only release/give Rx buffers once per napi_poll call.\n \n Signed-off-by: Brett Creeley \u003cbrett.creeley@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 9195c3e8d58a258344a33e0e05359564e8f8e538\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Wed Jul 31 17:27:56 2019 -0300\n\n docs: fs: porting.rst: fix a broken reference to another doc\n \n With all those document shifts, references to documents get\n broken.\n \n Fix one such occurrence at porting.rst.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit e9bb627561535dd584b43a8c0afe93a67bc6a2c5\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Wed Jul 31 17:08:53 2019 -0300\n\n docs: w1: convert to ReST and add to the kAPI group of docs\n \n The 1wire documentation was written with w1 developers in\n mind, so, it makes sense to add it together with the driver-api\n set.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit f139291c713069b5fa826ff509190efb5df83860\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Wed Jul 31 17:08:51 2019 -0300\n\n docs: fs: cifs: convert to ReST and add to admin-guide book\n \n The filenames for cifs documentation is not using the same\n convention as almost all Kernel documents is using. So,\n rename them to a more appropriate name. Then, manually convert\n the documentation files for CIFS to ReST.\n \n By doing a manual conversion, we can preserve the original\n author\u0027s style, while making it to look more like the other\n Kernel documents.\n \n Most of the conversion here is trivial. The most complex one was\n the README file (which was renamed to usage.rst).\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 60f2c82ed20bde57c362e66f796cf9e0e38a6dbb\nAuthor: Joonwon Kang \u003ckjw1627@gmail.com\u003e\nDate: Sun Jul 28 00:58:41 2019 +0900\n\n randstruct: Check member structs in is_pure_ops_struct()\n \n While no uses in the kernel triggered this case, it was possible to have\n a false negative where a struct contains other structs which contain only\n function pointers because of unreachable code in is_pure_ops_struct().\n \n Signed-off-by: Joonwon Kang \u003ckjw1627@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190727155841.GA13586@host\n Fixes: 313dd1b62921 (\"gcc-plugins: Add the randstruct plugin\")\n Cc: stable@vger.kernel.org\n Signed-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\n\ncommit 9cdd273e29f3b901712ec3c298b1d506861f48e3\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Wed Jul 31 17:08:50 2019 -0300\n\n spi: docs: convert to ReST and add it to the kABI bookset\n \n While there\u0027s one file there with briefily describes the uAPI,\n the documentation was written just like most subsystems: focused\n on kernel developers. So, add it together with driver-api books.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Acked-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e # for iio\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit d2fd3732e4acd6ab72960e7f8e00845438803f3e\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Wed Jul 31 17:08:49 2019 -0300\n\n docs: writing-schema.md: convert from markdown to ReST\n \n The documentation standard is ReST and not markdown.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Acked-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit aa95b4a960ab53234863d4ae0d1a09ad74253104\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Wed Jul 31 17:08:48 2019 -0300\n\n docs: fix a couple of new broken references\n \n Those are due to recent changes. Most of the issues\n can be automatically fixed with:\n \n $ ./scripts/documentation-file-ref-check --fix\n \n The only exception was the sound binding with required\n manual work.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 9475a77b57713e3763be90d0bdf4db5f42a5cb6b\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Fri Jul 26 14:15:12 2019 -0500\n\n drm/amdkfd: enable KFD support for navi14\n \n Same as navi10.\n \n Reviewed-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit dc4d716d4c0aa54392dfaca7a2e54b664edb5779\nAuthor: Dennis Li \u003cDennis.Li@amd.com\u003e\nDate: Tue Jul 23 18:23:44 2019 +0800\n\n drm/amdgpu: disable inject for failed subblocks of gfx\n \n some subblocks of gfx fail in inject test, disable them\n \n Signed-off-by: Dennis Li \u003cDennis.Li@amd.com\u003e\n Reviewed-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 83b0582c90576dd838bfbec84579f7e674cf59b2\nAuthor: Dennis Li \u003cDennis.Li@amd.com\u003e\nDate: Wed Jul 31 20:45:50 2019 +0800\n\n drm/amdgpu: support gfx ras error injection and err_cnt query\n \n check gfx error count in both ras querry function and\n ras interrupt handler.\n \n gfx ras is still disabled by default due to known stability\n issue found in gpu reset.\n \n Signed-off-by: Dennis Li \u003cDennis.Li@amd.com\u003e\n Reviewed-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2c960ea02f5863c910f9c839e8564ea01d868547\nAuthor: Dennis Li \u003cDennis.Li@amd.com\u003e\nDate: Wed Jul 31 20:42:15 2019 +0800\n\n drm/amdgpu: add RAS callback for gfx\n \n Add functions for RAS error inject and query error counter\n \n Signed-off-by: Dennis Li \u003cDennis.Li@amd.com\u003e\n Reviewed-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit dc23a08f0310babf8d2dfa4514edbc381a767410\nAuthor: Dennis Li \u003cDennis.Li@amd.com\u003e\nDate: Fri Jul 19 15:22:29 2019 +0800\n\n drm/amdgpu: add define for gfx ras subblock\n \n Signed-off-by: Dennis Li \u003cDennis.Li@amd.com\u003e\n Reviewed-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4bb6b8c758fe3a9225e3003a3fb35d16cac1baeb\nAuthor: Dennis Li \u003cDennis.Li@amd.com\u003e\nDate: Fri Jul 19 14:50:25 2019 +0800\n\n drm/amd/include: add define of TCP_EDC_CNT_NEW\n \n Signed-off-by: Dennis Li \u003cDennis.Li@amd.com\u003e\n Reviewed-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ca3f422f539551c8a20dc07aa1db87847de478d8\nAuthor: Dennis Li \u003cDennis.Li@amd.com\u003e\nDate: Fri Jul 19 14:42:49 2019 +0800\n\n drm/amd/include: add bitfield define for EDC registers\n \n Add EDC registers to support VEGA20 RAS\n \n Signed-off-by: Dennis Li \u003cDennis.Li@amd.com\u003e\n Reviewed-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7cdc2ee3004d9de1247fb6b66afa9f7429cbef9a\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Wed Jul 24 11:19:56 2019 +0800\n\n drm/amdgpu: remove ras_reserve_vram in ras injection\n \n error injection address is not in gpu address space\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Dennis Li \u003cdennis.li@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e10634938b66cff736f5d9b91f80815ef48645d3\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Tue Jul 23 13:07:24 2019 +0800\n\n drm/amdgpu: add check for ras error type\n \n only ue and ce errors are supported\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Dennis Li \u003cdennis.li@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 81e02619e9589474e815b78cbd0e098de6f0c05b\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Mon Jul 22 20:33:39 2019 +0800\n\n drm/amdgpu: update interrupt callback for all ras clients\n \n add err_data parameter in interrupt cb for ras clients\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Dennis Li \u003cdennis.li@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit cf04dfd0e91f11ac83d369f809ada8147aeaf3fe\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Mon Jul 22 20:27:25 2019 +0800\n\n drm/amdgpu: allow ras interrupt callback to return error data\n \n add error data as parameter for ras interrupt cb and process it\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Dennis Li \u003cdennis.li@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8c94810357fad9d583e37785534a8caec558bb24\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Wed Jul 24 21:43:45 2019 +0800\n\n drm/amdgpu: query umc ras error address\n \n query umc ras error address, translate it to gpu 4k page view\n and save it.\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Dennis Li \u003cdennis.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c2742aef4d17cca71346dc9327eef5840878a7d7\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Mon Jul 22 18:30:59 2019 +0800\n\n drm/amdgpu: add structures for umc error address translation\n \n add related registers, callback function and channel index table\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6f102dba808845836426dd08f1e5b19e2f70bec9\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Mon Jul 22 19:20:29 2019 +0800\n\n drm/amdgpu: add support for recording ras error address\n \n more than one error address may be recorded in one query\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Dennis Li \u003cdennis.li@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f1ed4afa130291be918e4b65a3759108d8fc868b\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Tue Jul 23 12:25:16 2019 +0800\n\n drm/amdgpu: update algorithm of umc uncorrectable error counting\n \n remove the check of ErrorCodeExt\n \n v2: refine the if condition for ue counting\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Dennis Li \u003cdennis.li@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 045c02165397c6c2c01ca5b8f68a9b642f4d244f\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Tue Jul 23 12:18:39 2019 +0800\n\n drm/amdgpu: switch to amdgpu_umc structure\n \n create new amdgpu_umc structure to for more umc\n settings in future and switch to the new structure\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Dennis Li \u003cdennis.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5bbfb64a177f36d3d208e39c61ce6df3968df4d4\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Tue Jul 23 11:57:15 2019 +0800\n\n drm/amdgpu: use 64bit operation macros for umc\n \n replace some 32bit macros with 64bit operations to simplify code\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Dennis Li \u003cdennis.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4fa1c6a679bb0d0bb92cf5bf9b7049ef98552848\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Wed Jul 24 15:13:27 2019 +0800\n\n drm/amdgpu: add RREG64/WREG64(_PCIE) operations\n \n add 64 bits register access functions\n \n v2: implement 64 bit functions in low level\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Dennis Li \u003cdennis.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 05a58345db4e075eb709d89ea3f43610b93bfb89\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Wed Jul 31 20:28:13 2019 +0800\n\n drm/amdgpu: add ras error count after each query (v2)\n \n v1: increase ras ce/ue error count\n v2: log the number of correctable and uncorrectable errors\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Dennis Li \u003cdennis.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 939e2258ce4c2e9f5bc3a4f42ca7290a015306c2\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Wed Jul 17 21:49:53 2019 +0800\n\n drm/amdgpu: querry umc error count\n \n check umc error count in both ras querry function and\n ras interrupt handler\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Dennis Li \u003cdennis.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5b6b35aaacca95616873db25ad521c6bb5ad39f6\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Wed Jul 17 21:47:44 2019 +0800\n\n drm/amdgpu: init umc v6_1 functions for vega20\n \n init umc callback function for vega20 in sw early init phase\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Dennis Li \u003cdennis.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9884c2b1c38c33a9152f5aff162473f348fe3acd\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Wed Jul 31 20:23:01 2019 +0800\n\n drm/amdgpu: add umc v6_1 query error count support\n \n Implement umc query_ras_error_count function to support querry\n both correctable and uncorrectable error\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Reviewed-by: Dennis Li \u003cdennis.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 03c9963f47a9efe204983fb0ea022814f8ce0084\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Wed Jul 24 14:36:49 2019 +0800\n\n drm/amdgpu: add umc v6_1_1 IP headers\n \n the change introduces IP headers for unified memory controller (umc)\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Dennis Li \u003cdennis.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 245219a66085332a30e4653db3542ea5654ff762\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Wed Jul 24 14:13:53 2019 +0800\n\n drm/amdgpu: add rsmu v_0_0_2 ip headers\n \n remote smu (rsmu) is a sub-block used as ip register interface,\n error handling, reset generation.etc\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Dennis Li \u003cdennis.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9e585a523baa7cdd7d47f95336bdb47bd64f4d7b\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Tue Jul 23 19:42:03 2019 +0800\n\n drm/amdgpu: add amdgpu_umc_functions structure\n \n This is common structure as UMC callback function\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Dennis Li \u003cdennis.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6501a771708d7fa66bc87b13efcc8ab837f1e2ed\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Wed Jul 17 17:52:28 2019 +0800\n\n drm/amdgpu: init RSMU and UMC ip base address for vega20\n \n the driver needs to program RSMU and UMC registers to\n support vega20 RAS feature\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Dennis Li \u003cdennis.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7af25d5b7ea70a7ac669a61c5d9317b2be27c2b9\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Wed Jul 17 17:34:46 2019 +0800\n\n drm/amdgpu: move some ras data structure to amdgpu_ras.h\n \n These are common structures that can be included by IP specific\n source files\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Dennis Li \u003cdennis.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2ef5a7f1482c21579d556f31a605a7ca32f251e7\nAuthor: Jarkko Sakkinen \u003cjarkko.sakkinen@linux.intel.com\u003e\nDate: Fri Jul 12 18:44:32 2019 +0300\n\n tpm: Document UEFI event log quirks\n \n There are some weird quirks when it comes to UEFI event log. Provide a\n brief introduction to TPM event log mechanism and describe the quirks\n and how they can be sorted out.\n \n Signed-off-by: Jarkko Sakkinen \u003cjarkko.sakkinen@linux.intel.com\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit fa1884f9d8016bbd94e5b6c7c9abb4bd3b223bad\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Wed Jul 31 10:47:26 2019 -0500\n\n drm/amdgpu: drop drmP.h from vcn_v2_5.c\n \n Unused.\n \n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9a2ffeb525a1c6f4e25afe0598a10ec43731cb6e\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Wed Jul 31 10:45:52 2019 -0500\n\n drm/amdgpu: drop drmP.h from vcn_v2_0.c\n \n And fix the fallout.\n \n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 75589f496d6d3f13e8e2da6c32a93b6b7902d4ad\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Wed Jul 31 10:43:40 2019 -0500\n\n drm/amdgpu: drop drmP.h from sdma_v5_0.c\n \n And fix the fallout.\n \n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e9eea90247fef9070e72b6a010535681c825c117\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Wed Jul 31 10:39:40 2019 -0500\n\n drm/amdgpu: drop drmP.h from nv.c\n \n And fix up the fallout.\n \n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b23b2e9e495799e7889b2cb64ea505a27553bdc3\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Wed Jul 31 10:34:39 2019 -0500\n\n drm/amdgpu: drop drmP.h from navi10_ih.c\n \n And fix the fallout.\n \n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 0a069bbe138797efb3669176888eefe0742690b6\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Wed Jul 31 10:31:44 2019 -0500\n\n drm/amdgpu: drop drmP.h in gfx_v10_0.c\n \n And fix the fallout.\n \n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 3b90f6ecdfa5fb5efd92c557e18266afc3eb93d6\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Wed Jul 31 10:27:57 2019 -0500\n\n drm/amdgpu: drop drmP.h from amdgpu_amdkfd_gfx_v10.c\n \n Unused.\n \n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 32978d8cfd3c0df8b909f9d4020718613819fe81\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Wed Jul 31 10:26:39 2019 -0500\n\n drm/amdgpu: drop drmP.h in amdgpu_amdkfd_arcturus.c\n \n Unused.\n \n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e15d5a53ea3239e537d3a0c4bdfce8a4ea4e0da6\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 09:51:34 2019 -0300\n\n docs: net: convert two README files to ReST format\n \n There are two README files there with doesn\u0027t have a .txt\n extension nor are at ReST format.\n \n In order to help with the docs conversion to ReST, rename those\n and manually convert them to ReST format.\n \n As there are lot more to be done for networking to be part of\n the documentation body, for now mark those two files with\n :orphan:, in order to supress a build warning.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 47359e5734f1df0301ee3f0c46da9954f15b7bd9\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 09:51:33 2019 -0300\n\n docs: nios2: add it to the main Documentation body\n \n Rename and add the nios2 documentation to the documentation\n body.\n \n The nios2 document is already on an ReST compatible format.\n All it needs is that the title of the document to be promoted\n one level.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit bfc8a222d3bca4c09a63de8b321bf7f825fa59ff\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 09:51:32 2019 -0300\n\n docs: hwmon: pxe1610: convert to ReST format and add to the index\n \n This document was recently introduced. Convert it to ReST\n just like the other hwmon documents, adding it to the hwmon index.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 80b1505893b60608e1525fc63caf16fd128f1b4c\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 09:51:31 2019 -0300\n\n docs: mips: add to the documentation body as ReST\n \n Manually convert the AU1xxx_IDE.README file to ReST and add\n to a MIPS book as part of the main documentation body.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Acked-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit ff497db295c8939c4badf9cbaf5f0679af8ada0a\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 09:51:30 2019 -0300\n\n docs: wimax: convert to ReST and add to admin-guide\n \n Manually convert wimax documentation to ReST and add theit\n to the Kernel doc body, inside the admin-guide.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 9c970ab54c336340429fcd2570f3c577610078e2\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 09:51:29 2019 -0300\n\n docs: index.rst: don\u0027t use genindex for pdf output\n \n The genindex logic is meant to be used only for html output, as\n pdf build has its own way to generate indexes.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Acked-by: Vinod Koul \u003cvkoul@kernel.org\u003e # dmaengine and soundwire\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 25b532cec59ee119bf1c4081fe06fac43722f61f\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 09:51:28 2019 -0300\n\n docs: fs: convert porting to ReST\n \n This file has its own proper style, except that, after a while,\n the coding style gets violated and whitespaces are placed on\n different ways.\n \n As Sphinx and ReST are very sentitive to whitespace differences,\n I had to opt if each entry after required/mandatory/... fields\n should start with zero spaces or with a tab. I opted to start them\n all from the zero position, in order to avoid needing to break lines\n with more than 80 columns, with would make harder for review.\n \n Most of the other changes at porting.rst were made to use an unified\n notation with works nice as a text file while also produce a good html\n output after being parsed.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit ec23eb54fbc7a07405d416d77e8115e575ce3adc\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 09:51:27 2019 -0300\n\n docs: fs: convert docs without extension to ReST\n \n There are 3 remaining files without an extension inside the fs docs\n dir.\n \n Manually convert them to ReST.\n \n In the case of the nfs/exporting.rst file, as the nfs docs\n aren\u0027t ported yet, I opted to convert and add a :orphan: there,\n with should be removed when it gets added into a nfs-specific\n part of the fs documentation.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 5a5e045bb3b839405e3a58b02a3333d33812214c\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 09:51:25 2019 -0300\n\n docs: isdn: convert to ReST and add to kAPI bookset\n \n The ISDN documentation is a mix of admin guide, uAPI and kAPI.\n \n Ideally, it should be split. Yet, not sure if it would worth\n the troble. Anyway, we have the same kind of mix on several\n drivers specific documentation. So, just like the others, keep\n the directory at the root Documentation/ tree, just adding a\n pointer to it at the kAPI section, as the documentation was\n written with the Kernel developers in mind.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 32fc3cd8ba2375b0ee385a42ba2a1aad5547816e\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 09:51:24 2019 -0300\n\n docs: openrisc: convert to ReST and add to documentation body\n \n Manually convert the two openRisc documents to ReST, adding them\n to the Linux documentation body.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Acked-by: Stafford Horne \u003cshorne@gmail.com\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit e77e9187ae1caf2d83dd5e7f0c1466254b644a4c\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 09:51:23 2019 -0300\n\n docs: parisc: convert to ReST and add to documentation body\n \n Manually convert the two PA-RISC documents to ReST, adding them\n to the Linux documentation body.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 6d6486a0c59759681e75d1a2bd6684c501fcbd0e\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 09:51:22 2019 -0300\n\n docs: README.buddha: convert to ReST and add to m68k book\n \n Adjust the file for it to be properly parsed by Sphinx, adding\n it to the index of the book it belongs.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 76b5a6e8427159ad2b3b8764ebd6f3f5213be97e\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 09:51:21 2019 -0300\n\n docs: admin-guide: add auxdisplay files to it after conversion to ReST\n \n Those two files describe userspace-faced information. While part of\n it might fit on uAPI, it sounds to me that the admin guide is the\n best place for them.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Acked-by: Miguel Ojeda \u003cmiguel.ojeda.sandonis@gmail.com\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 1ec779b9fabcdbfaa06971b5c2a4d9a6e4b45b3a\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 09:51:20 2019 -0300\n\n docs: packing: move it to core-api book and adjust markups\n \n The packing.txt file was misplaced, as docs should be part of\n a documentation book, and not at the root dir.\n \n So, move it to the core-api directory and add to its index.\n \n Also, ensure that the file will be properly parsed and the bitmap\n ascii artwork will use a monotonic font.\n \n Fixes: 554aae35007e (\"lib: Add support for generic packing operations\")\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Reviewed-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Tested-by: Vladimir Oltean \u003colteanv@gmail.com\u003e\n Reviewed-by: Mike Rapoport \u003crppt@linux.ibm.com\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit f6ae22d64433fd8e08654adad7966299da931bb9\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 09:51:19 2019 -0300\n\n docs: ipmb: place it at driver-api and convert to ReST\n \n No new doc should be added at the main Documentation/ directory.\n \n Instead, new docs should be added as ReST files, within the\n Kernel documentation body.\n \n Fixes: 51bd6f291583 (\"Add support for IPMB driver\")\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit ccf988b66d697efcd0ceccc2398e0d9b909cd17c\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 09:51:16 2019 -0300\n\n docs: i2c: convert to ReST and add to driver-api bookset\n \n Convert each file at I2C subsystem, renaming them to .rst and\n adding to the driver-api book.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Acked-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n Acked-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Acked-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 09f4c750a8c7d1fc0b7bb3a7aa1de55de897a375\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 09:51:14 2019 -0300\n\n docs: ubifs-authentication.md: convert to ReST\n \n The documentation standard is ReST and not markdown.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Acked-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit eaf7b46083a7e341a23ab3d6042e0ccc115b0914\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 09:51:12 2019 -0300\n\n docs: thermal: add it to the driver API\n \n The file contents mostly describes driver internals.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit fe13225fdc3f7b79e2921869a13386f48b30bf79\nAuthor: Phong Tran \u003ctranmanphong@gmail.com\u003e\nDate: Thu Jul 11 23:52:01 2019 +0700\n\n Documentation: coresight: convert txt to rst\n \n This changes from plain text to reStructuredText as suggestion\n in doc-guide [1]\n \n [1] https://www.kernel.org/doc/html/latest/doc-guide/sphinx.html\n \n Some adaptations such as: literal block, ``inline literal`` and\n alignment text,...\n \n Signed-off-by: Phong Tran \u003ctranmanphong@gmail.com\u003e\n Reviewed-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Acked-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 34d5f4f269a21c7b5d64b4db0391d36511f1ac1a\nAuthor: Shobhit Kukreti \u003cshobhitkukreti@gmail.com\u003e\nDate: Wed Jul 10 08:31:23 2019 -0700\n\n Documentation: filesystems: Convert ufs.txt to reStructuredText format\n \n This converts the plain text documentation of ufs.txt to\n reStructuredText format. Added to documentation build process\n and verified with make htmldocs\n \n Signed-off-by: Shobhit Kukreti \u003cshobhitkukreti@gmail.com\u003e\n Reviewed-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit ac841c4e457c1fae6f661108f811b554c8581976\nAuthor: Shobhit Kukreti \u003cshobhitkukreti@gmail.com\u003e\nDate: Wed Jul 10 08:29:01 2019 -0700\n\n Documentation: filesystems: Convert jfs.txt to\n \n This converts the plain text documentation of jfs.txt to reStructuredText\n format. Added to documentation build process and verified with\n make htmldocs\n \n Signed-off-by: Shobhit Kukreti \u003cshobhitkukreti@gmail.com\u003e\n Reviewed-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit e226b4f0e04f4cd5396041661a27eae5aa370bb3\nAuthor: Federico Vaga \u003cfederico.vaga@vaga.pv.it\u003e\nDate: Sat Jul 6 23:01:00 2019 +0200\n\n doc: email-clients miscellaneous fixes\n \n Fixed some style inconsistencies and remove old statement referring to\n kmail missing feature (saving email from the view window is possible).\n \n Signed-off-by: Federico Vaga \u003cfederico.vaga@vaga.pv.it\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit d129e3f42266a12b821868fc8c2407a4d80f0b18\nAuthor: Leon Romanovsky \u003cleonro@mellanox.com\u003e\nDate: Wed Jul 31 14:56:27 2019 +0300\n\n RDMA/mlx5: Remove DEBUG ODP code\n \n Delete DEBUG ODP dead code which is leftover from development\n stage and doesn\u0027t need to be part of the upstream kernel.\n \n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Link: https://lore.kernel.org/r/20190731115627.5433-1-leon@kernel.org\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 638b642f82bb8ee81fe00bdbb70b5ff0885df1a5\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 18:01:55 2019 -0300\n\n docs: riscv: convert boot-image-header.txt to ReST\n \n Convert this small file to ReST format by:\n - Using a proper markup for the document title;\n - marking a code block as such;\n - use tags for Author and date;\n - use tables for bit map fields.\n \n While here, fix a broken reference for a document with is\n planned but is not here yet.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Reviewed-by: Atish Patra \u003catish.patra@wdc.com\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 54bfe6feba0e9ead1172cf26381f7dce5df10e51\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 16:29:14 2019 -0300\n\n docs: zh_CN: howto.rst: fix a broken reference\n \n There\u0027s a broken reference there pointing to the long gone\n DocBook dir.\n \n While I don\u0027t read chinese, Google translator translates it\n to:\n \"The generated documentation will be placed in\n the Documentation/DocBook/ directory.\"\n \n Well, we know that the output will be Documentation/output\n dir. So, let\u0027s fix this one.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 23aa16489c06e6739c7c99e9fdccf723d2691a5d\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 16:29:13 2019 -0300\n\n docs: cgroup-v1/blkio-controller.rst: remove a CFQ left over\n \n changeset fb5772cbfe48 (\"blkio-controller.txt: Remove references to CFQ\")\n removed cgroup references to CFQ, but it kept one left. Get rid of\n it.\n \n Fixes: fb5772cbfe48 (\"blkio-controller.txt: Remove references to CFQ\")\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 7530c497592e04f5da69fe968c09622b489f2921\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 16:29:12 2019 -0300\n\n MAINTAINERS: fix a renamed DT reference\n \n Fix this rename:\n \n Documentation/devicetree/bindings/i2c/{i2c-mv64xxx.txt -\u003e marvell,mv64xxx-i2c.yaml}\n \n Fixes: f8bbde72ef44 (\"dt-bindings: i2c: mv64xxx: Add YAML schemas\")\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit b26bb0faf8f5d7fccb2e1827f33012177ac9d217\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 16:29:11 2019 -0300\n\n MAINTAINERS: fix reference to net phy ABI file\n \n The file sysfs-bus-mdio got removed in favor of sysfs-class-net-phydev,\n with contained a duplicated set of information.\n \n Fixes: a6cd0d2d493a (\"Documentation: net-sysfs: Remove duplicate PHY device documentation\")\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 370910efdb6ece4a391afbc047c39e2d1b61374c\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 16:29:10 2019 -0300\n\n MAINTAINERS: fix broken ref for ABI sysfs-bus-counter-ftm-quaddec\n \n There\u0027s a typo here:\n \n sysfs-bus-counter-ftm-quadddec -\u003e sysfs-bus-counter-ftm-quaddec\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 803deeaaea10a87b7d92bd2eccbc637711390a04\nAuthor: Federico Vaga \u003cfederico.vaga@vaga.pv.it\u003e\nDate: Mon Jul 29 23:58:56 2019 +0200\n\n doc:it_IT: align translation to mainline\n \n The patch translates the following patches in Italian:\n \n 1fb12b35e5ff kbuild: Raise the minimum required binutils version to 2.21\n 9c3c0c204814 isdn: remove isdn4linux\n \n Signed-off-by: Federico Vaga \u003cfederico.vaga@vaga.pv.it\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 074b326ee5e73c8e7a10d0eebbc6a1b56e1253d2\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Wed Jul 31 19:45:10 2019 +0800\n\n mailmap: add entry to connect my email addresses\n \n I\u0027ve used several email accounts to contribute codes, samsung\u0027s one\n is obsolete, so let me add entry to map them, in order to let people\n find me easily when they blame my codes.\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 257e26c6403c23714db69d6e21c58b0d92636c4e\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Wed Jul 31 11:02:11 2019 +0200\n\n docs: arm: Remove orphan sh-mobile directory\n \n This directory is empty, except for a .gitignore file, listing an\n executable file that can no longer be built since commit\n c6535e1e0361157e (\"Documentation: Remove ZBOOT MMC/SDHI utility and\n docs\").\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit c2052d6e242c33594db90751a361aca1065a054a\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Tue Jul 30 10:51:21 2019 -0700\n\n drm/i915/ehl: Ungate DDIC and DDID\n \n Specification states that DDI_CLK_SEL needs to be mapped to MG clock\n even if MG do not exist on EHL, this will ungate those DDIs.\n \n BSpec: 20845\n Cc: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Cc: Vivek Kasireddy \u003cvivek.kasireddy@intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Reviewed-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Tested-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730175121.16413-1-jose.souza@intel.com\n\ncommit c7aeb4d1b9bfdb4e07da1c77cb095f02e912a83f\nAuthor: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\nDate: Wed Jun 26 02:20:18 2019 -0700\n\n ice: Disable VFs until reset is completed\n \n This patch adds code to clear VFs enable status until reset is completed,\n and Tx/Rx rings are setup. Without this patch, the code flow request Tx\n queues to be disabled after reset, especially PFR - where VF VSI Tx rings\n have already been wiped off in the NVM and result to adminq error based on\n the call to disable Tx LAN queue in ice_reset_all_vfs function call.\n \n Signed-off-by: Akeem G Abodunrin \u003cakeem.g.abodunrin@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 6d5999467df03d1d7fd64ac761ffa20d00ce9e52\nAuthor: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\nDate: Wed Jun 26 02:20:17 2019 -0700\n\n ice: Do not configure port with no media\n \n The firmware reports an error when trying to configure a port with no\n media. Instead of always configuring the port, check for media before\n attempting to configure it. In the absence of media, turn off link and\n poll for media to become available before re-enabling link.\n \n Move ice_force_phys_link_state() up to avoid forward declaration.\n \n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 5c91ecfda5a8bfbc4697b35fe875cbc43a0f9100\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Wed Jun 26 02:20:16 2019 -0700\n\n ice: separate out control queue lock creation\n \n The ice_init_all_ctrlq and ice_shutdown_all_ctrlq functions create and\n destroy the locks used to protect the send and receive process of each\n control queue.\n \n This is problematic, as the driver may use these functions to shutdown\n and re-initialize the control queues at run time. For example, it may do\n this in response to a device reset.\n \n If the driver failed to recover from a reset, it might leave the control\n queues offline. In this case, the locks will no longer be initialized.\n A later call to ice_sq_send_cmd will then attempt to acquire a lock that\n has been destroyed.\n \n It is incorrect behavior to access a lock that has been destroyed.\n \n Indeed, ice_aq_send_cmd already tries to avoid accessing an offline\n control queue, but the check occurs inside the lock.\n \n The root of the problem is that the locks are destroyed at run time.\n \n Modify ice_init_all_ctrlq and ice_shutdown_all_ctrlq such that they no\n longer create or destroy the locks.\n \n Introduce new functions, ice_create_all_ctrlq and ice_destroy_all_ctrlq.\n Call these functions in ice_init_hw and ice_deinit_hw.\n \n Now, the control queue locks will remain valid for the life of the\n driver, and will not be destroyed until the driver unloads.\n \n This also allows removing a duplicate check of the sq.count and\n rq.count values when shutting down the controlqs. The ice_shutdown_ctrlq\n function already checks this value under the lock. Previously\n commit dec64ff10ed9 (\"ice: use [sr]q.count when checking if queue is\n initialized\") needed this check to happen outside the lock, because it\n prevented duplicate attempts at destroying the locks.\n \n The driver may now safely use ice_init_all_ctrlq and\n ice_shutdown_all_ctrlq while handling reset events, without causing the\n locks to be invalid.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit c31a5c25bb19c62d1cea69d3abcc7e0405bd4596\nAuthor: Brett Creeley \u003cbrett.creeley@intel.com\u003e\nDate: Wed Jun 26 02:20:15 2019 -0700\n\n ice: Always set prefena when configuring an Rx queue\n \n Currently we are always setting prefena to 0. This is causing the\n hardware to only fetch descriptors when there are none free in the cache\n for a received packet instead of prefetching when it has used the last\n descriptor regardless of incoming packets. Fix this by allowing the\n hardware to prefetch Rx descriptors.\n \n Signed-off-by: Brett Creeley \u003cbrett.creeley@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 17bc6d07212c8bc4521056a7f871d143192d385c\nAuthor: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\nDate: Wed Jun 26 02:20:14 2019 -0700\n\n ice: Move vector base setup to PF VSI\n \n When interrupt tracking was refactored, during rebuild, the call to\n ice_vsi_setup_vector_base() was inadvertently removed from the PF VSI\n instead of being removed from the VF VSI. During reset, the failure to\n properly setup the vector base generates a call trace. Correct this so\n that resets/rebuilds properly complete.\n \n Fixes: cbe66bfee6a0 (\"ice: Refactor interrupt tracking\")\n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 36517fd397f124acfa396e770468530136f4207d\nAuthor: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\nDate: Wed Jun 26 02:20:13 2019 -0700\n\n ice: track hardware stat registers past rollover\n \n Currently, ice_stat_update32 and ice_stat_update40 will limit the\n value of the software statistic to 32 or 40 bits wide, depending on\n which register is being read.\n \n This means that if a driver is running for a long time, the displayed\n software register values will roll over to zero at 40 bits or 32 bits.\n \n This occurs because the functions directly assign the difference between\n the previous value and current value of the hardware statistic.\n \n Instead, add this value to the current software statistic, and then\n update the previous value.\n \n In this way, each time ice_stat_update40 or ice_stat_update32 are\n called, they will increment the software tracking value by the\n difference of the hardware register from its last read. The software\n tracking value will correctly count up until it overflows a u64.\n \n The only requirement is that the ice_stat_update functions be called at\n least once each time the hardware register overflows.\n \n While we\u0027re fixing ice_stat_update40, modify it to use rd64 instead of\n two calls to rd32. Additionally, drop the now unnecessary hireg\n function parameter.\n \n Signed-off-by: Jacob Keller \u003cjacob.e.keller@intel.com\u003e\n Signed-off-by: Tony Nguyen \u003canthony.l.nguyen@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 5a056cd7ead2b72b00fea8a6819fb93eeb12e313\nAuthor: Paul Greenwalt \u003cpaul.greenwalt@intel.com\u003e\nDate: Wed Jun 26 02:20:12 2019 -0700\n\n ice: add lp_advertising flow control support\n \n Add support for reporting link partner advertising when\n ETHTOOL_GLINKSETTINGS defined. Get pause param reports the Tx/Rx\n pause configured, and then ethtool issues ETHTOOL_GSET ioctl and\n ice_get_settings_link_up reports the negotiated Tx/Rx pause. Negotiated\n pause frame report per IEEE 802.3-2005 table 288-3.\n \n $ ethtool --show-pause ens6f0\n Pause parameters for ens6f0:\n Autonegotiate: on\n RX: on\n TX: on\n RX negotiated: on\n TX negotiated: on\n \n $ ethtool ens6f0\n Settings for ens6f0:\n Supported ports: [ FIBRE ]\n Supported link modes: 25000baseCR/Full\n Supported pause frame use: Symmetric\n Supports auto-negotiation: Yes\n Supported FEC modes: None BaseR RS\n Advertised link modes: 25000baseCR/Full\n Advertised pause frame use: Symmetric Receive-only\n Advertised auto-negotiation: Yes\n Advertised FEC modes: None BaseR RS\n Link partner advertised link modes: Not reported\n Link partner advertised pause frame use: Symmetric\n Link partner advertised auto-negotiation: Yes\n Link partner advertised FEC modes: Not reported\n Speed: 25000Mb/s\n Duplex: Full\n Port: Direct Attach Copper\n PHYAD: 0\n Transceiver: internal\n Auto-negotiation: on\n Supports Wake-on: g\n Wake-on: g\n Current message level: 0x00000007 (7)\n drv probe link\n Link detected: yes\n \n When ETHTOOL_GLINKSETTINGS is not defined, get pause param reports the\n negotiated Tx/Rx pause.\n \n Signed-off-by: Paul Greenwalt \u003cpaul.greenwalt@intel.com\u003e\n Tested-by: Andrew Bowers \u003candrewx.bowers@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 09c7e8b21d67c3c78ab9701dbc0fb1e9f14a0ba5\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Fri Jul 26 23:19:44 2019 +0200\n\n x86/kvm: Use CONFIG_PREEMPTION\n \n CONFIG_PREEMPTION is selected by CONFIG_PREEMPT and by\n CONFIG_PREEMPT_RT. Both PREEMPT and PREEMPT_RT require the same\n functionality which today depends on CONFIG_PREEMPT.\n \n Switch the conditional for async pagefaults to use CONFIG_PREEMPTION.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\n Cc: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n Cc: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Steven Rostedt \u003crostedt@goodmis.org\u003e\n Link: http://lkml.kernel.org/r/20190726212124.789755413@linutronix.de\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit cb376c26971ff54f25980ec1f0ae2f06d6a69df0\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Fri Jul 26 23:19:43 2019 +0200\n\n x86/dumpstack: Indicate PREEMPT_RT in dumps\n \n Stack dumps print whether the kernel has preemption enabled or not. Extend\n it so a PREEMPT_RT enabled kernel can be identified.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\n Cc: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n Cc: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Steven Rostedt \u003crostedt@goodmis.org\u003e\n Link: http://lkml.kernel.org/r/20190726212124.699136351@linutronix.de\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 48593975aeee548f25e256c515fd1d1e3fb2cc20\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Fri Jul 26 23:19:42 2019 +0200\n\n x86: Use CONFIG_PREEMPTION\n \n CONFIG_PREEMPTION is selected by CONFIG_PREEMPT and by\n CONFIG_PREEMPT_RT. Both PREEMPT and PREEMPT_RT require the same\n functionality which today depends on CONFIG_PREEMPT.\n \n Switch the entry code, preempt and kprobes conditionals over to\n CONFIG_PREEMPTION.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\n Cc: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n Cc: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Steven Rostedt \u003crostedt@goodmis.org\u003e\n Link: http://lkml.kernel.org/r/20190726212124.608488448@linutronix.de\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 92616606368ee01f1163fcfc986116c810cd48ba\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Fri Jul 26 23:19:41 2019 +0200\n\n kprobes: Use CONFIG_PREEMPTION\n \n CONFIG_PREEMPTION is selected by CONFIG_PREEMPT and by\n CONFIG_PREEMPT_RT. Both PREEMPT and PREEMPT_RT require the same\n functionality which today depends on CONFIG_PREEMPT.\n \n Switch kprobes conditional over to CONFIG_PREEMPTION.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\n Cc: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n Cc: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Steven Rostedt \u003crostedt@goodmis.org\u003e\n Link: http://lkml.kernel.org/r/20190726212124.516286187@linutronix.de\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 30c937043b2db09ae3408f5534824f9ececdb581\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Fri Jul 26 23:19:40 2019 +0200\n\n tracing: Use CONFIG_PREEMPTION\n \n CONFIG_PREEMPTION is selected by CONFIG_PREEMPT and by\n CONFIG_PREEMPT_RT. Both PREEMPT and PREEMPT_RT require the same\n functionality which today depends on CONFIG_PREEMPT.\n \n Switch the conditionals in the tracer over to CONFIG_PREEMPTION.\n \n This is the first step to make the tracer work on RT. The other small\n tweaks are submitted separately.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\n Cc: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n Cc: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Steven Rostedt \u003crostedt@goodmis.org\u003e\n Link: http://lkml.kernel.org/r/20190726212124.409766323@linutronix.de\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 27972765bd0410fc2ef5e86a41de17c71440a2dd\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Fri Jul 26 23:19:39 2019 +0200\n\n locking/spinlocks: Use CONFIG_PREEMPTION\n \n CONFIG_PREEMPTION is selected by CONFIG_PREEMPT and by\n CONFIG_PREEMPT_RT. Both PREEMPT and PREEMPT_RT require the same\n functionality which today depends on CONFIG_PREEMPT.\n \n Adjust the comments in the locking code.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\n Cc: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n Cc: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Steven Rostedt \u003crostedt@goodmis.org\u003e\n Link: http://lkml.kernel.org/r/20190726212124.302995288@linutronix.de\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 01b1d88b09824bea1a75b0bac04dcf50d9893875\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Fri Jul 26 23:19:38 2019 +0200\n\n rcu: Use CONFIG_PREEMPTION\n \n CONFIG_PREEMPTION is selected by CONFIG_PREEMPT and by\n CONFIG_PREEMPT_RT. Both PREEMPT and PREEMPT_RT require the same\n functionality which today depends on CONFIG_PREEMPT.\n \n Switch the conditionals in RCU to use CONFIG_PREEMPTION.\n \n That\u0027s the first step towards RCU on RT. The further tweaks are work in\n progress. This neither touches the selftest bits which need a closer look\n by Paul.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\n Cc: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n Cc: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Steven Rostedt \u003crostedt@goodmis.org\u003e\n Link: http://lkml.kernel.org/r/20190726212124.210156346@linutronix.de\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit c1a280b68d4e6b6db4a65aa7865c22d8789ddf09\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Fri Jul 26 23:19:37 2019 +0200\n\n sched/preempt: Use CONFIG_PREEMPTION where appropriate\n \n CONFIG_PREEMPTION is selected by CONFIG_PREEMPT and by\n CONFIG_PREEMPT_RT. Both PREEMPT and PREEMPT_RT require the same\n functionality which today depends on CONFIG_PREEMPT.\n \n Switch the preemption code, scheduler and init task over to use\n CONFIG_PREEMPTION.\n \n That\u0027s the first step towards RT in that area. The more complex changes are\n coming separately.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Masami Hiramatsu \u003cmhiramat@kernel.org\u003e\n Cc: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n Cc: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Steven Rostedt \u003crostedt@goodmis.org\u003e\n Link: http://lkml.kernel.org/r/20190726212124.117528401@linutronix.de\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 56db1991690f076c2a7e3b2a226629cd10901690\nAuthor: Steve Magnani \u003csteve.magnani@digidescorp.com\u003e\nDate: Sun Jul 28 14:19:12 2019 -0500\n\n udf: prevent allocation beyond UDF partition\n \n The UDF bitmap allocation code assumes that a recorded\n Unallocated Space Bitmap is compliant with ECMA-167 4/13,\n which requires that pad bytes between the end of the bitmap\n and the end of a logical block are all zero.\n \n When a recorded bitmap does not comply with this requirement,\n for example one padded with FF to the block boundary instead\n of 00, the allocator may \"allocate\" blocks that are outside\n the UDF partition extent. This can result in UDF volume descriptors\n being overwritten by file data or by partition-level descriptors,\n and in extreme cases, even in scribbling on a subsequent disk partition.\n \n Add a check that the block selected by the allocator actually\n resides within the UDF partition extent.\n \n Signed-off-by: Steven J. Magnani \u003csteve@digidescorp.com\u003e\n \n Link: https://lore.kernel.org/r/1564341552-129750-1-git-send-email-steve@digidescorp.com\n Signed-off-by: Jan Kara \u003cjack@suse.cz\u003e\n\ncommit 228f855fb57ae25599eee64a3f3db4f7f405b34f\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:32 2019 -0700\n\n perf: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Will Deacon \u003cwill@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 4b9ace9c25dc5d672a548da34cebf3a39710017f\nAuthor: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\nDate: Thu Jul 4 11:53:20 2019 +0300\n\n perf/imx_ddr: Add MODULE_DEVICE_TABLE\n \n This is required for automatic probing when driver is built as a module.\n \n Fixes: 9a66d36cc7ac (\"drivers/perf: imx_ddr: Add DDR performance counter support to perf\")\n Acked-by: Frank Li \u003cfrank.li@nxp.com\u003e\n Signed-off-by: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 6a7ce95d752efa86a1a383385d4f8035c224dc3d\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Jul 31 18:02:19 2019 +0200\n\n staging/octeon: Fix build error without CONFIG_NETDEVICES\n \n While do COMPILE_TEST build without CONFIG_NETDEVICES,\n we get Kconfig warning:\n \n WARNING: unmet direct dependencies detected for PHYLIB\n Depends on [n]: NETDEVICES [\u003dn]\n Selected by [y]:\n - OCTEON_ETHERNET [\u003dy] \u0026\u0026 STAGING [\u003dy] \u0026\u0026 (CAVIUM_OCTEON_SOC \u0026\u0026 NETDEVICES [\u003dn] || COMPILE_TEST [\u003dy])\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Reported-by: Mark Brown \u003cbroonie@kernel.org\u003e\n Fixes: 171a9bae68c7 (\"staging/octeon: Allow test build on !MIPS\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1dc558923c5ceab0b4eb78dc0050ea4c4aa1accc\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Wed Jul 31 09:01:44 2019 +0100\n\n RDMA/core: fix spelling mistake \"Nelink\" -\u003e \"Netlink\"\n \n There is a spelling mistake in a warning message, fix it.\n \n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Reviewed-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Link: https://lore.kernel.org/r/20190731080144.18327-1-colin.king@canonical.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit cb560f5fd951cea1e5f4afd2820b1c0deb75dcb1\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:20 2019 -0700\n\n infiniband: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Doug Ledford \u003cdledford@redhat.com\u003e\n Cc: Jason Gunthorpe \u003cjgg@ziepe.ca\u003e\n Cc: linux-rdma@vger.kernel.org\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190730181557.90391-21-swboyd@chromium.org\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit d11786bb9664a5bed47e7839265f49bb26d54a1b\nAuthor: Petr Machata \u003cpetrm@mellanox.com\u003e\nDate: Wed Jul 31 10:30:27 2019 +0000\n\n selftests: mlxsw: Add a test for leftover DSCP rule\n \n Commit dedfde2fe1c4 (\"mlxsw: spectrum_dcb: Configure DSCP map as the last\n rule is removed\") fixed a problem in mlxsw where last DSCP rule to be\n removed remained in effect when DSCP rewrite was applied.\n \n Add a selftest that covers this problem.\n \n Signed-off-by: Petr Machata \u003cpetrm@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7700476f319844022f6e469e7acb578178ee29de\nAuthor: Petr Machata \u003cpetrm@mellanox.com\u003e\nDate: Wed Jul 31 10:30:26 2019 +0000\n\n selftests: mlxsw: Fix local variable declarations in DSCP tests\n \n These two tests have some problems in the global scope pollution and on\n contrary, contain unnecessary local declarations. Fix them.\n \n Signed-off-by: Petr Machata \u003cpetrm@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7084148854e75090712439f73fbe2f4d6bfd96e1\nAuthor: Ding Xiang \u003cdingxiang@cmss.chinamobile.com\u003e\nDate: Wed Jul 31 16:53:46 2019 +0800\n\n myri10ge: remove unneeded variable\n \n \"error\" is unneeded,just return 0\n \n Signed-off-by: Ding Xiang \u003cdingxiang@cmss.chinamobile.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a9d41e7b8b1fe43c0beeab9a45d4c886a5e4ee85\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Wed Jul 31 10:06:38 2019 +0200\n\n net: ag71xx: Slighly simplify code in \u0027ag71xx_rings_init()\u0027\n \n A few lines above, we have:\n tx_size \u003d BIT(tx-\u003eorder);\n \n So use \u0027tx_size\u0027 directly to be consistent with the way \u0027rx-\u003edescs_cpu\u0027 and\n \u0027rx-\u003edescs_dma\u0027 are computed below.\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7ff1154d459d25f8f674765a5ec575207516168c\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Mon Jun 17 16:35:47 2019 +0200\n\n ARM: dts: vexpress: Add missing newline at end of file\n \n \"git diff\" says:\n \n \\ No newline at end of file\n \n after modifying the file.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit 8d65f58021fbbc3fd5464697789bf015fcca03a4\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Wed Jul 31 15:55:47 2019 +0100\n\n ARM: dts: vexpress: add missing SPDX GPL-2.0 license identifier\n \n Commit b24413180f56 (\"License cleanup: add SPDX GPL-2.0 license\n identifier to files with no license\") seem to have missed this\n one file.\n \n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n\ncommit e17be6e1b713f188728a23ebfba15546a025c109\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:10 2019 -0700\n\n dmaengine: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Vinod Koul \u003cvkoul@kernel.org\u003e\n Cc: Dan Williams \u003cdan.j.williams@intel.com\u003e\n Cc: dmaengine@vger.kernel.org\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190730181557.90391-11-swboyd@chromium.org\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 7f5d7425748d1cc2dadb6966ec84363d3342b5a1\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Jul 29 17:52:21 2019 -0500\n\n dmaengine: imx-dma: Mark expected switch fall-through\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warning (Building: arm):\n \n drivers/dma/imx-dma.c: In function ‘imxdma_xfer_desc’:\n drivers/dma/imx-dma.c:542:6: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n if (slot \u003d\u003d IMX_DMA_2D_SLOT_A) {\n ^\n drivers/dma/imx-dma.c:559:2: note: here\n case IMXDMA_DESC_MEMCPY:\n ^~~~\n \n Notice that, in this particular case, the code comment is\n modified in accordance with what GCC is expecting to find.\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Link: https://lore.kernel.org/r/20190729225221.GA24269@embeddedor\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit a9afc9ea93995021f29025e62d83415a45c5f58b\nAuthor: Jonathan Hunter \u003cjonathanh@nvidia.com\u003e\nDate: Wed Jul 31 11:16:39 2019 +0100\n\n dmaengine: tegra210-adma: Don\u0027t program FIFO threshold\n \n The Tegra210 ADMA supports two modes for transferring data to a FIFO\n which are ...\n \n 1. Transfer data to/from the FIFO as soon as a single burst can be\n transferred.\n 2. Transfer data to/from the FIFO based upon FIFO thresholds, where\n the FIFO threshold is specified in terms on multiple bursts.\n \n Currently, the ADMA driver programs the FIFO threshold values in the\n FIFO_CTRL register, but never enables the transfer mode that uses\n these threshold values. Given that these have never been used so far,\n simplify the ADMA driver by removing the programming of these threshold\n values.\n \n Signed-off-by: Jonathan Hunter \u003cjonathanh@nvidia.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Link: https://lore.kernel.org/r/20190731101639.22755-1-jonathanh@nvidia.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 1b6c3c6d4607784c83160cf7fb9c94fae5c117a1\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Tue Jul 30 11:04:07 2019 -0700\n\n drm/i915: Move MOCS setup to intel_mocs.c\n \n Hide the details of MOCS setup from i915_gem by moving both current calls\n into one in intel_mocs_init.\n \n Cc: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190713010940.17711-21-lucas.demarchi@intel.com\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730180407.5993-6-lucas.demarchi@intel.com\n\ncommit a7a7a0e6ebde34e05793d390cc9303e06e8f8dd1\nAuthor: Michel Thierry \u003cmichel.thierry@intel.com\u003e\nDate: Tue Jul 30 11:04:06 2019 -0700\n\n drm/i915/tgl: Tigerlake only has global MOCS registers\n \n Until Icelake, each engine had its own set of 64 MOCS registers. In\n order to simplify, Tigerlake moves to only 64 Global MOCS registers,\n which are no longer part of the engine context. Since these registers\n are now global, they also only need to be initialized once.\n \n \u003eFrom Gen12 onwards, MOCS must specify the target cache (3:2) and LRU\n management (5:4) fields and cannot be programmed to \u0027use the value from\n Private PAT\u0027, because these fields are no longer part of the PPAT. Also\n cacheability control (1:0) field has changed, 00 no longer means \u0027use\n controls from page table\u0027, but uncacheable (UC).\n \n v2 (Lucas):\n - Move the changes to the fault registers to a separate commit - the\n old ones overlap with the range used by the new global MOCS\n (requested by Daniele)\n v3 (Lucas):\n - Clarify comment about setting the unused entries to the same value\n of index 0, that is the invalid entry (requested by Daniele)\n - Move changes to DONE_REG and ERROR_GEN6 to a separate commit\n (requested by Daniele)\n \n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Michel Thierry \u003cmichel.thierry@intel.com\u003e\n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Tomasz Lis \u003ctomasz.lis@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730180407.5993-5-lucas.demarchi@intel.com\n\ncommit 2ddf992179c45fb93de190b5c6ae16d2a4f4849a\nAuthor: Tomasz Lis \u003ctomasz.lis@intel.com\u003e\nDate: Tue Jul 30 11:04:05 2019 -0700\n\n drm/i915/tgl: Define MOCS entries for Tigerlake\n \n The MOCS table is published as part of bspec, and versioned. Entries\n are supposed to never be modified, but new ones can be added. Adding\n entries increases table version. The patch includes version 1 entries.\n \n Two of the 3 legacy entries used for gen9 are no longer expected to work.\n Although we are changing the gen11 table, those changes are supposed to\n be backward compatible since we are only touching previously undefined\n entries.\n \n v2: Add the missing entries in 49-51 range and replace \"HW reserved\"\n terminology to what it actually is: L1 is implicitly enabled\n (from Daniele)\n v3: Use a different table for Tiger Lake since entries 0 and 1 are not\n the same (from Daniele)\n \n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@intel.com\u003e\n Signed-off-by: Tomasz Lis \u003ctomasz.lis@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730180407.5993-4-lucas.demarchi@intel.com\n\ncommit 23dea05191be897a5f30416342c3cff101f2701c\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Tue Jul 30 11:04:04 2019 -0700\n\n drm/i915/tgl: stop using ERROR_GEN6 and DONE_REG\n \n These registers have been removed on gen12.\n \n v2: merge common branch for IS_GEN_RANGE(i915, 6, 11)\n \n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730180407.5993-3-lucas.demarchi@intel.com\n\ncommit 91b59cd98a970f566b81418b9bd6f14f918f62d8\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Tue Jul 30 11:04:03 2019 -0700\n\n drm/i915/tgl: Move fault registers to their new offset\n \n The fault registers moved to another offset. The old location is now\n taken by the global MOCS registers, to be added in a follow up change.\n \n Based on previous patches by Michel Thierry \u003cmichel.thierry@intel.com\u003e.\n \n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730180407.5993-2-lucas.demarchi@intel.com\n\ncommit 900c9173ca692f08de9332cf5e4f0ab9486db8b9\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Tue Jul 30 11:26:14 2019 -0700\n\n drm/i915: remove dangling forward declaration\n \n Commit 20a7f2fc4d7a (\"drm/i915: Convert intel_mocs_init_l3cc_table to\n intel_gt\") removed the only user.\n \n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730182614.14379-1-lucas.demarchi@intel.com\n\ncommit bed7a2182de6833f5d0f990a656bffb1c6000c70\nAuthor: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\nDate: Fri Jul 26 19:23:17 2019 +0200\n\n drm/radeon: Provide ddc symlink in connector sysfs directory\n \n Use the ddc pointer provided by the generic connector.\n \n Signed-off-by: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/0836f34238730afce3f4d6b13f5cf04f832b668a.1564161140.git.andrzej.p@collabora.com\n\ncommit 5b50fa2b35a4ddad11cb3d06231bf71759b49566\nAuthor: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\nDate: Fri Jul 26 19:23:16 2019 +0200\n\n drm/amdgpu: Provide ddc symlink in connector sysfs directory\n \n Use the ddc pointer provided by the generic connector.\n \n Signed-off-by: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/7fee0fa0d0f77af6595d283d5f3ae5d551475821.1564161140.git.andrzej.p@collabora.com\n\ncommit cfb444552926989f9f558f8863270e50e3efb4a3\nAuthor: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\nDate: Fri Jul 26 19:23:15 2019 +0200\n\n drm/bridge: ti-tfp410: Provide ddc symlink in connector sysfs directory\n \n Use the ddc pointer provided by the generic connector.\n \n Signed-off-by: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/3b61da77a6456805db0deffe6d1a2343dd784730.1564161140.git.andrzej.p@collabora.com\n\ncommit 9ebc4d2140adcdea2399b42d3f0d2f7e141ac1a8\nAuthor: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\nDate: Fri Jul 26 19:23:14 2019 +0200\n\n drm/bridge: dw-hdmi: Provide ddc symlink in connector sysfs directory\n \n Use the ddc pointer provided by the generic connector.\n \n Signed-off-by: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\n Acked-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/4bcf0f154c683c9787fa34f911ebc52de6b4a7a1.1564161140.git.andrzej.p@collabora.com\n\ncommit a4f9087e85de141e4e6d21ac2c583ae096cc9aba\nAuthor: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\nDate: Fri Jul 26 19:23:13 2019 +0200\n\n drm/bridge: dumb-vga-dac: Provide ddc symlink in connector sysfs directory\n \n Use the ddc pointer provided by the generic connector.\n \n Signed-off-by: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/ebb75e71b8b7c8d65d54a947a03fd21b8969fb3a.1564161140.git.andrzej.p@collabora.com\n\ncommit 350fd554ee44325661d0d3c3831f428f4fbb0f2d\nAuthor: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\nDate: Fri Jul 26 19:23:12 2019 +0200\n\n drm/ast: Provide ddc symlink in connector sysfs directory\n \n Use the ddc pointer provided by the generic connector.\n \n Signed-off-by: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\n Reviewed-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/ff968add8074d109aeed02f0708edd85138246c3.1564161140.git.andrzej.p@collabora.com\n\ncommit 9572ae176a10f3b2ecfd51bfcd2e949bff566786\nAuthor: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\nDate: Fri Jul 26 19:23:11 2019 +0200\n\n drm/mgag200: Provide ddc symlink in connector sysfs directory\n \n Use the ddc pointer provided by the generic connector.\n \n Signed-off-by: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\n Reviewed-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/d9e8d4281f6778e5598410a44f1b29c85df1c16d.1564161140.git.andrzej.p@collabora.com\n\ncommit 7058e76682d762e211e7681155c5e11709182271\nAuthor: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\nDate: Fri Jul 26 19:23:10 2019 +0200\n\n drm: sti: Provide ddc symlink in hdmi connector sysfs directory\n \n Use the ddc pointer provided by the generic connector.\n \n Signed-off-by: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\n Reviewed-by: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/510765aff8ef99683aa2da48bd08004376b1980a.1564161140.git.andrzej.p@collabora.com\n\ncommit 2ae7eb372ed42e480d55848d597b44d4e47b0ab6\nAuthor: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\nDate: Fri Jul 26 19:23:05 2019 +0200\n\n drm/imx: imx-tve: Provide ddc symlink in connector\u0027s sysfs\n \n Use the ddc pointer provided by the generic connector.\n \n Signed-off-by: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\n Acked-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/1c660776741760b8094484268b670a09da8a9042.1564161140.git.andrzej.p@collabora.com\n\ncommit be0ec35940bc477eefb6370f04a9e031692ba642\nAuthor: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\nDate: Fri Jul 26 19:23:04 2019 +0200\n\n drm/imx: imx-ldb: Provide ddc symlink in connector\u0027s sysfs\n \n Use the ddc pointer provided by the generic connector.\n \n Signed-off-by: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\n Acked-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/bcfe39a0333df951a7d160b3a946c2c32e7eac7f.1564161140.git.andrzej.p@collabora.com\n\ncommit 1e8f17855ff884d49bec491ab8f5404aaca919bf\nAuthor: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\nDate: Fri Jul 26 19:23:01 2019 +0200\n\n drm/sun4i: hdmi: Provide ddc symlink in sun4i hdmi connector sysfs directory\n \n Use the ddc pointer provided by the generic connector.\n \n Signed-off-by: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/b5a8881b226a9af7d31eb6a57d7fe0fccdcb3a47.1564161140.git.andrzej.p@collabora.com\n\ncommit 100163df420305b78153e6f5ec10c90d755acee3\nAuthor: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\nDate: Fri Jul 26 19:22:56 2019 +0200\n\n drm: Add drm_connector_init() variant with ddc\n \n Allow passing ddc adapter pointer to the init function. Even if\n drm_connector_init() sometime in the future decides to e.g. memset() all\n connector fields to zeros, the newly added function ensures that at its\n completion the ddc member of connector is correctly set.\n \n Signed-off-by: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\n Acked-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/3915224ae895240fd0973cf7f06b9d453e4d8520.1564161140.git.andrzej.p@collabora.com\n\ncommit e1a29c6c59553d80a8e17d63494c65a13fb8e241\nAuthor: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\nDate: Fri Jul 26 19:22:55 2019 +0200\n\n drm: Add ddc link in sysfs created by drm_connector\n \n Add generic code which creates symbolic links in sysfs, pointing to ddc\n interface used by a particular video output. For example:\n \n ls -l /sys/class/drm/card0-HDMI-A-1/ddc\n lrwxrwxrwx 1 root root 0 Jun 24 10:42 /sys/class/drm/card0-HDMI-A-1/ddc \\\n -\u003e ../../../../soc/13880000.i2c/i2c-2\n \n This makes it easy for user to associate a display with its ddc adapter\n and use e.g. ddcutil to control the chosen monitor.\n \n This patch adds an i2c_adapter pointer to struct drm_connector. Particular\n drivers can then use it instead of using their own private instance. If a\n connector contains a ddc, then create a symbolic link in sysfs.\n \n Signed-off-by: Andrzej Pietrasiewicz \u003candrzej.p@collabora.com\u003e\n Acked-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Reviewed-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/d470def6cd661b777faeee67b5838a4623c4010e.1564161140.git.andrzej.p@collabora.com\n\ncommit 8c57588313805d4e12a3d470dee3405c2236655f\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Jul 29 10:51:51 2019 -0500\n\n ALSA: hda/intel: stop probe if DMICS are detected on Skylake+ platforms\n \n The legacy HD-Audio driver cannot handle Skylake+ platforms with\n digital microphones. For those platforms, the SOF or SST drivers need\n to be used.\n \n This patch provides an automatic way of detecting the presence of\n DMICs using NHTL information reported by the BIOS. A kernel kconfig\n option or a kernel module parameter provide an opt-in means of\n stopping the probe. The kernel would then look for an alternate driver\n registered for the same PCI ID to probe.\n \n With this capability, distros no longer have to blacklist\n snd-hda-intel, but still need to make sure the SOF/SST drivers are\n functional by providing the relevant firmware and topology files in\n /lib/firmware/intel\n \n The coexistence between SOF and SST drivers and their dynamic\n detection is not addressed by this patch, different mechanisms need to\n be used, e.g. DMI-based quirks.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 1169cbf6b98e11397715d457e42070e066325fc0\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Jul 29 10:51:50 2019 -0500\n\n ASoC: Intel: Skylake: use common NHLT module\n \n No functionality change, only use common functions now.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 7a33ea70e1868ee578fe2e9a85dd300efa1a35d5\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Jul 29 10:51:49 2019 -0500\n\n ALSA: hda: intel-nhlt: handle NHLT VENDOR_DEFINED DMIC geometry\n \n The NHLT spec defines a VENDOR_DEFINED geometry, which requires\n reading additional information to figure out the number of\n microphones.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 303681f4356d322232dd5f6d9eb4bc62666064c5\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Jul 29 10:51:48 2019 -0500\n\n ALSA: hda: move parts of NHLT code to new module\n \n Move parts of the code outside of the Skylake driver to help detect\n the presence of DMICs (which are not supported by the HDaudio legacy\n driver).\n \n No functionality change (except for the removal of useless OR\n operations), only indentation and checkpatch fixes, making sure\n that the code compiles without ACPI and fixing an ACPI leak\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 63643b5902c4bf096b504b0563f5426ba5baef15\nAuthor: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\nDate: Mon Jul 29 10:51:47 2019 -0500\n\n ASoC: Intel: Skylake: move NHLT header to common directory\n \n Prepare move from NHLT code to common directory, starting with header.\n \n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 77b8cabf3d529725e59b71671a562556f6c84446\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Thu Jul 25 12:51:32 2019 +0200\n\n drm/gm12u320: Move driver to drm/tiny\n \n Move the driver to the new haven for tiny DRM drivers.\n \n Cc: Hans de Goede \u003chdegoede@redhat.com\u003e\n Suggested-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Acked-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725105132.22545-4-noralf@tronnes.org\n\ncommit 043386a0f094e0ae8bda6f824461debe9b708db5\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Thu Jul 25 12:51:31 2019 +0200\n\n drm/tinydrm: Rename folder to tiny\n \n The drm in tinydrm is superfluous so rename to tiny.\n \n Suggested-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Acked-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725105132.22545-3-noralf@tronnes.org\n\ncommit 28c47e16ea2a19adb47fe2c182cbd61cb854237c\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Thu Jul 25 12:51:30 2019 +0200\n\n drm/tinydrm/Kconfig: Remove menuconfig DRM_TINYDRM\n \n This makes the tiny drivers visible by default without having to enable a\n knob.\n \n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Acked-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Hans de Goede \u003chdegoede@redhat.com\u003e to it once\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725105132.22545-2-noralf@tronnes.org\n\ncommit 6492a4c12c658caca7d3e0ab12f2280aa62747a8\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Thu Jul 4 10:46:09 2019 +0200\n\n serial: sh-sci: use driver core functions, not sysfs ones.\n \n This is a driver, do not call \"raw\" sysfs functions, instead call driver\n core ones. Specifically convert the use of sysfs_create_file() and\n sysfs_remove_file() to use device_create_file() and device_remove_file()\n \n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-serial@vger.kernel.org\n Link: https://lore.kernel.org/r/20190704084617.3602-4-gregkh@linuxfoundation.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 9ba2986ff5799551e023d39fa3e8ebc6d0b148c3\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Wed Jul 31 10:47:36 2019 +0100\n\n staging: rtl8188eu: remove redundant assignment to variable rtstatus\n \n Variable rtstatus is being initialized with a value that is never read\n and rtstatus is being re-assigned a little later on. The assignment is\n redundant and hence can be removed. Also, make rtstatus a bool to\n match the function return type.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190731094736.28637-1-colin.king@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f7d120d9e0b9cbdb20d45ff7998d5fe4820d1cdd\nAuthor: Chris Chiu \u003cchiu@endlessm.com\u003e\nDate: Wed Jul 31 18:35:17 2019 +0800\n\n staging: rtl8723bs: indicate disconnection when disconnecting\n \n The Realtek RTL8723BS only connects successfully at the very first\n time, then it always fails attempting to switch to another AP. No\n authentication/association observed from the air capture for each\n attempt due to the cfg80211 believes the device is still connected.\n \n Fix this by forcing to indicate the disconnection events during\n disconnection so the cfg80211_connect can connect to a different\n AP without problem.\n \n Signed-off-by: Chris Chiu \u003cchiu@endlessm.com\u003e\n Link: https://lore.kernel.org/r/20190731103517.66903-1-chiu@endlessm.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 82cb54856874b1b374f18420be013ff4057700a9\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Mon Jul 29 12:55:21 2019 +0300\n\n asm-generic: make simd.h a mandatory include/asm header\n \n The generic aegis128 software crypto driver recently gained support\n for using SIMD intrinsics to increase performance, for which it\n uncondionally #include\u0027s the \u003casm/simd.h\u003e header. Unfortunately,\n this header does not exist on many architectures, resulting in\n build failures.\n \n Since asm-generic already has a version of simd.h, let\u0027s make it\n a mandatory header so that it gets instantiated on all architectures\n that don\u0027t provide their own version.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Acked-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit caa918ef14065b812737f3ee4ac349dcfff196e6\nAuthor: Andra Danciu \u003candradanciu1997@gmail.com\u003e\nDate: Wed Jul 31 14:19:30 2019 +0300\n\n ASoC: codecs: Add uda1334 codec driver\n \n The UDA1334BTS supports the I2S-bus data format with word lengths of up\n to 24 bits serial data format and has basic features such as de-emphasis\n (at 44.1 kHz sampling rate) and mute.\n \n Datasheet can be found at:\n https://www.nxp.com/docs/en/data-sheet/UDA1334BTS.pdf\n \n Cc: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Andra Danciu \u003candradanciu1997@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190731111930.20230-3-andradanciu1997@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit f39b07fdfb688724fedabf5507e15eaf398f2500\nAuthor: Shay Bar \u003cshay.bar@celeno.com\u003e\nDate: Wed Jul 3 16:18:48 2019 +0300\n\n mac80211: HE STA disassoc due to QOS NULL not sent\n \n In case of HE AP-STA link, ieee80211_send_nullfunc() will not\n send the QOS NULL packet to check if AP is still associated.\n \n In this case, probe_send_count will be non-zero and\n ieee80211_sta_work() will later disassociate the AP, even\n though no packet was ever sent.\n \n Fix this by decrementing probe_send_count and not calling\n ieee80211_send_nullfunc() in case of HE link, so that we\n still wait for some time for the AP beacon to reappear and\n don\u0027t disconnect right away.\n \n Signed-off-by: Shay Bar \u003cshay.bar@celeno.com\u003e\n Link: https://lore.kernel.org/r/20190703131848.22879-1-shay.bar@celeno.com\n [clarify commit message]\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit d6de65fde51644f6ed6b1c0c05fef6a2da5ff768\nAuthor: Andra Danciu \u003candradanciu1997@gmail.com\u003e\nDate: Wed Jul 31 14:19:29 2019 +0300\n\n dt-bindings: sound: Add bindings for UDA1334 codec\n \n The UDA1334 is an NXP audio codec, supports the I2S-bus data format\n and has basic features such as de-emphasis (at 44.1 kHz sampling\n rate) and mute. Product information can be found at:\n https://www.nxp.com/pages/low-power-audio-dac-with-pll:UDA1334\n \n Cc: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Andra Danciu \u003candradanciu1997@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190731111930.20230-2-andradanciu1997@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 10392fcad7dfc8ea38959b18327ff18b81b1c161\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:30 2019 +0800\n\n ASoC: meson: axg-tdm-formatter: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-27-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit a95c901248642d62dc7462a2d2190c8ac183c84f\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:29 2019 +0800\n\n ASoC: meson: g12a-tohdmitx: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-26-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 12a63c0fa03691328b948690601dc7dde8fc527b\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:28 2019 +0800\n\n ASoC: au1x: psc-i2s: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-25-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit b1abe8780041d568ae07013a3592519fca855a1c\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:27 2019 +0800\n\n ASoC: psc-ac97: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-24-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit b885c9fa363fa4604d78ea00adfed64db656fb78\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:26 2019 +0800\n\n ASoC: uniphier: aio-dma: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-23-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 4e5bc35988af2483138feb3c5c3bf5fb0cae4228\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:25 2019 +0800\n\n ASoC: uniphier: evea: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-22-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 2d1ffc7d0f58f6a39a34854e952f5020fc655af0\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:24 2019 +0800\n\n ASoC: inno_rk3036: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-21-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit a8dc106410a2c3c87917447fb8c132b66062443e\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:23 2019 +0800\n\n ASoC: jz4740: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-20-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 72d09322e709c5c2d7112d6a19ec9f3846d3609d\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:22 2019 +0800\n\n ASoC: mmp-sspa: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-19-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 94db63761576a98279e19f51173dec85f803bad4\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:21 2019 +0800\n\n ASoC: codecs: jz4725b: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-18-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit f62d6426718837663484bbf0ed94e56573cbc365\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:20 2019 +0800\n\n ASoC: codecs: msm8916-wcd: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-17-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 1c63223cda84fb10643938e4e4eb02110f8ae9a2\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:19 2019 +0800\n\n ASoC: stm32: sai: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-16-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 790b36578e23264ff05be2ba8466ebcf7d3db38a\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:38 2019 +0800\n\n ASoC: sun8i-codec: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-35-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 54dd39e4aca8edb231e13349ac3a3a44ef28d3d4\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:37 2019 +0800\n\n ASoC: sunxi: sun50i-codec-analog: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-34-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit e12fd61fd7c1bbbb758ede9fdf58ac2bd5de6b33\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:36 2019 +0800\n\n ASoC: sun8i-codec-analog: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-33-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit cf80aa2c1359f5d014981e251049bcc21a2217bc\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:35 2019 +0800\n\n ASoC: xlnx: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-32-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9b2089313156ad16bd257217cc67e10597a82923\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:34 2019 +0800\n\n ASoC: meson: axg-fifo: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-31-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c43fd289dd58951f3cfddb3bb66a6ed7747c5986\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:33 2019 +0800\n\n ASoC: meson: axg-spdifout: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-30-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit d61feec0f60c250c786a9e0069ce909979491c7b\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:32 2019 +0800\n\n ASoC: meson: axg-spdifin: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-29-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 2e73d98f2ecf61008926bb8a425c3281a89959c0\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:31 2019 +0800\n\n ASoC: meson: axg-pdm: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-28-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 0c03e37af47efcb8600f95f399783c082fcf2f93\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Wed Jul 24 08:00:23 2019 +0200\n\n ASoC: cs4271: Fix a typo in the CS4171_NR_RATIOS\n \n This should be CS4271_NR_RATIOS.\n Fix it and use it.\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Link: https://lore.kernel.org/r/20190724060023.31302-1-christophe.jaillet@wanadoo.fr\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 22581e7c51f50fbbadba70499bb5e2d09f830cbb\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Jul 29 17:15:34 2019 -0500\n\n ASoC: ti: Mark expected switch fall-throughs\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warning (Building: arm):\n \n sound/soc/ti/n810.c: In function ‘n810_ext_control’:\n sound/soc/ti/n810.c:48:10: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n line1l \u003d 1;\n ~~~~~~~^~~\n sound/soc/ti/n810.c:49:2: note: here\n case N810_JACK_HP:\n ^~~~\n \n sound/soc/ti/rx51.c: In function ‘rx51_ext_control’:\n sound/soc/ti/rx51.c:57:6: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n hs \u003d 1;\n ~~~^~~\n sound/soc/ti/rx51.c:58:2: note: here\n case RX51_JACK_HP:\n ^~~~\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Acked-by: Jarkko Nikula \u003cjarkko.nikula@bitmer.com\u003e\n Link: https://lore.kernel.org/r/20190729221534.GA18696@embeddedor\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit ebdd7be5415c7795c77609ad908222038e441835\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:18 2019 +0800\n\n ASoC: xtfpga-i2s: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-15-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 89dd38bf130e661d92c6ab7e3b003907f1d3f367\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:17 2019 +0800\n\n ASoC: kirkwood-i2s: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-14-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 0fb46f541bb3e6e5bbb155c08b9e75f92c21acb9\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:16 2019 +0800\n\n ASoC: spear: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-13-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 1327bfe2887c0e29bcdc0c5c3e445da107ccb4e2\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:15 2019 +0800\n\n ASoC: mxs-saif: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-12-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c9dbe688983b1d77a9a46cf6b7ac88348a1ca395\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:14 2019 +0800\n\n ASoC: sirf: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-11-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit d400b1b30cbff1c0962bbb3f895e839ccd120cb7\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:13 2019 +0800\n\n ASoC: bcm2835-i2s: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-10-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 959bb6b54d708652fe06097feecc3acd4d6b8c89\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:12 2019 +0800\n\n ASoC: fsl_audmix: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-9-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit d003e3081c1d069b94d1b5f4c3bc8234d4603282\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:11 2019 +0800\n\n ASoC: imx-audmux: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-8-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 61e799e3658952397da80f6c2e3581e7b65209a8\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:10 2019 +0800\n\n ASoC: mt6797: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-7-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit fceef72b68d63593d8543a8d608ccc8427e13d56\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:09 2019 +0800\n\n ASoC: mt8173: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-6-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 4b3540074361fc7b97f2da54365c55ceca6333d2\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:08 2019 +0800\n\n ASoC: ep93xx-i2s: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-5-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit f052172a5bb474fef2d0c3911655c9e107a3c024\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:07 2019 +0800\n\n ASoC: rockchip: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-4-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 7d22808641cdea50baeae636acf5d592a7aa1239\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:06 2019 +0800\n\n ASoC: tegra: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-3-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 21cc4ea85879c1b424fb01caa6e481c14286f7f0\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 23:07:05 2019 +0800\n\n ASoC: tegra20_das: use devm_platform_ioremap_resource() to simplify code\n \n Use devm_platform_ioremap_resource() to simplify the code a bit.\n This is detected by coccinelle.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190727150738.54764-2-yuehaibing@huawei.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 2c4956bc1e9062e5e3c5ea7612294f24e6d4fbdd\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Mon Jul 29 10:01:39 2019 +0200\n\n ASoC: meson: g12a-tohdmitx: override codec2codec params\n \n So far, forwarding the hw_params of the input to output relied on the\n .hw_params() callback of the cpu side of the codec2codec link to be called\n first. This is a bit weak.\n \n Instead, override the stream params of the codec2codec to link to set it up\n correctly.\n \n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Link: https://lore.kernel.org/r/20190729080139.32068-1-jbrunet@baylibre.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 931c940fa5643d3d8cd62ba7bfa0c61077eb7b59\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Fri Jul 26 21:19:04 2019 +0300\n\n crypto: s390/aes - fix name clash after AES library refactor\n \n The newly introduced AES library exposes aes_encrypt/aes_decrypt\n routines so rename existing occurrences of those identifiers in\n the s390 driver.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 7dcddef6f769d7e60691c732eb6d09cdb1d9df76\nAuthor: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nDate: Wed Jul 31 15:29:52 2019 +1000\n\n cpuidle: header file stubs must be \"static inline\"\n \n An x86_64 allmodconfig build produces these errors:\n \n x86_64-linux-gnu-ld: kernel/sched/core.o: in function `cpuidle_poll_time\u0027:\n core.c:(.text+0x230): multiple definition of `cpuidle_poll_time\u0027; arch/x86/\u003d\n kernel/process.o:process.c:(.text+0xc0): first defined here\n \n (and more)\n \n Fixes: 259231a04561 (\"cpuidle: add poll_limit_ns to cpuidle_device structure\")\n Signed-off-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 61059b7048ee1d635673f3014ea9f7e3e4b1532f\nAuthor: Ding Xiang \u003cdingxiang@cmss.chinamobile.com\u003e\nDate: Wed Jul 31 17:47:54 2019 +0800\n\n gpio: ixp4xx: remove redundant dev_err message\n \n devm_ioremap_resource already contains error message, so remove\n the redundant dev_err message\n \n Signed-off-by: Ding Xiang \u003cdingxiang@cmss.chinamobile.com\u003e\n Signed-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n\ncommit 4b8e1106dd95d4d7f2781258d3871b445f3928ce\nAuthor: Chengguang Xu \u003ccgxu519@zoho.com.cn\u003e\nDate: Wed Jul 24 13:32:16 2019 +0800\n\n quota: fix condition for resetting time limit in do_set_dqblk()\n \n We reset time limit when current usage is smaller\n or equal to soft limit in other place, so follow\n this rule in do_set_dqblk().\n \n Signed-off-by: Chengguang Xu \u003ccgxu519@zoho.com.cn\u003e\n Link: https://lore.kernel.org/r/20190724053216.19392-1-cgxu519@zoho.com.cn\n Signed-off-by: Jan Kara \u003cjack@suse.cz\u003e\n\ncommit b6aeffc5852f39db6e6e56da5327d0c43ac3c30a\nAuthor: Chengguang Xu \u003ccgxu519@zoho.com.cn\u003e\nDate: Tue Jul 23 19:21:55 2019 +0800\n\n ext2: code cleanup for ext2_free_blocks()\n \n Call ext2_data_block_valid() for block range validity.\n \n Signed-off-by: Chengguang Xu \u003ccgxu519@zoho.com.cn\u003e\n Link: https://lore.kernel.org/r/20190723112155.20329-2-cgxu519@zoho.com.cn\n Signed-off-by: Jan Kara \u003cjack@suse.cz\u003e\n\ncommit e5d395974e043cdcedcd84a0d41aaebb723786d8\nAuthor: Chengguang Xu \u003ccgxu519@zoho.com.cn\u003e\nDate: Tue Jul 23 19:21:54 2019 +0800\n\n ext2: fix block range in ext2_data_block_valid()\n \n For block validity we should check the block range\n from start_block to start_block + count - 1, so fix\n the range in ext2_data_block_valid() and also modify\n the count argument properly in calling place.\n \n Signed-off-by: Chengguang Xu \u003ccgxu519@zoho.com.cn\u003e\n Link: https://lore.kernel.org/r/20190723112155.20329-1-cgxu519@zoho.com.cn\n Signed-off-by: Jan Kara \u003cjack@suse.cz\u003e\n\ncommit 6fbacb8539a6659d446a9efabb538cfc007c1427\nAuthor: Steven J. Magnani \u003csteve.magnani@digidescorp.com\u003e\nDate: Thu Jul 11 08:38:52 2019 -0500\n\n udf: support 2048-byte spacing of VRS descriptors on 4K media\n \n Some UDF creators (specifically Microsoft, but perhaps others) mishandle\n the ECMA-167 corner case that requires descriptors within a Volume\n Recognition Sequence to be placed at 4096-byte intervals on media where\n the block size is 4K. Instead, the descriptors are placed at the 2048-\n byte interval mandated for media with smaller blocks. This nonconformity\n currently prevents Linux from recognizing the filesystem as UDF.\n \n Modify the driver to tolerate a misformatted VRS on 4K media.\n \n [JK: Simplified descriptor checking]\n Signed-off-by: Steven J. Magnani \u003csteve@digidescorp.com\u003e\n Tested-by: Steven J. Magnani \u003csteve@digidescorp.com\u003e\n Link: https://lore.kernel.org/r/20190711133852.16887-2-steve@digidescorp.com\n Signed-off-by: Jan Kara \u003cjack@suse.cz\u003e\n\ncommit ba54aef0313322d9eea0fca648170d5a3c906de4\nAuthor: Steven J. Magnani \u003csteve.magnani@digidescorp.com\u003e\nDate: Thu Jul 11 08:38:51 2019 -0500\n\n udf: refactor VRS descriptor identification\n \n Extract code that parses a Volume Recognition Sequence descriptor\n (component), in preparation for calling it twice against different\n locations in a block.\n \n Signed-off-by: Steven J. Magnani \u003csteve@digidescorp.com\u003e\n Link: https://lore.kernel.org/r/20190711133852.16887-1-steve@digidescorp.com\n Signed-off-by: Jan Kara \u003cjack@suse.cz\u003e\n\ncommit 63064d822c964c04107ead05b64eddccfa142005\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Tue Jul 30 16:07:40 2019 -0700\n\n drm/i915/uc: Move uC WOPCM setup in uc_init_hw\n \n The register we write are not WOPCM regs but uC ones related to how\n GuC and HuC are going to use the WOPCM, so it makes logical sense\n for them to be programmed as part of uc_init_hw. The WOPCM map on the\n other side is not uC-specific (although that is our main use-case), so\n keep that separate.\n \n v2: move write_and_verify to uncore, fix log, re-use err_out tag,\n add intel_wopcm_guc_base, fix log\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730230743.19542-2-daniele.ceraolospurio@intel.com\n\ncommit 602776f96bfa50a6107af017522501c43a82b29a\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Tue Jul 30 16:07:39 2019 -0700\n\n drm/i915/uc: Don\u0027t enable communication twice on resume\n \n When coming out of S3/S4 we sanitize and re-init the HW, which includes\n enabling communication during uc_init_hw. We therefore don\u0027t want to do\n that again in uc_resume and can just tell GuC to reload its state.\n \n v2: split uc_resume and uc_runtime_resume to match the suspend\n functions and to better differentiate the expected state in the 2\n scenarios (Chris)\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730230743.19542-1-daniele.ceraolospurio@intel.com\n\ncommit 1ced169cc1c2f3e054fa14974443383ee02a8b6a\nAuthor: John Crispin \u003cjohn@phrozen.org\u003e\nDate: Tue Jul 30 18:37:01 2019 +0200\n\n mac80211: allow setting spatial reuse parameters from bss_conf\n \n Store the OBSS PD parameters inside bss_conf when bringing up an AP and/or\n when a station connects to an AP. This allows the driver to configure the\n HW accordingly.\n \n Signed-off-by: John Crispin \u003cjohn@phrozen.org\u003e\n Link: https://lore.kernel.org/r/20190730163701.18836-3-john@phrozen.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 6d4dd4ef1a23cd2130d5c256aa698bf7e24d1e31\nAuthor: Johannes Berg \u003cjohannes.berg@intel.com\u003e\nDate: Wed Jul 31 10:58:20 2019 +0200\n\n nl80211: add strict start type\n \n Add a strict start type so all new attributes starting from\n NL80211_ATTR_HE_OBSS_PD are validated strictly.\n \n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 796e90f42b7e52cf1c88e978e1d5ee69c102d85d\nAuthor: John Crispin \u003cjohn@phrozen.org\u003e\nDate: Tue Jul 30 18:37:00 2019 +0200\n\n cfg80211: add support for parsing OBBS_PD attributes\n \n Add the data structure, policy and parsing code allowing userland to send\n the OBSS PD information into the kernel.\n \n Signed-off-by: John Crispin \u003cjohn@phrozen.org\u003e\n Link: https://lore.kernel.org/r/20190730163701.18836-2-john@phrozen.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 52dba8d7d5aba65f818bd27603ae10ebc006ab3b\nAuthor: Karthikeyan Periyasamy \u003cperiyasa@codeaurora.org\u003e\nDate: Wed Jul 24 14:46:10 2019 +0530\n\n mac80211: reject zero MAC address in add station\n \n This came up in fuzz testing, and really we don\u0027t consider\n all-zeroes to be a valid MAC address in most places, so\n also reject it here to avoid confusion later on.\n \n Signed-off-by: Karthikeyan Periyasamy \u003cperiyasa@codeaurora.org\u003e\n Link: https://lore.kernel.org/r/1563959770-21570-1-git-send-email-periyasa@codeaurora.org\n [rewrite commit message]\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 50508d941c180a105fdba802d5af1abf3d93a625\nAuthor: Johannes Berg \u003cjohannes.berg@intel.com\u003e\nDate: Mon Jul 29 16:31:09 2019 +0200\n\n cfg80211: use parallel_ops for genl\n \n Over time, we really need to get rid of all of our global locking.\n One of the things needed is to use parallel_ops. This isn\u0027t really\n the most important (RTNL is much more important) but OTOH we just\n keep adding uses of genl_family_attrbuf() now. Use .parallel_ops to\n disallow this.\n \n Reviewed-By: Denis Kenzior \u003cdenkenz@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190729143109.18683-1-johannes@sipsolutions.net\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 05d610af3e71a782fa28a1351b687da982d208ee\nAuthor: Johannes Berg \u003cjohannes.berg@intel.com\u003e\nDate: Mon Jul 29 18:06:05 2019 +0200\n\n mac80211_hwsim: fill boottime_ns in netlink RX path\n \n Give a proper boottime_ns value for netlink RX to avoid scan\n issues here.\n \n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n Link: https://lore.kernel.org/r/20190729160605.1074-1-johannes@sipsolutions.net\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit f12cac539fe07fe410f2776993e85d343c665141\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Tue Jul 30 15:32:05 2019 +0100\n\n mac80211: add missing null return check from call to ieee80211_get_sband\n \n The return from ieee80211_get_sband can potentially be a null pointer, so\n it seems prudent to add a null check to avoid a null pointer dereference\n on sband.\n \n Addresses-Coverity: (\"Dereference null return\")\n Fixes: 2ab45876756f (\"mac80211: add support for the ADDBA extension element\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190730143205.14261-1-colin.king@canonical.com\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit f277bc0c98a407732d933ac5c53d9cd1224653cb\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jul 31 09:11:26 2019 +0100\n\n drm/i915/selftests: Pass intel_context to igt_spinner\n \n Teach igt_spinner to only use our internal structs, decoupling the\n interface from the GEM contexts. This makes it easier to avoid\n requiring ce-\u003egem_context back references for kernel_context that may\n have them in future.\n \n v2: Lift engine lock to verify_wa() caller.\n v3: Less than v2, but more so\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190731081126.9139-1-chris@chris-wilson.co.uk\n\ncommit 51c98747113e93b6229f12d1a744a51fd59eff3a\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Thu Jul 25 15:26:55 2019 +0200\n\n drm/prime: Ditch gem_prime_res_obj hook\n \n Everyone is just using gem_object-\u003eresv now.\n \n Acked-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Cc: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725132655.11951-5-daniel.vetter@ffwll.ch\n\ncommit b2ad978fd00467e8109918849a817018aaaeed19\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Thu Jul 25 15:26:54 2019 +0200\n\n drm/amdgpu: Fill out gem_object-\u003eresv\n \n That way we can ditch our gem_prime_res_obj implementation. Since ttm\n absolutely needs the right reservation object all the boilerplate is\n already there and we just have to wire it up correctly.\n \n Note that gem/prime doesn\u0027t care when we do this, as long as we do it\n before the bo is registered and someone can call the handle2fd ioctl\n on it.\n \n Aside: ttm_buffer_object.ttm_resv could probably be ditched in favour\n of always passing a non-NULL resv to ttm_bo_init(). At least for gem\n drivers that would avoid having two of these, on in ttm_buffer_object\n and the other in drm_gem_object, one just there for confusion.\n \n Acked-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Cc: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Cc: \"Christian König\" \u003cchristian.koenig@amd.com\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: \"Michel Dänzer\" \u003cmichel.daenzer@amd.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Huang Rui \u003cray.huang@amd.com\u003e\n Cc: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\n Cc: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Cc: Evan Quan \u003cevan.quan@amd.com\u003e\n Cc: Sonny Jiang \u003csonny.jiang@amd.com\u003e\n Cc: Amber Lin \u003cAmber.Lin@amd.com\u003e\n Cc: \"Michał Mirosław\" \u003cmirq-linux@rere.qmqm.pl\u003e\n Cc: Junwei Zhang \u003cJerry.Zhang@amd.com\u003e\n Cc: Thomas Zimmermann \u003ccontact@tzimmermann.org\u003e\n Cc: Samuel Li \u003cSamuel.Li@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725132655.11951-4-daniel.vetter@ffwll.ch\n\ncommit aa77e7f6d5d1e8e22907370fdbfe7a3431b16bac\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Thu Jul 25 15:26:53 2019 +0200\n\n drm/nouveau: Fill out gem_object-\u003eresv\n \n That way we can ditch our gem_prime_res_obj implementation. Since ttm\n absolutely needs the right reservation object all the boilerplate is\n already there and we just have to wire it up correctly.\n \n Note that gem/prime doesn\u0027t care when we do this, as long as we do it\n before the bo is registered and someone can call the handle2fd ioctl\n on it.\n \n Aside: ttm_buffer_object.ttm_resv could probably be ditched in favour\n of always passing a non-NULL resv to ttm_bo_init(). At least for gem\n drivers that would avoid having two of these, on in ttm_buffer_object\n and the other in drm_gem_object, one just there for confusion.\n \n Acked-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Cc: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n Cc: nouveau@lists.freedesktop.org\n Reviewed-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725132655.11951-3-daniel.vetter@ffwll.ch\n\ncommit c549da2d217ecb7c6d385b4f93b7180cc3a59210\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Thu Jul 25 15:26:52 2019 +0200\n\n drm/radeon: Fill out gem_object-\u003eresv\n \n That way we can ditch our gem_prime_res_obj implementation. Since ttm\n absolutely needs the right reservation object all the boilerplate is\n already there and we just have to wire it up correctly.\n \n Note that gem/prime doesn\u0027t care when we do this, as long as we do it\n before the bo is registered and someone can call the handle2fd ioctl\n on it.\n \n Aside: ttm_buffer_object.ttm_resv could probably be ditched in favour\n of always passing a non-NULL resv to ttm_bo_init(). At least for gem\n drivers that would avoid having two of these, on in ttm_buffer_object\n and the other in drm_gem_object, one just there for confusion.\n \n Acked-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Cc: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Cc: \"Christian König\" \u003cchristian.koenig@amd.com\u003e\n Cc: \"David (ChunMing) Zhou\" \u003cDavid1.Zhou@amd.com\u003e\n Cc: amd-gfx@lists.freedesktop.org\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725132655.11951-2-daniel.vetter@ffwll.ch\n\ncommit cb0c43f30ca6a34cf9e796d6ca165668cbc2ec89\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jul 30 17:34:41 2019 +0100\n\n drm/i915: Avoid ce-\u003egem_context-\u003ei915\n \n My plan for the future is to have kernel contexts not to have a GEM\n context backpointer (as they will not belong to any GEM context). In a\n few places, we use ce-\u003egem_context to simply obtain the i915 backpointer,\n for which we can use ce-\u003eengine-\u003ei915 instead.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730163441.16477-1-chris@chris-wilson.co.uk\n\ncommit 9d4742136a2667774805b774eae12810857ad96e\nAuthor: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nDate: Wed Jul 31 16:05:57 2019 +1000\n\n xilinx_uartps.c: suppress \"may be used uninitialised\" warning\n \n A powerpc allyesconfig build produces this warning:\n \n In file included from include/linux/radix-tree.h:16,\n from include/linux/idr.h:15,\n from include/linux/kernfs.h:13,\n from include/linux/sysfs.h:16,\n from include/linux/kobject.h:20,\n from include/linux/device.h:16,\n from include/linux/platform_device.h:13,\n from drivers/tty/serial/xilinx_uartps.c:16:\n drivers/tty/serial/xilinx_uartps.c: In function \u0027cdns_uart_console_write\u0027:\n include/linux/spinlock.h:288:3: warning: \u0027flags\u0027 may be used uninitialized in this function [-Wmaybe-uninitialized]\n _raw_spin_unlock_irqrestore(lock, flags); \\\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/tty/serial/xilinx_uartps.c:1197:16: note: \u0027flags\u0027 was declared here\n unsigned long flags;\n ^~~~~\n \n It looks like gcc just can\u0027t track the relationship between \"locked\"\n and \"flags\", and it is obvious that \"flags\" won\u0027t be used when \"locked\"\n is zero, so the simplest thing is to initialise flags.\n \n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: Michal Simek \u003cmichal.simek@xilinx.com\u003e\n Signed-off-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Link: https://lore.kernel.org/r/20190731160557.6a09c3e1@canb.auug.org.au\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 9a675a13c3bbb6ff46414b4ffaca5fa6bb54b62a\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Wed Jul 31 07:28:04 2019 +0200\n\n Revert \"staging: kpc2000: Convert put_page() to put_user_page*()\"\n \n This reverts commit 8e7cb7352ec85e9e4fbbd7bfe6c5c5a6806f70e3.\n \n Bharath writes:\n Could you drop this patch from the staging-next tree? This is\n because John is making some changes to the put_user_page*()\n functions. He has submitted a patch recently removing\n put_user_page_dirty() function which is being used in this\n patch. This might break the kernel build if John\u0027s patch gets\n merged in.\n \n I ll submit a patch once the put_user_page*() apis are fixed.\n \n Reported-by: Bharath Vedartham \u003clinux.bhar@gmail.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 59de58f84f7f561fe67b2d78a6a2cae6c0b50c24\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Thu Jul 25 16:40:51 2019 +0800\n\n drm/amd/powerplay: determine the features to enable by pptable only\n \n Per current logics, the features to enable are determined together\n by driver and pptable. This is not efficient in co-debug with\n firmware team.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 861324983d4f7fac6a66ddc021511df7d4193e2b\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Thu Jul 25 17:22:01 2019 +0800\n\n drm/amdgpu: correct irq type used for sdma ecc\n \n we should pass irq type, instead of irq client id,\n to irq_get/put interface\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b4af964e75c4163fe3baf98193495f7921a4b3b7\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Tue Jul 30 22:52:37 2019 -0500\n\n drm/amd/powerplay: make power limit retrieval as asic specific\n \n The power limit retrieval should be done per asic. Since we may\n need to lookup in the pptable and that\u0027s really asic specific.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1f23cadbe0775559b3152a4f1fcb566d8cf6f571\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Tue Jul 23 20:28:14 2019 +0800\n\n drm/amd/powerplay: correct arcturus current clock level calculation\n \n There may be 1Mhz delta between target and actual frequency. That\n should be taken into consideration for current level check.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 60d435b73db607df560f3df04fe5f4ba96e37116\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Tue Jul 23 17:30:35 2019 +0800\n\n drm/amd/powerplay: support UMD PSTATE settings on arcturus\n \n Enable arcturus UMD PSTATE support.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4bf76e60b9c7737ea9f585795b6ca97bb03100b8\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Tue Jul 23 11:42:24 2019 +0800\n\n drm/amd/powerplay: fix arcturus real-time clock frequency retrieval\n \n Make sure we can still get the accurate gfxclk/uclk/socclk frequency\n even on dpm disabled.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 790ef68afcf5134ddc1354c29ef5dddb99a45bad\nAuthor: Kevin Wang \u003ckevin1.wang@amd.com\u003e\nDate: Fri Jul 19 16:06:29 2019 +0800\n\n drm/amd/powerplay: remove redundancy debug log in smu\n \n remove redundacy debug log in smu.\n eg:\n [ 6897.969447] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled\n [ 6897.969448] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled\n [ 6897.969448] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled\n [ 6899.024114] amdgpu: [powerplay] Unsupported SMU message: 38\n [ 6899.024151] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled\n [ 6899.024151] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled\n [ 6899.024152] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled\n [ 6900.078296] amdgpu: [powerplay] Unsupported SMU message: 38\n [ 6900.078332] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled\n [ 6900.078332] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled\n [ 6900.078333] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled\n [ 6901.133230] amdgpu: [powerplay] Unsupported SMU message: 38\n \n Signed-off-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8a856ced35d51c988497856dcf6f76174305e416\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Mon Jul 22 17:03:02 2019 +0800\n\n drm/amd/powerplay: correct the bitmask used in arcturus\n \n Those bitmask prefixed by \"SMU_\" should be used.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 55bf7e6243e72de5d042e76d5bde49f1984530c5\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Mon Jul 22 16:26:04 2019 +0800\n\n drm/amd/powerplay: add missing arcturus feature maps\n \n Add missing feature maps for arcturus.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d427cf8f7f9932aac3908b38ab25d4c4c7c91315\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Mon Jul 22 12:09:38 2019 +0800\n\n drm/amd/powerplay: support fan speed retrieval on arcturus\n \n Support arcturus fan speed retrieval.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 631807f091aeb8a3c9f5c7271753a1b257446c7e\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Fri Jul 19 17:18:34 2019 +0800\n\n drm/amd/powerplay: support real-time clock retrieval on arcturus\n \n Enable arcturus real-time clock retrieval.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ba74c8bf889cbfc02c90525a859bc65ddafc1368\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Mon Jul 29 13:18:37 2019 -0500\n\n drm/amd/powerplay: support sensor reading on arcturus\n \n Support sensor reading for gpu loading, power and\n temperatures.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 832a7062a06016d198039a566dc71073b7155881\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Mon Jul 22 15:55:52 2019 +0800\n\n drm/amd/powerplay: init arcturus SMU metrics table on bootup\n \n Initialize arcturus SMU metrics table.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1f96ecef6f0137791b5a37e005e65c42ce927e62\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Mon Jul 22 10:42:29 2019 +0800\n\n drm/amd/powerplay: correct UVD/VCE/VCN power status retrieval\n \n VCN should be used for Vega20 later ASICs while UVD and VCE\n are for previous ASICs.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5fa790f6c936c4705dea5883fa12da9e017ceb4f\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Mon Jul 22 10:27:21 2019 +0800\n\n drm/amd/powerplay: correct Navi10 VCN powergate control (v2)\n \n No VCN DPM bit check as that\u0027s different from VCN PG. Also\n no extra check for possible double enablement/disablement\n as that\u0027s already done by VCN.\n \n v2: check return value of smu_feature_set_enabled\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit bf2bf52383a09256e11278e7bcb67dcd912078c7\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Mon Jul 22 09:57:27 2019 +0800\n\n drm/amd/powerplay: support VCN powergate status retrieval for SW SMU\n \n Commonly used for VCN powergate status retrieval for SW SMU.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ab9e314886adef86ca6bf7b6c485abf427209295\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Mon Jul 22 09:55:36 2019 +0800\n\n drm/amd/powerplay: support VCN powergate status retrieval on Raven\n \n Enable VCN powergate status report on Raven.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9829e3d89b6ea9027542d094cdb8a434eef4b3aa\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Mon Jul 22 09:51:59 2019 +0800\n\n drm/amd/powerplay: add new sensor type for VCN powergate status\n \n VCN is widely used in new ASICs and different from tranditional\n UVD and VCE.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7d0e6329dfdcfe48311f8888d6a8dfa73bee00a9\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Jul 16 15:21:54 2019 +0800\n\n drm/amdgpu: update more sdma instances irq support\n \n Update for sdma ras ecc_irq and other minors.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9d4d7236efad8cf47e989be0c80dd37d68d55a5c\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Mon Jul 15 18:00:50 2019 +0800\n\n drm/amd/include: adjust base offset of SMUIO and THM for Arcturus\n \n Arcturus has different _BASE_IDX value in some HWIP_offset.h. To make source\n files like smu_v11_0.c and soc15.c that include HWIP_offset.h of Vega20\n reusable for Arcturus, align this base offset with Vega20.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 3ff101b8ab78e115feb52645dac35b3634ee5a77\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Wed Jul 17 09:34:13 2019 +0800\n\n drm/amd/powerplay: hold on the arcturus gfx dpm support in driver\n \n As for now, only \"Prefetcher\" is guarded to be working from\n SMU firmware.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7c16d24abe3525873ec7e62ec366d06022dc6367\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Tue Jul 16 11:03:10 2019 +0800\n\n drm/amdgpu: correct VCN powergate routine for acturus\n \n Arcturus VCN should powergate in the way as Navi.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Le Ma \u003cLe.Ma@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit fe089e1dd797c2c548425dc4756ee02e3682d78e\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Fri Jul 12 16:53:28 2019 +0800\n\n drm/amd/powerplay: enable arcturus powerplay\n \n Arcturus powerplay is ready to use.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit cca4fafc09042cc63e7e277ad8d40b03307dfe3d\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Fri Jul 12 16:50:52 2019 +0800\n\n drm/amd/powerplay: initialize arcturus MP1 and THM base address\n \n Initialize base address for those IPs which are used in powerplay.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 22e1831734716f9e3e4fb1f88060f0efaa59fc6f\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Tue Jul 30 23:30:07 2019 -0500\n\n drm/amd/powerplay: enable SW SMU routine support for arcturus\n \n Enable arcturus SW SMU routines.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a94235af11b3f7d33eed56c7aef1864ea6e16db4\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Mon Jul 29 12:43:28 2019 -0500\n\n drm/amd/powerplay: update arcturus_ppt.c/h V3\n \n Arcturus ASIC specific powerplay interfaces.\n \n V2: correct SMU msg naming\n drop unnecessary debugs\n \n V3: rebase (Alex)\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Le Ma \u003cLe.Ma@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c8893d5ce716459db532c72b9814b355abe23ac1\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Fri Jul 12 16:28:02 2019 +0800\n\n drm/amd/powerplay: update arcturus_ppsmc.h\n \n Correct header and fix typo.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Le Ma \u003cLe.Ma@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 22808306f2b2704be87378bf58e009a61598a5a7\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Fri Jul 12 16:24:34 2019 +0800\n\n drm/amd/powerplay: update smu11_driver_if_arcturus.h\n \n It guides how driver should interface with SMU in arcturus.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7c8bcaf4085af03365e86f11dba6a8363ef87c07\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Tue Jul 30 22:50:14 2019 -0500\n\n drm/amd/powerplay: add SW SMU interface for dumping pptable out (v2)\n \n This is especially useful in early bring up phase.\n \n v2: disabled by default (Alex)\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Le Ma \u003cLe.Ma@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4c35e77865a9037c32b0354663d23c33b08ae188\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Wed Jul 10 09:29:57 2019 +0800\n\n drm/amd/powerplay: add smcdpminfo table v4_6 support\n \n New smcdpminfo table used in arcturus.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1faa3b805473d7f4197b943419781d9fd21e4352\nAuthor: Jay Cornwall \u003cjay.cornwall@amd.com\u003e\nDate: Sun Jul 28 16:00:59 2019 -0500\n\n drm/amdkfd: Save/restore vcc on gfx10\n \n VCC moved out of user SGPR allocation in gfx10. It\u0027s now stored\n in SGPRs 106-107.\n \n Also fixes incorrect SGPR read offsets.\n \n Cc: Shaoyun Liu \u003cshaoyun.liu@amd.com\u003e\n Signed-off-by: Jay Cornwall \u003cjay.cornwall@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: shaoyunl \u003cshaoyun.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f9e346aba193d82286dbb3aab358906a7416568b\nAuthor: Jay Cornwall \u003cjay.cornwall@amd.com\u003e\nDate: Sun Jul 28 15:25:05 2019 -0500\n\n drm/amdkfd: Save/restore flat_scratch_lo/hi on gfx10\n \n These moved from SGPRs in gfx9 to HWREG in gfx10.\n \n Cc: Shaoyun Liu \u003cshaoyun.liu@amd.com\u003e\n Signed-off-by: Jay Cornwall \u003cjay.cornwall@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: shaoyunl \u003cshaoyun.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7ce55e0b6fbbbddf4984e8343fbb4254afc78dc9\nAuthor: Jay Cornwall \u003cjay.cornwall@amd.com\u003e\nDate: Sun Jul 28 15:24:40 2019 -0500\n\n drm/amdkfd: Fix gfx10 wave64 VGPR context restore\n \n Copy/paste error, first 4 VGPRs are separated by 64 dwords (256 bytes).\n \n Cc: Shaoyun Liu \u003cshaoyun.liu@amd.com\u003e\n Signed-off-by: Jay Cornwall \u003cjay.cornwall@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: shaoyunl \u003cshaoyun.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f16d523f9d83a73c274fb97a8fed17114d67b30e\nAuthor: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\nDate: Tue Jul 30 09:45:33 2019 -0400\n\n drm/amd/display: Support uclk switching for DCN2\n \n [Why]\n We were previously forcing the uclk for every state to max and reducing\n the switch time to prevent uclk switching from occuring. This workaround\n was previously needed in order to avoid hangs + underflow under certain\n display configurations.\n \n Now that DC has the proper fix complete we can drop the hacks and\n improve power for most display configurations.\n \n [How]\n We still need the function pointers hooked up to grab the real uclk\n states from pplib. The rest of the prior hack can be reverted.\n \n The key requirements here are really just DC support, updated firmware,\n and support for disabling p-state support when needed in pplib/smu.\n \n When these requirements are met uclk switching works without underflow\n or hangs.\n \n Fixes: 02316e963a5a (\"drm/amd/display: Force uclk to max for every state\")\n \n Signed-off-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit fb6959ae50176758a073687dbb081d26521f4576\nAuthor: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\nDate: Tue Jul 30 09:08:34 2019 -0400\n\n drm/amd/display: Embed DCN2 SOC bounding box\n \n [Why]\n In order to support uclk switching on NV10 the SOC bounding box\n needs to be updated.\n \n [How]\n We currently read the constants from the gpu info FW, but supporting\n workarounds in DC for different versions of the FW adds additional\n complexity to the codebase.\n \n NV10 has been released so it\u0027s cleanest to keep the bounding box and\n source code in sync by embedding the bounding box like we do for\n other ASICs.\n \n Fixes: 02316e963a5a (\"drm/amd/display: Force uclk to max for every state\")\n \n Signed-off-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1a2c29bce06083e0f53f3a328f0bb43a2856b622\nAuthor: Wang Xiayang \u003cxywang.sjtu@sjtu.edu.cn\u003e\nDate: Sat Jul 27 17:30:30 2019 +0800\n\n drm/amdgpu: fix a potential information leaking bug\n \n Coccinelle reports a path that the array \"data\" is never initialized.\n The path skips the checks in the conditional branches when either\n of callback functions, read_wave_vgprs and read_wave_sgprs, is not\n registered. Later, the uninitialized \"data\" array is read\n in the while-loop below and passed to put_user().\n \n Fix the path by allocating the array with kcalloc().\n \n The patch is simplier than adding a fall-back branch that explicitly\n calls memset(data, 0, ...). Also it does not need the multiplication\n 1024*sizeof(*data) as the size parameter for memset() though there is\n no risk of integer overflow.\n \n Signed-off-by: Wang Xiayang \u003cxywang.sjtu@sjtu.edu.cn\u003e\n Reviewed-by: Chunming Zhou \u003cdavid1.zhou@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 649412069501c1710f42ad50ab420e2fd5eaea70\nAuthor: Christian König \u003cchristian.koenig@amd.com\u003e\nDate: Tue Jul 30 11:17:03 2019 +0200\n\n drm/amdgpu: fix error handling in amdgpu_cs_process_fence_dep\n \n We always need to drop the ctx reference and should check\n for errors first and then dereference the fence pointer.\n \n Signed-off-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Chunming Zhou \u003cdavid1.zhou@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6e92e156aacce9b5d2f8003b92d71e676e5e9139\nAuthor: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\nDate: Mon Jul 29 17:51:55 2019 +0800\n\n drm/amdgpu/powerplay: provide the interface to disable uclk switch for DAL\n \n provide the interface for DAL to disable uclk switch on navi10.\n in this case, the uclk will be fixed to maximum.\n this is a workaround when display configuration causes underflow issue.\n \n Signed-off-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c74dbe44eacf00a5ccc229b5cc340a9b7f6851a0\nAuthor: Thong Thai \u003cthong.thai@amd.com\u003e\nDate: Thu Jul 25 11:26:56 2019 -0400\n\n drm/amd/amdgpu/vcn_v2_0: Move VCN 2.0 specific dec ring test to vcn_v2_0\n \n VCN 2.0 firmware now requires a packet start command to be sent before\n any other decode ring buffer command.\n \n Signed-off-by: Thong Thai \u003cthong.thai@amd.com\u003e\n Reviewed-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 3207dcf3afd6bffe5e887d483e7071616b97bd8f\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Fri Jul 26 14:14:15 2019 -0500\n\n drm/amdgpu/gfx10: update golden settings for navi14\n \n Updated settings for hw team.\n \n Reviewed-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 98eb03bbf0175f009a74c80ac12b91a9680292f4\nAuthor: Kevin Wang \u003ckevin1.wang@amd.com\u003e\nDate: Thu Jul 25 11:47:44 2019 +0800\n\n drm/amd/powerplay: implment sysfs feature status function in smu\n \n 1. Unified feature enable status format in sysfs\n 2. Rename ppfeature to pp_features to adapt other pp sysfs node name\n 3. this function support all asic, not asic related function.\n \n Signed-off-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Acked-by: Rui Huang \u003cray.huang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 26dd668155b24eb24401aed9d73e2f9c7b694ea6\nAuthor: Kevin Wang \u003ckevin1.wang@amd.com\u003e\nDate: Thu Jul 25 11:57:25 2019 +0800\n\n drm/amd/powerplay: move smu_feature_update_enable_state to up level\n \n this function is not ip or asic related function,\n so move it to top level as public api in smu.\n \n Signed-off-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit cb33363d0e85341c8781d9095341a981ab410a80\nAuthor: Kevin Wang \u003ckevin1.wang@amd.com\u003e\nDate: Thu Jul 25 11:08:42 2019 +0800\n\n drm/amd/powerplay: add smu feature name support\n \n add smu_get_feature_name support in smu.\n \n Signed-off-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6b294793e3846988ca1234c324ae256b25cee9cf\nAuthor: Kevin Wang \u003ckevin1.wang@amd.com\u003e\nDate: Thu Jul 25 10:32:48 2019 +0800\n\n drm/amd/powerplay: add smu message name support\n \n add smu_get_message_name support in smu.\n \n Signed-off-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 0ba5eda81a82719c229c819929e8297559522429\nAuthor: Kevin Wang \u003ckevin1.wang@amd.com\u003e\nDate: Thu Jul 25 09:59:46 2019 +0800\n\n drm/amd/powerplay: move smu types to smu_types.h\n \n move some enum type (message, feature, clock) to smu_types.h.\n these types is too long in amdgpu_smu.h, and not clearly.\n \n Signed-off-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2c8973180392d1835d07eafd361e821f4aa28dfa\nAuthor: Joseph Greathouse \u003cJoseph.Greathouse@amd.com\u003e\nDate: Fri Jul 26 15:52:05 2019 -0500\n\n drm/amdgpu: Default disable GDS for compute+gfx\n \n Units in the GDS block default to allowing all VMIDs access to all\n entries. Disable shader access to the GDS, GWS, and OA blocks from all\n compute and gfx VMIDs by default. For compute, HWS firmware will set\n up the access bits for the appropriate VMID when a compute queue\n requires access to these blocks.\n The driver will handle enabling access on-demand for graphics VMIDs.\n \n Leaving VMID0 with full access because otherwise HWS cannot save or\n restore values during task switch.\n \n v2: Fixed code and comment styling.\n \n Signed-off-by: Joseph Greathouse \u003cJoseph.Greathouse@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 333fe325febabe3df2bc3019d4b97f879d8cef73\nAuthor: Thong Thai \u003cthong.thai@amd.com\u003e\nDate: Thu Jul 25 11:21:58 2019 -0400\n\n drm/amd/amdgpu/vcn_v2_0: Mark RB commands as KMD commands\n \n Sets the CMD_SOURCE bit for VCN 2.0 Decoder Ring Buffer commands. This\n bit was previously set by the RBC HW on older firmware. Newer firmware\n uses a SW RBC and this bit has to be set by the driver.\n \n Signed-off-by: Thong Thai \u003cthong.thai@amd.com\u003e\n Reviewed-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d3b9f39d8417ee2f2cd87b5e5410015ce6f78491\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Fri Jul 26 11:04:39 2019 -0500\n\n drm/amdgpu/display: fix the build without CONFIG_DRM_AMD_DC_DSC_SUPPORT\n \n Some code was missing the CONFIG_DRM_AMD_DC_DSC_SUPPORT guard.\n \n Reviewed-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f2bd8a0ed7e79579d61cea01bab2dfb09099d379\nAuthor: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\nDate: Fri Jul 26 09:24:35 2019 -0400\n\n drm/amdgpu: Fix amdgpu_display_supported_domains logic.\n \n Add restriction to dissallow GTT domain if the relevant BO\n doesn\u0027t have USWC flag set to avoid the APU hang scenario.\n \n Signed-off-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 354e6e14ef947f07055d3570b4bd7a33196b57f6\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Thu Jul 25 22:28:58 2019 -0500\n\n drm/amdgpu/powerplay: use proper revision id for navi\n \n The PCI revision id determines the sku.\n \n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a3a09142f43c456fff7ddbc0ac867af8979a368c\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Thu Jul 25 11:44:59 2019 -0500\n\n drm/amdgpu: put the SMC into the proper state on reset/unload\n \n When doing a GPU reset or unloading the driver, we need to\n put the SMU into the apprpriate state for the re-init after\n the reset or unload to reliably work.\n \n I don\u0027t think this is necessary for BACO because the SMU actually\n controls the BACO state to it needs to be active.\n \n For suspend (S3), the asic is put into D3 so the SMU would be\n powered down so I don\u0027t think we need to put the SMU into\n any special state.\n \n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1c074a63834e7414031f33c23efd87bbb8c25cea\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Thu Jul 25 10:55:57 2019 -0500\n\n drm/amdgpu/powerplay: add set_mp1_state for vega12\n \n This sets the SMU into the proper state for various\n operations (shutdown, unload, GPU reset, etc.).\n \n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e254102d5079f83376b6b6b227e546bcde89683c\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Thu Jul 25 10:55:25 2019 -0500\n\n drm/amdgpu/powerplay: add set_mp1_state for vega10\n \n This sets the SMU into the proper state for various\n operations (shutdown, unload, GPU reset, etc.).\n \n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a906277d22f790418e68a2ad0feba6617b165d40\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Thu Jul 25 10:43:28 2019 -0500\n\n drm/amdgpu/powerplay: add set_mp1_state for vega20\n \n This sets the SMU into the proper state for various\n operations (shutdown, unload, GPU reset, etc.).\n \n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 479baeacd82841e44cd860d84fa66fe6240f1814\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Tue Jul 30 21:27:03 2019 -0500\n\n drm/amdgpu/powerplay: return success if set_mp1_state is not set\n \n Some asics (APUs) don\u0027t have this callback so we want to return\n success. Avoids spurious error messages on APUs.\n \n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a2c28e34f8c42e35f5f2990d558d367152e63c27\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Thu Jul 25 09:41:53 2019 -0500\n\n drm/amdgpu/powerplay: add a new interface to set the mp1 state\n \n This is required for certain cases such as various GPU resets\n (mode1, mode2), BACO, shutdown, unload, etc. to put the SMU into\n the appropriate state for when the hw is re-initialized.\n \n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2ddc6c3ef9d5f22ad004aad2e28a32810fe9c407\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Tue Jul 23 23:48:21 2019 -0500\n\n drm/amdgpu: add reset_method asic callback for navi\n \n Navi uses either mode1 or baco depending on various\n conditions.\n \n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ee360c0b7cc3148ff8216286f53f8b3577b2b1a0\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Tue Jul 23 23:47:06 2019 -0500\n\n drm/amdgpu: add reset_method asic callback for soc15\n \n APUs only support mode2 reset. dGPUs use either mode1 or\n baco depending on various conditions.\n \n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9bc1932f5ce5cb601278b8a37d36d7939f3e2482\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Tue Jul 23 23:46:12 2019 -0500\n\n drm/amdgpu: add reset_method asic callback for vi\n \n VI always uses the legacy pci based reset.\n \n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6d0f50dafed4771121f674aef2db2c13ef526ad0\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Tue Jul 23 23:45:39 2019 -0500\n\n drm/amdgpu: add reset_method asic callback for cik\n \n CIK always uses the legacy pci based reset.\n \n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit dd81eede7703fd7349be84f1830d7a1339918e38\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Tue Jul 23 23:44:54 2019 -0500\n\n drm/amdgpu: add reset_method asic callback for si\n \n SI always uses the legacy pci based reset.\n \n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 0cf3c64f294c1a03e7e1d826c6c9d1b126216c02\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Tue Jul 23 23:27:21 2019 -0500\n\n drm/amdgpu: add an asic callback to determine the reset method\n \n Sometimes the driver may have to behave differently depending\n on the method we are using to reset the GPU.\n \n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4d7fd9e20b0784b07777728316da5bcc13f9f2ab\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Thu Jul 25 10:12:42 2019 +0800\n\n drm/amd/powerplay: enable SW SMU reset functionality\n \n Move SMU irq handler register to sw_init as that\u0027s totally\n software related. Otherwise, it will prevent SMU reset working.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f0d2a7dc1154ed680a8422916ab6a38860800de4\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Thu Jul 25 12:10:34 2019 +0800\n\n drm/amd/powerplay: fix null pointer dereference around dpm state relates\n \n DPM state relates are not supported on the new SW SMU ASICs. But still\n it\u0027s not OK to trigger null pointer dereference on accessing them.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit fcd90fee8ac22da3bce1c6652cf36bc24e7a0749\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Wed Jul 24 14:06:09 2019 +0800\n\n drm/amd/powerplay: minor fixes around SW SMU power and fan setting\n \n Add checking for possible invalid input and null pointer. And\n drop redundant code.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1c4259159132ae4ceaf7c6db37a6cf76417f73d9\nAuthor: Shirish S \u003cshirish.s@amd.com\u003e\nDate: Tue Jul 16 14:49:48 2019 +0530\n\n drm/amd/display: enable S/G for RAVEN chip\n \n enables gpu_vm_support in dm and adds\n AMDGPU_GEM_DOMAIN_GTT as supported domain\n \n v2:\n Move BO placement logic into amdgpu_display_supported_domains\n \n v3:\n Use amdgpu_bo_validate_uswc in amdgpu_display_supported_domains.\n \n v4:\n amdgpu_bo_validate_uswc moved to sepperate patch.\n \n Signed-off-by: Shirish S \u003cshirish.s@amd.com\u003e\n Signed-off-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Reviewed-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ddcb7fc62f4be99faedfa1764c971a2f31468962\nAuthor: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\nDate: Wed Jul 24 11:09:03 2019 -0400\n\n drm/amdgpu: Add check for USWC support for amdgpu_display_supported_domains\n \n This verifies we don\u0027t add GTT as allowed domain for APUs when USWC\n is disabled.\n \n Signed-off-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Acked-by: Michel Dänzer \u003cmichel.daenzer@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 3d1b8ec76b8f4f91641fd87cc19cd980426a2060\nAuthor: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\nDate: Wed Jul 24 10:04:27 2019 -0400\n\n drm/amdgpu: Create helper to clear AMDGPU_GEM_CREATE_CPU_GTT_USWC\n \n Move the logic to clear AMDGPU_GEM_CREATE_CPU_GTT_USWC in\n amdgpu_bo_do_create into standalone helper so it can be reused\n in other functions.\n \n v4:\n Switch to return bool.\n \n v5: Fix typos.\n \n Signed-off-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Acked-by: Michel Dänzer \u003cmichel.daenzer@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e4c4073b0139d055d43a9568690fc560aab4fa5c\nAuthor: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\nDate: Mon Jul 15 18:04:08 2019 -0400\n\n drm/amdgpu: Fix hard hang for S/G display BOs.\n \n HW requires for caching to be unset for scanout BO\n mappings when the BO placement is in GTT memory.\n Usually the flag to unset is passed from user mode\n but for FB mode this was missing.\n \n v2:\n Keep all BO placement logic in amdgpu_display_supported_domains\n \n Suggested-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Tested-by: Shirish S \u003cshirish.s@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 24f9aacfb0fbe724c94f6ffe24cc518bfdca4b1d\nAuthor: Jonathan Kim \u003cjonathan.kim@amd.com\u003e\nDate: Thu Jul 11 13:14:02 2019 -0400\n\n drm/amdgpu: adding xgmi error monitoring\n \n monitor xgmi errors via mc pie status through fica registers.\n \n Signed-off-by: Jonathan Kim \u003cJonathan.Kim@amd.com\u003e\n Reviewed-by: Kent Russell \u003cKent.Russell@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 64671c0fdc9193978cb93aaa79965e45b3cce437\nAuthor: Jonathan Kim \u003cjonathan.kim@amd.com\u003e\nDate: Tue Jul 9 15:47:57 2019 -0400\n\n drm/amdgpu: add perfmon and fica atomics for df\n \n adding perfmon and fica atomic operations to adhere to data fabrics finite\n state machine requirements for indirect register access.\n \n Signed-off-by: Jonathan Kim \u003cJonathan.Kim@amd.com\u003e\n Reviewed-by: Kent Russell \u003cKent.Russell@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 306fc9c568518d18821adfa050147ab1a9241fbd\nAuthor: Jay Cornwall \u003cjay.cornwall@amd.com\u003e\nDate: Wed Jul 24 12:26:08 2019 -0500\n\n drm/amdkfd: Remove dead code from gfx8/gfx9 trap handlers\n \n Signed-off-by: Jay Cornwall \u003cjay.cornwall@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a36e89674000a2663421da57391ad5897081a2e1\nAuthor: Jay Cornwall \u003cjay.cornwall@amd.com\u003e\nDate: Wed Jul 24 12:23:42 2019 -0500\n\n drm/amdkfd: Replace gfx10 trap handler with correct branch\n \n Previously submitted code was taken from an incorrect branch and\n was non-functional.\n \n Cc: Oak Zeng \u003coak.zeng@amd.com\u003e\n Signed-off-by: Jay Cornwall \u003cjay.cornwall@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-By: Oak Zeng \u003coak.zeng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7c2eaf5cdb8797bae792fb3cd96ebc6c7f3c53e7\nAuthor: Jay Cornwall \u003cjay.cornwall@amd.com\u003e\nDate: Mon Jul 22 19:21:13 2019 -0500\n\n drm/amdkfd: Fix lost single step exceptions in gfx9 trap handler\n \n If the trap is entered due to MODE.DEBUG_EN\u003d1 and SAVECTX is raised\n concurrently the handler cannot identify the source of the exception.\n This causes the debugger to lose single step exception notification\n when a context save request arrives at the same time.\n \n When MODE.DEBUG_EN\u003d1 and STATUS.HALT\u003d0 (exception not already handled)\n jump to the second-level trap handler upon entering the trap. The\n second-level trap will set STATUS.HALT\u003d1 and return to the shader.\n If SAVECTX was raised then control flow will return to the trap, which\n will then handle the context save request.\n \n Cc: Tony Tye \u003ctony.tye@amd.com\u003e\n Cc: Laurent Morichetti \u003claurent.morichetti@amd.com\u003e\n Cc: Qingchuan Shi \u003cqingchuan.shi@amd.com\u003e\n Signed-off-by: Jay Cornwall \u003cjay.cornwall@amd.com\u003e\n Reviewed-by: Laurent Morichetti \u003claurent.morichetti@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8c7a5d9e6f141aa48473f15ac271c75688404ba8\nAuthor: Jay Cornwall \u003cjay.cornwall@amd.com\u003e\nDate: Tue Jul 23 14:40:07 2019 -0500\n\n drm/amdkfd: Use SQC when TCP would fail in gfx9 context save.\n \n When a wavefront raises TRAPSTS.XNACK_ERROR with STATUS.ALLOW_REPLAY\u003d0\n subsequent memory instructions have undefined behavior. In practice\n SQC stores continue to work but TCP stores do not.\n \n Context save is permitted to fail after XNACK error because the\n wavefront will be halted and subsequently terminated. However the\n debugger has an interest in retrieving the wavefront VGPR/LDS state.\n \n Detect the out-of-spec case and use SQC stores during context save\n in place of TCP stores.\n \n Signed-off-by: Jay Cornwall \u003cjay.cornwall@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5f4814deab509e4d0a52e4a0b016d49a75b2cf7c\nAuthor: tiancyin \u003ctianci.yin@amd.com\u003e\nDate: Tue Jul 16 18:25:01 2019 +0800\n\n drm/amdgpu/gmc10: fix pte mytpe field error for navi14\n \n navi14 share same PTE format with navi10.\n \n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: tiancyin \u003ctianci.yin@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6913848087e29103f7062376d0ebf7efcaf1a26b\nAuthor: James Zhu \u003cJames.Zhu@amd.com\u003e\nDate: Tue Jul 23 16:45:19 2019 -0400\n\n drm/amdgpu: use VCN firmware offset for cache window\n \n Since we are using the signed FW now, and also using PSP firmware loading,\n but it\u0027s still potential to break driver when loading FW directly\n instead of PSP, so we should add offset.\n \n Signed-off-by: James Zhu \u003cJames.Zhu@amd.com\u003e\n Reviewed-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 668a9469f11470868ad4577d75894f8812919314\nAuthor: Kevin Wang \u003ckevin1.wang@amd.com\u003e\nDate: Tue Jul 23 19:56:52 2019 +0800\n\n drm/amd/powerplay: fix temperature granularity error in smu11\n \n in this patch,\n drm/amd/powerplay: add callback function of get_thermal_temperature_range\n the driver missed temperature granularity change on other temperature.\n \n Signed-off-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 780f3a9c5b9f739d77cc15fac58a97c3ec6cad99\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Tue Jul 23 16:23:28 2019 +0800\n\n drm/amd/powerplay: some cosmetic fixes\n \n Drop redundant check, duplicate check, duplicate setting\n and fix the return value.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d3a593e97ecdff6bbebfd8b83079ffc5a332dccf\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Tue Jul 23 14:27:20 2019 +0800\n\n drm/amd/powerplay: no pptable transfer and dpms enabled with \"dpm\u003d0\"\n \n Honor the \u0027dpm\u0027 module parameter setting on SW SMU routine as what\n we did on previous ASICs. SMU FW loading is still proceeded even\n with \"dpm\u003d0\".\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 87abff77880da827b8c9acb37e241999183b406f\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Tue Jul 23 15:23:12 2019 +0100\n\n drm/amd/display: fix a missing null check on a failed kzalloc\n \n Currently the allocation of config may fail and a null pointer\n dereference on config can occur. Fix this by added a null\n check on a failed allocation of config.\n \n Addresses-Coverity: (\"Dereference null return\")\n Fixes: c2cd9d04ecf0 (\"drm/amd/display: Hook up calls to do stereo mux and dig programming to stereo control interface\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 59d788b18771f5f0f198e8da3cd4f38dfbfa64ba\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 19:10:08 2019 +0800\n\n drm/radeon: Use dev_get_drvdata where possible\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ef1de3618125e4a4d6511553f8b527f58829a8e6\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 17:04:22 2019 +0800\n\n drm/amd/display: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 911d8b3069ce87bcc874503d8f0b17b019e6e502\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 17:04:50 2019 +0800\n\n drm/amdgpu: Use dev_get_drvdata where possible\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 0f0727d971f6fdf8f1077180d495ddb9928f0c8b\nAuthor: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nDate: Mon Jul 22 15:31:05 2019 -0700\n\n drm/amd/display: readd -msse2 to prevent Clang from emitting libcalls to undefined SW FP routines\n \n arch/x86/Makefile disables SSE and SSE2 for the whole kernel. The\n AMDGPU drivers modified in this patch re-enable SSE but not SSE2. Turn\n on SSE2 to support emitting double precision floating point instructions\n rather than calls to non-existent (usually available from gcc_s or\n compiler_rt) floating point helper routines for Clang.\n \n This was originally landed in:\n commit 10117450735c (\"drm/amd/display: add -msse2 to prevent Clang from emitting libcalls to undefined SW FP routines\")\n but reverted in:\n commit 193392ed9f69 (\"Revert \"drm/amd/display: add -msse2 to prevent Clang from emitting libcalls to undefined SW FP routines\"\")\n due to bugreports from GCC builds. Add guards to only do so for Clang.\n \n Link: https://bugs.freedesktop.org/show_bug.cgi?id\u003d109487\n Link: https://github.com/ClangBuiltLinux/linux/issues/327\n \n Suggested-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Suggested-by: Sami Tolvanen \u003csamitolvanen@google.com\u003e\n Signed-off-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4cab85afe9d7903936ef86ef143e37440cb8fc98\nAuthor: Kent Russell \u003ckent.russell@amd.com\u003e\nDate: Tue Jul 23 10:18:01 2019 -0400\n\n drm/amdkfd: Fix byte align on VegaM\n \n This was missed during the addition of VegaM support\n \n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Kent Russell \u003ckent.russell@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7a81637105345dce2bae7bef69125f0d40570a36\nAuthor: Kevin Wang \u003ckevin1.wang@amd.com\u003e\nDate: Mon Jun 3 15:58:31 2019 +0800\n\n drm/amd/powerplay: add callback function of get_thermal_temperature_range\n \n 1. the thermal temperature is asic related data, move the code logic to\n xxx_ppt.c.\n 2. replace data structure PP_TemperatureRange with\n smu_temperature_range.\n 3. change temperature uint from temp*1000 to temp (temperature uint).\n \n Signed-off-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Acked-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a98bf57391a24a68ec8381b9d35b60c2bee79150\nAuthor: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\nDate: Tue Jul 30 14:03:00 2019 -0700\n\n tools: bpftool: add support for reporting the effective cgroup progs\n \n Takshak said in the original submission:\n \n With different bpf attach_flags available to attach bpf programs specially\n with BPF_F_ALLOW_OVERRIDE and BPF_F_ALLOW_MULTI, the list of effective\n bpf-programs available to any sub-cgroups really needs to be available for\n easy debugging.\n \n Using BPF_F_QUERY_EFFECTIVE flag, one can get the list of not only attached\n bpf-programs to a cgroup but also the inherited ones from parent cgroup.\n \n So a new option is introduced to use BPF_F_QUERY_EFFECTIVE query flag here\n to list all the effective bpf-programs available for execution at a specified\n cgroup.\n \n Reused modified test program test_cgroup_attach from tools/testing/selftests/bpf:\n # ./test_cgroup_attach\n \n With old bpftool:\n \n # bpftool cgroup show /sys/fs/cgroup/cgroup-test-work-dir/cg1/\n ID AttachType AttachFlags Name\n 271 egress multi pkt_cntr_1\n 272 egress multi pkt_cntr_2\n \n Attached new program pkt_cntr_4 in cg2 gives following:\n \n # bpftool cgroup show /sys/fs/cgroup/cgroup-test-work-dir/cg1/cg2\n ID AttachType AttachFlags Name\n 273 egress override pkt_cntr_4\n \n And with new \"effective\" option it shows all effective programs for cg2:\n \n # bpftool cgroup show /sys/fs/cgroup/cgroup-test-work-dir/cg1/cg2 effective\n ID AttachType AttachFlags Name\n 273 egress override pkt_cntr_4\n 271 egress override pkt_cntr_1\n 272 egress override pkt_cntr_2\n \n Compared to original submission use a local flag instead of global\n option.\n \n We need to clear query_flags on every command, in case batch mode\n wants to use varying settings.\n \n v2: (Takshak)\n - forbid duplicated flags;\n - fix cgroup path freeing.\n \n Signed-off-by: Takshak Chahande \u003cctakshak@fb.com\u003e\n Signed-off-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Reviewed-by: Quentin Monnet \u003cquentin.monnet@netronome.com\u003e\n Reviewed-by: Takshak Chahande \u003cctakshak@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit bf8ff0f8cfd73e850c01b453ddb79609bd83279c\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Tue Jul 30 11:05:41 2019 -0700\n\n selftests/bpf: fix clearing buffered output between tests/subtests\n \n Clear buffered output once test or subtests finishes even if test was\n successful. Not doing this leads to accumulation of output from previous\n tests and on first failed tests lots of irrelevant output will be\n dumped, greatly confusing things.\n \n v1-\u003ev2: fix Fixes tag, add more context to patch\n \n Fixes: 3a516a0a3a7b (\"selftests/bpf: add sub-tests support for test_progs\")\n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 91bc35789db4e1a489be7ab6e318e6265202e096\nAuthor: Petar Penkov \u003cppenkov@google.com\u003e\nDate: Mon Jul 29 09:59:18 2019 -0700\n\n selftests/bpf: add test for bpf_tcp_gen_syncookie\n \n Modify the existing bpf_tcp_check_syncookie test to also generate a\n SYN cookie, pass the packet to the kernel, and verify that the two\n cookies are the same (and both valid). Since cloned SKBs are skipped\n during generic XDP, this test does not issue a SYN cookie when run in\n XDP mode. We therefore only check that a valid SYN cookie was issued at\n the TC hook.\n \n Additionally, verify that the MSS for that SYN cookie is within\n expected range.\n \n Signed-off-by: Petar Penkov \u003cppenkov@google.com\u003e\n Reviewed-by: Lorenz Bauer \u003clmb@cloudflare.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 637f71c09ba22d1042f5b48b58c249ee5665d44d\nAuthor: Petar Penkov \u003cppenkov@google.com\u003e\nDate: Mon Jul 29 09:59:17 2019 -0700\n\n selftests/bpf: bpf_tcp_gen_syncookie-\u003ebpf_helpers\n \n Expose bpf_tcp_gen_syncookie to selftests.\n \n Signed-off-by: Petar Penkov \u003cppenkov@google.com\u003e\n Reviewed-by: Lorenz Bauer \u003clmb@cloudflare.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 3745ee18017e36be34f6f2a81e802a20e54e5e8b\nAuthor: Petar Penkov \u003cppenkov@google.com\u003e\nDate: Mon Jul 29 09:59:16 2019 -0700\n\n bpf: sync bpf.h to tools/\n \n Sync updated documentation for bpf_redirect_map.\n \n Sync the bpf_tcp_gen_syncookie helper function definition with the one\n in tools/uapi.\n \n Signed-off-by: Petar Penkov \u003cppenkov@google.com\u003e\n Reviewed-by: Lorenz Bauer \u003clmb@cloudflare.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 70d66244317e958092e9c971b08dd5b7fd29d9cb\nAuthor: Petar Penkov \u003cppenkov@google.com\u003e\nDate: Mon Jul 29 09:59:15 2019 -0700\n\n bpf: add bpf_tcp_gen_syncookie helper\n \n This helper function allows BPF programs to try to generate SYN\n cookies, given a reference to a listener socket. The function works\n from XDP and with an skb context since bpf_skc_lookup_tcp can lookup a\n socket in both cases.\n \n Signed-off-by: Petar Penkov \u003cppenkov@google.com\u003e\n Suggested-by: Eric Dumazet \u003cedumazet@google.com\u003e\n Reviewed-by: Lorenz Bauer \u003clmb@cloudflare.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 9349d600fb6a1ca0aaeb515523e1bb5409483d76\nAuthor: Petar Penkov \u003cppenkov@google.com\u003e\nDate: Mon Jul 29 09:59:14 2019 -0700\n\n tcp: add skb-less helpers to retrieve SYN cookie\n \n This patch allows generation of a SYN cookie before an SKB has been\n allocated, as is the case at XDP.\n \n Signed-off-by: Petar Penkov \u003cppenkov@google.com\u003e\n Reviewed-by: Lorenz Bauer \u003clmb@cloudflare.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 965112785e4bd4355262c6c5a32ea8f349adb401\nAuthor: Petar Penkov \u003cppenkov@google.com\u003e\nDate: Mon Jul 29 09:59:13 2019 -0700\n\n tcp: tcp_syn_flood_action read port from socket\n \n This allows us to call this function before an SKB has been\n allocated.\n \n Signed-off-by: Petar Penkov \u003cppenkov@google.com\u003e\n Reviewed-by: Lorenz Bauer \u003clmb@cloudflare.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 7f80fe207de9602aaff028c79345caa68c90cd31\nAuthor: Guo Ren \u003cren_guo@c-sky.com\u003e\nDate: Tue Jul 30 14:43:22 2019 +0800\n\n csky: Fixup dma_alloc_coherent with PAGE_SO attribute\n \n This bug is from commit: 2b070ccdf8c0 (fixup abiv2 mmap(... O_SYNC)\n failed). In that patch we remove the _PAGE_SO for memory noncache\n mapping and this will cause problem when drivers use dma descriptors\n to control the transcations without dma_w/rmb().\n \n After referencing other archs\u0027 implementation, pgprot_writecombine is\n introduced for mmap(... O_SYNC).\n \n Signed-off-by: Guo Ren \u003cren_guo@c-sky.com\u003e\n\ncommit 81b23ba645e6b2b446093b2d927c261a17f7dee3\nAuthor: Guo Ren \u003cren_guo@c-sky.com\u003e\nDate: Tue Jul 30 14:08:07 2019 +0800\n\n csky: Fixup mb() synchronization problem\n \n The mb() is the superset of dma and smp. Using bar.xxx to implement\n mb() will cause problem when sync data with dma device, becasue\n bar.xxx couldn\u0027t guarantee bus transactions finished at outside bus\n level.\n \n We must use sync.s instead of bar.xxx for dma data synchronization\n and it will guarantee retirement after getting the bus bresponse.\n \n Changes for V2:\n - Use sync.s for all mb, rmb, wmb, dma_wmb, dma_rmb.\n \n Signed-off-by: Guo Ren \u003cren_guo@c-sky.com\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n\ncommit 51b0c5c24484f7263ee34d2b48396b39734d0a1c\nAuthor: Matt Spinler \u003cspinler@us.ibm.com\u003e\nDate: Fri Jul 19 15:30:37 2019 -0500\n\n ARM: dts: aspeed: swift: Fix FSI GPIOs\n \n Change the FSI clock and data GPIOs to match what the hardware turned\n out to use.\n \n Fixes: 8e8fd0cbd7c5 (\"ARM: dts: aspeed: Add Swift BMC machine\")\n Signed-off-by: Matt Spinler \u003cspinler@us.ibm.com\u003e\n Reviewed-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Reviewed-by: Adriana Kobylak \u003canoo@us.ibm.com\u003e\n Signed-off-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n\ncommit a3fc5723397703a56fb6083b3e2f2ac601d1dfe0\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:12 2019 -0700\n\n extcon: adc-jack: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n [cw00.choi: Edit patch title and description for readability]\n Signed-off-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n\ncommit 2838bf941b120ec846a3903db13e319368d51b08\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 30 13:43:37 2019 +0300\n\n gpiolib-acpi: Move acpi_dev_add_driver_gpios() et al to consumer.h\n \n The API, which belongs to GPIO library, is foreign to ACPI headers. Earlier\n we moved out I²C out of the latter, and now it\u0027s time for\n acpi_dev_add_driver_gpios() et al.\n \n For time being the acpi_gpio_get_irq_resource() and acpi_dev_gpio_irq_get()\n are left untouched as they need more thought about.\n \n Note, it requires uninline acpi_dev_remove_driver_gpios() to keep purity of\n consumer.h.\n \n Cc: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Cc: Liam Girdwood \u003cliam.r.girdwood@linux.intel.com\u003e\n Cc: Jie Yang \u003cyang.jie@linux.intel.com\u003e\n Cc: Mark Brown \u003cbroonie@kernel.org\u003e\n Cc: alsa-devel@alsa-project.org (moderated list:INTEL ASoC DRIVERS)\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190730104337.21235-3-andriy.shevchenko@linux.intel.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 77cb907abe6c4b399ad77f9b6c1d77d44a435de3\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 30 13:43:36 2019 +0300\n\n gpiolib: acpi: Split ACPI stuff to gpiolib-acpi.h\n \n This is a follow up to the commit\n \n f626d6dfb709 (\"gpio: of: Break out OF-only code\")\n \n which broke down OF parts of GPIO library. Here we do the similar to ACPI.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190730104337.21235-2-andriy.shevchenko@linux.intel.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit fce04b1ce8e326bf2cafaea4a0c91dbc3e99e403\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 30 13:43:35 2019 +0300\n\n gpiolib: of: Reshuffle contents of consumer.h for new library layout\n \n Kernel build bot reported a compilation error after the commit\n \n f626d6dfb709 (\"gpio: of: Break out OF-only code\"):\n \n drivers/gpio/gpiolib-devres.o: In function `devm_gpiod_get_from_of_node\u0027:\n gpiolib-devres.c:(.text+0x19a): undefined reference to `gpiod_get_from_of_node\u0027\n \n This happens due to move the latter under umbrella of CONFIG_OF_GPIO while\n customer.h contains staled data.\n \n Fix it by reshuffling contents of consumer.h to satisfy build dependencies.\n \n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Fixes: f626d6dfb709 (\"gpio: of: Break out OF-only code\"):\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190730104337.21235-1-andriy.shevchenko@linux.intel.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 221f6df008ab39151600d4b09c85fcc730716bcb\nAuthor: Dexuan Cui \u003cdecui@microsoft.com\u003e\nDate: Fri Jun 14 18:42:30 2019 +0000\n\n hv_balloon: Reorganize the probe function\n \n Move the code that negotiates with the host to a new function\n balloon_connect_vsp() and improve the error handling.\n \n This makes the code more readable and paves the way for the\n support of hibernation in future.\n \n Makes no real logic change here.\n \n Signed-off-by: Dexuan Cui \u003cdecui@microsoft.com\u003e\n Reviewed-by: Michael Kelley \u003cmikelley@microsoft.com\u003e\n Signed-off-by: Sasha Levin \u003csashal@kernel.org\u003e\n\ncommit 1fed17df7e501b170f876b87eec11f930e3f1df5\nAuthor: Dexuan Cui \u003cdecui@microsoft.com\u003e\nDate: Fri Jun 14 18:42:17 2019 +0000\n\n hv_balloon: Use a static page for the balloon_up send buffer\n \n It\u0027s unnecessary to dynamically allocate the buffer.\n \n Signed-off-by: Dexuan Cui \u003cdecui@microsoft.com\u003e\n Reviewed-by: Michael Kelley \u003cmikelley@microsoft.com\u003e\n Signed-off-by: Sasha Levin \u003csashal@kernel.org\u003e\n\ncommit e66f840c08a237acfbb981592047b53e7411012f\nAuthor: Tristram Ha \u003cTristram.Ha@microchip.com\u003e\nDate: Mon Jul 29 19:49:47 2019 +0200\n\n net: dsa: ksz: Add Microchip KSZ8795 DSA driver\n \n Add Microchip KSZ8795 DSA driver.\n \n Signed-off-by: Tristram Ha \u003cTristram.Ha@microchip.com\u003e\n Signed-off-by: Marek Vasut \u003cmarex@denx.de\u003e\n Cc: Andrew Lunn \u003candrew@lunn.ch\u003e\n Cc: David S. Miller \u003cdavem@davemloft.net\u003e\n Cc: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Cc: Tristram Ha \u003cTristram.Ha@microchip.com\u003e\n Cc: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Cc: Woojung Huh \u003cwoojung.huh@microchip.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 016e43a26bab0126e33c9682f9d9d05eca9f0386\nAuthor: Tristram Ha \u003cTristram.Ha@microchip.com\u003e\nDate: Mon Jul 29 19:49:46 2019 +0200\n\n net: dsa: ksz: Add KSZ8795 tag code\n \n Add DSA tag code for Microchip KSZ8795 switch. The switch is simpler\n and the tag is only 1 byte, instead of 2 as is the case with KSZ9477.\n \n Signed-off-by: Tristram Ha \u003cTristram.Ha@microchip.com\u003e\n Signed-off-by: Marek Vasut \u003cmarex@denx.de\u003e\n Cc: Andrew Lunn \u003candrew@lunn.ch\u003e\n Cc: David S. Miller \u003cdavem@davemloft.net\u003e\n Cc: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Cc: Tristram Ha \u003cTristram.Ha@microchip.com\u003e\n Cc: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Cc: Woojung Huh \u003cwoojung.huh@microchip.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4c173472d051ac469f2dc816aa94484c730c7e61\nAuthor: Marek Vasut \u003cmarex@denx.de\u003e\nDate: Mon Jul 29 19:49:45 2019 +0200\n\n dt-bindings: net: dsa: ksz: document Microchip KSZ87xx family switches\n \n Document Microchip KSZ87xx family switches. These include\n KSZ8765 - 5 port switch\n KSZ8794 - 4 port switch\n KSZ8795 - 5 port switch\n \n Signed-off-by: Marek Vasut \u003cmarex@denx.de\u003e\n Cc: Andrew Lunn \u003candrew@lunn.ch\u003e\n Cc: David S. Miller \u003cdavem@davemloft.net\u003e\n Cc: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Tristram Ha \u003cTristram.Ha@microchip.com\u003e\n Cc: Vivien Didelot \u003cvivien.didelot@gmail.com\u003e\n Cc: Woojung Huh \u003cwoojung.huh@microchip.com\u003e\n Cc: devicetree@vger.kernel.org\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 0038ff357f05fee242aa4e5ec5e75f83fa1ed28c\nAuthor: Stefano Garzarella \u003csgarzare@redhat.com\u003e\nDate: Tue Jul 30 17:43:34 2019 +0200\n\n vsock/virtio: change the maximum packet size allowed\n \n Since now we are able to split packets, we can avoid limiting\n their sizes to VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE.\n Instead, we can use VIRTIO_VSOCK_MAX_PKT_BUF_SIZE as the max\n packet size.\n \n Signed-off-by: Stefano Garzarella \u003csgarzare@redhat.com\u003e\n Reviewed-by: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\n Acked-by: Michael S. Tsirkin \u003cmst@redhat.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6dbd3e66e7785a2f055bf84d98de9b8fd31ff3f5\nAuthor: Stefano Garzarella \u003csgarzare@redhat.com\u003e\nDate: Tue Jul 30 17:43:33 2019 +0200\n\n vhost/vsock: split packets to send using multiple buffers\n \n If the packets to sent to the guest are bigger than the buffer\n available, we can split them, using multiple buffers and fixing\n the length in the packet header.\n This is safe since virtio-vsock supports only stream sockets.\n \n Signed-off-by: Stefano Garzarella \u003csgarzare@redhat.com\u003e\n Reviewed-by: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\n Acked-by: Michael S. Tsirkin \u003cmst@redhat.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9632e9f61bc4191411c47933abe5f2d93c578f5e\nAuthor: Stefano Garzarella \u003csgarzare@redhat.com\u003e\nDate: Tue Jul 30 17:43:32 2019 +0200\n\n vsock/virtio: fix locking in virtio_transport_inc_tx_pkt()\n \n fwd_cnt and last_fwd_cnt are protected by rx_lock, so we should use\n the same spinlock also if we are in the TX path.\n \n Move also buf_alloc under the same lock.\n \n Signed-off-by: Stefano Garzarella \u003csgarzare@redhat.com\u003e\n Reviewed-by: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\n Acked-by: Michael S. Tsirkin \u003cmst@redhat.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b89d882dc9fc279c8acbf1df71d51b22394186d5\nAuthor: Stefano Garzarella \u003csgarzare@redhat.com\u003e\nDate: Tue Jul 30 17:43:31 2019 +0200\n\n vsock/virtio: reduce credit update messages\n \n In order to reduce the number of credit update messages,\n we send them only when the space available seen by the\n transmitter is less than VIRTIO_VSOCK_MAX_PKT_BUF_SIZE.\n \n Signed-off-by: Stefano Garzarella \u003csgarzare@redhat.com\u003e\n Reviewed-by: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\n Acked-by: Michael S. Tsirkin \u003cmst@redhat.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 473c7391ce731adb482c03e420964676ed8b494d\nAuthor: Stefano Garzarella \u003csgarzare@redhat.com\u003e\nDate: Tue Jul 30 17:43:30 2019 +0200\n\n vsock/virtio: limit the memory used per-socket\n \n Since virtio-vsock was introduced, the buffers filled by the host\n and pushed to the guest using the vring, are directly queued in\n a per-socket list. These buffers are preallocated by the guest\n with a fixed size (4 KB).\n \n The maximum amount of memory used by each socket should be\n controlled by the credit mechanism.\n The default credit available per-socket is 256 KB, but if we use\n only 1 byte per packet, the guest can queue up to 262144 of 4 KB\n buffers, using up to 1 GB of memory per-socket. In addition, the\n guest will continue to fill the vring with new 4 KB free buffers\n to avoid starvation of other sockets.\n \n This patch mitigates this issue copying the payload of small\n packets (\u003c 128 bytes) into the buffer of last packet queued, in\n order to avoid wasting memory.\n \n Signed-off-by: Stefano Garzarella \u003csgarzare@redhat.com\u003e\n Reviewed-by: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\n Acked-by: Michael S. Tsirkin \u003cmst@redhat.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b74494872555d1f7888dfd9225700a363f4a84fc\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Fri Jul 26 20:30:49 2019 +0200\n\n hrtimer: Remove task argument from hrtimer_init_sleeper()\n \n All callers hand in \u0027current\u0027 and that\u0027s the only task pointer which\n actually makes sense. Remove the task argument and set current in the\n function.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190726185752.791885290@linutronix.de\n\ncommit d1a55841ab24c6d1e4087e5c285601a9dffb8641\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:51 2019 -0700\n\n net: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n Cc: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n Cc: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n Cc: Felix Fietkau \u003cnbd@nbd.name\u003e\n Cc: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\n Cc: netdev@vger.kernel.org\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 65c84f148e359ed398dcc9ed736131103f40896b\nAuthor: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\nDate: Tue Jul 30 07:40:34 2019 -0700\n\n linux: Remove bvec page_offset, use bv_offset\n \n Now that page_offset is referenced through accessors, remove\n the union, and use bv_offset.\n \n Signed-off-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b54c9d5bd6e38edac9ce3a3f95f14a1292b5268d\nAuthor: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\nDate: Tue Jul 30 07:40:33 2019 -0700\n\n net: Use skb_frag_off accessors\n \n Use accessor functions for skb fragment\u0027s page_offset instead\n of direct references, in preparation for bvec conversion.\n \n Signed-off-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7240b60c98d6309363a9f8d5a4ecd5b0626f2aff\nAuthor: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\nDate: Tue Jul 30 07:40:32 2019 -0700\n\n linux: Add skb_frag_t page_offset accessors\n \n Add skb_frag_off(), skb_frag_off_add(), skb_frag_off_set(),\n and skb_frag_off_copy() accessors for page_offset.\n \n Signed-off-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a64e59c72ca6383149a19164abd29f81e640c08d\nAuthor: Xin Long \u003clucien.xin@gmail.com\u003e\nDate: Tue Jul 30 20:38:23 2019 +0800\n\n sctp: factor out sctp_connect_add_peer\n \n In this function factored out from sctp_sendmsg_new_asoc() and\n __sctp_connect(), it adds a peer with the other addr into the\n asoc after this asoc is created with the 1st addr.\n \n Signed-off-by: Xin Long \u003clucien.xin@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f26f995122f4c16c3a863aacbe85043135976632\nAuthor: Xin Long \u003clucien.xin@gmail.com\u003e\nDate: Tue Jul 30 20:38:22 2019 +0800\n\n sctp: factor out sctp_connect_new_asoc\n \n In this function factored out from sctp_sendmsg_new_asoc() and\n __sctp_connect(), it creates the asoc and adds a peer with the\n 1st addr.\n \n Signed-off-by: Xin Long \u003clucien.xin@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit dd8378b3af57840ef1cc87e416bd0bb35e60d8ec\nAuthor: Xin Long \u003clucien.xin@gmail.com\u003e\nDate: Tue Jul 30 20:38:21 2019 +0800\n\n sctp: clean up __sctp_connect\n \n __sctp_connect is doing quit similar things as sctp_sendmsg_new_asoc.\n To factor out common functions, this patch is to clean up their code\n to make them look more similar:\n \n 1. create the asoc and add a peer with the 1st addr.\n 2. add peers with the other addrs into this asoc one by one.\n \n while at it, also remove the unused \u0027addrcnt\u0027.\n \n Signed-off-by: Xin Long \u003clucien.xin@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f40f1177c38cb642b65af9f077bc56241e2b41c2\nAuthor: Xin Long \u003clucien.xin@gmail.com\u003e\nDate: Tue Jul 30 20:38:20 2019 +0800\n\n sctp: check addr_size with sa_family_t size in __sctp_setsockopt_connectx\n \n Now __sctp_connect() is called by __sctp_setsockopt_connectx() and\n sctp_inet_connect(), the latter has done addr_size check with size\n of sa_family_t.\n \n In the next patch to clean up __sctp_connect(), we will remove\n addr_size check with size of sa_family_t from __sctp_connect()\n for the 1st address.\n \n So before doing that, __sctp_setsockopt_connectx() should do\n this check first, as sctp_inet_connect() does.\n \n Signed-off-by: Xin Long \u003clucien.xin@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4c31bc6b1e2ee9c21608506431783dfa525b9989\nAuthor: Xin Long \u003clucien.xin@gmail.com\u003e\nDate: Tue Jul 30 20:38:19 2019 +0800\n\n sctp: only copy the available addr data in sctp_transport_init\n \n \u0027addr\u0027 passed to sctp_transport_init is not always a whole size\n of union sctp_addr, like the path:\n \n sctp_sendmsg() -\u003e\n sctp_sendmsg_new_asoc() -\u003e\n sctp_assoc_add_peer() -\u003e\n sctp_transport_new() -\u003e sctp_transport_init()\n \n In the next patches, we will also pass the address length of data\n only to sctp_assoc_add_peer().\n \n So sctp_transport_init() should copy the only available data from\n addr to peer-\u003eipaddr, instead of \u0027peer-\u003eipaddr \u003d *addr\u0027 which may\n cause slab-out-of-bounds.\n \n Signed-off-by: Xin Long \u003clucien.xin@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 2e8de0879cf662967c850edb4462f73f3baf8794\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Tue Jul 30 11:17:59 2019 -0700\n\n drm/i915: make i915_selftest.h self-contained\n \n Fix build breakage:\n \n In file included from \u003ccommand-line\u003e:\n ./drivers/gpu/drm/i915/i915_selftest.h:125:1: error: unknown type name ‘bool’\n 125 | bool __igt_timeout(unsigned long timeout, const char *fmt, ...);\n | ^~~~\n \n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730181759.26162-1-lucas.demarchi@intel.com\n\ncommit c1030cd456198a2c58f718c3c4b215698d635553\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:16 2019 -0700\n\n HSI: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Sebastian Reichel \u003csre@kernel.org\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit c6e0396124de72d6ab7f8c8e01af7f79ee1aa698\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 19:57:50 2019 +0300\n\n coda: Fix typo in the struct CodaCred documentation\n \n Documentation mistakenly refers to a different type while explaining\n the contents of the struct CodaCred.\n \n Fix the typo in the struct CodaCred description in the documentation.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 38a449ff533c9a21c254473a9f0cf59b6f420f50\nAuthor: Sheriff Esseson \u003csheriffesseson@gmail.com\u003e\nDate: Tue Jul 23 12:48:13 2019 +0100\n\n Documentation: filesystem: fix \"Removed Sysctls\" table\n \n the \"Removed Sysctls\" section is a table - bring it alive with ReST.\n \n Signed-off-by: Sheriff Esseson \u003csheriffesseson@gmail.com\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit da422ade5c879355f7410859069ef4f957c303d4\nAuthor: Mark Rutland \u003cmark.rutland@arm.com\u003e\nDate: Tue Jul 23 14:22:03 2019 +0100\n\n Documentation/features/locking: update lists\n \n The locking feature lists don\u0027t match reality as of v5.3-rc1:\n \n * arm64 moved to queued spinlocks in commit:\n c11090474d70590170cf5fa6afe85864ab494b37\n (\"arm64: locking: Replace ticket lock implementation with qspinlock\")\n \n * xtensa moved to queued spinlocks and rwlocks in commit:\n 579afe866f52adcd921272a224ab36733051059c\n (\"xtensa: use generic spinlock/rwlock implementation\")\n \n * architecture-specific rwsem support was removed in commit:\n 46ad0840b1584b92b5ff2cc3ed0b011dd6b8e0f1\n (\"locking/rwsem: Remove arch specific rwsem files\")\n \n So update the feature lists accordingly, and remove the now redundant\n rwsem-optimized list.\n \n Signed-off-by: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 7f4374e67b3046c9628cf0ab93a117704a38e95d\nAuthor: Quinn Tran \u003cqutran@marvell.com\u003e\nDate: Fri Jul 26 09:07:31 2019 -0700\n\n scsi: qla2xxx: Reject EH_{abort|device_reset|target_request}\n \n Reject eh_{abort|device_reset|target_reset} when rport is being torn down\n or chip is down.\n \n Signed-off-by: Quinn Tran \u003cquinn.tran@cavium.com\u003e\n Signed-off-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 5e5402c147083786db2238302e25c44b7a7dc5e9\nAuthor: Quinn Tran \u003cqutran@marvell.com\u003e\nDate: Fri Jul 26 09:07:30 2019 -0700\n\n scsi: qla2xxx: Skip FW dump on LOOP initialization error\n \n Firmware dump captured during LOOP Init error does not yield any\n significant information. This patch removes call to trigger firmware dump\n collection during Loop Initialization.\n \n Signed-off-by: Quinn Tran \u003cqutran@marvell.com\u003e\n Signed-off-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 178235f43ea142cf0f07dba67657494fcec21254\nAuthor: Quinn Tran \u003cqutran@marvell.com\u003e\nDate: Fri Jul 26 09:07:29 2019 -0700\n\n scsi: qla2xxx: Use Correct index for Q-Pair array\n \n For target mode, the default number of Q-Pairs allowed to use is 2. If the\n number of Q-Pairs allocated is lower than the default Q-Pairs, then lower\n value should be the set as default.\n \n [mkp: typo]\n \n Signed-off-by: Quinn Tran \u003cqutran@marvell.com\u003e\n Signed-off-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 0c6df59061b23c7a951836d23977be34e896d3da\nAuthor: Quinn Tran \u003cqutran@marvell.com\u003e\nDate: Fri Jul 26 09:07:28 2019 -0700\n\n scsi: qla2xxx: Fix abort timeout race condition.\n \n If an abort times out, the Abort IOCB completion and Abort timer can race\n against each other. This patch provides unique error code for timer path to\n allow proper cleanup.\n \n [mkp: typo]\n \n Signed-off-by: Quinn Tran \u003cqutran@marvell.com\u003e\n Signed-off-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit d376dbda187317d06d3a2d495b43a7983e4a3250\nAuthor: Quinn Tran \u003cqutran@marvell.com\u003e\nDate: Fri Jul 26 09:07:27 2019 -0700\n\n scsi: qla2xxx: Fix different size DMA Alloc/Unmap\n \n [ 17.177276] qla2xxx 0000:05:00.0: DMA-API: device driver frees DMA memory\n with different size [device address\u003d0x00000006198b0000] [map size\u003d32784 bytes]\n [unmap size\u003d8208 bytes]\n [ 17.177390] RIP: 0010:check_unmap+0x7a2/0x1750\n [ 17.177425] Call Trace:\n [ 17.177438] debug_dma_free_coherent+0x1b5/0x2d5\n [ 17.177470] dma_free_attrs+0x7f/0x140\n [ 17.177489] qla24xx_sp_unmap+0x1e2/0x610 [qla2xxx]\n [ 17.177509] qla24xx_async_gnnft_done+0x9c6/0x17d0 [qla2xxx]\n [ 17.177535] qla2x00_do_work+0x514/0x2200 [qla2xxx]\n \n Fixes: b5f3bc39a0e8 (\"scsi: qla2xxx: Fix inconsistent DMA mem alloc/free\")\n Signed-off-by: Quinn Tran \u003cqutran@marvell.com\u003e\n Signed-off-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 5d328de64d89400dcf9911125844d8adc0db697f\nAuthor: Himanshu Madhani \u003chmadhani@marvell.com\u003e\nDate: Fri Jul 26 09:07:26 2019 -0700\n\n scsi: qla2xxx: Fix DMA unmap leak\n \n With debug kernel we see following wanings indicating memory leak.\n \n [28809.523959] WARNING: CPU: 3 PID: 6790 at lib/dma-debug.c:978\n dma_debug_device_change+0x166/0x1d0\n [28809.523964] pci 0000:0c:00.6: DMA-API: device driver has pending DMA\n allocations while released from device [count\u003d5]\n [28809.523964] One of leaked entries details: [device\n address\u003d0x00000002aefe4000] [size\u003d8208 bytes] [mapped with DMA_BIDIRECTIONAL]\n [mapped as coherent]\n \n Fix this by unmapping DMA memory.\n \n Signed-off-by: Quinn Tran \u003cqutran@marvell.com\u003e\n Signed-off-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 8f55c307ab577af74efab066ba1e45b0e51e4bae\nAuthor: Minwoo Im \u003cminwoo.im@samsung.com\u003e\nDate: Sun Jul 28 03:53:37 2019 +0900\n\n scsi: mpt3sas: support target smid for [abort|query] task\n \n We can request task management IOCTL command(MPI2_FUNCTION_SCSI_TASK_MGMT)\n to /dev/mpt3ctl. If the given task_type is either abort task or query\n task, it may need a field named \"Initiator Port Transfer Tag to Manage\" in\n the IU.\n \n Current code does not support to check target IPTT tag from the tm_request.\n This patch introduces to check TaskMID given from the userspace as a target\n tag. We have a rule of relationship between\n (struct request *req-\u003etag) and smid in mpt3sas_base.c:\n \n 3318 u16\n 3319 mpt3sas_base_get_smid_scsiio(struct MPT3SAS_ADAPTER *ioc, u8 cb_idx,\n 3320 struct scsi_cmnd *scmd)\n 3321 {\n 3322 struct scsiio_tracker *request \u003d scsi_cmd_priv(scmd);\n 3323 unsigned int tag \u003d scmd-\u003erequest-\u003etag;\n 3324 u16 smid;\n 3325\n 3326 smid \u003d tag + 1;\n \n So if we want to abort a request tagged #X, then we can pass (X + 1) to\n this IOCTL handler. Otherwise, user space just can pass 0 TaskMID to abort\n the first outstanding smid which is legacy behaviour.\n \n Cc: Sreekanth Reddy \u003csreekanth.reddy@broadcom.com\u003e\n Cc: Suganath Prabu Subramani \u003csuganath-prabu.subramani@broadcom.com\u003e\n Cc: Sathya Prakash \u003csathya.prakash@broadcom.com\u003e\n Cc: James E.J. Bottomley \u003cjejb@linux.ibm.com\u003e\n Cc: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n Cc: MPT-FusionLinux.pdl@broadcom.com\n Signed-off-by: Minwoo Im \u003cminwoo.im@samsung.com\u003e\n Acked-by: Sreekanth Reddy \u003csreekanth.reddy@broadcom.com\u003e\n Signed-off-by: Minwoo Im \u003cminwoo.im.dev@gmail.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 9fba647cfedc15e1ac80d396c116e106da947e55\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Jul 29 06:29:02 2019 -0500\n\n scsi: ibmvscsi_tgt: Mark expected switch fall-throughs\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warnings (Building: powerpc allyesconfig):\n \n drivers/scsi/ibmvscsi_tgt/ibmvscsi_tgt.c: In function \u0027ibmvscsis_adapter_info\u0027:\n drivers/scsi/ibmvscsi_tgt/ibmvscsi_tgt.c:1582:6: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n if (connection_broken(vscsi))\n ^\n drivers/scsi/ibmvscsi_tgt/ibmvscsi_tgt.c:1584:2: note: here\n default:\n ^~~~~~~\n drivers/scsi/ibmvscsi_tgt/ibmvscsi_tgt.c: In function \u0027ibmvscsis_ping_response\u0027:\n drivers/scsi/ibmvscsi_tgt/ibmvscsi_tgt.c:2494:16: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n vscsi-\u003eflags |\u003d CLIENT_FAILED;\n drivers/scsi/ibmvscsi_tgt/ibmvscsi_tgt.c:2495:2: note: here\n case H_DROPPED:\n ^~~~\n drivers/scsi/ibmvscsi_tgt/ibmvscsi_tgt.c:2496:16: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n vscsi-\u003eflags |\u003d RESPONSE_Q_DOWN;\n drivers/scsi/ibmvscsi_tgt/ibmvscsi_tgt.c:2497:2: note: here\n case H_REMOTE_PARM:\n ^~~~\n \n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Reviewed-by: Bryant G. Ly \u003cbly@catalogicsoftware.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 657bd277c162580674ddb86a90c4aeb62639bff5\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Sun Jul 28 19:21:19 2019 -0500\n\n scsi: cxlflash: Mark expected switch fall-throughs\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warnings:\n \n drivers/scsi/cxlflash/main.c: In function \u0027send_afu_cmd\u0027:\n drivers/scsi/cxlflash/main.c:2347:6: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n if (rc) {\n ^\n drivers/scsi/cxlflash/main.c:2357:2: note: here\n case -EAGAIN:\n ^~~~\n drivers/scsi/cxlflash/main.c: In function \u0027term_intr\u0027:\n drivers/scsi/cxlflash/main.c:754:6: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n if (index \u003d\u003d PRIMARY_HWQ)\n ^\n drivers/scsi/cxlflash/main.c:756:2: note: here\n case UNMAP_TWO:\n ^~~~\n drivers/scsi/cxlflash/main.c:757:3: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n cfg-\u003eops-\u003eunmap_afu_irq(hwq-\u003ectx_cookie, 2, hwq);\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/cxlflash/main.c:758:2: note: here\n case UNMAP_ONE:\n ^~~~\n drivers/scsi/cxlflash/main.c:759:3: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n cfg-\u003eops-\u003eunmap_afu_irq(hwq-\u003ectx_cookie, 1, hwq);\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/cxlflash/main.c:760:2: note: here\n case FREE_IRQ:\n ^~~~\n drivers/scsi/cxlflash/main.c: In function \u0027cxlflash_remove\u0027:\n drivers/scsi/cxlflash/main.c:975:3: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n cxlflash_release_chrdev(cfg);\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/cxlflash/main.c:976:2: note: here\n case INIT_STATE_SCSI:\n ^~~~\n drivers/scsi/cxlflash/main.c:978:3: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n scsi_remove_host(cfg-\u003ehost);\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/cxlflash/main.c:979:2: note: here\n case INIT_STATE_AFU:\n ^~~~\n drivers/scsi/cxlflash/main.c:980:3: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n term_afu(cfg);\n ^~~~~~~~~~~~~\n drivers/scsi/cxlflash/main.c:981:2: note: here\n case INIT_STATE_PCI:\n ^~~~\n drivers/scsi/cxlflash/main.c:983:3: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n pci_disable_device(pdev);\n ^~~~~~~~~~~~~~~~~~~~~~~~\n drivers/scsi/cxlflash/main.c:984:2: note: here\n case INIT_STATE_NONE:\n ^~~~\n drivers/scsi/cxlflash/main.c: In function \u0027num_hwqs_store\u0027:\n drivers/scsi/cxlflash/main.c:3018:6: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n if (cfg-\u003estate \u003d\u003d STATE_NORMAL)\n ^\n drivers/scsi/cxlflash/main.c:3020:2: note: here\n default:\n ^~~~~~~\n \n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Acked-by: Matthew R. Ochs \u003cmrochs@linux.ibm.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 1de540a9263876a068a15ce216bd236e2c3394d6\nAuthor: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nDate: Fri Jul 26 09:52:05 2019 +0300\n\n scsi: mpt3sas: clean up a couple sizeof() uses\n \n There is a copy and paste bug here. It uses EVENT_TRIGGERS size instead of\n SCSI_TRIGGERS size but fortunately both size are 84 bytes so it doesn\u0027t\n affect runtime.\n \n These days the preferred style is to just say sizeof(object) instead of\n sizeof(type) so I have updated the function to the latest style as well.\n \n Signed-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Acked-by: Sreekanth Reddy \u003csreekanth.reddy@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 555233601407725065cf37e00c834394452c6780\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Thu Jul 25 16:48:13 2019 -0700\n\n drm/i915/tgl: handle DP aux interrupts\n \n For Tiger Lake the DE Port Interrupt Definition bits changed, so use the\n new bit definitions.\n \n Cc: Jose Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190713010940.17711-7-lucas.demarchi@intel.com\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725234813.27179-4-lucas.demarchi@intel.com\n\ncommit 48ef15d32a77042ee60fcb070dcf88ece72ec2e1\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Thu Jul 25 16:48:12 2019 -0700\n\n drm/i915/tgl: Update north display hotplug detection to TGL connections\n \n TGL has 3 combophys and 6 TC/TBT ports, so it has 2 more TC/TBT ports\n than ICL and the PORT_C on TGL is a combophy.\n So here adding a new hpd north table and function to detect long\n pulse for TGL.\n \n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190713010940.17711-6-lucas.demarchi@intel.com\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725234813.27179-3-lucas.demarchi@intel.com\n\ncommit 52dfdba0a987689c56bf7013304cf1041d841064\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Thu Jul 25 16:48:11 2019 -0700\n\n drm/i915/tgl: Add hpd interrupt handling\n \n Add hotdplug detection for all ports on TGP. icp_hpd_detection_setup()\n is refactored to be shared with TGP.\n \n While we increase the number of pins, add a BUILD_BUG_ON() to avoid\n going over the number of bits allowed.\n \n v2: use BITS_PER_TYPE and correct type for BUILD_BUG_ON() check\n (requested by Ville)\n \n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Jose Souza \u003cjose.souza@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725234813.27179-2-lucas.demarchi@intel.com\n\ncommit e008227eb34fe14eb5fdb1ca2a07ac018dd9f22a\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Jul 29 17:45:19 2019 -0500\n\n HSI: ssi_protocol: Mark expected switch fall-throughs\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warning (Building: arm):\n \n drivers/hsi/clients/ssi_protocol.c: In function ‘ssip_set_rxstate’:\n drivers/hsi/clients/ssi_protocol.c:291:6: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n if (atomic_read(\u0026ssi-\u003etx_usecnt))\n ^\n drivers/hsi/clients/ssi_protocol.c:294:2: note: here\n case RECEIVING:\n ^~~~\n \n drivers/hsi/clients/ssi_protocol.c: In function ‘ssip_keep_alive’:\n drivers/hsi/clients/ssi_protocol.c:466:7: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n if (atomic_read(\u0026ssi-\u003etx_usecnt) \u003d\u003d 0)\n ^\n drivers/hsi/clients/ssi_protocol.c:472:3: note: here\n case SEND_IDLE:\n ^~~~\n \n Notice that, in this particular case, the code comment is\n modified in accordance with what GCC is expecting to find.\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n\ncommit e0e712fe42ef67bdf45fc348767d1d0a4eeba77f\nAuthor: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\nDate: Tue Jul 30 11:50:24 2019 -0700\n\n drm/i915: Update DRIVER_DATE to 20190730\n \n Signed-off-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n\ncommit 04d15d5cadb8f764ccf978ddd33cf233dcc68e13\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:42 2019 -0700\n\n staging: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190730181557.90391-43-swboyd@chromium.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 509ce4c85bd055ee1013bc853b5d543428b0f017\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Mon Jul 29 00:27:39 2019 +0900\n\n ppdev: add header include guard\n \n Add a header include guard just in case.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Link: https://lore.kernel.org/r/20190728152739.9249-1-yamada.masahiro@socionext.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d4bad8118d2ab6587c508d531deac659bb8ad188\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:44 2019 -0700\n\n tty: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190730181557.90391-45-swboyd@chromium.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit ab81f3f386e2650b625c4dfe8ca092d30289c2a3\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 20:46:27 2019 +0800\n\n mei: Use dev_get_drvdata where possible\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190723124627.24671-1-hslester96@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 75080370459a717da17fc9df2e7541f201c63746\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:45 2019 -0700\n\n uio: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190730181557.90391-46-swboyd@chromium.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b33f37064b743d4c5771d3bfc65976790b7396cc\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Tue Jul 30 11:15:46 2019 -0700\n\n usb: Remove dev_err() usage after platform_get_irq()\n \n We don\u0027t need dev_err() messages when platform_get_irq() fails now that\n platform_get_irq() prints an error message itself when something goes\n wrong. Let\u0027s remove these prints with a simple semantic patch.\n \n // \u003csmpl\u003e\n @@\n expression ret;\n struct platform_device *E;\n @@\n \n ret \u003d\n (\n platform_get_irq(E, ...)\n |\n platform_get_irq_byname(E, ...)\n );\n \n if ( \\( ret \u003c 0 \\| ret \u003c\u003d 0 \\) )\n {\n (\n -if (ret !\u003d -EPROBE_DEFER)\n -{ ...\n -dev_err(...);\n -... }\n |\n ...\n -dev_err(...);\n )\n ...\n }\n // \u003c/smpl\u003e\n \n While we\u0027re here, remove braces on if statements that only have one\n statement (manually).\n \n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190730181557.90391-47-swboyd@chromium.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit dc7077f89ad9e16533ec1e507fabd8f427982f3e\nAuthor: Firoz Khan \u003cfiroz.khan@linaro.org\u003e\nDate: Wed Jan 2 20:26:17 2019 +0530\n\n mips: remove nargs from __SYSCALL\n \n The __SYSCALL macro\u0027s arguments are system call number,\n system call entry name and number of arguments for the\n system call.\n \n Argument- nargs in __SYSCALL(nr, entry, nargs) is neither\n calculated nor used anywhere. So it would be better to\n keep the implementaion as __SYSCALL(nr, entry). This will\n unifies the implementation with some other architetures\n too.\n \n Signed-off-by: Firoz Khan \u003cfiroz.khan@linaro.org\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Philippe Ombredanne \u003cpombredanne@nexb.com\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Kate Stewart \u003ckstewart@linuxfoundation.org\u003e\n Cc: y2038@lists.linaro.org\n Cc: linux-kernel@vger.kernel.org\n Cc: linux-arch@vger.kernel.org\n Cc: arnd@arndb.de\n Cc: deepa.kernel@gmail.com\n Cc: marcin.juszkiewicz@linaro.org\n\ncommit 7f428252dbd0adc91317ae7ea27d7e27b70df1f6\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Thu Jul 25 18:02:15 2019 -0400\n\n MIPS: jz4740: Drop dead code\n \n Remove all the source files that are not used anywhere anymore.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit d202742058b2ee5666d9820d539b67dc14f64712\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Thu Jul 25 18:02:13 2019 -0400\n\n hwmon: Drop obsolete JZ4740 driver\n \n The JZ4740 boards now use the iio-hwmon driver.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Acked-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit aea12071d6fcf143b8140bcdde80cdc419101299\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Thu Jul 25 18:02:12 2019 -0400\n\n power/supply: Drop obsolete JZ4740 driver\n \n It has been replaced with the more mature ingenic-battery driver.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Acked-by: Sebastian Reichel \u003csebastian.reichel@collabora.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit b58451e00f2b803961c3ec8b46f5fcfd8cf78bef\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Thu Jul 25 18:02:11 2019 -0400\n\n mtd: rawnand: Drop obsolete JZ4740 NAND driver\n \n It has been replaced with the newer Ingenic NAND driver.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Acked-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit 091c6104ade5527ce29035673ff262a24c4f036c\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Thu Jul 25 18:02:10 2019 -0400\n\n dma: Drop JZ4740 driver\n \n The newer and better JZ4780 driver is now used to provide DMA\n functionality on the JZ4740.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Acked-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit e305bff3a706f58eb65e91cc8fde9bb67ba9688f\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Thu Jul 25 18:02:09 2019 -0400\n\n video/fbdev: Drop JZ4740 driver\n \n The JZ4740 fbdev driver has been replaced with the ingenic-drm driver.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit 91c7db5e7c652108342e80cba4d924f091a92957\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Thu Jul 25 18:02:08 2019 -0400\n\n ASoC: jz4740: Drop lb60 board code\n \n The board now uses the simple-audio-card driver.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Acked-by: Mark Brown \u003cbroonie@kernel.org\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit fe9825146b2586f6747f1f91c5bde9ab7624dfc5\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Thu Jul 25 18:02:07 2019 -0400\n\n MIPS: configs: LB60: update defconfig\n \n Update the defconfig to select the new drivers instead of the old ones.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit 8ddebad15e9b0b8d5b9599618c0f7dca96222713\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Thu Jul 25 18:02:06 2019 -0400\n\n MIPS: qi_lb60: Migrate to devicetree\n \n Move all the platform data to devicetree.\n \n The only bit dropped is the PWM beeper, which requires the PWM driver\n to be updated. I figured it\u0027s okay to remove it here since it\u0027s really\n a non-critical device, and it\u0027ll be re-introduced soon enough.\n \n The other change is the CS line of the SPI is now set as active low. The\n SPI core would have forced \"active low\" anyway, unless the \u0027spi-cs-high\u0027\n property is set.\n \n In the process of moving to devicetree, we also switched to new drivers:\n - We use the simple-audio-card and simple-amplifier drivers instead of\n the custom ASoC code;\n - We use the new Ingenic DRM driver coupled with the GiantPlus GPM940B0\n DRM panel driver instead of the old framebuffer driver;\n - We use the new jz4780-dma driver instead of the old jz4740-dma one;\n - We use the ingenic-nand and jz4740-ecc drivers instead of the old\n jz4740-nand driver;\n - We use ingenic-battery instead of jz4740-battery;\n - We use iio-hwmon instead of jz4740-hwmon;\n - We use ingenic-iio instead of the old jz4740-adc MFD driver.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n [paul.burton@mips.com:\n Drop the unused \u0026 undocumented ili8960 spi@0 node.]\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit 1db88c5343712e411a2dd45375f27c477e33dc07\nAuthor: David Howells \u003cdhowells@redhat.com\u003e\nDate: Tue Jul 30 15:56:57 2019 +0100\n\n rxrpc: Fix -Wframe-larger-than\u003d warnings from on-stack crypto\n \n rxkad sometimes triggers a warning about oversized stack frames when\n building with clang for a 32-bit architecture:\n \n net/rxrpc/rxkad.c:243:12: error: stack frame size of 1088 bytes in function \u0027rxkad_secure_packet\u0027 [-Werror,-Wframe-larger-than\u003d]\n net/rxrpc/rxkad.c:501:12: error: stack frame size of 1088 bytes in function \u0027rxkad_verify_packet\u0027 [-Werror,-Wframe-larger-than\u003d]\n \n The problem is the combination of SYNC_SKCIPHER_REQUEST_ON_STACK() in\n rxkad_verify_packet()/rxkad_secure_packet() with the relatively large\n scatterlist in rxkad_verify_packet_1()/rxkad_secure_packet_encrypt().\n \n The warning does not show up when using gcc, which does not inline the\n functions as aggressively, but the problem is still the same.\n \n Allocate the cipher buffers from the slab instead, caching the allocated\n packet crypto request memory used for DATA packet crypto in the rxrpc_call\n struct.\n \n Fixes: 17926a79320a (\"[AF_RXRPC]: Provide secure RxRPC sockets for use by userspace and kernel both\")\n Reported-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Signed-off-by: David Howells \u003cdhowells@redhat.com\u003e\n Acked-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n cc: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 36ba3eae036242fdc36eb8e89d5001a219a7fb1d\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Thu Jul 25 18:02:05 2019 -0400\n\n MIPS: DTS: jz4740: Add missing nodes\n \n Add nodes for the MMC, AIC, ADC, CODEC, MUSB, LCD, memory,\n and BCH controllers.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit 8b6655d22b42880c5a68676738ee00f64acae216\nAuthor: Sai Prakash Ranjan \u003csaiprakash.ranjan@codeaurora.org\u003e\nDate: Fri Jul 12 19:46:27 2019 +0530\n\n coresight: cpu-debug: Add support for Qualcomm Kryo\n \n Add support for coresight CPU debug module on Qualcomm\n Kryo CPUs. This patch adds the UCI entries for Kryo CPUs\n found on MSM8996 which shares the same PIDs as ETMs.\n \n Without this, below error is observed on MSM8996:\n \n [ 5.429867] OF: graph: no port node found in /soc/debug@3810000\n [ 5.429938] coresight-etm4x: probe of 3810000.debug failed with error -22\n [ 5.435415] coresight-cpu-debug 3810000.debug: Coresight debug-CPU0 initialized\n [ 5.446474] OF: graph: no port node found in /soc/debug@3910000\n [ 5.448927] coresight-etm4x: probe of 3910000.debug failed with error -22\n [ 5.454681] coresight-cpu-debug 3910000.debug: Coresight debug-CPU1 initialized\n [ 5.487765] OF: graph: no port node found in /soc/debug@3a10000\n [ 5.488007] coresight-etm4x: probe of 3a10000.debug failed with error -22\n [ 5.493024] coresight-cpu-debug 3a10000.debug: Coresight debug-CPU2 initialized\n [ 5.501802] OF: graph: no port node found in /soc/debug@3b10000\n [ 5.512901] coresight-etm4x: probe of 3b10000.debug failed with error -22\n [ 5.513192] coresight-cpu-debug 3b10000.debug: Coresight debug-CPU3 initialized\n \n Signed-off-by: Sai Prakash Ranjan \u003csaiprakash.ranjan@codeaurora.org\u003e\n Tested-by: Leo Yan \u003cleo.yan@linaro.org\u003e\n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n\ncommit 914f79d6f9d54306cdda3eb36ed987168f3ee9d4\nAuthor: Sai Prakash Ranjan \u003csaiprakash.ranjan@codeaurora.org\u003e\nDate: Fri Jul 12 19:46:26 2019 +0530\n\n coresight: etm4x: Add ETM PIDs for SDM845 and MSM8996\n \n Instead of overriding the peripheral id(PID) check in AMBA\n by hardcoding them in DT, add the PIDs to the ETM4x driver.\n Here we use Unique Component Identifier(UCI) for MSM8996\n since the ETM and CPU debug module shares the same PIDs.\n SDM845 does not support CPU debug module.\n \n Signed-off-by: Sai Prakash Ranjan \u003csaiprakash.ranjan@codeaurora.org\u003e\n Reviewed-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n\ncommit 18e87cb700a9923a4173a8515d291ac1159df883\nAuthor: Markus Elfring \u003celfring@users.sourceforge.net\u003e\nDate: Fri Jul 5 19:33:26 2019 +0200\n\n coresight: etm4x: Two function calls less\n \n Avoid an extra function call in two function implementations\n by using a ternary operator instead of a conditional statement.\n \n This issue was detected by using the Coccinelle software.\n \n Signed-off-by: Markus Elfring \u003celfring@users.sourceforge.net\u003e\n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n\ncommit 939878fe06461dec3edfb31f7dd130dcea1d21e4\nAuthor: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\nDate: Mon Jul 29 19:11:52 2019 +0300\n\n staging: fsl-dpaa2/ethsw: add .ndo_fdb[add|del] callbacks\n \n Add the .ndo_fdb_[add|del] callbacks so that FDB entries not associated\n with a master device still end up offloaded.\n \n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Link: https://lore.kernel.org/r/1564416712-16946-6-git-send-email-ioana.ciornei@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit e1e6a04d97cf747ccbec3ae23008c9faf8557aee\nAuthor: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\nDate: Mon Jul 29 19:11:51 2019 +0300\n\n staging: fsl-dpaa2/ethsw: check added_by_user flag\n \n We do not want to offload FDB entries if not added by user as static\n entries. Check the added_by_user flag and break if not set.\n \n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Link: https://lore.kernel.org/r/1564416712-16946-5-git-send-email-ioana.ciornei@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit de01ac2e49d5b403c3edca17e7fb257060e5192c\nAuthor: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\nDate: Mon Jul 29 19:11:50 2019 +0300\n\n staging: fsl-dpaa2/ethsw: add .ndo_fdb_dump callback\n \n Implement the .ndo_fdb_dump callback for the switch net devices. The\n list of all offloaded FDB entries is retrieved through the dpsw_fdb_dump()\n firmware call. Filter the entries by the switch port on which the\n callback was called and for each of them create a new neighbour message.\n Also remove the requirement from the TODO list.\n \n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Link: https://lore.kernel.org/r/1564416712-16946-4-git-send-email-ioana.ciornei@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b380a4e6e5e1f25eb40ec2a2f40196d0c00e6832\nAuthor: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\nDate: Mon Jul 29 19:11:49 2019 +0300\n\n staging: fsl-dpaa2/ethsw: notify switchdev of offloaded entry\n \n Notify switchdev in case the FDB entry was successfully offloaded.\n This will help users to make the distinction between entries known to\n the HW switch and those that are held only on the software bridge.\n \n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Link: https://lore.kernel.org/r/1564416712-16946-3-git-send-email-ioana.ciornei@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit e2724c77ca219ef4e7ba2a5399e2d02f560d0c69\nAuthor: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\nDate: Mon Jul 29 19:11:48 2019 +0300\n\n staging: fsl-dpaa2/ethsw: remove unused structure\n \n The dpsw_cfg structure is only used when creating a new dpsw DPAA2\n object. In the DPAA2 architecture, objects are created at boot time by\n the firmware or dynamically from userspace while drivers on the fsl-mc\n bus only configure those objects.\n Remove the structure since it\u0027s of no use.\n \n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Link: https://lore.kernel.org/r/1564416712-16946-2-git-send-email-ioana.ciornei@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f066fa60640130eed62f628e382ad28f92d442a5\nAuthor: Fabian Mewes \u003carchitekt@coding4coffee.org\u003e\nDate: Mon Jul 29 07:08:34 2019 +0200\n\n MIPS: Kconfig: remove HAVE_LATENCYTOP_SUPPORT\n \n HAVE_LATENCYTOP_SUPPORT was removed all together in commit da48d094ce5d7\n (\"Kconfig: remove HAVE_LATENCYTOP_SUPPORT\"). This commit removes a\n leftover in the MIPS Kconfig.\n \n Signed-off-by: Fabian Mewes \u003carchitekt@coding4coffee.org\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: Sergei Shtylyov \u003csergei.shtylyov@cogentembedded.com\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n\ncommit 1623835507e343fbc30c5806c65cafc149e8fcc7\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Thu Jul 25 23:07:51 2019 +0300\n\n scsi: scsi_debugfs: Use for_each_set_bit to simplify code\n \n We can use for_each_set_bit() to slightly simplify the code.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 7ea502e1d885966cb5912b4488255d8c12645042\nAuthor: Zhou Yanjie \u003czhouyanjie@zoho.com\u003e\nDate: Tue Jul 30 19:30:11 2019 +0800\n\n MIPS: X1000: Add X1000 system type.\n \n Add X1000 system type for cat /proc/cpuinfo to give out X1000.\n \n Signed-off-by: Zhou Yanjie \u003czhouyanjie@zoho.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-mips@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: ralf@linux-mips.org\n Cc: paul@crapouillou.net\n Cc: jhogan@kernel.org\n Cc: fancer.lancer@gmail.com\n Cc: chenhc@lemote.com\n Cc: tglx@linutronix.de\n Cc: gregkh@linuxfoundation.org\n Cc: armijn@tjaldur.nl\n Cc: syq@debian.org\n Cc: jiaxun.yang@flygoat.com\n\ncommit 0486ad20e73d03c82208b802ac41e80b942b23f6\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Tue Jul 30 23:00:57 2019 +0800\n\n drm/rockchip: Make analogix_dp_atomic_check static\n \n Fix sparse warning:\n \n drivers/gpu/drm/bridge/analogix/analogix_dp_core.c:1151:5: warning:\n symbol \u0027analogix_dp_atomic_check\u0027 was not declared. Should it be static?\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730150057.57388-1-yuehaibing@huawei.com\n\ncommit 56cc8fae5f7e9f38cb367754c52491ba1645d1bf\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Thu Jul 18 15:45:18 2019 +0800\n\n scsi: qla2xxx: Replace vmalloc + memset with vzalloc\n \n Use vzalloc instead of using vmalloc to allocate memory and then zeroing it\n with memset. This simplifies the code.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Acked-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 4590c07462fbff4ecbfe1deec44431c16191bd19\nAuthor: Hans Verkuil \u003chverkuil@xs4all.nl\u003e\nDate: Sun Jul 28 10:43:41 2019 -0400\n\n media: via-camera: convert to the vb2 framework\n \n Convert the via-camera to the vb2 framework. With this change this\n driver passes all \u0027v4l2-compliance -s\u0027 tests on my OLPC 1.5.\n \n Also tested with the Sugar \u0027Record\u0027 application.\n \n All tests were done under the OLPC official 5.0.8 kernel.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 22d5aa3a4570b1419003d19992080c423dbf1734\nAuthor: Sergei Shtylyov \u003csergei.shtylyov@cogentembedded.com\u003e\nDate: Tue Jul 16 23:41:54 2019 +0300\n\n scsi: fdomain_isa: use CFG1_IRQ_MASK\n \n Commit 1697c6a64c49 (\"scsi: fdomain: Add register definitions\") somehow\n missed the uses of CFG1_IRQ_MASK in the Futire Domain ISA driver, leaving\n the magic numbers intact. Fix this issue (with no change in the generated\n object file), removing an excess empty line, while at it...\n \n Fixes: 1697c6a64c49 (\"scsi: fdomain: Add register definitions\")\n Signed-off-by: Sergei Shtylyov \u003csergei.shtylyov@cogentembedded.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 4e4f3b99b41ff3e5cd85e6a309e423b5721ca8ce\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Fri Jul 26 08:08:21 2019 -0400\n\n media: vivid: add support for new pixelformats\n \n New RGB444, RGB555 and RGB32 variants were recently added. Now also\n support them in vivid.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 7a1c0b79dbe3423214e70750a5ee38fb617168ab\nAuthor: Sergei Shtylyov \u003csergei.shtylyov@cogentembedded.com\u003e\nDate: Tue Jul 16 23:40:47 2019 +0300\n\n scsi: fdomain: use BSTAT_{MSG|CMD|IO} in fdomain_work()\n \n Commit 1697c6a64c49 (\"scsi: fdomain: Add register definitions\") somehow\n missed the masking of the \u0027status\u0027 variable with the SCSI phase mask in\n fdomain_work(), leaving the magic number intact. Fix this issue; while at\n it, change the order of BSTAT_{MSG|CMD|IO} bits in the MESSAGE IN\n phase *case* (with no change in the generated object file).\n \n Fixes: 1697c6a64c49 (\"scsi: fdomain: Add register definitions\")\n Signed-off-by: Sergei Shtylyov \u003csergei.shtylyov@cogentembedded.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 079a649c75252b76799713262835cc5200013149\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Fri Jul 26 08:08:20 2019 -0400\n\n media: v4l2-common: add support for new RGB32 pixelformats\n \n Four new variants of RGB32 were added. Add support for them to\n the v4l2_format_info() function.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit cad2991a950198f71520e8bbbe9d1ea9d001e7f0\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Fri Jul 26 08:08:19 2019 -0400\n\n media: v4l2-tpg: add support for new pixelformats\n \n Several new variations of RGB444, RGB555 and RGB32 were\n added. Add support for these.\n \n This is the list of new pixelformats:\n \n V4L2_PIX_FMT_RGBX444\n V4L2_PIX_FMT_RGBA444\n V4L2_PIX_FMT_XBGR444\n V4L2_PIX_FMT_ABGR444\n V4L2_PIX_FMT_BGRX444\n V4L2_PIX_FMT_BGRA444\n V4L2_PIX_FMT_RGBX555\n V4L2_PIX_FMT_RGBA555\n V4L2_PIX_FMT_XBGR555\n V4L2_PIX_FMT_ABGR555\n V4L2_PIX_FMT_BGRX555\n V4L2_PIX_FMT_BGRA555\n V4L2_PIX_FMT_RGBX32\n V4L2_PIX_FMT_BGRX32\n V4L2_PIX_FMT_RGBA32\n V4L2_PIX_FMT_BGRA32\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit a9651be7e4818d252e2a35f371bde79c556b1560\nAuthor: Sergei Shtylyov \u003csergei.shtylyov@cogentembedded.com\u003e\nDate: Tue Jul 16 23:39:15 2019 +0300\n\n scsi: fdomain: use BCTL_RST in fdomain_reset()\n \n Commit 1697c6a64c49 (\"scsi: fdomain: Add register definitions\") somehow\n missed the use of BCTL_RST in fdomain_reset(), leaving the magic number\n intact. Fix this issue (with no change in the generated object file).\n \n Fixes: 1697c6a64c49 (\"scsi: fdomain: Add register definitions\")\n Signed-off-by: Sergei Shtylyov \u003csergei.shtylyov@cogentembedded.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 1e42d384aa19e06410be52bd05cfc159722c0b7f\nAuthor: Paul Kocialkowski \u003cpaul.kocialkowski@bootlin.com\u003e\nDate: Thu Jul 25 14:55:59 2019 -0400\n\n media: v4l2-ctrl: Add a comment on why we zero out compound controls fields\n \n Since it\u0027s not necessarily very explicit why we need to zero some fields\n in std_validate_compound, add a comment before the function to explain\n why, which should help as a reminder.\n \n Signed-off-by: Paul Kocialkowski \u003cpaul.kocialkowski@bootlin.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit b217598890a26fe49c83f4f3124c343a6d990eb2\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Jul 11 22:15:39 2019 +0800\n\n scsi: aic94xx: Remove unnecessary null check\n \n kmem_cache_destroy() can handle NULL pointer correctly, so there is no need\n to check NULL pointer before calling kmem_cache_destroy().\n \n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 4b4d62b03191700f3245941146a8510d31ec66b3\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jul 24 07:27:06 2019 -0400\n\n media: videodev2.h.rst.exceptions: tymecode -\u003e timecode\n \n Fix typo.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 7c490e2561f271a59605ac31e2582d62c918715d\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jul 24 07:27:03 2019 -0400\n\n media: v4l2-ioctl.c: OR flags in v4l_fill_fmtdesc(), not don\u0027t overwrite\n \n If a driver sets a FMT flag in the enum_fmt op, then that will be\n ignored since v4l_fill_fmtdesc() overwrites it again.\n \n v4l_fill_fmtdesc() should OR its flag, not overwrite it.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 0b3b6fe299c471e44ed8713b7a602882626e693f\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Jul 11 22:13:17 2019 +0800\n\n scsi: qla2xxx: Remove unnecessary null check\n \n A null check before dma_pool_destroy is redundant, so remove it. This is\n detected by coccinelle.\n \n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Acked-by: Himanshu Madhani \u003chmadhani@marvell.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 64a38e840ce5940253208eaba40265c73decc4ee\nAuthor: Dave Wysochanski \u003cdwysocha@redhat.com\u003e\nDate: Fri Jul 26 18:33:01 2019 -0400\n\n SUNRPC: Track writers of the \u0027channel\u0027 file to improve cache_listeners_exist\n \n The sunrpc cache interface is susceptible to being fooled by a rogue\n process just reading a \u0027channel\u0027 file. If this happens the kernel\n may think a valid daemon exists to service the cache when it does not.\n For example, the following may fool the kernel:\n cat /proc/net/rpc/auth.unix.gid/channel\n \n Change the tracking of readers to writers when considering whether a\n listener exists as all valid daemon processes either open a channel\n file O_RDWR or O_WRONLY. While this does not prevent a rogue process\n from \"stealing\" a message from the kernel, it does at least improve\n the kernels perception of whether a valid process servicing the cache\n exists.\n \n Signed-off-by: Dave Wysochanski \u003cdwysocha@redhat.com\u003e\n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n\ncommit a912275d322e1f76bbedac2805b7d359acf0cf6e\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Jul 29 12:52:20 2019 -0700\n\n tty: serial: fsl_lpuart: Introduce lpuart_dma_shutdown()\n \n Last steps of .shutdown() code are identical for lpuart and lpuart32\n cases, so move it all into a standalone subroutine.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Bhuvanchandra DV \u003cbhuvanchandra.dv@toradex.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190729195226.8862-19-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d4580bc4b39912f4a7da7c7aa7cdaf80c724734e\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Jul 29 12:52:19 2019 -0700\n\n tty: serial: fsl_lpuart: Drop unnecessary lpuart*_stop_tx()\n \n By the time lpuart_shutdown() calls lpuart_stop_tx() UARTCR2_TE and\n UARTCR2_TIE (which the latter will clear) are already cleared, so that\n function call should effectively be a no-op. Moreso, lpuart_stop_tx()\n is expected to be executed with port spinlock held, which the caller\n doesn\u0027t. Given all that, drop the call to lpuart_stop_tx() in\n lpuart_shutdown().\n \n In case of lpuart32_shutdown()/lpuart32_stop_tx(), TIE won\u0027t even be\n set if lpuart_dma_tx_use is true. Drop it there as well.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Bhuvanchandra DV \u003cbhuvanchandra.dv@toradex.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190729195226.8862-18-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 8a58a7651c886e6d2cbd33a6489327f493c9391d\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Jul 29 12:52:18 2019 -0700\n\n tty: serial: fsl_lpuart: Introduce lpuart_stopped_or_empty()\n \n The check for\n \n uart_circ_empty(xmit) || uart_tx_stopped(\u0026sport-\u003eport)\n \n appears in multiple places in the driver. Move it into a helper\n function.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Bhuvanchandra DV \u003cbhuvanchandra.dv@toradex.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190729195226.8862-17-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 3ef58b18d6d423481ed3fad57edfe4023e4c3b80\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Jul 29 12:52:17 2019 -0700\n\n tty: serial: fsl_lpuart: Use cpu_relax() instead of barrier()\n \n Use cpu_relax() instead of barrier() in a tight polling loops to make\n them a bit more idiomatic. Should also improve things on ARM64 a bit\n since cpu_relax() will expand into \"yield\" instruction there.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Bhuvanchandra DV \u003cbhuvanchandra.dv@toradex.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190729195226.8862-16-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 32ee0fb0c0afb0db7e85370b65e4c8a007a2145a\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Jul 29 12:52:16 2019 -0700\n\n tty: serial: fsl_lpuart: Introduce lpuart_wait_bit_set()\n \n Busy polling on a bit in a register is used in multiple places in the\n driver. Move it into a shared function.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Bhuvanchandra DV \u003cbhuvanchandra.dv@toradex.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190729195226.8862-15-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit fa04d8c1c150528558ac7b42b27ac9380a80a406\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Jul 29 12:52:15 2019 -0700\n\n tty: serial: fsl_lpuart: Use appropriate lpuart32_* I/O funcs\n \n When dealing with 32-bit variant of LPUART IP block appropriate I/O\n helpers have to be used to properly deal with endianness\n differences. Change all of the offending code to do that.\n \n Fixes: a5fa2660d787 (\"tty/serial/fsl_lpuart: Add CONSOLE_POLL support\n for lpuart32.\")\n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Bhuvanchandra DV \u003cbhuvanchandra.dv@toradex.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190729195226.8862-14-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit c12a5975e095a71eede023cefd31d57c92edecf3\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Jul 29 12:52:14 2019 -0700\n\n tty: serial: fsl_lpuart: Clear CSTOPB unconditionally\n \n Clearing CSTOPB bit if it is set is functionally equivalent to jsut\n clearing it unconditionally. Drop unnecessary check.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Bhuvanchandra DV \u003cbhuvanchandra.dv@toradex.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190729195226.8862-13-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 414290fad17593d26c66d1460db98523f3336e93\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Jul 29 12:52:13 2019 -0700\n\n tty: serial: fsl_lpuart: Drop unnecessary extra parenthesis\n \n Drop unnecessary extra parenthesis in the driver.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Bhuvanchandra DV \u003cbhuvanchandra.dv@toradex.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190729195226.8862-12-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 97a3124ddf45396dded1a136ef6b529f5a3e5d81\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Jul 29 12:52:12 2019 -0700\n\n tty: serial: fsl_lpuart: Drop no-op bit opearation\n \n The check for termios-\u003ec_cflag \u0026 CRTSCTS ensure that if we reach else\n branch, CRTSCTS in termios-\u003ec_cflag is already going to be\n cleard. Doing so explicitly there is not necessary. Drop it.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Bhuvanchandra DV \u003cbhuvanchandra.dv@toradex.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190729195226.8862-11-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 48145e144256f250b84bf161681cda51981a758e\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Jul 29 12:52:11 2019 -0700\n\n tty: serial: fls_lpuart: Split shared TX IRQ handler into two\n \n While sharing code for Tx interrupt handler between 8 and 32 bit\n variant of the peripheral saves a bit of code duplication it also adds\n quite a number of lpuart_is_32() checks which makes it harder to\n understand. Move shared bits back into corresponding\n lpuart*_transmit_buffer functions, split lpuart_txint into\n lpuart_txint and lpuart32_txint so we can drop all extra\n lpuart_is_32() check and make the code flow more linear.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Bhuvanchandra DV \u003cbhuvanchandra.dv@toradex.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190729195226.8862-10-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit ec8aec1419366dfd17643749de8a159d610cb418\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Jul 29 12:52:10 2019 -0700\n\n tty: serial: fsl_lpuart: Fix issue in software flow control\n \n Although I haven\u0027t observed this bug in practice, it seems that the\n code for handling x_char of LPUART is pretty much identical to that of\n i.MX. So the fix found in commit 7e2fb5aa8d81 (\"serial: imx: Fix issue\n in software flow control\"):\n \n serial: imx: Fix issue in software flow control\n \n After send out x_char in UART driver, x_char needs to be cleared\n by UART driver itself, otherwise data in TXFIFO can no longer be\n sent out.\n Also tx counter needs to be increased to keep track of correct\n number of transmitted data.\n \n Signed-off-by: Jiada Wang \u003cjiada_wang@mentor.com\u003e\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n \n should apply here as well.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Bhuvanchandra DV \u003cbhuvanchandra.dv@toradex.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190729195226.8862-9-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 1b5e0815d2f3cbd02fcefd95b8974a4dc2658617\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Jul 29 12:52:09 2019 -0700\n\n tty: serial: fsl_lpuart: Drop unnecessary uart_write_wakeup()\n \n Uart_write_wakeup() will already be called as a part of\n lpuart*_transmit_buffer() call, so there doesn\u0027t seem to be a reason\n to call it again right after.\n \n It also appears that second uart_write_wakeup() might potentially\n cause unwanted write wakeup when transmitting an x_char. See commit\n 5e42e9a30cda (\"serial: imx: Fix x_char handling and tx flow control\")\n where this problem was fixed in a very similarly structured i.MX UART\n driver.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Bhuvanchandra DV \u003cbhuvanchandra.dv@toradex.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190729195226.8862-8-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit bdb60b10616f653f9e0a1db1817e9a23d7611738\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Jul 29 12:52:07 2019 -0700\n\n tty: serial: fsl_lpuart: Fix bogus indentation\n \n Fix bogus indentation in rx_dma_timer_init().\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Bhuvanchandra DV \u003cbhuvanchandra.dv@toradex.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190729195226.8862-6-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d1ec5bcc89053acc41a731abb8f7184842f2d6cc\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Jul 29 12:52:06 2019 -0700\n\n tty: serial: fsl_lpuart: Simplify RX/TX IRQ handlers\n \n It appears that lpuart_rxint, lpuart_txint and lpuart32_rxint were\n modelled after identical function found in UART driver for\n i.MX. However, while said functions are used as individual IRQ\n handlers in i.MX driver (in case of i.MX1), it is not the case for\n LPUART. Given that, there\u0027s no need for us to restrict the prototype\n of the handler to irqreturn_t foo(int, void *) and we can drop all of\n uneened boilerplate code by changing it void foo(struct lpuart_port *).\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Bhuvanchandra DV \u003cbhuvanchandra.dv@toradex.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190729195226.8862-5-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 04561fabd23454a4bf6f8952edb91d9c7be7242f\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Jul 29 12:52:05 2019 -0700\n\n tty: serial: fsl_lpuart: Flush HW FIFOs in .flush_buffer\n \n Switching baud rate might cause bogus data to appear in HW\n FIFO. Add code to do a HW FIFO flush to .flush_buffer callback to\n avoid that.\n \n Signed-off-by: Fugang Duan \u003cfugang.duan@nxp.com\u003e\n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Bhuvanchandra DV \u003cbhuvanchandra.dv@toradex.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190729195226.8862-4-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 55a49070ee901176e6ac543cefcb884dcb112772\nAuthor: Stefan Agner \u003cstefan.agner@toradex.com\u003e\nDate: Mon Jul 29 12:52:04 2019 -0700\n\n tty: serial: fsl_lpuart: flush receive FIFO after overruns\n \n After overruns the FIFO pointers become misaligned. This\n typically shows by characters still being stuck in the FIFO\n despite the empty flag being asserted. After the first\n assertion of the overrun flag the empty flag still seems to\n indicate FIFO state correctly and all data can be read.\n However, after another overrun assertion the FIFO seems to\n be off by one such that the last received character is still\n in the FIFO (despite the empty flag being asserted).\n \n Flushing the receive FIFO reinitializes pointers. Hence it\n is recommended to flush the FIFO after overruns, see also:\n https://community.nxp.com/thread/321175\n \n Hence, on assertion of the overrun flag read the remaining\n data from the FIFO and flush buffers.\n \n Signed-off-by: Stefan Agner \u003cstefan.agner@toradex.com\u003e\n Acked-by: Max Krummenacher \u003cmax.krummenacher@toradex.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Bhuvanchandra DV \u003cbhuvanchandra.dv@toradex.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190729195226.8862-3-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7927b970e758522c4c98420a1dd6f941bd0a3e56\nAuthor: Stefan Agner \u003cstefan.agner@toradex.com\u003e\nDate: Mon Jul 29 12:52:03 2019 -0700\n\n tty: serial: fsl_lpuart: fix framing error handling when using DMA\n \n When using DMA framing error get cleared properly. However, due\n to the additional read from the data register, an underflow in\n the receive FIFO buffer occurs (the FIFO pointer gets out of\n sync).\n \n Clear the FIFO in case an underflow has occurred. Also disable the\n receiver during this operation and when reading the data register to\n minimize potential interference.\n \n Signed-off-by: Stefan Agner \u003cstefan.agner@toradex.com\u003e\n Acked-by: Max Krummenacher \u003cmax.krummenacher@toradex.com\u003e\n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Bhuvanchandra DV \u003cbhuvanchandra.dv@toradex.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: linux-imx@nxp.com\n Cc: linux-serial@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Link: https://lore.kernel.org/r/20190729195226.8862-2-andrew.smirnov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit e7a81412af476f332a8ada185404587465c1e79f\nAuthor: Sergey Organov \u003csorganov@gmail.com\u003e\nDate: Fri Jul 26 21:52:41 2019 +0300\n\n serial: imx: get rid of imx_uart_rts_auto()\n \n Called in only one place, for RS232, it only obscures things, as it\n doesn\u0027t go well with 2 similar named functions,\n imx_uart_rts_inactive() and imx_uart_rts_active(), that both are\n RS485-specific.\n \n Reviewed-by: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\n Tested-by: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\n Signed-off-by: Sergey Organov \u003csorganov@gmail.com\u003e\n Link: https://lore.kernel.org/r/1564167161-3972-4-git-send-email-sorganov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit c53c1d876bf0b015690a7a5c7f8f9fe981956ee3\nAuthor: Sergey Organov \u003csorganov@gmail.com\u003e\nDate: Fri Jul 26 21:52:40 2019 +0300\n\n serial: imx: set_mctrl(): correctly restore autoRTS state\n \n imx_uart_set_mctrl() happened to set UCR2_CTSC bit whenever TIOCM_RTS\n was set, no matter if RTS/CTS handshake is enabled or not. Now fixed by\n turning handshake on only when CRTSCTS bit for the port is set.\n \n Reviewed-by: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\n Tested-by: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\n Signed-off-by: Sergey Organov \u003csorganov@gmail.com\u003e\n Link: https://lore.kernel.org/r/1564167161-3972-3-git-send-email-sorganov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 40c3d1cd7e6ef75a4e7f12ce5da1128125ac9555\nAuthor: Sergey Organov \u003csorganov@gmail.com\u003e\nDate: Fri Jul 26 21:52:39 2019 +0300\n\n serial: imx: set_termios(): do not enable autoRTS if RTS is unset\n \n Don\u0027t let receiver hardware automatically control RTS output if it\n was requested to be inactive.\n \n To ensure this, set_termios() shouldn\u0027t set UCR2_CTSC bit if UCR2_CTS\n (\u003dTIOCM_RTS) is cleared. Added corresponding check in imx_uart_rts_auto()\n to fix this.\n \n Acked-by: Uwe Kleine-König \u003cu.kleine-koenig@pengutronix.de\u003e\n Reviewed-by: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\n Tested-by: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\n Signed-off-by: Sergey Organov \u003csorganov@gmail.com\u003e\n Link: https://lore.kernel.org/r/1564167161-3972-2-git-send-email-sorganov@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7b0efe59f42a9aa0a1b7bf1f0fac663984e414f1\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Fri Jul 26 20:28:17 2019 +0300\n\n serial: 8250_pnp: Move to struct dev_pm_ops\n \n The established way to provide PM callbacks is through struct dev_pm_ops\n which is more generic.\n \n Convert driver to use it instead of legacy approach.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190726172817.73253-1-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a1c4423b02b2121108e3ea9580741e0f26309a48\nAuthor: Marcelo Tosatti \u003cmtosatti@redhat.com\u003e\nDate: Wed Jul 3 20:51:29 2019 -0300\n\n cpuidle-haltpoll: disable host side polling when kvm virtualized\n \n When performing guest side polling, it is not necessary to\n also perform host side polling.\n \n So disable host side polling, via the new MSR interface,\n when loading cpuidle-haltpoll driver.\n \n Signed-off-by: Marcelo Tosatti \u003cmtosatti@redhat.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 2cffe9f6b96fece065ee8522673c90e92ef2085d\nAuthor: Marcelo Tosatti \u003cmtosatti@redhat.com\u003e\nDate: Wed Jul 3 20:51:28 2019 -0300\n\n cpuidle: add haltpoll governor\n \n The cpuidle_haltpoll governor, in conjunction with the haltpoll cpuidle\n driver, allows guest vcpus to poll for a specified amount of time before\n halting.\n This provides the following benefits to host side polling:\n \n 1) The POLL flag is set while polling is performed, which allows\n a remote vCPU to avoid sending an IPI (and the associated\n cost of handling the IPI) when performing a wakeup.\n \n 2) The VM-exit cost can be avoided.\n \n The downside of guest side polling is that polling is performed\n even with other runnable tasks in the host.\n \n Results comparing halt_poll_ns and server/client application\n where a small packet is ping-ponged:\n \n host --\u003e 31.33\n halt_poll_ns\u003d300000 / no guest busy spin --\u003e 33.40 (93.8%)\n halt_poll_ns\u003d0 / guest_halt_poll_ns\u003d300000 --\u003e 32.73 (95.7%)\n \n For the SAP HANA benchmarks (where idle_spin is a parameter\n of the previous version of the patch, results should be the\n same):\n \n hpns \u003d\u003d halt_poll_ns\n \n idle_spin\u003d0/ idle_spin\u003d800/ idle_spin\u003d0/\n hpns\u003d200000 hpns\u003d0 hpns\u003d800000\n DeleteC06T03 (100 thread) 1.76 1.71 (-3%) 1.78 (+1%)\n InsertC16T02 (100 thread) 2.14 2.07 (-3%) 2.18 (+1.8%)\n DeleteC00T01 (1 thread) 1.34 1.28 (-4.5%) 1.29 (-3.7%)\n UpdateC00T03 (1 thread) 4.72 4.18 (-12%) 4.53 (-5%)\n \n Signed-off-by: Marcelo Tosatti \u003cmtosatti@redhat.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 7d4daeedd575bbc3c40c87fc6708a8b88c50fe7e\nAuthor: Marcelo Tosatti \u003cmtosatti@redhat.com\u003e\nDate: Wed Jul 3 20:51:27 2019 -0300\n\n governors: unify last_state_idx\n \n Since this field is shared by all governors, move it to\n cpuidle device structure.\n \n Signed-off-by: Marcelo Tosatti \u003cmtosatti@redhat.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit 259231a045616c4101d023a8f4dcc8379af265a6\nAuthor: Marcelo Tosatti \u003cmtosatti@redhat.com\u003e\nDate: Wed Jul 3 20:51:26 2019 -0300\n\n cpuidle: add poll_limit_ns to cpuidle_device structure\n \n Add a poll_limit_ns variable to cpuidle_device structure.\n \n Calculate and configure it in the new cpuidle_poll_time\n function, in case its zero.\n \n Individual governors are allowed to override this value.\n \n Signed-off-by: Marcelo Tosatti \u003cmtosatti@redhat.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit fa86ee90eb1111267de67cb4272b5ce711f18cbb\nAuthor: Marcelo Tosatti \u003cmtosatti@redhat.com\u003e\nDate: Wed Jul 3 20:51:25 2019 -0300\n\n add cpuidle-haltpoll driver\n \n Add a cpuidle driver that calls the architecture default_idle routine.\n \n To be used in conjunction with the haltpoll governor.\n \n Signed-off-by: Marcelo Tosatti \u003cmtosatti@redhat.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n\ncommit f5d974f9d2a811ef08c044b6fce95c94a6a6e19b\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jul 30 15:32:09 2019 +0100\n\n drm/i915/gt: Provide a local intel_context.vm\n \n Track the currently bound address space used by the HW context. Minor\n conversions to use the local intel_context.vm are made, leaving behind\n some more surgery required to make intel_context the primary through the\n selftests.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730143209.4549-2-chris@chris-wilson.co.uk\n\ncommit c082afac86cb31e2a5843336e81a9b89e3b0d664\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jul 30 15:32:08 2019 +0100\n\n drm/i915: Move aliasing_ppgtt underneath its i915_ggtt\n \n The aliasing_ppgtt provides a PIN_USER alias for the global gtt, so move\n it under the i915_ggtt to simplify later transformations to enable\n intel_context.vm.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190730143209.4549-1-chris@chris-wilson.co.uk\n\ncommit 621f7e354fd8f99db0e3f7a0e8cda238caee9f3a\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Wed Jul 24 17:38:48 2019 -0600\n\n PCI: Make pci_set_of_node(), etc private\n \n These interfaces:\n \n void pci_set_of_node(struct pci_dev *dev);\n void pci_release_of_node(struct pci_dev *dev);\n void pci_set_bus_of_node(struct pci_bus *bus);\n void pci_release_bus_of_node(struct pci_bus *bus);\n \n are only used in drivers/pci/ and do not need to be seen by the rest of the\n kernel. Move them to drivers/pci/pci.h so they\u0027re private to the PCI\n subsystem.\n \n Link: https://lore.kernel.org/r/20190724233848.73327-12-skunberg.kelsey@gmail.com\n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit ac6c26da29c12fa511c877c273ed5c939dc9e96c\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Wed Jul 24 17:38:47 2019 -0600\n\n PCI: Make pci_enable_ptm() private\n \n This interface:\n \n int pci_enable_ptm(struct pci_dev *dev, u8 *granularity);\n \n is only used in drivers/pci/ and does not need to be seen by the rest of\n the kernel. Move it to drivers/pci/pci.h so it\u0027s private to the PCI\n subsystem.\n \n Link: https://lore.kernel.org/r/20190724233848.73327-11-skunberg.kelsey@gmail.com\n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit 72bde9ced373ca1e27332fd020d248151319a3d6\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Wed Jul 24 17:38:46 2019 -0600\n\n PCI: Make pcie_set_ecrc_checking(), pcie_ecrc_get_policy() private\n \n These interfaces:\n \n void pcie_set_ecrc_checking(struct pci_dev *dev);\n void pcie_ecrc_get_policy(char *str);\n \n are only used in drivers/pci/ and do not need to be seen by the rest of the\n kernel. Move them to drivers/pci/pci.h so they\u0027re private to the PCI\n subsystem.\n \n Link: https://lore.kernel.org/r/20190724233848.73327-10-skunberg.kelsey@gmail.com\n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit b92b512a435da01b52de07e3dcc2f07a4ad404de\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Wed Jul 24 17:38:45 2019 -0600\n\n PCI: Make pci_ats_init() private\n \n This interface:\n \n void pci_ats_init(struct pci_dev *dev);\n \n is only used in drivers/pci/ and does not need to be seen by the rest of\n the kernel. Move it to drivers/pci/pci.h so it\u0027s private to the PCI\n subsystem.\n \n Link: https://lore.kernel.org/r/20190724233848.73327-9-skunberg.kelsey@gmail.com\n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit 5da78d95785daa9454336a88b2f86a8998f6c739\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Wed Jul 24 17:38:44 2019 -0600\n\n PCI: Make pcie_update_link_speed() private\n \n This interface:\n \n void pcie_update_link_speed(struct pci_bus *bus, u16 link_status);\n \n is only used in drivers/pci/ and does not need to be seen by the rest of\n the kernel. Move it to drivers/pci/pci.h so it\u0027s private to the PCI\n subsystem.\n \n Link: https://lore.kernel.org/r/20190724233848.73327-8-skunberg.kelsey@gmail.com\n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit ecd29c1a38af4ba6e61382591dfe93f06f14ba25\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Wed Jul 24 17:38:43 2019 -0600\n\n PCI: Make pci_bus_get(), pci_bus_put() private\n \n These interfaces:\n \n struct pci_bus *pci_bus_get(struct pci_bus *bus);\n void pci_bus_put(struct pci_bus *bus);\n \n are only used in drivers/pci/ and do not need to be seen by the rest of the\n kernel. Move them to drivers/pci/pci.h so they\u0027re private to the PCI\n subsystem.\n \n Link: https://lore.kernel.org/r/20190724233848.73327-7-skunberg.kelsey@gmail.com\n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit 003d3b2c5f835d897b3b10a13a9e1340630e93f1\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Wed Jul 24 17:38:42 2019 -0600\n\n PCI: Make pci_hotplug_io_size, mem_size, and bus_size private\n \n These symbols:\n \n extern unsigned long pci_hotplug_io_size;\n extern unsigned long pci_hotplug_mem_size;\n extern unsigned long pci_hotplug_bus_size;\n \n are only used in drivers/pci/ and do not need to be seen by the rest of the\n kernel. Move them to drivers/pci/pci.h so they\u0027re private to the PCI\n subsystem.\n \n Link: https://lore.kernel.org/r/20190724233848.73327-6-skunberg.kelsey@gmail.com\n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit 440589dd1068fb36b2fe10ccddf97e43267e3b8d\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Wed Jul 24 17:38:41 2019 -0600\n\n PCI: Make pci_save_vc_state(), pci_restore_vc_state(), etc private\n \n These Virtual Channel interfaces:\n \n int pci_save_vc_state(struct pci_dev *dev);\n void pci_restore_vc_state(struct pci_dev *dev);\n void pci_allocate_vc_save_buffers(struct pci_dev *dev);\n \n are only used in drivers/pci/ and do not need to be seen by the rest of the\n kernel. Move them to drivers/pci/pci.h so they\u0027re private to the PCI\n subsystem.\n \n Link: https://lore.kernel.org/r/20190724233848.73327-5-skunberg.kelsey@gmail.com\n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit 975e1ac173058b8710e5979e97fc1397233301f3\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Wed Jul 24 17:38:40 2019 -0600\n\n PCI: Make pci_get_host_bridge_device(), pci_put_host_bridge_device() private\n \n These interfaces:\n \n struct device *pci_get_host_bridge_device(struct pci_dev *dev);\n void pci_put_host_bridge_device(struct device *dev);\n \n are only used in drivers/pci/ and do not need to be seen by the rest of the\n kernel. Move them to drivers/pci/pci.h so they\u0027re private to the PCI\n subsystem.\n \n Link: https://lore.kernel.org/r/20190724233848.73327-4-skunberg.kelsey@gmail.com\n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit 669696ebbccc40e8096a2ee9809c028fc1538001\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Wed Jul 24 17:38:39 2019 -0600\n\n PCI: Make pci_check_pme_status(), pci_pme_wakeup_bus() private\n \n These interfaces:\n \n bool pci_check_pme_status(struct pci_dev *dev);\n void pci_pme_wakeup_bus(struct pci_bus *bus);\n \n are only used in drivers/pci/ and do not need to be seen by the rest of the\n kernel. Move them to drivers/pci/pci.h so they\u0027re private to the PCI\n subsystem.\n \n Link: https://lore.kernel.org/r/20190724233848.73327-3-skunberg.kelsey@gmail.com\n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit c776dd50196a07a0ef943fcb74e8c86f9a5a20a8\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Wed Jul 24 17:38:38 2019 -0600\n\n PCI: Make PCI_PM_* delay times private\n \n These delay time definitions:\n \n #define PCI_PM_D2_DELAY 200\n #define PCI_PM_D3_WAIT 10\n #define PCI_PM_D3COLD_WAIT 100\n #define PCI_PM_BUS_WAIT 50\n \n are only used in drivers/pci/ and do not need to be seen by the rest of the\n kernel. Move them to drivers/pci/pci.h so they\u0027re private to the PCI\n subsystem.\n \n Link: https://lore.kernel.org/r/20190724233848.73327-2-skunberg.kelsey@gmail.com\n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit 98051ba2b28be8a8ec5e22c01f464cff2d6d5f7c\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Mon Jul 29 22:38:45 2019 -0700\n\n coccinelle: Add script to check for platform_get_irq() excessive prints\n \n Add a coccinelle script to check for the usage of dev_err() after a call\n to platform_get_irq{,_byname}() as it\u0027s redundant now that the function\n already prints an error when it fails.\n \n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Rob Herring \u003crobh@kernel.org\u003e\n Cc: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Cc: Javier Martinez Canillas \u003cjavierm@redhat.com\u003e\n Cc: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Cc: Mark Brown \u003cbroonie@kernel.org\u003e\n Cc: Russell King - ARM Linux \u003clinux@armlinux.org.uk\u003e\n Cc: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Cc: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Cc: Andy Shevchenko \u003candy.shevchenko@gmail.com\u003e\n Cc: Markus Elfring \u003cMarkus.Elfring@web.de\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190730053845.126834-4-swboyd@chromium.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7723f4c5ecdb8d832f049f8483beb0d1081cedf6\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Mon Jul 29 22:38:43 2019 -0700\n\n driver core: platform: Add an error message to platform_get_irq*()\n \n A grep of the kernel shows that many drivers print an error message if\n they fail to get the irq they\u0027re looking for. Furthermore, those drivers\n all decide to print the device name, or not, and the irq they were\n requesting, or not, etc. Let\u0027s consolidate all these error messages into\n the API itself, allowing us to get rid of the error messages in each\n driver.\n \n Cc: Rob Herring \u003crobh@kernel.org\u003e\n Cc: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Cc: Javier Martinez Canillas \u003cjavierm@redhat.com\u003e\n Cc: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Cc: Mark Brown \u003cbroonie@kernel.org\u003e\n Cc: Russell King - ARM Linux \u003clinux@armlinux.org.uk\u003e\n Cc: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Cc: Markus Elfring \u003cMarkus.Elfring@web.de\u003e\n Reviewed-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Reviewed-by: Andy Shevchenko \u003candy.shevchenko@gmail.com\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190730053845.126834-2-swboyd@chromium.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit e3168b874321d04c160c9eb937919eb926ae232f\nAuthor: Daniel Drake \u003cdrake@endlessm.com\u003e\nDate: Mon Jul 29 16:27:39 2019 +0800\n\n platform/x86: asus-wmi: fix CPU fan control on recent products\n \n Previously, asus-wmi was using the AGFN interface and FAN_CTRL device\n for CPU fan control. However, this code has been found to be not fully\n working on some recent products, and having checked the spec, these\n interfaces are marked as being removed from future products currently\n in development.\n \n The replacement appears to be the CPU_FAN device, added in spec version\n 8.3 (March 2014) and present on many modern Asus laptops.\n \n Add support for this device, and use it whenever it is detected.\n The older approach based on AGFN and FAN_CTRL is used as a fallback\n on products that do not have such device.\n \n Other than switching between automatic and full speed, there is\n no fan speed control through this new interface.\n \n Signed-off-by: Daniel Drake \u003cdrake@endlessm.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit f1fc032192d2872caef69fb899e939bf7cc24fc7\nAuthor: Daniel Drake \u003cdrake@endlessm.com\u003e\nDate: Mon Jul 29 16:27:38 2019 +0800\n\n platform/x86: asus-wmi: add a helper for device presence\n \n Factor out the WLAN LED and lightbar LED presence checks into a\n helper function, which will also be used by the upcoming CPU fan device\n support.\n \n Signed-off-by: Daniel Drake \u003cdrake@endlessm.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 2889ffcfc2522d6d25e5bda704275064062bbb21\nAuthor: Daniel Drake \u003cdrake@endlessm.com\u003e\nDate: Mon Jul 29 16:27:37 2019 +0800\n\n platform/x86: asus-wmi: cleanup AGFN fan handling\n \n The asus-wmi driver currently uses the \"AGFN\" interface and\n the FAN_CTRL device for fan control. According to the spec, this\n interface is very dated and marked as pending removal from products\n currently in development.\n \n Clean up the way that the AGFN fan is detected and handled, also\n preparing the driver for the introduction of an alternate fan\n control method needed to support recent Asus products.\n \n Not anticipating further development of this interface, simplify\n the code by dropping any notion of being able to control multiple\n AGFN fans (this was already limited to just a single fan through only\n exposing a single fan in sysfs).\n \n Check for the presence of AGFN fans at probe time, simplifying the code\n flow in asus_hwmon_sysfs_is_visible().\n \n Signed-off-by: Daniel Drake \u003cdrake@endlessm.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 5037d4ddda31c2dbbb018109655f61054b1756dc\nAuthor: Enrico Weigelt \u003cinfo@metux.net\u003e\nDate: Mon Jul 29 16:39:26 2019 +0200\n\n platform/x86: pcengines-apuv2: wire up simswitch gpio as led\n \n The APU3+ boards have two SIM sockets, while only one of them\n can be routed to the mpcie slots at a time. Selection is done\n via simswap gpio.\n \n We currently don\u0027t have a fitting subsystem for those cases yet,\n so just wire it up to a LED for the time being. While this isn\u0027t\n really semantically correct, it\u0027s a good compromise.\n \n Signed-off-by: Enrico Weigelt \u003cinfo@metux.net\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 6a0694b6a244bc43ee79b65d0ae5435fb7b74401\nAuthor: Florian Eckert \u003cfe@dev.tdt.de\u003e\nDate: Fri Jul 26 08:21:15 2019 +0200\n\n platform/x86: pcengines-apuv2: add mpcie reset gpio export\n \n On APUx we have also mpcie2/mpcie3 reset pins. To make it possible to reset\n the ports from the userspace, add the definition to this platform\n device. The gpio can then be exported by the legancy gpio subsystem to\n toggle the mpcie reset pin.\n \n Signed-off-by: Florian Eckert \u003cfe@dev.tdt.de\u003e\n Acked-by: Enrico Weigelt \u003cinfo@metux.net\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit c03f282e506f1e873ef9b270cf245509d68dd53b\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Thu Jul 25 22:05:50 2019 +0300\n\n platform/x86: i2c-multi-instantiate: Use struct_size() helper\n \n One of the more common cases of allocation size calculations is finding\n the size of a structure that has a zero-sized array at the end, along\n with memory for some number of elements for that array.\n \n Make use of the struct_size() helper instead of an open-coded version\n in order to avoid any potential type mistakes.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n\ncommit aaed5c9c2a146640170e0fbc84ca97e352c065ce\nAuthor: Rhys Kidd \u003crhyskidd@gmail.com\u003e\nDate: Wed Jul 24 18:14:15 2019 +1000\n\n platform/x86: dell-wmi: Use existing defined KBD_LED_* magic values\n \n These values have already been defined in platform/x86/dell-smbios.h\n \n Signed-off-by: Rhys Kidd \u003crhyskidd@gmail.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 1be4311063a44ba8f4292a3a4a1f537be1f0b344\nAuthor: Rhys Kidd \u003crhyskidd@gmail.com\u003e\nDate: Wed Jul 24 18:14:14 2019 +1000\n\n platform/x86: dell-wmi: Ignore keyboard backlight change KBD_LED_AUTO_TOKEN\n \n There\u0027s a wmi event generated by dell-wmi when pressing keyboard backlight\n toggle key:\n [1224178.355650] dell_wmi: Unknown key with type 0x0011 and code 0x01e3 pressed\n \n This event is for notification purposes, let\u0027s ignore it.\n \n Signed-off-by: Rhys Kidd \u003crhyskidd@gmail.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit ff32e6e2dfd3cfdcaef63a0688bbfb9259aebe02\nAuthor: Rhys Kidd \u003crhyskidd@gmail.com\u003e\nDate: Wed Jul 24 18:14:13 2019 +1000\n\n platform/x86: dell-wmi: Ignore keyboard backlight change KBD_LED_ON_TOKEN\n \n There\u0027s a wmi event generated by dell-wmi when pressing keyboard backlight\n toggle key:\n [1224203.948894] dell_wmi: Unknown key with type 0x0011 and code 0x01e2 pressed\n \n This event is for notification purposes, let\u0027s ignore it.\n \n Signed-off-by: Rhys Kidd \u003crhyskidd@gmail.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 6fe9363b15767babf3f5b9eb41a664de2cff4030\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 23:34:11 2019 +0300\n\n platform/x86: acer-wmi: Switch to acpi_dev_get_first_match_dev()\n \n The acpi_dev_get_first_match_dev() helper will find and return\n an ACPI device pointer of the first registered device in the system\n by its HID.\n \n Use it instead of open coded variant.\n \n Cc: Chun-Yi Lee \u003cjlee@suse.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit bad9da86f9f10a2e68a1574dd34edde6f8556d5f\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Sun Jul 21 21:11:58 2019 -0600\n\n platform/x86: wmi: Remove acpi_has_method() call\n \n acpi_has_method() is unnecessary within __query_block() and should be\n removed to avoid extra work.\n \n wc_status is initialized to AE_ERROR before the acpi_has_method() call.\n acpi_has_method() and acpi_execute_simple_method() failing due to the\n method not existing will result in the same outcome from __query_block().\n \n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 5599e98fe06a88db55133ae8ac53f69216b82c6a\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Wed Jul 24 20:23:20 2019 +0800\n\n platform/x86: ISST: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 36f3313d6bff91ab2a9e47698c27d15363640a4e\nAuthor: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\nDate: Tue Jul 23 23:18:38 2019 +0100\n\n platform: Add platform_find_device_by_driver() helper\n \n Provide a helper to lookup platform devices by matching device\n driver in order to avoid drivers trying to use platform bus\n internals.\n \n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: \"Heiko Stübner\" \u003cheiko@sntech.de\u003e\n Cc: Inki Dae \u003cinki.dae@samsung.com\u003e\n Cc: \"Rafael J. Wysocki\" \u003crafael@kernel.org\u003e\n Cc: Sandy Huang \u003chjc@rock-chips.com\u003e\n Cc: Seung-Woo Kim \u003csw0312.kim@samsung.com\u003e\n Tested-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Signed-off-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Link: https://lore.kernel.org/r/20190723221838.12024-8-suzuki.poulose@arm.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 6bf85ba9e55f659ddc0747bf1bb504ec6d15f525\nAuthor: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\nDate: Tue Jul 23 23:18:37 2019 +0100\n\n drivers: Add generic helper to match any device\n \n Add a generic helper to match any/all devices. Using this\n introduce new wrappers {bus/driver/class}_find_next_device().\n \n Cc: Elie Morisse \u003csyniurge@gmail.com\u003e\n Cc: \"James E.J. Bottomley\" \u003cjejb@linux.ibm.com\u003e\n Cc: \"Martin K. Petersen\" \u003cmartin.petersen@oracle.com\u003e\n Cc: Nehal Shah \u003cnehal-bakulchandra.shah@amd.com\u003e\n Cc: \"Rafael J. Wysocki\" \u003crafael.j.wysocki@intel.com\u003e\n Cc: Shyam Sundar S K \u003cshyam-sundar.s-k@amd.com\u003e\n Signed-off-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Acked-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e # PCI\n Link: https://lore.kernel.org/r/20190723221838.12024-7-suzuki.poulose@arm.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 00500147cbd3fc51353d0d003eaa9d31c72c0d50\nAuthor: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\nDate: Tue Jul 23 23:18:36 2019 +0100\n\n drivers: Introduce device lookup variants by ACPI_COMPANION device\n \n Add a generic helper to match a device by the ACPI_COMPANION device\n and provide wrappers for the device lookup APIs.\n \n Cc: Len Brown \u003clenb@kernel.org\u003e\n Cc: linux-acpi@vger.kernel.org\n Cc: linux-spi@vger.kernel.org\n Cc: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n Cc: linux-i2c@vger.kernel.org\n Cc: \"Rafael J. Wysocki\" \u003crafael@kernel.org\u003e\n Signed-off-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Acked-by: Mark Brown \u003cbroonie@kernel.org\u003e\n Acked-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e # I2C parts\n Link: https://lore.kernel.org/r/20190723221838.12024-6-suzuki.poulose@arm.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 4495dfdd6193d9712b7b8f5d699d89d5996e6aaa\nAuthor: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\nDate: Tue Jul 23 23:18:35 2019 +0100\n\n drivers: Introduce device lookup variants by device type\n \n Add a helper to match a device by its type and provide wrappers\n for {bus/class/driver}_find_device() APIs.\n \n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\n Cc: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\n Cc: linux-usb@vger.kernel.org\n Cc: Oliver Neukum \u003coneukum@suse.com\u003e\n Cc: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Cc: Tomas Winkler \u003ctomas.winkler@intel.com\u003e\n Cc: \"Rafael J. Wysocki\" \u003crafael@kernel.org\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n Cc: Joe Perches \u003cjoe@perches.com\u003e\n Signed-off-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Link: https://lore.kernel.org/r/20190723221838.12024-5-suzuki.poulose@arm.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 67843bbaf36eb087714f40e783ee78e99e9e4b86\nAuthor: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\nDate: Tue Jul 23 23:18:34 2019 +0100\n\n drivers: Introduce device lookup variants by fwnode\n \n Add a helper to match the firmware node handle of a device and provide\n wrappers for {bus/class/driver}_find_device() APIs to avoid proliferation\n of duplicate custom match functions.\n \n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: Doug Ledford \u003cdledford@redhat.com\u003e\n Cc: Jason Gunthorpe \u003cjgg@ziepe.ca\u003e\n Cc: linux-usb@vger.kernel.org\n Cc: \"Rafael J. Wysocki\" \u003crafael@kernel.org\u003e\n Cc: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n Cc: Joe Perches \u003cjoe@perches.com\u003e\n Cc: Will Deacon \u003cwill.deacon@arm.com\u003e\n Cc: Joerg Roedel \u003cjoro@8bytes.org\u003e\n Signed-off-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Acked-by: Robin Murphy \u003crobin.murphy@arm.com\u003e\n Reviewed-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Reviewed-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190723221838.12024-4-suzuki.poulose@arm.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit cfba5de9b99f8bbb8b4ea11b3049784e78b8759b\nAuthor: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\nDate: Tue Jul 23 23:18:33 2019 +0100\n\n drivers: Introduce device lookup variants by of_node\n \n Introduce wrappers for {bus/driver/class}_find_device() to\n locate devices by its of_node.\n \n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: devicetree@vger.kernel.org\n Cc: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Cc: Frank Rowand \u003cfrowand.list@gmail.com\u003e\n Cc: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Cc: Liam Girdwood \u003clgirdwood@gmail.com\u003e\n Cc: linux-i2c@vger.kernel.org\n Cc: linux-rockchip@lists.infradead.org\n Cc: linux-spi@vger.kernel.org\n Cc: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Srinivas Kandagatla \u003csrinivas.kandagatla@linaro.org\u003e\n Cc: Takashi Iwai \u003ctiwai@suse.com\u003e\n Cc: Alan Tull \u003catull@kernel.org\u003e\n Cc: linux-fpga@vger.kernel.org\n Cc: Peter Rosin \u003cpeda@axentia.se\u003e\n Cc: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Cc: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: Andrew Lunn \u003candrew@lunn.ch\u003e\n Cc: Liam Girdwood \u003clgirdwood@gmail.com\u003e\n Cc: \"Rafael J. Wysocki\" \u003crafael@kernel.org\u003e\n Cc: Thor Thayer \u003cthor.thayer@linux.intel.com\u003e\n Cc: Jiri Slaby \u003cjslaby@suse.com\u003e\n Cc: Andrew Lunn \u003candrew@lunn.ch\u003e\n Cc: Peter Rosin \u003cpeda@axentia.se\u003e\n Signed-off-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Acked-by: Lee Jones \u003clee.jones@linaro.org\u003e\n Acked-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e # I2C part\n Acked-by: Moritz Fischer \u003cmdf@kernel.org\u003e # For FPGA part\n Acked-by: Mark Brown \u003cbroonie@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190723221838.12024-3-suzuki.poulose@arm.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 6cda08a20dbde45b021091230c8a359fa08c5103\nAuthor: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\nDate: Tue Jul 23 23:18:32 2019 +0100\n\n drivers: Introduce device lookup variants by name\n \n Add a helper to match the device name for device lookup. Also\n reuse this generic exported helper for the existing bus_find_device_by_name().\n and add similar variants for driver/class.\n \n Cc: Alessandro Zummo \u003ca.zummo@towertech.it\u003e\n Cc: Alexander Aring \u003calex.aring@gmail.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Arnd Bergmann \u003carnd@arndb.de\u003e\n Cc: Dan Murphy \u003cdmurphy@ti.com\u003e\n Cc: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\n Cc: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\n Cc: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Cc: Lee Jones \u003clee.jones@linaro.org\u003e\n Cc: linux-leds@vger.kernel.org\n Cc: linux-rtc@vger.kernel.org\n Cc: linux-usb@vger.kernel.org\n Cc: linux-wpan@vger.kernel.org\n Cc: Maxime Coquelin \u003cmcoquelin.stm32@gmail.com\u003e\n Cc: Pavel Machek \u003cpavel@ucw.cz\u003e\n Cc: Peter Oberparleiter \u003coberpar@linux.ibm.com\u003e\n Cc: \"Rafael J. Wysocki\" \u003crafael@kernel.org\u003e\n Cc: Stefan Schmidt \u003cstefan@datenfreihafen.org\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: \"Rafael J. Wysocki\" \u003crafael.j.wysocki@intel.com\u003e\n Signed-off-by: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Reviewed-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\n Acked-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190723221838.12024-2-suzuki.poulose@arm.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit dcf8f7ecb36185f4cf96ad1f098704eb690f792e\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Sun Jul 28 19:06:31 2019 -0500\n\n usb: phy: ab8500-usb: Mark expected switch fall-throughs\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warnings:\n \n drivers/usb/phy/phy-ab8500-usb.c: In function \u0027ab8500_usb_link_status_update\u0027:\n drivers/usb/phy/phy-ab8500-usb.c:424:9: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n event \u003d UX500_MUSB_RIDB;\n ~~~~~~^~~~~~~~~~~~~~~~~\n drivers/usb/phy/phy-ab8500-usb.c:425:2: note: here\n case USB_LINK_NOT_CONFIGURED_8500:\n ^~~~\n drivers/usb/phy/phy-ab8500-usb.c:440:9: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n event \u003d UX500_MUSB_RIDC;\n ~~~~~~^~~~~~~~~~~~~~~~~\n drivers/usb/phy/phy-ab8500-usb.c:441:2: note: here\n case USB_LINK_STD_HOST_NC_8500:\n ^~~~\n drivers/usb/phy/phy-ab8500-usb.c:459:9: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n event \u003d UX500_MUSB_RIDA;\n ~~~~~~^~~~~~~~~~~~~~~~~\n drivers/usb/phy/phy-ab8500-usb.c:460:2: note: here\n case USB_LINK_HM_IDGND_8500:\n ^~~~\n drivers/usb/phy/phy-ab8500-usb.c: In function \u0027ab8505_usb_link_status_update\u0027:\n drivers/usb/phy/phy-ab8500-usb.c:332:9: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n event \u003d UX500_MUSB_RIDB;\n ~~~~~~^~~~~~~~~~~~~~~~~\n drivers/usb/phy/phy-ab8500-usb.c:333:2: note: here\n case USB_LINK_NOT_CONFIGURED_8505:\n ^~~~\n drivers/usb/phy/phy-ab8500-usb.c:352:9: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n event \u003d UX500_MUSB_RIDC;\n ~~~~~~^~~~~~~~~~~~~~~~~\n drivers/usb/phy/phy-ab8500-usb.c:353:2: note: here\n case USB_LINK_STD_HOST_NC_8505:\n ^~~~\n drivers/usb/phy/phy-ab8500-usb.c:370:9: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n event \u003d UX500_MUSB_RIDA;\n ~~~~~~^~~~~~~~~~~~~~~~~\n drivers/usb/phy/phy-ab8500-usb.c:371:2: note: here\n case USB_LINK_HM_IDGND_8505:\n ^~~~\n \n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190729000631.GA24165@embeddedor\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 9e22cb82998253072ec05ee1f8b06daf5ca50800\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Jul 29 17:22:01 2019 -0500\n\n usb: host: ohci-tmio: Mark expected switch fall-throughs\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warning (Building: arm):\n \n drivers/usb/host/ohci-tmio.c: In function ‘tmio_stop_hc’:\n ./include/linux/device.h:1499:2: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n _dev_err(dev, dev_fmt(fmt), ##__VA_ARGS__)\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n drivers/usb/host/ohci-tmio.c:99:4: note: in expansion of macro ‘dev_err’\n dev_err(\u0026dev-\u003edev, \"Unsupported amount of ports: %d\\n\", ohci-\u003enum_ports);\n ^~~~~~~\n In file included from drivers/usb/host/ohci-hcd.c:1257:0:\n drivers/usb/host/ohci-tmio.c:100:3: note: here\n case 3:\n ^~~~\n drivers/usb/host/ohci-tmio.c:101:7: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n pm |\u003d CCR_PM_USBPW3;\n ^\n drivers/usb/host/ohci-tmio.c:102:3: note: here\n case 2:\n ^~~~\n drivers/usb/host/ohci-tmio.c:103:7: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n pm |\u003d CCR_PM_USBPW2;\n ^\n drivers/usb/host/ohci-tmio.c:104:3: note: here\n case 1:\n ^~~~\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Link: https://lore.kernel.org/r/20190729222201.GA19408@embeddedor\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 314de2f6b577a55772674dd2e4907a8c94b00975\nAuthor: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\nDate: Fri Jul 26 10:14:53 2019 +0200\n\n ARM: dts: exynos: Use standard arrays of generic PHYs for EHCI/OHCI devices\n \n Move USB PHYs to a standard arrays for Exynos EHCI/OHCI devices. This\n resolves the conflict between Exynos EHCI/OHCI sub-nodes and generic USB\n device bindings. Once the Exynos EHCI/OHCI sub-nodes are removed, the\n boards can finally provide sub-nodes for the USB devices using generic USB\n device bindings.\n \n Suggested-by: Måns Rullgård \u003cmans@mansr.com\u003e\n Signed-off-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Link: https://lore.kernel.org/r/20190726081453.9456-4-m.szyprowski@samsung.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 214b606e90dd087d190b7307183ec8c16f500371\nAuthor: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\nDate: Fri Jul 26 10:14:52 2019 +0200\n\n usb: exynos: add support for getting PHYs from the standard dt array\n \n Add the code for getting generic PHYs from standard device tree array\n from the main controller device node. This is a first step in resolving\n the conflict between Exynos EHCI/OHCI sub-nodes and generic USB device\n bindings. Later the sub-nodes currently used for assigning PHYs to root\n ports of the controller will be removed making a place for the generic\n USB device bindings nodes.\n \n Suggested-by: Måns Rullgård \u003cmans@mansr.com\u003e\n Signed-off-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Acked-by: Alan Stern \u003cstern@rowland.harvard.edu\u003e\n Acked-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190726081453.9456-3-m.szyprowski@samsung.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit c27989cc536b5e0ba179d7ab50ebc2a06fddf663\nAuthor: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\nDate: Fri Jul 26 10:14:51 2019 +0200\n\n dt-bindings: switch Exynos EHCI/OHCI bindings to use array of generic PHYs\n \n Commit 69bec7259853 (\"USB: core: let USB device know device node\") added\n support for attaching devicetree node for USB devices. Those nodes are\n children of their USB host controller. However Exynos EHCI and OHCI\n driver bindings already define child-nodes for each physical root hub\n port and assigns respective PHY controller and parameters to them. This\n leads to the conflict. A workaround for it has been merged as commit\n 01d4071486fe (\"usb: exynos: add workaround for the USB device bindings\n conflict\"), but it disabled support for USB device binding for Exynos\n EHCI/OHCI controllers.\n \n To resolve it properly, lets move PHYs from the sub-nodes to a standard\n array under the \u0027phys\u0027 property.\n \n Suggested-by: Måns Rullgård \u003cmans@mansr.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Acked-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190726081453.9456-2-m.szyprowski@samsung.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a56277216637c359e9125b9c16be09f21ed2a395\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Jul 29 12:37:20 2019 +0100\n\n drm/i915: Inline engine-\u003einit_context into its caller\n \n We only use the init_context vfunc once while recording the default\n context state, and we use the same sequence in each backend (eliding\n steps that do not apply). Remove the vfunc for simplicity and\n de-duplication.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190729113720.24830-1-chris@chris-wilson.co.uk\n\ncommit fb583c8eeeb1fd57e24ef41ed94c9112067aeac9\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Tue Jul 30 11:28:57 2019 +0200\n\n driver core: Fix creation of device links with PM-runtime flags\n \n After commit 515db266a9da (\"driver core: Remove device link creation\n limitation\"), if PM-runtime flags are passed to device_link_add(), it\n will fail (returning NULL) due to an overly restrictive flags check\n introduced by that commit.\n \n Fix this issue by extending the check in question to cover the\n PM-runtime flags too.\n \n Fixes: 515db266a9da (\"driver core: Remove device link creation limitation\")\n Reported-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Tested-by: Jon Hunter \u003cjonathanh@nvidia.com\u003e\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Tested-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Tested-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Link: https://lore.kernel.org/r/7674989.cD04D8YV3U@kreacher\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b33b556ce59b5239a7b73c495209f90ec990e60c\nAuthor: Christian König \u003cchristian.koenig@amd.com\u003e\nDate: Mon Jul 22 14:56:25 2019 +0200\n\n drm/syncobj: fix leaking dma_fence in drm_syncobj_query_ioctl\n \n We need to check the context number instead if the previous sequence to detect\n an error and if an error is detected we need to drop the reference to the\n current fence or otherwise would leak it.\n \n Signed-off-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Fixes: 27b575a9aa2f (\"drm/syncobj: add timeline payload query ioctl v6\")\n Link: https://patchwork.freedesktop.org/patch/319123/\n\ncommit 10a08fd65ec1a68ccd86b19ec822ed5f2e50113f\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Tue Jul 30 11:55:59 2019 +0200\n\n ACPI: PM: Set up EC GPE for system wakeup from drivers that need it\n \n The EC GPE needs to be set up for system wakeup only if there is a\n driver depending on it, either intel-hid or intel-vbtn, bound to a\n button device that is expected to wake up the system from sleep (such\n as the power button on some Dell systems, like the XPS13 9360). It\n doesn\u0027t need to be set up for waking up the system from sleep in any\n other cases and whether or not it is expected to wake up the system\n from sleep doesn\u0027t depend on whether or not the LPS0 device is\n present in the ACPI namespace.\n \n For this reason, rearrange the ACPI suspend-to-idle code to make the\n drivers depending on the EC GPE wakeup take care of setting it up and\n decouple that from the LPS0 device handling.\n \n While at it, make intel-hid and intel-vbtn prepare for system wakeup\n only if they are allowed to wake up the system from sleep by user\n space (via sysfs).\n \n [Note that acpi_ec_mark_gpe_for_wake() and acpi_ec_set_gpe_wake_mask()\n are there to prevent the EC GPE from being disabled by the\n acpi_enable_all_wakeup_gpes() call in acpi_s2idle_prepare(), so on\n systems with either intel-hid or intel-vbtn this change doesn\u0027t\n affect any interactions with the hardware or platform firmware.]\n \n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Reviewed-by: Andy Shevchenko \u003candy.shevchenko@gmail.com\u003e\n\ncommit 4de09ffcf4f622876668b76c8e416f4143b32a9e\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Jul 29 17:27:52 2019 -0500\n\n drm: sti: Mark expected switch fall-throughs\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warning (Building: arm):\n \n drivers/gpu/drm/sti/sti_hdmi.c: In function ‘hdmi_audio_configure’:\n drivers/gpu/drm/sti/sti_hdmi.c:851:13: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n audio_cfg |\u003d HDMI_AUD_CFG_CH78_VALID;\n drivers/gpu/drm/sti/sti_hdmi.c:852:2: note: here\n case 6:\n ^~~~\n drivers/gpu/drm/sti/sti_hdmi.c:853:13: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n audio_cfg |\u003d HDMI_AUD_CFG_CH56_VALID;\n drivers/gpu/drm/sti/sti_hdmi.c:854:2: note: here\n case 4:\n ^~~~\n drivers/gpu/drm/sti/sti_hdmi.c:855:13: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n audio_cfg |\u003d HDMI_AUD_CFG_CH34_VALID | HDMI_AUD_CFG_8CH;\n drivers/gpu/drm/sti/sti_hdmi.c:856:2: note: here\n case 2:\n ^~~~\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190729222752.GA20277@embeddedor\n\ncommit 0a05b3d7fee511de8a1c9dffc3d86f1df55e1737\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Wed Jul 24 03:04:06 2019 +0900\n\n arm64: dts: renesas: ulcb: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit b726a9e3ac06fa05e2b40d78750882348658a248\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Wed Jul 24 03:03:35 2019 +0900\n\n arm64: dts: renesas: ulcb-kf: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n [geert: Sort i2c slave nodes]\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 44d2266c8c6945fad704d62617d8772089d82b4e\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Wed Jul 24 03:03:10 2019 +0900\n\n arm64: dts: renesas: salvator-common: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n [geert: Sort i2c slave nodes]\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 93c3438f7604385f6a7dc6ccfc880acb6c0819a5\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Wed Jul 24 03:02:41 2019 +0900\n\n arm64: dts: renesas: r8a7796: salvator-xs: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 74b1435efb0b6e14eba9feeb5a5ec89d2e2652a5\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Wed Jul 24 03:02:08 2019 +0900\n\n arm64: dts: renesas: r8a7796: salvator-x: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit a23bc44c2deaf659c565b522bed6e78ce573a16d\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Tue Jul 23 02:23:08 2019 +0900\n\n arm64: dts: renesas: r8a7795: salvator-xs: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 480160ee3c05e2729a048139dc37af98739b3ddf\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Tue Jul 23 02:22:52 2019 +0900\n\n arm64: dts: renesas: r8a7795: salvator-x: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 34e776ea894a063670d61d2e6f16795d84a3ccc1\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Tue Jul 23 02:22:21 2019 +0900\n\n arm64: dts: renesas: r8a7795-es1: salvator-x: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit da3db1c84629b6a4e7b780dfdfb6ec2112ad0d56\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Thu Jul 18 23:45:37 2019 +0900\n\n arm64: dts: renesas: r8a77965: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 3bb350f2249277abedbc9b6cc1de3a715e981c3b\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Thu Jul 18 23:43:53 2019 +0900\n\n arm64: dts: renesas: r8a7795-es1: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit c7a895fc5dd58b9e9378f420b1bbecf6737aeb55\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Thu Jul 18 23:43:26 2019 +0900\n\n arm64: dts: renesas: r8a7795: Sort nodes\n \n Sort nodes.\n \n If node address is present\n * Sort by node address, grouping all nodes with the same compat string\n and sorting the group alphabetically.\n Else\n * Sort alphabetically\n \n This should not have any run-time effect.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 1032a2af93f58ff12996f6122a8a488979ea22a8\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Mon Jul 29 17:08:47 2019 +0300\n\n drm/i915: use upstream version of header tests\n \n Throw out our local hacks of header tests now that the more generic\n kbuild versions are upstream.\n \n At least for now, continue to keep the header tests behind\n CONFIG_DRM_I915_WERROR\u003dy knob.\n \n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190729140847.18557-1-jani.nikula@intel.com\n\ncommit d9e5ebac18a122f53ee87671f62d42b0d9e9a491\nAuthor: Jeremy Cline \u003cjcline@redhat.com\u003e\nDate: Mon Jul 29 10:37:30 2019 -0400\n\n Documentation: gpio: fix function links in the HTML docs\n \n The shorthand [_data] and [devm_] cause the HTML documentation to not\n link to the function documentation properly. This expands the references\n to the complete function names with the exception of\n devm_gpiochip_remove() which was dropped by commit 48207d7595d2 (\"gpio:\n drop devm_gpiochip_remove()\").\n \n Signed-off-by: Jeremy Cline \u003cjcline@redhat.com\u003e\n Signed-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n\ncommit 2c0e81ec21db971eb9eb6adf987f9f15aec75968\nAuthor: Michael Straube \u003cstraube.linux@gmail.com\u003e\nDate: Fri Jul 26 20:04:48 2019 +0200\n\n staging: rtl8188eu: cleanup comparsion to NULL in usb_halinit.c\n \n Use if(!x) instead of if(x \u003d\u003d NULL).\n Reported by checkpatch.\n \n Signed-off-by: Michael Straube \u003cstraube.linux@gmail.com\u003e\n Acked-by: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\n Link: https://lore.kernel.org/r/20190726180448.2290-6-straube.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 1f5bfba56f90afe8e158b655ecdd6eac78205dae\nAuthor: Michael Straube \u003cstraube.linux@gmail.com\u003e\nDate: Fri Jul 26 20:04:47 2019 +0200\n\n staging: rtl8188eu: add spaces around \u0027-\u0027 and \u0027*\u0027 in usb_halinit.c\n \n Add spaces around \u0027-\u0027 and \u0027*\u0027 to improve readability and follow kernel\n coding style. Reported by checkpatch.\n \n Signed-off-by: Michael Straube \u003cstraube.linux@gmail.com\u003e\n Acked-by: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\n Link: https://lore.kernel.org/r/20190726180448.2290-5-straube.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 8f4686cb7db0839c9d42e3b8e813bf55a9161d29\nAuthor: Michael Straube \u003cstraube.linux@gmail.com\u003e\nDate: Fri Jul 26 20:04:46 2019 +0200\n\n staging: rtl8188eu: add spaces around \u0027\u003c\u003c\u0027 and \u0027\u003e\u003e\u0027 in usb_halinit.c\n \n Add spaces around \u0027\u003c\u003c\u0027 and \u0027\u003e\u003e\u0027 to improve readability and follow\n kernel coding style. Reported by checkpatch.\n \n Signed-off-by: Michael Straube \u003cstraube.linux@gmail.com\u003e\n Acked-by: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\n Link: https://lore.kernel.org/r/20190726180448.2290-4-straube.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit fe6c0d20bb911b9c31ee5d724428551b3aa2f7a9\nAuthor: Michael Straube \u003cstraube.linux@gmail.com\u003e\nDate: Fri Jul 26 20:04:45 2019 +0200\n\n staging: rtl8188eu: add spaces around \u0027|\u0027 in usb_halinit.c\n \n Add spaces around \u0027|\u0027 to improve readability and follow kernel\n coding style. Reported by checkpatch.\n \n Signed-off-by: Michael Straube \u003cstraube.linux@gmail.com\u003e\n Acked-by: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\n Link: https://lore.kernel.org/r/20190726180448.2290-3-straube.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 6d7ff7361137cced8aaf7de568ad2fb8c00ec532\nAuthor: Michael Straube \u003cstraube.linux@gmail.com\u003e\nDate: Fri Jul 26 20:04:44 2019 +0200\n\n staging: rtl8188eu: add spaces around \u0027\u0026\u0027 in usb_halinit.c\n \n Add spaces around \u0027\u0026\u0027 to improve readability and follow kernel\n coding style. Reported by checkpatch.\n \n Signed-off-by: Michael Straube \u003cstraube.linux@gmail.com\u003e\n Acked-by: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\n Link: https://lore.kernel.org/r/20190726180448.2290-2-straube.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 3455db0e056516f576c3d4a8baf320d814a6e202\nAuthor: Michael Straube \u003cstraube.linux@gmail.com\u003e\nDate: Fri Jul 26 20:04:43 2019 +0200\n\n staging: rtl8188eu: add spaces around \u0027+\u0027 in usb_halinit.c\n \n Add spaces around \u0027+\u0027 to improve readability and follow kernel\n coding style. Reported by checkpatch.\n \n Signed-off-by: Michael Straube \u003cstraube.linux@gmail.com\u003e\n Acked-by: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\n Link: https://lore.kernel.org/r/20190726180448.2290-1-straube.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 25f12201a4ba67451a8be71db83dd734b2f62b30\nAuthor: Adham Abozaeid \u003cadham.abozaeid@microchip.com\u003e\nDate: Thu Jul 25 21:31:37 2019 +0000\n\n staging: wilc1000: remove manual sleep mode\n \n manual sleep mode was used to put the wilc1000 chip in sleep while in\n disconnected state. This is taken care of in the firmware\n \n Signed-off-by: Adham Abozaeid \u003cadham.abozaeid@microchip.com\u003e\n Link: https://lore.kernel.org/r/20190725213125.2810-7-adham.abozaeid@microchip.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7e0cf3e3bfe94d030c3d3d79f8af8f0e365fdf1c\nAuthor: Adham Abozaeid \u003cadham.abozaeid@microchip.com\u003e\nDate: Thu Jul 25 21:31:36 2019 +0000\n\n staging: wilc1000: apply power management regardless of open interfaces\n \n Don\u0027t ignore power management if 2 interfaces are open\n \n Signed-off-by: Adham Abozaeid \u003cadham.abozaeid@microchip.com\u003e\n Link: https://lore.kernel.org/r/20190725213125.2810-6-adham.abozaeid@microchip.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit ded11ab1818440a9e5b35f2cff78a76148762dbd\nAuthor: Adham Abozaeid \u003cadham.abozaeid@microchip.com\u003e\nDate: Thu Jul 25 21:31:35 2019 +0000\n\n staging: wilc1000: avoid overriding powersave state\n \n Don\u0027t override powersave state with respect to the open interfaces and\n let the firmware take care of when it\u0027s appropriate to do so\n \n Signed-off-by: Adham Abozaeid \u003cadham.abozaeid@microchip.com\u003e\n Link: https://lore.kernel.org/r/20190725213125.2810-5-adham.abozaeid@microchip.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 5a28d3fd407bf5225e7ab4eea1c5779a8a71c966\nAuthor: Adham Abozaeid \u003cadham.abozaeid@microchip.com\u003e\nDate: Thu Jul 25 21:31:34 2019 +0000\n\n staging: wilc1000: remove unused members\n \n remove obtaining_ip from struct wilc_vif\n \n Signed-off-by: Adham Abozaeid \u003cadham.abozaeid@microchip.com\u003e\n Link: https://lore.kernel.org/r/20190725213125.2810-4-adham.abozaeid@microchip.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 98af3e7ed6831f001fee70584c08f39264002b1d\nAuthor: Adham Abozaeid \u003cadham.abozaeid@microchip.com\u003e\nDate: Thu Jul 25 21:31:34 2019 +0000\n\n staging: wilc1000: remove ip timeout timer\n \n during_ip_timer is not required after removing the code that disables\n powersave while the ip is being obtained.\n Its handler clear_during_ip is also removed\n \n Signed-off-by: Adham Abozaeid \u003cadham.abozaeid@microchip.com\u003e\n Link: https://lore.kernel.org/r/20190725213125.2810-3-adham.abozaeid@microchip.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f68dc00ee710ba56ee2112ea1184d7c5d3c40370\nAuthor: Adham Abozaeid \u003cadham.abozaeid@microchip.com\u003e\nDate: Thu Jul 25 21:31:33 2019 +0000\n\n staging: wilc1000: remove inetaddr notifier\n \n Driver registers an inetaddr notifier to disable powersave while the ip\n address is being obtained which should be controlled only by cfg80211.\n \n Signed-off-by: Adham Abozaeid \u003cadham.abozaeid@microchip.com\u003e\n Link: https://lore.kernel.org/r/20190725213125.2810-2-adham.abozaeid@microchip.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 8e7cb7352ec85e9e4fbbd7bfe6c5c5a6806f70e3\nAuthor: Bharath Vedartham \u003clinux.bhar@gmail.com\u003e\nDate: Thu Jul 25 18:14:18 2019 +0530\n\n staging: kpc2000: Convert put_page() to put_user_page*()\n \n For pages that were retained via get_user_pages*(), release those pages\n via the new put_user_page*() routines, instead of via put_page().\n \n This is part a tree-wide conversion, as described in commit fc1d8e7cca2d\n (\"mm: introduce put_user_page*(), placeholder versions\").\n \n Cc: Ira Weiny \u003cira.weiny@intel.com\u003e\n Cc: John Hubbard \u003cjhubbard@nvidia.com\u003e\n Cc: Jérôme Glisse \u003cjglisse@redhat.com\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Matt Sickler \u003cMatt.Sickler@daktronics.com\u003e\n Cc: devel@driverdev.osuosl.org\n Cc: linux-kernel@vger.kernel.org\n Cc: linux-mm@kvack.org\n Reviewed-by: John Hubbard \u003cjhubbard@nvidia.com\u003e\n Signed-off-by: Bharath Vedartham \u003clinux.bhar@gmail.com\u003e\n Link: https://lore.kernel.org/r/1564058658-3551-1-git-send-email-linux.bhar@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 36475456f3530491460453baae591b0bfd2e94be\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Jul 26 13:48:03 2019 +0100\n\n staging: rtl8188eu: remove redundant assignment to rtstatus\n \n Variable rtstatus is initialized to a value that is never read and it\n is re-assigned later. The initialization is redundant and can be\n removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190726124803.11349-1-colin.king@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 91e590851f51064a2f4cd11cf6ea01c21c4490bc\nAuthor: Abhinav Jain \u003ccrazypsychild@gmail.com\u003e\nDate: Fri Jul 26 19:01:46 2019 +0000\n\n Staging: rtl8192e: fixed a function prototype definition issue\n \n Added the identifier name in the function prototype definition.\n \n Signed-off-by: Abhinav Jain \u003ccrazypsychild@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190726190146.10875-1-ubuntu@ip-172-31-129-142.ec2.internal\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit c47f77eed0da6f829042a266a2f33ef31c250969\nAuthor: Michael Straube \u003cstraube.linux@gmail.com\u003e\nDate: Thu Jul 25 19:09:22 2019 +0200\n\n staging: rtl8188eu: replace hal_EfusePgCheckAvailableAddr()\n \n Function hal_EfusePgCheckAvailableAddr() contains just a single if\n test. Remove the function and replace the call to it with the if test.\n \n Signed-off-by: Michael Straube \u003cstraube.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190725170922.16465-1-straube.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 9d6b3ee001cba29ad6d8d5f31953926504c67a1d\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Thu Jul 25 20:38:56 2019 +0200\n\n Staging: fbtft: Fix some typo. pdc8544 --\u003e pcd8544\n \n The driver is related to \u0027pcd8544\u0027.\n However, 2 strings are about pdc8544 (c and d switched)\n Fix it.\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Link: https://lore.kernel.org/r/20190725183856.17616-1-christophe.jaillet@wanadoo.fr\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a20eefaee64610541d125cd2c10fa25e278f0bee\nAuthor: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\nDate: Fri Jul 26 13:22:14 2019 +0200\n\n staging: most: Use DEFINE_SPINLOCK() instead of struct spinlock\n \n For spinlocks the type spinlock_t should be used instead of \"struct\n spinlock\".\n \n Use DEFINE_SPINLOCK() and spare the run time initialization\n \n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190704153803.12739-5-bigeasy@linutronix.de\n Link: https://lore.kernel.org/r/alpine.DEB.2.21.1907261319100.1791@nanos.tec.linutronix.de\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 5f251e986de05af8a177bfc632980cc64a4d3e7e\nAuthor: Jerry Lin \u003cwahahab11@gmail.com\u003e\nDate: Mon Jul 29 10:35:44 2019 +0800\n\n staging: olpc_dcon: Remove TODO item\n \n All uses of old GPIO API have been converted to new API.\n This item can be removed from TODO file.\n \n Signed-off-by: Jerry Lin \u003cwahahab11@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190729023544.GA25930@compute1\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit db0e6a573e9c54151cdc828e80ab82849802337f\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Sat Jul 27 09:35:24 2019 +0800\n\n staging: vc04_services: fix unused-but-set-variable warning\n \n Fix gcc used-but-set-variable warning:\n \n drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c: In function vchiq_release_internal:\n drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c:2827:16: warning:\n variable local_entity_uc set but not used [-Wunused-but-set-variable]\n drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c:2827:6: warning:\n variable local_uc set but not used [-Wunused-but-set-variable]\n \n Remove the unused variables \u0027local_entity_uc\u0027 and \u0027local_uc\u0027\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Acked-by: Stefan Wahren \u003cwahrenst@gmx.net\u003e\n Link: https://lore.kernel.org/r/20190727013524.33168-1-yuehaibing@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 149726d3d78a6053445aaff6361374a0b67b7ac1\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Jul 26 22:09:59 2019 +0800\n\n staging: rtl8723bs: os_dep: remove two set but not used variables\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/staging/rtl8723bs//os_dep/osdep_service.c: In function \u0027rtw_buf_free\u0027:\n drivers/staging/rtl8723bs//os_dep/osdep_service.c:321:6: warning:\n variable \u0027ori_len\u0027 set but not used [-Wunused-but-set-variable]\n drivers/staging/rtl8723bs//os_dep/ioctl_linux.c: In function \u0027rtw_ioctl_wext_private\u0027:\n drivers/staging/rtl8723bs//os_dep/ioctl_linux.c:4915:6: warning:\n variable \u0027num_priv\u0027 set but not used [-Wunused-but-set-variable]\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190726140959.15008-1-yuehaibing@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 30833b4b4bc6c3da04e3478552a25a7da4ebc92c\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Jul 26 22:07:34 2019 +0800\n\n staging: rtl8723bs: remove set but not used variable \u0027pszBBRegMpFile\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/staging/rtl8723bs/hal/rtl8723b_phycfg.c: In function phy_BB8723b_Config_ParaFile:\n drivers/staging/rtl8723bs/hal/rtl8723b_phycfg.c:436:77:\n warning: variable pszBBRegMpFile set but not used [-Wunused-but-set-variable]\n \n It is never used so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190726140734.39564-1-yuehaibing@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 07ca9aa4a4b7fdd3c29cc9f8b07a2d1b8e0b6ba5\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Jul 26 22:03:21 2019 +0800\n\n staging: rtl8723bs: remove set but not used variable \u0027bWifiBusy\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n In function halbtc8723b1ant_TdmaDurationAdjustForAcl:\n drivers/staging/rtl8723bs/hal/HalBtc8723b1Ant.c:1761:7: warning:\n variable bWifiBusy set but not used [-Wunused-but-set-variable]\n \n It is never used so can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190726140321.19200-1-yuehaibing@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 9091373ab7ea27cad381ce71aa37de6b9e687e81\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Tue Jul 30 14:43:47 2019 +0900\n\n gpio: remove less important #ifdef around declarations\n \n The whole struct/function declarations in this header are surrounded\n by #ifdef.\n \n As far as I understood, the motivation of this is probably to break\n the build earlier if a driver misses to select or depend on correct\n CONFIG options in Kconfig.\n \n Since commit 94bed2a9c4ae (\"Add -Werror-implicit-function-declaration\")\n no one cannot call functions that have not been declared.\n \n So, I see some benefit in doing this in the cost of uglier headers.\n \n In reality, it would not be so easy to catch missed \u0027select\u0027 or\n \u0027depends on\u0027 because GPIOLIB, GPIOLIB_IRQCHIP etc. are already selected\n by someone else eventually. So, this kind of error, if any, will be\n caught by randconfig bots.\n \n In summary, I am not a big fan of cluttered #ifdef nesting, and this\n does not matter for normal developers. The code readability wins.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n\ncommit 83554cb94515467cfff10f70ec858d60bbbab8ca\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Jul 29 20:49:44 2019 -0500\n\n ALSA: sparc: Mark expected switch fall-throughs\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warnings (Building: sparc64):\n \n sound/sparc/dbri.c: In function ‘reverse_bytes’:\n sound/sparc/dbri.c:582:5: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n b \u003d ((b \u0026 0xffff0000) \u003e\u003e 16) | ((b \u0026 0x0000ffff) \u003c\u003c 16);\n ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n sound/sparc/dbri.c:583:2: note: here\n case 16:\n ^~~~\n sound/sparc/dbri.c:584:5: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n b \u003d ((b \u0026 0xff00ff00) \u003e\u003e 8) | ((b \u0026 0x00ff00ff) \u003c\u003c 8);\n ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n sound/sparc/dbri.c:585:2: note: here\n case 8:\n ^~~~\n sound/sparc/dbri.c:586:5: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n b \u003d ((b \u0026 0xf0f0f0f0) \u003e\u003e 4) | ((b \u0026 0x0f0f0f0f) \u003c\u003c 4);\n ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n sound/sparc/dbri.c:587:2: note: here\n case 4:\n ^~~~\n sound/sparc/dbri.c:588:5: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n b \u003d ((b \u0026 0xcccccccc) \u003e\u003e 2) | ((b \u0026 0x33333333) \u003c\u003c 2);\n ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n sound/sparc/dbri.c:589:2: note: here\n case 2:\n ^~~~\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 8e774e023565c682c67617da5141c7395c8a5fac\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Jul 29 15:54:54 2019 -0500\n\n sound: dmasound_atari: Mark expected switch fall-through\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warning (Building: m68k):\n \n sound/oss/dmasound/dmasound_atari.c: warning: this statement may fall\n through [-Wimplicit-fallthrough\u003d]: \u003d\u003e 1449:24\n \n Notice that, in this particular case, the code comment is\n modified in accordance with what GCC is expecting to find.\n \n Reported-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n Tested-by: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 51dbe89768289c32406f47ad89957b79220c1b63\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Jul 29 17:49:33 2019 -0500\n\n cpufreq: ti-cpufreq: Mark expected switch fall-through\n \n Mark switch cases where we are expecting to fall through.\n \n This patch fixes the following warning (Building: arm):\n \n drivers/cpufreq/ti-cpufreq.c: In function ‘dra7_efuse_xlate’:\n drivers/cpufreq/ti-cpufreq.c:79:20: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n calculated_efuse |\u003d DRA7_EFUSE_HIGH_MPU_OPP;\n drivers/cpufreq/ti-cpufreq.c:80:2: note: here\n case DRA7_EFUSE_HAS_OD_MPU_OPP:\n ^~~~\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit b6fac0b46a1a76024698d240f0f9aac552f897b7\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Mon Sep 17 14:01:58 2018 +0300\n\n net/mlx5e: Protect tc flow table with mutex\n \n TC flow table is created when first flow is added, and destroyed when last\n flow is removed. This assumes that all accesses to the table are externally\n synchronized with rtnl lock. To remove dependency on rtnl lock, add new\n mutex mlx5e_tc_table-\u003et_lock and use it to protect the flow table.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Jianbo Liu \u003cjianbol@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit fa833bd52b7b58d8311bf6a0be29a32e62a1f289\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Tue Mar 12 11:40:12 2019 +0200\n\n net/mlx5e: Rely on rcu instead of rtnl lock when getting upper dev\n \n Function netdev_master_upper_dev_get() generates warning if caller doesn\u0027t\n hold rtnl lock. Modify rules update path to use rcu version of that\n function.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Jianbo Liu \u003cjianbol@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 0e18134f4f9f1e5c75b63b84d250b116c76d5116\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Tue Sep 11 15:48:59 2018 +0300\n\n net/mlx5e: Eswitch, use state_lock to synchronize vlan change\n \n esw-\u003estate_lock is already used to protect vlan vport configuration change.\n However, all preparation and correctness checks, and code that sets vport\n data are not protected by this lock and assume external synchronization by\n rtnl lock. In order to remove dependency on rtnl lock, extend\n esw-\u003estate_lock protection to whole eswitch vlan add/del functions.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Jianbo Liu \u003cjianbol@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 525e84bea52be145b0ae863ff8d242ce273e3bc6\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Sun Nov 18 13:12:59 2018 +0200\n\n net/mlx5e: Eswitch, change offloads num_flows type to atomic64\n \n Eswitch implements its own locking by means of state_lock mutex and\n multiple fine-grained lock in containing data structures, and is supposed\n to not rely on rtnl lock. However, eswitch offloads num_flows type is a\n regular long long integer and cannot be modified concurrently. This is an\n implicit assumptions that mlx5 tc is serialized (by rtnl lock or any other\n means). In order to remove implicit dependency on rtnl lock, change\n num_flows type to atomic64 to allow concurrent modifications.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Jianbo Liu \u003cjianbol@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit ad86755b18d5edf1956f6d25c844f27289216877\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Wed Mar 13 19:50:24 2019 +0200\n\n net/mlx5e: Protect unready flows with dedicated lock\n \n In order to remove dependency on rtnl lock for protecting unready_flows\n list when reoffloading unready flows on workqueue, extend representor\n uplink private structure with dedicated \u0027unready_flows_lock\u0027 mutex. Take\n the lock in all users of unready_flows list before accessing it. Implement\n helper functions to add and delete unready flow.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Jianbo Liu \u003cjianbol@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit c5d326b29603bad4e5342801414b7302ba010209\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Wed Oct 10 13:51:10 2018 +0300\n\n net/mlx5e: Protect tc flows hashtable with rcu\n \n In order to remove dependency on rtnl lock, access to tc flows hashtable\n must be explicitly protected from concurrent flows removal.\n \n Extend tc flow structure with rcu to allow concurrent parallel access. Use\n rcu read lock to safely lookup flow in tc flows hash table, and take\n reference to it. Use rcu free for flow deletion to accommodate concurrent\n stats requests.\n \n Add new DELETED flow flag. Imlement new flow_flag_test_and_set() helper\n that is used to set a flag and return its previous value. Use it to\n atomically set the flag in mlx5e_delete_flower() to guarantee that flow can\n only be deleted once, even when same flow is deleted concurrently by\n multiple tasks.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Jianbo Liu \u003cjianbol@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 226f2ca3075ab20abf69c68235e8ba9ac2dab596\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Thu Nov 8 20:01:35 2018 +0200\n\n net/mlx5e: Change flow flags type to unsigned long\n \n To remove dependency on rtnl lock and allow concurrent modification of\n \u0027flags\u0027 field of tc flow structure, change flow flag type to unsigned long\n and use atomic bit ops for reading and changing the flags. Implement\n auxiliary functions for setting, resetting and getting specific flag, and\n for checking most often used flag values.\n \n Always set flags with smp_mb__before_atomic() to ensure that all\n mlx5e_tc_flow are updated before concurrent readers can read new flags\n value. Rearrange all code paths to actually set flow-\u003erule[] pointers\n before setting the OFFLOADED flag. On read side, use smp_mb__after_atomic()\n when accessing flags to ensure that offload-related flow fields are only\n read after the flags.\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Jianbo Liu \u003cjianbol@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 5a7e5bcb663d46d9cfe7d86d5a8ede91338275cb\nAuthor: Vlad Buslov \u003cvladbu@mellanox.com\u003e\nDate: Thu Nov 8 17:46:06 2018 +0200\n\n net/mlx5e: Extend tc flow struct with reference counter\n \n With new classifier type that doesn\u0027t require rtnl lock, following\n invariant holds:\n - Filter with specified cookie created only once.\n - Filter with specified cookie deleted only once.\n - Stats updates can be performed in parallel to each other.\n \n Extend tc flow with rcu and reference counter. To protect from concurrent\n delete, get reference to tc flow when:\n - Reading flow stats.\n - Accessing flow in neigh update handler.\n - Accessing flow in neigh update used value handler.\n \n Only free flow when reference counter reached zero. Modify flow cleanup to\n account for flows that could be not fully initialized by checking if flow\n is actually in the list of corresponding mod_hdr, hairpin and encap\n entries. Don\u0027t cleanup flow directly in case of error to allow concurrent\n neigh update (neigh update will be modified to always take reference to\n flow when using it).\n \n Signed-off-by: Vlad Buslov \u003cvladbu@mellanox.com\u003e\n Reviewed-by: Jianbo Liu \u003cjianbol@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 233fd21211bd0c77109460811eefa8d10dcfc5e7\nAuthor: Eli Britstein \u003celibr@mellanox.com\u003e\nDate: Mon Jun 3 08:57:13 2019 +0000\n\n net/mlx5e: Simplify get_route_and_out_devs helper function\n \n The helper function has \"if\" branches that do the same. Merge them to\n simplify the code.\n \n Signed-off-by: Eli Britstein \u003celibr@mellanox.com\u003e\n Reviewed-by: Roi Dayan \u003croid@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit aae67158da3677cde569919d66c6521f7b3a4502\nAuthor: wenxu \u003cwenxu@ucloud.cn\u003e\nDate: Sat Jul 27 22:59:55 2019 +0800\n\n net/mlx5e: Fix unnecessary flow_block_cb_is_busy call\n \n When call flow_block_cb_is_busy. The indr_priv is guaranteed to\n NULL ptr. So there is no need to call flow_bock_cb_is_busy.\n \n Fixes: 0d4fd02e7199 (\"net: flow_offload: add flow_block_cb_is_busy() and use it\")\n Signed-off-by: wenxu \u003cwenxu@ucloud.cn\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 79ce39be1d63cc919062ffdacea583c90f3364f6\nAuthor: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\nDate: Tue Jun 25 11:17:18 2019 -0700\n\n net/mlx5e: Improve ethtool rxnfc callback structure\n \n Don\u0027t choose who implements the rxnfc \"get/set\" callbacks according to\n CONFIG_MLX5_EN_RXNFC, instead have the callbacks always available and\n delegate to a function of a different driver module when needed\n (en_fs_ethtool.c), have stubs in en/fs.h to fallback to when\n en_fs_ethtool.c is compiled out, to avoid complications and ifdefs in\n en_main.c.\n \n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 4240196776572a133a54d631ce18b447279af2b5\nAuthor: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\nDate: Fri May 24 12:24:43 2019 -0700\n\n net/mlx5e: Avoid warning print when not required\n \n When disabling CQE compression in favor of time-stamping, don\u0027t show a\n warning when CQE compression is already disabled.\n \n Reviewed-by: Tariq Toukan \u003ctariqt@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 842a2eb28f14b184c7a653448f402f5273e6ff5c\nAuthor: Huy Nguyen \u003chuyn@mellanox.com\u003e\nDate: Thu Jun 27 09:24:57 2019 -0500\n\n net/mlx5e: Print a warning when LRO feature is dropped or not allowed\n \n When user enables LRO via ethtool and if the RQ mode is legacy,\n mlx5e_fix_features drops the request without any explanation.\n Add netdev_warn to cover this case.\n \n Fixes: 6c3a823e1e9c (\"net/mlx5e: RX, Remove HW LRO support in legacy RQ\")\n Signed-off-by: Huy Nguyen \u003chuyn@mellanox.com\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 5d21595b17f693dce313b58f9617a8ea2a45b1b1\nAuthor: Zhou Yanjie \u003czhouyanjie@zoho.com\u003e\nDate: Sun Jul 14 11:53:56 2019 +0800\n\n pinctrl: Ingenic: Add pinctrl driver for X1500.\n \n Add support for probing the pinctrl-ingenic driver on the\n X1500 Soc from Ingenic.\n \n Signed-off-by: Zhou Yanjie \u003czhouyanjie@zoho.com\u003e\n Link: https://lore.kernel.org/r/1563076436-5338-7-git-send-email-zhouyanjie@zoho.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 9b6084d80a46496a1a8ff066b990a05582d6c811\nAuthor: Zhou Yanjie \u003czhouyanjie@zoho.com\u003e\nDate: Sun Jul 14 11:53:55 2019 +0800\n\n dt-bindings: pinctrl: Add X1500 bindings.\n \n Add the pinctrl bindings for the X1500 Soc from Ingenic.\n \n Signed-off-by: Zhou Yanjie \u003czhouyanjie@zoho.com\u003e\n Link: https://lore.kernel.org/r/1563076436-5338-6-git-send-email-zhouyanjie@zoho.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit fe1ad5eedc916bd18c678dfeae9438f8affd69de\nAuthor: Zhou Yanjie \u003czhouyanjie@zoho.com\u003e\nDate: Sun Jul 14 11:53:54 2019 +0800\n\n pinctrl: Ingenic: Add pinctrl driver for X1000 and X1000E.\n \n Add support for probing the pinctrl-ingenic driver on the\n X1000 Soc and the X1000E Soc from Ingenic.\n \n Signed-off-by: Zhou Yanjie \u003czhouyanjie@zoho.com\u003e\n Link: https://lore.kernel.org/r/1563076436-5338-5-git-send-email-zhouyanjie@zoho.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 6835ad54731f3ad4ca7edeef8b2fd1a0d1396044\nAuthor: Zhou Yanjie \u003czhouyanjie@zoho.com\u003e\nDate: Sun Jul 14 11:53:53 2019 +0800\n\n dt-bindings: pinctrl: Add X1000 and X1000E bindings.\n \n Add the pinctrl bindings for the X1000 Soc and\n the X1000E Soc from Ingenic.\n \n Signed-off-by: Zhou Yanjie \u003czhouyanjie@zoho.com\u003e\n Link: https://lore.kernel.org/r/1563076436-5338-4-git-send-email-zhouyanjie@zoho.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 0257595a5cf4ff32c1e2071692210b980e3bbb18\nAuthor: Zhou Yanjie \u003czhouyanjie@zoho.com\u003e\nDate: Sun Jul 14 11:53:52 2019 +0800\n\n pinctrl: Ingenic: Add pinctrl driver for JZ4760 and JZ4760B.\n \n Add support for probing the pinctrl-ingenic driver on the\n JZ4760 Soc and the JZ4760B Soc from Ingenic.\n \n Signed-off-by: Zhou Yanjie \u003czhouyanjie@zoho.com\u003e\n Link: https://lore.kernel.org/r/1563076436-5338-3-git-send-email-zhouyanjie@zoho.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit d0d6d8364ef6b2f46c4751a50690dc7aa4994f0d\nAuthor: Zhou Yanjie \u003czhouyanjie@zoho.com\u003e\nDate: Sun Jul 14 11:53:51 2019 +0800\n\n dt-bindings: pinctrl: Add JZ4760 and JZ4760B bindings.\n \n Add the pinctrl bindings for the JZ4760 Soc and\n the JZ4760B Soc from Ingenic.\n \n Signed-off-by: Zhou Yanjie \u003czhouyanjie@zoho.com\u003e\n Link: https://lore.kernel.org/r/1563076436-5338-2-git-send-email-zhouyanjie@zoho.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit a0b447c18c4279cf0aa67bd62f258a01295855e7\nAuthor: Andrew Jeffery \u003candrew@aj.id.au\u003e\nDate: Fri Jul 12 15:47:21 2019 +0930\n\n dt-bindings: pinctrl: aspeed: Strip unnecessary quotes\n \n Rob pointed out that we didn\u0027t need the noise of the quotes in a\n separate review[1], so strip them out for consistency and avoid setting\n a bad example.\n \n [1] https://lists.ozlabs.org/pipermail/linux-aspeed/2019-July/002009.html\n \n Signed-off-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Link: https://lore.kernel.org/r/20190712061721.26645-1-andrew@aj.id.au\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 123a039d0d54de6d5bafd551e7aa17569e13e315\nAuthor: Michael Petlan \u003cmpetlan@redhat.com\u003e\nDate: Fri Jul 19 12:08:37 2019 +0200\n\n perf vendor events power9: Added missing event descriptions\n \n Documentation source:\n \n https://wiki.raptorcs.com/w/images/6/6b/POWER9_PMU_UG_v12_28NOV2018_pub.pdf\n \n Signed-off-by: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Reviewed-by: Madhavan Srinivasan \u003cmaddy@linux.vnet.ibm.com\u003e\n Cc: Ananth N Mavinakayanahalli \u003cananth@linux.vnet.ibm.com\u003e\n Cc: Carl Love \u003ccel@us.ibm.com\u003e\n Cc: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n Cc: Naveen N. Rao \u003cnaveen.n.rao@linux.vnet.ibm.com\u003e\n Cc: Paul Clarke \u003cpc@us.ibm.com\u003e\n Cc: Sukadev Bhattiprolu \u003csukadev@linux.vnet.ibm.com\u003e\n Cc: linuxppc-dev@ozlabs.org\n LPU-Reference: 20190719100837.7503-1-mpetlan@redhat.com\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit f4f48e9c1adb49f732ac0abc4b2513f2b62a10cb\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:25:06 2019 +0200\n\n libperf: Initial documentation\n \n Add initial drafts of documentation files, hugely unfinished.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-80-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 02266a2d9cf7e04bf3d4b4457654839dc253f605\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:25:05 2019 +0200\n\n libperf: Add perf_evsel__enable/disable test\n \n Add simple perf_evsel enable/disable test together with evsel counter\n reading interface.\n \n Committer testing:\n \n # make -C tools/perf/lib tests\n make: Entering directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n LINK test-cpumap-a\n LINK test-threadmap-a\n LINK test-evlist-a\n LINK test-evsel-a\n LINK test-cpumap-so\n LINK test-threadmap-so\n LINK test-evlist-so\n LINK test-evsel-so\n running static:\n - running test-cpumap.c...OK\n - running test-threadmap.c...OK\n - running test-evlist.c...OK\n - running test-evsel.c...OK\n running dynamic:\n - running test-cpumap.c...OK\n - running test-threadmap.c...OK\n - running test-evlist.c...OK\n - running test-evsel.c...OK\n make: Leaving directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n #\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Tested-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-79-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 6bda376ff416cf713773d38b743953a1a9bc0603\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:25:04 2019 +0200\n\n libperf: Add perf_evlist__enable/disable test\n \n Add simple perf_evlist enable/disable test together with evlist counter\n reading interface.\n \n Committer testing:\n \n # make -C tools/perf/lib tests\n make: Entering directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n LINK test-cpumap-a\n LINK test-threadmap-a\n LINK test-evlist-a\n LINK test-evsel-a\n LINK test-cpumap-so\n LINK test-threadmap-so\n LINK test-evlist-so\n LINK test-evsel-so\n running static:\n - running test-cpumap.c...OK\n - running test-threadmap.c...OK\n - running test-evlist.c...OK\n - running test-evsel.c...OK\n running dynamic:\n - running test-cpumap.c...OK\n - running test-threadmap.c...OK\n - running test-evlist.c...OK\n - running test-evsel.c...OK\n make: Leaving directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n #\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Tested-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-78-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit bb5133ae4d404a8a08d8a94dfdea1c477c93e842\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:25:03 2019 +0200\n\n libperf: Add perf_evsel tests\n \n Add 2 simple perf_evsel tests to test counters reading interface through\n the struct evsel object.\n \n Committer testing:\n \n # make -C tools/perf/lib tests\n make: Entering directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n LINK test-cpumap-a\n LINK test-threadmap-a\n LINK test-evlist-a\n LINK test-evsel-a\n LINK test-cpumap-so\n LINK test-threadmap-so\n LINK test-evlist-so\n LINK test-evsel-so\n running static:\n - running test-cpumap.c...OK\n - running test-threadmap.c...OK\n - running test-evlist.c...OK\n - running test-evsel.c...OK\n running dynamic:\n - running test-cpumap.c...OK\n - running test-threadmap.c...OK\n - running test-evlist.c...OK\n - running test-evsel.c...OK\n make: Leaving directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n #\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Tested-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-77-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 8ded5425fa71e2f7f60eb59d64ecdba80582b641\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:25:02 2019 +0200\n\n libperf: Add perf_evlist test\n \n Add 2 simple perf_evlist tests to test counters reading interface\n through the struct evlist object.\n \n Committer testing:\n \n # make -C tools/perf/lib tests\n make: Entering directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n LINK test-cpumap-a\n LINK test-threadmap-a\n LINK test-evlist-a\n LINK test-cpumap-so\n LINK test-threadmap-so\n LINK test-evlist-so\n running static:\n - running test-cpumap.c...OK\n - running test-threadmap.c...OK\n - running test-evlist.c...OK\n running dynamic:\n - running test-cpumap.c...OK\n - running test-threadmap.c...OK\n - running test-evlist.c...OK\n make: Leaving directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n #\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Tested-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-76-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 43d6976365d5f90de487e8f9f49ab21775ae84f4\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:25:01 2019 +0200\n\n libperf: Add perf_thread_map test\n \n Add simple perf_thread_map tests.\n \n Committer testing:\n \n $ make O\u003d/tmp/build/perf -C tools/perf/lib tests\n make: Entering directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n LINK test-cpumap-a\n LINK test-threadmap-a\n LINK test-cpumap-so\n LINK test-threadmap-so\n running static:\n - running test-cpumap.c...OK\n - running test-threadmap.c...OK\n running dynamic:\n - running test-cpumap.c...OK\n - running test-threadmap.c...OK\n make: Leaving directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n $\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Tested-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-75-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit c0e730456ae8a864701f5ca4f6c2e23010e4b04a\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:25:00 2019 +0200\n\n libperf: Add perf_cpu_map test\n \n Add simple perf_cpu_map tests.\n \n Committer testing:\n \n One has to build it in the source tree, a limitation that should be\n fixed in followup patches:\n \n $ make O\u003d/tmp/build/perf -C tools/perf/lib\n make: Entering directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n LINK /tmp/build/perf/libperf.so.0.0.1\n GEN /tmp/build/perf/libperf.pc\n make: Leaving directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n $ make O\u003d/tmp/build/perf -C tools/perf/lib tests\n make: Entering directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n LINK test-cpumap-a\n gcc: error: ../libperf.a: No such file or directory\n make[1]: *** [Makefile:22: test-cpumap-a] Error 1\n make: *** [Makefile:115: tests] Error 2\n make: Leaving directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n [acme@quaco perf]$ make -C tools/perf/lib\n make: Entering directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n HOSTCC fixdep.o\n HOSTLD fixdep-in.o\n LINK fixdep\n CC core.o\n CC cpumap.o\n CC threadmap.o\n CC evsel.o\n CC evlist.o\n CC zalloc.o\n CC xyarray.o\n CC lib.o\n LD libperf-in.o\n AR libperf.a\n LINK libperf.so.0.0.1\n GEN libperf.pc\n make: Leaving directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n $ make O\u003d/tmp/build/perf -C tools/perf/lib tests\n make: Entering directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n LINK test-cpumap-a\n LINK test-cpumap-so\n running static:\n - running test-cpumap.c...OK\n running dynamic:\n - running test-cpumap.c...OK\n make: Leaving directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n $\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Tested-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-74-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 6a94b52a71b7d3ca3ec47c194f7916b306cb26ef\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:59 2019 +0200\n\n libperf: Add tests support\n \n Adding simple test framework, now empty.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-73-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 0a64d7091efde161a7d0fa385ed5c3bdb72ecdf9\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:58 2019 +0200\n\n libperf: Add install targets\n \n Add install targets (mostly copied from tools/lib/bpf), it\u0027s now\n possible to install libperf with:\n \n $ make DESTDIR\u003d/tmp/krava install\n INSTALL libperf.a\n INSTALL libperf.so\n INSTALL libperf.so.0\n INSTALL libperf.so.0.0.1\n INSTALL headers\n INSTALL libperf.pc\n \n $ find /tmp/krava/\n /tmp/krava/\n /tmp/krava/include\n /tmp/krava/include/perf\n /tmp/krava/include/perf/evsel.h\n /tmp/krava/include/perf/evlist.h\n /tmp/krava/include/perf/threadmap.h\n /tmp/krava/include/perf/cpumap.h\n /tmp/krava/include/perf/core.h\n /tmp/krava/lib64\n /tmp/krava/lib64/pkgconfig\n /tmp/krava/lib64/pkgconfig/libperf.pc\n /tmp/krava/lib64/libperf.so.0.0.1\n /tmp/krava/lib64/libperf.so.0\n /tmp/krava/lib64/libperf.so\n /tmp/krava/lib64/libperf.a\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Tested-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-72-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 384c4ad192a01fe235c7ac9e9fd8605d12a807e8\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:57 2019 +0200\n\n libperf: Add perf_evsel__attr() function\n \n Add a perf_evsel__attr() function to get attr pointer from a perf_evsel\n instance.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-71-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit fcc97c3e7a9d6fd3fda7674f52fb005f4e8453e7\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:56 2019 +0200\n\n libperf: Adopt perf_evlist__enable()/disable() functions from perf\n \n Adopt the following functions from tools/perf:\n \n perf_evlist__enable()\n perf_evlist__disable()\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-70-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 80dc2b3e257cbd62e1cd5b18a6581f231c828c81\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:55 2019 +0200\n\n libperf: Adopt simplified perf_evlist__open()/close() functions from tools/perf\n \n Add the following functions:\n \n perf_evlist__open()\n perf_evlist__close()\n \n It\u0027s a simplified version of perf\u0027s evlist__open() without the sampling\n id index calculations. We can try to merge it in the future when we need\n it in some new libperf user.\n \n Also adopt some helper evlist traversing macros. In the future we can\n remove them from util/evlist.h, but that requires also some other\n changes.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-69-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 0ff1a0fdf52cffa998eee4303e02780d39b2b09c\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:54 2019 +0200\n\n libperf: Add perf_evsel__cpus()/threads() functions\n \n Add the following functions:\n \n perf_evsel__cpus()\n perf_evsel__threads()\n \n to access the evsel\u0027s cpus and threads objects.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-68-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 09145d26b608e886415396e9277ae08f0617d21b\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:53 2019 +0200\n\n libperf: Add perf_cpu_map__for_each_cpu() macro\n \n Add the following macro to libperf:\n \n perf_cpu_map__for_each_cpu()\n \n And its related functions:\n \n perf_cpu_map__cpu()\n perf_cpu_map__nr()\n \n That will allow hiding how it is implemented.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-67-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit a00571fda6091b5268d05e87d0797efe2db1920a\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:52 2019 +0200\n\n libperf: Adopt perf_evsel__enable()/disable()/apply_filter() functions\n \n Move the following functions:\n \n evsel__enable()\n evsel__disable()\n evsel__apply_filter()\n \n to libperf with the following names:\n \n perf_evsel__enable()\n perf_evsel__disable()\n perf_evsel__apply_filter()\n \n Export only perf_evsel__enable()/disable(), keeping the\n perf_evsel__apply_filter() one private for the moment.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-66-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 5c30af92f2b1e9d844e1ae3243e4adcd7753d4c1\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:51 2019 +0200\n\n libperf: Adopt perf_evsel__read() function from tools/perf\n \n Move the perf_evsel__read() function to libperf as a public interface\n together with struct perf_counts_values for returning counter values.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-65-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 88761fa1f1e3fb2df86727ac99f88abf2ac7e00b\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:50 2019 +0200\n\n libperf: Adopt simplified perf_evsel__close() function from tools/perf\n \n Add perf_evsel__close() function to libperf while keeping a tools/perf\n specific evsel__close() to free ids.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-64-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 50a4e6fa450c4e5b688814a7ec8236d0de6e38bf\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:49 2019 +0200\n\n libperf: Adopt simplified perf_evsel__open() function from tools/perf\n \n Add a perf_evsel__open() function to libperf.\n \n It\u0027s a simplified version of evsel__open() without the fallback\n mechanism.\n \n We can try to merge it in the future to libperf, but it has many\n details, lets start simple, requiring the latest kernel, perf should\n continue using its evsel__open() version, continuing to support running\n on older kernels when possible.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-63-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit b8eca4d761c57fcf691a8063cd562f205645d11f\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:48 2019 +0200\n\n libperf: Adopt perf_evsel__alloc_fd() function from tools/perf\n \n Move the perf_evsel__alloc_fd() function from perf to libperf.\n \n It\u0027s not exported.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-62-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit c03538b1f1a7e0e996a8d6feb20cf001d4b14939\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:47 2019 +0200\n\n libperf: Adopt the readn()/writen() functions from tools/perf\n \n Move the readn()/writen() functions into libperf.\n \n Keep those non-namespaced names because they will be shared only between\n perf and libperf.\n \n Again, these are not exported functions.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-61-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 5643b1a59e581ac3f66d36caba8124313cc446c0\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:46 2019 +0200\n\n libperf: Move nr_members from perf\u0027s evsel to libperf\u0027s perf_evsel\n \n Move the nr_members member from perf\u0027s evsel to libperf\u0027s perf_evsel.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-60-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 9dfcb7599084382884fec6d0fd9ca33945fa7578\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:45 2019 +0200\n\n libperf: Move fd array from perf\u0027s evsel to lobperf\u0027s perf_evsel class\n \n Move the fd array from perf\u0027s evsel to libperf\u0027s perf_evsel class.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-59-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 4b247fa7314ce48282f3da400a9ffb7f3fd3f863\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:44 2019 +0200\n\n libperf: Adopt xyarray class from perf\n \n Move the xyarray class from perf to libperf, because it\u0027s going to be\n used in both.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-58-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 453fa03090a64c0e0a561f10dfd5e8747796949c\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:43 2019 +0200\n\n libperf: Add perf_evlist__set_maps() function\n \n Move the evlist__set_maps() function from tools/perf to libperf.\n \n Committer notes:\n \n Fix up reject due to earlier inversion in calling perf_evlist__init().\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-57-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 03617c22e31f32cbf0e4797e216db898fb898d90\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:42 2019 +0200\n\n libperf: Add threads to struct perf_evlist\n \n Move threads from tools/perf\u0027s evlist to libperf\u0027s perf_evlist struct.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-56-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit f72f901d90b00aaf2a6c1335b41311687b3f2dec\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:41 2019 +0200\n\n libperf: Add cpus to struct perf_evlist\n \n Move cpus from tools/perf\u0027s evlist to libperf\u0027s perf_evlist struct.\n \n Committer notes:\n \n Fixed up this one:\n \n tools/perf/arch/arm/util/cs-etm.c\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-55-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit ec903f264f0184a0aba62b42d7717c61f1893450\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:40 2019 +0200\n\n libperf: Add has_user_cpus to struct perf_evlist\n \n Move has_user_cpus from tools/perf\u0027s evlist to libbperf\u0027s perf_evlist struct.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-54-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit af663bd01beaff8d9514199fcc1b239902a77de5\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:39 2019 +0200\n\n libperf: Add threads to struct perf_evsel\n \n Move \u0027threads\u0027 from tools/perf\u0027s evsel to libperf\u0027s perf_evsel struct.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-53-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit fe1f61b37ffada9fc7ec2c9d4ca5376b5a797dbc\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:38 2019 +0200\n\n libperf: Add own_cpus to struct perf_evsel\n \n Move own_cpus from tools/perf\u0027s evsel to libbpf\u0027s perf_evsel.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-52-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit d400bd3abf2cc68df2df32047d3533faf690f404\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:37 2019 +0200\n\n libperf: Add cpus to struct perf_evsel\n \n Mov the \u0027cpus\u0027 field from tools/perf\u0027s evsel to libperf\u0027s perf_evsel.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-51-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit b9358ee95ec65fe7e2c4dc12e4d3da4aeee0d8fc\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:36 2019 +0200\n\n libperf: Add perf_evsel__delete() function\n \n Add the perf_evsel__delete() function to delete a perf_evsel instance.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-50-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 57f0c3b6e13ae822ba02dd37563c8e6956a47141\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:35 2019 +0200\n\n libperf: Add perf_evlist__delete() function\n \n Add the perf_evlist__delete() function to delete a perf_evlist instance.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-49-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 651bf38ce10a65ef8efb901fc33187127c023e97\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:34 2019 +0200\n\n libperf: Add perf_evlist__for_each_evsel() iterator\n \n Add a perf_evlist__for_each_evsel() macro to iterate perf_evsel objects\n in evlist.\n \n Introduce the perf_evlist__next() function to do that without exposing\n \u0027struct perf_evlist\u0027 internals.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-48-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 63bd5dfa69658c459d08a6ee6bfebbd4a91cf24d\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:33 2019 +0200\n\n libperf: Add perf_evsel__new() function\n \n Add a perf_evsel__new() function to create and init a perf_evsel struct\n dynamicaly.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-47-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 634912d61ccc6bfeebb87716c276fbea20f63bdc\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:32 2019 +0200\n\n libperf: Add perf_evlist__new() function\n \n Add perf_evlist__new() function to create and init a perf_evlist struct\n dynamicaly.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-46-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 93bce7e5bfcd570e9250c974b5c2c91d6b8332ef\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:31 2019 +0200\n\n libperf: Move zalloc.o into libperf\n \n We need it in both perf and libperf, thus moving it to libperf.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-45-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 9c3516d1b850ea938b074df33e4c86d721c77720\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:30 2019 +0200\n\n libperf: Add perf_cpu_map__new()/perf_cpu_map__read() functions\n \n Moving the following functions from tools/perf:\n \n cpu_map__new()\n cpu_map__read()\n \n to libperf with the following names:\n \n perf_cpu_map__new()\n perf_cpu_map__read()\n \n Committer notes:\n \n Fixed up this one:\n \n tools/perf/arch/arm/util/cs-etm.c\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-44-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 1fc632cef4ea137bc45fd0fc4cb902e374064163\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:29 2019 +0200\n\n libperf: Move perf_event_attr field from perf\u0027s evsel to libperf\u0027s perf_evsel\n \n Move the perf_event_attr struct fron \u0027struct evsel\u0027 to \u0027struct perf_evsel\u0027.\n \n Committer notes:\n \n Fixed up these:\n \n tools/perf/arch/arm/util/auxtrace.c\n tools/perf/arch/arm/util/cs-etm.c\n tools/perf/arch/arm64/util/arm-spe.c\n tools/perf/arch/s390/util/auxtrace.c\n tools/perf/util/cs-etm.c\n \n Also\n \n cc1: warnings being treated as errors\n tests/sample-parsing.c: In function \u0027do_test\u0027:\n tests/sample-parsing.c:162: error: missing initializer\n tests/sample-parsing.c:162: error: (near initialization for \u0027evsel.core.cpus\u0027)\n \n struct evsel evsel \u003d {\n .needs_swap \u003d false,\n - .core.attr \u003d {\n - .sample_type \u003d sample_type,\n - .read_format \u003d read_format,\n + .core \u003d {\n + . attr \u003d {\n + .sample_type \u003d sample_type,\n + .read_format \u003d read_format,\n + },\n \n [perfbuilder@a70e4eeb5549 /]$ gcc --version |\u0026 head -1\n gcc (GCC) 4.4.7\n \n Also we don\u0027t need to include perf_event.h in\n tools/perf/lib/include/perf/evsel.h, forward declaring \u0027struct\n perf_event_attr\u0027 is enough. And this even fixes the build in some\n systems where things are used somewhere down the include path from\n perf_event.h without defining __always_inline.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-43-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 6484d2f9dc3ecbf13f07100f7f771d1d779eda04\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:28 2019 +0200\n\n libperf: Add nr_entries to struct perf_evlist\n \n Move nr_entries count from \u0027struct perf\u0027 to into perf_evlist struct.\n \n Committer notes:\n \n Fix tools/perf/arch/s390/util/auxtrace.c case. And also the comment in\n tools/perf/util/annotate.h.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-42-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 52e22fb8af779e1a26b1cbde1db2f82f78b3ae68\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:27 2019 +0200\n\n libperf: Add perf_evlist__remove() function\n \n Adding perf_evlist__remove() function to remove a perf_evsel from\n a perf_evlist struct.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-41-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 9a5edde6d3a6fb26101406534f7a5d09a9fcd362\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:26 2019 +0200\n\n libperf: Add perf_evlist__add() function\n \n Add the perf_evlist__add() function to add a perf_evsel in a perf_evlist\n struct.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-40-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 4562a7393996bb28bf629277903a561bfefea177\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:25 2019 +0200\n\n libperf: Add perf_evlist__init() function\n \n Add the perf_evlist__init() function to initialize a perf_evlist struct.\n \n Committer testing:\n \n Fix a change in init ordering that was causing this backtrace:\n \n (gdb) run stat sleep 1\n Starting program: /root/bin/perf stat sleep 1\n Program received signal SIGSEGV, Segmentation fault.\n 0x00000000004f6b55 in __perf_evlist__propagate_maps (evlist\u003d0xbb34c0, evsel\u003d0x0) at util/evlist.c:161\n 161 if (!evsel-\u003eown_cpus || evlist-\u003ehas_user_cpus) {\n Missing separate debuginfos, use: dnf debuginfo-install bzip2-libs-1.0.6-29.fc30.x86_64 elfutils-libelf-0.176-3.fc30.x86_64 elfutils-libs-0.176-3.fc30.x86_64 glib2-2.60.4-1.fc30.x86_64 libbabeltrace-1.5.6-2.fc30.x86_64 libgcc-9.1.1-1.fc30.x86_64 libunwind-1.3.1-2.fc30.x86_64 libuuid-2.33.2-1.fc30.x86_64 libxcrypt-4.4.6-2.fc30.x86_64 libzstd-1.4.0-1.fc30.x86_64 numactl-libs-2.0.12-2.fc30.x86_64 pcre-8.43-2.fc30.x86_64 perl-libs-5.28.2-436.fc30.x86_64 popt-1.16-17.fc30.x86_64 python2-libs-2.7.16-2.fc30.x86_64 slang-2.3.2-5.fc30.x86_64 xz-libs-5.2.4-5.fc30.x86_64 zlib-1.2.11-15.fc30.x86_64\n (gdb) bt\n #0 0x00000000004f6b55 in __perf_evlist__propagate_maps (evlist\u003d0xbb34c0, evsel\u003d0x0) at util/evlist.c:161\n #1 0x00000000004f6c7a in perf_evlist__propagate_maps (evlist\u003d0xbb34c0) at util/evlist.c:178\n #2 0x00000000004f955e in perf_evlist__set_maps (evlist\u003d0xbb34c0, cpus\u003d0x0, threads\u003d0x0) at util/evlist.c:1128\n #3 0x00000000004f66f8 in evlist__init (evlist\u003d0xbb34c0, cpus\u003d0x0, threads\u003d0x0) at util/evlist.c:52\n #4 0x00000000004f6790 in evlist__new () at util/evlist.c:64\n #5 0x0000000000456071 in cmd_stat (argc\u003d3, argv\u003d0x7fffffffd670) at builtin-stat.c:1705\n #6 0x00000000004dd0fa in run_builtin (p\u003d0xa21e00 \u003ccommands+288\u003e, argc\u003d3, argv\u003d0x7fffffffd670) at perf.c:304\n #7 0x00000000004dd367 in handle_internal_command (argc\u003d3, argv\u003d0x7fffffffd670) at perf.c:356\n #8 0x00000000004dd4ae in run_argv (argcp\u003d0x7fffffffd4cc, argv\u003d0x7fffffffd4c0) at perf.c:400\n #9 0x00000000004dd81a in main (argc\u003d3, argv\u003d0x7fffffffd670) at perf.c:522\n (gdb) bt\n \n So move the initialization of the core evlist (calling\n perf_evlist__init()) to before perf_evlist__set_maps() in\n evlist__init().\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Tested-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-39-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit b04c597af761ccfd32f40ee3629843b6f3674fce\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:24 2019 +0200\n\n libperf: Add perf_evsel__init function\n \n Add the perf_evsel__init() function to initialize perf_evsel struct.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-38-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit ce9036a6e3bdfac6c7ccf8221aec9bcf9c2d355e\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:23 2019 +0200\n\n libperf: Include perf_evlist in evlist object\n \n Include perf_evlist in the evlist object, will continue to move other\n generic things into libperf\u0027s perf_evlist.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-37-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit b27c4ece725a7f2225f76ad05dc6f3f5463fe893\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:22 2019 +0200\n\n libperf: Include perf_evsel in evsel object\n \n Including perf_evsel in evsel object, will continue to move other\n generic things into libperf\u0027s perf_evsel struct.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-36-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 285a30c36d1e18e7e2afa24dae50ba5596be45e7\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:21 2019 +0200\n\n libperf: Add perf_evlist and perf_evsel structs\n \n Add the perf_evlist and perf_evsel structs to libperf.\n \n It\u0027s added as a declarations into:\n \n include/perf/evlist.h\n include/perf/evsel.h\n \n which will be included by users.\n \n The definitions are added into:\n \n include/internal/evlist.h\n include/internal/evsel.h\n \n which is not to be included by users, but shared\n within perf and libperf.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-35-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 7836e52e518b5e3fd695850f1d4081f756f58406\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:20 2019 +0200\n\n libperf: Add perf_thread_map__get()/perf_thread_map__put()\n \n Move the following functions:\n \n thread_map__get()\n thread_map__put()\n thread_map__comm()\n \n to libperf with the following names:\n \n perf_thread_map__get()\n perf_thread_map__put()\n perf_thread_map__comm()\n \n Add the perf_thread_map__comm() function for it to work/compile.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-34-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 4b49cce25e719587e934b745fe9bbb5bc8c4ba29\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:19 2019 +0200\n\n libperf: Add perf_thread_map__new_dummy() function\n \n Moving the following functions:\n \n thread_map__new_dummy()\n thread_map__realloc()\n thread_map__set_pid()\n \n to libperf with the following names:\n \n perf_thread_map__new_dummy()\n perf_thread_map__realloc()\n perf_thread_map__set_pid()\n \n the other 2 functions are dependencies of the\n perf_thread_map__new_dummy() function.\n \n The perf_thread_map__realloc() function is not exported.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-33-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 07acd22677ac6bb2db404d1d258e8c7d06ca7706\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:18 2019 +0200\n\n libperf: Add perf_thread_map struct\n \n Add perf_thread_map struct to libperf.\n \n It\u0027s added as a declaration into into:\n \n include/perf/threadmap.h\n \n which will be included by users.\n \n The perf_thread_map struct definition is added into:\n \n include/internal/threadmap.h\n \n which is not to be included by users, but shared within perf and\n libperf.\n \n We tried the total separation of the perf_thread_map struct in libperf,\n but it lead to complications and much bigger changes in perf code, so we\n decided to share the declaration.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-32-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 38f01d8da1d8d28678ea16a0a484f4d3eded34b2\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:17 2019 +0200\n\n libperf: Add perf_cpu_map__get()/perf_cpu_map__put()\n \n Moving the following functions:\n \n cpu_map__get()\n cpu_map__put()\n \n to libperf with following names:\n \n perf_cpu_map__get()\n perf_cpu_map__put()\n \n Committer notes:\n \n Added fixes for arm/arm64\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-31-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 397721e06e52d017cfdd403f63284ed0995d4caf\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:16 2019 +0200\n\n libperf: Add perf_cpu_map__dummy_new() function\n \n Move cpu_map__dummy_new() to libperf as perf_cpu_map__dummy_new() function.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-30-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 959b83c769389b24d64759f60e64c4c62620ff02\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:15 2019 +0200\n\n libperf: Add perf_cpu_map struct\n \n Add perf_cpu_map struct to libperf.\n \n It\u0027s added as a declaration into:\n \n include/perf/cpumap.h\n \n which will be included by users.\n \n The perf_cpu_map struct definition is added into:\n \n include/internal/cpumap.h\n \n which is not to be included by users, but shared within perf and\n libperf.\n \n We tried the total separation of the perf_cpu_map struct in libperf, but\n it lead to complications and much bigger changes in perf code, so we\n decided to share the declaration.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-29-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit a1556f8479ed58b8d5a33aef54578bad0165c7e7\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:14 2019 +0200\n\n libperf: Add debug output support\n \n Add the perf_set_print() function to allow setting an output function\n for warn/info/debug messages.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-28-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 5b7f445d684fc287a2101e29d42d1fee19ae14ff\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:13 2019 +0200\n\n libperf: Add perf/core.h header\n \n Add perf/core.h header to be used in header files coming in the\n following patches.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-27-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit a429dcb8feb60b8500fed81e2275c1944e3091fc\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:12 2019 +0200\n\n libperf: Add libperf to the python.so build\n \n Link libperf.a with python.so.\n \n Committer testing:\n \n Continues to work:\n \n # perf test python\n 18: \u0027import perf\u0027 in python : Ok\n #\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Tested-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-26-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 47f9bccc79cb067103ad5e9790e0d01c94839429\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:11 2019 +0200\n\n libperf: Add build version support\n \n Add a shared library version, generating the following files:\n \n $ ll tools/perf/lib/libperf.so*\n libperf.so -\u003e libperf.so.0.0.1\n libperf.so.0 -\u003e libperf.so.0.0.1\n libperf.so.0.0.1\n \n Committer testing:\n \n One has to build just libbperf to get this, building perf so far doesn\u0027t\n trigger this, i.e. I tried:\n \n $ make O\u003d/tmp/build/perf -C tools/perf\n \n And the files above were not created, so one has to do:\n \n $ make O\u003d/tmp/build/perf -C tools/perf/lib/\n make: Entering directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n LINK /tmp/build/perf/libperf.so.0.0.1\n make: Leaving directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n $ ls -la /tmp/build/perf/*.so.*\n lrwxrwxrwx. 1 acme acme 16 Jul 22 15:37 /tmp/build/perf/libperf.so.0 -\u003e libperf.so.0.0.1\n -rwxrwxr-x. 1 acme acme 16368 Jul 22 15:37 /tmp/build/perf/libperf.so.0.0.1\n $\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Tested-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-25-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 3143504918105156d03e8f927e127f7b9ea260d2\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:10 2019 +0200\n\n libperf: Make libperf.a part of the perf build\n \n Add an empty libperf.a under tools/perf/lib and link it with perf.\n \n It can also be built separately with:\n \n $ cd tools/perf/lib \u0026\u0026 make\n CC core.o\n LD libperf-in.o\n AR libperf.a\n LINK libperf.so\n \n Committer testing:\n \n $ make O\u003d/tmp/build/perf -C tools/perf/lib/\n make: Entering directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n LINK /tmp/build/perf/libperf.so\n make: Leaving directory \u0027/home/acme/git/perf/tools/perf/lib\u0027\n $ ls -la /tmp/build/perf/libperf.so\n -rwxrwxr-x. 1 acme acme 16232 Jul 22 15:30 /tmp/build/perf/libperf.so\n $ file /tmp/build/perf/libperf.so\n /tmp/build/perf/libperf.so: ELF 64-bit LSB shared object, x86-64, version 1 (SYSV), dynamically linked, BuildID[sha1]\u003d7a51d227d871b381ddb686dcf94145c4dd908221, not stripped\n $ git status tools/perf\n On branch perf/core\n nothing to commit, working tree clean\n $\n $ ls -lart tools/perf/lib/\n total 16\n drwxrwxr-x. 16 acme acme 4096 Jul 22 15:29 ..\n -rw-rw-r--. 1 acme acme 1633 Jul 22 15:29 Makefile\n -rw-rw-r--. 1 acme acme 0 Jul 22 15:29 core.c\n -rw-rw-r--. 1 acme acme 20 Jul 22 15:29 Build\n drwxrwxr-x. 2 acme acme 4096 Jul 22 15:29 .\n $\n \n Committer notes:\n \n Need to add -I$(srctree)/tools/arch/$(ARCH)/include/uapi\n -I$(srctree)/tools/include/uapi to tools/perf/lib/Makefile\u0027s INCLUDE\n variable to pick up the latest versions of kernel headers, even in older\n systems, this is in line with what is in tools/lib/bpf/Makefile.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Tested-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-24-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit e74676debaae7dcce20a34817ef145478887ba95\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:09 2019 +0200\n\n perf evlist: Rename perf_evlist__disable() to evlist__disable()\n \n Rename perf_evlist__disable() to evlist__disable(), so we don\u0027t have a\n name clash when we add perf_evlist__disable() in libperf.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-23-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 1c87f1654cc315fbeae0238a8dbf5bf3c498f3af\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:08 2019 +0200\n\n perf evlist: Rename perf_evlist__enable() to evlist__enable()\n \n Rename perf_evlist__enable() to evlist__enable(), so we don\u0027t have a\n name clash when we add perf_evlist__enable() in libperf.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-22-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 750b4edeb0527414fb17b0ee2a76d2dbbd9a199d\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:07 2019 +0200\n\n perf evlist: Rename perf_evlist__close() to evlist__close()\n \n Rename perf_evlist__close() to evlist__close(), so we don\u0027t have a name\n clash when we add perf_evlist__close() in libperf.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-21-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 474ddc4c46025a615d0ea791d37ce9038fa20229\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:06 2019 +0200\n\n perf evlist: Rename perf_evlist__open() to evlist__open()\n \n Rename perf_evlist__open() to evlist__open(), so we don\u0027t have a name\n clash when we add perf_evlist__open() in libperf.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-20-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit b49aca3e9ce60d00e5bf0694b2ff4c2cd40809e5\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:05 2019 +0200\n\n perf evsel: Rename perf_evsel__cpus() to evsel__cpus()\n \n Rename perf_evsel__cpus() to evsel__cpus(), so we don\u0027t have a name\n clash when we add perf_evsel__cpus() in libperf.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-19-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 24e376b24582648d363df4e0a6bcc2ffcecd211e\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:04 2019 +0200\n\n perf evsel: Rename perf_evsel__apply_filter() to evsel__apply_filter()\n \n Rename perf_evsel__apply_filter() to evsel__apply_filter(), so we don\u0027t\n have a name clash when we add perf_evsel__apply_filter() in libperf.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-18-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 9a10bb22897ae9c2aa0ac9c2071f539f406ef942\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:03 2019 +0200\n\n perf evsel: Rename perf_evsel__disable() to evsel__disable()\n \n Renaming perf_evsel__disable() to evsel__disable(), so we don\u0027t have a\n name clash when we add perf_evsel__disable() in libperf.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-17-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit ec7f24ef44fc5a4eb5cb71658c33db538ed66003\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:02 2019 +0200\n\n perf evsel: Rename perf_evsel__enable() to evsel__enable()\n \n Rename perf_evsel__enable() to evsel__enable(), so we don\u0027t have a name\n clash when we add perf_evsel__enable() in libperf.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-16-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 5972d1e07bd95c7458e2d7f484391d69008affc7\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:01 2019 +0200\n\n perf evsel: Rename perf_evsel__open() to evsel__open()\n \n Rename perf_evsel__open() to evsel__open(), so we don\u0027t have a name\n clash when we add perf_evsel__open() in libperf.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-15-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 1625102764a578b11fb407b8194cb0521129d919\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:24:00 2019 +0200\n\n perf evlist: Rename perf_evlist__remove() to evlist__remove()\n \n Rename perf_evlist__remove() to evlist__remove(), so we don\u0027t have a\n name clash when we add perf_evlist__remove() in libperf.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-14-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit a1cf3a75d3317ed893d453c222d220ca4d5f4c4e\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:23:59 2019 +0200\n\n perf evlist: Rename perf_evlist__add() to evlist__add()\n \n Rename perf_evlist__add() to evlist__add(), so we don\u0027t have a name\n clash when we add perf_evlist__add() in libperf.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-13-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 365c3ae7452ca95e0a8f1e4716dd806550ade706\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:23:58 2019 +0200\n\n perf evsel: Rename perf_evsel__new() to evsel__new()\n \n Rename perf_evsel__new() to evsel__new(), so we don\u0027t have a name clash\n when we add perf_evsel__new() in libperf.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-12-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 5eb2dd2ade8354dcbe4cef54cd1719622af8f2dc\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:23:57 2019 +0200\n\n perf evsel: Rename perf_evsel__delete() to evsel__delete()\n \n Remame perf_evsel__delete() to evsel__delete(), so we don\u0027t have a name\n clash when we add perf_evsel__delete() in libperf.\n \n Also renaming perf_evsel__delete_priv() to evsel__delete_priv().\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-11-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit c12995a55474e1cefac52da38c7fc47c024de067\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:23:56 2019 +0200\n\n perf evlist: Rename perf_evlist__delete() to evlist__delete()\n \n Rename perf_evlist__delete() to evlist__delete(), so we don\u0027t have a\n name clash when we add perf_evlist__delete() in libperf.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-10-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 0f98b11c616f240b54ee85629ff4d3650c7ccc7d\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:23:55 2019 +0200\n\n perf evlist: Rename perf_evlist__new() to evlist__new()\n \n Rename perf_evlist__new() to evlist__new(), so we don\u0027t have a name\n clash when we add perf_evlist__new() in libperf.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-9-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 52c86bca94b42239563b1510d5fc6329b0ec1a86\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:23:54 2019 +0200\n\n perf evlist: Rename perf_evlist__init() to evlist__init()\n \n Rename perf_evlist__init() to evlist__init(), so we don\u0027t have a name\n clash when we add perf_evlist__init() in libperf.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-8-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit b4b62ee688eb39151c9d8182c3e2f12c9d34602d\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:23:53 2019 +0200\n\n perf evsel: Rename perf_evsel__init() to evsel__init()\n \n Rename perf_evsel__init() to evsel__init(), so we don\u0027t have a name\n clash when we add perf_evsel__init() in libperf.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-7-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 63503dba87acfab49280d3b05df6705a6f327e8a\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:23:52 2019 +0200\n\n perf evlist: Rename struct perf_evlist to struct evlist\n \n Rename struct perf_evlist to struct evlist, so we don\u0027t have a name\n clash when we add struct perf_evlist in libperf.\n \n Committer notes:\n \n Added fixes to build on arm64, from Jiri and from me\n (tools/perf/util/cs-etm.c)\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-6-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 32dcd021d004038ca12ac17319da5aa4756e9312\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:23:51 2019 +0200\n\n perf evsel: Rename struct perf_evsel to struct evsel\n \n Rename struct perf_evsel to struct evsel, so we don\u0027t have a name clash\n when we add struct perf_evsel in libperf.\n \n Committer notes:\n \n Added fixes for arm64, provided by Jiri.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-5-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 9749b90e566ca1a235fc8e2118f99c5690969342\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:23:50 2019 +0200\n\n perf tools: Rename struct thread_map to struct perf_thread_map\n \n Rename struct thread_map to struct perf_thread_map, so it could be part\n of libperf.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-4-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit f854839ba2a546a888159667c5ade96793e5cd10\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:23:49 2019 +0200\n\n perf cpu_map: Rename struct cpu_map to struct perf_cpu_map\n \n Rename struct cpu_map to struct perf_cpu_map, so it could be part of\n libperf.\n \n Committer notes:\n \n Added fixes for arm64, provided by Jiri.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-3-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit df1d6856eaa7ec9ad1e670685b370f3e66326079\nAuthor: Jiri Olsa \u003cjolsa@kernel.org\u003e\nDate: Sun Jul 21 13:23:48 2019 +0200\n\n perf stat: Move loaded out of struct perf_counts_values\n \n Because we will make struct perf_counts_values public in following\n patches and \u0027loaded\u0027 is implementation related.\n \n No functional change is expected.\n \n Signed-off-by: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Cc: Alexey Budankov \u003calexey.budankov@linux.intel.com\u003e\n Cc: Andi Kleen \u003cak@linux.intel.com\u003e\n Cc: Michael Petlan \u003cmpetlan@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Link: http://lkml.kernel.org/r/20190721112506.12306-2-jolsa@kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit e4b00e930bf71ef32189716e6cb6b0565592f078\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Jul 17 20:32:13 2019 -0300\n\n perf trace: Add \"sendfile64\" alias to the \"sendfile\" syscall\n \n We were looking in tracefs for:\n \n /sys/kernel/debug/tracing/events/syscalls/sys_enter_sendfile/format when\n \n what is there is just\n \n /sys/kernel/debug/tracing/events/syscalls/sys_enter_sendfile/format\n \n Its the same id, 40 in x86_64, so just add an alias and let the existing\n logic take care of that.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-km2hmg7hru6u4pawi5fi903q@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit ad4153f964ebb756617e1586ba372156db0efeed\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Jul 17 18:27:33 2019 -0300\n\n perf trace: Reuse BPF augmenters from syscalls with similar args signature\n \n We have an augmenter for the \"open\" syscall, which has just one pointer,\n in the first argument, a \"const char *\", so any other syscall that has\n just one pointer and that is the first can reuse the \"open\" BPF\n augmenter program.\n \n Even more, syscalls that get two pointers with the first being a string\n can reuse \"open\"\u0027s BPF augmenter till we have an augmenter that better\n matches that syscall with two pointers.\n \n With this the few augmenters we have, for open (first arg is a string),\n openat (2nd arg is a string), renameat (2nd and 4th are strings) can be\n reused by a lot of syscalls, ditto for \"bind\" reusing \"connect\" because\n both have the 2nd argument as a sockaddr and the 3rd as its len.\n \n Lets see how this makes the \"bind\" syscall reuse the \"connect\" BPF prog\n augmenter found in tools/perf/examples/bpf/augmented_raw_syscalls.c:\n \n # perf trace -e bind,connect systemctl restart sshd\n connect(3, { .family: PF_LOCAL, path: /run/systemd/private }, 23) \u003d 0\n #\n \n Oh, it just connects to some daemon, so we better do it system wide and then\n stop/start sshd:\n \n # perf trace -e bind,connect\n systemctl/10124 connect(3, { .family: PF_LOCAL, path: /run/systemd/private }, 23) \u003d 0\n sshd/10102 connect(7, { .family: PF_LOCAL, path: /dev/log }, 110) \u003d 0\n systemctl/10126 connect(3, { .family: PF_LOCAL, path: /run/systemd/private }, 23) \u003d 0\n systemd/10128 ... [continued]: connect()) \u003d 0\n (sshd)/10128 connect(3, { .family: PF_LOCAL, path: /run/systemd/journal/stdout }, 30) ...\n sshd/10128 bind(3, { .family: PF_NETLINK }, 12) \u003d 0\n sshd/10128 connect(4, { .family: PF_LOCAL, path: /var/run/nscd/socket }, 110) \u003d -1 ENOENT (No such file or directory)\n sshd/10128 connect(3, { .family: PF_INET6, port: 22, addr: :: }, 28) \u003d 0\n sshd/10128 connect(3, { .family: PF_UNSPEC }, 16) \u003d 0\n sshd/10128 connect(3, { .family: PF_INET, port: 22, addr: 0.0.0.0 }, 16) \u003d 0\n sshd/10128 connect(3, { .family: PF_LOCAL, path: /var/run/nscd/socket }, 110) \u003d -1 ENOENT (No such file or directory)\n sshd/10128 connect(3, { .family: PF_LOCAL, path: /var/run/nscd/socket }, 110) \u003d -1 ENOENT (No such file or directory)\n sshd/10128 connect(5, { .family: PF_LOCAL, path: /var/run/nscd/socket }, 110) \u003d -1 ENOENT (No such file or directory)\n sshd/10128 connect(5, { .family: PF_LOCAL, path: /var/run/nscd/socket }, 110) \u003d -1 ENOENT (No such file or directory)\n sshd/10128 bind(4, { .family: PF_INET, port: 22, addr: 0.0.0.0 }, 16) \u003d 0\n sshd/10128 connect(6, { .family: PF_LOCAL, path: /dev/log }, 110) \u003d 0\n sshd/10128 bind(6, { .family: PF_INET6, port: 22, addr: :: }, 28) \u003d 0\n sshd/10128 connect(7, { .family: PF_LOCAL, path: /dev/log }, 110) \u003d 0\n ^C#\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-zfley2ghs4nim1uq4nu6ed3l@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 30a910d7d3e04dd920e4ca3e8dcabf10c67fb03e\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Thu Jul 18 20:19:30 2019 -0300\n\n perf trace: Preallocate the syscall table\n \n We\u0027ll continue reading its details from tracefs as we need it, but\n preallocate the whole thing otherwise we may realloc and end up with\n pointers to the previous buffer.\n \n I.e. in an upcoming algorithm we\u0027ll look for syscalls that have function\n signatures that are similar to a given syscall to see if we can reuse\n its BPF augmenter, so we may be at syscall 42, having a \u0027struct syscall\u0027\n pointing to that slot in trace-\u003esyscalls.table[] and try to read the\n slot for an yet unread syscall, which would realloc that table to read\n the info for syscall 43, say, which would trigger a realoc of\n trace-\u003esyscalls.table[], and then the pointer we had for syscall 42\n would be pointing to the previous block of memory. b00m.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Brendan Gregg \u003cbrendan.d.gregg@gmail.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-m3cjzzifibs13imafhkk77a0@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit b8b1033fcaa091d82289698d7179e84e28cbd92a\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Jul 17 20:21:37 2019 -0300\n\n perf trace: Mark syscall ids that are not allocated to avoid unnecessary error messages\n \n There are holes in syscall tables with IDs not associated with any\n syscall, mark those when trying to read information for syscalls, which\n could happen when iterating thru all syscalls from 0 to the highest\n numbered syscall id.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-cku9mpcrcsqaiq0jepu86r68@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 5d2bd88975117062766a48b5f36ce31d2c1a8269\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Jul 17 19:53:51 2019 -0300\n\n perf trace: Forward error codes when trying to read syscall info\n \n We iterate thru the syscall table produced from the kernel syscall\n tables reading info, propagate the error and add to the debug message.\n \n This helps in fixing further bugs, such as failing to read the\n \"sendfile\" syscall info when it really should try the aliasm\n \"sendfile64\".\n \n Problems reading syscall 40: 2 (No such file or directory)(sendfile) information\n \n # grep sendfile /tmp/build/perf/arch/x86/include/generated/asm/syscalls_64.c\n [40] \u003d \"sendfile\",\n #\n \n I.e. in the tracefs format file for the syscall tracepoints we have it\n as sendfile64:\n \n # find /sys -type f -name format | grep sendfile\n /sys/kernel/debug/tracing/events/syscalls/sys_enter_sendfile64/format\n /sys/kernel/debug/tracing/events/syscalls/sys_exit_sendfile64/format\n #\n \n But as \"sendfile\" in the file used to build the syscall table used in\n perf:\n \n $ grep sendfile arch/x86/entry/syscalls/syscall_64.tbl\n 40 common sendfile __x64_sys_sendfile64\n $\n \n So we need to add, in followup patches, aliases in \u0027perf trace\u0027 syscall\n data structures to cope with thie.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-w3eluap63x9je0bb8o3t79tz@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit cfa9ac73d6f9790f569959a729cbc9d52bff4270\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Wed Jul 17 11:47:37 2019 -0300\n\n perf trace beauty: Add BPF augmenter for the \u0027rename\u0027 syscall\n \n I.e. two strings:\n \n # perf trace -e rename\n systemd/1 rename(\"/run/systemd/units/.#invocation:dnf-makecache.service970761b7f2840dcc\", \"/run/systemd/units/invocation:dnf-makecache.service\") \u003d 0\n systemd-journa/715 rename(\"/run/systemd/journal/streams/.#9:17539785BJDblc\", \"/run/systemd/journal/streams/9:17539785\") \u003d 0\n mv/1936 rename(\"/tmp/build/perf/fd/.array.o.tmp\", \"/tmp/build/perf/fd/.array.o.cmd\") \u003d 0\n sh/1949 rename(\"/tmp/build/perf/.cpu.o.tmp\", \"/tmp/build/perf/.cpu.o.cmd\") \u003d 0\n mv/1954 rename(\"/tmp/build/perf/fs/.tracing_path.o.tmp\", \"/tmp/build/perf/fs/.tracing_path.o.cmd\") \u003d 0\n mv/1963 rename(\"/tmp/build/perf/common-cmds.h+\", \"/tmp/build/perf/common-cmds.h\") \u003d 0\n :1975/1975 rename(\"/tmp/build/perf/.exec-cmd.o.tmp\", \"/tmp/build/perf/.exec-cmd.o.cmd\") \u003d 0\n mv/1979 rename(\"/tmp/build/perf/fs/.fs.o.tmp\", \"/tmp/build/perf/fs/.fs.o.cmd\") \u003d 0\n mv/2005 rename(\"/tmp/build/perf/.debug.o.tmp\", \"/tmp/build/perf/.debug.o.cmd\") \u003d 0\n mv/2012 rename(\"/tmp/build/perf/.str_error_r.o.tmp\", \"/tmp/build/perf/.str_error_r.o.cmd\") \u003d 0\n mv/2019 rename(\"/tmp/build/perf/.help.o.tmp\", \"/tmp/build/perf/.help.o.cmd\") \u003d 0\n mv/2031 rename(\"/tmp/build/perf/.trace-seq.o.tmp\", \"/tmp/build/perf/.trace-seq.o.cmd\") \u003d 0\n make/2038 ... [continued]: rename()) \u003d 0\n :2038/2038 rename(\"/tmp/build/perf/.event-plugin.o.tmp\", \"/tmp/build/perf/.event-plugin.o.cmd\") ...\n ar/2035 rename(\"/tmp/build/perf/stzwBX3a\", \"/tmp/build/perf/libapi.a\") \u003d 0\n mv/2051 rename(\"/tmp/build/perf/.parse-utils.o.tmp\", \"/tmp/build/perf/.parse-utils.o.cmd\") \u003d 0\n mv/2069 rename(\"/tmp/build/perf/.subcmd-config.o.tmp\", \"/tmp/build/perf/.subcmd-config.o.cmd\") \u003d 0\n make/2080 rename(\"/tmp/build/perf/.parse-filter.o.tmp\", \"/tmp/build/perf/.parse-filter.o.cmd\") \u003d 0\n mv/2099 rename(\"/tmp/build/perf/.pager.o.tmp\", \"/tmp/build/perf/.pager.o.cmd\") \u003d 0\n :2124/2124 rename(\"/tmp/build/perf/.sigchain.o.tmp\", \"/tmp/build/perf/.sigchain.o.cmd\") \u003d 0\n make/2140 rename(\"/tmp/build/perf/.event-parse.o.tmp\", \"/tmp/build/perf/.event-parse.o.cmd\") \u003d 0\n mv/2164 rename(\"/tmp/build/perf/.kbuffer-parse.o.tmp\", \"/tmp/build/perf/.kbuffer-parse.o.cmd\") \u003d 0\n sh/2174 rename(\"/tmp/build/perf/.run-command.o.tmp\", \"/tmp/build/perf/.run-command.o.cmd\") \u003d 0\n mv/2190 rename(\"/tmp/build/perf/.tep_strerror.o.tmp\", \"/tmp/build/perf/.tep_strerror.o.cmd\") \u003d 0\n :2261/2261 rename(\"/tmp/build/perf/.event-parse-api.o.tmp\", \"/tmp/build/perf/.event-parse-api.o.cmd\") \u003d 0\n :2480/2480 rename(\"/tmp/build/perf/stLv3kG2\", \"/tmp/build/perf/libtraceevent.a\") \u003d 0\n ^C#\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-6hh2rl27uri6gsxhmk6q3hx5@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 247dd65b909f1cd62a178febe3f6f8d5efdd7dd2\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Jul 16 17:38:26 2019 -0300\n\n perf trace beauty: Beautify bind\u0027s sockaddr arg\n \n By reusing the \"connect\" BPF collector.\n \n Testing it system wide and stopping/starting sshd:\n \n # perf trace -e bind\n LLVM: dumping /home/acme/git/perf/tools/perf/examples/bpf/augmented_raw_syscalls.o\n DNS Res~er #18/15132 bind(243, { .family: PF_NETLINK }, 12) \u003d 0\n DNS Res~er #19/4833 bind(247, { .family: PF_NETLINK }, 12) \u003d 0\n DNS Res~er #19/4833 bind(238, { .family: PF_NETLINK }, 12) \u003d 0\n DNS Res~er #18/15132 bind(243, { .family: PF_NETLINK }, 12) \u003d 0\n DNS Res~er #18/10327 bind(258, { .family: PF_NETLINK }, 12) \u003d 0\n :6507/6507 bind(24, { .family: PF_NETLINK }, 12) \u003d 0\n DNS Res~er #19/4833 bind(238, { .family: PF_NETLINK }, 12) \u003d 0\n DNS Res~er #18/15132 bind(242, { .family: PF_NETLINK }, 12) \u003d 0\n sshd/6514 bind(3, { .family: PF_NETLINK }, 12) \u003d 0\n sshd/6514 bind(5, { .family: PF_INET, port: 22, addr: 0.0.0.0 }, 16) \u003d 0\n sshd/6514 bind(7, { .family: PF_INET6, port: 22, addr: :: }, 28) \u003d 0\n DNS Res~er #18/10327 bind(229, { .family: PF_NETLINK }, 12) \u003d 0\n DNS Res~er #18/15132 bind(231, { .family: PF_NETLINK }, 12) \u003d 0\n DNS Res~er #19/4833 bind(229, { .family: PF_NETLINK }, 12) \u003d 0\n ^C#\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-m2hmxqrckxxw2ciki0tu889u@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 3c475bc021be1f5e0a00dc1a13dc85ce7924a7d6\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Jul 16 17:33:39 2019 -0300\n\n perf trace beauty: Beautify \u0027sendto\u0027s sockaddr arg\n \n By just writing the collector in the augmented_raw_syscalls.c BPF\n program:\n \n # perf trace -e sendto\n \u003cSNIP\u003e\n ping/23492 sendto(3, 0x56253bbef700, 64, NONE, { .family: PF_INET, port: 0, addr: 10.10.161.32 }, 0x10) \u003d 64\n ping/23492 sendto(3, 0x56253bbef700, 64, NONE, { .family: PF_INET, port: 0, addr: 10.10.161.32 }, 0x10) \u003d 64\n ping/23492 sendto(3, 0x56253bbef700, 64, NONE, { .family: PF_INET, port: 0, addr: 10.10.161.32 }, 0x10) \u003d 64\n ping/23492 sendto(3, 0x56253bbef700, 64, NONE, { .family: PF_INET, port: 0, addr: 10.10.161.32 }, 0x10) \u003d 64\n Socket Thread/3573 sendto(247, 0x7fb32d49c000, 120, NONE, { .family: PF_UNSPEC }, NULL) \u003d 120\n DNS Res~er #18/11374 sendto(242, 0x7fb342cfe420, 20, NONE, { .family: PF_NETLINK }, 0xc) \u003d 20\n DNS Res~er #18/11374 sendto(242, 0x7fb342cfcca0, 42, MSG_NOSIGNAL, { .family: PF_UNSPEC }, NULL) \u003d 42\n DNS Res~er #18/11374 sendto(242, 0x7fb342cfcccc, 42, MSG_NOSIGNAL, { .family: PF_UNSPEC }, NULL) \u003d 42\n ping/23492 sendto(3, 0x56253bbef700, 64, NONE, { .family: PF_INET, port: 0, addr: 10.10.161.32 }, 0x10) \u003d 64\n Socket Thread/3573 sendto(242, 0x7fb308bb1c08, 296, NONE, { .family: PF_UNSPEC }, NULL) \u003d 296\n ping/23492 sendto(3, 0x56253bbef700, 64, NONE, { .family: PF_INET, port: 0, addr: 10.10.161.32 }, 0x10) \u003d 64\n ping/23492 sendto(3, 0x56253bbef700, 64, NONE, { .family: PF_INET, port: 0, addr: 10.10.161.32 }, 0x10) \u003d 64\n ping/23492 sendto(3, 0x56253bbef700, 64, NONE, { .family: PF_INET, port: 0, addr: 10.10.161.32 }, 0x10) \u003d 64\n ping/23492 sendto(3, 0x56253bbef700, 64, NONE, { .family: PF_INET, port: 0, addr: 10.10.161.32 }, 0x10) \u003d 64\n ^C\n #\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-p0l0rlvq19v5zf8qc2x2itow@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit ef969ca64d04161ccbde2aaf8b0767f91a6d32ff\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Jul 16 17:21:09 2019 -0300\n\n perf trace beauty: Do not try to use the fd-\u003epathname beautifier for bind/connect fd arg\n \n Doesn\u0027t make sense and also we now beautify the sockaddr, which provides\n enough info:\n \n # trace -e close,socket,connec* ssh www.bla.com\n \u003cSNIP\u003e\n close(5) \u003d 0\n socket(PF_INET, SOCK_DGRAM|CLOEXEC|NONBLOCK, IPPROTO_IP) \u003d 5\n connect(5, { .family: PF_INET, port: 53, addr: 192.168.44.1 }, 16) \u003d 0\n close(5) \u003d 0\n socket(PF_INET, SOCK_STREAM, IPPROTO_TCP) \u003d 5\n ^C#\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-h9drpb7ail808d2mh4n7tla4@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 79d725cdf24dec7bfe7ad27b107f5cb06cd3785a\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Jul 16 16:56:49 2019 -0300\n\n perf trace beauty: Disable fd-\u003epathname when close() not enabled\n \n As we invalidate the fd-\u003epathname table in the SCA_CLOSE_FD beautifier,\n if we don\u0027t have it we may end up keeping an fd-\u003epathname association\n that then gets misprinted.\n \n The previous behaviour continues when the close() syscall is enabled,\n which may still be a a problem if we lose records (i.e. we may lose a\n \u0027close\u0027 record and then get that fd reused by socket()) but then the\n tool will notify that records are being lost and the user will be warned\n that some of the heuristics will fall apart.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-b7t6h8sq9lebemvfy2zh3qq1@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 1d86275225b4c9db3fb426e992886df5051f0047\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Jul 16 16:34:27 2019 -0300\n\n perf trace beauty: Make connect\u0027s addrlen be printed as an int, not hex\n \n # perf trace -e connec* ssh www.bla.com\n connect(3\u003c/var/lib/sss/mc/passwd\u003e, { .family: PF_LOCAL, path: /var/run/nscd/socket }, 110) \u003d -1 ENOENT (No such file or directory)\n connect(3\u003c/var/lib/sss/mc/passwd\u003e, { .family: PF_LOCAL, path: /var/run/nscd/socket }, 110) \u003d -1 ENOENT (No such file or directory)\n connect(4\u003csocket:[16610959]\u003e, { .family: PF_LOCAL, path: /var/lib/sss/pipes/nss }, 110) \u003d 0\n connect(7, { .family: PF_LOCAL, path: /var/run/nscd/socket }, 110) \u003d -1 ENOENT (No such file or directory)\n connect(7, { .family: PF_LOCAL, path: /var/run/nscd/socket }, 110) \u003d -1 ENOENT (No such file or directory)\n connect(5, { .family: PF_LOCAL, path: /var/run/nscd/socket }, 110) \u003d -1 ENOENT (No such file or directory)\n connect(5\u003c/usr/lib64/libnss_mdns4_minimal.so.2\u003e, { .family: PF_LOCAL, path: /var/run/nscd/socket }, 110) \u003d -1 ENOENT (No such file or directory)\n connect(5\u003c/usr/lib64/libnss_mdns4_minimal.so.2\u003e, { .family: PF_INET, port: 53, addr: 192.168.44.1 }, 16) \u003d 0\n connect(5\u003c/usr/lib64/libnss_mdns4_minimal.so.2\u003e, { .family: PF_INET, port: 22, addr: 146.112.61.108 }, 16) \u003d 0\n connect(5\u003c/usr/lib64/libnss_mdns4_minimal.so.2\u003e, { .family: PF_INET6, port: 22, addr: ::ffff:146.112.61.108 }, 28) \u003d 0\n ^Cconnect(5\u003c/usr/lib64/libnss_mdns4_minimal.so.2\u003e, { .family: PF_INET, port: 22, addr: 146.112.61.108 }, 16) \u003d -1 (unknown) (INTERNAL ERROR: strerror_r(512, [buf], 128)\u003d22)\n #\n \n Argh, the SCA_FD needs to invalidate its cache when close is done...\n \n It works if the \u0027close\u0027 syscall is not filtered out ;-\\\n \n # perf trace -e close,connec* ssh www.bla.com\n close(3) \u003d 0\n close(3\u003c/usr/lib64/libpcre2-8.so.0.8.0\u003e) \u003d 0\n close(3) \u003d 0\n close(3\u003c/usr/lib64/libkrb5.so.3.3\u003e) \u003d 0\n close(3\u003c/usr/lib64/libkrb5.so.3.3\u003e) \u003d 0\n close(3) \u003d 0\n close(3\u003c/usr/lib64/libk5crypto.so.3.1\u003e) \u003d 0\n close(3\u003c/usr/lib64/libk5crypto.so.3.1\u003e) \u003d 0\n close(3\u003c/usr/lib64/libcom_err.so.2.1\u003e) \u003d 0\n close(3\u003c/usr/lib64/libcom_err.so.2.1\u003e) \u003d 0\n close(3) \u003d 0\n close(3\u003c/usr/lib64/libkrb5support.so.0.1\u003e) \u003d 0\n close(3\u003c/usr/lib64/libkrb5support.so.0.1\u003e) \u003d 0\n close(3\u003c/usr/lib64/libkeyutils.so.1.8\u003e) \u003d 0\n close(3\u003c/usr/lib64/libkeyutils.so.1.8\u003e) \u003d 0\n close(3) \u003d 0\n close(3) \u003d 0\n close(3) \u003d 0\n close(3) \u003d 0\n close(4) \u003d 0\n close(3) \u003d 0\n close(3) \u003d 0\n connect(3\u003c/etc/nsswitch.conf\u003e, { .family: PF_LOCAL, path: /var/run/nscd/socket }, 110) \u003d -1 ENOENT (No such file or directory)\n close(3\u003c/etc/nsswitch.conf\u003e) \u003d 0\n connect(3\u003c/usr/lib64/libnss_sss.so.2\u003e, { .family: PF_LOCAL, path: /var/run/nscd/socket }, 110) \u003d -1 ENOENT (No such file or directory)\n close(3\u003c/usr/lib64/libnss_sss.so.2\u003e) \u003d 0\n close(3\u003c/usr/lib64/libnss_sss.so.2\u003e) \u003d 0\n close(3) \u003d 0\n close(3) \u003d 0\n connect(4\u003csocket:[16616519]\u003e, { .family: PF_LOCAL, path: /var/lib/sss/pipes/nss }, 110) \u003d 0\n ^C\n #\n \n Will disable this beautifier when \u0027close\u0027 is filtered out...\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-ekuiciyx4znchvy95c8p1yyi@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 212b9ab6775b5f340de848b5b6eef6968ccf7f20\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Jul 16 16:28:14 2019 -0300\n\n perf augmented_raw_syscalls: Augment sockaddr arg in \u0027connect\u0027\n \n We already had a beautifier for an augmented sockaddr payload, but that\n was when we were hooking on each syscalls:sys_enter_foo tracepoints,\n since now we\u0027re almost doing that by doing a tail call from\n raw_syscalls:sys_enter, its almost the same, we can reuse it straight\n away.\n \n # perf trace -e connec* ssh www.bla.com\n connect(3\u003c/var/lib/sss/mc/passwd\u003e, { .family: PF_LOCAL, path: /var/run/nscd/socket }, 0x6e) \u003d -1 ENOENT (No such file or directory)\n connect(3\u003c/var/lib/sss/mc/passwd\u003e, { .family: PF_LOCAL, path: /var/run/nscd/socket }, 0x6e) \u003d -1 ENOENT (No such file or directory)\n connect(4\u003csocket:[16604782]\u003e, { .family: PF_LOCAL, path: /var/lib/sss/pipes/nss }, 0x6e) \u003d 0\n connect(7, { .family: PF_LOCAL, path: /var/run/nscd/socket }, 0x6e) \u003d -1 ENOENT (No such file or directory)\n connect(7, { .family: PF_LOCAL, path: /var/run/nscd/socket }, 0x6e) \u003d -1 ENOENT (No such file or directory)\n connect(5\u003c/etc/hosts\u003e, { .family: PF_LOCAL, path: /var/run/nscd/socket }, 0x6e) \u003d -1 ENOENT (No such file or directory)\n connect(5\u003c/etc/hosts\u003e, { .family: PF_LOCAL, path: /var/run/nscd/socket }, 0x6e) \u003d -1 ENOENT (No such file or directory)\n connect(5\u003c/etc/hosts\u003e, { .family: PF_INET, port: 53, addr: 192.168.44.1 }, 0x10) \u003d 0\n connect(5\u003c/etc/hosts\u003e, { .family: PF_INET, port: 22, addr: 146.112.61.108 }, 0x10) \u003d 0\n connect(5\u003c/etc/hosts\u003e, { .family: PF_INET6, port: 22, addr: ::ffff:146.112.61.108 }, 0x1c) \u003d 0\n ^C#\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-5xkrbcpjsgnr3zt1aqdd7nvc@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 6f563674935e6dc9e2190ce798c1917f51af6eed\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Jul 16 15:33:20 2019 -0300\n\n perf augmented_raw_syscalls: Rename augmented_args_filename to augmented_args_payload\n \n It\u0027ll get other stuff in there than just filenames, starting with\n sockaddr for \u0027connect\u0027 and \u0027bind\u0027.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-bsexidtsn91ehdpzcd6n5fm9@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 8b8044e5c9527cdfff3b0937f2d17470cc4acf9e\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Jul 16 15:24:58 2019 -0300\n\n perf trace: Look for default name for entries in the syscalls prog array\n \n I.e. just look for \"!syscalls:sys_enter_\" or \"exit_\" plus the syscall\n name, that way we need just to add entries to the\n augmented_raw_syscalls.c BPF source to add handlers.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-6xavwddruokp6ohs7tf4qilb@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 8d5da2649d8211e63c5f65ccf8945f2c46a9c0b8\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Jul 16 14:55:57 2019 -0300\n\n perf augmented_raw_syscalls: Support copying two string syscall args\n \n Starting with the renameat and renameat2 syscall, that both receive as\n second and fourth parameters a pathname:\n \n # perf trace -e rename* mv one ANOTHER\n LLVM: dumping /home/acme/git/perf/tools/perf/examples/bpf/augmented_raw_syscalls.o\n mv: cannot stat \u0027one\u0027: No such file or directory\n renameat2(AT_FDCWD, \"one\", AT_FDCWD, \"ANOTHER\", RENAME_NOREPLACE) \u003d -1 ENOENT (No such file or directory)\n #\n \n Since the per CPU scratch buffer map has space for two maximum sized\n pathnames, the verifier is satisfied that there will be no overrun.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-x2uboyg5kx2wqeru288209b6@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit bf134ca6c8eafd7ddc28840f767259b97e950bac\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Jul 16 12:34:59 2019 -0300\n\n perf augmented_raw_syscalls: Switch to using BPF_MAP_TYPE_PROG_ARRAY\n \n Trying to control what arguments to copy, which ones were strings, etc\n all from userspace via maps went nowhere, lots of difficulties to get\n the verifier satisfied, so use what the fine BPF guys designed for such\n a syscall handling mechanism: bpf_tail_call + BPF_MAP_TYPE_PROG_ARRAY.\n \n The series leading to this should have explained it thoroughly, but the\n end result, explained via gdb should help understand this:\n \n Breakpoint 1, syscall_arg__scnprintf_filename (bf\u003d0xc002b1 \"\", size\u003d2031, arg\u003d0x7fffffff7970) at builtin-trace.c:1268\n 1268 {\n (gdb) n\n 1269 unsigned long ptr \u003d arg-\u003eval;\n (gdb) n\n 1271 if (arg-\u003eaugmented.args)\n (gdb) n\n 1272 return syscall_arg__scnprintf_augmented_string(arg, bf, size);\n (gdb) s\n syscall_arg__scnprintf_augmented_string (arg\u003d0x7fffffff7970, bf\u003d0xc002b1 \"\", size\u003d2031) at builtin-trace.c:1251\n 1251 {\n (gdb) n\n 1252 struct augmented_arg *augmented_arg \u003d arg-\u003eaugmented.args;\n (gdb) n\n 1253 size_t printed \u003d scnprintf(bf, size, \"\\\"%.*s\\\"\", augmented_arg-\u003esize, augmented_arg-\u003evalue);\n (gdb) n\n 1258 int consumed \u003d sizeof(*augmented_arg) + augmented_arg-\u003esize;\n (gdb) p bf\n $1 \u003d 0xc002b1 \"\\\"/etc/ld.so.cache\\\"\"\n (gdb) bt\n #0 syscall_arg__scnprintf_augmented_string (arg\u003d0x7fffffff7970, bf\u003d0xc002b1 \"\\\"/etc/ld.so.cache\\\"\", size\u003d2031) at builtin-trace.c:1258\n #1 0x0000000000492634 in syscall_arg__scnprintf_filename (bf\u003d0xc002b1 \"\\\"/etc/ld.so.cache\\\"\", size\u003d2031, arg\u003d0x7fffffff7970) at builtin-trace.c:1272\n #2 0x0000000000493cd7 in syscall__scnprintf_val (sc\u003d0xc0de68, bf\u003d0xc002b1 \"\\\"/etc/ld.so.cache\\\"\", size\u003d2031, arg\u003d0x7fffffff7970, val\u003d140737354091036) at builtin-trace.c:1689\n #3 0x000000000049404f in syscall__scnprintf_args (sc\u003d0xc0de68, bf\u003d0xc002a7 \"AT_FDCWD, \\\"/etc/ld.so.cache\\\"\", size\u003d2041, args\u003d0x7ffff6cbf1ec \"\\234\\377\\377\\377\", augmented_args\u003d0x7ffff6cbf21c, augmented_args_size\u003d28, trace\u003d0x7fffffffa170,\n thread\u003d0xbff940) at builtin-trace.c:1756\n #4 0x0000000000494a97 in trace__sys_enter (trace\u003d0x7fffffffa170, evsel\u003d0xbe1900, event\u003d0x7ffff6cbf1a0, sample\u003d0x7fffffff7b00) at builtin-trace.c:1975\n #5 0x0000000000496ff1 in trace__handle_event (trace\u003d0x7fffffffa170, event\u003d0x7ffff6cbf1a0, sample\u003d0x7fffffff7b00) at builtin-trace.c:2685\n #6 0x0000000000497edb in __trace__deliver_event (trace\u003d0x7fffffffa170, event\u003d0x7ffff6cbf1a0) at builtin-trace.c:3029\n #7 0x000000000049801e in trace__deliver_event (trace\u003d0x7fffffffa170, event\u003d0x7ffff6cbf1a0) at builtin-trace.c:3056\n #8 0x00000000004988de in trace__run (trace\u003d0x7fffffffa170, argc\u003d2, argv\u003d0x7fffffffd660) at builtin-trace.c:3258\n #9 0x000000000049c2d3 in cmd_trace (argc\u003d2, argv\u003d0x7fffffffd660) at builtin-trace.c:4220\n #10 0x00000000004dcb6c in run_builtin (p\u003d0xa18e00 \u003ccommands+576\u003e, argc\u003d5, argv\u003d0x7fffffffd660) at perf.c:304\n #11 0x00000000004dcdd9 in handle_internal_command (argc\u003d5, argv\u003d0x7fffffffd660) at perf.c:356\n #12 0x00000000004dcf20 in run_argv (argcp\u003d0x7fffffffd4bc, argv\u003d0x7fffffffd4b0) at perf.c:400\n #13 0x00000000004dd28c in main (argc\u003d5, argv\u003d0x7fffffffd660) at perf.c:522\n (gdb)\n (gdb) continue\n Continuing.\n openat(AT_FDCWD, \"/etc/ld.so.cache\", O_RDONLY|O_CLOEXEC) \u003d 3\n \n Now its a matter of automagically assigning the BPF programs copying\n syscall arg pointers to functions that are \"open\"-like (i.e. that need\n only the first syscall arg copied as a string), or \"openat\"-like (2nd\n arg, etc).\n \n End result in tool output:\n \n # perf trace -e open* ls /tmp/notthere\n LLVM: dumping /home/acme/git/perf/tools/perf/examples/bpf/augmented_raw_syscalls.o\n openat(AT_FDCWD, \"/etc/ld.so.cache\", O_RDONLY|O_CLOEXEC) \u003d 3\n openat(AT_FDCWD, \"/lib64/libselinux.so.1\", O_RDONLY|O_CLOEXEC) \u003d 3\n openat(AT_FDCWD, \"/lib64/libcap.so.2\", O_RDONLY|O_CLOEXEC) \u003d 3\n openat(AT_FDCWD, \"/lib64/libc.so.6\", O_RDONLY|O_CLOEXEC) \u003d 3\n openat(AT_FDCWD, \"/lib64/libpcre2-8.so.0\", O_RDONLY|O_CLOEXEC) \u003d 3\n openat(AT_FDCWD, \"/lib64/libdl.so.2\", O_RDONLY|O_CLOEXEC) \u003d 3\n openat(AT_FDCWD, \"/lib64/libpthread.so.0\", O_RDONLY|O_CLOEXEC) \u003d 3\n openat(AT_FDCWD, \"\", O_RDONLY|O_CLOEXEC) \u003d 3\n openat(AT_FDCWD, \"/usr/share/locale/locale.alias\", O_RDONLY|O_CLOEXEC) \u003d 3\n openat(AT_FDCWD, \"/usr/share/locale/en_US.UTF-8/LC_MESSAGES/coreutils.mo\", O_RDONLY) \u003d ls: cannot access \u0027/tmp/notthere\u0027-1 ENOENT (No such file or directory)\n openat(AT_FDCWD, \"/usr/share/locale/en_US.utf8/LC_MESSAGES/coreutils.mo\", O_RDONLY) \u003d -1 ENOENT (No such file or directory)\n openat(AT_FDCWD, \"/usr/share/locale/en_US/LC_MESSAGES/coreutils.mo\", O_RDONLY) \u003d -1 ENOENT (No such file or directory)\n openat(AT_FDCWD, \"/usr/share/locale/en.UTF-8/LC_MESSAGES/coreutils.mo\", O_RDONLY) \u003d -1 ENOENT (No such file or directory)\n openat(AT_FDCWD, \"/usr/share/locale/en.utf8/LC_MESSAGES/coreutils.mo\", O_RDONLY) \u003d -1 ENOENT (No such file or directory)\n openat(AT_FDCWD, \"/usr/share/locale/en/LC_MESSAGES/coreutils.mo\", O_RDONLY: No such file or directory) \u003d\n -1 ENOENT (No such file or directory)\n openat(AT_FDCWD, \"/usr/share/locale/en_US.UTF-8/LC_MESSAGES/libc.mo\", O_RDONLY) \u003d -1 ENOENT (No such file or directory)\n openat(AT_FDCWD, \"/usr/share/locale/en_US.utf8/LC_MESSAGES/libc.mo\", O_RDONLY) \u003d -1 ENOENT (No such file or directory)\n openat(AT_FDCWD, \"/usr/share/locale/en_US/LC_MESSAGES/libc.mo\", O_RDONLY) \u003d -1 ENOENT (No such file or directory)\n openat(AT_FDCWD, \"/usr/share/locale/en.UTF-8/LC_MESSAGES/libc.mo\", O_RDONLY) \u003d -1 ENOENT (No such file or directory)\n openat(AT_FDCWD, \"/usr/share/locale/en.utf8/LC_MESSAGES/libc.mo\", O_RDONLY) \u003d -1 ENOENT (No such file or directory)\n openat(AT_FDCWD, \"/usr/share/locale/en/LC_MESSAGES/libc.mo\", O_RDONLY) \u003d -1 ENOENT (No such file or directory)\n #\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-snc7ry99cl6r0pqaspjim98x@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 236dd5838871024d58d354ff8d0dab00ee59a867\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Jul 16 12:31:10 2019 -0300\n\n perf augmented_raw_syscalls: Add handler for \"openat\"\n \n I.e. for a syscall that has its second argument being a string, its\n difficult these days to find \u0027open\u0027 being used in the wild :-)\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-yf3kbzirqrukd3fb2sp5qx4p@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit b119970aa541091e405373399690c24ead9d2920\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Jul 16 11:53:03 2019 -0300\n\n perf trace: Handle raw_syscalls:sys_enter just like the BPF_OUTPUT augmented event\n \n So, we use a PERF_COUNT_SW_BPF_OUTPUT to output the augmented sys_enter\n payload, i.e. to output more than just the raw syscall args, and if\n something goes wrong when handling an unfiltered syscall, we bail out\n and just return 1 in the bpf program associated with\n raw_syscalls:sys_enter, meaning, don\u0027t filter that tracepoint, in which\n case what will appear in the perf ring buffer isn\u0027t the BPF_OUTPUT\n event, but the original raw_syscalls:sys_enter event with its normal\n payload.\n \n Now that we\u0027re switching to using a bpf_tail_call +\n BPF_MAP_TYPE_PROG_ARRAY we\u0027re going to use this in the common case, so a\n bug where raw_syscalls:sys_enter wasn\u0027t being handled by\n trace__sys_enter() surfaced and for that case, instead of using the\n strace-like augmenter (trace__sys_enter()), we continued to use the\n normal generic tracepoint handler:\n \n (gdb) p evsel\n $2 \u003d (struct perf_evsel *) 0xc03e40\n (gdb) p evsel-\u003ename\n $3 \u003d 0xbc56c0 \"raw_syscalls:sys_enter\"\n (gdb) p ((struct perf_evsel *) 0xc03e40)-\u003ename\n $4 \u003d 0xbc56c0 \"raw_syscalls:sys_enter\"\n (gdb) p ((struct perf_evsel *) 0xc03e40)-\u003ehandler\n $5 \u003d (void *) 0x495eb3 \u003ctrace__event_handler\u003e\n \n This resulted in this:\n \n 0.027 raw_syscalls:sys_enter:NR 12 (0, 7fcfcac64c9b, 4d, 7fcfcac64c9b, 7fcfcac6ce00, 19)\n ... [continued]: brk()) \u003d 0x563b88677000\n \n I.e. only the sys_exit tracepoint was being properly handled, but since\n the sys_enter went to the generic trace__event_handler() we printed it\n using libtraceevent\u0027s formatter instead of \u0027perf trace\u0027s strace-like\n one.\n \n Fix it by setting trace__sys_enter() as the handler for\n raw_syscalls:sys_enter and setup the tp_field tracepoint field\n accessors.\n \n Now, to test it we just make raw_syscalls:sys_enter return 1 right after\n checking if the pid is filtered, making it not use\n bpf_perf_output_event() but rather ask for the tracepoint not to be\n filtered and the result is the expected one:\n \n brk(NULL) \u003d 0x556f42d6e000\n \n I.e. raw_syscalls:sys_enter returns 1, gets handled by\n trace__sys_enter() and gets it combined with the raw_syscalls:sys_exit\n in a strace-like way.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-0mkocgk31nmy0odknegcby4z@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 3803a229312de539d2878f2fc5c6ee0202ce6728\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Jul 16 11:27:03 2019 -0300\n\n perf trace: Put the per-syscall entry/exit prog_array BPF map infrastructure in place\n \n I.e. look for \"syscalls_sys_enter\" and \"syscalls_sys_exit\" BPF maps of\n type PROG_ARRAY and populate it with the handlers as specified per\n syscall, for now only \u0027open\u0027 is wiring it to something, in time all\n syscalls that need to copy arguments entering a syscall or returning\n from one will set these to the right handlers, reusing when possible\n pre-existing ones.\n \n Next step is to use bpf_tail_call() into that.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-t0p4u43i9vbpzs1xtowna3gb@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 6ff8fff45611e0b5ff4c0979cd0470b5cbc0a031\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Tue Jul 16 10:59:19 2019 -0300\n\n perf trace: Allow specifying the bpf prog to augment specific syscalls\n \n This is a step in the direction of being able to use a\n BPF_MAP_TYPE_PROG_ARRAY to handle syscalls that need to copy pointer\n payloads in addition to the raw tracepoint syscall args.\n \n There is a first example in\n tools/perf/examples/bpf/augmented_raw_syscalls.c for the \u0027open\u0027 syscall.\n \n Next step is to introduce the prog array map and use this \u0027open\u0027\n augmenter, then use that augmenter in other syscalls that also only copy\n the first arg as a string, and then show how to use with a syscall that\n reads more than one filename, like \u0027rename\u0027, etc.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-pys4v57x5qqrybb4cery2mc8@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 5834da7f10917245d191032f76f132e62e57197c\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Jul 15 17:51:43 2019 -0300\n\n perf trace: Add BPF handler for unaugmented syscalls\n \n Will be used to assign to syscalls that don\u0027t need augmentation, i.e.\n those with just integer args.\n \n All syscalls will be in a BPF_MAP_TYPE_PROG_ARRAY, and the\n bpf_tail_call() keyed by the syscall id will either find nothing in\n place, which means the syscall is being filtered, or a function that\n will either add things like filenames to the ring buffer, right after\n the raw syscall args, or be this unaugmented handler that will just\n return 1, meaning don\u0027t filter the original\n raw_syscalls:sys_{enter,exit} tracepoint.\n \n For now it is not really being used, this is just leg work to break the\n patch into smaller pieces.\n \n It introduces a trace__find_bpf_program_by_title() helper that in turn\n uses libbpf\u0027s bpf_object__find_program_by_title() on the BPF object with\n the __augmented_syscalls__ map. \"title\" is how libbpf calls the SEC()\n argument for functions, i.e. the ELF section that follows a convention\n to specify what BPF program (a function with this SEC() marking) should\n be connected to which tracepoint, kprobes, etc.\n \n In perf anything that is of the form SEC(\"sys:event_name\") will be\n connected to that tracepoint by perf\u0027s BPF loader.\n \n In this case its something that will be bpf_tail_call()ed from either\n the \"raw_syscalls:sys_enter\" or \"raw_syscall:sys_exit\" tracepoints, so\n its named \"!raw_syscalls:unaugmented\" to convey that idea, i.e. its not\n going to be directly attached to a tracepoint, thus it starts with a\n \"!\".\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-meucpjx2u0slpkayx56lxqq6@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 83e69b92b10c2a8b75e1919b7074338f8bdbacaf\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Jul 15 17:28:25 2019 -0300\n\n perf trace: Order -e syscalls table\n \n The ev_qualifier is an array with the syscall ids passed via -e on the\n command line, sort it as we\u0027ll search it when setting up the\n BPF_MAP_TYPE_PROG_ARRAY.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-c8hprylp3ai6e0z9burn2r3s@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 5ca0b7f5004a5f1a35c1cb47894790ad98e34ed1\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Jul 15 17:03:10 2019 -0300\n\n perf trace: Look up maps just on the __augmented_syscalls__ BPF object\n \n We can conceivably have multiple BPF object files for other purposes, so\n better look just on the BPF object containing the __augmented_syscalls__\n map for all things augmented_syscalls related.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Brendan Gregg \u003cbrendan.d.gregg@gmail.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-3jt8knkuae9lt705r1lns202@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit c8c805707ed4c5d976210821da3242af8610a533\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Jul 15 16:58:23 2019 -0300\n\n perf trace: Add pointer to BPF object containing __augmented_syscalls__\n \n So that we can use it when looking for other components of that object\n file, such as other programs to add to the BPF_MAP_TYPE_PROG_ARRAY and\n use with bpf_tail_call().\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-1ibmz7ouv6llqxajy7m8igtd@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit af4a0991f40a1e50e5caff0317f152df2c82bdeb\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Jul 15 16:22:57 2019 -0300\n\n perf evsel: Store backpointer to attached bpf_object\n \n We may want to get to this bpf_object, to search for other BPF programs,\n etc.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-3y8hrb6lszjfi23vjlic3cib@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 2620b7e3696a9470c7cda0a08e55813fd5e57e5c\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Jul 15 15:29:34 2019 -0300\n\n perf bpf: Do not attach a BPF prog to a tracepoint if its name starts with !\n \n With BPF_MAP_TYPE_PROG_ARRAY + bpf_tail_call() we want to have BPF\n programs, i.e. functions in a object file that perf\u0027s BPF loader\n shouldn\u0027t try to attach to anything, i.e. \"!syscalls:sys_enter_open\"\n should just stay there, not be attached to a tracepoint with that name,\n it\u0027ll be used by, for instance, \u0027perf trace\u0027 to associate with syscalls\n that copy, in addition to the syscall raw args, a filename pointed by\n the first arg, i.e. multiple syscalls that need copying the same pointer\n arg in the same way, as a filename, for instance, will share the same\n BPF program/function.\n \n Right now when perf\u0027s BPF loader sees a function with a name\n \"sys:name\" it\u0027ll look for a tracepoint and will associate that BPF\n program with it, say:\n \n SEC(\"raw_syscalls:sys_enter\")\n int sys_enter(struct syscall_enter_args *args)\n {\n //SNIP\n }\n \n Will crate a perf_evsel tracepoint event and then associate with it that\n BPF program.\n \n This convention at some point will switch to the one used by the BPF\n loader in libbpf, but to experiment with BPF_MAP_TYPE_PROG_ARRAY in\n \u0027perf trace\u0027 lets do this, that will not require changing too much\n stuff.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-lk6dasjr1yf9rtvl292b2hpc@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit 941a7658e065e339ebdaf27b9a7702f9935d1d4a\nAuthor: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nDate: Mon Jul 15 15:25:41 2019 -0300\n\n perf include bpf: Add bpf_tail_call() prototype\n \n Will be used together with BPF_MAP_TYPE_PROG_ARRAY in\n tools/perf/examples/bpf/augmented_raw_syscalls.c.\n \n Cc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\n Cc: Jiri Olsa \u003cjolsa@kernel.org\u003e\n Cc: Luis Cláudio Gonçalves \u003clclaudio@redhat.com\u003e\n Cc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\n Link: https://lkml.kernel.org/n/tip-pd1bpy8i31nta6jqwdex871g@git.kernel.org\n Signed-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n\ncommit fa21cbea926c82333a315d21a06619e21cebb787\nAuthor: Marc Gonzalez \u003cmarc.w.gonzalez@free.fr\u003e\nDate: Thu Jul 4 10:57:06 2019 +0200\n\n pinctrl: msm8998: Squash TSIF pins together\n \n TSIF is the Transport Stream Interface.\n First, rename tsif1 to tsif0, and tsif2 to tsif1.\n Then squash all 5 tsif0 pins into a single function.\n Same for tsif1.\n \n Signed-off-by: Marc Gonzalez \u003cmarc.w.gonzalez@free.fr\u003e\n Link: https://lore.kernel.org/r/503b2ae8-ead6-70cd-7b21-ce5f5166a23a@free.fr\n Reviewed-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Acked-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 0470e5e38c9d97e7ce1e804cd54cf59accdf3ac2\nAuthor: Qian Cai \u003ccai@lca.pw\u003e\nDate: Mon Jul 22 14:34:42 2019 -0400\n\n net/mlx5: fix -Wtype-limits compilation warnings\n \n The commit b9a7ba556207 (\"net/mlx5: Use event mask based on device\n capabilities\") introduced a few compilation warnings due to it bumps\n MLX5_EVENT_TYPE_MAX from 0x27 to 0x100 which is always greater than\n an \"struct {mlx5_eqe|mlx5_nb}.type\" that is an \"u8\".\n \n drivers/net/ethernet/mellanox/mlx5/core/eq.c: In function\n \u0027mlx5_eq_notifier_register\u0027:\n drivers/net/ethernet/mellanox/mlx5/core/eq.c:948:21: warning: comparison\n is always false due to limited range of data type [-Wtype-limits]\n if (nb-\u003eevent_type \u003e\u003d MLX5_EVENT_TYPE_MAX)\n ^~\n drivers/net/ethernet/mellanox/mlx5/core/eq.c: In function\n \u0027mlx5_eq_notifier_unregister\u0027:\n drivers/net/ethernet/mellanox/mlx5/core/eq.c:959:21: warning: comparison\n is always false due to limited range of data type [-Wtype-limits]\n if (nb-\u003eevent_type \u003e\u003d MLX5_EVENT_TYPE_MAX)\n \n Fix them by removing unnecessary checkings.\n \n Fixes: b9a7ba556207 (\"net/mlx5: Use event mask based on device capabilities\")\n Signed-off-by: Qian Cai \u003ccai@lca.pw\u003e\n Signed-off-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n\ncommit 49c98421e6ab33665e8ee7901218a712f5b0db2e\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Mon Jul 29 06:10:33 2019 -0400\n\n bnxt_en: Add PCI IDs for 57500 series NPAR devices.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1dc88b97a020148c0eea6c595d511a19c2fab347\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Mon Jul 29 06:10:32 2019 -0400\n\n bnxt_en: Support all variants of the 5750X chip family.\n \n Define the 57508, 57504, and 57502 chip IDs that are all part of the\n BNXT_CHIP_P5 family of chips.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7c3809181468a219aa2abd25910bd3b02b89b0de\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Mon Jul 29 06:10:31 2019 -0400\n\n bnxt_en: Refactor bnxt_init_one() and turn on TPA support on 57500 chips.\n \n With the new TPA feature in the 57500 chips, we need to discover the\n feature first before setting up the netdev features. Refactor the\n the firmware probe and init logic more cleanly into 2 functions and\n and make these calls before setting up the netdev features.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 78e7b86605b460e8b40622d16d840f9276d58627\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Mon Jul 29 06:10:30 2019 -0400\n\n bnxt_en: Support TPA counters on 57500 chips.\n \n Support the new expanded TPA v2 counters on 57500 B0 chips for\n ethtool -S.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4e7485066373f3e9a87fa063b65d0838990753e5\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Mon Jul 29 06:10:29 2019 -0400\n\n bnxt_en: Allocate the larger per-ring statistics block for 57500 chips.\n \n The new TPA implemantation has additional TPA counters that extend the\n per-ring statistics block. Allocate the proper size accordingly.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ee79566e65945dcf557bcfb9335e46fac67fb002\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Mon Jul 29 06:10:28 2019 -0400\n\n bnxt_en: Refactor ethtool ring statistics logic.\n \n The current code assumes that the per ring statistics counters are\n fixed. In newer chips that support a newer version of TPA, the\n TPA counters are also changed. Refactor the code by defining these\n counter names in arrays so that it is easy to add a new array for\n a new set of counters supported by the newer chips.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 67912c366d4bb0a9d108459e7c845cc7ba83f76f\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Mon Jul 29 06:10:27 2019 -0400\n\n bnxt_en: Add hardware GRO setup function for 57500 chips.\n \n Add a more optimized hardware GRO function to setup the SKB on 57500\n chips. Some workaround code is no longer needed on 57500 chips and\n the pseudo checksum is also calculated in hardware, so no need to\n do the software pseudo checksum in the driver.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ec4d8e7cf024e42def027531676918048e5c7982\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Mon Jul 29 06:10:26 2019 -0400\n\n bnxt_en: Add TPA ID mapping logic for 57500 chips.\n \n The new TPA feature on 57500 supports a larger number of concurrent TPAs\n (up to 1024) divided among the functions. We need to add some logic to\n map the hardware TPA ID to a software index that keeps track of each TPA\n in progress. A 1:1 direct mapping without translation would be too\n wasteful as we would have to allocate 1024 TPA structures for each RX\n ring on each PCI function.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit bfcd8d791ec18496772d117774398e336917f56e\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Mon Jul 29 06:10:25 2019 -0400\n\n bnxt_en: Add fast path logic for TPA on 57500 chips.\n \n With all the previous refactoring, the TPA fast path can now be\n modified slightly to support TPA on the new chips. The main\n difference is that the agg completions are retrieved differently using\n the bnxt_get_tpa_agg_p5() function on the new chips.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f45b7b78c619cd73c7ca25b68c6ba9653b8e4a0a\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Mon Jul 29 06:10:24 2019 -0400\n\n bnxt_en: Set TPA GRO mode flags on 57500 chips properly.\n \n On 57500 chips, hardware GRO mode cannot be determined from the TPA\n end, so we need to check bp-\u003eflags to determine if we are in hardware\n GRO mode or not. Modify bnxt_set_features so that the TPA flags\n in bp-\u003eflags don\u0027t change until the device is closed. This will ensure\n that the fast path can safely rely on bp-\u003eflags to determine the\n TPA mode.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit bee5a188b71657092dc9eb1a529b4e502fe51444\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Mon Jul 29 06:10:23 2019 -0400\n\n bnxt_en: Refactor tunneled hardware GRO logic.\n \n The 2 GRO functions to set up the hardware GRO SKB fields for 2\n different hardware chips have practically identical logic for\n tunneled packets. Refactor the logic into a separate bnxt_gro_tunnel()\n function that can be used by both functions.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 8fe88ce7ab3181a11989eb7a8bb00c42a2b7b3b0\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Mon Jul 29 06:10:22 2019 -0400\n\n bnxt_en: Handle standalone RX_AGG completions.\n \n On the new 57500 chips, these new RX_AGG completions are not coalesced\n at the TPA_END completion. Handle these by storing them in the\n array in the bnxt_tpa_info struct, as they are seen when processing\n the CMPL ring.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 79632e9ba38671215fb193346ef6fb8db582744d\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Mon Jul 29 06:10:21 2019 -0400\n\n bnxt_en: Expand bnxt_tpa_info struct to support 57500 chips.\n \n Add an aggregation array to bnxt_tpa_info struct to keep track of the\n aggregation completions. The aggregation completions are not\n completed at the TPA_END completion on 57500 chips so we need to\n keep track of them. The array is only allocated on the new chips\n when required. An agg_count field is also added to keep track of the\n number of these completions.\n \n The maximum concurrent TPA is now discovered from firmware instead of\n the hardcoded 64. Add a new bp-\u003emax_tpa to keep track of maximum\n configured TPA.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4a228a3a5e58e5c05c6ffb5b430e5cb936865a8b\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Mon Jul 29 06:10:20 2019 -0400\n\n bnxt_en: Refactor TPA logic.\n \n Refactor the TPA logic slightly, so that the code can be more easily\n extended to support TPA on the new 57500 chips. In particular, the\n logic to get the next aggregation completion is refactored into a\n new function bnxt_get_agg() so that this operation is made more\n generalized. This operation will be different on the new chip in TPA\n mode. The logic to recycle the aggregation buffers has a new start\n index parameter added for the same purpose.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 218a8a71d91ab9e52205f4098cf1fe121c98850e\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Mon Jul 29 06:10:19 2019 -0400\n\n bnxt_en: Add TPA structure definitions for BCM57500 chips.\n \n The new chips have a slightly modified TPA interface for LRO/GRO_HW.\n Modify the TPA structures so that the same structures can also be\n used on the new chips.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 2792b5b95ed5f38279da08f467a490687332324d\nAuthor: Michael Chan \u003cmichael.chan@broadcom.com\u003e\nDate: Mon Jul 29 06:10:18 2019 -0400\n\n bnxt_en: Update firmware interface spec. to 1.10.0.89.\n \n Among the changes are new CoS discard counters and new ctx_hw_stats_ext\n struct for the latest 5750X B0 chips.\n \n Signed-off-by: Michael Chan \u003cmichael.chan@broadcom.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 473d924d7d46cb57aa4c1863261d18366af345af\nAuthor: Oliver Hartkopp \u003csocketcan@hartkopp.net\u003e\nDate: Mon Jul 29 22:40:56 2019 +0200\n\n can: fix ioctl function removal\n \n Commit 60649d4e0af (\"can: remove obsolete empty ioctl() handler\") replaced the\n almost empty can_ioctl() function with sock_no_ioctl() which always returns\n -EOPNOTSUPP.\n \n Even though we don\u0027t have any ioctl() functions on socket/network layer we need\n to return -ENOIOCTLCMD to be able to forward ioctl commands like SIOCGIFINDEX\n to the network driver layer.\n \n This patch fixes the wrong return codes in the CAN network layer protocols.\n \n Reported-by: kernel test robot \u003crong.a.chen@intel.com\u003e\n Fixes: 60649d4e0af (\"can: remove obsolete empty ioctl() handler\")\n Signed-off-by: Oliver Hartkopp \u003csocketcan@hartkopp.net\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 301efe96f777c0197502b9399694ae83779579b3\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Mon Jul 29 11:26:12 2019 +0000\n\n drm/i915/uc: Don\u0027t fail on HuC firmware failure\n \n HuC is usually not a critical component, so we can safely ignore\n firmware load or authentication failures unless HuC was explicitly\n requested by the user.\n \n v2: add convenient way to disable loading (Chris)\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e #v1\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190729112612.37476-1-michal.wajdeczko@intel.com\n\ncommit 1375dc4a4579d5e767dd8c2d2abcd929ff59d0a7\nAuthor: Toke Høiland-Jørgensen \u003ctoke@redhat.com\u003e\nDate: Fri Jul 26 18:06:58 2019 +0200\n\n tools: Add definitions for devmap_hash map type\n \n This adds selftest and bpftool updates for the devmap_hash map type.\n \n Signed-off-by: Toke Høiland-Jørgensen \u003ctoke@redhat.com\u003e\n Acked-by: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit e42346192c9f179a9a47845e4663ad2f453d2b7b\nAuthor: Toke Høiland-Jørgensen \u003ctoke@redhat.com\u003e\nDate: Fri Jul 26 18:06:57 2019 +0200\n\n tools/libbpf_probes: Add new devmap_hash type\n \n This adds the definition for BPF_MAP_TYPE_DEVMAP_HASH to libbpf_probes.c in\n tools/lib/bpf.\n \n Signed-off-by: Toke Høiland-Jørgensen \u003ctoke@redhat.com\u003e\n Acked-by: Yonghong Song \u003cyhs@fb.com\u003e\n Acked-by: Jesper Dangaard Brouer \u003cbrouer@redhat.com\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 10fbe21163fc6f05f5991d9cf9a2da9bdee4e834\nAuthor: Toke Høiland-Jørgensen \u003ctoke@redhat.com\u003e\nDate: Fri Jul 26 18:06:56 2019 +0200\n\n tools/include/uapi: Add devmap_hash BPF map type\n \n This adds the devmap_hash BPF map type to the uapi headers in tools/.\n \n Signed-off-by: Toke Høiland-Jørgensen \u003ctoke@redhat.com\u003e\n Acked-by: Yonghong Song \u003cyhs@fb.com\u003e\n Acked-by: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 6f9d451ab1a33728adb72d7ff66a7b374d665176\nAuthor: Toke Høiland-Jørgensen \u003ctoke@redhat.com\u003e\nDate: Fri Jul 26 18:06:55 2019 +0200\n\n xdp: Add devmap_hash map type for looking up devices by hashed index\n \n A common pattern when using xdp_redirect_map() is to create a device map\n where the lookup key is simply ifindex. Because device maps are arrays,\n this leaves holes in the map, and the map has to be sized to fit the\n largest ifindex, regardless of how many devices actually are actually\n needed in the map.\n \n This patch adds a second type of device map where the key is looked up\n using a hashmap, instead of being used as an array index. This allows maps\n to be densely packed, so they can be smaller.\n \n Signed-off-by: Toke Høiland-Jørgensen \u003ctoke@redhat.com\u003e\n Acked-by: Yonghong Song \u003cyhs@fb.com\u003e\n Acked-by: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit fca16e51078e8e5c0af839426b3d2dcd2bede135\nAuthor: Toke Høiland-Jørgensen \u003ctoke@redhat.com\u003e\nDate: Fri Jul 26 18:06:53 2019 +0200\n\n xdp: Refactor devmap allocation code for reuse\n \n The subsequent patch to add a new devmap sub-type can re-use much of the\n initialisation and allocation code, so refactor it into separate functions.\n \n Signed-off-by: Toke Høiland-Jørgensen \u003ctoke@redhat.com\u003e\n Acked-by: Yonghong Song \u003cyhs@fb.com\u003e\n Acked-by: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 6dbff13ca8a2ad2fddd904c2e789dd5e59a8644c\nAuthor: Toke Høiland-Jørgensen \u003ctoke@redhat.com\u003e\nDate: Fri Jul 26 18:06:52 2019 +0200\n\n include/bpf.h: Remove map_insert_ctx() stubs\n \n When we changed the device and CPU maps to use linked lists instead of\n bitmaps, we also removed the need for the map_insert_ctx() helpers to keep\n track of the bitmaps inside each map. However, it seems I forgot to remove\n the function definitions stubs, so remove those here.\n \n Signed-off-by: Toke Høiland-Jørgensen \u003ctoke@redhat.com\u003e\n Acked-by: Yonghong Song \u003cyhs@fb.com\u003e\n Acked-by: Jesper Dangaard Brouer \u003cbrouer@redhat.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit af7b65054952a9898534d966f2e548d7692b7016\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Jul 29 12:23:56 2019 +0530\n\n leds: netxbig: Add of_node_put() in netxbig_leds_get_of_pdata()\n \n The variable gpio_ext_np in the function netxbig_leds_get_of_pdata takes\n the value returned by of_parse_phandle; hence, it must be put in order\n to prevent a memory leak. Add an of_node_put for gpio_ext_np before a\n return statement, and move a pre-existing of_node_put statement to right\n after the last usage of this variable.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 156189a6d7a797d6fa0b13dd6b4f32b11d234a99\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Mon Jul 29 01:40:15 2019 +0900\n\n leds: netxbig: remove legacy board-file support\n \n Since commit ebc278f15759 (\"ARM: mvebu: remove static LED setup for\n netxbig boards\"), no one in upstream passes in the platform data to\n this driver.\n \n Squash leds-kirkwood-netxbig.h into the driver, and remove the legacy\n board-file support.\n \n Link: https://lkml.org/lkml/2019/7/20/83\n Suggested-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 1cb9dfca39eb406960f8f84864ddd6ba329ec321\nAuthor: Rasmus Villemoes \u003crasmus.villemoes@prevas.dk\u003e\nDate: Mon Jul 22 23:37:26 2019 +0000\n\n net: dsa: mv88e6xxx: avoid some redundant vtu load/purge operations\n \n We have an ERPS (Ethernet Ring Protection Switching) setup involving\n mv88e6250 switches which we\u0027re in the process of switching to a BSP\n based on the mainline driver. Breaking any link in the ring works as\n expected, with the ring reconfiguring itself quickly and traffic\n continuing with almost no noticable drops. However, when plugging back\n the cable, we see 5+ second stalls.\n \n This has been tracked down to the userspace application in charge of\n the protocol missing a few CCM messages on the good link (the one that\n was not unplugged), causing it to broadcast a \"signal fail\". That\n message eventually reaches its link partner, which responds by\n blocking the port. Meanwhile, the first node has continued to block\n the port with the just plugged-in cable, breaking the network. And the\n reason for those missing CCM messages has in turn been tracked down to\n the VTU apparently being too busy servicing load/purge operations that\n the normal lookups are delayed.\n \n Initial state, the link between C and D is blocked in software.\n \n _____________________\n / \\\n | |\n A ----- B ----- C *---- D\n \n Unplug the cable between C and D.\n \n _____________________\n / \\\n | |\n A ----- B ----- C * * D\n \n Reestablish the link between C and D.\n _____________________\n / \\\n | |\n A ----- B ----- C *---- D\n \n Somehow, enough VTU/ATU operations happen inside C that prevents\n the application from receving the CCM messages from B in a timely\n manner, so a Signal Fail message is sent by C. When B receives\n that, it responds by blocking its port.\n \n _____________________\n / \\\n | |\n A ----- B *---* C *---- D\n \n Very shortly after this, the signal fail condition clears on the\n BC link (some CCM messages finally make it through), so C\n unblocks the port. However, a guard timer inside B prevents it\n from removing the blocking before 5 seconds have elapsed.\n \n It is not unlikely that our userspace ERPS implementation could be\n smarter and/or is simply buggy. However, this patch fixes the symptoms\n we see, and is a small optimization that should not break anything\n (knock wood). The idea is simply to avoid doing an VTU load of an\n entry identical to the one already present. To do that, we need to\n know whether mv88e6xxx_vtu_get() actually found an existing entry, or\n has just prepared a struct mv88e6xxx_vtu_entry for us to load. To that\n end, let vlan-\u003evalid be an output parameter. The other two callers of\n mv88e6xxx_vtu_get() are not affected by this patch since they pass\n new\u003dfalse.\n \n Signed-off-by: Rasmus Villemoes \u003crasmus.villemoes@prevas.dk\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ef14358546b115d9ea4108f21c6badd97334ad18\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Sun Jul 28 11:25:19 2019 +0200\n\n r8169: make use of xmit_more\n \n There was a previous attempt to use xmit_more, but the change had to be\n reverted because under load sometimes a transmit timeout occurred [0].\n Maybe this was caused by a missing memory barrier, the new attempt\n keeps the memory barrier before the call to netif_stop_queue like it\n is used by the driver as of today. The new attempt also changes the\n order of some calls as suggested by Eric.\n \n [0] https://lkml.org/lkml/2019/2/10/39\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 16e9111e9ee3edfcf6df120080378afc620cb4d3\nAuthor: Gal Pressman \u003cgalpress@amazon.com\u003e\nDate: Thu Jul 25 16:03:53 2019 +0300\n\n RDMA/efa: Expose device statistics\n \n Expose hardware statistics through the sysfs api:\n /sys/class/infiniband/efa_0/hw_counters/*.\n /sys/class/infiniband/efa_0/ports/1/hw_counters/*.\n \n Reviewed-by: Firas JahJah \u003cfirasj@amazon.com\u003e\n Reviewed-by: Yossi Leybovich \u003csleybo@amazon.com\u003e\n Signed-off-by: Gal Pressman \u003cgalpress@amazon.com\u003e\n Link: https://lore.kernel.org/r/20190725130353.11544-1-galpress@amazon.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit bda9045a200c173b2883a878b68b8b9218ae65c6\nAuthor: Parav Pandit \u003cparav@mellanox.com\u003e\nDate: Fri Jul 26 13:26:52 2019 -0500\n\n IB/bnxt_re: Do not notifify GID change event\n \n GID table entry operations such as add/remove/modify are triggered\n by the IB core for RoCE ports.\n Hence, remove GID change notification from hw driver.\n \n Reviewed-by: Selvin Xavier \u003cselvin.xavier@broadcom.com\u003e\n Tested-by: Selvin Xavier \u003cselvin.xavier@broadcom.com\u003e\n Signed-off-by: Parav Pandit \u003cparav@mellanox.com\u003e\n Link: https://lore.kernel.org/r/20190726182652.50037-1-parav@mellanox.com\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 171a9bae68c72f2d1260c3825203760856e6793b\nAuthor: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\nDate: Fri Jul 26 10:44:25 2019 -0700\n\n staging/octeon: Allow test build on !MIPS\n \n Add compile test support by moving all includes of files under\n asm/octeon into octeon-ethernet.h, and if we\u0027re not on MIPS,\n stub out all the calls into the octeon support code in octeon-stubs.h\n \n Signed-off-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n Acked-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c51ab067c5312695331129806cb9a42791903305\nAuthor: Ding Xiang \u003cdingxiang@cmss.chinamobile.com\u003e\nDate: Mon Jul 29 17:01:22 2019 +0800\n\n net: ag71xx: use resource_size for the ioremap size\n \n use resource_size to calcuate ioremap size and make\n the code simpler.\n \n Signed-off-by: Ding Xiang \u003cdingxiang@cmss.chinamobile.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7c288a5b27934281d9ea8b5807bc727268b7001a\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Jul 2 17:16:16 2019 +0100\n\n iommu/arm-smmu-v3: Operate directly on low-level queue where possible\n \n In preparation for rewriting the command queue insertion code to use a\n new algorithm, rework many of our queue macro accessors and manipulation\n functions so that they operate on the arm_smmu_ll_queue structure where\n possible. This will allow us to call these helpers on local variables\n without having to construct a full-blown arm_smmu_queue on the stack.\n \n No functional change.\n \n Tested-by: Ganapatrao Kulkarni \u003cgkulkarni@marvell.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 52be86374f48a0d34015257f3e06965910b5b774\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Jul 2 17:16:08 2019 +0100\n\n iommu/arm-smmu-v3: Move low-level queue fields out of arm_smmu_queue\n \n In preparation for rewriting the command queue insertion code to use a\n new algorithm, introduce a new arm_smmu_ll_queue structure which contains\n only the information necessary to perform queue arithmetic for a queue\n and will later be extended so that we can perform complex atomic\n manipulation on some of the fields.\n \n No functional change.\n \n Tested-by: Ganapatrao Kulkarni \u003cgkulkarni@marvell.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 8a073da07bac169601a1874606e09bdb62811978\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Jul 2 17:15:50 2019 +0100\n\n iommu/arm-smmu-v3: Drop unused \u0027q\u0027 argument from Q_OVF macro\n \n The Q_OVF macro doesn\u0027t need to access the arm_smmu_queue structure, so\n drop the unused macro argument.\n \n No functional change.\n \n Tested-by: Ganapatrao Kulkarni \u003cgkulkarni@marvell.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 2a8868f16e6b1987cf43f1f46d2a12b7b6ddcd88\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Jul 2 17:12:24 2019 +0100\n\n iommu/arm-smmu-v3: Separate s/w and h/w views of prod and cons indexes\n \n In preparation for rewriting the command queue insertion code to use a\n new algorithm, separate the software and hardware views of the prod and\n cons indexes so that manipulating the software state doesn\u0027t\n automatically update the hardware state at the same time.\n \n No functional change.\n \n Tested-by: Ganapatrao Kulkarni \u003cgkulkarni@marvell.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 3951c41af4a65ba418e6b1b973d398552bedb84f\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Jul 2 16:45:15 2019 +0100\n\n iommu/io-pgtable: Pass struct iommu_iotlb_gather to -\u003etlb_add_page()\n \n With all the pieces in place, we can finally propagate the\n iommu_iotlb_gather structure from the call to unmap() down to the IOMMU\n drivers\u0027 implementation of -\u003etlb_add_page(). Currently everybody ignores\n it, but the machinery is now there to defer invalidation.\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit a2d3a382d6c682e22b263c9e7f0d857c3fa6c9d6\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Jul 2 16:44:58 2019 +0100\n\n iommu/io-pgtable: Pass struct iommu_iotlb_gather to -\u003eunmap()\n \n Update the io-pgtable -\u003eunmap() function to take an iommu_iotlb_gather\n pointer as an argument, and update the callers as appropriate.\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit e953f7f2fa78d1c7fd064171f88457c6b1e21af9\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Jul 2 16:44:50 2019 +0100\n\n iommu/io-pgtable: Remove unused -\u003etlb_sync() callback\n \n The -\u003etlb_sync() callback is no longer used, so it can be removed.\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit abfd6fe0cd535d31ee83b668be6eb59ce6a8469d\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Jul 2 16:44:41 2019 +0100\n\n iommu/io-pgtable: Replace -\u003etlb_add_flush() with -\u003etlb_add_page()\n \n The -\u003etlb_add_flush() callback in the io-pgtable API now looks a bit\n silly:\n \n - It takes a size and a granule, which are always the same\n - It takes a \u0027bool leaf\u0027, which is always true\n - It only ever flushes a single page\n \n With that in mind, replace it with an optional -\u003etlb_add_page() callback\n that drops the useless parameters.\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 10b7a7d912697afd681a0bcfced9e05543aded35\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Jul 2 16:44:32 2019 +0100\n\n iommu/io-pgtable-arm: Call -\u003etlb_flush_walk() and -\u003etlb_flush_leaf()\n \n Now that all IOMMU drivers using the io-pgtable API implement the\n -\u003etlb_flush_walk() and -\u003etlb_flush_leaf() callbacks, we can use them in\n the io-pgtable code instead of -\u003etlb_add_flush() immediately followed by\n -\u003etlb_sync().\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 05aed9412b0bd0d9a985d94010c42ff0a5c6cc29\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Jul 2 16:44:25 2019 +0100\n\n iommu/io-pgtable: Hook up -\u003etlb_flush_walk() and -\u003etlb_flush_leaf() in drivers\n \n Hook up -\u003etlb_flush_walk() and -\u003etlb_flush_leaf() in drivers using the\n io-pgtable API so that we can start making use of them in the page-table\n code. For now, they can just wrap the implementations of -\u003etlb_add_flush\n and -\u003etlb_sync pending future optimisation in each driver.\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 3445545b2248300319b6965208e77140c960c3fd\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Jul 2 16:44:16 2019 +0100\n\n iommu/io-pgtable: Introduce tlb_flush_walk() and tlb_flush_leaf()\n \n In preparation for deferring TLB flushes to iommu_tlb_sync(), introduce\n two new synchronous invalidation helpers to the io-pgtable API, which\n allow the unmap() code to force invalidation in cases where it cannot be\n deferred (e.g. when replacing a table with a block or when TLBI_ON_MAP\n is set).\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 56f8af5e9d38f120cba2c2adb0786fa2dbc901a4\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Jul 2 16:44:06 2019 +0100\n\n iommu: Pass struct iommu_iotlb_gather to -\u003eunmap() and -\u003eiotlb_sync()\n \n To allow IOMMU drivers to batch up TLB flushing operations and postpone\n them until -\u003eiotlb_sync() is called, extend the prototypes for the\n -\u003eunmap() and -\u003eiotlb_sync() IOMMU ops callbacks to take a pointer to\n the current iommu_iotlb_gather structure.\n \n All affected IOMMU drivers are updated, but there should be no\n functional change since the extra parameter is ignored for now.\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit f61be89fed56780750be2ee873d78f2bd8f38625\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Jul 29 09:17:19 2019 -0700\n\n iomap: zero newly allocated mapped blocks\n \n File systems like gfs2 don\u0027t support delayed allocations or unwritten\n extents and thus allocate normal mapped blocks to fill holes. To\n cover the case of such file systems allocating new blocks to fill holes\n also zero out mapped blocks with the new flag.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit 84be4599c1a66d3c29c8261166c9bae3879c17f2\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Jul 29 09:17:19 2019 -0700\n\n xfs: set IOMAP_F_NEW more carefully\n \n Don\u0027t set IOMAP_F_NEW if we COW over and existing allocated range, as\n these aren\u0027t strictly new allocations. This is required to be able to\n use IOMAP_F_NEW to zero newly allocated blocks, which is required for\n the iomap code to fully support file systems that don\u0027t do delayed\n allocations or use unwritten extents.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit 6f0af1def491c836bba4050fd5bcc9f74d617f6a\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Jul 29 09:17:18 2019 -0700\n\n iomap: warn on inline maps in iomap_writepage_map\n \n And inline mapping should never mark the page dirty and thus never end up\n in writepages. Add a check for that condition and warn if it happens.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit 4b45a4b5c1f5f52728cec1ce60a3a8f5bb9521b4\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Jul 29 09:17:18 2019 -0700\n\n iomap: add tracing for the address space operations\n \n Lift the xfs code for tracing address space operations to the iomap\n layer.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n [darrick: remove deletion of xfs writeback tracepoints]\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit a12c4b4602f1a1ebe17276415926268216b6fdf7\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Jul 29 09:17:17 2019 -0700\n\n iomap: copy the xfs writeback code to iomap.c\n \n Takes the xfs writeback code and copies it to iomap.c. A new structure\n with three methods is added as the abstraction from the generic\n writeback code to the file system. These methods are used to map\n blocks, submit an ioend, and cancel a page that encountered an error\n before it was added to an ioend.\n \n Note that we temporarily lose the writepage tracing, but that will\n be added back soon.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n [darrick: create the new iomap code, we\u0027ll delete the xfs code separately]\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit 31c20dede460435b2e4b96914e1ab5aec41628aa\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Mon Jul 29 09:17:17 2019 -0700\n\n list.h: add list_pop and list_pop_entry helpers\n \n We have a very common pattern where we want to delete the first entry\n from a list and return it as the properly typed container structure.\n \n Add two helpers to implement this behavior.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n Reviewed-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n Signed-off-by: Darrick J. Wong \u003cdarrick.wong@oracle.com\u003e\n\ncommit 7c4e91c0959bf42a86ba4887425fc944d0d28595\nAuthor: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\nDate: Wed Jun 12 15:05:34 2019 +0200\n\n s390/zcrypt: new sysfs attributes serialnr and mkvps\n \n This patch extends the sysfs interface with two new attributes for the\n CEX4, CEX5 and CEX6 crypto cards/queues in coprocessor (\u0027CCA\u0027) mode:\n /sys/devices/ap/cardxx/serialnr\n /sys/devices/ap/cardxx/xx.yyyy/mkvps\n \n The serialnr attribute is card based and shows the 8 character ASCII\n serial number string which should unique identify the card.\n \n The mkvps is queue based and displays 3 lines of information about the\n new, current and old master key register:\n AES NEW: \u003cnew_aes_mk_state\u003e \u003cnew_aes_mk_mkvp\u003e\n AES CUR: \u003ccur_aes_mk_state\u003e \u003ccur_aes_mk_mkvp\u003e\n AES OLD: \u003cold_aes_mk_state\u003e \u003cold_aes_mk_mkvp\u003e\n with\n \u003cnew_aes_mk_state\u003e: \u0027empty\u0027 or \u0027partial\u0027 or \u0027full\u0027\n \u003ccur_aes_mk_state\u003e: \u0027valid\u0027 or \u0027invalid\u0027\n \u003cold_aes_mk_state\u003e: \u0027valid\u0027 or \u0027invalid\u0027\n \u003cnew_aes_mk_mkvp\u003e, \u003ccur_aes_mk_mkvp\u003e, \u003cold_aes_mk_mkvp\u003e\n 8 byte hex string with leading 0x\n MKVP means Master Key Verification Pattern and is a folded hash over\n the key value. Only the states \u0027full\u0027 and \u0027valid\u0027 result in displaying\n a useful mkvp, otherwise a mkvp of all bytes zero is shown. If for any\n reason the FQ fails and the (cached) information is not available, the\n state \u0027-\u0027 will be shown with the mkvp value also \u0027-\u0027. The values shown\n here are the very same as the cca panel tools displays. As of now only\n the AES master keys states and verification patterns are shown. A CCA\n APQN also has similar master key registers for DES, RSA and ECC. So\n the content of this attribute may get extended.\n \n Reading the sysfs attribute automatically triggers an FQ CPRB to be\n sent to the queue as long as the queue is (soft-) online. For the\n serialnr attribute the queue with the default domain id is addressed\n (if available and valid). This is reasonable as it is assumed that\n this sysfs interface is not performance critical and on the other side\n a master key change should be visiable as soon as possible. When a\n queue is (soft-) offline however, the cached values are displayed. If\n no cached values are available, the serial number string will be empty\n and the mkvp lines will show state \u0027-\u0027 and mkvp value \u0027-\u0027.\n \n Signed-off-by: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\n Reviewed-by: Ingo Franzki \u003cifranzki@linux.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 94bbd34e51bc7809357e24e4011b169d321a8272\nAuthor: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\nDate: Tue Jun 11 14:10:47 2019 +0200\n\n s390/zcrypt: add base code for cca crypto card info support\n \n This patch widens the information held for cca crypto apqns.\n Currently the current and old master key verification pattern\n is used by the existing code. Now the new master key registers\n mkvp, the 8 byte serial number and state info about each master\n key register is part of the cca info cache.\n \n In a next step this information will be used to provide some\n additional attributes in sysfs for each CCA crypto adapter.\n \n Signed-off-by: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\n Reviewed-by: Ingo Franzki \u003cifranzki@linux.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit efc598e6c8a93491868ef7dc7c6fd28508898af2\nAuthor: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\nDate: Tue Jun 11 11:16:56 2019 +0200\n\n s390/zcrypt: move cca misc functions to new code file\n \n Rework of the pkey code. Moved all the cca generic code\n away from pkey_api.c into a new file zcrypt_ccamisc.c.\n This new file is now part of the zcrypt device driver\n and exports a bunch of cca functions to pkey and may\n be called from other kernel modules as well.\n \n The pkey ioctl API is unchanged.\n \n Signed-off-by: Harald Freudenberger \u003cfreude@linux.ibm.com\u003e\n Reviewed-by: Ingo Franzki \u003cifranzki@linux.ibm.com\u003e\n Signed-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\n\ncommit 8f6920ac0050c3156ce617d78a188dd1f40be93c\nAuthor: Sedat Dilek \u003csedat.dilek@credativ.de\u003e\nDate: Mon Jul 29 16:35:14 2019 +0300\n\n NFC: nxp-nci: Fix recommendation for NFC_NXP_NCI_I2C Kconfig\n \n This is a simple cleanup to the Kconfig help text as discussed in [1].\n \n [1] https://marc.info/?t\u003d155774435600001\u0026r\u003d1\u0026w\u003d2\n \n Suggested-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Suggested-by: Oleg Zhurakivskyy \u003coleg.zhurakivskyy@intel.com\u003e\n Signed-off-by: Sedat Dilek \u003csedat.dilek@credativ.de\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6f713f49b4b845d4c7a99b35e50cab5aed3dbbc0\nAuthor: Sedat Dilek \u003csedat.dilek@credativ.de\u003e\nDate: Mon Jul 29 16:35:13 2019 +0300\n\n NFC: nxp-nci: Clarify on supported chips\n \n This patch clarifies on the supported NXP NCI chips and families\n and lists PN547 and PN548 separately which are known as NPC100\n respectively NPC300.\n \n This helps to find informations and identify drivers on vendor\u0027s\n support websites.\n \n For details see the discussion in [1] and [2].\n \n [1] https://marc.info/?t\u003d155774435600001\u0026r\u003d1\u0026w\u003d2\n [2] https://patchwork.kernel.org/project/linux-wireless/list/?submitter\u003d33142\n \n Suggested-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Suggested-by: Oleg Zhurakivskyy \u003coleg.zhurakivskyy@intel.com\u003e\n Signed-off-by: Sedat Dilek \u003csedat.dilek@credativ.de\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Oleg Zhurakivskyy \u003coleg.zhurakivskyy@intel.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 826a99ebd4e24da32984cbe847ccc86d4c9a7437\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Jul 29 16:35:12 2019 +0300\n\n NFC: nxp-nci: Remove \u0027default n\u0027 for the core\n \n It seems contributors follow the style of Kconfig entries where explicit\n \u0027default n\u0027 is present. The default \u0027default\u0027 is \u0027n\u0027 already, thus, drop\n these lines from Kconfig to make it more clear.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Tested-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 3b11dc5712c30cf56b73e9f67a815e7c728c67be\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Jul 29 16:35:11 2019 +0300\n\n NFC: nxp-nci: Remove unused macro pr_fmt()\n \n The macro had never been used.\n \n The driver uses mostly the nfc_err(), which, with other macros in the family,\n is backed by corresponding dev_err(). pr_fmt() is not used for dev_err()\n macro. Moreover, there is no need to print the module name which is part of the\n device instance name anyway.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Tested-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 41bd9cee8a68b61743e096ac92a82de1c88f96be\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Jul 29 16:35:10 2019 +0300\n\n NFC: nxp-nci: Drop comma in terminator lines\n \n There is no need to have a comma after terminator entry\n in the arrays of IDs.\n \n This may prevent the misguided addition behind the terminator\n without compiler notice.\n \n Drop the comma in terminator lines for good.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Tested-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit da05208a0cb62d539470dd7f951f06cf0092e322\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Jul 29 16:35:09 2019 +0300\n\n NFC: nxp-nci: Drop of_match_ptr() use\n \n There is no need to guard OF device ID table with of_match_ptr().\n Otherwise we would get a defined but not used data.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Tested-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 52c2ea049142c75166456fed055883e22d6ba983\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Jul 29 16:35:08 2019 +0300\n\n NFC: nxp-nci: Constify acpi_device_id\n \n The content of acpi_device_id is not supposed to change at runtime.\n All functions working with acpi_device_id provided by \u003clinux/acpi.h\u003e\n work with const acpi_device_id. So mark the non-const structs as const.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Tested-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4f1cbf24fc523167e50f375d3a635431a96a4b03\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Jul 29 16:35:07 2019 +0300\n\n NFC: nxp-nci: Get rid of useless label\n \n Return directly in -\u003eprobe() since there no special cleaning is needed.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Tested-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ad0acfd69add44ab00abcce2240a7523f75bd315\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Jul 29 16:35:06 2019 +0300\n\n NFC: nxp-nci: Get rid of code duplication in -\u003eprobe()\n \n Since OF and ACPI case almost the same get rid of code duplication\n by moving gpiod_get() calls directly to -\u003eprobe().\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Tested-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 099d03f02dbffe1575d86063b2bf416502faa5e9\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Jul 29 16:35:05 2019 +0300\n\n NFC: nxp-nci: Add GPIO ACPI mapping table\n \n In order to unify GPIO resource request prepare gpiod_get_index()\n to behave correctly when there is no mapping provided by firmware.\n \n Here we add explicit mapping between _CRS GpioIo() resources and\n their names used in the driver.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Tested-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 43201767b44cbd873c60dbd2acd370147588cb18\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Jul 29 16:35:04 2019 +0300\n\n NFC: nxp-nci: Convert to use GPIO descriptor\n \n Since we got rid of platform data, the driver may use\n GPIO descriptor directly.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Tested-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 3b0b278312ba7d6c1eb8b2fb48d459fb7f341a20\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Jul 29 16:35:03 2019 +0300\n\n NFC: nxp-nci: Get rid of platform data\n \n Legacy platform data must go away. We are on the safe side here since\n there are no users of it in the kernel.\n \n If anyone by any odd reason needs it the GPIO lookup tables and\n built-in device properties at your service.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Tested-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1b14a37565d9e88b82b8a227690e5fbc0079b61b\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Mon Jul 29 16:35:02 2019 +0300\n\n NFC: nxp-nci: Add NXP1001 to the ACPI ID table\n \n It seems a lot of laptops are equipped with NXP NFC300 chip with\n the ACPI ID NXP1001 as per DSDT.\n \n Append it to the driver\u0027s ACPI ID table.\n \n Reported-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Tested-by: Sedat Dilek \u003csedat.dilek@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 18917d51472fe3b126a3a8f756c6b18085eb8130\nAuthor: Andrey Konovalov \u003candreyknvl@google.com\u003e\nDate: Mon Jul 29 16:35:01 2019 +0300\n\n NFC: fix attrs checks in netlink interface\n \n nfc_genl_deactivate_target() relies on the NFC_ATTR_TARGET_INDEX\n attribute being present, but doesn\u0027t check whether it is actually\n provided by the user. Same goes for nfc_genl_fw_download() and\n NFC_ATTR_FIRMWARE_NAME.\n \n This patch adds appropriate checks.\n \n Found with syzkaller.\n \n Signed-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 3e1f000ff74627c1adb99ee513f29ec2522ee309\nAuthor: Parav Pandit \u003cparav@mellanox.com\u003e\nDate: Tue Jul 23 10:31:17 2019 +0300\n\n IB/mlx5: Support per device q counters in switchdev mode\n \n When parent mlx5_core_dev is in switchdev mode, q_counters are not\n applicable to multiple non uplink vports.\n Hence, have make them limited to device level.\n \n While at it, correct __mlx5_ib_qp_set_counter() and\n __mlx5_ib_modify_qp() to use u16 set_id as defined by the device.\n \n Signed-off-by: Parav Pandit \u003cparav@mellanox.com\u003e\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Link: https://lore.kernel.org/r/20190723073117.7175-3-leon@kernel.org\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 5dcecbc96755793e9417a9135b290d16639ec6aa\nAuthor: Parav Pandit \u003cparav@mellanox.com\u003e\nDate: Tue Jul 23 10:31:16 2019 +0300\n\n IB/mlx5: Refactor code for counters allocation\n \n To support per device counters in switchdev mode (instead of\n per port counter), refactor query routines to work on mlx5_ib_counter\n structure instead of mlx5_ib_port structure.\n \n Signed-off-by: Parav Pandit \u003cparav@mellanox.com\u003e\n Reviewed-by: Daniel Jurgens \u003cdanielj@mellanox.com\u003e\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Link: https://lore.kernel.org/r/20190723073117.7175-2-leon@kernel.org\n Signed-off-by: Doug Ledford \u003cdledford@redhat.com\u003e\n\ncommit 35b7c80827e9352b88a2379f06cbb9aa3dd64c14\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Thu Jul 4 00:30:00 2019 +0800\n\n platform/x86: thinkpad_acpi: Use kmemdup rather than duplicating its implementation\n \n kmemdup is introduced to duplicate a region of memory in a neat way.\n Rather than kmalloc/kzalloc + memcpy, which the programmer needs to\n write the size twice (sometimes lead to mistakes), kmemdup improves\n readability, leads to smaller code and also reduce the chances of mistakes.\n Suggestion to use kmemdup rather than using kmalloc/kzalloc + memcpy.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 340f25ff1d4fd3bfe0d3e0769261fec8fb32ae9f\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Thu Jul 4 00:29:51 2019 +0800\n\n platform/x86: asus-wmi: Use kmemdup rather than duplicating its implementation\n \n kmemdup is introduced to duplicate a region of memory in a neat way.\n Rather than kmalloc/kzalloc + memcpy, which the programmer needs to\n write the size twice (sometimes lead to mistakes), kmemdup improves\n readability, leads to smaller code and also reduce the chances of mistakes.\n Suggestion to use kmemdup rather than using kmalloc/kzalloc + memcpy.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 4215484f8e33c8a4664aecb6db5ce2c544d3b908\nAuthor: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\nDate: Wed Jul 10 13:33:26 2019 +0800\n\n platform/x86: hp_accel: Add support for HP ZBook 17 G5\n \n HP ZBook 17 G5 needs a non-standard mapping, x_inverted.\n \n Signed-off-by: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit 08d80a4c90bdaa83680d1bdb58403fd8fe101885\nAuthor: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\nDate: Mon Jul 29 10:53:31 2019 +0800\n\n net: hns3: use dev_info() instead of pr_info()\n \n dev_info() is more appropriate for printing messages when driver\n initialization done, so switch to dev_info().\n \n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit dbba6da0c67ca99721b74760d3cc69df8a5a8230\nAuthor: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\nDate: Mon Jul 29 10:53:30 2019 +0800\n\n net: hns3: Add support for using order 1 pages with a 4K buffer\n \n Hardware supports 0.5K, 1K, 2K, 4K RX buffer size, the\n RX buffer can not be reused because the hns3_page_order\n return 0 when page size and RX buffer size are both 4096.\n \n So this patch changes the hns3_page_order to return 1 when\n RX buffer is greater than half of the page size and page size\n is less the 8192, and dev_alloc_pages has already been used\n to allocate the compound page for RX buffer.\n \n This patch also changes hnae3_* to hns3_* for page order\n and RX buffer size calculation because they are used in\n hns3 module.\n \n Signed-off-by: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 0812545487eca362126cc1ef6b7798ecac693629\nAuthor: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\nDate: Mon Jul 29 10:53:29 2019 +0800\n\n net: hns3: add interrupt affinity support for misc interrupt\n \n The misc interrupt is used to schedule the reset and mailbox\n subtask, and service_task delayed_work is used to do periodic\n management work each second.\n \n This patch sets the above three subtask\u0027s affinity using the\n misc interrupt\u0027 affinity.\n \n Also this patch setups a affinity notify for misc interrupt to\n allow user to change the above three subtask\u0027s affinity.\n \n Signed-off-by: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\n Signed-off-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7be1b9f3e99f6213d053d16ed2438126931d8351\nAuthor: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\nDate: Mon Jul 29 10:53:28 2019 +0800\n\n net: hns3: make hclge_service use delayed workqueue\n \n Use delayed work instead of using timers to trigger the\n hclge_serive.\n \n Simplify the code with one less middle function and in order\n to support misc irq affinity.\n \n Signed-off-by: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\n Reviewed-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1c822948fa600d2d2e7926640279b16d1f4a7423\nAuthor: Yonglong Liu \u003cliuyonglong@huawei.com\u003e\nDate: Mon Jul 29 10:53:27 2019 +0800\n\n net: hns3: add debug messages to identify eth down cause\n \n Some times just see the eth interface have been down/up via\n dmesg, but can not know why the eth down. So adds some debug\n messages to identify the cause for this.\n \n Signed-off-by: Yonglong Liu \u003cliuyonglong@huawei.com\u003e\n Signed-off-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 923713730db9795ac0658cc207935c11b2bbd705\nAuthor: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\nDate: Mon Jul 29 10:53:26 2019 +0800\n\n net: hns3: modify firmware version display format\n \n This patch modifies firmware version display format in\n hclge(vf)_cmd_init() and hns3_get_drvinfo(). Also, adds\n some optimizations for firmware version display format.\n \n Signed-off-by: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\n Signed-off-by: Peng Li \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d659f9f60f6a119654c65bfa58aa6fa762c0824d\nAuthor: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\nDate: Mon Jul 29 10:53:25 2019 +0800\n\n net: hns3: change GFP flag during lock period\n \n When allocating memory, the GFP_KERNEL cannot be used during the\n spin_lock period. This is because it may cause scheduling when holding\n spin_lock. This patch changes GFP flag to GFP_ATOMIC in this case.\n \n Fixes: dd74f815dd41 (\"net: hns3: Add support for rule add/delete for flow director\")\n Signed-off-by: Yufeng Mo \u003cmoyufeng@huawei.com\u003e\n Signed-off-by: lipeng 00277521 \u003clipeng321@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 8e9eee78316037f07fd391ddaef99e9c29b53b4b\nAuthor: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\nDate: Mon Jul 29 10:53:24 2019 +0800\n\n net: hns3: remove upgrade reset level when reset fail\n \n Currently, hclge_reset_err_handle() will assert a global reset\n when the failing count is smaller than MAX_RESET_FAIL_CNT, which\n will affect other running functions.\n \n So this patch removes this upgrading, and uses re-scheduling reset\n task to do it.\n \n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Reviewed-by: Yunsheng Lin \u003clinyunsheng@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit aa3253b8899d685811d159288f21cd2a2dbb5f0b\nAuthor: Guangbin Huang \u003chuangguangbin@huawei.com\u003e\nDate: Mon Jul 29 10:53:23 2019 +0800\n\n net: hns3: add a check for get_reset_level\n \n For some cases, ops-\u003eget_reset_level may not be implemented, so we\n should check whether it is NULL before calling get_reset_level.\n \n Signed-off-by: Guangbin Huang \u003chuangguangbin@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 44950d28cccf3049696e02d0adebb10e112cee24\nAuthor: Jian Shen \u003cshenjian15@huawei.com\u003e\nDate: Mon Jul 29 10:53:22 2019 +0800\n\n net: hns3: add reset checking before set channels\n \n hns3_set_channels() should check the resetting status firstly,\n since the device will reinitialize when resetting. If the\n reset has not completed, the hns3_set_channels() may access\n invalid memory.\n \n Signed-off-by: Jian Shen \u003cshenjian15@huawei.com\u003e\n Signed-off-by: Huazhong Tan \u003ctanhuazhong@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 60a4233a4952729089e4df152e730f8f4d0e82ce\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Jul 29 14:24:12 2019 +0100\n\n drm/i915: Flush the i915_vm_release before ggtt shutdown\n \n As the gen6_ppgtt may refer back to the GGTT for their page-directory\n slots, make sure those __i915_vm_release are completed prior to shutting\n down the GGTT.\n \n Fixes: b32fa8111563 (\"drm/i915/gtt: Defer address space cleanup to an RCU worker\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190729132412.23380-1-chris@chris-wilson.co.uk\n\ncommit 2ab45876756fb6c132ae801b0939e0474f84c426\nAuthor: John Crispin \u003cjohn@phrozen.org\u003e\nDate: Mon Jul 29 12:45:12 2019 +0200\n\n mac80211: add support for the ADDBA extension element\n \n HE allows peers to negotiate the aggregation fragmentation level to be used\n during transmission. The level can be 1-3. The Ext element is added behind\n the ADDBA request inside the action frame. The responder will then reply\n with the same level or a lower one if the requested one is not supported.\n This patch only handles the negotiation part as the ADDBA frames get passed\n to the ATH11k firmware, which does the rest of the magic for us aswell as\n generating the requests.\n \n Signed-off-by: Shashidhar Lakkavalli \u003cslakkavalli@datto.com\u003e\n Signed-off-by: John Crispin \u003cjohn@phrozen.org\u003e\n Link: https://lore.kernel.org/r/20190729104512.27615-1-john@phrozen.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 697f6c507c74991057eb6df3cfb46579ca136467\nAuthor: John Crispin \u003cjohn@phrozen.org\u003e\nDate: Mon Jul 29 12:23:42 2019 +0200\n\n mac80211: propagate HE operation info into bss_conf\n \n Upon a successful assoc a station shall store the content of the HE\n operation element inside bss_conf so that the driver can setup the\n hardware accordingly.\n \n Signed-off-by: Shashidhar Lakkavalli \u003cslakkavalli@datto.com\u003e\n Signed-off-by: John Crispin \u003cjohn@phrozen.org\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n Link: https://lore.kernel.org/r/20190729102342.8659-2-john@phrozen.org\n [use struct copy]\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 90d4962cfc87a6994a19db0d76e1fa214dd67267\nAuthor: John Crispin \u003cjohn@phrozen.org\u003e\nDate: Mon Jul 29 12:23:41 2019 +0200\n\n mac80211: fix ieee80211_he_oper_size() comment\n \n Johannes mentioned that the comment should not reference mac80211 as other\n subsystems might call the helper.\n \n Signed-off-by: John Crispin \u003cjohn@phrozen.org\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n Link: https://lore.kernel.org/r/20190729102342.8659-1-john@phrozen.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit d34990bbc25559fa1af5e23759c65a3951cbc956\nAuthor: Michael Vassernis \u003cmichael.vassernis@tandemg.com\u003e\nDate: Mon Jul 29 06:01:16 2019 +0000\n\n cfg80211: fix dfs channels remain DFS_AVAILABLE after ch_switch\n \n Depending on the regulatory domain, leaving a DFS channel requires\n a new CAC to be performed when returning back to that channel.\n If needed, update dfs states after a driver channel switch.\n \n Signed-off-by: Michael Vassernis \u003cmichael.vassernis@tandemg.com\u003e\n Link: https://lore.kernel.org/r/20190729060024.5660-1-michael.vassernis@tandemg.com\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 0afd425b1b64251f19b5d8d8b49bf56fefbc643f\nAuthor: Sergey Matyukevich \u003csergey.matyukevich.os@quantenna.com\u003e\nDate: Fri Jul 26 16:39:34 2019 +0000\n\n cfg80211: fix duplicated scan entries after channel switch\n \n When associated BSS completes channel switch procedure, its channel\n record needs to be updated. The existing mac80211 solution was\n extended to cfg80211 in commit 5dc8cdce1d72 (\"mac80211/cfg80211:\n update bss channel on channel switch\").\n \n However that solution still appears to be incomplete as it may lead\n to duplicated scan entries for associated BSS after channel switch.\n The root cause of the problem is as follows. Each BSS entry is\n included into the following data structures:\n - bss list rdev-\u003ebss_list\n - bss search tree rdev-\u003ebss_tree\n Updating BSS channel record without rebuilding bss_tree may break\n tree search since cmp_bss considers all of the following: channel,\n bssid, ssid. When BSS channel is updated, but its location in bss_tree\n is not updated, then subsequent search operations may fail to locate\n this BSS since they will be traversing bss_tree in wrong direction.\n As a result, for scan performed after associated BSS channel switch,\n cfg80211_bss_update may add the second entry for the same BSS to both\n bss_list and bss_tree, rather then update the existing one.\n \n To summarize, if BSS channel needs to be updated, then bss_tree should\n be rebuilt in order to put updated BSS entry into a proper location.\n \n This commit suggests the following straightforward solution:\n - if new entry has been already created for BSS after channel switch,\n then use its IEs to update known BSS entry and then remove new\n entry completely\n - use rb_erase/rb_insert_bss reinstall updated BSS in bss_tree\n - for nontransmit BSS entry, the whole transmit BSS hierarchy\n is updated\n \n Signed-off-by: Sergey Matyukevich \u003csergey.matyukevich.os@quantenna.com\u003e\n Link: https://lore.kernel.org/r/20190726163922.27509-3-sergey.matyukevich.os@quantenna.com\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 3ab8227d3e7d1d2bf1829675d3197e3cb600e9f6\nAuthor: Sergey Matyukevich \u003csergey.matyukevich.os@quantenna.com\u003e\nDate: Fri Jul 26 16:39:32 2019 +0000\n\n cfg80211: refactor cfg80211_bss_update\n \n This patch implements minor refactoring for cfg80211_bss_update function.\n Code path for updating known BSS is extracted into dedicated\n cfg80211_update_known_bss function.\n \n Signed-off-by: Sergey Matyukevich \u003csergey.matyukevich.os@quantenna.com\u003e\n Link: https://lore.kernel.org/r/20190726163922.27509-2-sergey.matyukevich.os@quantenna.com\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 0ed0eb0171d7c19310d9f7b06a6dee82a4fb8ab6\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Mon Jun 17 14:00:56 2019 +0200\n\n soc: renesas: rcar-sysc: Use [] to denote a flexible array member\n \n Flexible array members should be denoted using [] instead of [0], else\n gcc will not warn when they are no longer at the end of the structure.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n\ncommit 8cb7ec14188649cf2151464050413e2814fd7cf1\nAuthor: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\nDate: Fri Jun 14 12:53:30 2019 +0100\n\n dt-bindings: can: rcar_can: Complete documentation for RZ/G2[EM]\n \n Add missing RZ/G2[EM] CANFD clock specific documentation.\n \n Fixes: 868b7c0f43e61f22 (\"dt-bindings: can: rcar_can: Add r8a774a1 support\")\n Fixes: d703a52eb1ebeeba (\"dt-bindings: can: rcar_can: Add r8a774c0 support\")\n Signed-off-by: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\n Reviewed-by: Chris Paterson \u003cChris.Paterson2@renesas.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit a7aa356e2c9ad1e874460ac5ee499ed1690e80cb\nAuthor: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\nDate: Fri Jun 14 12:53:29 2019 +0100\n\n dt-bindings: can: rcar_canfd: document r8a774a1 support\n \n Document the support for rcar_canfd on R8A774A1 SoC devices.\n \n Signed-off-by: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\n Reviewed-by: Chris Paterson \u003cChris.Paterson2@renesas.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit a44efeaa0bbf699f7ad397f5130e2834d52f621e\nAuthor: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\nDate: Fri Jul 5 14:39:38 2019 +0100\n\n arm64: dts: renesas: r8a774a1: Add SSIU support for sound\n \n Add SSIU support to the SoC DT as the sound driver supports\n it now, and also since the sound driver can now handle\n BUSIF0-7 via SSIU remove the no longer needed \"rxu\" and \"txu\"\n properties.\n \n Based on similar work from Kuninori Morimoto and Simon Horman in commits\n 8d14bfa074db (\"arm64: dts: renesas: r8a7796: add SSIU support for\n sound\") and 10bd03fa896e (\"arm64: dts: renesas: r8a7796: remove BUSIF0\n settings from rcar_sound,ssi\").\n \n Signed-off-by: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit da245a5066aa4c96c153a4ede3dab68d6b2306e8\nAuthor: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\nDate: Fri Jul 5 14:39:37 2019 +0100\n\n arm64: dts: renesas: r8a774a1: Use extended audio dmac registers\n \n Basic audio dmac register only supports busif from 0 to 3,\n in order to use busif4 ~ busif7 extended audio dmac registers\n need to be used.\n \n Based on similar work from Jiada Wang in commit 7a516e49d975 (\"arm64:\n dts: renesas: use extended audio dmac register\").\n \n Signed-off-by: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 176f936a1e887a53d78f05248dc3c9bbd2c7f4c8\nAuthor: Biju Das \u003cbiju.das@bp.renesas.com\u003e\nDate: Fri Jul 5 11:15:20 2019 +0100\n\n arm64: dts: renesas: hihope-common: Add WLAN support\n \n This patch enables WLAN support for the HiHope RZ/G2[MN] boards.\n \n Signed-off-by: Biju Das \u003cbiju.das@bp.renesas.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit d112c20925443b4a8876b101e5b7b13ef105057e\nAuthor: Biju Das \u003cbiju.das@bp.renesas.com\u003e\nDate: Fri Jul 5 11:15:19 2019 +0100\n\n arm64: dts: renesas: hihope-common: Add BT support\n \n This patch enables BT support for the HiHope RZ/G2[MN] boards.\n \n Signed-off-by: Biju Das \u003cbiju.das@bp.renesas.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit cd3e43be14189542d8934faa94eff0392da32536\nAuthor: Biju Das \u003cbiju.das@bp.renesas.com\u003e\nDate: Fri Jul 5 11:15:18 2019 +0100\n\n arm64: dts: renesas: hihope-common: Add PCA9654 I/O expander\n \n Enable PCA9654 GPIO expander, so that we can configure its GPIOs later.\n \n Signed-off-by: Biju Das \u003cbiju.das@bp.renesas.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 816c5248df0fdcee33e56444232394d597a1894b\nAuthor: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\nDate: Fri Jun 14 12:53:34 2019 +0100\n\n arm64: dts: renesas: hihope-rzg2-ex: Enable CAN interfaces\n \n This patch enables both CAN0 and CAN1, both exposed via\n connectors found on the expansion board.\n \n Signed-off-by: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\n Reviewed-by: Chris Paterson \u003cChris.Paterson2@renesas.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 5b971c71dd64c08c73c37f754f60a7d777ea70e9\nAuthor: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\nDate: Fri Jun 14 12:53:33 2019 +0100\n\n arm64: dts: renesas: r8a774a1: Add CANFD support\n \n Add CANFD support to the SoC specific dtsi.\n \n Signed-off-by: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\n Reviewed-by: Chris Paterson \u003cChris.Paterson2@renesas.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 0a930f64a1cc36a34797a9e54f3040ad1d5833f2\nAuthor: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\nDate: Fri Jun 14 12:53:32 2019 +0100\n\n arm64: dts: renesas: r8a774a1: Add missing assigned-clocks for CAN[01]\n \n Define \"assigned-clocks\" and \"assigned-clock-rates\" properties\n for CAN[01] DT nodes, as required by the dt-bindings.\n \n Fixes: eccc40002972c424 (\"arm64: dts: renesas: r8a774a1: Add clkp2 clock to CAN nodes\")\n Signed-off-by: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\n Reviewed-by: Chris Paterson \u003cChris.Paterson2@renesas.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit e8efd2a8e20a9d7a7bd701950254a0ae04a8ce27\nAuthor: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\nDate: Fri Jun 14 12:53:31 2019 +0100\n\n arm64: dts: renesas: r8a774c0: Add missing assigned-clocks for CAN[01]\n \n Define \"assigned-clocks\" and \"assigned-clock-rates\" properties\n for CAN[01] DT nodes, as required by the dt-bindings.\n \n Fixes: 036bc85c1d06ef0a (\"arm64: dts: renesas: r8a774c0: Add clkp2 clock to CAN nodes\")\n Signed-off-by: Fabrizio Castro \u003cfabrizio.castro@bp.renesas.com\u003e\n Reviewed-by: Chris Paterson \u003cChris.Paterson2@renesas.com\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 56d651e890f3befd616b6962a862f5ffa1a514fa\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Tue Jun 18 05:18:16 2019 +0900\n\n arm64: dts: renesas: r8a77995: Fix register range of display node\n \n Since the R8A77995 SoC uses DU{0,1}, the range from the base address to\n the 0x4000 address is used.\n This patch fixed it.\n \n Fixes: 18f1a773e3f9e6d1 (\"arm64: dts: renesas: r8a77995: add DU support\")\n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Reviewed-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 3ed1db9071fde0ba9c4ce22a9b404887c0dbe909\nAuthor: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\nDate: Mon Jun 24 12:52:24 2019 +0200\n\n arm64: dts: renesas: r8a77995: Add cpg reset for DU\n \n Add CPG reset properties to DU node of D3 (r8a77995) SoC.\n \n According to Laurent Pinchart, R-Car Gen3 reset is handled at the group\n level so specifying one reset entry per group is sufficient.\n \n This patch was inspired by a patch in the BSP by\n Takeshi Kihara \u003ctakeshi.kihara.df@renesas.com\u003e.\n \n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Signed-off-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 4193a39240fbeda2ee35232bd0a1deedd41d31aa\nAuthor: Takeshi Kihara \u003ctakeshi.kihara.df@renesas.com\u003e\nDate: Mon Jun 24 12:52:23 2019 +0200\n\n arm64: dts: renesas: r8a77990: Add cpg reset for DU\n \n Add CPG reset properties to DU node of E3 (r8a77990) SoC.\n \n According to Laurent Pinchart, R-Car Gen3 reset is handled at the group\n level so specifying one reset entry per group is sufficient.\n \n Signed-off-by: Takeshi Kihara \u003ctakeshi.kihara.df@renesas.com\u003e\n Signed-off-by: Yoshihiro Kaneko \u003cykaneko0929@gmail.com\u003e\n Signed-off-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit e96c7612315a1183e12d5b6ebd523a3a93617510\nAuthor: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\nDate: Thu Jul 25 18:42:38 2019 +0200\n\n clk: meson: remove clk input helper\n \n The clk input function which allows clock controllers to register a bypass\n clock from a clock producer is no longer needed anymore since meson clock\n controllers have migrated to a new parent allocation method.\n \n Signed-off-by: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n\ncommit b11cfaba5b4d6e287540a3d64c403e5b26dd2728\nAuthor: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\nDate: Thu Jul 25 18:42:37 2019 +0200\n\n clk: meson: remove ee input bypass clocks\n \n During probe, bypass clocks (i.e. ee-in-xtal) are made from device-tree\n inputs to provide input clocks which can be access through global name.\n The cons of this method are the duplicated clocks, means more string\n comparison.\n \n Specify parent directly with device-tree clock name.\n \n Remove the bypass clock registration from the ee probe function.\n \n Signed-off-by: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n\ncommit 3a36044e7f3909c7ddb7ddfc727ab8104a563439\nAuthor: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\nDate: Thu Jul 25 18:42:36 2019 +0200\n\n clk: meson: clk-regmap: migrate to new parent description method\n \n This clock controller use the string comparison method to describe parent\n relation between the clocks, which is not optimized.\n \n Migrate to the new way by using .parent_hws where possible (ie. when\n all clocks are local to the controller) and use .parent_data otherwise.\n \n Signed-off-by: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n\ncommit 4b5b85c0e6505c50d4a986f75effe5b88d923737\nAuthor: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\nDate: Thu Jul 25 18:42:35 2019 +0200\n\n clk: meson: meson8b: migrate to the new parent description method\n \n This clock controller use the string comparison method to describe parent\n relation between the clocks, which is not optimized.\n \n Migrate to the new way by using .parent_hws where possible (ie. when\n all clocks are local to the controller) and use .parent_data otherwise.\n \n Signed-off-by: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Tested-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n\ncommit cc132d113dc589d8449fe2b53043b0f17029acac\nAuthor: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\nDate: Thu Jul 25 18:42:34 2019 +0200\n\n clk: meson: axg: migrate to the new parent description method\n \n This clock controller use the string comparison method to describe parent\n relation between the clocks, which is not optimized.\n \n Migrate to the new way by using .parent_hws where possible (ie. when\n all clocks are local to the controller) and use .parent_data otherwise.\n \n Signed-off-by: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n\ncommit 0dea3f35996f4571eacbf6cac889d57bfd20822a\nAuthor: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\nDate: Thu Jul 25 18:42:33 2019 +0200\n\n clk: meson: gxbb: migrate to the new parent description method\n \n This clock controller use the string comparison method to describe parent\n relation between the clocks, which is not optimized.\n \n Migrate to the new way by using .parent_hws where possible (ie. when\n all clocks are local to the controller) and use .parent_data otherwise.\n \n Signed-off-by: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n\ncommit 25e682a02d91e2d09f47c089a4b42bba726106b9\nAuthor: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\nDate: Thu Jul 25 18:42:31 2019 +0200\n\n clk: meson: g12a: migrate to the new parent description method\n \n This clock controller use the string comparison method to describe parent\n relation between the clocks, which is not optimized.\n \n Migrate to the new way by using .parent_hws where possible (ie. when\n all clocks are local to the controller) and use .parent_data otherwise.\n \n Signed-off-by: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n\ncommit 072a043f5a2e02441002fff34e3885e6026714eb\nAuthor: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\nDate: Thu Jul 25 18:41:26 2019 +0200\n\n clk: meson: remove ao input bypass clocks\n \n During probe, bypass clocks (i.e. ao-in-xtal) are made from device-tree\n inputs to provide input clocks which can be access through global name.\n The cons of this method are the duplicated clocks, means more string\n comparison.\n \n Specify parent directly with device-tree clock name.\n \n Function to regiter bypass clocks is removed.\n \n Input parameters from meson aoclk data structure are deprecated and then\n deleted since all aoclk files are migrated.\n \n Signed-off-by: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n\ncommit b90ec1e344a2dd4c1afebd75c1ce05afaf9e116b\nAuthor: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\nDate: Thu Jul 25 18:41:25 2019 +0200\n\n clk: meson: axg-aoclk: migrate to the new parent description method\n \n This clock controller use the string comparison method to describe parent\n relation between the clocks, which is not optimized.\n \n Migrate to the new way by using .parent_hws where possible (when parent\n clocks are localy declared in the controller) and use .parent_data\n otherwise.\n \n Remove clk input helper and all bypass clocks (declared in probe function)\n which are no longer used since we are able to use device-tree clock name\n directly.\n \n Signed-off-by: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n\ncommit 6e2bfc352e7a3a9b22f13c36627545d5f4caf3e9\nAuthor: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\nDate: Thu Jul 25 18:41:24 2019 +0200\n\n clk: meson: gxbb-aoclk: migrate to the new parent description method\n \n This clock controller use the string comparison method to describe parent\n relation between the clocks, which is not optimized.\n \n Migrate to the new way by using .parent_hws where possible (when parent\n clocks are localy declared in the controller) and use .parent_data\n otherwise.\n \n Remove clk input helper and all bypass clocks (declared in probe function)\n which are no longer used since we are able to use device-tree clock name\n directly.\n \n Signed-off-by: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n\ncommit ba626081107dceacff554e5ca2efc7cea7326b6e\nAuthor: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\nDate: Thu Jul 25 18:41:23 2019 +0200\n\n clk: meson: g12a-aoclk: migrate to the new parent description method\n \n This clock controller use the string comparison method to describe parent\n relation between the clocks, which is not optimized.\n \n Migrate to the new way by using .parent_hws where possible (when parent\n clocks are localy declared in the controller) and use .parent_data\n otherwise.\n \n Remove clk input helper and all bypass clocks (declared in probe function)\n which are no longer used since we are able to use device-tree clock name\n directly.\n \n Signed-off-by: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n\ncommit 282420eed23f237963f2033b2f2bedb90fbcc5e1\nAuthor: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\nDate: Thu Jul 25 18:40:23 2019 +0200\n\n clk: meson: axg-audio: migrate to the new parent description method\n \n This clock controller use the string comparison method to describe parent\n relation between the clocks, which is not optimized. A recent patch [0]\n allows parents to be specified without string names or with device-tree\n clock name by using a new assignment structure.\n \n Migrate to the new way by using .parent_hws where possible (when parent\n clocks are localy declared in the controller) and use .parent_data\n otherwise.\n \n Remove clk input helper and all bypass clocks (declared in probe function)\n which are no longer used since we are able to use device-tree clock name\n directly.\n \n [0] commit fc0c209c147f (\"clk: Allow parents to be specified without string names\")\n \n Signed-off-by: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\n [jbrunet@baylibre.com: remove CLK_SET_RATE_PARENT from mst muxes]\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n\ncommit 76c5399fe2e4d7a5013280629de05c8ab6301311\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Jul 29 09:59:44 2019 +0100\n\n drm/i915/selftests: Careful not to flush hang_fini on error setups\n \n Smatch spotted that we test at the start of hang_fini for a valid (h-\u003egt\n is only set after a request is created) but then used it regardless\n later on.\n \n v2: Alternatively, we do not need to check as we now always prime h-\u003egt\n in hang_init()\n \n References: cb823ed9915b (\"drm/i915/gt: Use intel_gt as the primary object for handling resets\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190729085944.2179-1-chris@chris-wilson.co.uk\n\ncommit f474808acb3c4b30552d9c59b181244e0300d218\nAuthor: Johannes Berg \u003cjohannes@sipsolutions.net\u003e\nDate: Mon Jul 29 09:14:22 2019 +0200\n\n ALSA: aoa: onyx: always initialize register read value\n \n A lot of places in the driver use onyx_read_register() without\n checking the return value, and it\u0027s been working OK for ~10 years\n or so, so probably never fails ... Rather than trying to check the\n return value everywhere, which would be relatively intrusive, at\n least make sure we don\u0027t use an uninitialized value.\n \n Fixes: f3d9478b2ce4 (\"[ALSA] snd-aoa: add snd-aoa\")\n Reported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Johannes Berg \u003cjohannes@sipsolutions.net\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 57dbd0e4b97d3bc9f257159a8853badff2dfb0d1\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 22:07:57 2019 +0300\n\n dmaengine: stm32-mdma: Switch to use device_property_count_u32()\n \n Use use device_property_count_u32() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190723190757.67351-2-andriy.shevchenko@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 2cb114c4fac7a9c3e8fd8c39ceac1e6ca030b412\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 22:07:56 2019 +0300\n\n dmaengine: stm32-dmamux: Switch to use device_property_count_u32()\n \n Use use device_property_count_u32() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190723190757.67351-1-andriy.shevchenko@linux.intel.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit ffbb569b9cee1fa0277a48d7925dbed1356dbb32\nAuthor: Simon Horman \u003chorms+renesas@verge.net.au\u003e\nDate: Wed Jul 24 13:49:46 2019 +0200\n\n dt-bindings: dmaengine: shdma: Rename bindings documentation file\n \n Rename the bindings documentation file for shdma\n from shdma.txt to renesas,shdma.txt.\n \n This is part of an ongoing effort to name bindings documentation files for\n Renesas IP blocks consistently, in line with the compat strings they\n document.\n \n Signed-off-by: Simon Horman \u003chorms+renesas@verge.net.au\u003e\n Link: https://lore.kernel.org/r/20190724114946.14021-1-horms+renesas@verge.net.au\n [vkoul: add subsystem name]\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit edd14218bd663e4dd2f1c8f4eed84f92ada93c9d\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Sat Jul 20 11:26:07 2019 +0200\n\n dt-bindings: dmaengine: Convert Allwinner A31 and A64 DMA to a schema\n \n The newer Allwinner SoCs have a DMA controller supported in Linux, with a\n matching Device Tree binding.\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for that controller over to a YAML schemas.\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190720092607.31095-3-maxime.ripard@bootlin.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 545a29c811f6a07433c2725a94243f56ae28a81f\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Sat Jul 20 11:26:06 2019 +0200\n\n dt-bindings: dmaengine: Convert Allwinner A10 DMA to a schema\n \n The older Allwinner SoCs have a DMA controller supported in Linux, with a\n matching Device Tree binding.\n \n Now that we have the DT validation in place, let\u0027s convert the device tree\n bindings for that controller over to a YAML schemas.\n \n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190720092607.31095-2-maxime.ripard@bootlin.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit b37e3534ac4275c1f0d0e08d1b6f672f6cc740c3\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Sat Jul 20 11:26:05 2019 +0200\n\n dt-bindings: dmaengine: Add YAML schemas for the generic DMA bindings\n \n The DMA controllers and consumers have a bunch of generic properties that\n are needed in a device tree. Add a YAML schemas for those.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n \n Link: https://lore.kernel.org/r/20190720092607.31095-1-maxime.ripard@bootlin.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit aa3c6ce4eab8fb0e967954be1ba1cad3b715f63b\nAuthor: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\nDate: Tue Jul 16 11:26:55 2019 +0300\n\n dmaengine: ti: edma: Support for polled (memcpy) completion\n \n When a DMA client driver does not set the DMA_PREP_INTERRUPT because it\n does not want to use interrupts for DMA completion or because it can not\n rely on DMA interrupts due to executing the memcpy when interrupts are\n disabled it will poll the status of the transfer.\n \n Since we can not tell from any EDMA register that the transfer is\n completed, we can only know that the paRAM set has been sent to TPTC for\n processing we need to check the residue of the transfer, if it is 0 then\n the transfer is completed.\n \n Signed-off-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Link: https://lore.kernel.org/r/20190716082655.1620-4-peter.ujfalusi@ti.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 097ffdc75259139ba157b7f924cfeb0d6b00559e\nAuthor: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\nDate: Tue Jul 16 11:26:54 2019 +0300\n\n dmaengine: ti: edma: Correct the residue calculation (fix for memcpy)\n \n For memcpy we never stored the start address of the transfer for the pset\n which rendered the memcpy residue calculation completely broken.\n \n In the edma_residue() function we also need to to some correction for the\n calculations:\n Instead waiting for all EDMA channels to be idle (in a busy system it can\n take few iteration to hit a point when all queues are idle) wait for the\n event pending on the given channel (SH_ER for hw synchronized channels,\n SH_ESR for manually triggered channels).\n \n If the position returned by EMDA is 0 it implies that the last paRAM set\n has been consumed and we are at the closing dummy set, thus we can conclude\n that the transfer is completed and we can return 0 as residue.\n \n Signed-off-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n [vkoul: fixed typo in commit log]\n Link: https://lore.kernel.org/r/20190716082655.1620-3-peter.ujfalusi@ti.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit e96b1f64ee2885acb8fb26325eb9743ad6c64696\nAuthor: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\nDate: Tue Jul 16 11:26:53 2019 +0300\n\n dmaengine: ti: edma: Clean up the 2x32bit array register accesses\n \n Introduce defines for getting the array index and the bit number within the\n 64bit array register pairs.\n \n Signed-off-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Link: https://lore.kernel.org/r/20190716082655.1620-2-peter.ujfalusi@ti.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 4689d35c765c696bdf0535486a990038b242a26b\nAuthor: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\nDate: Tue Jul 16 11:24:59 2019 +0300\n\n dmaengine: ti: omap-dma: Improved memcpy polling support\n \n When a DMA client driver does not set the DMA_PREP_INTERRUPT because it\n does not want to use interrupts for DMA completion or because it can not\n rely on DMA interrupts due to executing the memcpy when interrupts are\n disabled it will poll the status of the transfer.\n \n If the interrupts are enabled then the cookie will be set completed in the\n interrupt handler so only check in HW completion when the polling is really\n needed.\n \n Signed-off-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Link: https://lore.kernel.org/r/20190716082459.1222-3-peter.ujfalusi@ti.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit aac8670369dc017609b8e8870975641ad3143448\nAuthor: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\nDate: Tue Jul 16 11:24:58 2019 +0300\n\n dmaengine: ti: omap-dma: Readability cleanup in omap_dma_tx_status()\n \n The tx_status is most likely going to be asked for the current transfer, so\n check that first then try to fall back to lookup of non started transfers.\n \n In this way the code is a bit more readable and in most cases we will avoid\n to run vchan_find_desc() all the time.\n \n Signed-off-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Link: https://lore.kernel.org/r/20190716082459.1222-2-peter.ujfalusi@ti.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 21be848ebc5f3cb714b23c2528fbe73f69073c32\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Fri Jul 26 15:18:20 2019 +0300\n\n extcon: axp288: Use for_each_set_bit() in axp288_extcon_log_rsi()\n \n This simplifies and standardizes axp288_extcon_log_rsi()\n by using for_each_set_bit() library function.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n\ncommit d72e3dc7915fc6c54645772c13f4afc0e676c7e2\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Fri Jul 26 15:18:19 2019 +0300\n\n extcon: axp288: Add missed error check\n \n It seems from the very beginning the error check has been missed\n in axp288_extcon_log_rsi(). Add it here.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n\ncommit 8a1d0f9cacc997bedc017056a94f35dc823394ed\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Mon Jul 22 09:26:22 2019 -0700\n\n fs-verity: add data verification hooks for -\u003ereadpages()\n \n Add functions that verify data pages that have been read from a\n fs-verity file, against that file\u0027s Merkle tree. These will be called\n from filesystems\u0027 -\u003ereadpage() and -\u003ereadpages() methods.\n \n Since data verification can block, a workqueue is provided for these\n methods to enqueue verification work from their bio completion callback.\n \n See the \"Verifying data\" section of\n Documentation/filesystems/fsverity.rst for more information.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Reviewed-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit c1d9b584e2cf3f0562d8fcf34574c044d17853a1\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Mon Jul 22 09:26:22 2019 -0700\n\n fs-verity: add the hook for file -\u003esetattr()\n \n Add a function fsverity_prepare_setattr() which filesystems that support\n fs-verity must call to deny truncates of verity files.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Reviewed-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit fd2d1acfcadfe2e42567afaec5e989b38061a7d2\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Mon Jul 22 09:26:22 2019 -0700\n\n fs-verity: add the hook for file -\u003eopen()\n \n Add the fsverity_file_open() function, which prepares an fs-verity file\n to be read from. If not already done, it loads the fs-verity descriptor\n from the filesystem and sets up an fsverity_info structure for the inode\n which describes the Merkle tree and contains the file measurement. It\n also denies all attempts to open verity files for writing.\n \n This commit also begins the include/linux/fsverity.h header, which\n declares the interface between fs/verity/ and filesystems.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Reviewed-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 5585f2af737ae3d7454cb0ae77b995cd3ac7e43c\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Mon Jul 22 09:26:21 2019 -0700\n\n fs-verity: add inode and superblock fields\n \n Analogous to fs/crypto/, add fields to the VFS inode and superblock for\n use by the fs/verity/ support layer:\n \n - -\u003es_vop: points to the fsverity_operations if the filesystem supports\n fs-verity, otherwise is NULL.\n \n - -\u003ei_verity_info: points to cached fs-verity information for the inode\n after someone opens it, otherwise is NULL.\n \n - S_VERITY: bit in -\u003ei_flags that identifies verity inodes, even when\n they haven\u0027t been opened yet and thus still have NULL -\u003ei_verity_info.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Reviewed-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 671e67b47e9fffd12c8f69eda853a202cb5b3fc5\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Mon Jul 22 09:26:21 2019 -0700\n\n fs-verity: add Kconfig and the helper functions for hashing\n \n Add the beginnings of the fs/verity/ support layer, including the\n Kconfig option and various helper functions for hashing. To start, only\n SHA-256 is supported, but other hash algorithms can easily be added.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Reviewed-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit fe9918d3b228b3e8c726849d1486933f46b9069e\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Mon Jul 22 09:26:21 2019 -0700\n\n fs: uapi: define verity bit for FS_IOC_GETFLAGS\n \n Add FS_VERITY_FL to the flags for FS_IOC_GETFLAGS, so that applications\n can easily determine whether a file is a verity file at the same time as\n they\u0027re checking other file flags. This flag will be gettable only;\n FS_IOC_SETFLAGS won\u0027t allow setting it, since an ioctl must be used\n instead to provide more parameters.\n \n This flag matches the on-disk bit that was already allocated for ext4.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Reviewed-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 085771ec14b9bdb843fe9283d4703ced395d1b0b\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Mon Jul 22 09:26:20 2019 -0700\n\n fs-verity: add UAPI header\n \n Add the UAPI header for fs-verity, including two ioctls:\n \n - FS_IOC_ENABLE_VERITY\n - FS_IOC_MEASURE_VERITY\n \n These ioctls are documented in the \"User API\" section of\n Documentation/filesystems/fsverity.rst.\n \n Examples of using these ioctls can be found in fsverity-utils\n (https://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/fsverity-utils.git).\n \n I\u0027ve also written xfstests that test these ioctls\n (https://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/xfstests-dev.git/log/?h\u003dfsverity).\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Reviewed-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit c9570b2ced0305ad2e69ff57012641c584bfa2a9\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Mon Jul 22 09:26:20 2019 -0700\n\n fs-verity: add MAINTAINERS file entry\n \n fs-verity will be jointly maintained by Eric Biggers and Theodore Ts\u0027o.\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Reviewed-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 6ff2deb2e8f6f2e23959724bab03ac60a050710b\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Mon Jul 22 09:26:20 2019 -0700\n\n fs-verity: add a documentation file\n \n Add a documentation file for fs-verity, covering:\n \n - Introduction\n - Use cases\n - User API\n - FS_IOC_ENABLE_VERITY\n - FS_IOC_MEASURE_VERITY\n - FS_IOC_GETFLAGS\n - Accessing verity files\n - File measurement computation\n - Merkle tree\n - fs-verity descriptor\n - Built-in signature verification\n - Filesystem support\n - ext4\n - f2fs\n - Implementation details\n - Verifying data\n - Pagecache\n - Block device based filesystems\n - Userspace utility\n - Tests\n - FAQ\n \n Reviewed-by: Theodore Ts\u0027o \u003ctytso@mit.edu\u003e\n Reviewed-by: Jaegeuk Kim \u003cjaegeuk@kernel.org\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n\ncommit 0ce242ad2ec17ddab1e3e3185b33c78f70df080a\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Wed Jul 24 10:38:28 2019 +0200\n\n pinctrl: qcom: Pass irqchip when adding gpiochip\n \n We need to convert all old gpio irqchips to pass the irqchip\n setup along when adding the gpio_chip.\n \n For chained irqchips this is a pretty straight-forward\n conversion.\n \n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n Reviewed-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Co-developed-by: Lina Iyer \u003cilina@codeaurora.org\u003e\n Link: https://lore.kernel.org/r/20190724083828.7496-1-linus.walleij@linaro.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit b3a2b136011ef567f2b9cf55f559f38816c8a64d\nAuthor: Luca Ceresoli \u003cluca@lucaceresoli.net\u003e\nDate: Fri Jul 5 16:30:43 2019 +0200\n\n docs/pinctrl: fix compile errors in example code\n \n The code in the example does not build for a few trivial errors: type\n mismatch in callback, missing semicolon. Fix them to help newcomers using\n the example as a starting point.\n \n Signed-off-by: Luca Ceresoli \u003cluca@lucaceresoli.net\u003e\n Link: https://lore.kernel.org/r/20190705143043.1929-1-luca@lucaceresoli.net\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit a299726da44fd679ce805aa80d7d6b559bac9874\nAuthor: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\nDate: Wed Jul 3 10:46:01 2019 +0200\n\n gpio: max77650: add MODULE_ALIAS()\n \n Define a MODULE_ALIAS() in the gpio sub-driver for max77650 so that\n the appropriate module gets loaded together with the core mfd driver.\n \n Signed-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n Link: https://lore.kernel.org/r/20190703084601.9276-1-brgl@bgdev.pl\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit f626d6dfb709852595e6fb24710819f9d21bc5ae\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Wed Jul 17 09:10:01 2019 +0200\n\n gpio: of: Break out OF-only code\n \n The core gpiolib should not contain any OF/device tree-only\n code. Try to break out the main part of it and push it down\n into the optional gpiolib-of.c part of the library.\n \n Create a local gpiolib-of.h header and move stuff around a\n bit to get a clean cut.\n \n Reviewed-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190717071001.3858-1-linus.walleij@linaro.org\n\ncommit 9e4d5c1be21f0c00e747e92186784f3298309b3e\nAuthor: Geoffrey D. Bennett \u003cg@b4.vu\u003e\nDate: Mon Jul 29 00:42:45 2019 +0930\n\n ALSA: usb-audio: Scarlett Gen 2 mixer interface\n \n Add mixer quirk for the Focusrite Scarlett 6i6, 18i8, and 18i20 Gen 2\n audio interfaces. Although the interfaces are USB compliant,\n additional input/output level controls and hardware routing/mixing\n functionality are available using proprietary USB requests.\n \n Signed-off-by: Geoffrey D. Bennett \u003cg@b4.vu\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit f8650b8ff6fd96075391e72bb3b6d0a99f27829d\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Sat Jun 29 13:33:06 2019 +0100\n\n gpio: bd70528: remove redundant assignment to variable ret\n \n Variable ret is being initialized with a value that is never read\n and ret is being re-assigned a little later on. The assignment is\n redundant and hence can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190629123306.12519-1-colin.king@canonical.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit cbf2be749076c266f51cafc71dc9045bcad9ed05\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Jun 28 17:14:19 2019 +0100\n\n gpio: bd70528: fix spelling misstake \"debouce\" -\u003e \"debounce\"\n \n There is a spelling mistake in a dev_err message. Fix it.\n \n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190628161419.26439-1-colin.king@canonical.com\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit c394159310d02defcfe63c617f13b0c2974725c5\nAuthor: Maximilian Luz \u003cluzmaximilian@gmail.com\u003e\nDate: Sat Jul 27 11:25:38 2019 +0300\n\n Input: soc_button_array - add support for newer surface devices\n \n Power and volume button support for 5th and 6th generation Microsoft\n Surface devices via soc_button_array.\n \n Note that these devices use the same MSHW0040 device as on the Surface\n Pro 4, however the implementation is different (GPIOs vs. ACPI\n notifications). Thus some checking is required to ensure we only load\n this driver on the correct devices.\n \n Signed-off-by: Maximilian Luz \u003cluzmaximilian@gmail.com\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 64dd243d735655d1ff6bc7450dd5203cf57c9dfb\nAuthor: Maximilian Luz \u003cluzmaximilian@gmail.com\u003e\nDate: Sun Jul 28 12:55:48 2019 +0300\n\n platform/x86: surfacepro3_button: Fix device check\n \n Do not use the surfacepro3_button driver on newer Microsoft Surface\n models, only use it on the Surface Pro 3 and 4. Newer models (5th, 6th\n and possibly future generations) use the same device as the Surface Pro\n 4 to represent their volume and power buttons (MSHW0040), but their\n actual implementation is significantly different. This patch ensures\n that the surfacepro3_button driver is only used on the Pro 3 and 4\n models, allowing a different driver to bind on other models.\n \n Signed-off-by: Maximilian Luz \u003cluzmaximilian@gmail.com\u003e\n Acked-by: Andy Shevchenko \u003candy.shevchenko@gmail.com\u003e\n Acked-by: Chen Yu \u003cyu.c.chen@intel.com\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 2229a3618abdaac40c9e32754509357ed78c90f5\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Wed Jun 26 11:34:18 2019 +0200\n\n unicore: Drop pointless include\n \n I can\u0027t see why this file includes \u003clinux/gpio.h\u003e, it is not\n using any of the interfaces. Lots of things are named \"gpio\"\n in the file but it is an irqchip driver and has nothing to\n do with the GPIO interfaces.\n \n Cc: Guan Xuetao \u003cgxt@pku.edu.cn\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190626093418.6263-1-linus.walleij@linaro.org\n\ncommit a444fa597e2a130d2c56f1bfbe131c62625d7b55\nAuthor: Beniamin Bia \u003cbeniamin.bia@analog.com\u003e\nDate: Thu Jul 18 09:27:34 2019 +0300\n\n iio: adc: ad7606: Add debug mode for ad7616\n \n Support for register access was added for devices which have software\n mode.\n \n Signed-off-by: Beniamin Bia \u003cbeniamin.bia@analog.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit f2a22e1e172f7dc486325216ec0890f301cb09e2\nAuthor: Beniamin Bia \u003cbeniamin.bia@analog.com\u003e\nDate: Thu Jul 18 09:27:33 2019 +0300\n\n iio: adc: ad7606: Add support for software mode for ad7616\n \n Support for ad7616 running in software was added. In order\n to activate the software mode, HW_RNGSEL pins must be pulled low.\n Oversampling and input ranges are now configured in corresponding\n registers. Ad7616 has multiple scale options when it is configured\n in software mode.\n Also, in order to support multiple devices in software mode, the spi\n calculation of registers address must be generic. Because\n the length of address and bit which specifies the read/write operation is\n different for every device, calculation of address was made generic.\n \n Signed-off-by: Beniamin Bia \u003cbeniamin.bia@analog.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit c5d93137af93c228392d133b3b00a0edae80dd40\nAuthor: Beniamin Bia \u003cbeniamin.bia@analog.com\u003e\nDate: Thu Jul 18 09:27:32 2019 +0300\n\n iio: adc: ad7606: Allow reconfigration after reset\n \n According to datasheet, ad7616 require at least 15ms after a restart\n to fully reconfigure and being able to receive new commands via spi.\n \n Signed-off-by: Beniamin Bia \u003cbeniamin.bia@analog.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit a0c648c07ce50744d0f6ddb58b483e87907e9122\nAuthor: Beniamin Bia \u003cbeniamin.bia@analog.com\u003e\nDate: Thu Jul 18 09:27:31 2019 +0300\n\n iio: adc: ad7606: Move spi dependent features to spi file\n \n Because software mode and register access are only available in spi, they\n were moved in spi file and are accessbile via bops structure.\n The write_os/scale will be overwritten by sw_mode_config function.\n This patch was made in order to support devices in software mode without\n making the driver dependent to spi and increase the abstraction of the\n core.\n \n Signed-off-by: Beniamin Bia \u003cbeniamin.bia@analog.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 7677f73158d3dac530daca7ce834603989c05b18\nAuthor: Beniamin Bia \u003cbeniamin.bia@analog.com\u003e\nDate: Thu Jul 18 09:27:30 2019 +0300\n\n iio: adc: ad7606: Move common channel definition to header\n \n The common channel definition which are going to be used by both core\n file but also spi file, were moved in header file. Some devices have\n different channel definitions when are used in software mode, feature\n available only with spi, and those definitions will be added in spi file.\n \n Signed-off-by: Beniamin Bia \u003cbeniamin.bia@analog.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 06ae56b9e723d653b6c14da947794d440554dff4\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Thu Feb 21 18:02:47 2019 +0100\n\n iio: document bindings for mounting matrices\n \n The mounting matrix for sensors was introduced in\n commit dfc57732ad38 (\"iio:core: mounting matrix support\")\n \n However the device tree bindings are very terse and since this is\n a widely applicable property, we need a proper binding for it\n that the other bindings can reference. This will also be useful\n for other operating systems and sensor engineering at large.\n \n I think all 3D sensors should support it, the current situation\n is probably that the mounting information is confined in magic\n userspace components rather than using the mounting matrix, which\n is not good for portability and reuse.\n \n Cc: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Cc: Gregor Boirie \u003cgregor.boirie@parrot.com\u003e\n Cc: Sebastian Reichel \u003csre@kernel.org\u003e\n Cc: Samu Onkalo \u003csamu.onkalo@intel.com\u003e\n Cc: devicetree@vger.kernel.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: H. Nikolaus Schaller \u003chns@goldelico.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit b207edfe4e021f3d992ec8a277edee103840dfd9\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Sat Jul 27 20:25:31 2019 -0700\n\n selftests/bpf: convert send_signal.c to use subtests\n \n Convert send_signal set of tests to be exposed as three sub-tests.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 51436ed78d59d0a0b7e64a2a2b997ac66a6c050e\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Sat Jul 27 20:25:30 2019 -0700\n\n selftests/bpf: convert bpf_verif_scale.c to sub-tests API\n \n Expose each BPF verifier scale test as individual sub-test to allow\n independent results output and test selection.\n \n Test run results now look like this:\n \n $ sudo ./test_progs -t verif/\n #3/1 loop3.o:OK\n #3/2 test_verif_scale1.o:OK\n #3/3 test_verif_scale2.o:OK\n #3/4 test_verif_scale3.o:OK\n #3/5 pyperf50.o:OK\n #3/6 pyperf100.o:OK\n #3/7 pyperf180.o:OK\n #3/8 pyperf600.o:OK\n #3/9 pyperf600_nounroll.o:OK\n #3/10 loop1.o:OK\n #3/11 loop2.o:OK\n #3/12 strobemeta.o:OK\n #3/13 strobemeta_nounroll1.o:OK\n #3/14 strobemeta_nounroll2.o:OK\n #3/15 test_sysctl_loop1.o:OK\n #3/16 test_sysctl_loop2.o:OK\n #3/17 test_xdp_loop.o:OK\n #3/18 test_seg6_loop.o:OK\n #3 bpf_verif_scale:OK\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 3a516a0a3a7b21ec038db83ffb0d3cddc42514c9\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Sat Jul 27 20:25:29 2019 -0700\n\n selftests/bpf: add sub-tests support for test_progs\n \n Allow tests to have their own set of sub-tests. Also add ability to do\n test/subtest selection using `-t \u003ctest-name\u003e/\u003csubtest-name\u003e` and `-n\n \u003ctest-nums-set\u003e/\u003csubtest-nums-set\u003e`, as an extension of existing -t/-n\n selector options. For the \u003ctest-num-set\u003e format: it\u0027s a comma-separated\n list of either individual test numbers (1-based), or range of test\n numbers. E.g., all of the following are valid sets of test numbers:\n - 10\n - 1,2,3\n - 1-3\n - 5-10,1,3-4\n \n \u0027/\u003csubtest\u0027 part is optional, but has the same format. E.g., to select\n test #3 and its sub-tests #10 through #15, use: -t 3/10-15.\n \n Similarly, to select tests by name, use `-t verif/strobe`:\n \n $ sudo ./test_progs -t verif/strobe\n #3/12 strobemeta.o:OK\n #3/13 strobemeta_nounroll1.o:OK\n #3/14 strobemeta_nounroll2.o:OK\n #3 bpf_verif_scale:OK\n Summary: 1/3 PASSED, 0 FAILED\n \n Example of using subtest API is in the next patch, converting\n bpf_verif_scale.c tests to use sub-tests.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 0ff97e56c0986ea6633083c3487d9231bbbd881b\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Sat Jul 27 20:25:28 2019 -0700\n\n selftests/bpf: abstract away test log output\n \n This patch changes how test output is printed out. By default, if test\n had no errors, the only output will be a single line with test number,\n name, and verdict at the end, e.g.:\n \n #31 xdp:OK\n \n If test had any errors, all log output captured during test execution\n will be output after test completes.\n \n It\u0027s possible to force output of log with `-v` (`--verbose`) option, in\n which case output won\u0027t be buffered and will be output immediately.\n \n To support this, individual tests are required to use helper methods for\n logging: `test__printf()` and `test__vprintf()`.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 329e38f76cc2a77085264ce6e0dbe902c33fd7a3\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Sat Jul 27 20:25:27 2019 -0700\n\n selftest/bpf: centralize libbpf logging management for test_progs\n \n Make test_progs test runner own libbpf logging. Also introduce two\n levels of verbosity: -v and -vv. First one will be used in subsequent\n patches to enable test log output always. Second one increases verbosity\n level of libbpf logging further to include debug output as well.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit e87fd8bae44c3eaa6205c9c81419e773896dc157\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Sat Jul 27 20:25:26 2019 -0700\n\n libbpf: return previous print callback from libbpf_set_print\n \n By returning previously set print callback from libbpf_set_print, it\u0027s\n possible to restore it, eventually. This is useful when running many\n independent test with one default print function, but overriding log\n verbosity for particular subset of tests.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 8160bae21fc29de0ec795abcd209cdd7cc144e8e\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Sat Jul 27 20:25:25 2019 -0700\n\n selftests/bpf: add test selectors by number and name to test_progs\n \n Add ability to specify either test number or test name substring to\n narrow down a set of test to run.\n \n Usage:\n sudo ./test_progs -n 1\n sudo ./test_progs -t attach_probe\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 766f2a59323a7881613af577718bb46cc1267b1f\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Sat Jul 27 20:25:24 2019 -0700\n\n selftests/bpf: revamp test_progs to allow more control\n \n Refactor test_progs to allow better control on what\u0027s being run.\n Also use argp to do argument parsing, so that it\u0027s easier to keep adding\n more options.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 61098e89e6c80d6a141774ef8ee41e38471b069e\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Sat Jul 27 20:25:23 2019 -0700\n\n selftests/bpf: prevent headers to be compiled as C code\n \n Apprently listing header as a normal dependency for a binary output\n makes it go through compilation as if it was C code. This currently\n works without a problem, but in subsequent commits causes problems for\n differently generated test.h for test_progs. Marking those headers as\n order-only dependency solves the issue.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit ae7b02ad2f32d39d1434655f346c04a16c1aa703\nAuthor: Fabien Lahoudere \u003cfabien.lahoudere@collabora.com\u003e\nDate: Tue Jul 16 11:11:06 2019 +0200\n\n iio: common: cros_ec_sensors: Expose cros_ec_sensors frequency range via iio sysfs\n \n Embedded controller return minimum and maximum frequencies, unfortunately\n we have no way to know the step for all available frequencies.\n Even if not complete, we can return a list of known values using the\n standard read_avail callback (IIO_CHAN_INFO_SAMP_FREQ) to provide them to\n userland.\n \n Now cros_ec_* sensors provides frequencies values in sysfs like this:\n \"0 min max\". 0 is always true to disable the sensor.\n \n Default frequencies are provided for earlier protocol.\n \n Signed-off-by: Nick Vaccaro \u003cnvaccaro@chromium.org\u003e\n Signed-off-by: Fabien Lahoudere \u003cfabien.lahoudere@collabora.com\u003e\n [rebased on top of iio/testing and solved conflicts]\n Signed-off-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 9566cb1ddb81c4f232d08cb2c7aff5b947959381\nAuthor: Gwendal Grignou \u003cgwendal@chromium.org\u003e\nDate: Mon Jul 15 16:14:54 2019 -0700\n\n iio: cros_ec_accel_legacy: Add support for veyron-minnie\n \n Veyron minnie embedded controller presents 2 accelerometers using an\n older interface. Add function to query the data in cros_ec_accel.\n \n Verify accelerometers on veyron-minnie are presented and working.\n \n Signed-off-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit d96e267762fc723079b8089172077e94a4cf8dd2\nAuthor: Gwendal Grignou \u003cgwendal@chromium.org\u003e\nDate: Mon Jul 15 16:14:53 2019 -0700\n\n iio: cros_ec_accel_legacy: Use cros_ec_sensors_core\n \n Remove duplicate code in cros-ec-accel-legacy,\n use cros-ec-sensors-core functions and structures when possible.\n \n On glimmer, check the 2 accelerometers are presented and working.\n \n Signed-off-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Reviewed-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit b9a0a64e3b35934128551a25513d91bd634736ed\nAuthor: Gwendal Grignou \u003cgwendal@chromium.org\u003e\nDate: Mon Jul 15 16:14:52 2019 -0700\n\n iio: cros_ec_accel_legacy: Fix incorrect channel setting\n \n INFO_SCALE is set both for each channel and all channels.\n iio is using all channel setting, so the error was not user visible.\n \n Signed-off-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 12bf745c9afb6755acba186091bcbb61229f4165\nAuthor: Gwendal Grignou \u003cgwendal@chromium.org\u003e\nDate: Mon Jul 15 16:14:51 2019 -0700\n\n iio: cros_ec: Add sign vector in core for backward compatibility\n \n To allow cros_ec iio core library to be used with legacy device, add a\n vector to rotate sensor data if necessary: legacy devices are not\n reporting data in HTML5/Android sensor referential.\n \n Check the data is not rotated on recent chromebooks that use the HTML5\n standard to present sensor data.\n \n Signed-off-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Reviewed-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 3cf9df00957fb27cb116761cd4d656aee06e40d3\nAuthor: Fabien Lahoudere \u003cfabien.lahoudere@collabora.com\u003e\nDate: Tue Jul 2 10:49:38 2019 +0200\n\n iio: common: cros_ec_sensors: determine protocol version\n \n This patch adds a function to determine which version of the\n protocol is used to communicate with EC.\n \n Signed-off-by: Nick Vaccaro \u003cnvaccaro@chromium.org\u003e\n Signed-off-by: Fabien Lahoudere \u003cfabien.lahoudere@collabora.com\u003e\n Reviewed-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Tested-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n [rebased on top of iio/testing and solved conflicts]\n Signed-off-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 3e4dabacf967535e87bb1b254c08471a2148faea\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Fri Jul 26 21:39:16 2019 +0800\n\n counter/ftm-quaddec: Use device-managed registration API\n \n Make use of devm_counter_register.\n Then we can remove redundant unregistration API\n usage to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit ec76d918f23034f9f662539ca9c64e2ae3ba9fba\nAuthor: Sean Nyekjaer \u003csean@geanix.com\u003e\nDate: Mon Jul 15 09:07:15 2019 +0200\n\n iio: imu: st_lsm6dsx: flip irq return logic\n \n No need for using reverse logic in the irq return,\n fix this by flip things around.\n \n Signed-off-by: Sean Nyekjaer \u003csean@geanix.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 2660b0080bb26506d2f12983485deb9d0cf2747c\nAuthor: Vitor Soares \u003cVitor.Soares@synopsys.com\u003e\nDate: Fri Jul 19 15:30:55 2019 +0200\n\n iio: imu: st_lsm6dsx: add i3c basic support for LSM6DSO and LSM6DSR\n \n For today the st_lsm6dsx driver support LSM6DSO and LSM6DSR sensor only in\n spi and i2c mode.\n \n The LSM6DSO and LSM6DSR are also i3c capable so let\u0027s give i3c support to\n them.\n \n Signed-off-by: Vitor Soares \u003cvitor.soares@synopsys.com\u003e\n Acked-by: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 3f2cde742632226a86b54b2f8ed9cf5c4f50638c\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Sun Jul 21 23:35:33 2019 +0200\n\n iio: magnetometer: mmc35240: Fix a typo in the name of a constant\n \n Everything is about mmc35240_ except MMC53240_WAIT_SET_RESET (3 and 5\n switched).\n \n This is likely a typo. Define and use MMC35240_WAIT_SET_RESET instead.\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit b4a339baf35995cd9443439120de60d4605384b1\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:13 2019 +0200\n\n iio: pressure: hp03: convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 48cfd4ef0d52c1e02152949556492011b2524555\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:12 2019 +0200\n\n iio: light: veml6070: convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit d96dbb3f1b74f5d521e9f6148e6d467be9892dcf\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:11 2019 +0200\n\n iio: light: cm36651: convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 1c667c4d99d7a42bc6bdcafc2896045e8c17b910\nAuthor: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\nDate: Tue Jul 23 10:36:40 2019 +0300\n\n dt-bindings: iio: imu: add bindings for ADIS16460\n \n This change adds device-tree bindings for the ADIS16460.\n \n Signed-off-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit db6ed4d23dd10a5bfbe922f23fc284bd4b4b0a1d\nAuthor: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\nDate: Tue Jul 23 10:36:39 2019 +0300\n\n iio: imu: Add support for the ADIS16460 IMU\n \n The ADIS16460 device is a complete inertial system that includes a triaxial\n gyroscope and a triaxial accelerometer. It\u0027s more simplified design than\n that of the ADIS16480, and does not offer the triaxial magnetometers \u0026\n pressure sensors. It does also have a temperature sensor (like the\n ADIS16480).\n Since it is part of the ADIS16XXX family, it re-uses parts of the ADIS\n library.\n \n Naturally, the register map is different and much more simplified than the\n ADIS16480 subfamily, so it cannot be integrated into that driver. A major\n difference is that the registers are not paged.\n \n One thing that is particularly special about it, is that it requires a\n higher delay between CS changes (i.e. when CS goes up, the spec recommends\n that it be brought down after a minimum of 16 uS).\n Other ADIS chips require (via spec) a minimum of 2 uS between CS changes.\n The kernel\u0027s 10 uS default should be fine for those other chips; they\n haven\u0027t been tested with lower CS change delays yet.\n \n Datasheet:\n https://www.analog.com/media/en/technical-documentation/data-sheets/adis16460.pdf\n \n Signed-off-by: Dragos Bogdan \u003cdragos.bogdan@analog.com\u003e\n Signed-off-by: Michael Hennerich \u003cmichael.hennerich@analog.com\u003e\n Signed-off-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit dca39af8831e65adc13e0f9f8bdca3a746185072\nAuthor: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\nDate: Tue Jul 23 10:36:38 2019 +0300\n\n iio: imu: adis: Add support for SPI transfer cs_change_delay\n \n The ADIS16460 requires a higher delay before the next transfer. Since the\n SPI framework supports configuring the delay before the next transfer, this\n driver will become the first user of it.\n \n The support for this functionality in ADIS16460 requires an addition to the\n ADIS lib to support the `cs_change_delay` functionality from the SPI\n subsystem.\n \n Signed-off-by: Michael Hennerich \u003cmichael.hennerich@analog.com\u003e\n Signed-off-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 6e4707ed3411d0cb3c1d9a0a093f77fbdf768fa5\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Fri Jul 26 14:36:16 2019 +0800\n\n iio: mxc4005: Use device-managed APIs\n \n Use device-managed APIs to simplify the code.\n The remove function is redundant now and can\n be deleted.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 23f1ce3c08aad743cd828ca5a36946f043ecf0f1\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Fri Jul 26 16:02:55 2019 +0800\n\n iio: humidity: Use device-managed APIs\n \n Use device-managed APIs to simplify the code.\n The remove functions are redundant now and can\n be deleted.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 857ba268f918dec28470c2074e7c69ccad52547e\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Fri Jul 26 19:07:42 2019 +0800\n\n staging: iio: ad2s1210: Use device-managed API\n \n Use device-managed API to simplify the code.\n The remove function is redundant now and can\n be deleted.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Reviewed-by: Alexandru Ardelean \u003calexandru.ardelean@analog.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 8ecc8b2b78e6af55aa366db4da79be1c772d5e55\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Fri Jul 26 18:49:50 2019 +0800\n\n iio: maxim_thermocouple: Use device-managed APIs\n \n Use device-managed APIs to simplify the code.\n The remove functions are redundant now and can\n be deleted.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 062809ef7733209312562e87cefc84a470430929\nAuthor: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\nDate: Thu Jul 18 15:53:53 2019 -0700\n\n iio: make st_sensors drivers use regmap\n \n This patch is meant to replace the i2c/spi transfer functions with\n regmap. SPI framework requires DMA safe buffers so let\u0027s add GFP_DMA\n flag for memory allocation used by bulk_read functions.\n \n Signed-off-by: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 1ecd245e0eb23d1c3803474eba75589743d0d1fe\nAuthor: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\nDate: Thu Jul 18 15:53:52 2019 -0700\n\n iio: move 3-wire spi initialization to st_sensors_spi\n \n Some devices need to be configured with special bit in order to\n use spi 3-wire. This was done during device identification phase.\n Instead, let\u0027s move this part as spi specific.\n Doing this the check_device_support function becomes a simple\n device id check, so let\u0027s rename it.\n \n Signed-off-by: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 570c2c55ef364ef6aed531e554be4c989fb1a49c\nAuthor: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\nDate: Thu Jul 18 15:53:51 2019 -0700\n\n iio:pressure: device settings are set immediately during probe\n \n This patch set pressure settings right after probe start. This is\n done in preparation of regmap that needs different configuration\n based on multiread bit value.\n \n Signed-off-by: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 291d83f2f4ef40e1d2d9504d29ecdf7000a655fb\nAuthor: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\nDate: Thu Jul 18 15:53:50 2019 -0700\n\n iio:magn: device settings are set immediately during probe\n \n This patch set magn settings right after probe start. This is\n done in preparation of regmap that needs different configuration\n based on multiread bit value.\n \n Signed-off-by: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 3c05484c79071f3a20acb7242630590ba596b99e\nAuthor: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\nDate: Thu Jul 18 15:53:49 2019 -0700\n\n iio:gyro: device settings are set immediately during probe\n \n This patch set gyro settings right after probe start. This is\n done in preparation of regmap that needs different configuration\n based on multiread bit value.\n \n Signed-off-by: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit feba6b18c926ad59cb7ea97e6023a38c6bf74310\nAuthor: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\nDate: Thu Jul 18 15:53:48 2019 -0700\n\n iio:accel: device settings are set immediately during probe\n \n This patch set accel settings right after probe start. This is\n done in preparation of regmap that needs different configuration\n based on multiread bit value.\n \n Signed-off-by: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 539b25fa700acf8985ea1a273fcccef5e52d52e4\nAuthor: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\nDate: Thu Jul 18 15:53:47 2019 -0700\n\n iio:pressure: introduce st_press_get_settings() function\n \n The function is introduced to retrieve press device settings. It will be\n used by probe in order to configure regmap.\n \n Signed-off-by: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit aa4e75c85076c056d944b92ed94b76277430d263\nAuthor: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\nDate: Thu Jul 18 15:53:46 2019 -0700\n\n iio:magn: introduce st_magn_get_settings() function\n \n The function is introduced to retrieve magn device settings. It will be\n used by probe in order to configure regmap.\n \n Signed-off-by: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 2acca26bba3a7487e941e234e8ee33a998eecf53\nAuthor: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\nDate: Thu Jul 18 15:53:45 2019 -0700\n\n iio:gyro: introduce st_gyro_get_settings() function\n \n The function is introduced to retrieve gyro device settings. It will be\n used by probe in order to configure regmap.\n \n Signed-off-by: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 823a7adac4f1bde2828d18f3bd8fa75285d00f4d\nAuthor: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\nDate: Thu Jul 18 15:53:44 2019 -0700\n\n iio:accel: introduce st_accel_get_settings() function\n \n The function is introduced to retrieve accel device settings. It will be\n used by probe in order to configure regmap.\n \n Signed-off-by: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit a090965b882333500d8780e2c1762e17782f413f\nAuthor: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\nDate: Thu Jul 18 15:53:43 2019 -0700\n\n iio:common:st_sensors: add st_sensors_get_settings_index() helper function\n \n Extract from st_sensors_check_device_support() function the code that\n is used to get the specific settings for a device. This will be used\n as generic extractor by each ST driver.\n \n Signed-off-by: Denis Ciocca \u003cdenis.ciocca@st.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit f53199c0bc62657f7bc253b5dfc9f3d81ed2ca28\nAuthor: Gwendal Grignou \u003cgwendal@chromium.org\u003e\nDate: Thu Jul 18 15:22:37 2019 -0700\n\n iio: cros_ec: Remove replacing error code with -EIO\n \n Due to an API misread, error code can be different for -EIO when reading\n a sysfs entry. Return the error reported by the cros_ec stack.\n \n Check the proper error message (protocol error, not supported) is\n reported when there is an error returned by the EC stack.\n \n Signed-off-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 7079676d0931e30f4f4949d1580fc87fde4ddc5b\nAuthor: Jiri Pirko \u003cjiri@mellanox.com\u003e\nDate: Sat Jul 27 20:32:57 2019 +0300\n\n mlxsw: spectrum_flower: Forbid to offload match on reserved TCP flags bits\n \n Matching on reserved TCP flags bits is only supported using custom\n parser. Since the usecase for that is not known now, just forbid to\n offload rules that match on these bits.\n \n Reported-by: Alex Kushnarov \u003calexanderk@mellanox.com\u003e\n Signed-off-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c9588e28123c563a355964c6ee8ca5dd28aebd6a\nAuthor: Jiri Pirko \u003cjiri@mellanox.com\u003e\nDate: Sat Jul 27 20:32:56 2019 +0300\n\n mlxsw: spectrum_acl: Track rules that forbid egress block bind\n \n Some matches and actions are not supported on egress. Track such rules\n and forbid a bind of block which contains them to egress.\n \n With this patch, the kernel tells the user he cannot do that:\n $ tc qdisc add dev ens16np1 ingress_block 22 clsact\n $ tc filter add block 22 protocol 802.1q pref 2 handle 101 flower vlan_id 100 skip_sw action pass\n $ tc qdisc add dev ens16np2 egress_block 22 clsact\n Error: mlxsw_spectrum: Block cannot be bound to egress because it contains unsupported rules.\n \n Signed-off-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 185556f092491120ea2bd7eab6f9f78ff6c4d6d0\nAuthor: Jiri Pirko \u003cjiri@mellanox.com\u003e\nDate: Sat Jul 27 20:32:55 2019 +0300\n\n mlxsw: spectrum_flower: Forbid to offload mirred redirect on egress\n \n Spectrum ASIC does not support redirection on egress, so refuse to\n insert such flows:\n \n $ tc qdisc add dev ens16np1 clsact\n $ tc filter add dev ens16np1 egress protocol all pref 1 handle 101 flower skip_sw action mirred egress redirect dev ens16np2\n Error: mlxsw_spectrum: Redirect action is not supported on egress.\n \n Signed-off-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 93681cd7d94f83903cb3f0f95433d10c28a7e9a5\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Fri Jul 26 21:51:36 2019 +0200\n\n r8169: enable HW csum and TSO\n \n Enable HW csum and TSO per default except on known buggy chip versions.\n Realtek confirmed that RTL8168evl has a HW issue with TSO.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 96ea772ef24114e6aa3ce39599474ae794be625f\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Fri Jul 26 21:50:34 2019 +0200\n\n r8169: remove r8169_csum_workaround\n \n The loop in r8169_csum_workaround is called only if in\n msdn_giant_send_check a copy of the skb header needs to be made and\n we don\u0027t have enough memory. Let\u0027s simply drop the packet in that case\n so that we can remove r8169_csum_workaround.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit e64e0c89749969b22415fd3e766b33b37264c6ea\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Fri Jul 26 21:49:22 2019 +0200\n\n r8169: implement callback ndo_features_check\n \n Implement callback ndo_features_check and move all feature checks there.\n This will allow us to get rid of r8169_csum_workaround() completely in\n a subsequent step. Like in the vendor driver disable HW csum for short\n packets on RTL8168b.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 0170d594ded8297760ca9d8eb7eb6a9aff378059\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Fri Jul 26 21:48:32 2019 +0200\n\n r8169: set GSO size and segment limits\n \n Set GSO max size and max segment number as in the vendor driver.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 280b0b8e89ade4277147e598d5806de12bff5fbc\nAuthor: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\nDate: Fri Jul 26 12:16:09 2019 -0700\n\n ipv6: remove printk\n \n ipv6_find_hdr() prints a non-rate limited error message\n when it cannot find an ipv6 header at a specific offset.\n This could be used as a DoS, so just remove it.\n \n Signed-off-by: Jonathan Lemon \u003cjonathan.lemon@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit edcde3ee579bcac75ceb47758f18d2b1d26a39f8\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Fri Jul 26 20:56:20 2019 +0200\n\n r8169: align setting PME with vendor driver\n \n Align setting PME with the vendor driver. PMEnable is writable on\n RTL8169 only, on later chip versions it\u0027s read-only. PME_SIGNAL is\n used on chip versions from RTL8168evl with the exception of the\n RTL8168f family.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 91c6bfb8315b05d0c3aa8c2cfc172888d05e31d5\nAuthor: Davide Caratti \u003cdcaratti@redhat.com\u003e\nDate: Fri Jul 26 20:18:12 2019 +0200\n\n mlx4/en_netdev: allow offloading VXLAN over VLAN\n \n ConnectX-3 Pro can offload transmission of VLAN packets with VXLAN inside:\n enable tunnel offloads in dev-\u003evlan_features, like it\u0027s done with other\n NIC drivers (e.g. be2net and ixgbe).\n \n It\u0027s no more necessary to change dev-\u003ehw_enc_features when VXLAN are added\n or removed, since .ndo_features_check() already checks for VXLAN packet\n where the UDP destination port matches the configured value. Just set\n dev-\u003ehw_enc_features when the NIC is initialized, so that overlying VLAN\n can correctly inherit the tunnel offload capabilities.\n \n Changes since v1:\n - avoid flipping hw_enc_features, instead of calling netdev notifiers,\n thanks to Saeed Mahameed\n - squash two patches into a single one\n \n CC: Paolo Abeni \u003cpabeni@redhat.com\u003e\n CC: Marcelo Ricardo Leitner \u003cmarcelo.leitner@gmail.com\u003e\n Signed-off-by: Davide Caratti \u003cdcaratti@redhat.com\u003e\n Reviewed-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 85d2c5cde158ce00fe965561cfc57404fbefb0a7\nAuthor: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\nDate: Fri Jul 26 09:20:37 2019 -0700\n\n drivers: net: xgene: Move status variable declaration into CONFIG_ACPI block\n \n When CONFIG_ACPI is unset (arm allyesconfig), status is unused.\n \n drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c:383:14: warning:\n unused variable \u0027status\u0027 [-Wunused-variable]\n acpi_status status;\n ^\n drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c:440:14: warning:\n unused variable \u0027status\u0027 [-Wunused-variable]\n acpi_status status;\n ^\n drivers/net/ethernet/apm/xgene/xgene_enet_hw.c:697:14: warning: unused\n variable \u0027status\u0027 [-Wunused-variable]\n acpi_status status;\n ^\n \n Move the declaration into the CONFIG_ACPI block so that there are no\n compiler warnings.\n \n Fixes: 570d785ba46b (\"drivers: net: xgene: Remove acpi_has_method() calls\")\n Signed-off-by: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ddfbee9e3204a0158774bbe9df0f555573e81f43\nAuthor: Thierry Reding \u003ctreding@nvidia.com\u003e\nDate: Fri Jul 26 12:27:41 2019 +0200\n\n net: stmmac: Do not request stmmaceth clock\n \n The stmmaceth clock is specified by the slave_bus and apb_pclk clocks in\n the device tree bindings for snps,dwc-qos-ethernet-4.10 compatible nodes\n of this IP.\n \n The subdrivers for these bindings will be requesting the stmmac clock\n correctly at a later point, so there is no need to request it here and\n cause an error message to be printed to the kernel log.\n \n Signed-off-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1a981c0586c038710227eb740350f291e77ce365\nAuthor: Thierry Reding \u003ctreding@nvidia.com\u003e\nDate: Fri Jul 26 12:27:40 2019 +0200\n\n net: stmmac: Make MDIO bus reset optional\n \n The Tegra EQOS driver already resets the MDIO bus at probe time via the\n reset GPIO specified in the phy-reset-gpios device tree property. There\n is no need to reset the bus again later on.\n \n This avoids the need to query the device tree for the snps,reset GPIO,\n which is not part of the Tegra EQOS device tree bindings. This quiesces\n an error message from the generic bus reset code if it doesn\u0027t find the\n snps,reset related delays.\n \n Signed-off-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f530eed65bcaf9f74c312bbea09a36a27c48e06c\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Jul 26 10:46:11 2019 +0100\n\n net: neigh: remove redundant assignment to variable bucket\n \n The variable bucket is being initialized with a value that is never\n read and it is being updated later with a new value in a following\n for-loop. The initialization is redundant and can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ed1f2e85da79274f3dc4092953f1359eb732f0c6\nAuthor: Gwendal Grignou \u003cgwendal@chromium.org\u003e\nDate: Thu Jul 18 16:28:24 2019 -0700\n\n iio: cros_ec: Add calibscale for 3d MEMS\n \n Add calibration scale support to accel, gyro and magnetometer.\n \n Check on eve with current firmware, check reading calibscale returns 1.0,\n check with newer firmware values are applied.\n \n Signed-off-by: Gwendal Grignou \u003cgwendal@chromium.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 50e5bae0c1f976e881eecf4d1ad70ce1593021c9\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Sun Jul 21 12:49:18 2019 +0200\n\n iio: light: apds9960: Fix a typo\n \n s/ADPS9960/APDS9960/\n (P and D switched)\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 80343f5b8a203c5643dfdcc87a6e7a6d4db81226\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Sun Jul 21 12:53:53 2019 +0200\n\n iio: sca3000: Fix a typo\n \n All #define are about SCA3000_... except the last one.\n Make it consistent.\n \n s/SAC3000/SCA3000/\n \n This #define is apparently unused up to now.\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit b299d00420e2cc4289b1909943bb1e68634ef0bb\nAuthor: Benjamin Gaignard \u003cbenjamin.gaignard@st.com\u003e\nDate: Tue May 7 11:12:24 2019 +0200\n\n IIO: stm32: Remove quadrature related functions from trigger driver\n \n Quadrature feature is now hosted on it own framework.\n Remove quadrature related code from stm32-trigger driver to avoid\n code duplication and simplify the ABI.\n \n Signed-off-by: Benjamin Gaignard \u003cbenjamin.gaignard@st.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 0eec1f38c7bb53b9dd2221ac982739cce7aab7fc\nAuthor: Bárbara Fernandes \u003cbarbara.fernandes@usp.br\u003e\nDate: Fri Jun 28 16:49:22 2019 -0300\n\n staging: iio: ad7192: create of_device_id array\n \n Create list of compatible device ids to be matched with those stated in\n the device tree.\n \n Signed-off-by: Bárbara Fernandes \u003cbarbara.fernandes@usp.br\u003e\n Signed-off-by: Wilson Sales \u003cspoonm@spoonm.org\u003e\n Co-developed by: Wilson Sales \u003cspoonm@spoonm.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 0ebcdf3dfa3359c6a301ff1bc22cf67291d69113\nAuthor: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\nDate: Sat Jun 29 12:13:54 2019 +0200\n\n dt-bindings: iio: imu: st_lsm6dsx: add lsm6ds3tr-c device bindings\n \n Signed-off-by: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit dbcd2088a718875b4ef214099e9b3f378c47c2f4\nAuthor: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\nDate: Sat Jun 29 12:13:53 2019 +0200\n\n iio: imu: st_lsm6dsx: add support to LSM6DS3TR-C\n \n Add support to STM LSM6DS3TR-C 6-axis (acc + gyro) Mems sensor\n https://www.st.com/resource/en/datasheet/lsm6ds3tr-c.pdf\n \n Signed-off-by: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit daa2b5c6229d568b7fdef5f8d33cf271fecf8b8d\nAuthor: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\nDate: Mon Jul 1 15:56:31 2019 +0200\n\n Documentation: ABI: iio: Add missing newline at end of file\n \n \"git diff\" says:\n \n \\ No newline at end of file\n \n after modifying the files.\n \n Signed-off-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 3a81bde5457ec3292cf05c83c9a0b731b1b1531e\nAuthor: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\nDate: Wed Jul 3 12:08:15 2019 +0200\n\n iio: adc: stm32-adc: add analog switches supply control\n \n On stm32h7 and stm32mp1, the ADC inputs are multiplexed with analog\n switches which have reduced performances when their supply is below 2.7V\n (vdda by default):\n - 3.3V embedded booster can be used, to get full ADC performances\n (increases power consumption).\n - vdd supply can be selected if above 2.7V by setting ANASWVDD syscfg bit,\n on STM32MP1 only.\n \n Make this optional, since this is a trade-off between analog performance\n and power consumption.\n \n Signed-off-by: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit a85a43e04523df3b0f1e5f3100592806e29966a8\nAuthor: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\nDate: Wed Jul 3 12:08:14 2019 +0200\n\n dt-bindings: iio: adc: stm32: add analog switches supply control\n \n On stm32h7 and stm32mp1, the ADC inputs are multiplexed with analog\n switches which have reduced performances when their supply is below 2.7V\n (vdda by default). Booster or vdd can be used, to get full ADC\n performances.\n \n Add documentation for:\n - optional booster-supply that can be used on stm32h7 and stm32mp1.\n - optional vdd-supply that can be used on stm32mp1 to supply analog\n circuitry, selected by syscfg.\n Mark these as optional, since this is a trade-off between analog\n performance and power consumption.\n \n Signed-off-by: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit c3a686070670d9efad9a4af354ca88031cd7e7f9\nAuthor: Luca Weiss \u003cluca@z3ntu.xyz\u003e\nDate: Wed Jul 3 20:05:59 2019 +0200\n\n iio: light: stk3310: Add device tree support\n \n Add device tree support for the stk33xx family of ambient light sensors.\n \n Tested-by: Martijn Braam \u003cmartijn@brixit.nl\u003e\n Signed-off-by: Luca Weiss \u003cluca@z3ntu.xyz\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 04a14aed81144787e05f56faa7c07206b0c7308a\nAuthor: Luca Weiss \u003cluca@z3ntu.xyz\u003e\nDate: Wed Jul 3 20:05:58 2019 +0200\n\n dt-bindings: iio: light: add stk33xx\n \n Add binding documentation for the stk33xx family of ambient light\n sensors.\n \n Signed-off-by: Luca Weiss \u003cluca@z3ntu.xyz\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 82efbabfe89718215ac9d6f9c0f21ce4e3c4a9cb\nAuthor: Luca Weiss \u003cluca@z3ntu.xyz\u003e\nDate: Wed Jul 3 20:05:57 2019 +0200\n\n dt-bindings: Add vendor prefix for sensortek\n \n Sensortek Technology Corp. produces Proximity Sensors with ALS and\n Accelerometers.\n \n Signed-off-by: Luca Weiss \u003cluca@z3ntu.xyz\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit d284feaf982aa881ca78cb4d81954d7240d5d2b9\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Mon Jul 8 20:33:41 2019 +0800\n\n iio: imu: mpu6050: Replace devm_add_action() followed by failure action with devm_add_action_or_reset()\n \n devm_add_action_or_reset() is introduced as a helper function which\n internally calls devm_add_action(). If devm_add_action() fails\n then it will execute the action mentioned and return the error code.\n This reduce source code size (avoid writing the action twice)\n and reduce the likelyhood of bugs.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 46916ded7602039f7ecc02a49dd349a90a493d8c\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Mon Jul 8 20:32:43 2019 +0800\n\n iio: adc: sc27xx: Replace devm_add_action() followed by failure action with devm_add_action_or_reset()\n \n devm_add_action_or_reset() is introduced as a helper function which\n internally calls devm_add_action(). If devm_add_action() fails\n then it will execute the action mentioned and return the error code.\n This reduce source code size (avoid writing the action twice)\n and reduce the likelyhood of bugs.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit a0b4dddaa131089cedbb4dea182b136e6aa1f21e\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Mon Jul 8 20:32:21 2019 +0800\n\n iio: adc: sc27xx: Introduce local variable \u0027struct device *dev\u0027\n \n Introduce local variable \u0027struct device *dev\u0027 and use it instead of\n dereferencing it repeatly.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 8e97d4c8a4633f6d826e1872448435894df1794c\nAuthor: Tomasz Duszynski \u003ctduszyns@gmail.com\u003e\nDate: Wed Jul 10 21:21:55 2019 +0200\n\n MAINTAINERS: add entry for plantower pms7003 driver\n \n Add myself as a plantower pms7003 driver maintainer.\n \n Signed-off-by: Tomasz Duszynski \u003ctduszyns@gmail.com\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit 433514fc7002bafcfafd222bf4341c391ccffa25\nAuthor: Tomasz Duszynski \u003ctduszyns@gmail.com\u003e\nDate: Wed Jul 10 21:21:54 2019 +0200\n\n dt-bindings: iio: chemical: pms7003: convert bindings to yaml\n \n Convert existing device tree bindings to yaml.\n \n Signed-off-by: Tomasz Duszynski \u003ctduszyns@gmail.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Jonathan Cameron \u003cJonathan.Cameron@huawei.com\u003e\n\ncommit a5021c4597913aad16d9d176cbd1de7fa514477e\nAuthor: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\nDate: Sat Jul 27 16:27:35 2019 +0200\n\n ARM: dts: rockchip: Cleanup style around assignment operator\n \n Use a space before and after assignment operator to have consistent\n style.\n \n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit 5f68056ca50fdd3954a93ae66fea7452abddb66f\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Sat Jul 27 10:03:14 2019 -0400\n\n autofs_lookup(): hold -\u003ed_lock over playing with -\u003ed_flags\n \n ... as well as setting -\u003ed_fsdata, etc. Make all of that\n atomic.\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit c4931db9b08c18005fb21ab201e7137ba0547df5\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Sat Jul 27 10:00:33 2019 -0400\n\n get rid of autofs_info-\u003eactive_count\n \n autofs_add_active() is always called only once (and on a dentry\n with freshly allocated ino, at that). autofs_del_active() is\n never called more than once.\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit cd10afbc932dd0dc556378f892479647b131e2c9\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Tue Jul 23 20:30:36 2019 +0900\n\n sh: remove unneeded uapi asm-generic wrappers\n \n These are listed in include/uapi/asm-generic/Kbuild, so Kbuild will\n automatically generate them.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Yoshinori Sato \u003cysato@users.sourceforge.jp\u003e\n\ncommit cbfc6edb6a4a46aeb5ff25c4702d73708cfc8928\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Tue Jul 23 16:49:43 2019 +0900\n\n sh: use __builtin_constant_p() directly instead of IS_IMMEDIATE()\n \n __builtin_constant_p(nr) is used everywhere now. It does not make\n much sense to define IS_IMMEDIATE() as its alias.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Yoshinori Sato \u003cysato@users.sourceforge.jp\u003e\n\ncommit a5de8865cb3edc9c5b719dc45651b38226805f9e\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Tue Jul 23 19:21:06 2019 +0900\n\n h8300: move definition of __kernel_size_t etc. to posix_types.h\n \n These types should be defined in posix_types.h, not in bitsperlong.h .\n \n With these defines moved, h8300-specific bitsperlong.h is no longer\n needed since Kbuild will automatically create a wrapper of\n include/uapi/asm-generic/bitsperlong.h\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Yoshinori Sato \u003cysato@users.sourceforge.jp\u003e\n\ncommit 62336cc666b0e0f1f2bebd682c1741b9a439095a\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Sat Jul 27 10:54:41 2019 +0100\n\n drm/i915/uc: Fixup kerneldoc after params were flipped and renamed\n \n drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c:194: warning: Function parameter or member \u0027i915\u0027 not described in \u0027intel_uc_fw_fetch\u0027\n drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c:194: warning: Excess function parameter \u0027dev_priv\u0027 description in \u0027intel_uc_fw_fetch\u0027\n \n Fixes: 97dee74bb34a (\"drm/i915/uc: Reorder params in intel_uc_fw_fetch\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190727101055.5300-1-chris@chris-wilson.co.uk\n\ncommit 8dfa20fcfbeb245642dfe3a43f8a3735d9aed42a\nAuthor: Eric Biggers \u003cebiggers@google.com\u003e\nDate: Fri Jul 19 23:09:18 2019 -0700\n\n crypto: ghash - add comment and improve help text\n \n To help avoid confusion, add a comment to ghash-generic.c which explains\n the convention that the kernel\u0027s implementation of GHASH uses.\n \n Also update the Kconfig help text and module descriptions to call GHASH\n a \"hash function\" rather than a \"message digest\", since the latter\n normally means a real cryptographic hash function, which GHASH is not.\n \n Cc: Pascal Van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Signed-off-by: Eric Biggers \u003cebiggers@google.com\u003e\n Reviewed-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Acked-by: Pascal Van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 065cf577135a4977931c7a1e1edf442bfd9773dd\nAuthor: Daniel Jordan \u003cdaniel.m.jordan@oracle.com\u003e\nDate: Fri Jul 19 15:04:44 2019 -0400\n\n padata: purge get_cpu and reorder_via_wq from padata_do_serial\n \n With the removal of the padata timer, padata_do_serial no longer\n needs special CPU handling, so remove it.\n \n Signed-off-by: Daniel Jordan \u003cdaniel.m.jordan@oracle.com\u003e\n Cc: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n Cc: Steffen Klassert \u003csteffen.klassert@secunet.com\u003e\n Cc: linux-crypto@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit b3553effafa027feca33e0aa357fce08b3c80b02\nAuthor: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\nDate: Fri Jul 19 10:09:24 2019 +0300\n\n crypto: bcm - check assoclen for rfc4543/rfc4106\n \n Validated assoclen for RFC4543 which expects an assoclen\n of 16 or 20, the same as RFC4106.\n Based on seqiv, IPsec ESP and RFC4543/RFC4106 the assoclen is sizeof\n IP Header (spi, seq_no, extended seq_no) and IV len. This can be 16 or\n 20 bytes.\n \n Signed-off-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit b93ecf4296742c36192621645f59e4a607263233\nAuthor: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\nDate: Fri Jul 19 10:09:23 2019 +0300\n\n crypto: ccree - check assoclen for rfc4543\n \n Check assoclen to solve the extra tests that expect -EINVAL to be\n returned when the associated data size is not valid.\n \n Validated assoclen for RFC4543 which expects an assoclen\n of 16 or 20, the same as RFC4106.\n Based on seqiv, IPsec ESP and RFC4543/RFC4106 the assoclen is sizeof\n IP Header (spi, seq_no, extended seq_no) and IV len. This can be 16 or\n 20 bytes.\n \n Signed-off-by: Iuliana Prodan \u003ciuliana.prodan@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 6fc4dbcf0276279d488c5fbbfabe94734134f4fa\nAuthor: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nDate: Thu Jul 18 23:01:46 2019 +0800\n\n padata: Replace delayed timer with immediate workqueue in padata_reorder\n \n The function padata_reorder will use a timer when it cannot progress\n while completed jobs are outstanding (pd-\u003ereorder_objects \u003e 0). This\n is suboptimal as if we do end up using the timer then it would have\n introduced a gratuitous delay of one second.\n \n In fact we can easily distinguish between whether completed jobs\n are outstanding and whether we can make progress. All we have to\n do is look at the next pqueue list.\n \n This patch does that by replacing pd-\u003eprocessed with pd-\u003ecpu so\n that the next pqueue is more accessible.\n \n A work queue is used instead of the original try_again to avoid\n hogging the CPU.\n \n Note that we don\u0027t bother removing the work queue in\n padata_flush_queues because the whole premise is broken. You\n cannot flush async crypto requests so it makes no sense to even\n try. A subsequent patch will fix it by replacing it with a ref\n counting scheme.\n \n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 97ac82d9135d989c81268cb0a02d336993b6f9cf\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Thu Jul 18 15:50:04 2019 +0200\n\n crypto: aegis - fix badly optimized clang output\n \n Clang sometimes makes very different inlining decisions from gcc.\n In case of the aegis crypto algorithms, it decides to turn the innermost\n primitives (and, xor, ...) into separate functions but inline most of\n the rest.\n \n This results in a huge amount of variables spilled on the stack, leading\n to rather slow execution as well as kernel stack usage beyond the 32-bit\n warning limit when CONFIG_KASAN is enabled:\n \n crypto/aegis256.c:123:13: warning: stack frame size of 648 bytes in function \u0027crypto_aegis256_encrypt_chunk\u0027 [-Wframe-larger-than\u003d]\n crypto/aegis256.c:366:13: warning: stack frame size of 1264 bytes in function \u0027crypto_aegis256_crypt\u0027 [-Wframe-larger-than\u003d]\n crypto/aegis256.c:187:13: warning: stack frame size of 656 bytes in function \u0027crypto_aegis256_decrypt_chunk\u0027 [-Wframe-larger-than\u003d]\n crypto/aegis128l.c:135:13: warning: stack frame size of 832 bytes in function \u0027crypto_aegis128l_encrypt_chunk\u0027 [-Wframe-larger-than\u003d]\n crypto/aegis128l.c:415:13: warning: stack frame size of 1480 bytes in function \u0027crypto_aegis128l_crypt\u0027 [-Wframe-larger-than\u003d]\n crypto/aegis128l.c:218:13: warning: stack frame size of 848 bytes in function \u0027crypto_aegis128l_decrypt_chunk\u0027 [-Wframe-larger-than\u003d]\n crypto/aegis128.c:116:13: warning: stack frame size of 584 bytes in function \u0027crypto_aegis128_encrypt_chunk\u0027 [-Wframe-larger-than\u003d]\n crypto/aegis128.c:351:13: warning: stack frame size of 1064 bytes in function \u0027crypto_aegis128_crypt\u0027 [-Wframe-larger-than\u003d]\n crypto/aegis128.c:177:13: warning: stack frame size of 592 bytes in function \u0027crypto_aegis128_decrypt_chunk\u0027 [-Wframe-larger-than\u003d]\n \n Forcing the primitives to all get inlined avoids the issue and the\n resulting code is similar to what gcc produces.\n \n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Acked-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit bfb5eb084ae93eb86d3b088c301500db138963e5\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Thu Jul 18 21:16:09 2019 +0800\n\n crypto: ccp - Replace dma_pool_alloc + memset with dma_pool_zalloc\n \n Use dma_pool_zalloc instead of using dma_pool_alloc to allocate\n memory and then zeroing it with memset 0.\n This simplifies the code.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Acked-by: Gary R Hook \u003cgary.hook@amd.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 6ed01097f4378199614fe625495657af7c601cea\nAuthor: Vakul Garg \u003cvakul.garg@nxp.com\u003e\nDate: Thu Jul 18 11:29:16 2019 +0000\n\n crypto: caam/qi2 - Increase napi budget to process more caam responses\n \n While running ipsec processing for traffic through multiple network\n interfaces, it is observed that caam driver gets less time to poll\n responses from caam block compared to ethernet driver. This is because\n ethernet driver has as many napi instances per cpu as the number of\n ethernet interfaces in system. Therefore, caam driver\u0027s napi executes\n lesser than the ethernet driver\u0027s napi instances. This results in\n situation that we end up submitting more requests to caam (which it is\n able to finish off quite fast), but don\u0027t dequeue the responses at same\n rate. This makes caam response FQs bloat with large number of frames. In\n some situations, it makes kernel crash due to out-of-memory. To prevent\n it We increase the napi budget of dpseci driver to a big value so that\n caam driver is able to drain its response queues at enough rate.\n \n Signed-off-by: Vakul Garg \u003cvakul.garg@nxp.com\u003e\n Reviewed-by: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit f2f1d75ab725106c4958bda30f87bd39a0e3040a\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jul 17 17:04:38 2019 +0800\n\n hwrng: mxc-rnga - use devm_platform_ioremap_resource() to simplify code\n \n Use the new helper devm_platform_ioremap_resource() which wraps the\n platform_get_resource() and devm_ioremap_resource() together, to\n simplify the code.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit d10d094cbfd338f361a83a82e5d99feb7933008d\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jul 17 17:04:37 2019 +0800\n\n hwrng: imx-rngc - use devm_platform_ioremap_resource() to simplify code\n \n Use the new helper devm_platform_ioremap_resource() which wraps the\n platform_get_resource() and devm_ioremap_resource() together, to\n simplify the code.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Acked-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Reviewed-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 72c8117adfced37df101c8c0b3f363e0906f83f0\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Jul 12 10:59:24 2019 +0200\n\n crypto: ccp - Reduce maximum stack usage\n \n Each of the operations in ccp_run_cmd() needs several hundred\n bytes of kernel stack. Depending on the inlining, these may\n need separate stack slots that add up to more than the warning\n limit, as shown in this clang based build:\n \n drivers/crypto/ccp/ccp-ops.c:871:12: error: stack frame size of 1164 bytes in function \u0027ccp_run_aes_cmd\u0027 [-Werror,-Wframe-larger-than\u003d]\n static int ccp_run_aes_cmd(struct ccp_cmd_queue *cmd_q, struct ccp_cmd *cmd)\n \n The problem may also happen when there is no warning, e.g. in the\n ccp_run_cmd()-\u003eccp_run_aes_cmd()-\u003eccp_run_aes_gcm_cmd() call chain with\n over 2000 bytes.\n \n Mark each individual function as \u0027noinline_for_stack\u0027 to prevent\n this from happening, and move the calls to the two special cases for aes\n into the top-level function. This will keep the actual combined stack\n usage to the mimimum: 828 bytes for ccp_run_aes_gcm_cmd() and\n at most 524 bytes for each of the other cases.\n \n Fixes: 63b945091a07 (\"crypto: ccp - CCP device driver and interface support\")\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 934d24a5e1508e73c0001afb54a3916e4270428f\nAuthor: Vitor Soares \u003cVitor.Soares@synopsys.com\u003e\nDate: Fri Jul 19 15:30:54 2019 +0200\n\n i3c: move i3c_device_match_id to device.c and export it\n \n Some I3C device drivers need to know which entry matches the\n i3c_device object passed to the probe function\n \n Let\u0027s move i3c_device_match_id() to device.c and export it so it can be\n used by drivers.\n \n Signed-off-by: Vitor Soares \u003cvitor.soares@synopsys.com\u003e\n Signed-off-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n\ncommit cdab7e2c73d5b499f2e33e87423707dc28929850\nAuthor: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nDate: Fri Jul 26 17:29:44 2019 +0100\n\n mips: vdso: Fix flip/flop vdso building bug\n \n Running \"make\" on an already compiled kernel tree will rebuild the\n vdso library even if this has not been modified.\n \n $ make\n GEN Makefile\n Using linux as source for kernel\n CALL linux/scripts/atomic/check-atomics.sh\n CALL linux/scripts/checksyscalls.sh\n \u003cstdin\u003e:1511:2: warning: #warning syscall clone3 not implemented [-Wcpp]\n CHK include/generated/compile.h\n VDSO arch/mips/vdso/vdso.so.dbg.raw\n OBJCOPY arch/mips/vdso/vdso.so.raw\n GENVDSO arch/mips/vdso/vdso-image.c\n CC arch/mips/vdso/vdso-image.o\n AR arch/mips/vdso/built-in.a\n AR arch/mips/built-in.a\n CHK include/generated/autoksyms.h\n GEN .version\n CHK include/generated/compile.h\n UPD include/generated/compile.h\n CC init/version.o\n AR init/built-in.a\n LD vmlinux.o\n MODPOST vmlinux.o\n MODINFO modules.builtin.modinfo\n KSYM .tmp_kallsyms1.o\n KSYM .tmp_kallsyms2.o\n LD vmlinux\n SORTEX vmlinux\n SYSMAP System.map\n Building modules, stage 2.\n ITS arch/mips/boot/vmlinux.gz.its\n OBJCOPY arch/mips/boot/vmlinux.bin\n MODPOST 7 modules\n GZIP arch/mips/boot/vmlinux.bin.gz\n ITB arch/mips/boot/vmlinux.gz.itb\n \n The issue is generated by the fact that \"if_changed\" is called twice\n in a single target.\n \n Fix the build bug merging the two commands into a single function.\n \n Signed-off-by: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit b4c0f7fa5308930d7eee4d114d89fb4380d6248f\nAuthor: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nDate: Fri Jul 26 17:29:43 2019 +0100\n\n mips: vdso: Fix source path\n \n The vdso library for o32 and n32 does not compile compile correctly when\n building outside of the source tree due to a wrong inclusion path for\n config-n32-o32-env.c resulting in the error below:\n \n cc1: fatal error: arch/mips/vdso/config-n32-o32-env.c:\n No such file or dnirectory\n compilation terminated.\n arch/mips/vdso/Makefile:153: recipe for target\n \u0027arch/mips/vdso/vgettimeofday-o32.o\u0027 failed\n make[3]: *** [arch/mips/vdso/vgettimeofday-o32.o] Error 1\n scripts/Makefile.build:490: recipe for target \u0027arch/mips/vdso\u0027 failed\n \n Fix the config-n32-o32-env.c inclusion path prepending the $(srctree)\n variable.\n \n Signed-off-by: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\n [paul.burton@mips.com: Note that this occurs specifically when building\n outside of the source tree.]\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit aaf70b90a4f196b1abd67f97dace64a60308234e\nAuthor: Mahesh Kumar \u003cmahesh1.kumar@intel.com\u003e\nDate: Fri Jul 12 18:09:21 2019 -0700\n\n drm/i915/tgl: update ddi/tc clock_off bits\n \n In GEN 12 PORT_C DDI clk_off bit is not equally distanced to A/B,\n it\u0027s at offset 24. Similarly TC port (5/6) clk off bits are at\n offset 22/23. Extend the macros to cover the additional ports.\n \n Cc: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Signed-off-by: Mahesh Kumar \u003cmahesh1.kumar@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Reviewed-by: Matt Atwood \u003cmatthew.s.atwood@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190713010940.17711-4-lucas.demarchi@intel.com\n\ncommit df16b6361d888409a9e2f9ae777c8d5dde2e76b8\nAuthor: Mahesh Kumar \u003cmahesh1.kumar@intel.com\u003e\nDate: Fri Jul 12 18:09:20 2019 -0700\n\n drm/i915/tgl: select correct bit for port select\n \n Bit definitions for port-select got changed for TRANS_CLK_SEL \u0026\n TRANS_DDI_FUNC_CTL registers in TGL.\n \n v2 (Lucas):\n - Nuke TRANS_DDI_PORT_NONE since it\u0027s 0: we are already clearing\n {TGL_,}TRANS_DDI_PORT_MASK (suggested by Ville)\n - Also cover haswell_get_ddi_port_state() in intel_display.c that was\n missing\n - Define macros using the _SHIFT macros so we don\u0027t lose other users\n \n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Mahesh Kumar \u003cmahesh1.kumar@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Reviewed-by: Matt Atwood \u003cmatthew.s.atwood@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190713010940.17711-3-lucas.demarchi@intel.com\n\ncommit 98a5c2a3582a86d5acdcdeabbe0e6278e712201e\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Fri Jul 12 18:09:19 2019 -0700\n\n drm/i915/tgl: skip setting PORT_CL_DW12_* on initialization\n \n According to the spec when initializing the display in TGL we should not\n set PORT_CL_DW12 for the Aux channel of the combo PHYs. We will re-use the\n power well hooks from ICL so only set this register on gen \u003c 12.\n \n v2: Generalize check for gen 12 (suggested by José)\n v3: Rebase after enum phy introduction\n \n Cc: Imre Deak \u003cimre.deak@intel.com\u003e\n Cc: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Reviewed-by: Matt Atwood \u003cmatthew.s.atwood@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190713010940.17711-2-lucas.demarchi@intel.com\n\ncommit cd8ae2073963eff4c318d0a1f0e91fc35f0c6a83\nAuthor: Sergej Benilov \u003csergej.benilov@googlemail.com\u003e\nDate: Thu Jul 25 21:48:06 2019 +0200\n\n sis900: add support for ethtool\u0027s EEPROM dump\n \n Implement ethtool\u0027s EEPROM dump command (ethtool -e|--eeprom-dump).\n \n Thx to Andrew Lunn for comments.\n \n Signed-off-by: Sergej Benilov \u003csergej.benilov@googlemail.com\u003e\n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1fbf400b58fa70c35bf671ff640b83799e45388d\nAuthor: David S. Miller \u003cdavem@davemloft.net\u003e\nDate: Fri Jul 26 14:10:30 2019 -0700\n\n staging: octeon: Fix build failure due to typo.\n \n drivers/staging/octeon/ethernet-tx.c:287:23: error: implicit declaration of function \u0027skb_drag_size\u0027; did you mean \u0027skb_frag_size\u0027? [-Werror\u003dimplicit-function-declaration]\n \n From kernelci report:\n \n https://kernelci.org/build/id/5d3943f859b514103f688918/logs/\n \n Fixes: 92493a2f8a8d (\"Build fixes for skb_frag_size conversion\")\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 32630d526b9faaec02fa1d26582d91c01f6d476c\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 11:52:34 2019 -0300\n\n MAINTAINERS: add entries for some documentation scripts\n \n There are some documentation scripts I wrote with doesn\u0027t\n have any maintainer at maintainer\u0027s file.\n \n Add them to the DOCUMENTATION entry, in order to have\n Jon and linux-doc ML c/c on those patches, plus a new\n entry to ensure that I\u0027ll be c/c when people send patches\n to those.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n\ncommit 08f0e4a7ecb9e20f27b9d410bcba84e662fffe3b\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Fri Jul 26 18:42:12 2019 +0000\n\n drm/i915/uc: Remove redundant RSA offset definition\n \n According to Firmware layout definition, RSA signature is located\n after CSS header and uCode so actual RSA offset in the blob can be\n easily calculated when needed (and we need it only once).\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190726184212.1836-3-michal.wajdeczko@intel.com\n\ncommit 5de51fa0b9399e53a352437a9fdbd50a2e5912c5\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Fri Jul 26 18:42:11 2019 +0000\n\n drm/i915/uc: Remove redundant ucode offset definition\n \n According to Firmware layout definition, uCode is located right\n after CSS header, so ucode offset is always same as header size.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190726184212.1836-2-michal.wajdeczko@intel.com\n\ncommit 3a8c63d28a12dec948439dcb1fd0e60066dc3897\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Fri Jul 26 18:42:10 2019 +0000\n\n drm/i915/uc: Remove redundant header_offset/size definitions\n \n According to Firmware layout definition, CSS header is located\n in front of the firmware blob, so header offset is always 0.\n Similarly, size of the CSS header is constant and currently\n used version is exactly 128.\n \n While here, move type/status enums up and keep them together.\n \n v2: use sizeof consistently (Daniele), update commit message\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190726184212.1836-1-michal.wajdeczko@intel.com\n\ncommit 63b87c310a8326c2d0721ab6c981493d590aa542\nAuthor: Sean Paul \u003cseanpaul@chromium.org\u003e\nDate: Fri Jul 26 10:20:47 2019 -0400\n\n drm/mst: Fix sphinx warnings in drm_dp_msg_connector register functions\n \n Fixes the following warnings:\n \n ../drivers/gpu/drm/drm_dp_mst_topology.c:1593: warning: Excess function parameter \u0027drm_connector\u0027 description in \u0027drm_dp_mst_connector_late_register\u0027\n ../drivers/gpu/drm/drm_dp_mst_topology.c:1613: warning: Excess function parameter \u0027drm_connector\u0027 description in \u0027drm_dp_mst_connector_early_unregister\u0027\n ../drivers/gpu/drm/drm_dp_mst_topology.c:1594: warning: Function parameter or member \u0027connector\u0027 not described in \u0027drm_dp_mst_connector_late_register\u0027\n ../drivers/gpu/drm/drm_dp_mst_topology.c:1614: warning: Function parameter or member \u0027connector\u0027 not described in \u0027drm_dp_mst_connector_early_unregister\u0027\n \n Fixes: 562836a269e3 (\"drm/dp_mst: Enable registration of AUX devices for MST ports\")\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Leo Li \u003csunpeng.li@amd.com\u003e\n Cc: Lyude Paul \u003clyude@redhat.com\u003e\n Cc: Harry Wentland \u003charry.wentland@amd.com\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: dri-devel@lists.freedesktop.org\n Reviewed-by: Lyude Paul \u003clyude@redhat.com\u003e\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190726142057.224121-1-sean@poorly.run\n\ncommit c87fb38df19da3362a0e20df1aad852100995ead\nAuthor: Sean Paul \u003cseanpaul@chromium.org\u003e\nDate: Tue Jun 11 12:08:25 2019 -0400\n\n drm/rockchip: Use drm_atomic_helper_commit_tail_rpm\n \n Now that we use the drm psr helpers, we no longer need to hand-roll our\n atomic_commit_tail implementation. So use the helper\n \n Changes in v2:\n - None\n Changes in v3:\n - None\n Changes in v4:\n - None\n Changes in v5:\n - None\n \n Link to v1: https://patchwork.freedesktop.org/patch/msgid/20190228210939.83386-6-sean@poorly.run\n Link to v2: https://patchwork.freedesktop.org/patch/msgid/20190326204509.96515-5-sean@poorly.run\n Link to v3: https://patchwork.freedesktop.org/patch/msgid/20190502194956.218441-11-sean@poorly.run\n Link to v4: https://patchwork.freedesktop.org/patch/msgid/20190508160920.144739-12-sean@poorly.run\n \n Cc: Zain Wang \u003cwzz@rock-chips.com\u003e\n Cc: Tomasz Figa \u003ctfiga@chromium.org\u003e\n Tested-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Reviewed-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190611160844.257498-12-sean@poorly.run\n\ncommit bed030a49f3ece8af221debe4e435b282c4d58e5\nAuthor: Sean Paul \u003cseanpaul@chromium.org\u003e\nDate: Tue Jun 11 12:08:24 2019 -0400\n\n drm/rockchip: Don\u0027t fully disable vop on self refresh\n \n Instead of fully disabling and re-enabling the vop on self refresh\n transitions, only disable the active windows. This will speed up\n self refresh exits substantially and is still a power-savings win.\n \n This patch integrates portions of Zain\u0027s patch from here:\n https://patchwork.kernel.org/patch/9615063/\n \n Changes in v2:\n - None\n Changes in v3:\n - None\n Changes in v4:\n - Adjust for preceding vop_win_disable changes\n Changes in v5:\n - None\n \n Link to v1: https://patchwork.freedesktop.org/patch/msgid/20190228210939.83386-5-sean@poorly.run\n Link to v2: https://patchwork.freedesktop.org/patch/msgid/20190326204509.96515-4-sean@poorly.run\n Link to v3: https://patchwork.freedesktop.org/patch/msgid/20190502194956.218441-10-sean@poorly.run\n Link to v4: https://patchwork.freedesktop.org/patch/msgid/20190508160920.144739-11-sean@poorly.run\n \n Cc: Zain Wang \u003cwzz@rock-chips.com\u003e\n Cc: Tomasz Figa \u003ctfiga@chromium.org\u003e\n Cc: Kristian H. Kristensen \u003choegsberg@chromium.org\u003e\n Tested-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Reviewed-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190611160844.257498-11-sean@poorly.run\n\ncommit 2b60e11d0babddc0d5eec0d13abfd2cd5a2f29f0\nAuthor: Sean Paul \u003cseanpaul@chromium.org\u003e\nDate: Tue Jun 11 12:08:23 2019 -0400\n\n drm/rockchip: Use vop_win in vop_win_disable instead of vop_win_data\n \n Change the argument to vop_win_disable to vop_win to accomodate future\n changes to the function.\n \n Changes in v4:\n - Added to the patchset\n Changes in v5:\n - None\n \n Link to v4: https://patchwork.freedesktop.org/patch/msgid/20190508160920.144739-10-sean@poorly.run\n \n Tested-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Reviewed-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190611160844.257498-10-sean@poorly.run\n\ncommit 6c836d965bad5e13803256e3d316b50248531f4e\nAuthor: Sean Paul \u003cseanpaul@chromium.org\u003e\nDate: Tue Jun 11 12:08:22 2019 -0400\n\n drm/rockchip: Use the helpers for PSR\n \n Instead of rolling our own implementation for tracking when PSR should\n be [in]active, use the new self refresh helpers to do the heavy lifting.\n \n Changes in v2:\n - updated to reflect changes made in the helpers\n Changes in v3:\n - use the new atomic hooks to inspect crtc state instead of needing conn state (Daniel)\n Changes in v4:\n - Use Laurent\u0027s get_new_connector_for_encoder helper (Daniel)\n - Exit vop disable early if it\u0027s already off\n Changes in v5:\n - Rebase on latest drm-misc-next\n - Resolve conflict with s/edp_vsc_psr/dp_sdp/ rename\n - Resolve conflict with drm_atomic.h header inclusion\n \n Link to v1: https://patchwork.freedesktop.org/patch/msgid/20190228210939.83386-4-sean@poorly.run\n Link to v2: https://patchwork.freedesktop.org/patch/msgid/20190326204509.96515-3-sean@poorly.run\n Link to v3: https://patchwork.freedesktop.org/patch/msgid/20190502194956.218441-9-sean@poorly.run\n Link to v4: https://patchwork.freedesktop.org/patch/msgid/20190508160920.144739-9-sean@poorly.run\n \n Cc: Zain Wang \u003cwzz@rock-chips.com\u003e\n Cc: Tomasz Figa \u003ctfiga@chromium.org\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Tested-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Reviewed-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n [seanpaul resolved some conflicts with drmP.h work and Helen\u0027s async fixes]\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190611160844.257498-9-sean@poorly.run\n\ncommit ad309284a52be47c8b3126c9376358bf381861bc\nAuthor: Sean Paul \u003cseanpaul@chromium.org\u003e\nDate: Tue Jun 11 12:08:21 2019 -0400\n\n drm/rockchip: Check for fast link training before enabling psr\n \n Once we start shutting off the link during PSR, we\u0027re going to want fast\n training to work. If the display doesn\u0027t support fast training, don\u0027t\n enable psr.\n \n Changes in v2:\n - None\n Changes in v3:\n - None\n Changes in v4:\n - None\n Changes in v5:\n - None\n \n Link to v1: https://patchwork.freedesktop.org/patch/msgid/20190228210939.83386-3-sean@poorly.run\n Link to v2: https://patchwork.freedesktop.org/patch/msgid/20190326204509.96515-2-sean@poorly.run\n Link to v3: https://patchwork.freedesktop.org/patch/msgid/20190502194956.218441-9-sean@poorly.run\n Link to v4: https://patchwork.freedesktop.org/patch/msgid/20190508160920.144739-8-sean@poorly.run\n \n Cc: Zain Wang \u003cwzz@rock-chips.com\u003e\n Cc: Tomasz Figa \u003ctfiga@chromium.org\u003e\n Tested-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Reviewed-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190611160844.257498-8-sean@poorly.run\n\ncommit 82b78cad0c810caaa5237dfa7bc088e54d8102b7\nAuthor: Guido Günther \u003cagx@sigxcpu.org\u003e\nDate: Fri Jul 26 15:14:39 2019 +0200\n\n drm/panel: jh057n00900: Use drm_panel_{unprepare, disable} consistently\n \n We were already using the generic functions in our debugfs code, do the\n same in jh057n_shutdown. This was suggested by Sam Ravnborg.\n \n Signed-off-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/a37dd5083462064f437ff62fd84e6576d8a7c8dc.1564146727.git.agx@sigxcpu.org\n\ncommit 04eedeb46fa2e961baa1e9386b16dd46289f7b69\nAuthor: Guido Günther \u003cagx@sigxcpu.org\u003e\nDate: Fri Jul 26 15:14:38 2019 +0200\n\n drm/panel: jh057n00900: Print error code on all DRM_DEV_ERROR()s\n \n Most of them had these already but two mere missing. This eases\n debugging.\n \n Signed-off-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/6b237a570cb368dc4471fb8feb3a0441813cd576.1564146727.git.agx@sigxcpu.org\n\ncommit 6a6fc457b30f12a0ea781e2c0bbea3bff0b5e71c\nAuthor: Guido Günther \u003cagx@sigxcpu.org\u003e\nDate: Fri Jul 26 15:14:37 2019 +0200\n\n drm/panel: jh057n00900: Move mipi_dsi_dcs_set_display_off to disable()\n \n This makes it symmetric with the panel init happening in enable().\n \n Signed-off-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/e2c31d34ce1f917065d590297e5115a4ca954874.1564146727.git.agx@sigxcpu.org\n\ncommit 66a4e0ef4343bf03ce50d5813428bc0f2bc5ff36\nAuthor: Guido Günther \u003cagx@sigxcpu.org\u003e\nDate: Fri Jul 26 15:14:36 2019 +0200\n\n drm/panel: jh057n00900: Move panel DSI init to enable()\n \n If the panel is wrapped in a panel_bridge it gets prepar()ed before the\n upstream DSI bridge which can cause hangs (e.g. with imx-nwl since clocks\n are not enabled yet). To avoid this move the panel\u0027s first DSI access to\n enable() so the upstream bridge can prepare the DSI host controller in\n it\u0027s pre_enable().\n \n This is also in line with other panel drivers.\n \n Signed-off-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/12c3495b234952aafe11980a9e06cfd246134660.1564146727.git.agx@sigxcpu.org\n\ncommit 340c4c8daad95cc1064d88901a03bcb68d0f0185\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Jul 25 14:14:46 2019 +0100\n\n drm/i915/gt: Add to timeline requires the timeline mutex\n \n Modifying a remote context requires careful serialisation with requests\n on that context, and that serialisation requires us to take their\n timeline-\u003emutex. Make it so.\n \n Note that while struct_mutex rules, we can\u0027t create more than one\n request in parallel, but that age is soon coming to an end.\n \n v2: Though it doesn\u0027t affect the current users, contexts may share\n timelines so check if we already hold the right mutex.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725131447.27515-1-chris@chris-wilson.co.uk\n\ncommit f91bf7382856eb9b35664f92aab4c49b0407ac02\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Thu Jul 25 20:51:06 2019 +0000\n\n drm/i915/uc: Don\u0027t sanitize guc_log_level modparam\n \n We are already storing runtime value of log level in private\n field, so there is no need to modify modparam.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725205106.36148-1-michal.wajdeczko@intel.com\n\ncommit 3f99a61441e5d44b6dc00461a74d759c78191682\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Thu Jul 25 13:50:56 2019 +0100\n\n drm/i915: Do not rely on for loop caching the mask\n \n for_each_engine_masked caches the engine mask but what does the caller\n know.\n \n Cache it explicitly for clarity and while at it correct the type to match.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725125056.11942-1-tvrtko.ursulin@linux.intel.com\n\ncommit 096769ea95380ef3048765aef676e8947bc368d4\nAuthor: Amadeusz Sławiński \u003camadeuszx.slawinski@linux.intel.com\u003e\nDate: Fri Jul 26 11:09:29 2019 +0200\n\n ASoC: Intel: Skylake: Remove static table index when parsing topology\n \n Currently when we remove and reload driver we use previous ref_count\n value to start iterating over skl-\u003emodules which leads to out of table\n access. To fix this just inline the function and calculate indexes\n everytime we parse UUID token.\n \n Signed-off-by: Amadeusz Sławiński \u003camadeuszx.slawinski@linux.intel.com\u003e\n Reviewed-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190726090929.27946-2-amadeuszx.slawinski@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c8415833ec242b9ddf73bf9e1057e12f9b0fcd16\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Jul 26 13:33:27 2019 +0100\n\n ASoC: codec2codec: fix missing return of error return code\n \n Currently in function snd_soc_dai_link_event_pre_pmu the error return\n code in variable err is being set but this is not actually being returned,\n the function just returns zero even when there are failures. Fix this by\n returning the error return code.\n \n Addresses-Coverity: (\"Unused value\")\n Fixes: 3dcfb397dad2 (\"ASoC: codec2codec: deal with params when necessary\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190726123327.10467-1-colin.king@canonical.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit cc374377a19d2a49d693997b62dc3a6f5fac6d61\nAuthor: Ralph Campbell \u003crcampbell@nvidia.com\u003e\nDate: Thu Jul 25 17:56:50 2019 -0700\n\n mm/hmm: remove hmm_range vma\n \n Since hmm_range_fault() doesn\u0027t use the struct hmm_range vma field, remove\n it.\n \n Link: https://lore.kernel.org/r/20190726005650.2566-8-rcampbell@nvidia.com\n Suggested-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit f527688d5d8a80d2d1b2c02779105747c2f4f705\nAuthor: Ralph Campbell \u003crcampbell@nvidia.com\u003e\nDate: Thu Jul 25 17:56:49 2019 -0700\n\n mm/hmm: remove hugetlbfs check in hmm_vma_walk_pmd\n \n walk_page_range() will only call hmm_vma_walk_hugetlb_entry() for\n hugetlbfs pages and doesn\u0027t call hmm_vma_walk_pmd() in this case.\n \n Therefore, it is safe to remove the check for vma-\u003evm_flags \u0026 VM_HUGETLB\n in hmm_vma_walk_pmd().\n \n Link: https://lore.kernel.org/r/20190726005650.2566-7-rcampbell@nvidia.com\n Signed-off-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 4a6e4ae254e0ad5940067b4576a94ca51d6cea9e\nAuthor: Douglas Anderson \u003cdianders@chromium.org\u003e\nDate: Mon Jul 22 11:24:39 2019 -0700\n\n video: amba-clcd: Spout an error if of_get_display_timing() gives an error\n \n In the patch (\"video: of: display_timing: Don\u0027t yell if no timing node\n is present\") we\u0027ll stop spouting an error directly in\n of_get_display_timing() if no node is present. Presumably amba-clcd\n should take charge of spouting its own error now.\n \n NOTE: we\u0027ll print two errors if the node was present but there were\n problems parsing the timing node (one in of_parse_display_timing() and\n this new one). Since this is a fatal error for the driver\u0027s probe\n (and presumably someone will be debugging), this should be OK.\n \n Signed-off-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: Russell King \u003clinux@armlinux.org.uk\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190722182439.44844-5-dianders@chromium.org\n\ncommit c2d7e47ad190189213b3e01956a6b5055ba8de24\nAuthor: Douglas Anderson \u003cdianders@chromium.org\u003e\nDate: Mon Jul 22 11:24:38 2019 -0700\n\n drm: panel-lvds: Spout an error if of_get_display_timing() gives an error\n \n In the patch (\"video: of: display_timing: Don\u0027t yell if no timing node\n is present\") we\u0027ll stop spouting an error directly in\n of_get_display_timing() if no node is present. Presumably panel-lvds\n should take charge of spouting its own error now.\n \n NOTE: we\u0027ll print two errors if the node was present but there were\n problems parsing the timing node (one in of_parse_display_timing() and\n this new one). Since this is a fatal error for the driver\u0027s probe\n (and presumably someone will be debugging), this should be OK.\n \n Signed-off-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: Russell King \u003clinux@armlinux.org.uk\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190722182439.44844-4-dianders@chromium.org\n\ncommit 892e8ba38f28d68a17122a9e14cf36ea756aa6b8\nAuthor: Douglas Anderson \u003cdianders@chromium.org\u003e\nDate: Mon Jul 22 11:24:37 2019 -0700\n\n video: of: display_timing: Don\u0027t yell if no timing node is present\n \n There may be cases (like in panel-simple.c) where we have a sane\n fallback if no timings are specified in the device tree. Let\u0027s get\n rid of the unconditional pr_err(). We can add error messages in\n individual drivers if it makes sense.\n \n NOTE: we\u0027ll still print errors if the node is present but there are\n problems parsing the timings.\n \n Fixes: b8a2948fa2b3 (\"drm/panel: simple: Add ability to override typical timing\")\n Reported-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: Russell King \u003clinux@armlinux.org.uk\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190722182439.44844-3-dianders@chromium.org\n\ncommit 4faba50edbcc1df467f8f308893edc3fdd95536e\nAuthor: Douglas Anderson \u003cdianders@chromium.org\u003e\nDate: Mon Jul 22 11:24:36 2019 -0700\n\n video: of: display_timing: Add of_node_put() in of_get_display_timing()\n \n From code inspection it can be seen that of_get_display_timing() is\n lacking an of_node_put(). Add it.\n \n Fixes: ffa3fd21de8a (\"videomode: implement public of_get_display_timing()\")\n Signed-off-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: Russell King \u003clinux@armlinux.org.uk\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190722182439.44844-2-dianders@chromium.org\n\ncommit ae6fa4d5e94ea520506b691140ebcb5dc6bf0a17\nAuthor: Denis Kenzior \u003cdenkenz@gmail.com\u003e\nDate: Mon Jul 22 06:33:12 2019 -0500\n\n nl80211: Include wiphy address setup in NEW_WIPHY\n \n Include wiphy address setup in wiphy dumps and new wiphy events. The\n wiphy permanent address is exposed as ATTR_MAC. If addr_mask is setup,\n then it is included as ATTR_MAC_MASK attribute. If multiple addresses\n are available, then their are exposed in a nested ATTR_MAC_ADDRS array.\n \n This information is already exposed via sysfs, but it makes sense to\n include it in the wiphy dump as well.\n \n Signed-off-by: Denis Kenzior \u003cdenkenz@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190722113312.14031-3-denkenz@gmail.com\n [use just nla_nest_start(), this is new functionality]\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 7a113110fc8cdda14023c0bffc7bd8b5f3da1edf\nAuthor: Denis Kenzior \u003cdenkenz@gmail.com\u003e\nDate: Mon Jul 22 06:33:10 2019 -0500\n\n nl80211: document uapi for CMD_FRAME_WAIT_CANCEL\n \n Commit 1c38c7f22068 (\"nl80211: send event when CMD_FRAME duration\n expires\") added the possibility of NL80211_CMD_FRAME_WAIT_CANCEL\n being sent whenever the off-channel wait time associated with a\n CMD_FRAME completes. Document this in the uapi/linux/nl80211.h file.\n \n Signed-off-by: Denis Kenzior \u003cdenkenz@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190722113312.14031-1-denkenz@gmail.com\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 60ad72da55ac74a67d0eae5fb57327d7b4967786\nAuthor: Sven Eckelmann \u003cseckelmann@datto.com\u003e\nDate: Wed Jul 24 18:33:56 2019 +0200\n\n mac80211: implement HE support for mesh\n \n Implement the basics required for supporting high efficiency with mesh:\n include HE information elements in beacons, probe responses, and peering\n action frames, and check for compatible HE configurations when peering.\n \n Signed-off-by: Sven Eckelmann \u003cseckelmann@datto.com\u003e\n \n Forwarded: https://patchwork.kernel.org/patch/11029299/\n Link: https://lore.kernel.org/r/20190724163359.3507-2-sven@narfation.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit a0b4496a43681cbeec03a38e1b685c80c0d7405d\nAuthor: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\nDate: Tue Jul 16 00:09:19 2019 +0200\n\n mac80211: add IEEE80211_KEY_FLAG_GENERATE_MMIE to ieee80211_key_flags\n \n Add IEEE80211_KEY_FLAG_GENERATE_MMIE flag to ieee80211_key_flags in order\n to allow the driver to notify mac80211 to generate MMIE and that it\n requires sequence number generation only.\n This is a preliminary patch to add BIP_CMAC_128 hw support to mt7615\n driver\n \n Signed-off-by: Lorenzo Bianconi \u003clorenzo@kernel.org\u003e\n Link: https://lore.kernel.org/r/dfe275f9aa0f1cc6b33085f9efd5d8447f68ad13.1563228405.git.lorenzo@kernel.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit ef11a931bd1c57b02fe2603ff95a392a73041f9e\nAuthor: John Crispin \u003cjohn@phrozen.org\u003e\nDate: Tue Jun 18 08:19:14 2019 +0200\n\n mac80211: HE: add Spatial Reuse element parsing support\n \n Add support to mac80211 for parsing SPR elements as per\n P802.11ax_D4.0 section 9.4.2.241.\n \n Signed-off-by: Shashidhar Lakkavalli \u003cslakkavalli@datto.com\u003e\n Signed-off-by: John Crispin \u003cjohn@phrozen.org\u003e\n Link: https://lore.kernel.org/r/20190618061915.7102-2-john@phrozen.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 3d07ffcaf32006486f8743ef35c4706d4c776661\nAuthor: John Crispin \u003cjohn@phrozen.org\u003e\nDate: Sun Jul 14 17:44:16 2019 +0200\n\n mac80211: add struct ieee80211_tx_status support to ieee80211_add_tx_radiotap_header\n \n Add support to ieee80211_add_tx_radiotap_header() for handling rates\n reported via ieee80211_tx_status. This allows us to also report HE rates.\n \n Signed-off-by: John Crispin \u003cjohn@phrozen.org\u003e\n Link: https://lore.kernel.org/r/20190714154419.11854-4-john@phrozen.org\n [remove text about 60 GHz, mac80211 doesn\u0027t support it, fix endianness issue]\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 5cca503817d0e96f8d902c6598d5105dca5e718a\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 26 14:14:58 2019 +0100\n\n drm/i915/perf: Initialise err to 0 before looping over ce-\u003eengines\n \n Smatch warning that the loop may be empty causing us to check err before\n it had been set. Ensure that it is initialised to 0, just in case.\n \n v2: Refactor the inner loop for better scooping and clarity\n \n Fixes: a9877da2d629 (\"drm/i915/oa: Reconfigure contexts on the fly\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190726131458.8310-1-chris@chris-wilson.co.uk\n\ncommit d45d464b118f428229d91769c8a3cc1e2e0bb4d5\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Thu Jul 25 17:56:47 2019 -0700\n\n mm/hmm: merge hmm_range_snapshot into hmm_range_fault\n \n Add a HMM_FAULT_SNAPSHOT flag so that hmm_range_snapshot can be merged\n into the almost identical hmm_range_fault function.\n \n Link: https://lore.kernel.org/r/20190726005650.2566-5-rcampbell@nvidia.com\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 9a4903e49e495bfd2650862dfae4178bebe4db9c\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Thu Jul 25 17:56:46 2019 -0700\n\n mm/hmm: replace the block argument to hmm_range_fault with a flags value\n \n This allows easier expansion to other flags, and also makes the callers a\n little easier to read.\n \n Link: https://lore.kernel.org/r/20190726005650.2566-4-rcampbell@nvidia.com\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Signed-off-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit d2e8d551165ccb6669a0d0273be01ac99d61deba\nAuthor: Ralph Campbell \u003crcampbell@nvidia.com\u003e\nDate: Thu Jul 25 17:56:45 2019 -0700\n\n mm/hmm: a few more C style and comment clean ups\n \n A few more comments and minor programming style clean ups. There should\n be no functional changes.\n \n Link: https://lore.kernel.org/r/20190726005650.2566-3-rcampbell@nvidia.com\n Signed-off-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Reviewed-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 1f961807925032daa90267d8a23ff730e7ede07a\nAuthor: Ralph Campbell \u003crcampbell@nvidia.com\u003e\nDate: Thu Jul 25 17:56:44 2019 -0700\n\n mm/hmm: replace hmm_update with mmu_notifier_range\n \n The hmm_mirror_ops callback function sync_cpu_device_pagetables() passes a\n struct hmm_update which is a simplified version of struct\n mmu_notifier_range. This is unnecessary so replace hmm_update with\n mmu_notifier_range directly.\n \n Link: https://lore.kernel.org/r/20190726005650.2566-2-rcampbell@nvidia.com\n Signed-off-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Reviewed: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n [jgg: white space tuning]\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 2127c01b7f63b06a21559f56a8c81a3c6535bd1a\nAuthor: Jia-Ju Bai \u003cbaijiaju1990@gmail.com\u003e\nDate: Fri Jul 26 10:14:42 2019 +0800\n\n ALSA: i2c: ak4xxx-adda: Fix a possible null pointer dereference in build_adc_controls()\n \n In build_adc_controls(), there is an if statement on line 773 to check\n whether ak-\u003eadc_info is NULL:\n if (! ak-\u003eadc_info ||\n ! ak-\u003eadc_info[mixer_ch].switch_name)\n \n When ak-\u003eadc_info is NULL, it is used on line 792:\n knew.name \u003d ak-\u003eadc_info[mixer_ch].selector_name;\n \n Thus, a possible null-pointer dereference may occur.\n \n To fix this bug, referring to lines 773 and 774, ak-\u003eadc_info\n and ak-\u003eadc_info[mixer_ch].selector_name are checked before being used.\n \n This bug is found by a static analysis tool STCheck written by us.\n \n Signed-off-by: Jia-Ju Bai \u003cbaijiaju1990@gmail.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit dd65f7e19c6961ba6a69f7c925021b7a270cb950\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Fri Jul 26 11:42:34 2019 +0200\n\n ALSA: hda - Show the fatal CORB/RIRB error more clearly\n \n The last fallback of CORB/RIRB communication error recovery is to turn\n on the single command mode, and this last resort usually means that\n something is really screwed up. Instead of a normal dev_err(), show\n the error more clearly with dev_WARN() with the caller stack trace.\n \n Also, show the bus-reset fallback also as an error, too.\n \n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit cd5e1cbe1f0af7ab02b1b2e08c9d34f9158cedf4\nAuthor: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\nDate: Mon Jul 8 09:58:11 2019 -0700\n\n drm/panel: simple: Add support for Sharp LD-D5116Z01B panel\n \n The Sharp LD-D5116Z01B is a 12.3\" eDP panel with a 1920X1280 resolution.\n \n Signed-off-by: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190708165811.46370-1-jeffrey.l.hugo@gmail.com\n\ncommit d8481155a3219ef427c6384022931758fbbe8ebe\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Fri Jul 26 08:47:27 2019 -0300\n\n ASoC: sunxi: fix a sound binding broken reference\n \n Address this rename:\n Documentation/devicetree/bindings/sound/{sun4i-i2s.txt -\u003e allwinner,sun4i-a10-i2s.yaml}\n \n Fixes: 0a0ca8e94ca3 (\"dt-bindings: sound: Convert Allwinner I2S binding to YAML\")\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Link: https://lore.kernel.org/r/9932608f32030c886d906ea656eda8408c544776.1564140865.git.mchehab+samsung@kernel.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 2448c813b12cf1fb08da1d20bdb2ae583cda2d6f\nAuthor: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\nDate: Fri Jul 26 09:42:44 2019 +0300\n\n ASoC: ti: davinci-mcasp: Support for correct symmetric sample bits\n \n Implement custom snd_pcm_hw_rule to filter the available formats for the\n second stream to make it symmetric and allow only formats which require\n the same amount of bits on the bus as the running stream.\n \n A simple constraint is not working correctly because for example:\n the first stream is started with S24_LE\n \n If we place 24 as constraint for the SAMPLE_BITS then the second stream\n can not use S24_LE as it is physically 32bits.\n \n If we would place 32 as constraint (physical width) then S32_LE would have\n been allowed, but S24_3LE is not.\n \n Signed-off-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Link: https://lore.kernel.org/r/20190726064244.3762-3-peter.ujfalusi@ti.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 2a03e3a50a61494fd26d82eabd43a26be0f736cc\nAuthor: Hook, Gary \u003cGary.Hook@amd.com\u003e\nDate: Wed Jul 10 21:45:37 2019 +0000\n\n crypto: ccp - Include the module name in system log messages\n \n Redefine pr_fmt so that the module name is prefixed to every\n log message produced by the ccp-crypto module\n \n Signed-off-by: Gary R Hook \u003cgary.hook@amd.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 4f22f32356629ab2843e873887d34868b96a82b4\nAuthor: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\nDate: Wed Jul 10 13:32:18 2019 +0800\n\n crypto: Remove orphan tools/crypto directory\n \n The directory tools/crypto and the only file under it never gets\n built anywhere. This program should instead be incorporated into\n one of the existing user-space projects, crconf or libkcapi.\n \n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit a7268c4d4205b7f92eb77ac99d64099152b0f738\nAuthor: Phani Kiran Hemadri \u003cphemadri@marvell.com\u003e\nDate: Tue Jul 9 15:24:24 2019 +0000\n\n crypto: cavium/nitrox - Add support for loading asymmetric crypto firmware\n \n This patch adds support to load Asymmetric crypto firmware on\n AE cores of CNN55XX device. Firmware is stored on UCD block 2\n and all available AE cores are tagged to group 0.\n \n Signed-off-by: Phani Kiran Hemadri \u003cphemadri@marvell.com\u003e\n Reviewed-by: Srikanth Jampala \u003cjsrikanth@marvell.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit a7c2647034fa34e03d3f4e2cf66315440db2aa14\nAuthor: Hook, Gary \u003cGary.Hook@amd.com\u003e\nDate: Tue Jul 9 15:07:35 2019 +0000\n\n crypto: ccp - Add a module parameter to control registration for DMA\n \n The CCP driver is able to act as a DMA engine. Add a module parameter that\n allows this feature to be enabled/disabled.\n \n Signed-off-by: Gary R Hook \u003cgary.hook@amd.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 90773bc1ab407bc2198d7e9fc1c6d6b5ef27a24f\nAuthor: Hook, Gary \u003cGary.Hook@amd.com\u003e\nDate: Tue Jul 9 15:07:29 2019 +0000\n\n crypto: ccp - module parameter to limit the number of enabled CCPs\n \n Provide the ability to constrain the total number of enabled devices in\n the system. Once max_devs devices have been configured, subsequently\n probed devices are ignored.\n \n The max_devs parameter may be zero, in which case all CCPs are disabled.\n PSPs are always enabled and active.\n \n Disabling the CCPs also disables DMA and RNG registration.\n \n Signed-off-by: Gary R Hook \u003cgary.hook@amd.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit c4a89279086e9c98eba659d7665b1732e0d5e3e3\nAuthor: Hook, Gary \u003cGary.Hook@amd.com\u003e\nDate: Tue Jul 9 15:07:22 2019 +0000\n\n crypto: ccp - Add a module parameter to specify a queue count\n \n Add a module parameter to limit the number of queues per CCP. The default\n value (nqueues\u003d0) is to set up every available queue on each device.\n \n The count of queues starts from the first one found on the device (which\n varies based on the device ID).\n \n Signed-off-by: Gary R Hook \u003cgary.hook@amd.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 93308baf075039dccf347391491773f272cc24f6\nAuthor: Hook, Gary \u003cGary.Hook@amd.com\u003e\nDate: Tue Jul 9 15:07:15 2019 +0000\n\n crypto: ccp - Make CCP debugfs support optional\n \n Add a config option to exclude DebugFS support in the CCP driver.\n \n Signed-off-by: Gary R Hook \u003cgary.hook@amd.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 91b05a7e7d8033a90a64f5fc0e3808db423e420a\nAuthor: Ondrej Mosnacek \u003comosnace@redhat.com\u003e\nDate: Tue Jul 9 13:11:24 2019 +0200\n\n crypto: user - make NETLINK_CRYPTO work inside netns\n \n Currently, NETLINK_CRYPTO works only in the init network namespace. It\n doesn\u0027t make much sense to cut it out of the other network namespaces,\n so do the minor plumbing work necessary to make it work in any network\n namespace. Code inspired by net/core/sock_diag.c.\n \n Tested using kcapi-dgst from libkcapi [1]:\n Before:\n # unshare -n kcapi-dgst -c sha256 \u003c/dev/null | wc -c\n libkcapi - Error: Netlink error: sendmsg failed\n libkcapi - Error: Netlink error: sendmsg failed\n libkcapi - Error: NETLINK_CRYPTO: cannot obtain cipher information for hmac(sha512) (is required crypto_user.c patch missing? see documentation)\n 0\n \n After:\n # unshare -n kcapi-dgst -c sha256 \u003c/dev/null | wc -c\n 32\n \n [1] https://github.com/smuellerDD/libkcapi\n \n Signed-off-by: Ondrej Mosnacek \u003comosnace@redhat.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 31fb084c4eb556fac30115d99bc518c7944887d2\nAuthor: Pascal van Leeuwen \u003cpascalvanl@gmail.com\u003e\nDate: Fri Jul 5 09:36:31 2019 +0200\n\n crypto: inside-secure -reduce hash byte counters to 64 bits\n \n This patch recognises the fact that the hardware cannot ever process more\n than 2,199,023,386,111 bytes of hash or HMAC payload, so there is no point\n in maintaining 128 bit wide byte counters, 64 bits is more than sufficient\n \n Signed-off-by: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 0e17e3621a28a67871a8c69f5fccdf7d009fd939\nAuthor: Pascal van Leeuwen \u003cpascalvanl@gmail.com\u003e\nDate: Fri Jul 5 08:49:24 2019 +0200\n\n crypto: inside-secure - add support for authenc(hmac(sha*),rfc3686(ctr(aes))) suites\n \n This patch adds support for the following AEAD ciphersuites:\n - authenc(hmac(sha1),rfc3686(ctr(aes)))\n - authenc(hmac(sha224),rfc3686(ctr(aes)))\n - authenc(hmac(sha256),rfc3686(ctr(aes)))\n - authenc(hmac(sha384),rfc3686(ctr(aes)))\n - authenc(hmac(sha512),rfc3686(ctr(aes)))\n \n Signed-off-by: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 54f9e8fa6668366620655df4e6ce5e7b9bf0de8c\nAuthor: Pascal van Leeuwen \u003cpascalvanl@gmail.com\u003e\nDate: Fri Jul 5 08:49:23 2019 +0200\n\n crypto: inside-secure - added support for rfc3686(ctr(aes))\n \n Signed-off-by: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 85266d751551aa5c5c08a2145873f82e0c9ee04f\nAuthor: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\nDate: Mon Jul 8 09:57:53 2019 -0700\n\n dt-bindings: panel: Add Sharp LD-D5116Z01B\n \n The Sharp LD-D5116Z01B is a 12.3\" eDP panel with a 1920X1280 resolution.\n \n Signed-off-by: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190708165753.46275-1-jeffrey.l.hugo@gmail.com\n\ncommit afd6d4f5a52c16e1483328ac074abb1cde92c29f\nAuthor: Navid Emamdoost \u003cnavid.emamdoost@gmail.com\u003e\nDate: Wed Jul 24 14:55:34 2019 -0500\n\n drm/panel: check failure cases in the probe func\n \n The following function calls may fail and return NULL, so the null check\n is added.\n of_graph_get_next_endpoint\n of_graph_get_remote_port_parent\n of_graph_get_remote_port\n \n Update: Thanks to Sam Ravnborg, for suggession on the use of goto to avoid\n leaking endpoint.\n \n Signed-off-by: Navid Emamdoost \u003cnavid.emamdoost@gmail.com\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190724195534.9303-1-navid.emamdoost@gmail.com\n\ncommit e8547e12d5d6a40a16263cc8678d7025e54c379a\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Fri Jul 12 22:59:39 2019 +0200\n\n ARM: dts: gemini: Mount root from mtdblock3\n \n The third mtdblock device named \"Application\" is where we\n want to mount our root filesystem.\n \n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit b5a923f8c7395bea1e4b9f1de75a8ed76978e33f\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Fri Jul 12 22:57:11 2019 +0200\n\n ARM: dts: gemini: Switch to redboot partition parsing\n \n This switches the kernel to parse the Redboot partitions\n in the SL93512r and the NAS4220B properly using the\n right compatible string instead of using hard-coded\n partitions.\n \n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 47ef78b09952eb8498457753f4ba351a8366b80f\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Sat Jul 13 11:52:29 2019 +0200\n\n ARM: dts: gemini: Fix up confused pin settings\n \n The SL93512r board has its pin muxing set up for the wrong\n ASIC: SL3516 instead of SL3512 that it is using. Fix it\n up and reference the right GPIO for the WPS button.\n \n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit b7b2e8caa08c30d3ad2dcdb2133991b44db2913d\nAuthor: John Crispin \u003cjohn@phrozen.org\u003e\nDate: Sun Jul 14 17:44:15 2019 +0200\n\n mac80211: propagate struct ieee80211_tx_status into ieee80211_tx_monitor()\n \n This will allow use to report HE rates on the radiotap interface.\n \n Signed-off-by: John Crispin \u003cjohn@phrozen.org\u003e\n Link: https://lore.kernel.org/r/20190714154419.11854-3-john@phrozen.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit cbe77dde4757446bbe333299b0c91d48b8d575a2\nAuthor: John Crispin \u003cjohn@phrozen.org\u003e\nDate: Sun Jul 14 17:44:14 2019 +0200\n\n mac80211: add xmit rate to struct ieee80211_tx_status\n \n Right now struct ieee80211_tx_rate cannot hold HE rates. Lets use\n struct ieee80211_tx_status instead. This will also make the code\n future-proof for when we have EHT.\n \n Signed-off-by: John Crispin \u003cjohn@phrozen.org\u003e\n Link: https://lore.kernel.org/r/20190714154419.11854-2-john@phrozen.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 2aa485e1148557215337731b2c79f5569edcbbab\nAuthor: John Crispin \u003cjohn@phrozen.org\u003e\nDate: Sat Jul 13 18:36:41 2019 +0200\n\n mac80211: add support for parsing ADDBA_EXT IEs\n \n ADDBA_EXT IEs can be used to negotiate the BA fragmentation level.\n \n Signed-off-by: John Crispin \u003cjohn@phrozen.org\u003e\n Link: https://lore.kernel.org/r/20190713163642.18491-2-john@phrozen.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 60d7dfea00e14d87bdfd94cb7cca1f7592069fd4\nAuthor: Johannes Berg \u003cjohannes.berg@intel.com\u003e\nDate: Wed Jul 3 15:38:23 2019 +0200\n\n cfg80211: give all multi-BSSID BSS entries the same timestamp\n \n If we just read jiffies over and over again, a non-transmitting\n entry may have a newer timestamp than the transmitting one,\n leading to possible confusion on expiry. Give them all the same\n timestamp when creating them.\n \n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n Reviewed-by: Sergey Matyukevich \u003csergey.matyukevich.os@quantenna.com\u003e\n Link: https://lore.kernel.org/r/20190703133823.10530-3-johannes@sipsolutions.net\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit b0d1d7ffc55571185e626a6ab2983400fc493cc2\nAuthor: Johannes Berg \u003cjohannes.berg@intel.com\u003e\nDate: Wed Jul 3 15:38:22 2019 +0200\n\n cfg80211: don\u0027t parse MBSSID if transmitting BSS isn\u0027t created\n \n Don\u0027t parse the multi-BSSID structures if we couldn\u0027t even create\n their transmitting BSS, this would confuse all of our tracking.\n \n This also means that non_tx_data-\u003etx_bss will never be NULL, so\n we can clean up a little bit.\n \n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n Reviewed-by: Sergey Matyukevich \u003csergey.matyukevich.os@quantenna.com\u003e\n Link: https://lore.kernel.org/r/20190703133823.10530-2-johannes@sipsolutions.net\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 84f1772bc0c7bc26db9dbdbeabc4b1fa0ea1ace3\nAuthor: Johannes Berg \u003cjohannes.berg@intel.com\u003e\nDate: Wed Jul 3 15:38:21 2019 +0200\n\n cfg80211: clean up cfg80211_inform_single_bss_frame_data()\n \n cfg80211_inform_single_bss_frame_data() doesn\u0027t need the\n non_tx_data data argument since it\u0027s always NULL.\n \n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n Reviewed-by: Sergey Matyukevich \u003csergey.matyukevich.os@quantenna.com\u003e\n Link: https://lore.kernel.org/r/20190703133823.10530-1-johannes@sipsolutions.net\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit d82574a8e5a469b72efaaa134f953a71642217e8\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Wed Jul 3 09:01:42 2019 +0200\n\n cfg80211: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Link: https://lore.kernel.org/r/20190703070142.GA29993@kroah.com\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit dc3998ec5cf2d377f2e85ba16b6a15affec98a0a\nAuthor: Alexander Wetzel \u003calexander@wetzel-home.de\u003e\nDate: Sat Jun 29 21:50:14 2019 +0200\n\n mac80211: AMPDU handling for rekeys with Extended Key ID\n \n Extended Key ID allows A-MPDU sessions while rekeying as long as each\n A-MPDU aggregates only MPDUs with one keyid together.\n \n Drivers able to segregate MPDUs accordingly can tell mac80211 to not\n stop A-MPDU sessions when rekeying by setting the new flag\n IEEE80211_HW_AMPDU_KEYBORDER_SUPPORT.\n \n Signed-off-by: Alexander Wetzel \u003calexander@wetzel-home.de\u003e\n Link: https://lore.kernel.org/r/20190629195015.19680-3-alexander@wetzel-home.de\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 3e47bf1ca4c363ba8b1f99c4c3dcda13d2979954\nAuthor: Alexander Wetzel \u003calexander@wetzel-home.de\u003e\nDate: Sat Jun 29 21:50:13 2019 +0200\n\n mac80211: Simplify Extended Key ID API\n \n 1) Drop IEEE80211_HW_EXT_KEY_ID_NATIVE and let drivers directly set\n the NL80211_EXT_FEATURE_EXT_KEY_ID flag.\n \n 2) Drop IEEE80211_HW_NO_AMPDU_KEYBORDER_SUPPORT and simply assume all\n drivers are unable to handle A-MPDU key borders.\n \n The new Extended Key ID API now requires all mac80211 drivers to set\n NL80211_EXT_FEATURE_EXT_KEY_ID when they implement set_key() and can\n handle Extended Key ID. For drivers not providing set_key() mac80211\n itself enables Extended Key ID support, using the internal SW crypto\n services.\n \n Signed-off-by: Alexander Wetzel \u003calexander@wetzel-home.de\u003e\n Link: https://lore.kernel.org/r/20190629195015.19680-2-alexander@wetzel-home.de\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 70f69f481b87da0614360e57e05a3ba962d60439\nAuthor: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\nDate: Fri Jul 26 15:20:53 2019 +0800\n\n spi: sprd: Change the hwlock support to be optional\n \n No need to add hardware spinlock proctection due to add multiple\n msater channel, so change it to be optional in documentation.\n \n Signed-off-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Link: https://lore.kernel.org/r/23d51f5d9c9cc647ad0c5a1fb950d3d9fb9c1303.1564125131.git.baolin.wang@linaro.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 5d29050b409d961df34a7290270ba53f0c025152\nAuthor: Alexander Wetzel \u003calexander@wetzel-home.de\u003e\nDate: Sat Jun 29 21:50:12 2019 +0200\n\n mac80211_hwsim: Extended Key ID API update\n \n Prepare hwsim Extended Key ID support for a mac80211 API change.\n \n The mac80211 flag IEEE80211_HW_EXT_KEY_ID_NATIVE is being replaced by\n NL80211_EXT_FEATURE_EXT_KEY_ID which only must be set by drivers when\n they support HW crypto.\n \n This reverts commit cfe7007a9b4c (\"mac80211_hwsim: Extended Key ID support\").\n \n Signed-off-by: Alexander Wetzel \u003calexander@wetzel-home.de\u003e\n Link: https://lore.kernel.org/r/20190629195015.19680-1-alexander@wetzel-home.de\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit f9adf61e983f5d03334d841bc30f0e967a340b37\nAuthor: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\nDate: Fri Jul 26 15:20:52 2019 +0800\n\n spi: sprd: adi: Change hwlock to be optional\n \n Now Spreadtrum ADI controller supplies multiple master accessing channel\n to support multiple subsystems accessing, instead of using a hardware\n spinlock to synchronize between the multiple subsystems.\n \n To keep backward compatibility, we should change the hardware spinlock\n to be optional. Moreover change to use of_hwspin_lock_get_id() function\n which return -ENOENT error number to indicate no hwlock support.\n \n Signed-off-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Link: https://lore.kernel.org/r/2abe7dcf210e4197f8c5ece7fc6d6cc1eda8c655.1564125131.git.baolin.wang@linaro.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit e6d722ca09c1eebba5660e58b6bcac30c1ccc7ca\nAuthor: Sherry Zong \u003csherry.zong@unisoc.com\u003e\nDate: Fri Jul 26 15:20:51 2019 +0800\n\n spi: sprd: adi: Add a reset reason for watchdog mode\n \n When the system was rebooted by watchdog, now we did not save the watchdog\n reset mode which will make system enter a incorrect mode after rebooting.\n \n Thus we should set the watchdog reset mode as default when opening the\n watchdog configuration, that means if the system was rebooted by other\n reason through the restart_handler(), then we will clear the default\n watchdog reset mode to save the correct reset mode.\n \n Signed-off-by: Sherry Zong \u003csherry.zong@unisoc.com\u003e\n Signed-off-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Link: https://lore.kernel.org/r/1563f3de43c6c2262d597a25d6138b5de61ea23d.1564125131.git.baolin.wang@linaro.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9d9aa1cc91d68e2140e5948c15947b8254e008e2\nAuthor: Sherry Zong \u003csherry.zong@unisoc.com\u003e\nDate: Fri Jul 26 15:20:50 2019 +0800\n\n spi: sprd: adi: Add a reset reason for factory test mode\n \n Add a new reset flag to indicate that the system need enter factory test\n mode after restarting system.\n \n Signed-off-by: Sherry Zong \u003csherry.zong@unisoc.com\u003e\n Signed-off-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Link: https://lore.kernel.org/r/8ae5651e876b527920ff878721a8a8ef47b099ac.1564125131.git.baolin.wang@linaro.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit cc6b3431b36a1109d44cb8e4015cc68dddc75208\nAuthor: Chenxu Wei \u003cweicx@spreadst.com\u003e\nDate: Fri Jul 26 15:20:49 2019 +0800\n\n spi: sprd: adi: Add a reset reason for TOS panic\n \n Add a new reset flag to indicate the reset reason is caused by TOS.\n \n Signed-off-by: Chenxu Wei \u003cweicx@spreadst.com\u003e\n Signed-off-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Link: https://lore.kernel.org/r/97583aad1f2b849d69b4e76e8d29113da72a9fff.1564125131.git.baolin.wang@linaro.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c627c58acdc48055a9e4d40d6f9f1b434222a68d\nAuthor: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\nDate: Fri Jul 26 15:20:48 2019 +0800\n\n spi: sprd: adi: Remove redundant address bits setting\n \n The ADI default transfer address bits is 12bit on Spreadtrum SC9860\n platform, thus there is no need to set again, remove it.\n \n Signed-off-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Link: https://lore.kernel.org/r/3cb57b8aadb7747a9f833e9b4fe8596ba738d9f6.1564125131.git.baolin.wang@linaro.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit fb0e76abe34bd67756dbdf4d5982b7dc54afa1d8\nAuthor: Erik Stromdahl \u003cerik.stromdahl@gmail.com\u003e\nDate: Mon Jun 17 22:01:39 2019 +0200\n\n mac80211: add tx dequeue function for process context\n \n Since ieee80211_tx_dequeue() must not be called with softirqs enabled\n (i.e. from process context without proper disable of bottom halves),\n we add a wrapper that disables bottom halves before calling\n ieee80211_tx_dequeue()\n \n The new function is named ieee80211_tx_dequeue_ni() just as all other\n from-process-context versions found in mac80211.\n \n The documentation of ieee80211_tx_dequeue() is also updated so it\n mentions that the function should not be called from process context.\n \n Signed-off-by: Erik Stromdahl \u003cerik.stromdahl@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190617200140.6189-1-erik.stromdahl@gmail.com\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit a11e2f85481c2f08b55c06467445602a2330ee5b\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Mon Jun 17 11:19:01 2019 +0200\n\n lib80211: use crypto API ccm(aes) transform for CCMP processing\n \n Instead of open coding the CCM aead mode in the driver, and invoking\n the AES block cipher block by block, use a ccm(aes) aead transform\n which already encapsulates this functionality. This is a cleaner use\n of the crypto API, and permits optimized implementations to be used,\n which are typically much faster and deal more efficiently with the\n SIMD register file, which usually needs to be preserved/restored in\n order to use special AES instructions.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190617091901.7063-1-ard.biesheuvel@linaro.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 612fcfd9b31f08858d2a2e1279adda367e1ade00\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Wed Jun 12 16:26:58 2019 +0200\n\n mac80211: remove unused and unneeded remove_sta_debugfs callback\n \n The remove_sta_debugfs callback in struct rate_control_ops is no longer\n used by any driver, as there is no need for it (the debugfs directory is\n already removed recursivly by the mac80211 core.) Because no one needs\n it, just remove it to keep anyone else from accidentally using it in the\n future.\n \n Cc: Johannes Berg \u003cjohannes@sipsolutions.net\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: linux-wireless@vger.kernel.org\n Cc: netdev@vger.kernel.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Link: https://lore.kernel.org/r/20190612142658.12792-5-gregkh@linuxfoundation.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 09e1946cb7590d1a7a314534420f05c7abdf1f55\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Wed Jun 12 16:26:57 2019 +0200\n\n iwlwifi: mvm: remove unused .remove_sta_debugfs callback\n \n The .remove_sta_debugfs callback was not doing anything in this driver,\n so remove it as it is not needed.\n \n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Link: https://lore.kernel.org/r/20190612142658.12792-4-gregkh@linuxfoundation.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 49175fe63ee8433dffeb64aee685bf8d439d5698\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Wed Jun 12 16:26:56 2019 +0200\n\n iwlwifi: dvm: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. This driver was saving the debugfs file away to be\n removed at a later time. However, the 80211 core would delete the whole\n directory that the debugfs files are created in, after it asks the\n driver to do the deletion, so just rely on the 80211 core to do all of\n the cleanup for us, making us not need to keep a pointer to the dentries\n around at all.\n \n This cleans up the structure of the driver data a bit and makes the code\n a tiny bit smaller.\n \n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Link: https://lore.kernel.org/r/20190612142658.12792-3-gregkh@linuxfoundation.org\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 5db4c4b9559f8cddd5f7f74e58c7b8f172120e6d\nAuthor: Emmanuel Grumbach \u003cemmanuel.grumbach@intel.com\u003e\nDate: Tue Jul 23 21:00:01 2019 +0300\n\n mac80211: pass the vif to cancel_remain_on_channel\n \n This low level driver can find it useful to get the vif\n when a remain on channel session is cancelled.\n \n iwlwifi will need this soon.\n \n Signed-off-by: Emmanuel Grumbach \u003cemmanuel.grumbach@intel.com\u003e\n Link: https://lore.kernel.org/r/20190723180001.5828-1-emmanuel.grumbach@intel.com\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 690c4509e980e7c9652945f5ae47e90f00f25ffc\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Thu Jul 25 21:13:28 2019 +0200\n\n mac80211_hwsim: Fix a typo in the name of function \u0027mac80211_hswim_he_capab()\u0027\n \n This function name should be \u0027mac80211_hwsim_he_capab()\u0027 (s wand w\n switched) to be consistent with the rest of the file.\n Fix and use it.\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Link: https://lore.kernel.org/r/20190725191328.18010-1-christophe.jaillet@wanadoo.fr\n Signed-off-by: Johannes Berg \u003cjohannes.berg@intel.com\u003e\n\ncommit 67de843e315845bb9f9bede75ca8cbb04a57c417\nAuthor: Je Yen Tam \u003cje.yen.tam@ni.com\u003e\nDate: Fri Jul 26 15:40:12 2019 +0800\n\n serial/8250: Add support for NI-Serial PXI/PXIe+485 devices\n \n Add support for NI-Serial PXIe-RS232, PXI-RS485 and PXIe-RS485 devices.\n \n Signed-off-by: Je Yen Tam \u003cje.yen.tam@ni.com\u003e\n Link: https://lore.kernel.org/r/20190726074012.2590-1-je.yen.tam@ni.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit edad25476ba353cfc0f3b34b150888b46f18e8a6\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Thu Jul 25 10:46:55 2019 -0700\n\n drm/i915/guc: init submission structures as part of guc_init\n \n guc-\u003estage_desc_pool is required as part of the init parameters and\n there is no reason we have to init them after HuC. This fixes a NULL\n ptr dereference due to guc-\u003estage_desc_pool not being set (no fixes\n tag since GuC submission can\u0027t be enabled yet).\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725174655.24382-1-daniele.ceraolospurio@intel.com\n\ncommit a409906003a2b5418e6e60ac2524948ea80819f2\nAuthor: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\nDate: Thu Jul 25 12:41:32 2019 +0200\n\n dt-bindings: opp: qcom-nvmem: Make speedbin related properties optional\n \n Not all Qualcomm platforms need to care about the speedbin efuse,\n nor the value blown into the speedbin efuse.\n Therefore, make the nvmem-cells and opp-supported-hw properties\n optional.\n \n Signed-off-by: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n Reviewed-by: Ilia Lin \u003cilia.lin@kernel.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit 8cfda0df3a3020454848f55ed23a781169770c99\nAuthor: Sricharan R \u003csricharan@codeaurora.org\u003e\nDate: Thu Jul 25 12:41:30 2019 +0200\n\n dt-bindings: opp: Re-organise kryo cpufreq to use it for other nvmem based qcom socs\n \n The kryo cpufreq driver reads the nvmem cell and uses that data to\n populate the opps. There are other qcom cpufreq socs like krait which\n does similar thing. Except for the interpretation of the read data,\n rest of the driver is same for both the cases. So pull the common things\n out for reuse.\n \n Signed-off-by: Sricharan R \u003csricharan@codeaurora.org\u003e\n [niklas.cassel@linaro.org: split dt-binding into a separate patch and\n do not rename the compatible string.]\n Signed-off-by: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n Reviewed-by: Ilia Lin \u003cilia.lin@kernel.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit 27a84f76738cc4a9509ec9234a7021de34a88e58\nAuthor: Yue Hu \u003chuyue2@yulong.com\u003e\nDate: Fri Jul 26 14:07:47 2019 +0800\n\n PM / OPP: Correct Documentation about library location\n \n OPP library is now located in drivers/opp/ directory.\n \n Signed-off-by: Yue Hu \u003chuyue2@yulong.com\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit 4527551750e85f9be9296f2b30b19bb257f342e5\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Tue Jul 9 16:00:13 2019 +0800\n\n opp: of: Support multiple suspend OPPs defined in DT\n \n With property \"opp-supported-hw\" introduced, the OPP table\n in DT could be a large OPP table and ONLY a subset of OPPs\n are available, based on the version of the hardware running\n on. That introduces restriction of using \"opp-suspend\"\n property to define the suspend OPP, as we are NOT sure if the\n OPP containing \"opp-suspend\" property is available for the\n hardware running on, and the of opp core does NOT allow multiple\n suspend OPPs defined in DT OPP table.\n \n To eliminate this restrition, make of opp core allow multiple\n suspend OPPs defined in DT, and pick the OPP with highest rate\n and with \"opp-suspend\" property present to be suspend OPP, it\n can speed up the suspend/resume process.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit 518c6880ffc965d59349bf064ff00153cf05f033\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Tue Jul 9 16:00:12 2019 +0800\n\n dt-bindings: opp: Support multiple opp-suspend properties\n \n Update opp-suspend property\u0027s description to support multiple\n opp-suspend properties defined in DT, the OPP with highest opp-hz\n and with opp-suspend property present will be used as suspend opp.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit 7f93ff73f7c8c8bfa6be33bcc16470b0b44682aa\nAuthor: k.konieczny@partner.samsung.com \u003ck.konieczny@partner.samsung.com\u003e\nDate: Fri Jul 19 17:05:32 2019 +0200\n\n opp: core: add regulators enable and disable\n \n Add enable regulators to dev_pm_opp_set_regulators() and disable\n regulators to dev_pm_opp_put_regulators(). Even if bootloader\n leaves regulators enabled, they should be enabled in kernel in\n order to increase the reference count.\n \n Signed-off-by: Kamil Konieczny \u003ck.konieczny@partner.samsung.com\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit 11e1a1648298656722fe206a53e555e2e98668ae\nAuthor: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\nDate: Wed Jul 3 15:03:14 2019 +0530\n\n opp: Don\u0027t decrement uninitialized list_kref\n \n The list_kref was added for static OPPs and to track their users. The\n kref is initialized while the static OPPs are added, but removed\n unconditionally even if the static OPPs were never added. This causes\n refcount mismatch warnings currently.\n \n Fix that by always initializing the kref when the OPP table is first\n initialized. The refcount is later incremented only for the second user\n onwards.\n \n Fixes: d0e8ae6c26da (\"OPP: Create separate kref for static OPPs list\")\n Reported-by: Rajendra Nayak \u003crnayak@codeaurora.org\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit 71419d84c216cee8da3b19fb843b4242f112cde4\nAuthor: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\nDate: Thu Jul 25 12:41:29 2019 +0200\n\n opp: Add dev_pm_opp_find_level_exact()\n \n Since the performance states in the OPP table are unique, implement a\n dev_pm_opp_find_level_exact() in order to be able to fetch a specific OPP.\n \n Signed-off-by: Niklas Cassel \u003cniklas.cassel@linaro.org\u003e\n [ Viresh: Updated commit log ]\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit 17a8f868ae3e85a173843b1ac65e744e8585bc5a\nAuthor: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\nDate: Mon Jul 8 11:24:56 2019 +0530\n\n opp: Return genpd virtual devices from dev_pm_opp_attach_genpd()\n \n The cpufreq drivers don\u0027t need to do runtime PM operations on the\n virtual devices returned by dev_pm_domain_attach_by_name() and so the\n virtual devices weren\u0027t shared with the callers of\n dev_pm_opp_attach_genpd() earlier.\n \n But the IO device drivers would want to do that. This patch updates the\n prototype of dev_pm_opp_attach_genpd() to accept another argument to\n return the pointer to the array of genpd virtual devices.\n \n Reported-by: Rajendra Nayak \u003crnayak@codeaurora.org\u003e\n Tested-by: Rajendra Nayak \u003crnayak@codeaurora.org\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit baea35e4db17a72145c84a401f70d496c8ebf833\nAuthor: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\nDate: Wed Jul 17 11:20:17 2019 +0530\n\n opp: Not all power-domains are scalable\n \n A device may have multiple power-domains and not all of them may be\n scalable (i.e. support performance states). But\n dev_pm_opp_attach_genpd() doesn\u0027t take that into account currently.\n \n Fix that by not verifying the names argument with \"power-domain-names\"\n DT property and finding the index into the required-opps array. The\n names argument will anyway get verified later on when we call\n dev_pm_domain_attach_by_name().\n \n Fixes: 6319aee10e53 (\"opp: Attach genpds to devices from within OPP core\")\n Reported-by: Rajendra Nayak \u003crnayak@codeaurora.org\u003e\n Tested-by: Rajendra Nayak \u003crnayak@codeaurora.org\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit 42374b94d70cea72af94202ff5a467f036bc083e\nAuthor: Edward Matijević \u003cmotolav@gmail.com\u003e\nDate: Mon Jun 17 00:26:46 2019 -0500\n\n MIPS: BCM47XX: Add support for Netgear R6200 V1\n \n The Netgear R6200 v1 uses a BCM4718A1 SOC and a BCM4352/BCM4360 for 5GHz\n wireless. This patch adds support for detecting this model board and\n registers the 3 buttons.\n \n I have tested that the device can boot kernels 4.14 and 4.19 under\n OpenWRT.\n \n There is one issue that the LEDs on the device are controlled by a\n 74HC164 that uses bit-banging instead of SPI so it isn\u0027t accessible to\n the kernel without adding a workaround. Without any workaround the\n device on boot will flash all LEDs once then the power LED will remain\n amber as all other LEDs stay off.\n \n Signed-off-by: Edward Matijevic \u003cmotolav@gmail.com\u003e\n Acked-by: Hauke Mehrtens \u003chauke@hauke-m.de\u003e\n [paul.burton@mips.com:\n - Sort bcm47xx_board_list_board_id alphabetically by board type.\n - Fix whitespace.\n - Wrap commit message \u0026 drop OpenWRT-based justification for\n bcm47xx_board_list_board_id being mis-sorted.]\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit 77cdd4efe571345e9c116e65f64a616969e0bc35\nAuthor: Pascal van Leeuwen \u003cpascalvanl@gmail.com\u003e\nDate: Fri Jul 5 08:49:22 2019 +0200\n\n crypto: inside-secure - add support for authenc(hmac(sha1),cbc(des3_ede))\n \n Signed-off-by: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 5c9254ad7ae32a124fb084badd0cb3720f7c49cb\nAuthor: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\nDate: Thu Jul 4 17:37:57 2019 +0200\n\n crypto: ux500 - Use spinlock_t instead of struct spinlock\n \n For spinlocks the type spinlock_t should be used instead of \"struct\n spinlock\".\n \n Use spinlock_t for spinlock\u0027s definition.\n \n Cc: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: linux-crypto@vger.kernel.org\n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit cc2a58f14fb88e753fbc38b658e320c3a62d8c47\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Thu Jul 4 00:27:08 2019 +0800\n\n crypto: drivers - Use kmemdup rather than duplicating its implementation\n \n kmemdup is introduced to duplicate a region of memory in a neat way.\n Rather than kmalloc/kzalloc + memcpy, which the programmer needs to\n write the size twice (sometimes lead to mistakes), kmemdup improves\n readability, leads to smaller code and also reduce the chances of mistakes.\n Suggestion to use kmemdup rather than using kmalloc/kzalloc + memcpy.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Reviewed-by: Horia Geantă \u003choria.geanta@nxp.com\u003e\n Acked-by: Michael S. Tsirkin \u003cmst@redhat.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 97bcb161995548ad319c78b2533f998a7b92ab4c\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Wed Jul 3 10:55:12 2019 +0200\n\n crypto: tcrypt - add a speed test for AEGIS128\n \n Reviewed-by: Ondrej Mosnacek \u003comosnace@redhat.com\u003e\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit ecc8bc81f2fb3976737ef312f824ba6053aa3590\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Wed Jul 3 10:55:11 2019 +0200\n\n crypto: aegis128 - provide a SIMD implementation based on NEON intrinsics\n \n Provide an accelerated implementation of aegis128 by wiring up the\n SIMD hooks in the generic driver to an implementation based on NEON\n intrinsics, which can be compiled to both ARM and arm64 code.\n \n This results in a performance of 2.2 cycles per byte on Cortex-A53,\n which is a performance increase of ~11x compared to the generic\n code.\n \n Reviewed-by: Ondrej Mosnacek \u003comosnace@redhat.com\u003e\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 7cdc0ddbf74a19cecb2f0e9efa2cae9d3c665189\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Wed Jul 3 10:55:10 2019 +0200\n\n crypto: aegis128 - add support for SIMD acceleration\n \n Add some plumbing to allow the AEGIS128 code to be built with SIMD\n routines for acceleration.\n \n Reviewed-by: Ondrej Mosnacek \u003comosnace@redhat.com\u003e\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 521cdde758bf331d4e264ef3deef5a26d5ce0b4f\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Wed Jul 3 10:55:09 2019 +0200\n\n crypto: aegis - avoid prerotated AES tables\n \n The generic AES code provides four sets of lookup tables, where each\n set consists of four tables containing the same 32-bit values, but\n rotated by 0, 8, 16 and 24 bits, respectively. This makes sense for\n CISC architectures such as x86 which support memory operands, but\n for other architectures, the rotates are quite cheap, and using all\n four tables needlessly thrashes the D-cache, and actually hurts rather\n than helps performance.\n \n Since x86 already has its own implementation of AEGIS based on AES-NI\n instructions, let\u0027s tweak the generic implementation towards other\n architectures, and avoid the prerotated tables, and perform the\n rotations inline. On ARM Cortex-A53, this results in a ~8% speedup.\n \n Acked-by: Ondrej Mosnacek \u003comosnace@redhat.com\u003e\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 368b1bdc0a5983c8c908ce2001229e2291eac583\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Wed Jul 3 10:55:08 2019 +0200\n\n crypto: aegis128 - drop empty TFM init/exit routines\n \n TFM init/exit routines are optional, so no need to provide empty ones.\n \n Reviewed-by: Ondrej Mosnacek \u003comosnace@redhat.com\u003e\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 520c1993bbe620e39fd93de1a01b9e0dc0b97aa6\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Wed Jul 3 10:55:07 2019 +0200\n\n crypto: aegis128l/aegis256 - remove x86 and generic implementations\n \n Three variants of AEGIS were proposed for the CAESAR competition, and\n only one was selected for the final portfolio: AEGIS128.\n \n The other variants, AEGIS128L and AEGIS256, are not likely to ever turn\n up in networking protocols or other places where interoperability\n between Linux and other systems is a concern, nor are they likely to\n be subjected to further cryptanalysis. However, uninformed users may\n think that AEGIS128L (which is faster) is equally fit for use.\n \n So let\u0027s remove them now, before anyone starts using them and we are\n forced to support them forever.\n \n Note that there are no known flaws in the algorithms or in any of these\n implementations, but they have simply outlived their usefulness.\n \n Reviewed-by: Ondrej Mosnacek \u003comosnace@redhat.com\u003e\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 5cb97700beaa005ceb2a127b6f53536a4544c9d8\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Wed Jul 3 10:55:06 2019 +0200\n\n crypto: morus - remove generic and x86 implementations\n \n MORUS was not selected as a winner in the CAESAR competition, which\n is not surprising since it is considered to be cryptographically\n broken [0]. (Note that this is not an implementation defect, but a\n flaw in the underlying algorithm). Since it is unlikely to be in use\n currently, let\u0027s remove it before we\u0027re stuck with it.\n \n [0] https://eprint.iacr.org/2019/172.pdf\n \n Reviewed-by: Ondrej Mosnacek \u003comosnace@redhat.com\u003e\n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit f248caf9a5ccda733456baddb234bce7431152ba\nAuthor: Hannah Pan \u003channahpan@google.com\u003e\nDate: Tue Jul 2 15:16:02 2019 -0700\n\n crypto: testmgr - add tests for lzo-rle\n \n Add self-tests for the lzo-rle algorithm.\n \n Signed-off-by: Hannah Pan \u003channahpan@google.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit b46033fdd25abb30b253cd283b551323b723a2db\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:50 2019 +0200\n\n crypto: arm/aes-scalar - unexport en/decryption routines\n \n The scalar table based AES routines are not used by other drivers, so\n let\u0027s keep it that way and unexport the symbols.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 642a88fbe94441de7f2c807263599d5ef61b70fb\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:49 2019 +0200\n\n crypto: arm64/aes-cipher - switch to shared AES inverse Sbox\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 8de6dd3386003a3fe4509304494061b3554fe516\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:48 2019 +0200\n\n crypto: arm/aes-cipher - switch to shared AES inverse Sbox\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 58144b8d03fca6209a5097e85bcd63a6a956e75f\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:47 2019 +0200\n\n crypto: arm64/aes-neon - switch to shared AES Sboxes\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 9467a3150cf4afca638673e099af71e8c493a3a0\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:46 2019 +0200\n\n crypto: lib/aes - export sbox and inverse sbox\n \n There are a few copies of the AES S-boxes floating around, so export\n the ones from the AES library so that we can reuse them in other\n modules.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 1e25ca02a0619c10bc0eae975846926902381ce5\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:45 2019 +0200\n\n crypto: aes-generic - unexport last-round AES tables\n \n The versions of the AES lookup tables that are only used during the last\n round are never used outside of the driver, so there is no need to\n export their symbols.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 571c47ab98ad8c85e44b28f2f0c41b0e27312a9b\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:44 2019 +0200\n\n crypto: chelsio - replace AES cipher calls with library calls\n \n Replace a couple of occurrences where the \"aes-generic\" cipher is\n instantiated explicitly and only used for encryption of a single block.\n Use AES library calls instead.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 6273fd7a5a99549bc5b90ce8a70df88cc6b570c3\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:43 2019 +0200\n\n crypto: ccp - move to AES library for CMAC key derivation\n \n Use the AES library instead of the cipher interface to perform\n the single block of AES processing involved in updating the key\n of the cmac(aes) hash.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit da3e7a9715ea041299a9e1eae1e73405b110a333\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:42 2019 +0200\n\n crypto: amcc - switch to AES library for GCM key derivation\n \n The AMCC code for GCM key derivation allocates a AES cipher to\n perform a single block encryption. So let\u0027s switch to the new\n and more lightweight AES library instead.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 28a220aac596ff37b564b2d6dd6a9b7e2be29217\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:41 2019 +0200\n\n bluetooth: switch to AES library\n \n The bluetooth code uses a bare AES cipher for the encryption operations.\n Given that it carries out a set_key() operation right before every\n encryption operation, this is clearly not a hot path, and so the use of\n the cipher interface (which provides the best implementation available\n on the system) is not really required.\n \n In fact, when using a cipher like AES-NI or AES-CE, both the set_key()\n and the encrypt() operations involve en/disabling preemption as well as\n stacking and unstacking the SIMD context, and this is most certainly\n not worth it for encrypting 16 bytes of data.\n \n So let\u0027s switch to the new lightweight library interface instead.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 0a5dff9882e56f80ff021e039aee93e167f16b9e\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:40 2019 +0200\n\n crypto: arm/ghash - provide a synchronous version\n \n GHASH is used by the GCM mode, which is often used in contexts where\n only synchronous ciphers are permitted. So provide a synchronous version\n of GHASH based on the existing code. This requires a non-SIMD fallback\n to deal with invocations occurring from a context where SIMD instructions\n may not be used.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit e5f050402f1870317826990e968493e4411a0d53\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:39 2019 +0200\n\n crypto: arm/aes-neonbs - provide a synchronous version of ctr(aes)\n \n AES in CTR mode is used by modes such as GCM and CCM, which are often\n used in contexts where only synchronous ciphers are permitted. So\n provide a synchronous version of ctr(aes) based on the existing code.\n This requires a non-SIMD fallback to deal with invocations occurring\n from a context where SIMD instructions may not be used. We have a\n helper for this now in the AES library, so wire that up.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 5eedf315f27d3990f4d00c7cd3b9751e184a83b1\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:38 2019 +0200\n\n crypto: arm/aes-ce - provide a synchronous version of ctr(aes)\n \n AES in CTR mode is used by modes such as GCM and CCM, which are often\n used in contexts where only synchronous ciphers are permitted. So\n provide a synchronous version of ctr(aes) based on the existing code.\n This requires a non-SIMD fallback to deal with invocations occurring\n from a context where SIMD instructions may not be used. We have a\n helper for this now in the AES library, so wire that up.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit fafb1dca6fad0efeeaaaeae5067cf7e194a47027\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:37 2019 +0200\n\n crypto: arm/aes - use native endiannes for key schedule\n \n Align ARM\u0027s hw instruction based AES implementation with other versions\n that keep the key schedule in native endianness. This will allow us to\n merge the various implementations going forward.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 4d3f9d89c745ea5494ea23e545385af1b6f81363\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:36 2019 +0200\n\n crypto: arm64/aes-ce-cipher - use AES library as fallback\n \n Instead of calling into the table based scalar AES code in situations\n where the SIMD unit may not be used, use the generic AES code, which\n is more appropriate since it is less likely to be susceptible to\n timing attacks.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit ff6f4115cb953c5be8d7a76b2ec1877df2f4c2c0\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:35 2019 +0200\n\n crypto: aes - move sync ctr(aes) to AES library and generic helper\n \n In preparation of duplicating the sync ctr(aes) functionality to modules\n under arch/arm, move the helper function from a inline .h file to the\n AES library, which is already depended upon by the drivers that use this\n fallback.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit d9ec772d9550b6e513c51bc4e6fa1e3ffb50181c\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:34 2019 +0200\n\n crypto: ctr - add helper for performing a CTR encryption walk\n \n Add a static inline helper modeled after crypto_cbc_encrypt_walk()\n that can be reused for SIMD algorithms that need to implement a\n non-SIMD fallback for performing CTR encryption.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 5bb12d7825adf0e80b849a273834f3131a6cc4e1\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:33 2019 +0200\n\n crypto: aes-generic - drop key expansion routine in favor of library version\n \n Drop aes-generic\u0027s version of crypto_aes_expand_key(), and switch to\n the key expansion routine provided by the AES library. AES key expansion\n is not performance critical, and it is better to have a single version\n shared by all AES implementations.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit c184472902d87189082e5a349051197e252ae9af\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:32 2019 +0200\n\n crypto: arm64/aes-ce - switch to library version of key expansion routine\n \n Switch to the new AES library that also provides an implementation of\n the AES key expansion routine. This removes the dependency on the\n generic AES cipher, allowing it to be omitted entirely in the future.\n \n While at it, remove some references to the table based arm64 version\n of AES and replace them with AES library calls as well.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit f68df54307aaae01df66977511ea719158201188\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:31 2019 +0200\n\n crypto: arm64/aes-neonbs - switch to library version of key expansion routine\n \n Switch to the new AES library that also provides an implementation of\n the AES key expansion routine. This removes the dependency on the\n generic AES cipher, allowing it to be omitted entirely in the future.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit c59a6dffa3cdf826cba92f1b967f8dba2fafec53\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:30 2019 +0200\n\n crypto: arm64/aes-ccm - switch to AES library\n \n The CCM code calls directly into the scalar table based AES cipher for\n arm64 from the fallback path, and since this implementation is known to\n be non-time invariant, doing so from a time invariant SIMD cipher is a\n bit nasty.\n \n So let\u0027s switch to the AES library - this makes the code more robust,\n and drops the dependency on the generic AES cipher, allowing us to\n omit it entirely in the future.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit aa6e2d2b355f89e22c30b41d7e6653c8189235b8\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:29 2019 +0200\n\n crypto: arm/aes-neonbs - switch to library version of key expansion routine\n \n Switch to the new AES library that also provides an implementation of\n the AES key expansion routine. This removes the dependency on the\n generic AES cipher, allowing it to be omitted entirely in the future.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit fe3b99b64909e8994f2606120e4703c9a1e8c080\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:28 2019 +0200\n\n crypto: arm64/ghash - switch to AES library\n \n The GHASH code uses the generic AES key expansion routines, and calls\n directly into the scalar table based AES cipher for arm64 from the\n fallback path, and since this implementation is known to be non-time\n invariant, doing so from a time invariant SIMD cipher is a bit nasty.\n \n So let\u0027s switch to the AES library - this makes the code more robust,\n and drops the dependency on the generic AES cipher, allowing us to\n omit it entirely in the future.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 363a90c2d517e69776dcf71cc3d6fcaee9fef868\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:27 2019 +0200\n\n crypto: safexcel/aes - switch to library version of key expansion routine\n \n Switch to the new AES library that also provides an implementation of\n the AES key expansion routine. This removes the dependency on the\n generic AES cipher, allowing it to be omitted entirely in the future.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 18d8b96daded4123d6f1ec00a390b16e4cf1d40c\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:26 2019 +0200\n\n crypto: cesa/aes - switch to library version of key expansion routine\n \n Switch to the new AES library that also provides an implementation of\n the AES key expansion routine. This removes the dependency on the\n generic AES cipher, allowing it to be omitted entirely in the future.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 8131878db76c2bd3065bd3b75cb6615390a393e6\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:25 2019 +0200\n\n crypto: padlock/aes - switch to library version of key expansion routine\n \n Switch to the new AES library that also provides an implementation of\n the AES key expansion routine. This removes the dependency on the\n generic AES cipher, allowing it to be omitted entirely in the future.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 1d2c3279311e4f03fcf164e1366f2fda9f4bfccf\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:24 2019 +0200\n\n crypto: x86/aes - drop scalar assembler implementations\n \n The AES assembler code for x86 isn\u0027t actually faster than code\n generated by the compiler from aes_generic.c, and considering\n the disproportionate maintenance burden of assembler code on\n x86, it is better just to drop it entirely. Modern x86 systems\n will use AES-NI anyway, and given that the modules being removed\n have a dependency on aes_generic already, we can remove them\n without running the risk of regressions.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 2c53fd11f7624658222d175ec27e6c07b20b63d0\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:23 2019 +0200\n\n crypto: x86/aes-ni - switch to generic for fallback and key routines\n \n The AES-NI code contains fallbacks for invocations that occur from a\n context where the SIMD unit is unavailable, which really only occurs\n when running in softirq context that was entered from a hard IRQ that\n was taken while running kernel code that was already using the FPU.\n \n That means performance is not really a consideration, and we can just\n use the new library code for this use case, which has a smaller\n footprint and is believed to be time invariant. This will allow us to\n drop the non-SIMD asm routines in a subsequent patch.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit e59c1c98745637796df824c0177f279b6e9cad94\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:22 2019 +0200\n\n crypto: aes - create AES library based on the fixed time AES code\n \n Take the existing small footprint and mostly time invariant C code\n and turn it into a AES library that can be used for non-performance\n critical, casual use of AES, and as a fallback for, e.g., SIMD code\n that needs a secondary path that can be taken in contexts where the\n SIMD unit is off limits (e.g., in hard interrupts taken from kernel\n context)\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit b158fcbba857c71ffb05ab254aff3b32b5e3cfc3\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:21 2019 +0200\n\n crypto: aes/fixed-time - align key schedule with other implementations\n \n The fixed time AES code mangles the key schedule so that xoring the\n first round key with values at fixed offsets across the Sbox produces\n the correct value. This primes the D-cache with the entire Sbox before\n any data dependent lookups are done, making it more difficult to infer\n key bits from timing variances when the plaintext is known.\n \n The downside of this approach is that it renders the key schedule\n incompatible with other implementations of AES in the kernel, which\n makes it cumbersome to use this implementation as a fallback for SIMD\n based AES in contexts where this is not allowed.\n \n So let\u0027s tweak the fixed Sbox indexes so that they add up to zero under\n the xor operation. While at it, increase the granularity to 16 bytes so\n we cover the entire Sbox even on systems with 16 byte cachelines.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 724ecd3c0eb7040d423b22332a60d097e2666820\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:20 2019 +0200\n\n crypto: aes - rename local routines to prevent future clashes\n \n Rename some local AES encrypt/decrypt routines so they don\u0027t clash with\n the names we are about to introduce for the routines exposed by the\n generic AES library.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 20bb4ef038a97b8bb5c07d2a1125019a93f618b3\nAuthor: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\nDate: Tue Jul 2 21:41:19 2019 +0200\n\n crypto: arm/aes-ce - cosmetic/whitespace cleanup\n \n Rearrange the aes_algs[] array for legibility.\n \n Signed-off-by: Ard Biesheuvel \u003card.biesheuvel@linaro.org\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 85b36ee8e9a8b18b3c09b8a8b6ac3dc694584bab\nAuthor: Pascal van Leeuwen \u003cpvanleeuwen@insidesecure.com\u003e\nDate: Tue Jul 2 16:40:00 2019 +0200\n\n crypto: inside-secure - add support for 0 length HMAC messages\n \n This patch adds support for the specific corner case of performing HMAC\n on an empty string (i.e. payload length is zero). This solves the last\n failing cryptomgr extratests for HMAC.\n \n Signed-off-by: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 41abed7d72c93fd31b0c1d51f42606216f1ea882\nAuthor: Pascal van Leeuwen \u003cpvanleeuwen@insidesecure.com\u003e\nDate: Tue Jul 2 16:39:59 2019 +0200\n\n crypto: inside-secure - add support for arbitrary size hash/HMAC updates\n \n This patch fixes an issue with hash and HMAC operations that perform\n \"large\" intermediate updates (i.e. combined size \u003e 2 hash blocks) by\n actually making use of the hardware\u0027s hash continue capabilities.\n The original implementation would cache these updates in a buffer that\n was 2 hash blocks in size and fail if all update calls combined would\n overflow that buffer. Which caused the cryptomgr extra tests to fail.\n \n Signed-off-by: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 85695b093d559bcd2393a65373355390ac62d67b\nAuthor: Pascal van Leeuwen \u003cpvanleeuwen@insidesecure.com\u003e\nDate: Tue Jul 2 16:39:58 2019 +0200\n\n crypto: inside-secure - let HW deal with initial hash digest\n \n The driver was loading the initial digest for hash operations into\n the hardware explicitly, but this is not needed as the hardware can\n handle that by itself, which is more efficient and avoids any context\n record coherence issues.\n \n Signed-off-by: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit dc5268b65dfc147850e0d2c49b000cb385d7b026\nAuthor: Pascal van Leeuwen \u003cpvanleeuwen@insidesecure.com\u003e\nDate: Tue Jul 2 16:39:57 2019 +0200\n\n crypto: inside-secure: back out parts of earlier HMAC update workaround\n \n This patch backs out some changes done with commit 082ec2d48467 -\n \"add support for HMAC updates\" as that update just works around the\n issue for the basic tests by providing twice the amount of buffering,\n but this does not solve the case of much larger data blocks such as\n those performed by the extra tests.\n This is in preparation of an actual solution in the next patch(es),\n which does not actually require any extra buffering at all.\n \n Signed-off-by: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit a74d850f787ef9578fa2829bfaecb222269cb754\nAuthor: Pascal van Leeuwen \u003cpvanleeuwen@insidesecure.com\u003e\nDate: Tue Jul 2 16:39:56 2019 +0200\n\n crypto: inside-secure - fix EINVAL error (buf overflow) for AEAD decrypt\n \n This patch fixes a buffer overflow error returning -EINVAL for AEAD\n decrypt operations by NOT appending the (already verified) ICV to\n the output packet (which is not expected by the API anyway).\n With this fix, all testmgr AEAD (extra) tests now pass.\n \n Signed-off-by: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 19b347b32bb195c50aea56b1ebf524e461e39827\nAuthor: Pascal van Leeuwen \u003cpvanleeuwen@insidesecure.com\u003e\nDate: Tue Jul 2 16:39:55 2019 +0200\n\n crypto: inside-secure - fix scatter/gather list to descriptor conversion\n \n Fixed issues with the skcipher and AEAD scatter/gather list to engine\n descriptor conversion code which caused either too much or too little\n buffer space to be provided to the hardware. This caused errors with the\n testmgr extra tests, either kernel panics (on x86-EIP197-FPGA) or engine\n descriptor errors 0x1, 0x8 or 0x9 (on Macchiatobin e.g. Marvell A8K).\n With this patch in place, all skcipher and AEAD (extra) tests pass.\n \n Signed-off-by: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 5bdb6e6aa53bf5c6e6082a4bb44e1106a22d68b8\nAuthor: Pascal van Leeuwen \u003cpvanleeuwen@insidesecure.com\u003e\nDate: Tue Jul 2 16:39:54 2019 +0200\n\n crypto: inside-secure - fix incorrect skcipher output IV\n \n This patch fixes corruption issues with the skcipher output IV\n witnessed on x86+EIP197-FPGA (devboard). The original fix, commit\n 57660b11d5ad (\"crypto: inside-secure - implement IV retrieval\"),\n attempted to write out the result IV through the context record.\n However, this is not a reliable mechanism as there is no way of\n knowing the hardware context update actually arrived in memory, so\n it is possible to read the old contents instead of the updated IV.\n (and indeed, this failed for the x86/FPGA case)\n \n The alternative approach used here recognises the fact that the\n result IV for CBC is actually the last cipher block, which is the last\n input block in case of decryption and the last output block in case\n of encryption. So the result IV is taken from the input data buffer\n respectively the output data buffer instead, which *is* reliable.\n \n Signed-off-by: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 384ce433884af04c2063ec8a0ec0464b3fb6f1d2\nAuthor: Pascal van Leeuwen \u003cpvanleeuwen@insidesecure.com\u003e\nDate: Tue Jul 2 16:39:53 2019 +0200\n\n crypto: inside-secure - silently return -EINVAL for input error cases\n \n Driver was printing an error message for certain input error cases that\n should just return -EINVAL, which caused the related testmgr extra tests\n to flood the kernel message log. Ensured those cases remain silent while\n making some other device-specific errors a bit more verbose.\n \n Signed-off-by: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 36160aadb1034787d4d30b6af7805cb3dd636aac\nAuthor: Pascal van Leeuwen \u003cpvanleeuwen@insidesecure.com\u003e\nDate: Tue Jul 2 16:39:52 2019 +0200\n\n crypto: inside-secure - keep ivsize for DES ECB modes at 0\n \n The driver incorrectly advertised the IV size for DES and 3DES ECB\n mode as being the DES blocksize of 8. This is incorrect as ECB mode\n does not need any IV.\n \n Signed-off-by: Pascal van Leeuwen \u003cpvanleeuwen@verimatrix.com\u003e\n Acked-by: Antoine Tenart \u003cantoine.tenart@bootlin.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 452c53d7868b3f95658ebb2f346acca96eaa2aed\nAuthor: Gilad Ben-Yossef \u003cgilad@benyossef.com\u003e\nDate: Tue Jul 2 14:39:21 2019 +0300\n\n crypto: ccree - notify TEE on FIPS tests errors\n \n Register a FIPS test failure notifier and use it to notify\n TEE side of FIPS test failures on our side prior to panic.\n \n Signed-off-by: Gilad Ben-Yossef \u003cgilad@benyossef.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 9552389c465ed1ded39edf4a5642a861b53c2955\nAuthor: Gilad Ben-Yossef \u003cgilad@benyossef.com\u003e\nDate: Tue Jul 2 14:39:20 2019 +0300\n\n crypto: fips - add FIPS test failure notification chain\n \n Crypto test failures in FIPS mode cause an immediate panic, but\n on some system the cryptographic boundary extends beyond just\n the Linux controlled domain.\n \n Add a simple atomic notification chain to allow interested parties\n to register to receive notification prior to us kicking the bucket.\n \n Signed-off-by: Gilad Ben-Yossef \u003cgilad@benyossef.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 76a95bd8f9e10cade9c4c8df93b5c20ff45dc0f5\nAuthor: Gilad Ben-Yossef \u003cgilad@benyossef.com\u003e\nDate: Tue Jul 2 14:39:19 2019 +0300\n\n crypto: ccree - account for TEE not ready to report\n \n When ccree driver runs it checks the state of the Trusted Execution\n Environment CryptoCell driver before proceeding. We did not account\n for cases where the TEE side is not ready or not available at all.\n Fix it by only considering TEE error state after sync with the TEE\n side driver.\n \n Signed-off-by: Gilad Ben-Yossef \u003cgilad@benyossef.com\u003e\n Fixes: ab8ec9658f5a (\"crypto: ccree - add FIPS support\")\n CC: stable@vger.kernel.org # v4.17+\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit e6e6600c001c3aa3760aeb83dbd4630063e926ee\nAuthor: Gilad Ben-Yossef \u003cgilad@benyossef.com\u003e\nDate: Tue Jul 2 14:39:18 2019 +0300\n\n crypto: ccree - drop legacy ivgen support\n \n ccree had a mechanism for IV generation which was not compatible\n with the Linux seqiv or echainiv iv generator and was never used\n in any of the upstream versions so drop all the code implementing it.\n \n Signed-off-by: Gilad Ben-Yossef \u003cgilad@benyossef.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit aca24d48cee232a88839c8895ac6fcd2fcad6588\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Mon Jul 1 14:08:14 2019 +0100\n\n crypto: ccree - fix spelling mistake \"configration\" -\u003e \"configuration\"\n \n There is a spelling mistake in a dev_err message. Fix it.\n \n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Herbert Xu \u003cherbert@gondor.apana.org.au\u003e\n\ncommit 1f66c45db3302386ca3bb3b5ed05248434237fc9\nAuthor: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nDate: Fri Jun 21 10:52:48 2019 +0100\n\n mips: Add clock_gettime64 entry point\n \n With the release of Linux 5.1 has been added a new syscall,\n clock_gettime64, that provided a 64 bit time value for a specified\n clock_ID to make the kernel Y2038 safe on 32 bit architectures.\n \n Update the mips32 specific vDSO library accordingly with what it has\n been done for the kernel syscall exposing the clock_gettime64 entry\n point.\n \n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: Paul Burton \u003cpaul.burton@mips.com\u003e\n Signed-off-by: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit abed3d826f2fb723c61b1251720348a42357c4e5\nAuthor: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nDate: Fri Jun 21 10:52:47 2019 +0100\n\n mips: Add clock_getres entry point\n \n The generic vDSO library provides an implementation of clock_getres()\n that can be leveraged by each architecture.\n \n Add clock_getres() entry point on mips.\n \n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: Paul Burton \u003cpaul.burton@mips.com\u003e\n Signed-off-by: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit 24640f233b466051ad3a5d2786d2951e43026c9d\nAuthor: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\nDate: Fri Jun 21 10:52:46 2019 +0100\n\n mips: Add support for generic vDSO\n \n The mips vDSO library requires some adaptations to take advantage of the\n newly introduced generic vDSO library.\n \n Introduce the following changes:\n - Modification of vdso.c to be compliant with the common vdso datapage\n - Use of lib/vdso for gettimeofday\n \n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: Paul Burton \u003cpaul.burton@mips.com\u003e\n Signed-off-by: Vincenzo Frascino \u003cvincenzo.frascino@arm.com\u003e\n [paul.burton@mips.com: Prepend $(src) to config-n32-o32-env.c path.]\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit e853ae776a58633492b59badab04f53a6b730d62\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Thu Jul 25 15:52:31 2019 -0700\n\n selftests/bpf: support BPF_FLOW_DISSECTOR_F_STOP_AT_ENCAP\n \n Exit as soon as we found that packet is encapped when\n BPF_FLOW_DISSECTOR_F_STOP_AT_ENCAP is passed.\n Add appropriate selftest cases.\n \n v2:\n * Subtract sizeof(struct iphdr) from .iph_inner.tot_len (Willem de Bruijn)\n \n Acked-by: Petar Penkov \u003cppenkov@google.com\u003e\n Acked-by: Willem de Bruijn \u003cwillemb@google.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Cc: Song Liu \u003csongliubraving@fb.com\u003e\n Cc: Willem de Bruijn \u003cwillemb@google.com\u003e\n Cc: Petar Penkov \u003cppenkov@google.com\u003e\n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 71c99e32b926159ea628352751f66383d7d04d17\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Thu Jul 25 15:52:30 2019 -0700\n\n bpf/flow_dissector: support ipv6 flow_label and BPF_FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL\n \n Add support for exporting ipv6 flow label via bpf_flow_keys.\n Export flow label from bpf_flow.c and also return early when\n BPF_FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL is passed.\n \n Acked-by: Petar Penkov \u003cppenkov@google.com\u003e\n Acked-by: Willem de Bruijn \u003cwillemb@google.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Cc: Song Liu \u003csongliubraving@fb.com\u003e\n Cc: Willem de Bruijn \u003cwillemb@google.com\u003e\n Cc: Petar Penkov \u003cppenkov@google.com\u003e\n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit ae173a915785e55574c1fc54edf58b9b87b28c22\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Thu Jul 25 15:52:29 2019 -0700\n\n selftests/bpf: support BPF_FLOW_DISSECTOR_F_PARSE_1ST_FRAG\n \n bpf_flow.c: exit early unless BPF_FLOW_DISSECTOR_F_PARSE_1ST_FRAG is\n passed in flags. Also, set ip_proto earlier, this makes sure we have\n correct value with fragmented packets.\n \n Add selftest cases to test ipv4/ipv6 fragments and skip eth_get_headlen\n tests that don\u0027t have BPF_FLOW_DISSECTOR_F_PARSE_1ST_FRAG flag.\n \n eth_get_headlen calls flow dissector with\n BPF_FLOW_DISSECTOR_F_PARSE_1ST_FRAG flag so we can\u0027t run tests that\n have different set of input flags against it.\n \n v2:\n * sefltests -\u003e selftests (Willem de Bruijn)\n * Reword a comment about eth_get_headlen flags (Song Liu)\n \n Acked-by: Petar Penkov \u003cppenkov@google.com\u003e\n Acked-by: Willem de Bruijn \u003cwillemb@google.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Cc: Song Liu \u003csongliubraving@fb.com\u003e\n Cc: Willem de Bruijn \u003cwillemb@google.com\u003e\n Cc: Petar Penkov \u003cppenkov@google.com\u003e\n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 57debff23c4cd47f25850a5f42a903aebe535e95\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Thu Jul 25 15:52:28 2019 -0700\n\n tools/bpf: sync bpf_flow_keys flags\n \n Export bpf_flow_keys flags to tools/libbpf/selftests.\n \n Acked-by: Petar Penkov \u003cppenkov@google.com\u003e\n Acked-by: Willem de Bruijn \u003cwillemb@google.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Cc: Song Liu \u003csongliubraving@fb.com\u003e\n Cc: Willem de Bruijn \u003cwillemb@google.com\u003e\n Cc: Petar Penkov \u003cppenkov@google.com\u003e\n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit b2ca4e1cfa7d3d755e1ec637d1235f89af9bd01f\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Thu Jul 25 15:52:27 2019 -0700\n\n bpf/flow_dissector: support flags in BPF_PROG_TEST_RUN\n \n This will allow us to write tests for those flags.\n \n v2:\n * Swap kfree(data) and kfree(user_ctx) (Song Liu)\n \n Acked-by: Petar Penkov \u003cppenkov@google.com\u003e\n Acked-by: Willem de Bruijn \u003cwillemb@google.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Cc: Song Liu \u003csongliubraving@fb.com\u003e\n Cc: Willem de Bruijn \u003cwillemb@google.com\u003e\n Cc: Petar Penkov \u003cppenkov@google.com\u003e\n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 1ac6b126dbe8108c1fe6e29d023a0f2989cd3fea\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Thu Jul 25 15:52:26 2019 -0700\n\n bpf/flow_dissector: document flags\n \n Describe what each input flag does and who uses it.\n \n Acked-by: Petar Penkov \u003cppenkov@google.com\u003e\n Acked-by: Willem de Bruijn \u003cwillemb@google.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Cc: Song Liu \u003csongliubraving@fb.com\u003e\n Cc: Willem de Bruijn \u003cwillemb@google.com\u003e\n Cc: Petar Penkov \u003cppenkov@google.com\u003e\n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 086f95682114fd2d1790bd3226e76cbae9a2d192\nAuthor: Stanislav Fomichev \u003csdf@google.com\u003e\nDate: Thu Jul 25 15:52:25 2019 -0700\n\n bpf/flow_dissector: pass input flags to BPF flow dissector program\n \n C flow dissector supports input flags that tell it to customize parsing\n by either stopping early or trying to parse as deep as possible. Pass\n those flags to the BPF flow dissector so it can make the same\n decisions. In the next commits I\u0027ll add support for those flags to\n our reference bpf_flow.c\n \n v3:\n * Export copy of flow dissector flags instead of moving (Alexei Starovoitov)\n \n Acked-by: Petar Penkov \u003cppenkov@google.com\u003e\n Acked-by: Willem de Bruijn \u003cwillemb@google.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Cc: Song Liu \u003csongliubraving@fb.com\u003e\n Cc: Willem de Bruijn \u003cwillemb@google.com\u003e\n Cc: Petar Penkov \u003cppenkov@google.com\u003e\n Signed-off-by: Stanislav Fomichev \u003csdf@google.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 03cd1d1a493e92a80d60040d6aa8160aff239942\nAuthor: Allan Zhang \u003callanzhang@google.com\u003e\nDate: Tue Jul 23 17:07:25 2019 -0700\n\n selftests/bpf: Add selftests for bpf_perf_event_output\n \n Software event output is only enabled by a few prog types.\n This test is to ensure that all supported types are enabled for\n bpf_perf_event_output successfully.\n \n Signed-off-by: Allan Zhang \u003callanzhang@google.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 7c4b90d79d0f4ee6f2c01a114ef0a83a09dfc900\nAuthor: Allan Zhang \u003callanzhang@google.com\u003e\nDate: Tue Jul 23 17:07:24 2019 -0700\n\n bpf: Allow bpf_skb_event_output for a few prog types\n \n Software event output is only enabled by a few prog types right now (TC,\n LWT out, XDP, sockops). Many other skb based prog types need\n bpf_skb_event_output to produce software event.\n \n Added socket_filter, cg_skb, sk_skb prog types to generate sw event.\n \n Test bpf code is generated from code snippet:\n \n struct TMP {\n uint64_t tmp;\n } tt;\n tt.tmp \u003d 5;\n bpf_perf_event_output(skb, \u0026connection_tracking_event_map, 0,\n \u0026tt, sizeof(tt));\n return 1;\n \n the bpf assembly from llvm is:\n 0: b7 02 00 00 05 00 00 00 r2 \u003d 5\n 1: 7b 2a f8 ff 00 00 00 00 *(u64 *)(r10 - 8) \u003d r2\n 2: bf a4 00 00 00 00 00 00 r4 \u003d r10\n 3: 07 04 00 00 f8 ff ff ff r4 +\u003d -8\n 4: 18 02 00 00 00 00 00 00 00 00 00 00 00 00 00 00 r2 \u003d 0ll\n 6: b7 03 00 00 00 00 00 00 r3 \u003d 0\n 7: b7 05 00 00 08 00 00 00 r5 \u003d 8\n 8: 85 00 00 00 19 00 00 00 call 25\n 9: b7 00 00 00 01 00 00 00 r0 \u003d 1\n 10: 95 00 00 00 00 00 00 00 exit\n \n Signed-off-by: Allan Zhang \u003callanzhang@google.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 00c33afbf9dd06f77a2f15117cd4bdc2a54b51d7\nAuthor: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\nDate: Thu Jul 25 07:48:04 2019 +0000\n\n net: mvneta: use devm_platform_ioremap_resource() to simplify code\n \n devm_platform_ioremap_resource() wraps platform_get_resource() and\n devm_ioremap_resource() in a single helper, let\u0027s use that helper to\n simplify the code.\n \n Signed-off-by: Jisheng Zhang \u003cJisheng.Zhang@synaptics.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 97dee74bb34a1364c6b3f3f2e3f31165063d3c6e\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Thu Jul 25 21:03:14 2019 +0000\n\n drm/i915/uc: Reorder params in intel_uc_fw_fetch\n \n All intel_uc_fw_* functions are taking uc_fw as first param\n except intel_uc_fw_fetch() which is taking i915. Fix that.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725210314.21188-1-michal.wajdeczko@intel.com\n\ncommit 79c7a28e1f3a74b95ae2eae36ed0046fc8e6c7fd\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Jul 25 23:38:43 2019 +0100\n\n drm/i915: Capture vma contents outside of spinlock\n \n Currently we use the engine-\u003eactive.lock to ensure that the request is\n not retired as we capture the data. However, we only need to ensure that\n the vma are not removed prior to use acquiring their contents, and\n since we have already relinquished our stop-machine protection, we\n assume that the user will not be overwriting the contents before we are\n able to record them.\n \n In order to capture the vma outside of the spinlock, we acquire a\n reference and mark the vma as active to prevent it from being unbound.\n However, since it is tricky allocate an entry in the fence tree (doing\n so would require taking a mutex) while inside the engine spinlock, we\n use an atomic bit and special case the handling for i915_active_wait.\n \n The core benefit is that we can use some non-atomic methods for mapping\n the device pages, we can remove the slow compression phase out of atomic\n context (i.e. stop antagonising the nmi-watchdog), and no we longer need\n large reserves of atomic pages.\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111215\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.william.auld@gmail.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725223843.8971-1-chris@chris-wilson.co.uk\n\ncommit 2320bcdae62887555701ea78a46b640ff6b63868\nAuthor: Tuong Lien \u003ctuong.t.lien@dektech.com.au\u003e\nDate: Wed Jul 24 08:56:12 2019 +0700\n\n tipc: fix changeover issues due to large packet\n \n In conjunction with changing the interfaces\u0027 MTU (e.g. especially in\n the case of a bonding) where the TIPC links are brought up and down\n in a short time, a couple of issues were detected with the current link\n changeover mechanism:\n \n 1) When one link is up but immediately forced down again, the failover\n procedure will be carried out in order to failover all the messages in\n the link\u0027s transmq queue onto the other working link. The link and node\n state is also set to FAILINGOVER as part of the process. The message\n will be transmited in form of a FAILOVER_MSG, so its size is plus of 40\n bytes (\u003d the message header size). There is no problem if the original\n message size is not larger than the link\u0027s MTU - 40, and indeed this is\n the max size of a normal payload messages. However, in the situation\n above, because the link has just been up, the messages in the link\u0027s\n transmq are almost SYNCH_MSGs which had been generated by the link\n synching procedure, then their size might reach the max value already!\n When the FAILOVER_MSG is built on the top of such a SYNCH_MSG, its size\n will exceed the link\u0027s MTU. As a result, the messages are dropped\n silently and the failover procedure will never end up, the link will\n not be able to exit the FAILINGOVER state, so cannot be re-established.\n \n 2) The same scenario above can happen more easily in case the MTU of\n the links is set differently or when changing. In that case, as long as\n a large message in the failure link\u0027s transmq queue was built and\n fragmented with its link\u0027s MTU \u003e the other link\u0027s one, the issue will\n happen (there is no need of a link synching in advance).\n \n 3) The link synching procedure also faces with the same issue but since\n the link synching is only started upon receipt of a SYNCH_MSG, dropping\n the message will not result in a state deadlock, but it is not expected\n as design.\n \n The 1) \u0026 3) issues are resolved by the last commit that only a dummy\n SYNCH_MSG (i.e. without data) is generated at the link synching, so the\n size of a FAILOVER_MSG if any then will never exceed the link\u0027s MTU.\n \n For the 2) issue, the only solution is trying to fragment the messages\n in the failure link\u0027s transmq queue according to the working link\u0027s MTU\n so they can be failovered then. A new function is made to accomplish\n this, it will still be a TUNNEL PROTOCOL/FAILOVER MSG but if the\n original message size is too large, it will be fragmented \u0026 reassembled\n at the receiving side.\n \n Acked-by: Ying Xue \u003cying.xue@windriver.com\u003e\n Acked-by: Jon Maloy \u003cjon.maloy@ericsson.com\u003e\n Signed-off-by: Tuong Lien \u003ctuong.t.lien@dektech.com.au\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4929a932be334d68d333089872bc67e4f1d97475\nAuthor: Tuong Lien \u003ctuong.t.lien@dektech.com.au\u003e\nDate: Wed Jul 24 08:56:11 2019 +0700\n\n tipc: optimize link synching mechanism\n \n This commit along with the next one are to resolve the issues with the\n link changeover mechanism. See that commit for details.\n \n Basically, for the link synching, from now on, we will send only one\n single (\"dummy\") SYNCH message to peer. The SYNCH message does not\n contain any data, just a header conveying the synch point to the peer.\n \n A new node capability flag (\"TIPC_TUNNEL_ENHANCED\") is introduced for\n backward compatible!\n \n Acked-by: Ying Xue \u003cying.xue@windriver.com\u003e\n Acked-by: Jon Maloy \u003cjon.maloy@ericsson.com\u003e\n Suggested-by: Jon Maloy \u003cjon.maloy@ericsson.com\u003e\n Signed-off-by: Tuong Lien \u003ctuong.t.lien@dektech.com.au\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 67fe62dcf713c36f4766c0218cc14796ee9536e1\nAuthor: Yue Hu \u003chuyue2@yulong.com\u003e\nDate: Thu Jul 25 11:52:39 2019 +0800\n\n drm: Switch to use DEVFREQ_GOV_SIMPLE_ONDEMAND constant\n \n Since governor name is defined by DEVFREQ framework internally, use the\n macro definition instead of using the name directly.\n \n Signed-off-by: Yue Hu \u003chuyue2@yulong.com\u003e\n Reviewed-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n Acked-by: Jordan Crouse \u003cjcrouse@codeaurora.org\u003e for the msm part.\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725035239.1192-1-zbestahu@gmail.com\n\ncommit 4bced8bea0943bb3c4b97012d2221d3aca5b9384\nAuthor: Steven Price \u003csteven.price@arm.com\u003e\nDate: Wed Jul 24 11:56:26 2019 +0100\n\n drm/panfrost: Export all GPU feature registers\n \n Midgard/Bifrost GPUs have a bunch of feature registers providing details\n of what the hardware supports. Panfrost already reads these, this patch\n exports them all to user space so that the jobs created by the user space\n driver can be tuned for the particular hardware implementation.\n \n Signed-off-by: Steven Price \u003csteven.price@arm.com\u003e\n Reviewed-by: Alyssa Rosenzweig \u003calyssa.rosenzweig@collabora.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190724105626.53552-1-steven.price@arm.com\n\ncommit a1b27e99229a448075abc3ca006043d4785f14a4\nAuthor: Leo Li \u003csunpeng.li@amd.com\u003e\nDate: Tue Jul 23 19:28:08 2019 -0400\n\n drm/amd/display: Implement MST Aux device registration\n \n Implement late_register and early_unregister hooks for MST connectors.\n Call drm helpers for MST connector registration, which registers the\n AUX devices.\n \n Cc: Jerry Zuo \u003cJerry.Zuo@amd.com\u003e\n Cc: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Cc: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Reviewed-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Signed-off-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190723232808.28128-10-sunpeng.li@amd.com\n\ncommit db432cd0960cd32d02ac8ef64deeed6dfe75b7cd\nAuthor: Leo Li \u003csunpeng.li@amd.com\u003e\nDate: Tue Jul 23 19:28:04 2019 -0400\n\n drm/amd/display: Use connector kdev as aux device parent\n \n Set the connector\u0027s kernel device as the parent for the aux kernel\n device. This allows udev rules to access connector attributes when\n creating symlinks to aux devices.\n \n For example, the following udev rule:\n \n SUBSYSTEM\u003d\u003d\"drm_dp_aux_dev\", SUBSYSTEMS\u003d\u003d\"drm\", ATTRS{edid}\u003d\u003d\"*\",\n SYMLINK+\u003d\"drm_dp_aux/by-name/$id\"\n \n Will create the following symlinks using the connector\u0027s name:\n \n $ ls /dev/drm_dp_aux/by-name/\n card0-DP-1 card0-DP-2 card0-DP-3\n \n Cc: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Cc: Jerry (Fangzhi) Zuo \u003cJerry.Zuo@amd.com\u003e\n Cc: Harry Wentland \u003charry.wentland@amd.com\u003e\n Signed-off-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Reviewed-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190723232808.28128-6-sunpeng.li@amd.com\n\ncommit 7713c0f10dd52ff91a136566b91e7bcaabd4bda9\nAuthor: Leo Li \u003csunpeng.li@amd.com\u003e\nDate: Tue Jul 23 19:28:02 2019 -0400\n\n drm/nouveau: Use connector kdev as aux device parent\n \n Set the connector\u0027s kernel device as the parent for the aux kernel\n device. This allows udev rules to access connector attributes when\n creating symlinks to aux devices.\n \n Cc: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n Signed-off-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Reviewed-by: Lyude Paul \u003clyude@redhat.com\u003e\n Signed-off-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190723232808.28128-4-sunpeng.li@amd.com\n\ncommit 4a11458611fb17b4f21e5e7b2f24445c00e900e1\nAuthor: Matthias Kaehlcke \u003cmka@chromium.org\u003e\nDate: Thu Jul 25 09:26:42 2019 -0700\n\n ARM: dts: rockchip: add veyron-tiger board\n \n Also known as the AOpen Chromebase Mini.\n \n tiger and fievel are share the same board, tiger has a display and\n touchscreen, fievel not. Use the fievel .dts as base and add the\n extra bits.\n \n Signed-off-by: Matthias Kaehlcke \u003cmka@chromium.org\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit 0067692b662e7dd99afa2f871e79e3e5aa5c3c5b\nAuthor: Matthias Kaehlcke \u003cmka@chromium.org\u003e\nDate: Thu Jul 25 09:26:41 2019 -0700\n\n ARM: dts: rockchip: add veyron-fievel board\n \n Also known as AOpen Chromebox Mini.\n \n Signed-off-by: Matthias Kaehlcke \u003cmka@chromium.org\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit c6855f7bb66b6c0c53acc15d2012ebd15a681159\nAuthor: Matthias Kaehlcke \u003cmka@chromium.org\u003e\nDate: Thu Jul 25 09:26:40 2019 -0700\n\n dt-bindings: ARM: dts: rockchip: Add bindings for rk3288-veyron-{fievel,tiger}\n \n Fievel is a Chromebox and Tiger a Chromebase with a 10\" display and\n touchscreen. Tiger and Fievel are based on the same board.\n \n Signed-off-by: Matthias Kaehlcke \u003cmka@chromium.org\u003e\n Reviewed-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit 562836a269e363cdb74b551e3be7021c9d228378\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Tue Jul 23 19:28:01 2019 -0400\n\n drm/dp_mst: Enable registration of AUX devices for MST ports\n \n All available downstream ports - physical and logical - are exposed for\n each MST device. They are listed in /dev/, following the same naming\n scheme as SST devices by appending an incremental ID.\n \n Although all downstream ports are exposed, only some will work as\n expected. Consider the following topology:\n \n +---------+\n | ASIC |\n +---------+\n Conn-0|\n |\n +----v----+\n +----| MST HUB |----+\n | +---------+ |\n | |\n |Port-1 Port-2|\n +-----v-----+ +-----v-----+\n | MST | | SST |\n | Display | | Display |\n +-----------+ +-----------+\n |Port-1\n x\n \n MST Path | MST Device\n ----------+----------------------------------\n sst:0 | MST Hub\n mst:0-1 | MST Display\n mst:0-1-1 | MST Display\u0027s disconnected DP out\n mst:0-1-8 | MST Display\u0027s internal sink\n mst:0-2 | SST Display\n \n On certain MST displays, the upstream physical port will ACK DPCD reads.\n However, reads on the local logical port to the internal sink will\n *NAK*. i.e. reading mst:0-1 ACKs, but mst:0-1-8 NAKs.\n \n There may also be duplicates. Some displays will return the same GUID\n when reading DPCD from both mst:0-1 and mst:0-1-8.\n \n There are some device-dependent behavior as well. The MST hub used\n during testing will actually *ACK* read requests on a disconnected\n physical port, whereas the MST displays will NAK.\n \n In light of these discrepancies, it\u0027s simpler to expose all downstream\n ports - both physical and logical - and let the user decide what to use.\n \n v3 changes:\n * Change WARN_ON_ONCE -\u003e DRM_ERROR on dpcd read errors\n * Docstring and cosmetic fixes\n \n v2 changes:\n \n Moved remote aux device (un)registration to new mst connector late\n register and early unregister helpers. Drivers should call these from\n their own mst connector function hooks.\n \n This is to solve an issue during driver unload, where mst connector\n devices are unregistered before the remote aux devices are. In a setup\n where aux devices are created as children of connector devices, the aux\n device would be removed too early, and uncleanly. Doing so in\n early_unregister solves this issue, as that is called before connector\n unregistration.\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Reviewed-by: Lyude Paul \u003clyude@redhat.com\u003e\n Signed-off-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190723232808.28128-3-sunpeng.li@amd.com\n\ncommit 6b381a8e2c5f9667918cc1ed0e3daeec869b76b1\nAuthor: Matthias Kaehlcke \u003cmka@chromium.org\u003e\nDate: Thu Jul 25 09:26:39 2019 -0700\n\n ARM: dts: rockchip: consolidate veyron panel and backlight settings\n \n veyron jaq, jerry, minnie and speedy have mostly redundant regulator\n and pinctrl configurations for the panel/backlight. Consolidate these\n pieces in the eDP .dtsi.\n \n Also change the default power supply for the panel to\n \u0027panel_regulator\u0027, instead of overriding it in all the board files.\n pinky is the only device that uses \u0027vcc33_lcd\u0027 (the prior default),\n so overwrite it in this case. pinky doesn\u0027t have a complete display\n configuration, to keep things as they were delete the common nodes\n that didn\u0027t exist previously in pinky\u0027s board file.\n \n Reviewed-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n Signed-off-by: Matthias Kaehlcke \u003cmka@chromium.org\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit abf30f235390b8da1cd28b73e7ec3f0ff96ac450\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Thu Jul 25 14:13:08 2019 +0000\n\n drm/i915/uc: Move uc firmware layout definitions to dedicated file\n \n Generic uc firmware layout definitions are unlikely to change and\n are separate to other GuC specific definitions.\n \n v2: reordered\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725141308.24660-3-michal.wajdeczko@intel.com\n\ncommit 199ddded380a99940b1f155e19cc885f57bfb852\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Thu Jul 25 14:13:07 2019 +0000\n\n drm/i915/uc: Update drawing for firmware layout\n \n Sphinx was rendering firmware layout as html table, but since\n we want to add sizes relations switch to plain text graphics.\n \n v2: also update text and do it before move (Daniele)\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725141308.24660-2-michal.wajdeczko@intel.com\n\ncommit dbbff8c3f641a2c014ab1e441d822b6fc9844384\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Thu Jul 25 14:13:06 2019 +0000\n\n drm/i915: Fix GuC documentation links\n \n We moved GuC related files to new location but we missed to update\n .rst file with links.\n \n References: commit 0f261b241d9c (\"drm/i915/uc: move GuC and HuC files under gt/uc/\")\n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725141308.24660-1-michal.wajdeczko@intel.com\n\ncommit ab9640000d92d29a342656424d763c138d1dd66b\nAuthor: Matthias Kaehlcke \u003cmka@chromium.org\u003e\nDate: Thu Jul 25 09:26:38 2019 -0700\n\n ARM: dts: rockchip: move rk3288-veryon display settings into a separate file\n \n The chromebook .dtsi file contains common settings for veyron\n Chromebooks with eDP displays. Some veyron devices with a display\n aren\u0027t Chromebooks (e.g. \u0027tiger\u0027 aka \u0027AOpen Chromebase Mini\u0027), move\n display related bits from the chromebook .dtsi into a separate file\n to avoid redundant DT settings.\n \n The new file is included from the chromebook .dtsi and can be\n included by non-Chromebook devices with a display.\n \n Signed-off-by: Matthias Kaehlcke \u003cmka@chromium.org\u003e\n Reviewed-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit 31ed9d9d71ec396d38090a85f5b3dbd6936650b4\nAuthor: Matthias Kaehlcke \u003cmka@chromium.org\u003e\nDate: Tue Jul 23 15:52:58 2019 -0700\n\n ARM: dts: rockchip: Limit WiFi TX power on rk3288-veyron-jerry\n \n The downstream Chrome OS 3.14 kernel for jerry limits WiFi TX power\n through calibration data in the device tree [1]. Add a DT node for\n the WiFi chip and use the downstream calibration data.\n \n Not all calibration data entries have the length specified in the\n binding (Documentation/devicetree/bindings/net/wireless/marvell-8xxx.txt),\n however this is the data used by the downstream (\u0027official\u0027) kernel\n and the binding mentions that \"the length can vary between hw\n versions\".\n \n [1] https://crrev.com/c/271237\n \n Signed-off-by: Matthias Kaehlcke \u003cmka@chromium.org\u003e\n Reviewed-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit 3935ec4a1184a32696251d19d87849981f4d189d\nAuthor: Leo Li \u003csunpeng.li@amd.com\u003e\nDate: Tue Jul 23 19:28:00 2019 -0400\n\n drm/dp: Use non-cyclic idr\n \n In preparation for adding aux devices for DP MST, make the IDR\n non-cyclic. That way, hotplug cycling MST devices won\u0027t needlessly\n increment the minor version index.\n \n Signed-off-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Reviewed-by: Lyude Paul \u003clyude@redhat.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190723232808.28128-2-sunpeng.li@amd.com\n\ncommit 5882d65c169152178e6fb2fa2e5e3a72fcb08248\nAuthor: Andrius Štikonas \u003candrius@stikonas.eu\u003e\nDate: Wed Jul 24 20:17:15 2019 +0100\n\n arm64: dts: rockchip: Add PWM fan for RockPro64\n \n RockPro64 has a dedicated circuit for driving a 12V fan from PWM1.\n \n At the moment this makes fan spin at full speed. fancontrol can be used\n to control fan speed. E.g. the following config file works well:\n \n INTERVAL\u003d10\n DEVPATH\u003dhwmon0\u003ddevices/platform/pwm-fan\n DEVNAME\u003dhwmon0\u003dpwmfan\n FCTEMPS\u003dhwmon0/device/pwm1\u003d../thermal/thermal_zone0/temp\n MINTEMP\u003dhwmon0/device/pwm1\u003d40\n MAXTEMP\u003dhwmon0/device/pwm1\u003d60\n MINSTART\u003dhwmon0/device/pwm1\u003d100\n MINSTOP\u003dhwmon0/device/pwm1\u003d70\n \n In the future it would be nice to define trip points in dts file,\n so that kernel could adjust fan speed itself.\n \n Signed-off-by: Andrius Štikonas \u003candrius@stikonas.eu\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit e709accc7670a07b2e5186449f0640c2416662ec\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Wed Jul 24 08:52:58 2019 +0200\n\n mm/hmm: comment on VM_FAULT_RETRY semantics in handle_mm_fault\n \n The magic dropping of mmap_sem when handle_mm_fault returns VM_FAULT_RETRY\n is rather subtile. Add a comment explaining it.\n \n Link: https://lore.kernel.org/r/20190724065258.16603-8-hch@lst.de\n Tested-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n [hch: wrote a changelog]\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n\ncommit f32471e2cf87112b8f5dc10469b27c39c1a41722\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Wed Jul 24 08:52:57 2019 +0200\n\n mm/hmm: remove the legacy hmm_pfn_* APIs\n \n Switch the one remaining user in nouveau over to its replacement, and\n remove all the wrappers.\n \n Link: https://lore.kernel.org/r/20190724065258.16603-7-hch@lst.de\n Tested-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 1b88b99bdff85e75df1cfaa418ed2ef8e1bffeb2\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Wed Jul 24 08:52:56 2019 +0200\n\n nouveau: return -EBUSY when hmm_range_wait_until_valid fails\n \n -EAGAIN has a magic meaning for non-blocking faults, so don\u0027t overload it.\n Given that the caller doesn\u0027t check for specific error codes this change\n is purely cosmetic.\n \n Link: https://lore.kernel.org/r/20190724065258.16603-6-hch@lst.de\n Tested-by: Ralph Campbell \u003crcampbell@nvidia.com\u003e\n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 7114fbdaa2aad9ff00c4acc1fbba3eb07bac8889\nAuthor: Nathan Huckleberry \u003cnhuck@google.com\u003e\nDate: Thu Jun 27 15:22:20 2019 -0700\n\n clk: rockchip: Fix -Wunused-const-variable in rv1108 clk driver\n \n Clang produces the following warning\n \n drivers/clk/rockchip/clk-rv1108.c:125:7: warning: unused variable\n \u0027mux_pll_src_3plls_p\u0027 [-Wunused-const-variable]\n PNAME(mux_pll_src_3plls_p) \u003d { \"apll\", \"gpll\", \"dpll\" };\n \n Looks like this variable was never used. Deleting it to remove the\n warning.\n \n Cc: clang-built-linux@googlegroups.com\n Link: https://github.com/ClangBuiltLinux/linux/issues/524\n Signed-off-by: Nathan Huckleberry \u003cnhuck@google.com\u003e\n Reviewed-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit 37f7c66f4560b154cfa7a8d8f1902e222e38adba\nAuthor: Ding Xiang \u003cdingxiang@cmss.chinamobile.com\u003e\nDate: Tue Jul 23 16:54:05 2019 +0800\n\n ptp: ptp_dte: remove redundant dev_err message\n \n devm_ioremap_resource already contains error message, so remove\n the redundant dev_err message\n \n Signed-off-by: Ding Xiang \u003cdingxiang@cmss.chinamobile.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit fc25996e6f46ac05378f45691d9c6ea08c2037b9\nAuthor: Ido Schimmel \u003cidosch@mellanox.com\u003e\nDate: Tue Jul 23 10:57:42 2019 +0300\n\n mlxsw: spectrum_router: Increase scale of IPv6 nexthop groups\n \n Unlike IPv4, the kernel does not consolidate IPv6 nexthop groups. To\n avoid exhausting the device\u0027s adjacency table - where nexthops are\n stored - the driver does this consolidation instead.\n \n Each nexthop group is hashed by XOR-ing the interface indexes of all the\n member nexthop devices. However, the ifindex itself is not hashed, which\n can result in identical keys used for different groups and finally an\n -EBUSY error from rhashtable due to too long objects list.\n \n Improve the situation by hashing the ifindex itself.\n \n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b06689cc1b0446b959ed1a41b4f8dd892aa6b331\nAuthor: Amit Cohen \u003camitc@mellanox.com\u003e\nDate: Tue Jul 23 10:57:41 2019 +0300\n\n mlxsw: spectrum: Expose KVD size for Spectrum-2\n \n Unlike Spectrum-1, the KVD (Key-value database) of Spectrum-2 is not\n partitioned, so only expose the entire KVD size. This enables users to\n query the total size of the KVD.\n \n Signed-off-by: Amit Cohen \u003camitc@mellanox.com\u003e\n Acked-by: Jiri Pirko \u003cjiri@mellanox.com\u003e\n Signed-off-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 3123c5c4ca157edc518102ee5385f60cb90f93f5\nAuthor: Thor Thayer \u003cthor.thayer@linux.intel.com\u003e\nDate: Fri Jul 12 13:28:43 2019 -0500\n\n edac: altera: Move Stratix10 SDRAM ECC to peripheral\n \n ARM32 SoCFPGAs had separate IRQs for SDRAM. ARM64 SoCFPGAs\n send all DBEs to SError so filtering by source is necessary.\n \n The Stratix10 SDRAM ECC is a better match with the generic\n Altera peripheral ECC framework because the linked list can\n be searched to find the ECC block offset and printout\n the DBE Address.\n \n Signed-off-by: Thor Thayer \u003cthor.thayer@linux.intel.com\u003e\n Acked-by: James Morse \u003cjames.morse@arm.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit c93496e9f7680741c25bab866e729e9606bfae3e\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:35 2019 +0200\n\n net: sfc: falcon: convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(). So, we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Acked-by: Edward Cree \u003cecree@solarflare.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 246eab59eefc26706e59520cf20d2315377efa08\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 23:14:59 2019 +0300\n\n leds: lm3697: Switch to use fwnode_property_count_uXX()\n \n Use fwnode_property_count_uXX() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit ebefec8d9fc07eafa7434df1a1c2447b0ea96e50\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 23:14:58 2019 +0300\n\n leds: lm36274: Switch to use fwnode_property_count_uXX()\n \n Use fwnode_property_count_uXX() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit cc93c86370086519f69aefbbac18375f0b489a0e\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 23:14:57 2019 +0300\n\n leds: lm3532: Switch to use fwnode_property_count_uXX()\n \n Use fwnode_property_count_uXX() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 1817208e14c6f7b451c64d7127e130e5ed926f54\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n leds: an30259a: Use generic support for composing LED names\n \n Switch to using generic LED support for composing LED class\n device name.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Cc: Simon Shields \u003csimon@lineageos.org\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n\ncommit 91f1be8a83273cd9069e07a7f65955c6d5dfad6a\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n dt-bindings: an30259a: Add function and color properties\n \n Refer to new \"function\" and \"color\" properties and mark \"label\"\n as deprecated.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Cc: Simon Shields \u003csimon@lineageos.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit d7235f5feaa0d53e27a369a3198827c921103c24\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n leds: gpio: Use generic support for composing LED names\n \n Switch to using generic LED support for composing LED class\n device name.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Cc: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n\ncommit a1c60ec95d15e50aef47e22197181dc6fd1dd01f\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n dt-bindings: leds-gpio: Add function and color properties\n \n Refer to new \"function\" and \"color\" properties and mark \"label\"\n as deprecated.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Cc: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit b1f24498f187cf807faf3b677ffee183a6579a20\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n leds: as3645a: Use generic support for composing LED names\n \n Switch to using generic LED support for composing LED class\n device name.\n \n After switching to new led_classdev_register_ext() the validity\n of struct led_classdev\u0027s name property is no longer guaranteed,\n and therefore rely on struct device\u0027s kobj.name instead.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Cc: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n\ncommit 5b95cb0cbf6451d80c1385e1e3198e2730a3a59c\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n dt-bindings: as3645a: Add function and color properties\n \n Refer to new \"function\" and \"color\" properties and mark \"label\"\n as deprecated.\n \n Also, fix malformed syntax of address-cells and size-cells\n in the example.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Cc: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 8103f558527f77e7d36e11cddfbef8bf8c4bae13\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n leds: aat1290: Use generic support for composing LED names\n \n Switch to using generic LED support for composing LED class\n device name.\n \n After switching to new led_classdev_register_ext() the validity\n of struct led_classdev\u0027s name property is no longer guaranteed,\n and therefore rely on struct device\u0027s kobj.name instead.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n\ncommit d60a3d30c2858872ba97093689a741f1966b4b00\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n dt-bindings: aat1290: Add function and color properties\n \n Refer to new \"function\" and \"color\" properties and mark \"label\"\n as deprecated.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 889003c20602bbebc82a6f32e24e7d0785a7b324\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n leds: cr0014114: Use generic support for composing LED names\n \n Switch to using generic LED support for composing LED class\n device name.\n \n At this occassion remove initialization of struct led_classdev\u0027s\n dev-\u003eof_node property in the driver, since now it will be taken from\n fwnode assigned to struct led_init_data and passed to the new\n devm_led_classdev_register_ext() API.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Cc: Oleh Kravchenko \u003coleg@kaa.org.ua\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n\ncommit 6c01a5cc36ea9245816df8bd89245116f34fc0cb\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n dt-bindings: cr0014114: Add function and color properties\n \n Refer to new \"function\" and \"color\" properties and mark \"label\"\n as deprecated.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Cc: Oleh Kravchenko \u003coleg@kaa.org.ua\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit e322b75a301b8cc3915f1e1cbd0e48958b717f6c\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n leds: lm3601x: Use generic support for composing LED names\n \n Switch to using generic LED support for composing LED class\n device name.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Tested-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n\ncommit 28e12cf498ec39f23cd719b37f123301ab7ab7aa\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n dt-bindings: lm36010: Add function and color properties\n \n Refer to new \"function\" and \"color\" properties and mark \"label\"\n as deprecated.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Reviewed-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit a50ff28348934913c46feb7945571329e46c70b3\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n leds: lm3692x: Use generic support for composing LED names\n \n Switch to using generic LED support for composing LED class\n device name.\n \n At this occassion remove initialization of struct led_classdev\u0027s\n dev-\u003eof_node property in the driver, since now it will be taken from\n fwnode assigned to struct led_init_data and passed to the new\n devm_led_classdev_register_ext() API.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Tested-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n\ncommit 4dcbc8f8c59f4b618d651f5ba884ee5bf562c8de\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n dt-bindings: lm3692x: Add function and color properties\n \n Refer to new \"function\" and \"color\" properties and mark \"label\"\n as deprecated.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Reviewed-by: Dan Murphy \u003cdmurphy@ti.com\u003eA\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 99ca0ea5730971035caa847e5cacf324a2b5598a\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n leds: lp8860: Use generic support for composing LED names\n \n Switch to using generic LED support for composing LED class\n device name.\n \n While at it, avoid iterating through available child of nodes\n in favor of obtaining single expected child node using single\n call to of_get_next_available_child().\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Tested-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n\ncommit 0ec39e8cd8f60fa0b93c298a48c3a9279f70375b\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n dt-bindings: lp8860: Add function and color properties\n \n Refer to new \"function\" and \"color\" properties and mark \"label\"\n as deprecated.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n\ncommit 68e192078e022b8dc0cec57cfdd0e369f62d7ea5\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n leds: lt3593: Use generic support for composing LED names\n \n Switch to using generic LED support for composing LED class\n device name.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Cc: Daniel Mack \u003cdaniel@zonque.org\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n\ncommit e12e91a4129be7a11e7b6912910c6fb1a71a0fb3\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n dt-bindings: lt3593: Add function and color properties\n \n Refer to new \"function\" and \"color\" properties and mark \"label\"\n as deprecated.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Cc: Daniel Mack \u003cdaniel@zonque.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 5fdf85a0911a4109abbb2c006da7d8407cb60078\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n leds: sc27xx-blt: Use generic support for composing LED names\n \n Switch to using generic LED support for composing LED class\n device name.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Reviewed-by: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n\ncommit 0f0f25797a7d20e9d969a46ba952a329b270cf09\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n dt-bindings: sc27xx-blt: Add function and color properties\n \n Refer to new \"function\" and \"color\" properties and mark \"label\"\n as deprecated.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Cc: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit bb4e9af0348dfeafd66c7e7f82e8a0983fe5390c\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n leds: core: Add support for composing LED class device names\n \n Add generic support for composing LED class device name. The newly\n introduced led_compose_name() function composes device name according\n to either \u003ccolor:function\u003e or \u003cdevicename:color:function\u003e pattern,\n depending on the configuration of initialization data.\n \n Backward compatibility with in-driver hard-coded LED class device\n names is assured thanks to the default_label and devicename properties\n of newly introduced struct led_init_data.\n \n In case none of the aforementioned properties was found, then, for OF\n nodes, the node name is adopted for LED class device name.\n \n At the occassion of amending the Documentation/leds/leds-class.txt\n unify spelling: colour -\u003e color.\n \n Alongside these changes added is a new tool - tools/leds/get_led_device_info.sh.\n The tool allows retrieving details of a LED class device\u0027s parent device,\n which proves that using vendor or product name for devicename part\n of LED name doesn\u0027t convey any added value since that information had been\n already available in sysfs. The script performs also basic validation\n of a LED class device name.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Cc: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Cc: Dan Murphy \u003cdmurphy@ti.com\u003e\n Cc: Daniel Mack \u003cdaniel@zonque.org\u003e\n Cc: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Cc: Oleh Kravchenko \u003coleg@kaa.org.ua\u003e\n Cc: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Cc: Simon Shields \u003csimon@lineageos.org\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n\ncommit c5d18dd6b64e09dd6984bda9bdd55160af537a8c\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n dt-bindings: leds: Add properties for LED name construction\n \n Introduce dedicated properties for conveying information about\n LED function and color. Mark old \"label\" property as deprecated.\n \n Additionally function-enumerator property is being provided\n for the cases when neither function nor color can be used\n for LED differentiation.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Cc: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Cc: Pavel Machek \u003cpavel@ucw.cz\u003e\n Cc: Dan Murphy \u003cdmurphy@ti.com\u003e\n Cc: Daniel Mack \u003cdaniel@zonque.org\u003e\n Cc: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Cc: Oleh Kravchenko \u003coleg@kaa.org.ua\u003e\n Cc: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Cc: Simon Shields \u003csimon@lineageos.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 2f430310f7b4216408aaae57c55a596091c47846\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n dt-bindings: leds: Add LED_FUNCTION definitions\n \n Add initial set of common LED function definitions.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Cc: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Cc: Daniel Mack \u003cdaniel@zonque.org\u003e\n Cc: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Cc: Oleh Kravchenko \u003coleg@kaa.org.ua\u003e\n Cc: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Cc: Simon Shields \u003csimon@lineageos.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Reviewed-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n\ncommit 853a78a7d6c7734f5aaa9c2a866cbd1f2a658892\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:04 2019 +0200\n\n dt-bindings: leds: Add LED_COLOR_ID definitions\n \n Add common LED color identifiers.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Cc: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Cc: Daniel Mack \u003cdaniel@zonque.org\u003e\n Cc: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Cc: Oleh Kravchenko \u003coleg@kaa.org.ua\u003e\n Cc: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Cc: Simon Shields \u003csimon@lineageos.org\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Reviewed-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n\ncommit b2b998c0f944993c9ef435569651e407d607af41\nAuthor: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\nDate: Sun Jun 9 20:19:03 2019 +0200\n\n leds: class: Improve LED and LED flash class registration API\n \n Replace of_led_classdev_register() with led_classdev_register_ext(), which\n accepts easily extendable struct led_init_data, instead of the fixed\n struct device_node argument. The latter can be now passed in an fwnode\n property of the struct led_init_data.\n \n The modification is driven by the need for passing additional arguments\n required for the forthcoming generic mechanism for composing LED names.\n Currently the LED name is conveyed in the \"name\" char pointer property of\n the struct led_classdev. This is redundant since LED class device name\n is accessible throughout the whole LED class device life time via\n associated struct device\u0027s kobj-\u003ename property.\n \n The change will not break any existing clients since the patch alters\n also existing led_classdev{_flash}_register() macro wrappers, that pass\n NULL in place of init_data, which leads to using legacy name\n initialization path basing on the struct led_classdev\u0027s \"name\" property.\n \n Three existing users of devm_of_led_classdev_registers() are modified\n to use devm_led_classdev_register(), which will not impact their\n operation since they in fact didn\u0027t need to pass struct device_node on\n registration from the beginning.\n \n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n Cc: Baolin Wang \u003cbaolin.wang@linaro.org\u003e\n Cc: Dan Murphy \u003cdmurphy@ti.com\u003e\n Cc: Daniel Mack \u003cdaniel@zonque.org\u003e\n Cc: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Cc: Oleh Kravchenko \u003coleg@kaa.org.ua\u003e\n Cc: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Cc: Simon Shields \u003csimon@lineageos.org\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n\ncommit 0058eb589881056b49a4ba15dfa3f1b8db53991c\nAuthor: Michal Kalderon \u003cmichal.kalderon@marvell.com\u003e\nDate: Tue Jul 9 17:17:33 2019 +0300\n\n qed*: Change dpi_addr to be denoted with __iomem\n \n Several casts were required around dpi_addr parameter in qed_rdma_if.h\n This is an address on the doorbell bar and should therefore be marked with\n __iomem.\n \n Link: https://lore.kernel.org/r/20190709141735.19193-5-michal.kalderon@marvell.com\n Reported-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Ariel Elior \u003cariel.elior@marvell.com\u003e\n Signed-off-by: Michal Kalderon \u003cmichal.kalderon@marvell.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 3dcfb397dad2ad55bf50de3c5d5a57090d35a18a\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Thu Jul 25 18:59:46 2019 +0200\n\n ASoC: codec2codec: deal with params when necessary\n \n When there is an event on codec to codec dai_link, we only need to deal\n with params if the event is SND_SOC_DAPM_PRE_PMU, when .hw_params() is\n called. For the other events, it is useless.\n \n Also, dealing with the codec to codec params just before calling\n .hw_params() callbacks give change to either party on the link to alter\n params content in .startup(), which might be useful in some cases\n \n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Link: https://lore.kernel.org/r/20190725165949.29699-4-jbrunet@baylibre.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 054d65004c6a008dfefbdae4fc1b46a3ad4e94c1\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Thu Jul 25 18:59:45 2019 +0200\n\n ASoC: codec2codec: name link using stream direction\n \n At the moment, codec to codec dai link widgets are named after the\n cpu dai and the 1st codec valid on the link. This might be confusing\n if there is multiple valid codecs on the link for one stream\n direction.\n \n Instead, use the dai link name and the stream direction to name the\n the dai link widget\n \n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Link: https://lore.kernel.org/r/20190725165949.29699-3-jbrunet@baylibre.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 68c907f10cd816cad2287167a1a1d77914a6d466\nAuthor: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\nDate: Thu Jul 25 18:59:44 2019 +0200\n\n ASoC: codec2codec: run callbacks in order\n \n When handling dai_link events on codec to codec links, run all .startup()\n callbacks on sinks and sources before running any .hw_params(). Same goes\n for hw_free() and shutdown(). This is closer to the behavior of regular\n dai links\n \n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Link: https://lore.kernel.org/r/20190725165949.29699-2-jbrunet@baylibre.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 1c38bdc96941fcf8ce9c32d1c45eca5547abd45e\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Thu Jul 25 12:13:11 2019 -0400\n\n MAINTAINERS: update EDAC entry to reflect current tree and maintainers\n \n Tony will start to officially maintain EDAC trees. Also, we\u0027ll\n be using a single tree for the EDAC development.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Acked-by: Borislav Petkov \u003cbp@suse.de\u003e\n Acked-by: Tony Luck \u003ctony.luck@intel.com\u003e\n\ncommit 8b38c538d460f46680d0fb8524efa15a8652b2d1\nAuthor: Max Gurtovoy \u003cmaxg@mellanox.com\u003e\nDate: Tue Jul 23 10:04:12 2019 +0300\n\n IB/mlx5: Add CREATE_PSV/DESTROY_PSV for devx interface\n \n Limit the number of PSV\u0027s created through devx to 1, to create a symmetry\n between create/destroy cmds. In the kernel, one can create up to 4 PSV\u0027s\n using CREATE_PSV cmd but the destruction is one by one. Add a protection\n for this a-symmetric definition for devx.\n \n Link: https://lore.kernel.org/r/20190723070412.6385-1-leon@kernel.org\n Signed-off-by: Max Gurtovoy \u003cmaxg@mellanox.com\u003e\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 1d2fedd8561dc469a7503855ee602f4bb3eccfa7\nAuthor: Parav Pandit \u003cparav@mellanox.com\u003e\nDate: Tue Jul 23 10:02:05 2019 +0300\n\n RDMA/core: Support netlink commands in non init_net net namespaces\n \n Now that IB core supports RDMA device binding with specific net namespace,\n enable IB core to accept netlink commands in non init_net namespaces.\n \n This is done by having per net namespace netlink socket.\n \n At present only netlink device handling client RDMA_NL_NLDEV supports\n device handling in multiple net namespaces. Hence do not accept netlink\n messages for other clients in non init_net net namespaces.\n \n Link: https://lore.kernel.org/r/20190723070205.6247-1-leon@kernel.org\n Signed-off-by: Parav Pandit \u003cparav@mellanox.com\u003e\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 913df8c35322dcd15d993cabc8cc92814dee3d50\nAuthor: Leon Romanovsky \u003cleonro@mellanox.com\u003e\nDate: Thu Jul 4 16:09:36 2019 +0300\n\n RDMA/mlx4: Annotate boolean arguments as bool and not int\n \n Information provided by qp_has_rq() and used latter is boolean, so update\n callers to proper type.\n \n Link: https://lore.kernel.org/r/20190704130936.8705-3-leon@kernel.org\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 089b645d19b2de7c9b541828d445b8c9a5dba792\nAuthor: Leon Romanovsky \u003cleonro@mellanox.com\u003e\nDate: Thu Jul 4 16:09:35 2019 +0300\n\n RDMA/mlx4: Separate creation of RWQ and QP\n \n The mlx4 WQ is implemented with HW QP without special HW object. Current\n implementation which tried to reuse the code did it with common QP\n creation flows. Such decision caused to the absence of mlx4_ib_wq struct,\n which is needed to ensure proper allocation of ib_wq inside of IB/core.\n \n Separate create_qp_common() to pure QP flow and to create_rq() for RWQ.\n \n Link: https://lore.kernel.org/r/20190704130936.8705-2-leon@kernel.org\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 4f96061b92da395ecd5325a87368670e1b1a7cc9\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 19:49:28 2019 +0800\n\n IB/usnic: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata, use dev_get_drvdata to make\n the code simpler.\n \n Link: https://lore.kernel.org/r/20190723114928.18424-1-hslester96@gmail.com\n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 39289bfc221423b27570e7c9157b690828e786cb\nAuthor: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\nDate: Mon Jul 22 17:01:30 2019 +0000\n\n RDMA: Make most headers compile stand alone\n \n So that rdma can work with CONFIG_KERNEL_HEADER_TEST and\n CONFIG_HEADERS_CHECK.\n \n Link: https://lore.kernel.org/r/20190722170126.GA16453@ziepe.ca\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit cf167e5eb92c143ac2b0300ea65ed9f0ce572c3b\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Tue Jul 16 23:07:12 2019 +0530\n\n RDMA/qedr: Remove Unneeded variable rc\n \n Fix the below warning reported by coccicheck:\n \n drivers/infiniband/hw/qedr/verbs.c:2454:5-7: Unneeded variable: \"rc\".\n Return \"0\" on line 2499\n \n Link: https://lore.kernel.org/r/20190716173712.GA12949@hari-Inspiron-1545\n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Acked-by: Michal Kalderon \u003cmichal.kalderon@marvell.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 4cc315c53f959c5d71b96111a7d317f40771558e\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Tue Jul 16 22:59:25 2019 +0530\n\n RDMA/qib: Unneeded variable ret\n \n Fix the below warning reported by coccicheck:\n \n drivers/infiniband/hw/qib/qib_file_ops.c:1792:5-8: Unneeded variable:\n \"ret\". Return \"0\" on line 1876\n \n Link: https://lore.kernel.org/r/20190716172924.GA12241@hari-Inspiron-1545\n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 1889c6e6ac6e15b3631d7e320f740219aa46504a\nAuthor: kbuild test robot \u003clkp@intel.com\u003e\nDate: Wed Jul 24 17:22:36 2019 +0800\n\n regulator: act8865: fix ptr_ret.cocci warnings\n \n drivers/regulator/act8865-regulator.c:447:8-14: WARNING: PTR_ERR_OR_ZERO can be used\n \n Use PTR_ERR_OR_ZERO rather than if(IS_ERR(...)) + PTR_ERR\n \n Generated by: scripts/coccinelle/api/ptr_ret.cocci\n \n Fixes: 2d09a79bf637 (\"regulator: act8865: Add support for act8600 charger\")\n CC: Maarten ter Huurne \u003cmaarten@treewalker.org\u003e\n Signed-off-by: kbuild test robot \u003clkp@intel.com\u003e\n Link: https://lore.kernel.org/r/20190724092236.witxtfmubun25l2t@1905cc33b6dd\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 6535e831b4c867467d838f2c7866b4dce1d74679\nAuthor: Stuart Henderson \u003cstuarth@opensource.wolfsonmicro.com\u003e\nDate: Thu Jul 25 17:39:31 2019 +0100\n\n ASoC: cs47l92: Add codec driver for Cirrus Logic CS47L92\n \n Adds the codec driver for the CS47L92 SmartCodec. This is a\n multi-functional codec based on the Cirrus Logic Madera platform.\n \n Signed-off-by: Stuart Henderson \u003cstuarth@opensource.wolfsonmicro.com\u003e\n Signed-off-by: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n Link: https://lore.kernel.org/r/20190725163931.24964-3-ckeepax@opensource.cirrus.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 9cba2d6a144741ad99bfa379337bb984a8ed94bd\nAuthor: Richard Fitzgerald \u003crf@opensource.wolfsonmicro.com\u003e\nDate: Thu Jul 25 17:39:30 2019 +0100\n\n ASoC: cs47l15: Add codec driver for Cirrus Logic CS47L15\n \n Adds the codec driver for the CS47L15 SmartCodec. This is a\n multi-functional codec based on the Cirrus Logic Madera platform.\n \n Signed-off-by: Richard Fitzgerald \u003crf@opensource.wolfsonmicro.com\u003e\n Signed-off-by: Jaswinder Jassal \u003cjjassal@opensource.wolfsonmicro.com\u003e\n Signed-off-by: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n Link: https://lore.kernel.org/r/20190725163931.24964-2-ckeepax@opensource.cirrus.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 01ec57a4371e573bfcfa898307af098a8c6f9dcf\nAuthor: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\nDate: Thu Jul 25 17:39:29 2019 +0100\n\n ASoC: wm_adsp: Allow bus error handler to be called directly\n \n There is no need for end drivers to add helper functions to allow the\n bus error handler to be called, simply update the prototype so it can be\n called directly.\n \n Signed-off-by: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n Link: https://lore.kernel.org/r/20190725163931.24964-1-ckeepax@opensource.cirrus.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 5af9e972ac72b1c9f13cd35ccfb4687741c53f3e\nAuthor: Jaroslav Kysela \u003cperex@perex.cz\u003e\nDate: Thu Jul 25 15:37:43 2019 +0200\n\n ASoC: SOF: Makefile - fix the top-level kernel module names (add snd- prefix)\n \n Use the proper module name. The objs assignments are already there.\n \n Signed-off-by: Jaroslav Kysela \u003cperex@perex.cz\u003e\n Cc: Liam Girdwood \u003cliam.r.girdwood@linux.intel.com\u003e\n Cc: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Cc: Takashi Iwai \u003ctiwai@suse.de\u003e\n Cc: Mark Brown \u003cbroonie@kernel.org\u003e0\n Link: https://lore.kernel.org/r/20190725133743.22145-1-perex@perex.cz\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit a8f910ec66583bfb61558c3f333195b3960d832d\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Mon Jul 15 17:06:42 2019 -0400\n\n media: v4l2-core: Cleanup Makefile\n \n Use the videodev-$(CONFIG_FOO) syntax to simplify the Makefile.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 620428820d0d76de5b29314457a3b27c0634b4f2\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Thu Jul 25 12:25:52 2019 -0400\n\n media: staging: hantro: avoid future namespace collisions\n \n Rename:\n vp8_dec_mc_filter -\u003e hantro_vp8_dec_mc_filter\n \n As other drivers may end implementing something with the same\n name.\n \n So, prepend driver\u0027s name here, in order to make symbol namespace\n cleaner.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 73fb5bfa53e972f4fcf53c1c230fcc07a8cf6801\nAuthor: Jeffy Chen \u003cjeffy.chen@rock-chips.com\u003e\nDate: Thu Jul 25 10:17:56 2019 -0400\n\n media: hantro: Support RK3399 VP8 decoding\n \n Rockchip RK3399 SoC has the same Hantro G1 IP block\n as RK3288, but the registers are entirely different.\n \n In a similar fashion as MPEG-2 decoding, it\u0027s simpler\n to just add a separate implementation.\n \n Signed-off-by: Jeffy Chen \u003cjeffy.chen@rock-chips.com\u003e\n Signed-off-by: Tomasz Figa \u003ctfiga@chromium.org\u003e\n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 613326d62f7975e4f7899ffbf745f7cb52882461\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Thu Jul 25 10:17:55 2019 -0400\n\n media: hantro: Move VP8 common code\n \n In order to introduce support for RK3399 VP8 decoding,\n move some common VP8 code. This will be reused by\n the RK3399 implementation, reducing code duplication.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 6389a62ff798e781567645c0b0ca3dd7b8a4289d\nAuthor: Liu, Changcheng \u003cchangcheng.liu@intel.com\u003e\nDate: Thu Jul 25 21:22:09 2019 +0800\n\n usbip: tools: fix GCC8 warning for strncpy\n \n GCC8 started emitting warning about using strncpy with number of bytes\n exactly equal destination size which could lead to non-zero terminated\n string being copied. Use \"SYSFS_PATH_MAX - 1\" \u0026 \"SYSFS_BUS_ID_SIZE - 1\"\n as number of bytes to ensure name is always zero-terminated.\n \n Signed-off-by: Changcheng Liu \u003cchangcheng.liu@aliyun.com\u003e\n Acked-by: Shuah Khan \u003cskhan@linuxfoundation.org\u003e\n Link: https://lore.kernel.org/r/20190725132209.GA27590@jerryopenix\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 932a9317ac492d4b93227357db4e3e2af28f0fbb\nAuthor: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\nDate: Thu Jul 25 10:17:54 2019 -0400\n\n media: hantro: Add helpers to prepare/finish a run\n \n And use them where appropriate.\n \n We might want to move hantro_{prepare,finish}_run() calls to\n device_run() and have a 2-step approach similar to cedrus (prepare +\n trigger) at some point, but let\u0027s keep that for later.\n \n Signed-off-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit deff5c37c88921e61a336f1c4eba1f7ad438cebe\nAuthor: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\nDate: Thu Jul 25 10:17:53 2019 -0400\n\n media: hantro: Add hantro_get_{src, dst}_buf() helpers\n \n And replace all calls to v4l2_m2m_next_{src,dst}_buf() by\n hantro_get_{src,dst}_buf() one.\n \n Signed-off-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 61e95d81aa28f411ff1238f56702dc2add2fac00\nAuthor: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\nDate: Thu Jul 25 10:17:52 2019 -0400\n\n media: hantro: Constify the control array\n \n controls[] is not supposed to be modified at runtime, let\u0027s make it\n explicit by adding a const specifier.\n \n Signed-off-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 9530bbe726a14fea2acd972a13b5d526edc94e87\nAuthor: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\nDate: Thu Jul 25 10:17:51 2019 -0400\n\n media: hantro: Simplify the controls creation logic\n \n v4l2_ctrl_new_custom() should work for any kind of control, including\n standard ones. With that change, we automatically get support for\n menu controls.\n \n Signed-off-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit c3c3509b86810293df5c524ef61421d8affc8bf0\nAuthor: Francois Buergisser \u003cfbuergisser@chromium.org\u003e\nDate: Thu Jul 25 10:17:50 2019 -0400\n\n media: hantro: Set DMA max segment size\n \n The Hantro codec is typically used in platforms with an IOMMU,\n so we need to set a proper DMA segment size. Devices without an\n IOMMU will still fallback to default 64KiB segments.\n \n Cc: stable@vger.kernel.org\n Fixes: 775fec69008d3 (\"media: add Rockchip VPU JPEG encoder driver\")\n Signed-off-by: Francois Buergisser \u003cfbuergisser@chromium.org\u003e\n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit aa0e9e72733b069a655a3fe9167f69de103830cc\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Wed Jul 24 23:06:02 2019 -0400\n\n media: staging: media: hantro: Remove call to memset after dma_alloc_coherent\n \n fix below issue reported by coccicheck\n /drivers/staging/media/hantro/hantro_vp8.c:149:16-34: WARNING:\n dma_alloc_coherent use in aux_buf -\u003e cpu already zeroes out memory, so\n memset is not needed\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 5e291a908869dd4fe70832a89678c6b0c1b8c481\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Thu Jul 25 10:09:24 2019 +0800\n\n ALSA: echoaudio: Replace kmalloc + memcpy with kmemdup\n \n Instead of using kmalloc + memcpy, use kmemdup\n to simplify the code.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 24f6f93a19e86779da94d97f4adccf3ab2662a77\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Mon Jul 15 17:23:15 2019 -0400\n\n media: Don\u0027t default-enable \"ancillary driver autoselect\" if EMBEDDED\n \n Admittedly, it\u0027s not easy to say what CONFIG_EMBEDDED really should be\n doing, just as it\u0027s not easy to say what is \"embedded\".\n \n In any case, the very description of MEDIA_SUBDRV_AUTOSELECT\n specifies that embedded systems usually don\u0027t want\n to enable this autoselect option.\n \n Therefore, drop the default-yes, when the CONFIG_EMBEDDED is chosen.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 6b4633310ae8a10f8ffacdc3eb6272b013c2adef\nAuthor: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\nDate: Thu Jul 4 13:16:02 2019 -0400\n\n media: rcar-vin: Always setup controls when opening video device\n \n Now that both Gen2 (video device centric) and Gen3 (media device\n centric) modes of the driver have controls it is required to always\n setup the controls when opening the devices. Remove the check which only\n calls v4l2_ctrl_handler_setup() for Gen2 and call it unconditionally.\n \n Signed-off-by: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 1d99e68ce7fd9b5f5ab01d72455e8acd699e49cc\nAuthor: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\nDate: Thu Jul 4 13:16:01 2019 -0400\n\n media: rcar-vin: Add support for RGB formats with alpha component\n \n The R-Car VIN module supports V4L2_PIX_FMT_ARGB555 and\n V4L2_PIX_FMT_ABGR32 pixel formats. Add the hardware register setup and\n allow the alpha component to be changed while streaming using the\n V4L2_CID_ALPHA_COMPONENT control.\n \n Signed-off-by: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n [hverkuil-cisco@xs4all.nl: fix checkpatch warning for macro VNDMR_A8BIT]\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 5720c7331ff7e376a6d612eb8c2920d1789d78f7\nAuthor: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\nDate: Thu Jul 4 13:16:00 2019 -0400\n\n media: rcar-vin: Add control for alpha component\n \n In preparation to adding support for RGB pixel formats with an alpha\n component add a control to allow the user to control which alpha value\n should be used.\n \n Signed-off-by: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 19ab1f649ca43bd16042389fa06524c3c58d7bfc\nAuthor: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\nDate: Thu Jul 4 13:15:59 2019 -0400\n\n media: rcar-vin: Rename VNDMR_DTMD_ARGB1555 to VNDMR_DTMD_ARGB\n \n The value have nothing to do with ARGB1555, it controls if the alpha\n component should be filled in for ARGB1555 or ARGB888. Rename it to\n reflect this.\n \n Signed-off-by: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\n Reviewed-by: Ulrich Hecht \u003culi+renesas@fpond.eu\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 049e684f2de9daa31a32423e16d2d3bd6bbd286f\nAuthor: Hans Verkuil \u003chverkuil@xs4all.nl\u003e\nDate: Thu Jul 25 08:48:21 2019 -0400\n\n media: v4l2-dev: fix WARN_ON(!vdev-\u003edevice_caps)\n \n Device nodes of type VFL_TYPE_SUBDEV have a 0 device_caps, so they\n trigger this warning. Add a check against VFL_TYPE_SUBDEV.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Fixes: 3c1350501c21 (\"media: v4l2-dev/ioctl: require non-zero device_caps, verify sane querycap results\")\n Reported-by: Jan Kotas \u003cjank@cadence.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 0005cbda438fa846650ae52ce693eeaff6d16b92\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 23 22:09:44 2019 +0200\n\n drm/via: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header.\n While touching the files divide include files in blocks\n and sort the files alphabetically.\n \n v2:\n - Replace all uses of DRM_WAIT_ON() with VIA_WAIT_ON()\n and thus avoiding to pull in drm_os_linux.h\n \n v3:\n - DRM_WAIT_ON replacement moved to earlier patch (Emil)\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Kevin Brace \u003ckevinbrace@gmx.com\u003e\n Cc: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Cc: \"Gustavo A. R. Silva\" \u003cgustavo@embeddedor.com\u003e\n Cc: Mike Marshall \u003chubcap@omnibond.com\u003e\n Cc: Ira Weiny \u003cira.weiny@intel.com\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: Michel Dänzer \u003cmichel@daenzer.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190723200944.17285-5-sam@ravnborg.org\n\ncommit 2299b52d139138e2ab91d3e835d7d18fdbde803d\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 23 22:09:43 2019 +0200\n\n drm/via: make via_drv.h self-contained\n \n Added include of header files to make via_drv.h self-contained.\n \n v3:\n - Reworded changelog a little - to reflect that more than one\n header files are added\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Kevin Brace \u003ckevinbrace@gmx.com\u003e\n Cc: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Cc: \"Gustavo A. R. Silva\" \u003cgustavo@embeddedor.com\u003e\n Cc: Mike Marshall \u003chubcap@omnibond.com\u003e\n Cc: Ira Weiny \u003cira.weiny@intel.com\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: Michel Dänzer \u003cmichel@daenzer.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190723200944.17285-4-sam@ravnborg.org\n\ncommit 9154e60c4e057296983ef08d96ab5db76c1574c2\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 23 22:09:42 2019 +0200\n\n drm/via: copy DRM_WAIT_ON as VIA_WAIT_ON and use it\n \n VIA_WAIT_ON() is a direct copy of DRM_WAIT_ON() from\n drm_os_linux.h.\n The copy is made so we can avoid the dependency on the legacy header.\n A more involved approach had been to introduce wait_event_* but for this\n legacy driver the simpler and more safe approach with a copy of the\n macro was selected.\n Added the relevant header files for the functions used in VIA_WAIT_ON.\n \n v3:\n - Updated users of DRM_WAIT_ON \u003d\u003e VIA_WAIT_ON (Emil)\n - Updated $subject (Emil)\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Kevin Brace \u003ckevinbrace@gmx.com\u003e\n Cc: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Cc: \"Gustavo A. R. Silva\" \u003cgustavo@embeddedor.com\u003e\n Cc: Mike Marshall \u003chubcap@omnibond.com\u003e\n Cc: Ira Weiny \u003cira.weiny@intel.com\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: Michel Dänzer \u003cmichel@daenzer.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190723200944.17285-3-sam@ravnborg.org\n\ncommit 3bf2a06e3612575284202c63de4f584006a85146\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 23 22:09:41 2019 +0200\n\n drm/via: drop use of DRM(READ|WRITE) macros\n \n The DRM_READ, DRM_WRITE macros comes from the deprecated drm_os_linux.h\n header file. Remove their use to remove this dependency.\n \n Replace the use of the macros with static inline variants.\n \n v4:\n - Use a more standard via_write8_mask() function (Emil)\n \n v3:\n - Use static inline functions, rather than macros (Emil)\n - Use dedicated mask variants for byte access (Emil)\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Kevin Brace \u003ckevinbrace@gmx.com\u003e\n Cc: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Cc: \"Gustavo A. R. Silva\" \u003cgustavo@embeddedor.com\u003e\n Cc: Mike Marshall \u003chubcap@omnibond.com\u003e\n Cc: Ira Weiny \u003cira.weiny@intel.com\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: Michel Dänzer \u003cmichel@daenzer.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190723200944.17285-2-sam@ravnborg.org\n\ncommit 33db6f94847c33ebdeb2e73ef5cd8fd04fb58e86\nAuthor: Yixian Liu \u003cliuyixian@huawei.com\u003e\nDate: Mon Jul 8 21:41:25 2019 +0800\n\n RDMA/hns: Refactor eq table init for hip08\n \n To make the code more readable, move the part of naming irq and request\n irq out of eq table init into a separate function.\n \n Link: https://lore.kernel.org/r/1562593285-8037-10-git-send-email-oulijun@huawei.com\n Signed-off-by: Yixian Liu \u003cliuyixian@huawei.com\u003e\n Signed-off-by: Lijun Ou \u003coulijun@huawei.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit d7019c0f47ae497bae27be5397287ca72c704bf4\nAuthor: Lijun Ou \u003coulijun@huawei.com\u003e\nDate: Mon Jul 8 21:41:24 2019 +0800\n\n RDMA/hns: Refactor hem table mhop check and calculation\n \n The calculation of mhop for hem is duplicated in hns_roce_init_hem_table\n and hns_roce_calc_hem_mhop, extracting it from them to a separate\n function. Moreover, this patch refactors hns_roce_check_whether_mhop to\n reduce complexity.\n \n Link: https://lore.kernel.org/r/1562593285-8037-9-git-send-email-oulijun@huawei.com\n Signed-off-by: Yixian Liu \u003cliuyixian@huawei.com\u003e\n Signed-off-by: Lijun Ou \u003coulijun@huawei.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 3ee0e170d72c05172b0aa4e46d58795fdc19081e\nAuthor: Lijun Ou \u003coulijun@huawei.com\u003e\nDate: Mon Jul 8 21:41:23 2019 +0800\n\n RDMA/hns: Package for hns_roce_rereg_user_mr function\n \n Move some code of the hns_roce_rereg_user_mr() function into an\n independent function in oder to improve readability.\n \n Link: https://lore.kernel.org/r/1562593285-8037-8-git-send-email-oulijun@huawei.com\n Signed-off-by: Lang Cheng \u003cchenglang@huawei.com\u003e\n Signed-off-by: Lijun Ou \u003coulijun@huawei.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 89b4b70b974c98139f57eeb06facab40473268d3\nAuthor: chenglang \u003cchenglang@huawei.com\u003e\nDate: Mon Jul 8 21:41:22 2019 +0800\n\n RDMA/hns: Optimize hns_roce_mhop_alloc function.\n \n Move some lines for allocating multi-hop addressing into independent\n functions in order to improve readability.\n \n Link: https://lore.kernel.org/r/1562593285-8037-7-git-send-email-oulijun@huawei.com\n Signed-off-by: Lang Cheng \u003cchenglang@huawei.com\u003e\n Signed-off-by: Lijun Ou \u003coulijun@huawei.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 99441ab552f1e1713a562838215c080387a94267\nAuthor: Xi Wang \u003cwangxi11@huawei.com\u003e\nDate: Mon Jul 8 21:41:21 2019 +0800\n\n RDMA/hns: optimize the duplicated code for qpc setting flow\n \n Currently, more than 20 lines of duplicate code exist in function\n \u0027modify_qp_init_to_init\u0027 and function \u0027modify_qp_reset_to_init\u0027, which\n affects the readability of the code. Consolidate them.\n \n Link: https://lore.kernel.org/r/1562593285-8037-6-git-send-email-oulijun@huawei.com\n Signed-off-by: Xi Wang \u003cwangxi11@huawei.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 947441eadb909b031fc0793d049e1af0d5855a21\nAuthor: Lijun Ou \u003coulijun@huawei.com\u003e\nDate: Mon Jul 8 21:41:20 2019 +0800\n\n RDMA/hns: Use a separated function for setting extend sge paramters\n \n Moves the related lines of setting extended sge size into a separate\n function as well as remove the unused variables.\n \n Link: https://lore.kernel.org/r/1562593285-8037-5-git-send-email-oulijun@huawei.com\n Signed-off-by: Lijun Ou \u003coulijun@huawei.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 606bf89e98efb1174891c4d96afd39c1a974ce3d\nAuthor: Lijun Ou \u003coulijun@huawei.com\u003e\nDate: Mon Jul 8 21:41:19 2019 +0800\n\n RDMA/hns: Refactor for hns_roce_v2_modify_qp function\n \n Move some lines which exist hns_roce_v2_modify_qp function into a new\n function. The code refactored mainly includes some absolute fields of qp\n context and some optional fields of qp context.\n \n Link: https://lore.kernel.org/r/1562593285-8037-4-git-send-email-oulijun@huawei.com\n Signed-off-by: Lijun Ou \u003coulijun@huawei.com\u003e\n Signed-off-by: Xi Wang \u003cwangxi11@huawei.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 2a2f1887e08964cf155bd67fc094eeb5b02c2f5c\nAuthor: Lijun Ou \u003coulijun@huawei.com\u003e\nDate: Mon Jul 8 21:41:18 2019 +0800\n\n RDMA/hns: Refactor the code of creating srq\n \n Move the related codes of creating user srq and kernel srq into two\n independent functions as well as remove some unused code and\n simplifications.\n \n Link: https://lore.kernel.org/r/1562593285-8037-3-git-send-email-oulijun@huawei.com\n Signed-off-by: Lijun Ou \u003coulijun@huawei.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 4f8f0d5e33ddb0d102cf6e61c8e8f497eb1c3866\nAuthor: Lijun Ou \u003coulijun@huawei.com\u003e\nDate: Mon Jul 8 21:41:17 2019 +0800\n\n RDMA/hns: Package the flow of creating cq\n \n Moves the related code of creating cq into separate functions in order to\n improve comprehensibility.\n \n Link: https://lore.kernel.org/r/1562593285-8037-2-git-send-email-oulijun@huawei.com\n Signed-off-by: Lijun Ou \u003coulijun@huawei.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit a5c9c299d1e1f3c6fb421d73367cc2f793af1a2d\nAuthor: Parav Pandit \u003cparav@mellanox.com\u003e\nDate: Tue Jul 23 09:57:31 2019 +0300\n\n IB/mlx5: Avoid unnecessary typecast\n \n IB device pointer is already available while deallocating IB device,\n Hence do not typecast it.\n \n Link: https://lore.kernel.org/r/20190723065733.4899-9-leon@kernel.org\n Signed-off-by: Parav Pandit \u003cparav@mellanox.com\u003e\n Reviewed-by: Daniel Jurgens \u003cdanielj@mellanox.com\u003e\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 56594ae1d250f839945cda89d0138eece46dd607\nAuthor: Parav Pandit \u003cparav@mellanox.com\u003e\nDate: Tue Jul 23 09:57:24 2019 +0300\n\n RDMA/core: Annotate destroy of mutex to ensure that it is released as unlocked\n \n While compiled with CONFIG_DEBUG_MUTEXES, the kernel ensures that mutex is\n not held during destroy. Hence add mutex_destroy() for mutexes used in\n RDMA modules.\n \n Link: https://lore.kernel.org/r/20190723065733.4899-2-leon@kernel.org\n Signed-off-by: Parav Pandit \u003cparav@mellanox.com\u003e\n Signed-off-by: Leon Romanovsky \u003cleonro@mellanox.com\u003e\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 2c93346698bacd5e1a7fb59b7a5ab84cae1cc5da\nAuthor: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\nDate: Mon Mar 4 06:29:43 2019 -0500\n\n media: ipu3-cio2: Parse information from firmware without using callbacks\n \n Instead of using the convenience function\n v4l2_async_notifier_parse_fwnode_endpoints(), parse the endpoints and set\n up the async sub-devices without using callbacks. While this adds a little\n bit of code, it makes parsing the endpoints quite a bit more simple and\n gives more control to the driver over the process. The parsing assumes\n D-PHY instead of letting the V4L2 fwnode framework guess it.\n \n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Reviewed-by: Jacopo Mondi \u003cjacopo+renesas@jmondi.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 706c0cffaf2a1b5844fc99e62c09a21f7a7acfdf\nAuthor: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\nDate: Mon Mar 4 04:45:57 2019 -0500\n\n media: ipu3-cio2: Proceed with notifier init even if there are no subdevs\n \n The notifier may be registered even if there are no subdevs. Do that to\n simplify the code.\n \n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit f6a5242b7cf20bf34e5362896df310a35ba3b90b\nAuthor: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\nDate: Mon Mar 4 04:36:40 2019 -0500\n\n media: ipu3-cio2: Clean up notifier\u0027s subdev list if parsing endpoints fails\n \n The notifier must be cleaned up whenever parsing endpoints fails. Do that\n to avoid a memory leak in that case.\n \n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit d1e164289e55de90fcc5680f459a43972430a28d\nAuthor: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\nDate: Mon Mar 4 04:29:49 2019 -0500\n\n media: v4l2-async: Safely clean up an uninitialised notifier\n \n Make the V4L2 async framework a bit more robust by allowing to clean up an\n uninitialised notifier. Otherwise the result would be a (close to) NULL\n pointer dereference.\n \n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Tested-by: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 2ea4cfc9a7163b48514361cb02d46b0546dfcc86\nAuthor: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\nDate: Fri Mar 1 08:08:02 2019 -0500\n\n media: omap3isp: Rework OF endpoint parsing\n \n Rework OF endpoint parsing for the omap3isp driver. This does add some\n lines of code. The benefits are still clear:\n \n - the great complication related to callbacks in endpoint parsing is gone;\n instead endpoints are obtained port by port and\n \n - endpoints may now have a default bus configuration which was not\n possible while using callbacks. This driver does not benefit from that\n feature, but as the omap3isp is one of the exemplary drivers, this works\n as an example for driver developers.\n \n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Tested-by: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit e77cc85ee3902303c71b3923f772680f4cb3de99\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Wed Jul 24 10:19:05 2019 +0530\n\n arm64: dts: qcom: sdm845: remove macro from unit name\n \n Unit name is supposed to be a number, using a macro with hex value is\n not recommended, so add the value in unit name.\n \n arch/arm64/boot/dts/qcom/pm8998.dtsi:81.18-84.6: Warning (unit_address_format): /soc/spmi@c440000/pmic@0/adc@3100/adc-chan@0x06: unit name should not have leading \"0x\"\n arch/arm64/boot/dts/qcom/pm8998.dtsi:81.18-84.6: Warning (unit_address_format): /soc/spmi@c440000/pmic@0/adc@3100/adc-chan@0x06: unit name should not have leading 0s\n \n Reviewed-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Reviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 2833d79c28b8c7687fd5d2974f84bbadf2dbe1a1\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Wed Jul 24 10:19:06 2019 +0530\n\n arm64: dts: qcom: sdm845-cheza: remove macro from unit name\n \n Unit address is supposed to be a number, using a macro with hex value is\n not recommended, so add the value in unit name.\n \n arch/arm64/boot/dts/qcom/sdm845-cheza.dtsi:966.16-969.4: Warning (unit_address_format): /soc@0/spmi@c440000/pmic@0/adc@3100/adc-chan@0x4d: unit name should not have leading \"0x\"\n arch/arm64/boot/dts/qcom/sdm845-cheza.dtsi:971.16-974.4: Warning (unit_address_format): /soc@0/spmi@c440000/pmic@0/adc@3100/adc-chan@0x4e: unit name should not have leading \"0x\"\n arch/arm64/boot/dts/qcom/sdm845-cheza.dtsi:976.16-979.4: Warning (unit_address_format): /soc@0/spmi@c440000/pmic@0/adc@3100/adc-chan@0x4f: unit name should not have leading \"0x\"\n arch/arm64/boot/dts/qcom/sdm845-cheza.dtsi:981.16-984.4: Warning (unit_address_format): /soc@0/spmi@c440000/pmic@0/adc@3100/adc-chan@0x50: unit name should not have leading \"0x\"\n arch/arm64/boot/dts/qcom/sdm845-cheza.dtsi:986.16-989.4: Warning (unit_address_format): /soc@0/spmi@c440000/pmic@0/adc@3100/adc-chan@0x51: unit name should not have leading \"0x\"\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Reviewed-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 19e684e835f6321b7c1d633f22cb4c4889f50747\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Wed Jul 24 10:19:04 2019 +0530\n\n arm64: dts: qcom: sdm845: remove unit name for thermal trip points\n \n The thermal trip points have unit name but no reg property, so we can\n remove them\n \n arch/arm64/boot/dts/qcom/sdm845.dtsi:2824.31-2828.7: Warning (unit_address_vs_reg): /thermal-zones/cpu0-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:2830.31-2834.7: Warning (unit_address_vs_reg): /thermal-zones/cpu0-thermal/trips/trip-point@1: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:2868.31-2872.7: Warning (unit_address_vs_reg): /thermal-zones/cpu1-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:2874.31-2878.7: Warning (unit_address_vs_reg): /thermal-zones/cpu1-thermal/trips/trip-point@1: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:2912.31-2916.7: Warning (unit_address_vs_reg): /thermal-zones/cpu2-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:2918.31-2922.7: Warning (unit_address_vs_reg): /thermal-zones/cpu2-thermal/trips/trip-point@1: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:2956.31-2960.7: Warning (unit_address_vs_reg): /thermal-zones/cpu3-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:2962.31-2966.7: Warning (unit_address_vs_reg): /thermal-zones/cpu3-thermal/trips/trip-point@1: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3000.31-3004.7: Warning (unit_address_vs_reg): /thermal-zones/cpu4-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3006.31-3010.7: Warning (unit_address_vs_reg): /thermal-zones/cpu4-thermal/trips/trip-point@1: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3044.31-3048.7: Warning (unit_address_vs_reg): /thermal-zones/cpu5-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3050.31-3054.7: Warning (unit_address_vs_reg): /thermal-zones/cpu5-thermal/trips/trip-point@1: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3088.31-3092.7: Warning (unit_address_vs_reg): /thermal-zones/cpu6-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3094.31-3098.7: Warning (unit_address_vs_reg): /thermal-zones/cpu6-thermal/trips/trip-point@1: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3132.31-3136.7: Warning (unit_address_vs_reg): /thermal-zones/cpu7-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3138.31-3142.7: Warning (unit_address_vs_reg): /thermal-zones/cpu7-thermal/trips/trip-point@1: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3176.32-3180.7: Warning (unit_address_vs_reg): /thermal-zones/aoss0-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3191.35-3195.7: Warning (unit_address_vs_reg): /thermal-zones/cluster0-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3211.35-3215.7: Warning (unit_address_vs_reg): /thermal-zones/cluster1-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3231.31-3235.7: Warning (unit_address_vs_reg): /thermal-zones/gpu-thermal-top/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3246.31-3250.7: Warning (unit_address_vs_reg): /thermal-zones/gpu-thermal-bottom/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3261.32-3265.7: Warning (unit_address_vs_reg): /thermal-zones/aoss1-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3276.35-3280.7: Warning (unit_address_vs_reg): /thermal-zones/q6-modem-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3291.30-3295.7: Warning (unit_address_vs_reg): /thermal-zones/mem-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3306.31-3310.7: Warning (unit_address_vs_reg): /thermal-zones/wlan-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3321.33-3325.7: Warning (unit_address_vs_reg): /thermal-zones/q6-hvx-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3336.33-3340.7: Warning (unit_address_vs_reg): /thermal-zones/camera-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3351.32-3355.7: Warning (unit_address_vs_reg): /thermal-zones/video-thermal/trips/trip-point@0: node has a unit name, but no reg property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:3366.32-3370.7: Warning (unit_address_vs_reg): /thermal-zones/modem-thermal/trips/trip-point@0: node has a unit name, but no reg property\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Reviewed-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 81a7b51177fc864a5c6bab885deae7b8f4103f21\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Wed Jul 24 10:19:03 2019 +0530\n\n arm64: dts: qcom: sdm845: remove unnecessary properties for dsi nodes\n \n We get a warning about unnecessary properties of\n \n arch/arm64/boot/dts/qcom/sdm845.dtsi:2211.22-2257.6: Warning (avoid_unnecessary_addr_size): /soc/mdss@ae00000/dsi@ae94000: unnecessary #address-cells/#size-cells without \"ranges\" or child \"reg\" property\n arch/arm64/boot/dts/qcom/sdm845.dtsi:2278.22-2324.6: Warning (avoid_unnecessary_addr_size): /soc/mdss@ae00000/dsi@ae96000: unnecessary #address-cells/#size-cells without \"ranges\" or child \"reg\" property\n \n So, remove these properties\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Reviewed-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit a1875bf98290e538de2484c4db957ace5a8ac578\nAuthor: Vinod Koul \u003cvkoul@kernel.org\u003e\nDate: Wed Jul 24 10:19:02 2019 +0530\n\n arm64: dts: qcom: sdm845: Add unit name to soc node\n \n We get a warning about missing unit name for soc node, so add it.\n \n arch/arm64/boot/dts/qcom/sdm845.dtsi:623.11-2814.4: Warning (unit_address_vs_reg): /soc: node has a reg or ranges property, but no unit name\n \n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n Reviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Reviewed-by: Amit Kucheria \u003camit.kucheria@linaro.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 36a80df44b6f06f33b0ba10ff01bc87e352257fa\nAuthor: Malathi Gottam \u003cmgottam@codeaurora.org\u003e\nDate: Tue Jul 2 17:42:29 2019 +0530\n\n arm64: dts: sdm845: Add video nodes\n \n This adds video nodes to sdm845 based on the examples\n in the bindings.\n \n Tested-by: An\\355bal Lim\\363n \u003canibal.limon@linaro.org\u003e\n Reviewed-by: Rajendra Nayak \u003crnayak@codeaurora.org\u003e\n Signed-off-by: Malathi Gottam \u003cmgottam@codeaurora.org\u003e\n Co-developed-by: Aniket Masule \u003camasule@codeaurora.org\u003e\n Signed-off-by: Aniket Masule \u003camasule@codeaurora.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 43931d350f30c6cd8c2f498d54ef7d65750abc92\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:30 2019 +0200\n\n x86/apic/x2apic: Implement IPI shorthands support\n \n All callers of apic-\u003esend_IPI_all() and apic-\u003esend_IPI_allbutself() contain\n the decision logic for shorthand invocation already and invoke\n send_IPI_mask() if the prereqisites are not satisfied.\n \n Implement shorthand support for x2apic.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105221.134696837@linutronix.de\n\ncommit 2510d09e9dabc265341f164e0b45b2dfdcb7ef36\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:29 2019 +0200\n\n x86/apic/flat64: Remove the IPI shorthand decision logic\n \n All callers of apic-\u003esend_IPI_all() and apic-\u003esend_IPI_allbutself() contain\n the decision logic for shorthand invocation already and invoke\n send_IPI_mask() if the prereqisites are not satisfied.\n \n Remove the now redundant decision logic in the APIC code and the duplicate\n helper in probe_64.c.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105221.042964120@linutronix.de\n\ncommit dea978632e8400b84888bad20df0cd91c18f0aec\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:28 2019 +0200\n\n x86/apic: Share common IPI helpers\n \n The 64bit implementations need the same wrappers around\n __default_send_IPI_shortcut() as 32bit.\n \n Move them out of the 32bit section.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105220.951534451@linutronix.de\n\ncommit 1f0ad660488b8eb2450d1834af6a156104281194\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:27 2019 +0200\n\n x86/apic: Remove the shorthand decision logic\n \n All callers of apic-\u003esend_IPI_all() and apic-\u003esend_IPI_allbutself() contain\n the decision logic for shorthand invocation already and invoke\n send_IPI_mask() if the prereqisites are not satisfied.\n \n Remove the now redundant decision logic in the 32bit implementation.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105220.860244707@linutronix.de\n\ncommit 832df3d47badcbc860aef617105b6bc1c9459304\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:26 2019 +0200\n\n x86/smp: Enhance native_send_call_func_ipi()\n \n Nadav noticed that the cpumask allocations in native_send_call_func_ipi()\n are noticeable in microbenchmarks.\n \n Use the new cpumask_or_equal() function to simplify the decision whether\n the supplied target CPU mask is either equal to cpu_online_mask or equal to\n cpu_online_mask except for the CPU on which the function is invoked.\n \n cpumask_or_equal() or\u0027s the target mask and the cpumask of the current CPU\n together and compares it to cpu_online_mask.\n \n If the result is false, use the mask based IPI function, otherwise check\n whether the current CPU is set in the target mask and invoke either the\n send_IPI_all() or the send_IPI_allbutselt() APIC callback.\n \n Make the shorthand decision also depend on the static key which enables\n shorthand mode. That allows to remove the extra cpumask comparison with\n cpu_callout_mask.\n \n Reported-by: Nadav Amit \u003cnamit@vmware.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105220.768238809@linutronix.de\n\ncommit d0a7166bc7ac4feac5c482ebe8b2417aa3302ef4\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:25 2019 +0200\n\n x86/smp: Move smp_function_call implementations into IPI code\n \n Move it where it belongs. That allows to keep all the shorthand logic in\n one place.\n \n No functional change.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105220.677835995@linutronix.de\n\ncommit 22ca7ee933a39f542ff6f81fc64f8036eff56519\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:23 2019 +0200\n\n x86/apic: Provide and use helper for send_IPI_allbutself()\n \n To support IPI shorthands wrap invocations of apic-\u003esend_IPI_allbutself()\n in a helper function, so the static key controlling the shorthand mode is\n only in one place.\n \n Fixup all callers.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105220.492691679@linutronix.de\n\ncommit 6a1cb5f5c6413222b8532722562dd1edb5fdfd38\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:22 2019 +0200\n\n x86/apic: Add static key to Control IPI shorthands\n \n The IPI shorthand functionality delivers IPI/NMI broadcasts to all CPUs in\n the system. This can have similar side effects as the MCE broadcasting when\n CPUs are waiting in the BIOS or are offlined.\n \n The kernel tracks already the state of offlined CPUs whether they have been\n brought up at least once so that the CR4 MCE bit is set to make sure that\n MCE broadcasts can\u0027t brick the machine.\n \n Utilize that information and compare it to the cpu_present_mask. If all\n present CPUs have been brought up at least once then the broadcast side\n effect is mitigated by disabling regular interrupt/IPI delivery in the APIC\n itself and by the cpu offline check at the begin of the NMI handler.\n \n Use a static key to switch between broadcasting via shorthands or sending\n the IPI/NMI one by one.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105220.386410643@linutronix.de\n\ncommit bdda3b93e66085abf0b2c16bcdf471176e3c816a\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:21 2019 +0200\n\n x86/apic: Move no_ipi_broadcast() out of 32bit\n \n For the upcoming shorthand support for all APIC incarnations the command\n line option needs to be available for 64 bit as well.\n \n While at it, rename the control variable, make it static and mark it\n __ro_after_init.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105220.278327940@linutronix.de\n\ncommit bd82dba2fa6ae91061e5d31399d61fe65028f714\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:20 2019 +0200\n\n x86/apic: Add NMI_VECTOR wait to IPI shorthand\n \n To support NMI shorthand broadcasts add the safe wait for ICR idle for NMI\n vector delivery.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105220.185838026@linutronix.de\n\ncommit 3994ff90acc3b115734fe532720c37a499c502ce\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:19 2019 +0200\n\n x86/apic: Remove dest argument from __default_send_IPI_shortcut()\n \n The SDM states:\n \n \"The destination shorthand field of the ICR allows the delivery mode to be\n by-passed in favor of broadcasting the IPI to all the processors on the\n system bus and/or back to itself (see Section 10.6.1, Interrupt Command\n Register (ICR)). Three destination shorthands are supported: self, all\n excluding self, and all including self. The destination mode is ignored\n when a destination shorthand is used.\"\n \n So there is no point to supply the destination mode to the shorthand\n delivery function.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105220.094613426@linutronix.de\n\ncommit 60dcaad5736faff5a6b1abba5a292499f57197fe\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Wed Jul 24 17:25:52 2019 +0200\n\n x86/hotplug: Silence APIC and NMI when CPU is dead\n \n In order to support IPI/NMI broadcasting via the shorthand mechanism side\n effects of shorthands need to be mitigated:\n \n Shorthand IPIs and NMIs hit all CPUs including unplugged CPUs\n \n Neither of those can be handled on unplugged CPUs for obvious reasons.\n \n It would be trivial to just fully disable the APIC via the enable bit in\n MSR_APICBASE. But that\u0027s not possible because clearing that bit on systems\n based on the 3 wire APIC bus would require a hardware reset to bring it\n back as the APIC would lose track of bus arbitration. On systems with FSB\n delivery APICBASE could be disabled, but it has to be guaranteed that no\n interrupt is sent to the APIC while in that state and it\u0027s not clear from\n the SDM whether it still responds to INIT/SIPI messages.\n \n Therefore stay on the safe side and switch the APIC into soft disabled mode\n so it won\u0027t deliver any regular vector to the CPU.\n \n NMIs are still propagated to the \u0027dead\u0027 CPUs. To mitigate that add a check\n for the CPU being offline on early nmi entry and if so bail.\n \n Note, this cannot use the stop/restart_nmi() magic which is used in the\n alternatives code. A dead CPU cannot invoke nmi_enter() or anything else\n due to RCU and other reasons.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/alpine.DEB.2.21.1907241723290.1791@nanos.tec.linutronix.de\n\ncommit 9c92374b631d233abf5bd355cb4253d3d83d5578\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:17 2019 +0200\n\n x86/cpu: Move arch_smt_update() to a neutral place\n \n arch_smt_update() will be used to control IPI/NMI broadcasting via the\n shorthand mechanism. Keeping it in the bugs file and calling the apic\n function from there is possible, but not really intuitive.\n \n Move it to a neutral place and invoke the bugs function from there.\n \n No functional change.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105219.910317273@linutronix.de\n\ncommit 82e574782345aa634e1544e80da85d71a9dbde19\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:15 2019 +0200\n\n x86/apic/uv: Make x2apic_extra_bits static\n \n Not used outside of the UV apic source.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105219.725264153@linutronix.de\n\ncommit c94f0718fb1c171d6dfdd69cb6001fa0d8206710\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:14 2019 +0200\n\n x86/apic: Consolidate the apic local headers\n \n Now there are three small local headers. Some contain functions which are\n only used in one source file.\n \n Move all the inlines and declarations into a single local header and the\n inlines which are only used in one source file into that.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105219.618612624@linutronix.de\n\ncommit ba77b2a02e0099ab0021bc3169b8f674c6be19f0\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:13 2019 +0200\n\n x86/apic: Move apic_flat_64 header into apic directory\n \n Only used locally.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105219.526508168@linutronix.de\n\ncommit 8b542da372875373db9688477671151df3418acb\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:12 2019 +0200\n\n x86/apic: Move ipi header into apic directory\n \n Only used locally.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105219.434738036@linutronix.de\n\ncommit 521b82fee98c1e334ba3a2459ba3739d459e9e4e\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:11 2019 +0200\n\n x86/apic: Cleanup the include maze\n \n All of these APIC files include the world and some more. Remove the\n unneeded cruft.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105219.342631201@linutronix.de\n\ncommit cdc86c9d1f825d13cef85d9ebd3e73572602fb48\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:10 2019 +0200\n\n x86/apic: Move IPI inlines into ipi.c\n \n No point in having them in an header file.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105219.252225936@linutronix.de\n\ncommit cc8bf191378c1da8ad2b99cf470ee70193ace84e\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:09 2019 +0200\n\n x86/apic: Make apic_pending_intr_clear() more robust\n \n In course of developing shorthand based IPI support issues with the\n function which tries to clear eventually pending ISR bits in the local APIC\n were observed.\n \n 1) O-day testing triggered the WARN_ON() in apic_pending_intr_clear().\n \n This warning is emitted when the function fails to clear pending ISR\n bits or observes pending IRR bits which are not delivered to the CPU\n after the stale ISR bit(s) are ACK\u0027ed.\n \n Unfortunately the function only emits a WARN_ON() and fails to dump\n the IRR/ISR content. That\u0027s useless for debugging.\n \n Feng added spot on debug printk\u0027s which revealed that the stale IRR\n bit belonged to the APIC timer interrupt vector, but adding ad hoc\n debug code does not help with sporadic failures in the field.\n \n Rework the loop so the full IRR/ISR contents are saved and on failure\n dumped.\n \n 2) The loop termination logic is interesting at best.\n \n If the machine has no TSC or cpu_khz is not known yet it tries 1\n million times to ack stale IRR/ISR bits. What?\n \n With TSC it uses the TSC to calculate the loop termination. It takes a\n timestamp at entry and terminates the loop when:\n \n (rdtsc() - start_timestamp) \u003e\u003d (cpu_hkz \u003c\u003c 10)\n \n That\u0027s roughly one second.\n \n Both methods are problematic. The APIC has 256 vectors, which means\n that in theory max. 256 IRR/ISR bits can be set. In practice this is\n impossible and the chance that more than a few bits are set is close\n to zero.\n \n With the pure loop based approach the 1 million retries are complete\n overkill.\n \n With TSC this can terminate too early in a guest which is running on a\n heavily loaded host even with only a couple of IRR/ISR bits set. The\n reason is that after acknowledging the highest priority ISR bit,\n pending IRRs must get serviced first before the next round of\n acknowledge can take place as the APIC (real and virtualized) does not\n honour EOI without a preceeding interrupt on the CPU. And every APIC\n read/write takes a VMEXIT if the APIC is virtualized. While trying to\n reproduce the issue 0-day reported it was observed that the guest was\n scheduled out long enough under heavy load that it terminated after 8\n iterations.\n \n Make the loop terminate after 512 iterations. That\u0027s plenty enough\n in any case and does not take endless time to complete.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105219.158847694@linutronix.de\n\ncommit 2640da4cccf5cc613bf26f0998b9e340f4b5f69c\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:08 2019 +0200\n\n x86/apic: Soft disable APIC before initializing it\n \n If the APIC was already enabled on entry of setup_local_APIC() then\n disabling it soft via the SPIV register makes a lot of sense.\n \n That masks all LVT entries and brings it into a well defined state.\n \n Otherwise previously enabled LVTs which are not touched in the setup\n function stay unmasked and might surprise the just booting kernel.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105219.068290579@linutronix.de\n\ncommit 39c89dff9c366ad98d2e5598db41ff9b1bdb9e88\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:07 2019 +0200\n\n x86/apic: Invoke perf_events_lapic_init() after enabling APIC\n \n If the APIC is soft disabled then unmasking an LVT entry does not work and\n the write is ignored. perf_events_lapic_init() tries to do so.\n \n Move the invocation after the point where the APIC has been enabled.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105218.962517234@linutronix.de\n\ncommit 2591bc4e8d70b4e1330d327fb7e3921f4e070a51\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:06 2019 +0200\n\n x86/kgbd: Use NMI_VECTOR not APIC_DM_NMI\n \n apic-\u003esend_IPI_allbutself() takes a vector number as argument.\n \n APIC_DM_NMI is clearly not a vector number. It\u0027s defined to 0x400 which is\n outside the vector space.\n \n Use NMI_VECTOR instead as that\u0027s what it is intended to be.\n \n Fixes: 82da3ff89dc2 (\"x86: kgdb support\")\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105218.855189979@linutronix.de\n\ncommit 747d5a1bf293dcb33af755a6d285d41b8c1ea010\nAuthor: Grzegorz Halat \u003cghalat@redhat.com\u003e\nDate: Fri Jun 28 14:28:13 2019 +0200\n\n x86/reboot: Always use NMI fallback when shutdown via reboot vector IPI fails\n \n A reboot request sends an IPI via the reboot vector and waits for all other\n CPUs to stop. If one or more CPUs are in critical regions with interrupts\n disabled then the IPI is not handled on those CPUs and the shutdown hangs\n if native_stop_other_cpus() is called with the wait argument set.\n \n Such a situation can happen when one CPU was stopped within a lock held\n section and another CPU is trying to acquire that lock with interrupts\n disabled. There are other scenarios which can cause such a lockup as well.\n \n In theory the shutdown should be attempted by an NMI IPI after the timeout\n period elapsed. Though the wait loop after sending the reboot vector IPI\n prevents this. It checks the wait request argument and the timeout. If wait\n is set, which is true for sys_reboot() then it won\u0027t fall through to the\n NMI shutdown method after the timeout period has finished.\n \n This was an oversight when the NMI shutdown mechanism was added to handle\n the \u0027reboot IPI is not working\u0027 situation. The mechanism was added to deal\n with stuck panic shutdowns, which do not have the wait request set, so the\n \u0027wait request\u0027 case was probably not considered.\n \n Remove the wait check from the post reboot vector IPI wait loop and enforce\n that the wait loop in the NMI fallback path is invoked even if NMI IPIs are\n disabled or the registration of the NMI handler fails. That second wait\n loop will then hang if not all CPUs shutdown and the wait argument is set.\n \n [ tglx: Avoid the hard to parse line break in the NMI fallback path,\n add comments and massage the changelog ]\n \n Fixes: 7d007d21e539 (\"x86/reboot: Use NMI to assist in shutting down if IRQ fails\")\n Signed-off-by: Grzegorz Halat \u003cghalat@redhat.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Don Zickus \u003cdzickus@redhat.com\u003e\n Link: https://lkml.kernel.org/r/20190628122813.15500-1-ghalat@redhat.com\n\ncommit a1dc0446d64966dc0ae756aebdc449f335742c13\nAuthor: Qian Cai \u003ccai@lca.pw\u003e\nDate: Fri Jul 19 21:23:19 2019 -0400\n\n sched/core: Silence a warning in sched_init()\n \n Compiling a kernel with both FAIR_GROUP_SCHED\u003dn and RT_GROUP_SCHED\u003dn\n will generate a compiler warning:\n \n kernel/sched/core.c: In function \u0027sched_init\u0027:\n kernel/sched/core.c:5906:32: warning: variable \u0027ptr\u0027 set but not used\n \n It is unnecessary to have both \"alloc_size\" and \"ptr\", so just combine\n them.\n \n Signed-off-by: Qian Cai \u003ccai@lca.pw\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: valentin.schneider@arm.com\n Link: https://lkml.kernel.org/r/20190720012319.884-1-cai@lca.pw\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit a07db5c0865799ebed1f88be0df50c581fb65029\nAuthor: Juri Lelli \u003cjuri.lelli@redhat.com\u003e\nDate: Fri Jul 19 08:34:55 2019 +0200\n\n sched/core: Fix CPU controller for !RT_GROUP_SCHED\n \n On !CONFIG_RT_GROUP_SCHED configurations it is currently not possible to\n move RT tasks between cgroups to which CPU controller has been attached;\n but it is oddly possible to first move tasks around and then make them\n RT (setschedule to FIFO/RR).\n \n E.g.:\n \n # mkdir /sys/fs/cgroup/cpu,cpuacct/group1\n # chrt -fp 10 $$\n # echo $$ \u003e /sys/fs/cgroup/cpu,cpuacct/group1/tasks\n bash: echo: write error: Invalid argument\n # chrt -op 0 $$\n # echo $$ \u003e /sys/fs/cgroup/cpu,cpuacct/group1/tasks\n # chrt -fp 10 $$\n # cat /sys/fs/cgroup/cpu,cpuacct/group1/tasks\n 2345\n 2598\n # chrt -p 2345\n pid 2345\u0027s current scheduling policy: SCHED_FIFO\n pid 2345\u0027s current scheduling priority: 10\n \n Also, as Michal noted, it is currently not possible to enable CPU\n controller on unified hierarchy with !CONFIG_RT_GROUP_SCHED (if there\n are any kernel RT threads in root cgroup, they can\u0027t be migrated to the\n newly created CPU controller\u0027s root in cgroup_update_dfl_csses()).\n \n Existing code comes with a comment saying the \"we don\u0027t support RT-tasks\n being in separate groups\". Such comment is however stale and belongs to\n pre-RT_GROUP_SCHED times. Also, it doesn\u0027t make much sense for\n !RT_GROUP_ SCHED configurations, since checks related to RT bandwidth\n are not performed at all in these cases.\n \n Make moving RT tasks between CPU controller groups viable by removing\n special case check for RT (and DEADLINE) tasks.\n \n Signed-off-by: Juri Lelli \u003cjuri.lelli@redhat.com\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Reviewed-by: Michal Koutný \u003cmkoutny@suse.com\u003e\n Reviewed-by: Daniel Bristot de Oliveira \u003cbristot@redhat.com\u003e\n Acked-by: Tejun Heo \u003ctj@kernel.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: lizefan@huawei.com\n Cc: longman@redhat.com\n Cc: luca.abeni@santannapisa.it\n Cc: rostedt@goodmis.org\n Link: https://lkml.kernel.org/r/20190719063455.27328-1-juri.lelli@redhat.com\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 710da3c8ea7dfbd327920afd3831d8c82c42789d\nAuthor: Juri Lelli \u003cjuri.lelli@redhat.com\u003e\nDate: Fri Jul 19 16:00:00 2019 +0200\n\n sched/core: Prevent race condition between cpuset and __sched_setscheduler()\n \n No synchronisation mechanism exists between the cpuset subsystem and\n calls to function __sched_setscheduler(). As such, it is possible that\n new root domains are created on the cpuset side while a deadline\n acceptance test is carried out in __sched_setscheduler(), leading to a\n potential oversell of CPU bandwidth.\n \n Grab cpuset_rwsem read lock from core scheduler, so to prevent\n situations such as the one described above from happening.\n \n The only exception is normalize_rt_tasks() which needs to work under\n tasklist_lock and can\u0027t therefore grab cpuset_rwsem. We are fine with\n this, as this function is only called by sysrq and, if that gets\n triggered, DEADLINE guarantees are already gone out of the window\n anyway.\n \n Tested-by: Dietmar Eggemann \u003cdietmar.eggemann@arm.com\u003e\n Signed-off-by: Juri Lelli \u003cjuri.lelli@redhat.com\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: bristot@redhat.com\n Cc: claudio@evidence.eu.com\n Cc: lizefan@huawei.com\n Cc: longman@redhat.com\n Cc: luca.abeni@santannapisa.it\n Cc: mathieu.poirier@linaro.org\n Cc: rostedt@goodmis.org\n Cc: tj@kernel.org\n Cc: tommaso.cucinotta@santannapisa.it\n Link: https://lkml.kernel.org/r/20190719140000.31694-9-juri.lelli@redhat.com\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 1a763fd7c6335e3122c1cc09576ef6c99ada4267\nAuthor: Juri Lelli \u003cjuri.lelli@redhat.com\u003e\nDate: Fri Jul 19 15:59:59 2019 +0200\n\n rcu/tree: Call setschedule() gp ktread to SCHED_FIFO outside of atomic region\n \n sched_setscheduler() needs to acquire cpuset_rwsem, but it is currently\n called from an invalid (atomic) context by rcu_spawn_gp_kthread().\n \n Fix that by simply moving sched_setscheduler_nocheck() call outside of\n the atomic region, as it doesn\u0027t actually require to be guarded by\n rcu_node lock.\n \n Suggested-by: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Tested-by: Dietmar Eggemann \u003cdietmar.eggemann@arm.com\u003e\n Signed-off-by: Juri Lelli \u003cjuri.lelli@redhat.com\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: bristot@redhat.com\n Cc: claudio@evidence.eu.com\n Cc: lizefan@huawei.com\n Cc: longman@redhat.com\n Cc: luca.abeni@santannapisa.it\n Cc: mathieu.poirier@linaro.org\n Cc: rostedt@goodmis.org\n Cc: tj@kernel.org\n Cc: tommaso.cucinotta@santannapisa.it\n Link: https://lkml.kernel.org/r/20190719140000.31694-8-juri.lelli@redhat.com\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit d74b27d63a8bebe2fe634944e4ebdc7b10db7a39\nAuthor: Juri Lelli \u003cjuri.lelli@redhat.com\u003e\nDate: Fri Jul 19 15:59:58 2019 +0200\n\n cgroup/cpuset: Change cpuset_rwsem and hotplug lock order\n \n cpuset_rwsem is going to be acquired from sched_setscheduler() with a\n following patch. There are however paths (e.g., spawn_ksoftirqd) in\n which sched_scheduler() is eventually called while holding hotplug lock;\n this creates a dependecy between hotplug lock (to be always acquired\n first) and cpuset_rwsem (to be always acquired after hotplug lock).\n \n Fix paths which currently take the two locks in the wrong order (after\n a following patch is applied).\n \n Tested-by: Dietmar Eggemann \u003cdietmar.eggemann@arm.com\u003e\n Signed-off-by: Juri Lelli \u003cjuri.lelli@redhat.com\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: bristot@redhat.com\n Cc: claudio@evidence.eu.com\n Cc: lizefan@huawei.com\n Cc: longman@redhat.com\n Cc: luca.abeni@santannapisa.it\n Cc: mathieu.poirier@linaro.org\n Cc: rostedt@goodmis.org\n Cc: tj@kernel.org\n Cc: tommaso.cucinotta@santannapisa.it\n Link: https://lkml.kernel.org/r/20190719140000.31694-7-juri.lelli@redhat.com\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 1243dc518c9da467da6635313a2dbb41b8ffc275\nAuthor: Juri Lelli \u003cjuri.lelli@redhat.com\u003e\nDate: Fri Jul 19 15:59:57 2019 +0200\n\n cgroup/cpuset: Convert cpuset_mutex to percpu_rwsem\n \n Holding cpuset_mutex means that cpusets are stable (only the holder can\n make changes) and this is required for fixing a synchronization issue\n between cpusets and scheduler core. However, grabbing cpuset_mutex from\n setscheduler() hotpath (as implemented in a later patch) is a no-go, as\n it would create a bottleneck for tasks concurrently calling\n setscheduler().\n \n Convert cpuset_mutex to be a percpu_rwsem (cpuset_rwsem), so that\n setscheduler() will then be able to read lock it and avoid concurrency\n issues.\n \n Tested-by: Dietmar Eggemann \u003cdietmar.eggemann@arm.com\u003e\n Signed-off-by: Juri Lelli \u003cjuri.lelli@redhat.com\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: bristot@redhat.com\n Cc: claudio@evidence.eu.com\n Cc: lizefan@huawei.com\n Cc: longman@redhat.com\n Cc: luca.abeni@santannapisa.it\n Cc: mathieu.poirier@linaro.org\n Cc: rostedt@goodmis.org\n Cc: tj@kernel.org\n Cc: tommaso.cucinotta@santannapisa.it\n Link: https://lkml.kernel.org/r/20190719140000.31694-6-juri.lelli@redhat.com\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 59d06cea1198d665ba11f7e8c5f45b00ff2e4812\nAuthor: Juri Lelli \u003cjuri.lelli@redhat.com\u003e\nDate: Fri Jul 19 15:59:56 2019 +0200\n\n sched/deadline: Fix bandwidth accounting at all levels after offline migration\n \n If a task happens to be throttled while the CPU it was running on gets\n hotplugged off, the bandwidth associated with the task is not correctly\n migrated with it when the replenishment timer fires (offline_migration).\n \n Fix things up, for this_bw, running_bw and total_bw, when replenishment\n timer fires and task is migrated (dl_task_offline_migration()).\n \n Tested-by: Dietmar Eggemann \u003cdietmar.eggemann@arm.com\u003e\n Signed-off-by: Juri Lelli \u003cjuri.lelli@redhat.com\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: bristot@redhat.com\n Cc: claudio@evidence.eu.com\n Cc: lizefan@huawei.com\n Cc: longman@redhat.com\n Cc: luca.abeni@santannapisa.it\n Cc: mathieu.poirier@linaro.org\n Cc: rostedt@goodmis.org\n Cc: tj@kernel.org\n Cc: tommaso.cucinotta@santannapisa.it\n Link: https://lkml.kernel.org/r/20190719140000.31694-5-juri.lelli@redhat.com\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit f9a25f776d780bfa3279f0b6e5f5cf3224997976\nAuthor: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\nDate: Fri Jul 19 15:59:55 2019 +0200\n\n cpusets: Rebuild root domain deadline accounting information\n \n When the topology of root domains is modified by CPUset or CPUhotplug\n operations information about the current deadline bandwidth held in the\n root domain is lost.\n \n This patch addresses the issue by recalculating the lost deadline\n bandwidth information by circling through the deadline tasks held in\n CPUsets and adding their current load to the root domain they are\n associated with.\n \n Tested-by: Dietmar Eggemann \u003cdietmar.eggemann@arm.com\u003e\n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Signed-off-by: Juri Lelli \u003cjuri.lelli@redhat.com\u003e\n [ Various additional modifications. ]\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: bristot@redhat.com\n Cc: claudio@evidence.eu.com\n Cc: lizefan@huawei.com\n Cc: longman@redhat.com\n Cc: luca.abeni@santannapisa.it\n Cc: rostedt@goodmis.org\n Cc: tj@kernel.org\n Cc: tommaso.cucinotta@santannapisa.it\n Link: https://lkml.kernel.org/r/20190719140000.31694-4-juri.lelli@redhat.com\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 4b211f2b129dd1f6a6956bbc76e2f232c1ec3ad8\nAuthor: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\nDate: Fri Jul 19 15:59:54 2019 +0200\n\n sched/core: Streamle calls to task_rq_unlock()\n \n Calls to task_rq_unlock() are done several times in the\n __sched_setscheduler() function. This is fine when only the rq lock needs to be\n handled but not so much when other locks come into play.\n \n This patch streamlines the release of the rq lock so that only one\n location need to be modified when dealing with more than one lock.\n \n No change of functionality is introduced by this patch.\n \n Tested-by: Dietmar Eggemann \u003cdietmar.eggemann@arm.com\u003e\n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Reviewed-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n Acked-by: Tejun Heo \u003ctj@kernel.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: bristot@redhat.com\n Cc: claudio@evidence.eu.com\n Cc: lizefan@huawei.com\n Cc: longman@redhat.com\n Cc: luca.abeni@santannapisa.it\n Cc: tommaso.cucinotta@santannapisa.it\n Link: https://lkml.kernel.org/r/20190719140000.31694-3-juri.lelli@redhat.com\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit c22645f4c8f021fb1c5e7189eb1f968132cc0844\nAuthor: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\nDate: Fri Jul 19 15:59:53 2019 +0200\n\n sched/topology: Add partition_sched_domains_locked()\n \n Introduce the partition_sched_domains_locked() function by taking\n the mutex locking code out of the original function. That way\n the work done by partition_sched_domains_locked() can be reused\n without dropping the mutex lock.\n \n No change of functionality is introduced by this patch.\n \n Tested-by: Dietmar Eggemann \u003cdietmar.eggemann@arm.com\u003e\n Signed-off-by: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Acked-by: Tejun Heo \u003ctj@kernel.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: bristot@redhat.com\n Cc: claudio@evidence.eu.com\n Cc: lizefan@huawei.com\n Cc: longman@redhat.com\n Cc: luca.abeni@santannapisa.it\n Cc: rostedt@goodmis.org\n Cc: tommaso.cucinotta@santannapisa.it\n Link: https://lkml.kernel.org/r/20190719140000.31694-2-juri.lelli@redhat.com\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 60e17f5cef838e9ca7946ced208ceddcec6c315d\nAuthor: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\nDate: Tue Jun 4 12:31:52 2019 +0530\n\n sched/fair: Introduce fits_capacity()\n \n The same formula to check utilization against capacity (after\n considering capacity_margin) is already used at 5 different locations.\n \n This patch creates a new macro, fits_capacity(), which can be used from\n all these locations without exposing the details of it and hence\n simplify code.\n \n All the 5 code locations are updated as well to use it..\n \n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Vincent Guittot \u003cvincent.guittot@linaro.org\u003e\n Link: https://lkml.kernel.org/r/b477ac75a2b163048bdaeb37f57b4c3f04f75a31.1559631700.git.viresh.kumar@linaro.org\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit e0e8d4911ed2695b12c3a01c15634000ede9bc73\nAuthor: Wanpeng Li \u003cwanpengli@tencent.com\u003e\nDate: Fri Jun 28 16:51:41 2019 +0800\n\n sched/isolation: Prefer housekeeping CPU in local node\n \n In real product setup, there will be houseeking CPUs in each nodes, it\n is prefer to do housekeeping from local node, fallback to global online\n cpumask if failed to find houseeking CPU from local node.\n \n Signed-off-by: Wanpeng Li \u003cwanpengli@tencent.com\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Reviewed-by: Srikar Dronamraju \u003csrikar@linux.vnet.ibm.com\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/1561711901-4755-2-git-send-email-wanpengli@tencent.com\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 65d74e91694e1afac40c96fb64a9ef120757729e\nAuthor: Yi Wang \u003cwang.yi59@zte.com.cn\u003e\nDate: Fri Jul 5 12:35:07 2019 +0800\n\n sched/stats: Fix unlikely() use of sched_info_on()\n \n sched_info_on() is called with unlikely hint, however, the test\n is to be a constant(1) on which compiler will do nothing when\n make defconfig, so remove the hint.\n \n Also, fix a lack of {}.\n \n Signed-off-by: Yi Wang \u003cwang.yi59@zte.com.cn\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: up2wing@gmail.com\n Cc: wang.liang82@zte.com.cn\n Cc: xue.zhihong@zte.com.cn\n Link: https://lkml.kernel.org/r/1562301307-43002-1-git-send-email-wang.yi59@zte.com.cn\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 7b3c92b85a65c2db1f542265bc98e1f9e3056eba\nAuthor: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\nDate: Thu Jul 4 15:13:23 2019 -0700\n\n sched/core: Convert get_task_struct() to return the task\n \n Returning the pointer that was passed in allows us to write\n slightly more idiomatic code. Convert a few users.\n \n Signed-off-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190704221323.24290-1-willy@infradead.org\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 3c29e651e16dd3b3179cfb2d055ee9538e37515c\nAuthor: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\nDate: Wed Jun 26 10:36:30 2019 +0530\n\n sched/fair: Fall back to sched-idle CPU if idle CPU isn\u0027t found\n \n We try to find an idle CPU to run the next task, but in case we don\u0027t\n find an idle CPU it is better to pick a CPU which will run the task the\n soonest, for performance reason.\n \n A CPU which isn\u0027t idle but has only SCHED_IDLE activity queued on it\n should be a good target based on this criteria as any normal fair task\n will most likely preempt the currently running SCHED_IDLE task\n immediately. In fact, choosing a SCHED_IDLE CPU over a fully idle one\n shall give better results as it should be able to run the task sooner\n than an idle CPU (which requires to be woken up from an idle state).\n \n This patch updates both fast and slow paths with this optimization.\n \n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Vincent Guittot \u003cvincent.guittot@linaro.org\u003e\n Cc: chris.redpath@arm.com\n Cc: quentin.perret@linaro.org\n Cc: songliubraving@fb.com\n Cc: steven.sistare@oracle.com\n Cc: subhra.mazumdar@oracle.com\n Cc: tkjos@google.com\n Link: https://lkml.kernel.org/r/eeafa25fdeb6f6edd5b2da716bc8f0ba7708cbcf.1561523542.git.viresh.kumar@linaro.org\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 43e9f7f231e40e4534fc3a735da152911a085c16\nAuthor: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\nDate: Wed Jun 26 10:36:29 2019 +0530\n\n sched/fair: Start tracking SCHED_IDLE tasks count in cfs_rq\n \n Track how many tasks are present with SCHED_IDLE policy in each cfs_rq.\n This will be used by later commits.\n \n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Daniel Lezcano \u003cdaniel.lezcano@linaro.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Vincent Guittot \u003cvincent.guittot@linaro.org\u003e\n Cc: chris.redpath@arm.com\n Cc: quentin.perret@linaro.org\n Cc: songliubraving@fb.com\n Cc: steven.sistare@oracle.com\n Cc: subhra.mazumdar@oracle.com\n Cc: tkjos@google.com\n Link: https://lkml.kernel.org/r/0d3cdc427fc68808ad5bccc40e86ed0bf9da8bb4.1561523542.git.viresh.kumar@linaro.org\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 84ec3a0787086fcd25f284f59b3aa01fd6fc0a5d\nAuthor: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\nDate: Tue Jun 25 09:52:38 2019 -0700\n\n time/tick-broadcast: Fix tick_broadcast_offline() lockdep complaint\n \n time/tick-broadcast: Fix tick_broadcast_offline() lockdep complaint\n \n The TASKS03 and TREE04 rcutorture scenarios produce the following\n lockdep complaint:\n \n WARNING: inconsistent lock state\n 5.2.0-rc1+ #513 Not tainted\n --------------------------------\n inconsistent {IN-HARDIRQ-W} -\u003e {HARDIRQ-ON-W} usage.\n migration/1/14 [HC0[0]:SC0[0]:HE1:SE1] takes:\n (____ptrval____) (tick_broadcast_lock){?...}, at: tick_broadcast_offline+0xf/0x70\n {IN-HARDIRQ-W} state was registered at:\n lock_acquire+0xb0/0x1c0\n _raw_spin_lock_irqsave+0x3c/0x50\n tick_broadcast_switch_to_oneshot+0xd/0x40\n tick_switch_to_oneshot+0x4f/0xd0\n hrtimer_run_queues+0xf3/0x130\n run_local_timers+0x1c/0x50\n update_process_times+0x1c/0x50\n tick_periodic+0x26/0xc0\n tick_handle_periodic+0x1a/0x60\n smp_apic_timer_interrupt+0x80/0x2a0\n apic_timer_interrupt+0xf/0x20\n _raw_spin_unlock_irqrestore+0x4e/0x60\n rcu_nocb_gp_kthread+0x15d/0x590\n kthread+0xf3/0x130\n ret_from_fork+0x3a/0x50\n irq event stamp: 171\n hardirqs last enabled at (171): [\u003cffffffff8a201a37\u003e] trace_hardirqs_on_thunk+0x1a/0x1c\n hardirqs last disabled at (170): [\u003cffffffff8a201a53\u003e] trace_hardirqs_off_thunk+0x1a/0x1c\n softirqs last enabled at (0): [\u003cffffffff8a264ee0\u003e] copy_process.part.56+0x650/0x1cb0\n softirqs last disabled at (0): [\u003c0000000000000000\u003e] 0x0\n \n [...]\n \n To reproduce, run the following rcutorture test:\n \n $ tools/testing/selftests/rcutorture/bin/kvm.sh --duration 5 --kconfig \"CONFIG_DEBUG_LOCK_ALLOC\u003dy CONFIG_PROVE_LOCKING\u003dy\" --configs \"TASKS03 TREE04\"\n \n It turns out that tick_broadcast_offline() was an innocent bystander.\n After all, interrupts are supposed to be disabled throughout\n take_cpu_down(), and therefore should have been disabled upon entry to\n tick_offline_cpu() and thus to tick_broadcast_offline(). This suggests\n that one of the CPU-hotplug notifiers was incorrectly enabling interrupts,\n and leaving them enabled on return.\n \n Some debugging code showed that the culprit was sched_cpu_dying().\n It had irqs enabled after return from sched_tick_stop(). Which in turn\n had irqs enabled after return from cancel_delayed_work_sync(). Which is a\n wrapper around __cancel_work_timer(). Which can sleep in the case where\n something else is concurrently trying to cancel the same delayed work,\n and as Thomas Gleixner pointed out on IRC, sleeping is a decidedly bad\n idea when you are invoked from take_cpu_down(), regardless of the state\n you leave interrupts in upon return.\n \n Code inspection located no reason why the delayed work absolutely\n needed to be canceled from sched_tick_stop(): The work is not\n bound to the outgoing CPU by design, given that the whole point is\n to collect statistics without disturbing the outgoing CPU.\n \n This commit therefore simply drops the cancel_delayed_work_sync() from\n sched_tick_stop(). Instead, a new -\u003estate field is added to the tick_work\n structure so that the delayed-work handler function sched_tick_remote()\n can avoid reposting itself. A cpu_is_offline() check is also added to\n sched_tick_remote() to avoid mucking with the state of an offlined CPU\n (though it does appear safe to do so). The sched_tick_start() and\n sched_tick_stop() functions also update -\u003estate, and sched_tick_start()\n also schedules the delayed work if -\u003estate indicates that it is not\n already in flight.\n \n Signed-off-by: Paul E. McKenney \u003cpaulmck@linux.ibm.com\u003e\n [ paulmck: Apply Peter Zijlstra and Frederic Weisbecker atomics feedback. ]\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Reviewed-by: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190625165238.GJ26519@linux.ibm.com\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit f6cad8df6b30a5d2bbbd2e698f74b4cafb9fb82b\nAuthor: Vincent Guittot \u003cvincent.guittot@linaro.org\u003e\nDate: Mon Jul 1 17:47:02 2019 +0200\n\n sched/fair: Fix imbalance due to CPU affinity\n \n The load_balance() has a dedicated mecanism to detect when an imbalance\n is due to CPU affinity and must be handled at parent level. In this case,\n the imbalance field of the parent\u0027s sched_group is set.\n \n The description of sg_imbalanced() gives a typical example of two groups\n of 4 CPUs each and 4 tasks each with a cpumask covering 1 CPU of the first\n group and 3 CPUs of the second group. Something like:\n \n { 0 1 2 3 } { 4 5 6 7 }\n * * * *\n \n But the load_balance fails to fix this UC on my octo cores system\n made of 2 clusters of quad cores.\n \n Whereas the load_balance is able to detect that the imbalanced is due to\n CPU affinity, it fails to fix it because the imbalance field is cleared\n before letting parent level a chance to run. In fact, when the imbalance is\n detected, the load_balance reruns without the CPU with pinned tasks. But\n there is no other running tasks in the situation described above and\n everything looks balanced this time so the imbalance field is immediately\n cleared.\n \n The imbalance field should not be cleared if there is no other task to move\n when the imbalance is detected.\n \n Signed-off-by: Vincent Guittot \u003cvincent.guittot@linaro.org\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/1561996022-28829-1-git-send-email-vincent.guittot@linaro.org\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 9434f9f5d117302cc7ddf038e7879f6871dc7a81\nAuthor: Valentin Schneider \u003cvalentin.schneider@arm.com\u003e\nDate: Mon Jul 15 11:25:08 2019 +0100\n\n sched/fair: Change task_numa_work() storage to static\n \n There are no callers outside of fair.c.\n \n Signed-off-by: Valentin Schneider \u003cvalentin.schneider@arm.com\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: mgorman@suse.de\n Cc: riel@surriel.com\n Link: https://lkml.kernel.org/r/20190715102508.32434-4-valentin.schneider@arm.com\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit b34920d4ce6e6fc9424c20a4be98676eb543122f\nAuthor: Valentin Schneider \u003cvalentin.schneider@arm.com\u003e\nDate: Mon Jul 15 11:25:07 2019 +0100\n\n sched/fair: Move task_numa_work() init to init_numa_balancing()\n \n We only need to set the callback_head worker function once, do it\n during sched_fork().\n \n While at it, move the comment regarding double task_work addition to\n init_numa_balancing(), since the double add sentinel is first set there.\n \n Suggested-by: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Signed-off-by: Valentin Schneider \u003cvalentin.schneider@arm.com\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: mgorman@suse.de\n Cc: riel@surriel.com\n Link: https://lkml.kernel.org/r/20190715102508.32434-3-valentin.schneider@arm.com\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit d35927a144641700c8328d707d1c89d305b4ecb8\nAuthor: Valentin Schneider \u003cvalentin.schneider@arm.com\u003e\nDate: Mon Jul 15 11:25:06 2019 +0100\n\n sched/fair: Move init_numa_balancing() below task_numa_work()\n \n To reference task_numa_work() from within init_numa_balancing(), we\n need the former to be declared before the latter. Do just that.\n \n This is a pure code movement.\n \n Signed-off-by: Valentin Schneider \u003cvalentin.schneider@arm.com\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: mgorman@suse.de\n Cc: riel@surriel.com\n Link: https://lkml.kernel.org/r/20190715102508.32434-2-valentin.schneider@arm.com\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 0c09ab96fc820109d63097a2adcbbd20836b655f\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Tue Jul 9 16:23:40 2019 +0200\n\n cpu/hotplug: Cache number of online CPUs\n \n Re-evaluating the bitmap wheight of the online cpus bitmap in every\n invocation of num_online_cpus() over and over is a pretty useless\n exercise. Especially when num_online_cpus() is used in code paths\n like the IPI delivery of x86 or the membarrier code.\n \n Cache the number of online CPUs in the core and just return the cached\n variable. The accessor function provides only a snapshot when used without\n protection against concurrent CPU hotplug.\n \n The storage needs to use an atomic_t because the kexec and reboot code\n (ab)use set_cpu_online() in their \u0027shutdown\u0027 handlers without any form of\n serialization as pointed out by Mathieu. Regular CPU hotplug usage is\n properly serialized.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Mathieu Desnoyers \u003cmathieu.desnoyers@efficios.com\u003e\n Link: https://lkml.kernel.org/r/alpine.DEB.2.21.1907091622590.1634@nanos.tec.linutronix.de\n\ncommit b9fa6442f7043e2cdd247905d4f3b80f2e9605cb\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:24 2019 +0200\n\n cpumask: Implement cpumask_or_equal()\n \n The IPI code of x86 needs to evaluate whether the target cpumask is equal\n to the cpu_online_mask or equal except for the calling CPU.\n \n To replace the current implementation which requires the usage of a\n temporary cpumask, which might involve allocations, add a new function\n which compares a cpumask to the result of two other cpumasks which are\n or\u0027ed together before comparison.\n \n This allows to make the required decision in one go and the calling code\n then can check for the calling CPU being set in the target mask with\n cpumask_test_cpu().\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105220.585449120@linutronix.de\n\ncommit e797bda3fd29137f6c151dfa10ea6a61c17895ce\nAuthor: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nDate: Mon Jul 22 20:47:16 2019 +0200\n\n smp/hotplug: Track booted once CPUs in a cpumask\n \n The booted once information which is required to deal with the MCE\n broadcast issue on X86 correctly is stored in the per cpu hotplug state,\n which is perfectly fine for the intended purpose.\n \n X86 needs that information for supporting NMI broadcasting via shortcuts,\n but retrieving it from per cpu data is cumbersome.\n \n Move it to a cpumask so the information can be checked against the\n cpu_present_mask quickly.\n \n No functional change intended.\n \n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Acked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Link: https://lkml.kernel.org/r/20190722105219.818822855@linutronix.de\n\ncommit 8c779229d0f4fe83ead90bdcbbf08b02989aa200\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Mon Jul 22 11:24:43 2019 -0700\n\n locking/lockdep: Report more stack trace statistics\n \n Report the number of stack traces and the number of stack trace hash\n chains. These two numbers are useful because these allow to estimate\n the number of stack trace hash collisions.\n \n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Waiman Long \u003clongman@redhat.com\u003e\n Cc: Will Deacon \u003cwill.deacon@arm.com\u003e\n Link: https://lkml.kernel.org/r/20190722182443.216015-5-bvanassche@acm.org\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 12593b7467f9130b64a6d4b6a26ed4ec217b6784\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Mon Jul 22 11:24:42 2019 -0700\n\n locking/lockdep: Reduce space occupied by stack traces\n \n Although commit 669de8bda87b (\"kernel/workqueue: Use dynamic lockdep keys\n for workqueues\") unregisters dynamic lockdep keys when a workqueue is\n destroyed, a side effect of that commit is that all stack traces\n associated with the lockdep key are leaked when a workqueue is destroyed.\n Fix this by storing each unique stack trace once. Other changes in this\n patch are:\n \n - Use NULL instead of { .nr_entries \u003d 0 } to represent \u0027no trace\u0027.\n - Store a pointer to a stack trace in struct lock_class and struct\n lock_list instead of storing \u0027nr_entries\u0027 and \u0027offset\u0027.\n \n This patch avoids that the following program triggers the \"BUG:\n MAX_STACK_TRACE_ENTRIES too low!\" complaint:\n \n #include \u003cfcntl.h\u003e\n #include \u003cunistd.h\u003e\n \n int main()\n {\n for (;;) {\n int fd \u003d open(\"/dev/infiniband/rdma_cm\", O_RDWR);\n close(fd);\n }\n }\n \n Suggested-by: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Reported-by: Eric Biggers \u003cebiggers@kernel.org\u003e\n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Waiman Long \u003clongman@redhat.com\u003e\n Cc: Will Deacon \u003cwill.deacon@arm.com\u003e\n Cc: Yuyang Du \u003cduyuyang@gmail.com\u003e\n Link: https://lkml.kernel.org/r/20190722182443.216015-4-bvanassche@acm.org\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit a2970421640bd9b6a78f2685d7750a791abdfd4e\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Mon Jul 22 11:24:41 2019 -0700\n\n stacktrace: Constify \u0027entries\u0027 arguments\n \n Make it clear to humans and to the compiler that the stack trace\n (\u0027entries\u0027) arguments are not modified.\n \n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Waiman Long \u003clongman@redhat.com\u003e\n Cc: Will Deacon \u003cwill.deacon@arm.com\u003e\n Link: https://lkml.kernel.org/r/20190722182443.216015-3-bvanassche@acm.org\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit 364f6afc4f5537b79cf454eb35cae92920676075\nAuthor: Bart Van Assche \u003cbvanassche@acm.org\u003e\nDate: Mon Jul 22 11:24:40 2019 -0700\n\n locking/lockdep: Make it clear that what lock_class::key points at is not modified\n \n This patch does not change the behavior of the lockdep code.\n \n Signed-off-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Signed-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\n Cc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Waiman Long \u003clongman@redhat.com\u003e\n Cc: Will Deacon \u003cwill.deacon@arm.com\u003e\n Link: https://lkml.kernel.org/r/20190722182443.216015-2-bvanassche@acm.org\n Signed-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n\ncommit bbe70e4e421116db0af3acf84cbce3c3f6cac53e\nAuthor: Jia-Ju Bai \u003cbaijiaju1990@gmail.com\u003e\nDate: Wed Jul 24 10:22:42 2019 +0800\n\n fs: kernfs: Fix possible null-pointer dereferences in kernfs_path_from_node_locked()\n \n In kernfs_path_from_node_locked(), there is an if statement on line 147\n to check whether buf is NULL:\n if (buf)\n \n When buf is NULL, it is used on line 151:\n len +\u003d strlcpy(buf + len, parent_str, ...)\n and line 158:\n len +\u003d strlcpy(buf + len, \"/\", ...)\n and line 160:\n len +\u003d strlcpy(buf + len, kn-\u003ename, ...)\n \n Thus, possible null-pointer dereferences may occur.\n \n To fix these possible bugs, buf is checked before being used.\n If it is NULL, -EINVAL is returned.\n \n These bugs are found by a static analysis tool STCheck written by us.\n \n Signed-off-by: Jia-Ju Bai \u003cbaijiaju1990@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190724022242.27505-1-baijiaju1990@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 2fd60da46da76948b8cc561ae8cc902209686253\nAuthor: Peng Wang \u003crocking@whu.edu.cn\u003e\nDate: Mon Jul 8 23:16:11 2019 +0800\n\n kernfs: fix potential null pointer dereference\n \n Get root safely after kn is ensureed to be not null.\n \n Signed-off-by: Peng Wang \u003crocking@whu.edu.cn\u003e\n Acked-by: Tejun Heo \u003ctj@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190708151611.13242-1-rocking@whu.edu.cn\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f4c255f1a747497a21748619d909cadabcfa060e\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Sun Jul 14 17:55:04 2019 -0400\n\n dmaengine: dma-jz4780: Break descriptor chains on JZ4740\n \n The current driver works perfectly fine on every generation of the\n JZ47xx SoCs, except on the JZ4740.\n \n There, when hardware descriptors are chained together (with the LINK\n bit set), the next descriptor isn\u0027t automatically fetched as it should -\n instead, an interrupt is raised, even if the TIE bit (Transfer Interrupt\n Enable) bit is cleared. When it happens, the DMA transfer seems to be\n stopped (it doesn\u0027t chain), and it\u0027s uncertain how many bytes have\n actually been transferred.\n \n Until somebody smarter than me can figure out how to make chained\n descriptors work on the JZ4740, we now disable chained descriptors on\n that particular SoC.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Link: https://lore.kernel.org/r/20190714215504.10877-1-paul@crapouillou.net\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit b5e29aa880be84c271be8d0726cec4018bfbfd74\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Mon Jul 22 13:47:13 2019 +0200\n\n mfd: davinci_voicecodec: Remove pointless #include\n \n When building davinci as multiplatform, we get a build error\n in this file:\n \n drivers/mfd/davinci_voicecodec.c:22:10: fatal error: \u0027mach/hardware.h\u0027 file not found\n \n The header is only used to access the io_v2p() macro, but the\n result is already known because that comes from the resource\n a little bit earlier.\n \n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Acked-by: Sekhar Nori \u003cnsekhar@ti.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit 197df18f7038178ef9c5806db1eff8e494381fa6\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Jul 9 23:01:32 2019 +0530\n\n mfd: max77620: Add of_node_put() before return\n \n Each iteration of for_each_child_of_node puts the previous node, but in\n the case of a return from the middle of the loop, there is no put, thus\n causing a memory leak. Hence add an of_node_put before the return.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Lee Jones \u003clee.jones@linaro.org\u003e\n\ncommit b7989e271e9f7fd4edb4166cd2a79b980c4b6136\nAuthor: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\nDate: Thu Jul 25 11:34:32 2019 +0300\n\n ASoC: ti: davinci-mcasp: Improve serializer handling in multi AXR setups\n \n When multiple serializers are used we need to track the number of\n serializers used by the other stream direction to avoid killing data lines\n when the first stream used more serializers than the second would need.\n We are still protected against the case when the second stream uses more\n serializers which had affected the running stream as well.\n \n To take advantage of the improved serializer logic we need to modify the\n channel constraints rule as well to allow the use of multiple serializers\n for the second stream as additional ones will not affect the FS/BCLK on\n the bus.\n \n Signed-off-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Link: https://lore.kernel.org/r/20190725083432.7419-1-peter.ujfalusi@ti.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 816fe206138a7e3895c3a74fbd326a51df003f70\nAuthor: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\nDate: Thu Jul 25 11:34:11 2019 +0300\n\n ASoC: ti: davinci-mcasp: Add support for RIGHT_J format\n \n The formater unit\u0027s rotation needs to be programmed differently for right\n aligned bus format to have the data moved to the correct place.\n \n Take the opportunity and simplify the formater unit setup code.\n \n Signed-off-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Link: https://lore.kernel.org/r/20190725083411.7211-1-peter.ujfalusi@ti.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit f2c8b57014038ae064a0f918f939af510f0aed02\nAuthor: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\nDate: Thu Jul 25 11:33:21 2019 +0300\n\n ASoC: pcm3168a: Allow all channels in case of parallel DIN/DOUT setup\n \n If multi DIN/DOUT mode is selected (tdm_slots \u003d\u003d 2) then configure the\n channel constraint to allow all channels.\n \n Signed-off-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Link: https://lore.kernel.org/r/20190725083321.6776-1-peter.ujfalusi@ti.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit b6297d9e078a4127fb608ede4d0944855dde667d\nAuthor: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\nDate: Fri May 24 11:15:32 2019 +0200\n\n clk: meson: g12a: fix hifi typo in mali parent_names\n \n Replace hihi by hifi in the mali parent_names of the g12a SoC family.\n \n Fixes: 085a4ea93d54 (\"clk: meson: g12a: add peripheral clock controller\")\n Signed-off-by: Alexandre Mergnat \u003camergnat@baylibre.com\u003e\n Acked-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Martin Blumenstingl \u003cmartin.blumenstingl@googlemail.com\u003e\n Signed-off-by: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n\ncommit 41db5f8397eee75afff82655a4884b5786a1d302\nAuthor: Jorge Ramirez-Ortiz \u003cjorge.ramirez-ortiz@linaro.org\u003e\nDate: Fri Jul 5 10:13:03 2019 +0200\n\n misc: fastrpc: fix memory leak when out of memory\n \n Do the necessary house-keeping if the allocated memory wont be used\n \n Signed-off-by: Jorge Ramirez-Ortiz \u003cjorge.ramirez-ortiz@linaro.org\u003e\n Link: https://lore.kernel.org/r/20190705081303.14170-1-jorge.ramirez-ortiz@linaro.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit e59fc52529118aaad9aa666f25530dde6e2180e9\nAuthor: J. Bruce Fields \u003cbfields@redhat.com\u003e\nDate: Wed Jul 10 18:59:34 2019 -0400\n\n staging: rtl8192*: display ESSIDs using %pE\n \n Everywhere else in the kernel ESSIDs are printed using %pE, and I can\u0027t\n see why there should be an exception here.\n \n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n Link: https://lore.kernel.org/r/1562799574-13315-1-git-send-email-bfields@redhat.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 0854d589aa91c68f676f55e087be549eacaaa408\nAuthor: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nDate: Tue Jul 9 07:34:37 2019 -0300\n\n docs: ABI: remove extension from sysfs-class-mic.txt\n \n Files under ABI doesn\u0027t end with a .txt extension.\n \n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Link: https://lore.kernel.org/r/5113d1ae9d2dd3bc685bbb65655abd67e26f48c4.1562668471.git.mchehab+samsung@kernel.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit bc83f79bd2119230888fb8574639d5a51b38f903\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Jul 12 11:24:09 2019 +0200\n\n mic: avoid statically declaring a \u0027struct device\u0027.\n \n Generally, declaring a platform device as a static variable is\n a bad idea and can cause all kinds of problems, in particular\n with the DMA configuration and lifetime rules.\n \n A specific problem we hit here is from a bug in clang that warns\n about certain (otherwise valid) macros when used in static variables:\n \n drivers/misc/mic/card/mic_x100.c:285:27: warning: shift count \u003e\u003d width of type [-Wshift-count-overflow]\n static u64 mic_dma_mask \u003d DMA_BIT_MASK(64);\n ^~~~~~~~~~~~~~~~\n include/linux/dma-mapping.h:141:54: note: expanded from macro \u0027DMA_BIT_MASK\u0027\n #define DMA_BIT_MASK(n) (((n) \u003d\u003d 64) ? ~0ULL : ((1ULL\u003c\u003c(n))-1))\n ^ ~~~\n \n A slightly better way here is to create the platform device dynamically\n and set the dma mask in the probe function.\n This avoids the warning and some other problems, but is still not ideal\n because the device creation should really be separated from the driver,\n and the fact that the device has no parent means we have to force\n the dma mask rather than having it set up from the bus that the device\n is actually on.\n \n Fixes: dd8d8d44df64 (\"misc: mic: MIC card driver specific changes to enable SCIF\")\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Link: https://lore.kernel.org/r/20190712092426.872625-1-arnd@arndb.de\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit dcfecd4d7a551906595351a26e1db91774b8563d\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Thu Jul 4 07:56:45 2019 +0200\n\n fpga: dfl: use driver core functions, not sysfs ones.\n \n This is a driver, do not call \"raw\" sysfs functions, instead call driver\n core ones. Specifically convert the use of sysfs_create_files() and\n sysfs_remove_files() to use device_add_groups() and\n device_remove_groups()\n \n Cc: Wu Hao \u003chao.wu@intel.com\u003e\n Cc: Alan Tull \u003catull@kernel.org\u003e\n Cc: Moritz Fischer \u003cmdf@kernel.org\u003e\n Cc: linux-fpga@vger.kernel.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Link: https://lore.kernel.org/r/20190704055645.GA15471@kroah.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 1670e678a8a72e8b09774e21986d1d77d4610f96\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Jul 9 22:50:30 2019 +0530\n\n mtd: rawnand: ingenic: Add of_node_put() before return\n \n Each iteration of for_each_child_of_node puts the previous node, but in\n the case of a return from the middle of the loop, there is no put, thus\n causing a memory leak. Hence add an of_node_put before the return.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n\ncommit 60be51f4be49b1b1a40f2d51a17ea54217d3dcde\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Jul 9 22:46:40 2019 +0530\n\n mtd: rawnand: meson: Add of_node_put() before return\n \n Each iteration of for_each_child_of_node puts the previous node, but in\n the case of a return from the middle of the loop, there is no put, thus\n causing a memory leak. Hence add an of_node_put before the return.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Acked-by: Liang Yang \u003cliang.yang@amlogic.com\u003e\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n\ncommit a08e42987118952c98707f0b29cf0e5f4fd2e00b\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Jul 9 22:43:16 2019 +0530\n\n mtd: rawnand: tango: Add of_node_put() before return\n \n Each iteration of for_each_child_of_node puts the previous node, but in\n the case of a return from the middle of the loop, there is no put, thus\n causing a memory leak. Hence add an of_node_put before the return.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Acked-by: Mans Rullgard \u003cmans@mansr.com\u003e\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n\ncommit c436f68beeb20f2f92937677db1d9069b0dd2a3d\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Jul 9 22:40:16 2019 +0530\n\n mtd: rawnand: oxnas: Add of_node_put()\n \n Each iteration of for_each_child_of_node puts the previous node, but in\n the case of a goto from the middle of the loop, there is no put, thus\n causing a memory leak. Hence add an of_node_put under a new goto to put\n the node at a loop exit.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n\ncommit 4902e87faf26067d661e2a9285ce93457884da6d\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Jul 9 22:38:37 2019 +0530\n\n mtd: rawnand: vf610: Add of_node_put() before goto\n \n Each iteration of for_each_available_child_of_node puts the previous\n node, but in the case of a goto from the middle of the loop, there is no\n put, thus causing a memory leak. Hence add an of_node_put before the\n goto.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Reviewed-by: Stefan Agner \u003cstefan@agner.ch\u003e\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n\ncommit b410f4eb01a1950ed73ae40859d0978b1a924380\nAuthor: Christophe Kerello \u003cchristophe.kerello@st.com\u003e\nDate: Tue Jul 9 11:41:45 2019 +0200\n\n mtd: rawnand: stm32_fmc2: avoid warnings when building with W\u003d1 option\n \n This patch solves warnings detected by setting W\u003d1 when building.\n \n Warnings type detected:\n drivers/mtd/nand/raw/stm32_fmc2_nand.c: In function ‘stm32_fmc2_calc_timings’:\n drivers/mtd/nand/raw/stm32_fmc2_nand.c:1417:23: warning: comparison is\n always false due to limited range of data type [-Wtype-limits]\n else if (tims-\u003etwait \u003e FMC2_PMEM_PATT_TIMING_MASK)\n \n Signed-off-by: Christophe Kerello \u003cchristophe.kerello@st.com\u003e\n Cc: stable@vger.kernel.org\n Fixes: 2cd457f328c1 (\"mtd: rawnand: stm32_fmc2: add STM32 FMC2 NAND flash controller driver\")\n Signed-off-by: Miquel Raynal \u003cmiquel.raynal@bootlin.com\u003e\n\ncommit 17224cf29cbb5c496d08ffecc9db3c5e6cf09551\nAuthor: Olivier Moysan \u003colivier.moysan@st.com\u003e\nDate: Wed Jul 3 10:04:12 2019 +0200\n\n drm/bridge: sii902x: add audio graph card support\n \n Implement get_dai_id callback of audio HDMI codec\n to support ASoC audio graph card.\n HDMI audio output has to be connected to sii902x port 3.\n get_dai_id callback maps this port to ASoC DAI index 0.\n \n Signed-off-by: Olivier Moysan \u003colivier.moysan@st.com\u003e\n Reviewed-by: Philippe Cornu \u003cphilippe.cornu@st.com\u003e\n Acked-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Reviewed-by: Jyri Sarha \u003cjsarha@ti.com\u003e\n Signed-off-by: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/1562141052-26221-1-git-send-email-olivier.moysan@st.com\n\ncommit 820342aca05188c9af4b468d6c41b3327161f7ad\nAuthor: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\nDate: Thu Feb 28 08:25:28 2019 -0500\n\n media: v4l2-async: Add v4l2_async_notifier_add_fwnode_remote_subdev\n \n v4l2_async_notifier_add_fwnode_remote_subdev is a convenience function for\n parsing information on V4L2 fwnode subdevs.\n \n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Tested-by: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 016413d967061fc2eb6798a487b3022bef7698a6\nAuthor: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\nDate: Thu Apr 4 19:43:29 2019 -0400\n\n media: v4l2-async: Get fwnode reference when putting it to the notifier\u0027s list\n \n The v4l2_async_notifier_add_fwnode_subdev() did not take a reference of\n the added fwnode, relying on the caller to handle that instead, in essence\n putting the fwnode to be added if there was an error.\n \n As the reference is eventually released during the notifier cleanup, this\n is not intuitive nor logical. Improve this by always getting a reference\n when the function succeeds, and the caller releasing the reference when it\n does not *itself* need it anymore.\n \n Luckily, perhaps, there were just a handful of callers using the function.\n \n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Reviewed-by: Jacopo Mondi \u003cjacopo+renesas@jmondi.org\u003e\n Tested-by: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 6a76404b13ba7101216967dc447b77cf69e56adf\nAuthor: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\nDate: Thu Apr 4 18:41:19 2019 -0400\n\n media: davinci-vpif: Don\u0027t dereference endpoint after putting it, fix refcounting\n \n The davinci-vpif driver dereferences its local endpoints after releasing\n the reference to them.\n \n The driver also puts its endpoints explicitly while the\n of_graph_get_next_endpoint() does that, too, leading to obtaining a\n reference once and releasing it twice.\n \n Both are fixed by this patch.\n \n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 43e4cb942e88e756b41ea4d30249a47973880508\nAuthor: Pavel Begunkov \u003casml.silence@gmail.com\u003e\nDate: Wed Jul 24 20:16:31 2019 +0300\n\n locks: Fix procfs output for file leases\n \n Since commit 778fc546f749c588aa2f (\"locks: fix tracking of inprogress\n lease breaks\"), leases break don\u0027t change @fl_type but modifies\n @fl_flags. However, procfs\u0027s part haven\u0027t been updated.\n \n Previously, for a breaking lease the target type was printed (see\n target_leasetype()), as returns fcntl(F_GETLEASE). But now it\u0027s always\n \"READ\", as F_UNLCK no longer means \"breaking\". Unlike the previous\n one, this behaviour don\u0027t provide a complete description of the lease.\n \n There are /proc/pid/fdinfo/ outputs for a lease (the same for READ and\n WRITE) breaked by O_WRONLY.\n -- before:\n lock: 1: LEASE BREAKING READ 2558 08:03:815793 0 EOF\n -- after:\n lock: 1: LEASE BREAKING UNLCK 2558 08:03:815793 0 EOF\n \n Signed-off-by: Pavel Begunkov \u003casml.silence@gmail.com\u003e\n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n\ncommit 515db266a9dace92b0cbaed9a6044dd5304b8ca9\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Tue Jul 16 17:21:06 2019 +0200\n\n driver core: Remove device link creation limitation\n \n If device_link_add() is called for a consumer/supplier pair with an\n existing device link between them and the existing link\u0027s type is\n not in agreement with the flags passed to that function by its\n caller, NULL will be returned. That is seriously inconvenient,\n because it forces the callers of device_link_add() to worry about\n what others may or may not do even if that is not relevant to them\n for any other reasons.\n \n It turns out, however, that this limitation can be made go away\n relatively easily.\n \n The underlying observation is that if DL_FLAG_STATELESS has been\n passed to device_link_add() in flags for the given consumer/supplier\n pair at least once, calling either device_link_del() or\n device_link_remove() to release the link returned by it should work,\n but there are no other requirements associated with that flag. In\n turn, if at least one of the callers of device_link_add() for the\n given consumer/supplier pair has not passed DL_FLAG_STATELESS to it\n in flags, the driver core should track the status of the link and act\n on it as appropriate (ie. the link should be treated as \"managed\").\n This means that DL_FLAG_STATELESS needs to be set for managed device\n links and it should be valid to call device_link_del() or\n device_link_remove() to drop references to them in certain\n sutiations.\n \n To allow that to happen, introduce a new (internal) device link flag\n called DL_FLAG_MANAGED and make device_link_add() set it automatically\n whenever DL_FLAG_STATELESS is not passed to it. Also make it take\n additional references to existing device links that were previously\n stateless (that is, with DL_FLAG_STATELESS set and DL_FLAG_MANAGED\n unset) and will need to be managed going forward and initialize\n their status (which has been DL_STATE_NONE so far).\n \n Accordingly, when a managed device link is dropped automatically\n by the driver core, make it clear DL_FLAG_MANAGED, reset the link\u0027s\n status back to DL_STATE_NONE and drop the reference to it associated\n with DL_FLAG_MANAGED instead of just deleting it right away (to\n allow it to stay around in case it still needs to be released\n explicitly by someone).\n \n With that, since setting DL_FLAG_STATELESS doesn\u0027t mean that the\n device link in question is not managed any more, replace all of the\n status-tracking checks against DL_FLAG_STATELESS with analogous\n checks against DL_FLAG_MANAGED and update the documentation to\n reflect these changes.\n \n While at it, make device_link_add() reject flags that it does not\n recognize, including DL_FLAG_MANAGED.\n \n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Reviewed-by: Saravana Kannan \u003csaravanak@google.com\u003e\n Tested-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Review-by: Saravana Kannan \u003csaravanak@google.com\u003e\n Link: https://lore.kernel.org/r/2305283.AStDPdUUnE@kreacher\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7bdd9695970eb8cd5b987e08f99cb8f0c84f30fd\nAuthor: Jia-Ju Bai \u003cbaijiaju1990@gmail.com\u003e\nDate: Wed Jul 24 17:04:26 2019 +0800\n\n char: ppdev: Fix a possible null-pointer dereference in pp_release()\n \n In pp_release(), there is an if statement on line 730 to check whether\n pp-\u003epdev is NULL:\n else if ((pp-\u003eflags \u0026 PP_CLAIMED) \u0026\u0026 pp-\u003epdev \u0026\u0026 ...)\n \n When pp-\u003epdev is NULL, it is used on line 743:\n info \u003d \u0026pp-\u003epdev-\u003eport-\u003eieee1284;\n and on line 748:\n parport_release(pp-\u003epdev);\n \n Thus, a possible null-pointer dereference may occur.\n \n To fix this bug, pp-\u003epdev is checked on line 740.\n \n This bug is found by a static analysis tool STCheck written by us.\n \n Signed-off-by: Jia-Ju Bai \u003cbaijiaju1990@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190724090426.1401-1-baijiaju1990@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 092f47c383adbff4d873dcb7f0887954bd407a08\nAuthor: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\nDate: Fri Jul 5 17:14:25 2019 +0300\n\n intel_th: msu: Preserve pre-existing buffer configuration\n \n The MSU configuration registers may contain buffer address/size set by\n the BIOS or an external hardware debugger, which may want to take over\n tracing from the driver when the driver is not actively tracing.\n \n Preserve these settings when not actively tracing.\n \n Signed-off-by: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Reviewed-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190705141425.19894-6-alexander.shishkin@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit ed5207afc580937cd6cc671357fe1119f1b5b0f5\nAuthor: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\nDate: Fri Jul 5 17:14:24 2019 +0300\n\n intel_th: msu: Prevent freeing buffers while locked windows exist\n \n We already prevent freeing buffers via sysfs interface in case there are\n existing users or if trace is active. Treat the existence of locked windows\n similarly and return -EBUSY on attempts to free the buffer. When the last\n window is unlocked, the freeing will succeed.\n \n Signed-off-by: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Reviewed-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190705141425.19894-5-alexander.shishkin@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 57b69a1f66486d632e188fb167950fb4bdf84221\nAuthor: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\nDate: Fri Jul 5 17:14:23 2019 +0300\n\n intel_th: msu: Get rid of the window size limit\n \n Currently, the window size is limited to the maximum number of sg entries\n in one table. This is because the code addresses individual blocks within\n the window by their numeric index. In reality, though, the blocks most\n often are iterated through sequentially. By rewriting the logic to use sg\n pointers instead of block indices we loose the necessity to dereference\n them directly and gain the ability to use multiple chained tables if\n necessary.\n \n Get rid of the limitation by replacing index-based block accesses with\n sequential block accesses.\n \n Signed-off-by: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Reviewed-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190705141425.19894-4-alexander.shishkin@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f220df66f67684246ae1bf4a4e479efc7c2f325a\nAuthor: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\nDate: Fri Jul 5 17:14:22 2019 +0300\n\n intel_th: msu-sink: An example msu buffer \"sink\"\n \n This patch adds an example MSU buffer \"sink\", which consumes trace\n data from MSC buffers.\n \n Functionally, it acts similarly to \"multi\" mode with automatic window\n switching.\n \n Signed-off-by: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Reviewed-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190705141425.19894-3-alexander.shishkin@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 615c164da0eb42cbfb1688cb429cc4d5039db5d8\nAuthor: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\nDate: Fri Jul 5 17:14:21 2019 +0300\n\n intel_th: msu: Introduce buffer interface\n \n Introduces a concept of external buffers, which is a mechanism for creating\n trace sinks that would receive trace data from MSC buffers and transfer it\n elsewhere.\n \n A external buffer can implement its own window allocation/deallocation if\n it has to. It must provide a callback that\u0027s used to notify it when a\n window fills up, so that it can then start a DMA transaction from that\n window \u0027elsewhere\u0027. This window remains in a \u0027locked\u0027 state and won\u0027t be\n used for storing new trace data until the buffer \u0027unlocks\u0027 it with a\n provided API call, at which point the window can be used again for storing\n trace data.\n \n This relies on a functional \"last block\" interrupt, so not all versions of\n Trace Hub can use this feature, which does not reflect on existing users.\n \n Signed-off-by: Alexander Shishkin \u003calexander.shishkin@linux.intel.com\u003e\n Reviewed-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190705141425.19894-2-alexander.shishkin@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 6ddb094a9f8c451c42bc7c58cfed22275e8a1e38\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Wed Jul 24 08:22:19 2019 -0400\n\n media: staging/intel-ipu3: Use dev_get_drvdata where possible\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 3be9cc6ad3faaf403550aea3879743be7d1c307d\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 16:21:28 2019 -0400\n\n media: v4l2-fwnode: Switch to use fwnode_property_count_uXX()\n \n Use fwnode_property_count_uXX() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit bfe655d159af96632f4930c718e0fc5689460f98\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 07:50:14 2019 -0400\n\n media: pci: Use dev_get_drvdata where possible\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 913c55a86400e0b3c99164a71a826ffb860c26c4\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Sun Jul 21 15:04:23 2019 -0400\n\n media: ov2680: fix a typo in a function name\n \n All functions in this file starts with \u0027ov2680_\u0027, except ov2860_parse_dt().\n \n This is likely a typo.\n rename it to \u0027ov2680_parse_dt()\u0027 (6 and 8)\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 050ff2ad1cca6fe6fcb6edde6e3b91f7e47d1c9c\nAuthor: Jan Kotas \u003cjank@cadence.com\u003e\nDate: Mon Jul 22 04:22:23 2019 -0400\n\n media: Add support for Cadence CSI2TX 2.1\n \n This patch adds support for CSI2TX v2.1 version of the controller.\n \n Signed-off-by: Jan Kotas \u003cjank@cadence.com\u003e\n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 6ded416d4ac4ecbf104b897661cdfa2cdacf022a\nAuthor: Jan Kotas \u003cjank@cadence.com\u003e\nDate: Mon Jul 22 04:22:22 2019 -0400\n\n media: Fix Lane mapping in Cadence CSI2TX\n \n This patch fixes mapping of lanes in DPHY_CFG register\n of the controller. In the register, bit 0 means first data lane.\n In Linux we currently assume lane 0 is clock.\n \n Signed-off-by: Jan Kotas \u003cjank@cadence.com\u003e\n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit bf9df90b3557ec6d5d92914da6a61453741d3e13\nAuthor: Jan Kotas \u003cjank@cadence.com\u003e\nDate: Mon Jul 22 04:22:21 2019 -0400\n\n media: Add lane checks for Cadence CSI2TX\n \n This patch adds lane checks for CSI2TX, to prevent clock lane\n being used as a data lane.\n \n Signed-off-by: Jan Kotas \u003cjank@cadence.com\u003e\n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit c0a7c0025369b3f3c3d18be76d250024faac6459\nAuthor: Jan Kotas \u003cjank@cadence.com\u003e\nDate: Mon Jul 22 04:22:20 2019 -0400\n\n media: dt-bindings: Update bindings for Cadence CSI2TX version 2.1\n \n This patch adds a DT bindings documentation for\n Cadence CSI2TX v2.1 controller.\n \n Signed-off-by: Jan Kotas \u003cjank@cadence.com\u003e\n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 24c8ac890130fda9aa0368dabba1b08d57d3d99e\nAuthor: Fabio Estevam \u003cfestevam@gmail.com\u003e\nDate: Fri Jun 28 07:00:35 2019 -0400\n\n media: i2c: ov5640: Fix the order for enabling regulators\n \n According to the OV5640 datasheet the following sequence needs to be\n followed when powering the OV5640 supplies:\n \n - DOVDD\n - AVDD\n - DVDD\n \n So follow this order inside the ov5640_supply_name[] array, so that\n the regulator_bulk() functions can enable the regulator in the\n correct sequence.\n \n Signed-off-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 8791a102ce579346cea9d2f911afef1c1985213c\nAuthor: Fabio Estevam \u003cfestevam@gmail.com\u003e\nDate: Fri Jun 28 07:00:34 2019 -0400\n\n media: i2c: ov5640: Check for devm_gpiod_get_optional() error\n \n The power down and reset GPIO are optional, but the return value\n from devm_gpiod_get_optional() needs to be checked and propagated\n in the case of error, so that probe deferral can work.\n \n Signed-off-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 4962bb38fb61518c3c5e9c8f4aedcf2e3486fc4b\nAuthor: Fabio Estevam \u003cfestevam@gmail.com\u003e\nDate: Thu Jun 27 14:22:31 2019 -0400\n\n media: ov5645: Use regulator_bulk() functions\n \n The code can be simplified by using the regulator_bulk() functions,\n so switch to it.\n \n Signed-off-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 45ffbd15ede6add6e4fb150e4bab7a27cfe62552\nAuthor: Fabio Estevam \u003cfestevam@gmail.com\u003e\nDate: Thu Jun 27 14:22:30 2019 -0400\n\n media: ov5645: Remove unneeded regulator_set_voltage()\n \n There is no need to call regulator_set_voltage() for each regulator\n that powers the camera.\n \n The voltage value for each regulator should be retrieved from the\n device tree, so remove the unneeded regulator_set_voltage().\n \n Signed-off-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit f61c0ca4b402af1ee619e989efadcf678051ee47\nAuthor: Hans Verkuil \u003chverkuil@xs4all.nl\u003e\nDate: Thu Jul 25 03:52:19 2019 -0400\n\n media: vivid: fix typo: use -\u003e user\n \n Fix typo in two comments.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 4ee222406740f0c39e00b12f51eda8133e35ee83\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Thu Jul 25 03:30:23 2019 -0400\n\n media: staging: davinci: remove vpfe driver\n \n The davinci_vpfe driver was merged into staging back in 2012 by Manjunath\n Hadli from TI, with a long TODO list.\n \n For all I can tell, since then it has only seen fixes for compile-time\n issues and global cleanups, but nobody has actually worked on the items\n on the TODO list.\n \n To make things worse, the driver in its current form is incompatible with\n the platform code in arch/arm/mach-davinci, i.e. the driver expects to\n get its platform_data passed to the device as a \u0027struct vpfe_config\u0027,\n but uses a differnet definition for that structure compared to what the\n platform uses.\n \n Finally, there is another driver for the same device in\n drivers/media/platform/davinci/vpfe_capture.c. From all I can tell, the\n staging version was originally a copy of a more featureful driver in TI\u0027s\n downstream kernels. However, that kernel no longer supports dm365 after\n linux-2.6.37, and the mainline version moved in a different direction.\n \n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 757c2bf597a8e9f859f8efceda9c5108607fd98d\nAuthor: Hans Verkuil \u003chverkuil@xs4all.nl\u003e\nDate: Thu Jul 25 02:31:53 2019 -0400\n\n media: staging/media/bcm2048: remove driver\n \n This driver has seen no substantial effort to move it out of\n staging for 6 years. And after asking around nobody stepped up\n to do the remaining work. So remove it.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 79e85d1d2c16ba4907bb9d6a4381516b729ff341\nAuthor: Guillaume Tucker \u003cguillaume.tucker@collabora.com\u003e\nDate: Wed Jul 24 11:19:22 2019 -0400\n\n media: vivid: fix device init when no_error_inj\u003d1 and fb disabled\n \n Add an extra condition to add the video output control class when the\n device has some hdmi outputs defined. This is required to then always\n be able to add the display present control, which is enabled when\n there are some hdmi outputs.\n \n This fixes the corner case where no_error_inj is enabled and the\n device has no frame buffer but some hdmi outputs, as otherwise the\n video output control class would be added anyway. Without this fix,\n the sanity checks fail in v4l2_ctrl_new() as name is NULL.\n \n Fixes: c533435ffb91 (\"media: vivid: add display present control\")\n Cc: stable@vger.kernel.org # for 5.3\n Signed-off-by: Guillaume Tucker \u003cguillaume.tucker@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 8baf50109929c9b407be756c52f4f80a659cb705\nAuthor: Hans Verkuil \u003chverkuil@xs4all.nl\u003e\nDate: Tue Jul 23 08:59:47 2019 -0400\n\n media: am437x: remove unused struct vpfe_pixel_format\n \n struct vpfe_pixel_format was defined, but never used. Remove it.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Acked-by: Lad, Prabhakar \u003cprabhakar.csengg@gmail.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit ee484875af0080a0c8621e58facf06f92c658a4b\nAuthor: Hans Verkuil \u003chverkuil@xs4all.nl\u003e\nDate: Tue Jul 23 08:58:01 2019 -0400\n\n media: davinci/vpfe_capture.c: drop unused format descriptions\n \n Simplify vpfe_pixel_format to just contain the pixelformat and bpp fields.\n All others are unused.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Acked-by: Lad, Prabhakar \u003cprabhakar.csengg@gmail.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 610fce5372216652cdcf6854c48d2876d1eeaad3\nAuthor: Hans Verkuil \u003chverkuil@xs4all.nl\u003e\nDate: Tue Jul 23 08:25:49 2019 -0400\n\n media: sh_veu: convert to struct v4l2_fh\n \n This driver didn\u0027t use struct v4l2_fh, so add it.\n \n This is a very basic conversion. I can\u0027t test this on real hardware,\n so I didn\u0027t dare to also convert the driver to start using the\n v4l2-mem2mem ioctl/fop helpers.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 3c1350501c21db8e3b1a38d9e97db29694305c3b\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Tue Jul 23 04:21:25 2019 -0400\n\n media: v4l2-dev/ioctl: require non-zero device_caps, verify sane querycap results\n \n Now that all V4L2 drivers set device_caps in struct video_device, we can add\n a check for this to ensure all future drivers fill this in.\n \n Also verify that when the querycap ioctl is called the driver didn\u0027t mess\n with the device_caps value and that capabilities is a superset of device_caps.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Reviewed-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n [hverkuil-cisco@xs4all.nl: fix too-long line]\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 08aac0e32fe44b0afe67da96b01c5c0aa27c7b67\nAuthor: Lubomir Rintel \u003clkundrak@v3.sk\u003e\nDate: Mon Jul 22 07:33:10 2019 -0400\n\n media: marvell-ccic: mmp: add MODULE_DEVICE_TABLE\n \n This fixes autoloading the module by the OF compatible string.\n \n Fixes: 83c40e6611ec (\"media: marvell-ccic/mmp: add devicetree support\")\n Signed-off-by: Lubomir Rintel \u003clkundrak@v3.sk\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 173f6eacc8a89ae4b08b166735930e6c46951a81\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Sat Jul 20 07:47:07 2019 -0400\n\n media: v4l: ctrls: Add debug messages\n \n Currently, the v4l2 control code is a bit silent on errors.\n Add debug messages on (hopefully) most of the error paths.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 156a599b0716ab3ee3869ff26119e3b5d46d91c8\nAuthor: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nDate: Fri Jul 5 18:05:19 2019 +0300\n\n dmaengine: tegra-apb: Support per-burst residue granularity\n \n Tegra\u0027s APB DMA engine updates words counter after each transferred burst\n of data, hence it can report transfer\u0027s residual with more fidelity which\n may be required in cases like audio playback. In particular this fixes\n audio stuttering during playback in a chromium web browser. The patch is\n based on the original work that was made by Ben Dooks and a patch from\n downstream kernel. It was tested on Tegra20 and Tegra30 devices.\n \n Link: https://lore.kernel.org/lkml/20190424162348.23692-1-ben.dooks@codethink.co.uk/\n Link: https://nv-tegra.nvidia.com/gitweb/?p\u003dlinux-4.4.git;a\u003dcommit;h\u003dc7bba40c6846fbf3eaad35c4472dcc7d8bbc02e5\n Inspired-by: Ben Dooks \u003cben.dooks@codethink.co.uk\u003e\n Reviewed-by: Jon Hunter \u003cjonathanh@nvidia.com\u003e\n Signed-off-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190705150519.18171-1-digetx@gmail.com\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit d4a6a9537bc32811486282206ecfb7c53754b74d\nAuthor: Luke Nowakowski-Krijger \u003clnowakow@eng.ucsd.edu\u003e\nDate: Wed Jul 17 10:19:46 2019 -0400\n\n media: hdpvr: Add device num check and handling\n \n Add hdpvr device num check and error handling\n \n We need to increment the device count atomically before we checkout a\n device to make sure that we do not reach the max count, otherwise we get\n out-of-bounds errors as reported by syzbot.\n \n Reported-and-tested-by: syzbot+aac8d0d7205f112045d2@syzkaller.appspotmail.com\n \n Signed-off-by: Luke Nowakowski-Krijger \u003clnowakow@eng.ucsd.edu\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 9e037bdf743cc081858423ad4123824e846b2358\nAuthor: Joe Perches \u003cjoe@perches.com\u003e\nDate: Wed Jul 10 01:04:24 2019 -0400\n\n media: staging: media: cedrus: Fix misuse of GENMASK macro\n \n Arguments are supposed to be ordered high then low.\n \n Signed-off-by: Joe Perches \u003cjoe@perches.com\u003e\n Acked-by: Paul Kocialkowski \u003cpaul.kocialkowski@bootlin.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 21a816e7ee5d4b217da14374af1c91eb725e665c\nAuthor: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\nDate: Fri Jul 5 00:43:50 2019 -0400\n\n media: rcar-vin: Centralize black listing of pixel formats\n \n Instead of scattering black listing of pixel formats who are not\n supported on particular platforms move it to a central location.\n \n Signed-off-by: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\n Reviewed-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit df2f84dfa709a4d44331ba6caddb31cd0654f37e\nAuthor: Erwan Le Ray \u003cerwan.leray@st.com\u003e\nDate: Thu Jun 13 15:49:55 2019 +0200\n\n serial: stm32: Use __maybe_unused instead of #if CONFIG_PM_SLEEP\n \n Use __maybe_unused for power management related functionsinstead of\n \n fixes: 270e5a74fe4c (\"serial: stm32: add wakeup mechanism\")\n Signed-off-by: Erwan Le Ray \u003cerwan.leray@st.com\u003e\n \n Link: https://lore.kernel.org/r/1560433800-12255-6-git-send-email-erwan.leray@st.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7bb4d2543364a10c6cc255309470d7b201a58d30\nAuthor: Erwan Le Ray \u003cerwan.leray@st.com\u003e\nDate: Thu Jun 13 15:49:54 2019 +0200\n\n serial: stm32: add pm_runtime support\n \n Use pm_runtime for clock management.\n \n Signed-off-by: Bich Hemon \u003cbich.hemon@st.com\u003e\n Signed-off-by: Erwan Le Ray \u003cerwan.leray@st.com\u003e\n \n Link: https://lore.kernel.org/r/1560433800-12255-5-git-send-email-erwan.leray@st.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b5500d9d651d76f29bdce016a5e53575f9df5d9b\nAuthor: Erwan Le Ray \u003cerwan.leray@st.com\u003e\nDate: Thu Jun 13 15:49:53 2019 +0200\n\n serial: stm32: select pinctrl state in each suspend/resume function\n \n Select either pinctrl sleep state in suspend function or default state in\n resume function.\n \n Signed-off-by: Bich Hemon \u003cbich.hemon@st.com\u003e\n Signed-off-by: Erwan Le Ray \u003cerwan.leray@st.com\u003e\n \n Link: https://lore.kernel.org/r/1560433800-12255-4-git-send-email-erwan.leray@st.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 23e343b5e432b35c96d1d798a4da33389fdf7511\nAuthor: Erwan Le Ray \u003cerwan.leray@st.com\u003e\nDate: Thu Jun 13 15:49:51 2019 +0200\n\n dt-bindings: serial: stm32: add wakeup option\n \n Add a note for enabling wakeup capabilities of usart\n \n Signed-off-by: Bich Hemon \u003cbich.hemon@st.com\u003e\n Signed-off-by: Erwan Le Ray \u003cerwan.leray@st.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n \n Link: https://lore.kernel.org/r/1560433800-12255-2-git-send-email-erwan.leray@st.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 0bd465765f8d1300040de627f0a0971a8849d654\nAuthor: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\nDate: Tue Jul 2 13:42:58 2019 -0400\n\n media: rcar-vin: Clean up correct notifier in error path\n \n The parallel input initialization error path cleans up the wrong\n async notifier, fix this by cleaning up the correct notifier.\n \n Fixes: 9863bc8695bc36e3 (\"media: rcar-vin: Cleanup notifier in error path\")\n Signed-off-by: Niklas Söderlund \u003cniklas.soderlund+renesas@ragnatech.se\u003e\n Reviewed-by: Jacopo Mondi \u003cjacopo+renesas@jmondi.org\u003e\n Tested-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Reviewed-by: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 78b3e86b5a75b6fb3122b11d507023f1b194f663\nAuthor: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\nDate: Sun Jun 23 13:07:26 2019 -0400\n\n media: dt: bindings: tegra-vde: Document new optional IOMMU property\n \n All NVIDIA Tegra SoC generations provide IOMMU support for the video\n decoder engine. Document new optional device-tree property that connects\n VDE with the IOMMU provider.\n \n Signed-off-by: Dmitry Osipenko \u003cdigetx@gmail.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit f5d6aadf3b6434f11393e33be9fd25a56d0bc872\nAuthor: Navid Emamdoost \u003cnavid.emamdoost@gmail.com\u003e\nDate: Fri Jul 19 12:48:45 2019 -0500\n\n 8250_lpss: check null return when calling pci_ioremap_bar\n \n pci_ioremap_bar may return null. This is eventually de-referenced at\n drivers/dma/dw/core.c:1154 and drivers/dma/dw/core.c:1168. A null check\n is needed to prevent null de-reference. I am adding the check and in\n case of failure. Thanks to Andy Shevchenko for the hint on the necessity\n of pci_iounmap when exiting.\n \n Signed-off-by: Navid Emamdoost \u003cnavid.emamdoost@gmail.com\u003e\n Reviewed-by: Andy Shevchenko \u003candy.shevchenko@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190719174848.24216-1-navid.emamdoost@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit be2847391ac3498581590a84c25a582cd14e9cea\nAuthor: Fugang Duan \u003cfugang.duan@nxp.com\u003e\nDate: Wed Jul 17 13:19:30 2019 +0800\n\n tty: serial: fsl_lpuart: correct the FIFO depth size\n \n VF610/LS1021a/i.MX7ULP/i.MX8QXP reference manual describe the\n TXFIFOSIZE/RXFIFOSIZE field as below.\n \n 000b - FIFO/Buffer depth \u003d 1 dataword.\n 001b - FIFO/Buffer depth \u003d 4 datawords.\n 010b - FIFO/Buffer depth \u003d 8 datawords.\n 011b - FIFO/Buffer depth \u003d 16 datawords.\n 100b - FIFO/Buffer depth \u003d 32 datawords.\n 101b - FIFO/Buffer depth \u003d 64 datawords.\n 110b - FIFO/Buffer depth \u003d 128 datawords.\n 111b - FIFO/Buffer depth \u003d 256 datawords. (Reserved for VF610)\n \n So the FIFO depth should be: 0x1 \u003c\u003c (val ? (val + 1) : 0)\n \n Signed-off-by: Fugang Duan \u003cfugang.duan@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190717051930.15514-6-fugang.duan@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d9aa9ab4fe6b5c43b9ccb8a0811dadcfe40ea27f\nAuthor: Fugang Duan \u003cfugang.duan@nxp.com\u003e\nDate: Wed Jul 17 13:19:29 2019 +0800\n\n tty: serial: fsl_lpuart: remove sg_set_buf() for sport-\u003erx_sgl\n \n Since .sg_init_one() already set sg entry page like below code.\n sg_init_one()\n sg_init_table(sg, 1);\n sg_set_buf(sg, buf, buflen);\n \n So it should not set sg entry page again, remove the redundant code.\n \n Signed-off-by: Fugang Duan \u003cfugang.duan@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190717051930.15514-5-fugang.duan@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 5791a830470a917836fd8edce4a92a812a4aedfa\nAuthor: Fugang Duan \u003cfugang.duan@nxp.com\u003e\nDate: Wed Jul 17 13:19:28 2019 +0800\n\n tty: serial: fsl_lpuart: use kzalloc() instead of kmalloc()\n \n Use kzalloc() instead of kmalloc() to get clean rx buffer\n that is useful for DMA mode debug to check the data moving\n validity.\n \n Signed-off-by: Fugang Duan \u003cfugang.duan@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190717051930.15514-4-fugang.duan@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a017d5deed22ffa709da28e57ee50cdecfd54405\nAuthor: Fugang Duan \u003cfugang.duan@nxp.com\u003e\nDate: Wed Jul 17 13:19:27 2019 +0800\n\n tty: serial: fsl_lpuart: add earlycon for imx8qxp platform\n \n Add earlycon support for imx8qxp platform.\n \n Signed-off-by: Fugang Duan \u003cfugang.duan@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190717051930.15514-3-fugang.duan@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 56acec623410cac89128761a781dc641b2ab6645\nAuthor: Fugang Duan \u003cfugang.duan@nxp.com\u003e\nDate: Wed Jul 17 13:19:26 2019 +0800\n\n tty: serial: fsl_lpuart: remove the dev.coherent_dma_mask zero setting\n \n By default, .of_dma_configure() init dev.coherent_dma_mask to BIT(32) that\n match the eDMA address range. If re-init dev.coherent_dma_mask to zero, then\n streaming dma mapping will go swiotlb dma_map, if swiotlb is not initalized\n then it causes mapping failed.\n \n Signed-off-by: Fugang Duan \u003cfugang.duan@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190717051930.15514-2-fugang.duan@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 78219b6c68ffafe85aa79431f6a6d59767ce146e\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Sun Jul 21 18:01:35 2019 +0300\n\n serial: 8250_exar: Use struct_size() helper\n \n One of the more common cases of allocation size calculations is finding\n the size of a structure that has a zero-sized array at the end, along\n with memory for some number of elements for that array.\n \n Make use of the struct_size() helper instead of an open-coded version\n in order to avoid any potential type mistakes.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190721150135.82065-1-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 026bd4a5b3c0f15acff88c0d7b73c6eca7560234\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Wed Jul 24 21:18:25 2019 +0800\n\n tty: nozomi: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190724131825.1875-1-hslester96@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit e56732b536110c5449b2978b4dd1e3eb4bc77968\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Wed Jul 24 21:17:58 2019 +0800\n\n serial: 8250: Use dev_get_drvdata where possible\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190724131758.1764-1-hslester96@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit ec7885e072e4f1d77f38d2772736591b7268045f\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Tue Jul 23 16:03:14 2019 +0100\n\n tty/isicom: remove redundant assignment to variable word_count\n \n The variable word_count is being assigned a value that is never read before\n a return, the assignment is redundant and can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Acked-by: Jiri Slaby \u003cjslaby@suse.cz\u003e\n Link: https://lore.kernel.org/r/20190723150314.14513-1-colin.king@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 44fe5ea44bf2d61ed7b891b2bba30427cfd6da8c\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Mon Jul 15 11:20:01 2019 +0800\n\n tty: serial: Remove call to memset after pci_alloc_consistent\n \n pci_alloc_consistent calls dma_alloc_coherent directly.\n In commit 518a2f1925c3\n (\"dma-mapping: zero memory returned from dma_alloc_*\"),\n dma_alloc_coherent has already zeroed the memory.\n So memset is not needed.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190715032001.7212-1-huangfq.daxian@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit fc78bd1fbd2df9102e4dc17dd22025f5d4bbf2d3\nAuthor: Joe Perches \u003cjoe@perches.com\u003e\nDate: Thu Jul 4 16:57:47 2019 -0700\n\n tty: hvcs: Fix odd use of strlcpy\n \n Use the typical style of array, not the equivalent \u0026array[0].\n \n Signed-off-by: Joe Perches \u003cjoe@perches.com\u003e\n Link: https://lore.kernel.org/r/1be432798311378947ec4e6051cad1235b6eccbb.1562283944.git.joe@perches.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 978946e428409bc1530fd80df5e63ffae9508273\nAuthor: Christophe Kerello \u003cchristophe.kerello@st.com\u003e\nDate: Fri Jun 21 16:49:49 2019 +0200\n\n ARM: dts: stm32: enable FMC2 NAND controller on stm32mp157c-ev1\n \n This patch enables FMC2 NAND controller used on stm32mp157c-ev1.\n \n Signed-off-by: Christophe Kerello \u003cchristophe.kerello@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit 52ded6f9ce84f15275a9b583bee55bd0361f9911\nAuthor: Christophe Kerello \u003cchristophe.kerello@st.com\u003e\nDate: Fri Jun 21 16:49:48 2019 +0200\n\n ARM: dts: stm32: add FMC2 NAND controller pins muxing on stm32mp157c-ev1\n \n This patch adds FMC2 NAND controller pins muxing used on stm32mp157c-ev1.\n \n Signed-off-by: Christophe Kerello \u003cchristophe.kerello@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit aafa0ae33586a2e3efcd76cfe5acaf292e95b9e1\nAuthor: Christophe Kerello \u003cchristophe.kerello@st.com\u003e\nDate: Fri Jun 21 16:49:47 2019 +0200\n\n ARM: dts: stm32: add FMC2 NAND controller support on stm32mp157c\n \n This patch adds FMC2 NAND controller support used by stm32mp157c SOC.\n \n Signed-off-by: Christophe Kerello \u003cchristophe.kerello@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit 6b72c7a98bd631533c4f4ab293c8082abbb42601\nAuthor: Olivier Moysan \u003colivier.moysan@st.com\u003e\nDate: Mon Jul 22 18:06:00 2019 +0200\n\n drm/bridge: sii902x: make audio mclk optional\n \n The master clock on i2s bus is not mandatory,\n as sii902X internal PLL can be used instead.\n Make use of mclk optional.\n \n Signed-off-by: Olivier Moysan \u003colivier.moysan@st.com\u003e\n Reviewed-by: Jyri Sarha \u003cjsarha@ti.com\u003e\n Acked-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Signed-off-by: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/1563811560-29589-4-git-send-email-olivier.moysan@st.com\n\ncommit 512ec50a97d80f604f166da08258395c6f50a607\nAuthor: Olivier Moysan \u003colivier.moysan@st.com\u003e\nDate: Mon Jul 22 18:05:59 2019 +0200\n\n dt-bindings: display: sii902x: Change audio mclk binding\n \n As stated in SiL9022/24 datasheet, master clock is not required for I2S.\n Make mclk property optional in DT bindings.\n \n Fixes: 3f18021f43a3 (\"dt-bindings: display: sii902x: Add HDMI audio bindings\")\n \n Signed-off-by: Olivier Moysan \u003colivier.moysan@st.com\u003e\n Reviewed-by: Jyri Sarha \u003cjsarha@ti.com\u003e\n Acked-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Signed-off-by: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/1563811560-29589-3-git-send-email-olivier.moysan@st.com\n\ncommit 8d34ce5cde995014ebbf588d61ffdb3367673d37\nAuthor: Tony Lindgren \u003ctony@atomide.com\u003e\nDate: Tue Jul 23 04:54:00 2019 -0700\n\n serial: 8250_omap: Fix idling for unloaded serdev drivers\n \n For many years omap variants have been setting the runtime PM\n autosuspend delay to -1 to prevent unsafe policy with lossy first\n character on wake-up. The user must specifically enable the timeout\n for UARTs if desired.\n \n We must not enable the workaround for serdev devices though. It leads\n into UARTs not idling if no serdev devices are loaded and there is no\n sysfs entry to configure the UART in that case. And this means that\n my PM may not work unless the serdev modules are loaded.\n \n We can detect a serdev device being configured based on a dts child\n node, and we can simply skip the workround in that case. And the\n serdev driver can idle the port during runtime when suitable if an\n out-of-band wake-up GPIO line exists for example.\n \n Let\u0027s also add some comments to the workaround while at it.\n \n Cc: Johan Hovold \u003cjohan@kernel.org\u003e\n Signed-off-by: Tony Lindgren \u003ctony@atomide.com\u003e\n Link: https://lore.kernel.org/r/20190723115400.46432-1-tony@atomide.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 365d28c92f8cd3d3860f8dd057a8c136e24b3698\nAuthor: Olivier Moysan \u003colivier.moysan@st.com\u003e\nDate: Mon Jul 22 18:05:58 2019 +0200\n\n drm/bridge: sii902x: fix missing reference to mclk clock\n \n Add devm_clk_get call to retrieve reference to master clock.\n \n Fixes: ff5781634c41 (\"drm/bridge: sii902x: Implement HDMI audio support\")\n \n Signed-off-by: Olivier Moysan \u003colivier.moysan@st.com\u003e\n Reviewed-by: Jyri Sarha \u003cjsarha@ti.com\u003e\n Acked-by: Andrzej Hajda \u003ca.hajda@samsung.com\n Signed-off-by: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/1563811560-29589-2-git-send-email-olivier.moysan@st.com\n\ncommit 8fabc9c3109a71b3577959a05408153ae69ccd8d\nAuthor: Ahmad Fatoum \u003ca.fatoum@pengutronix.de\u003e\nDate: Fri Jul 12 10:42:28 2019 +0200\n\n drm/stm: attach gem fence to atomic state\n \n To properly synchronize with other devices the fence from the GEM\n object backing the framebuffer needs to be attached to the atomic\n state, so the commit work can wait on fence signaling.\n \n Signed-off-by: Ahmad Fatoum \u003ca.fatoum@pengutronix.de\u003e\n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Acked-by: Philippe Cornu \u003cphilippe.cornu@st.com\u003e\n Tested-by: Philippe Cornu \u003cphilippe.cornu@st.com\u003e\n Signed-off-by: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712084228.8338-1-l.stach@pengutronix.de\n\ncommit 2c4313cb1162a12801d49b20fc263a12e3d93596\nAuthor: Martin Hundebøll \u003cmartin@geanix.com\u003e\nDate: Wed Jul 10 21:26:55 2019 +0200\n\n tty: n_gsm: add helpers to convert mux-num to/from tty-base\n \n Make it obvious how the gsm mux number relates to the virtual tty lines\n by using helper functions instead of shifting 6 bits.\n \n Signed-off-by: Martin Hundebøll \u003cmartin@geanix.com\u003e\n Link: https://lore.kernel.org/r/20190710192656.60381-3-martin@geanix.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit a022e71ba40482e548ceafcda5a2687f09f85ad4\nAuthor: Martin Hundebøll \u003cmartin@geanix.com\u003e\nDate: Wed Jul 10 21:26:54 2019 +0200\n\n tty: n_gsm: update doc example to use header for N_GSM0710 define\n \n There is no reason to gues the line discipline number when it is\n available from tty.h\n \n Signed-off-by: Martin Hundebøll \u003cmartin@geanix.com\u003e\n Link: https://lore.kernel.org/r/20190710192656.60381-2-martin@geanix.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 2e8d58c1f14ff2b535b3b4e19513ed8520d5ad2e\nAuthor: Martin Hundebøll \u003cmartin@geanix.com\u003e\nDate: Wed Jul 10 21:26:53 2019 +0200\n\n tty: n_gsm: remove obsolete mknod doc example\n \n The n_gsm driver handles registration of /dev/gsmttyX nodes, so there\u0027s\n no need to do mknod manually.\n \n Signed-off-by: Martin Hundebøll \u003cmartin@geanix.com\u003e\n Link: https://lore.kernel.org/r/20190710192656.60381-1-martin@geanix.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 2fa278e32b9f9489c85659c7da6fd1e76851ba9b\nAuthor: Ludovic Barre \u003cludovic.barre@st.com\u003e\nDate: Thu Jun 27 14:09:05 2019 +0200\n\n ARM: dts: stm32: activate dma for qspi on stm32mp157\n \n This patch activates dma for qspi on stm32mp157.\n \n Signed-off-by: Ludovic Barre \u003cludovic.barre@st.com\u003e\n Signed-off-by: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n\ncommit 34d0545978b65f0a4ced613eb99cd75b2a6132e9\nAuthor: Chunfeng Yun \u003cchunfeng.yun@mediatek.com\u003e\nDate: Wed Jun 12 13:55:17 2019 +0800\n\n dt-bindings: usb: mtu3: fix typo of DMA clock name\n \n Fix typo of dma_ck\n \n Signed-off-by: Chunfeng Yun \u003cchunfeng.yun@mediatek.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://lore.kernel.org/r/5e06482a0be15476c7b5825f155accf98275afa8.1560246390.git.chunfeng.yun@mediatek.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit ed7fe551575cd6dbbe31282a0869682aeea2627c\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Thu Jul 4 16:03:41 2019 +0100\n\n usb: chipidea: udc: remove redundant assignment to variable retval\n \n The variable retval is being initialized with a value that is never\n read and it is being updated later with a new value. The\n initialization is redundant and can be removed. Also remove a blank\n line.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190704150341.759-1-colin.king@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 6269e4c76eacabaea0d0099200ae1a455768d208\nAuthor: Ruslan Bilovol \u003cruslan.bilovol@gmail.com\u003e\nDate: Sun Jul 7 15:17:19 2019 +0300\n\n usb: host: xhci-hub: fix extra endianness conversion\n \n Don\u0027t do extra cpu_to_le32 conversion for\n put_unaligned_le32 because it is already implemented\n in this function.\n \n Fixes sparse error:\n xhci-hub.c:1152:44: warning: incorrect type in argument 1 (different base types)\n xhci-hub.c:1152:44: expected unsigned int [usertype] val\n xhci-hub.c:1152:44: got restricted __le32 [usertype]\n \n Fixes: 395f540 \"xhci: support new USB 3.1 hub request to get extended port status\"\n Cc: Mathias Nyman \u003cmathias.nyman@linux.intel.com\u003e\n Signed-off-by: Ruslan Bilovol \u003cruslan.bilovol@gmail.com\u003e\n Link: https://lore.kernel.org/r/1562501839-26522-1-git-send-email-ruslan.bilovol@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 3114bc89c0217254d3df10a42eab6b6ca00f547c\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Mon Jul 15 11:20:10 2019 +0800\n\n usb: host: Remove call to memset after dma_alloc_coherent\n \n In commit 518a2f1925c3\n (\"dma-mapping: zero memory returned from dma_alloc_*\"),\n dma_alloc_coherent has already zeroed the memory.\n So memset is not needed.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190715032010.7258-1-huangfq.daxian@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 6443c671d5cbc60932b31f19f48091cddb2285eb\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sun Jul 21 23:49:09 2019 +0900\n\n usb: host: oxu210hp-hcd: squash oxu210hp.h into oxu210hp-hcd.c\n \n The header, oxu210hp.h is only included from oxu210hp-hcd.c\n so squash it.\n \n When I moved the code, I also fixed the following warnings from\n scripts/checkpatch.pl:\n \n drivers/usb/host/oxu210hp-hcd.c:117: warning: __packed is preferred over __attribute__((packed))\n drivers/usb/host/oxu210hp-hcd.c:196: warning: __packed is preferred over __attribute__((packed))\n drivers/usb/host/oxu210hp-hcd.c:221: warning: __packed is preferred over __attribute__((packed))\n drivers/usb/host/oxu210hp-hcd.c:266: warning: __aligned(size) is preferred over __attribute__((aligned(size)))\n drivers/usb/host/oxu210hp-hcd.c:336: warning: Prefer \u0027unsigned int\u0027 to bare use of \u0027unsigned\u0027\n drivers/usb/host/oxu210hp-hcd.c:354: warning: __aligned(size) is preferred over __attribute__((aligned(size)))\n drivers/usb/host/oxu210hp-hcd.c:385: warning: __aligned(size) is preferred over __attribute__((aligned(size)))\n drivers/usb/host/oxu210hp-hcd.c:393: warning: __aligned(size) is preferred over __attribute__((aligned(size)))\n drivers/usb/host/oxu210hp-hcd.c:429: warning: Prefer \u0027unsigned int\u0027 to bare use of \u0027unsigned\u0027\n drivers/usb/host/oxu210hp-hcd.c:432: warning: Prefer \u0027unsigned int\u0027 to bare use of \u0027unsigned\u0027\n drivers/usb/host/oxu210hp-hcd.c:436: warning: Prefer \u0027unsigned int\u0027 to bare use of \u0027unsigned\u0027\n drivers/usb/host/oxu210hp-hcd.c:451: warning: Prefer \u0027unsigned int\u0027 to bare use of \u0027unsigned\u0027\n drivers/usb/host/oxu210hp-hcd.c:461: warning: Prefer \u0027unsigned int\u0027 to bare use of \u0027unsigned\u0027\n drivers/usb/host/oxu210hp-hcd.c:467: warning: please, no space before tabs\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Link: https://lore.kernel.org/r/20190721144909.5295-2-yamada.masahiro@socionext.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 110f87a6a5f65e825852632b9557f463e65251d7\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Sun Jul 21 23:49:08 2019 +0900\n\n usb: host: oxu210hp-hcd: remove include/linux/oxu210hp.h\n \n struct oxu210hp_platform_data is defined, but not used at all.\n \n $ git grep oxu210hp_platform_data\n include/linux/oxu210hp.h:struct oxu210hp_platform_data {\n \n include/linux/oxu210hp.h exists just for defining an unused structure,\n so it can go away.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Link: https://lore.kernel.org/r/20190721144909.5295-1-yamada.masahiro@socionext.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 6ce9ba07b1b78ca3412eac808cd59eea09c80468\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 22:47:26 2019 +0300\n\n usb: typec: mux: Switch to use fwnode_property_count_uXX()\n \n Use fwnode_property_count_uXX() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190723194726.69545-1-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit bc7514925cf50bf81f514d0f2d7521b25973a639\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 22:37:50 2019 +0300\n\n usb: typec: tcpm: Switch to use fwnode_property_count_uXX()\n \n Use use fwnode_property_count_uXX() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Link: https://lore.kernel.org/r/20190723193750.69038-1-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 41b57327b270935de8023f5473520dd6a81bd013\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 22:17:04 2019 +0300\n\n usb: dwc3: Switch to use device_property_count_u32()\n \n Use use device_property_count_u32() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190723191704.67945-1-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 07e803ec9f6be4b216b89f8530ea30d2a397696d\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 22:16:39 2019 +0300\n\n usb: dwc2: Switch to use device_property_count_u32()\n \n Use use device_property_count_u32() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Minas Harutyunyan \u003chminas@synopsys.com\u003e\n Link: https://lore.kernel.org/r/20190723191639.67883-1-andriy.shevchenko@linux.intel.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 57cd166afc4b430d7b98dca71911bfef478d236b\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Wed Jul 24 21:18:38 2019 +0800\n\n USB: core: hcd-pci: Use dev_get_drvdata where possible\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190724131838.1931-1-hslester96@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f27e43b46dae63a20ad01ba76f1ccb7cc46b28db\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Mon Jul 22 12:43:12 2019 +0200\n\n MAINTAINERS: Remove tinydrm entry\n \n tinydrm is just a collection of tiny drivers now.\n Add T: drm-misc entry for tinydrm drivers that lacks it.\n \n Cc: David Lechner \u003cdavid@lechnology.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Acked-by: David Lechner \u003cdavid@lechnology.com\u003e\n Acked-by: Eric Anholt \u003ceric@anholt.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190722104312.16184-10-noralf@tronnes.org\n\ncommit 174102f4de230a1bf85e6ef2f8c83e50b3ba22c9\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Mon Jul 22 12:43:11 2019 +0200\n\n drm/tinydrm: Move mipi-dbi\n \n This moves mipi-dbi to be a core helper with the name drm_mipi_dbi.\n \n Fixup include\u0027s in drivers.\n Move the docs entry and delete tinydrm.rst.\n Delete the last tinydrm todo entry.\n \n v2: Make DRM_MIPI_DBI tristate to enable it being built as a module.\n \n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: David Lechner \u003cdavid@lechnology.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Acked-by: David Lechner \u003cdavid@lechnology.com\u003e\n Acked-by: Eric Anholt \u003ceric@anholt.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190722104312.16184-9-noralf@tronnes.org\n\ncommit 2d2bb61a6ce121902f0f621f88def07b4e59da0a\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Mon Jul 22 12:43:10 2019 +0200\n\n drm/tinydrm/mipi-dbi: Select DRM_KMS_HELPER\n \n mipi-dbi uses several KMS helper functions but that build dependency is\n not expressed. Select DRM_KMS_HELPER to fix that.\n \n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Reviewed-by: David Lechner \u003cdavid@lechnology.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190722104312.16184-8-noralf@tronnes.org\n\ncommit 3389669ac5ea598562673c04971d7bb0fab0e9f1\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Mon Jul 22 12:43:09 2019 +0200\n\n drm/tinydrm/Kconfig: drivers: Select BACKLIGHT_CLASS_DEVICE\n \n The mipi_dbi helper is missing a dependency on DRM_KMS_HELPER and putting\n that in revealed this problem:\n \n drivers/video/fbdev/Kconfig:12:error: recursive dependency detected!\n drivers/video/fbdev/Kconfig:12: symbol FB is selected by DRM_KMS_FB_HELPER\n drivers/gpu/drm/Kconfig:75: symbol DRM_KMS_FB_HELPER depends on DRM_KMS_HELPER\n drivers/gpu/drm/Kconfig:69: symbol DRM_KMS_HELPER is selected by TINYDRM_MIPI_DBI\n drivers/gpu/drm/tinydrm/Kconfig:11: symbol TINYDRM_MIPI_DBI is selected by TINYDRM_HX8357D\n drivers/gpu/drm/tinydrm/Kconfig:15: symbol TINYDRM_HX8357D depends on BACKLIGHT_CLASS_DEVICE\n drivers/video/backlight/Kconfig:144: symbol BACKLIGHT_CLASS_DEVICE is selected by FB_BACKLIGHT\n drivers/video/fbdev/Kconfig:187: symbol FB_BACKLIGHT depends on FB\n \n A symbol that selects DRM_KMS_HELPER can not depend on\n BACKLIGHT_CLASS_DEVICE. The reason for this is that DRM_KMS_FB_HELPER\n selects FB instead of depending on it.\n \n The tinydrm drivers have somehow gotten away with depending on\n BACKLIGHT_CLASS_DEVICE because DRM_TINYDRM selects DRM_KMS_HELPER and the\n drivers depend on that symbol.\n \n An audit shows that all DRM drivers that select DRM_KMS_HELPER and use\n BACKLIGHT_CLASS_DEVICE, selects it:\n DRM_TILCDC, DRM_GMA500, DRM_SHMOBILE, DRM_NOUVEAU, DRM_FSL_DCU,\n DRM_I915, DRM_RADEON, DRM_AMDGPU, DRM_PARADE_PS8622\n \n Documentation/kbuild/kconfig-language.txt has a note regarding select:\n 1. \u0027select should be used with care since it doesn\u0027t visit dependencies.\u0027\n This is not a problem since BACKLIGHT_CLASS_DEVICE doesn\u0027t have any\n dependencies.\n 2. \u0027In general use select only for non-visible symbols\u0027\n BACKLIGHT_CLASS_DEVICE is user visible.\n \n The real solution to this would be to have DRM_KMS_FB_HELPER depend on the\n user visible symbol FB. That is a can of worms I\u0027m not willing to tackle.\n I fear that such a change will result in me handling difficult fallouts\n for the next weeks. So I\u0027m following DRM suite here.\n \n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Reviewed-by: David Lechner \u003cdavid@lechnology.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190722104312.16184-7-noralf@tronnes.org\n\ncommit 7c9f1312cfca15ce1f1ecf68133139fb096a8b4d\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Mon Jul 22 12:43:08 2019 +0200\n\n drm/tinydrm/mipi-dbi: Remove CMA helper dependency\n \n mipi-dbi depends on the CMA helper through its use of\n drm_fb_cma_get_gem_obj(). This is an unnecessary dependency to drag in for\n drivers that only want to use the MIPI DBI interface part.\n Avoid this by open coding the function.\n \n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Reviewed-by: David Lechner \u003cdavid@lechnology.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190722104312.16184-6-noralf@tronnes.org\n\ncommit 84137b866e834ac937582b04ae9bed0a72356a6a\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Mon Jul 22 12:43:07 2019 +0200\n\n drm/tinydrm: Split struct mipi_dbi in two\n \n Split struct mipi_dbi into an interface part and a display pipeline part.\n The interface part can be used by drivers that need to initialize the\n controller, but that won\u0027t upload the framebuffer over this interface.\n \n MIPI DBI supports 3 interface types:\n - A. Motorola 6800 type parallel bus\n - B. Intel 8080 type parallel bus\n - C. SPI type with 3 options:\n \n I\u0027ve embedded the SPI type specifics in the mipi_dbi struct to avoid\n adding unnecessary complexity. If more interface types will be supported\n in the future, the type specifics might have to be split out.\n \n Rename functions to match the new struct mipi_dbi_dev:\n - drm_to_mipi_dbi() -\u003e drm_to_mipi_dbi_dev().\n - mipi_dbi_init*() -\u003e mipi_dbi_dev_init*().\n \n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: David Lechner \u003cdavid@lechnology.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Acked-by: David Lechner \u003cdavid@lechnology.com\u003e\n Acked-by: Eric Anholt \u003ceric@anholt.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190722104312.16184-5-noralf@tronnes.org\n\ncommit 440961d20959e8528a86e054b0a762d9ca9c7e91\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Mon Jul 22 12:43:06 2019 +0200\n\n drm/tinydrm: Rename remaining variable mipi -\u003e dbidev\n \n struct mipi_dbi is going to be split into an interface part and a display\n pipeline part. The interface part can be used by drivers that need to\n initialize the controller, but that won\u0027t upload the framebuffer over\n this interface.\n \n tinydrm uses the variable name \u0027mipi\u0027 but this is not a good name since\n MIPI refers to a lot of standards. This patch changes the variable name\n to \u0027dbidev\u0027 where it refers to the pipeline part of struct mipi_dbi.\n \n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: David Lechner \u003cdavid@lechnology.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Acked-by: David Lechner \u003cdavid@lechnology.com\u003e\n Acked-by: Eric Anholt \u003ceric@anholt.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190722104312.16184-4-noralf@tronnes.org\n\ncommit 36b5057216236ae00ad8ab2f5232e02bbad6a00e\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Mon Jul 22 12:43:05 2019 +0200\n\n drm/tinydrm: Rename variable mipi -\u003e dbi\n \n struct mipi_dbi is going to be split into an interface part and a display\n pipeline part. The interface part can be used by drivers that need to\n initialize the controller, but that won\u0027t upload the framebuffer over\n this interface.\n \n tinydrm uses the variable name \u0027mipi\u0027 but this is not a good name since\n MIPI refers to a lot of standards. This patch changes the variable name\n to \u0027dbi\u0027 where it refers to the interface part of struct mipi_dbi.\n \n Functions that use both future parts will have both variables temporarily\n pointing to the same structure.\n \n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: David Lechner \u003cdavid@lechnology.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Acked-by: David Lechner \u003cdavid@lechnology.com\u003e\n Acked-by: Eric Anholt \u003ceric@anholt.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190722104312.16184-3-noralf@tronnes.org\n\ncommit 771ea160e68f69723c074a13110459327896d35e\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Mon Jul 22 12:43:04 2019 +0200\n\n drm/tinydrm/mipi-dbi: Move cmdlock mutex init\n \n cmdlock protects command execution so put it in mipi_dbi_spi_init()\n where it conceptually belongs.\n This is prep work for the splitting of struct mipi_dbi.\n \n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Eric Anholt \u003ceric@anholt.net\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190722104312.16184-2-noralf@tronnes.org\n\ncommit 3b51c44bd6936e86a7180abd9aebc4387a479253\nAuthor: Atif Niyaz \u003catifniyaz@google.com\u003e\nDate: Wed Jul 24 22:26:31 2019 +0300\n\n Input: allow drivers specify timestamp for input events\n \n Currently, evdev stamps events with timestamps acquired in evdev_events()\n However, this timestamping may not be accurate in terms of measuring\n when the actual event happened.\n \n Let\u0027s allow individual drivers specify timestamp in order to provide a more\n accurate sense of time for the event. It is expected that drivers will set the\n timestamp in their hard interrupt routine.\n \n Signed-off-by: Atif Niyaz \u003catifniyaz@google.com\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit db4ad308a50620c8a24419dcc6556fabaf594c0e\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 20:46:49 2019 +0800\n\n misc: alcor_pci: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190723124649.24728-1-hslester96@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 2495aeeca24df3dac97ebd9722862b0d96aee852\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:16 2019 +0200\n\n misc: eeprom: max6875: convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Link: https://lore.kernel.org/r/20190722172616.3982-3-wsa+renesas@sang-engineering.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 80257121f435a4b8cb9b1e976ebeab186c2357eb\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:15 2019 +0200\n\n misc: eeprom: ee1004: convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Link: https://lore.kernel.org/r/20190722172616.3982-2-wsa+renesas@sang-engineering.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 0f073df7e731b7b6c30c473dfdd28e0fd3cd489d\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Thu Jul 25 08:53:25 2019 +0530\n\n staging: rtl8723bs: hal: Remove unneeded variable pU1Tmp\n \n Both pu8 and pU1Tmp are of same data type u8. So replace pU1Tmp with u8.\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190725032325.GA16473@hari-Inspiron-1545\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 4da947b27cdb71b1e5718d202e0028c0aeb9bdfe\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Jul 25 10:56:45 2019 +0530\n\n staging: rtl8723bs: Remove unused function prototype ODM_SetAntenna()\n \n Remove unused function prototype ODM_SetAntenna.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190725052645.2372-9-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f79670abd541d6abeda238ce91ac5859aa309515\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Jul 25 10:56:44 2019 +0530\n\n staging: rtl8723bs: Replace function ODM_TXPowerTrackingCheck()\n \n Remove function ODM_TXPowerTrackingCheck as all it does is call\n odm_TXPowerTrackingCheckCE.\n Rename odm_TXPowerTrackingCheckCE to ODM_TXPowerTrackingCheck to\n maintain compatibility with call site.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190725052645.2372-8-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 4e8cc1c2df86f4245a37708bc007f8d530e235a1\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Jul 25 10:56:43 2019 +0530\n\n staging: rtl8723bs: Replace function odm_TXPowerTrackingInit()\n \n Remove function odm_TXPowerTrackingInit as all it does is call\n odm_TXPowerTrackingThermalMeterInit.\n Rename odm_TXPowerTrackingThermalMeterInit to\n odm_TXPowerTrackingInit to maintain compatibility with call sites.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190725052645.2372-7-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d40e29d7e219cd214eef295149fd2678871ac982\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Jul 25 10:56:42 2019 +0530\n\n staging: rtl8723bs: Replace function beacon_timing_control()\n \n Remove function beacon_timing_control as it does nothing except call\n rtw_hal_bcn_related_reg_setting.\n Rename rtw_hal_bcn_related_reg_setting to beacon_timing_control to\n maintain compatibility with call site.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190725052645.2372-6-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7a9a7332b62b802ca1a71f3e1351767129f3b53e\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Jul 25 10:56:41 2019 +0530\n\n staging: rtl8723bs: hal: Replace function rtl8723b_set_ap_wowlan_cmd()\n \n Remove function rtl8723b_set_ap_wowlan_cmd as all it does is call\n rtl8723b_set_AP_FwWoWlanRelated_cmd.\n Rename rtl8723b_set_AP_FwWoWlanRelated_cmd to\n rtl8723b_set_ap_wowlan_cmd to maintain compatibility with call site.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190725052645.2372-5-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 0528096df3778c51a99ae14ee36e1c13f6fd76e8\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Jul 25 10:56:40 2019 +0530\n\n staging: rtl8723bs: hal: Replace function rtl8723b_set_wowlan_cmd()\n \n Remove function rtl8723b_set_wowlan_cmd as all it does is call\n rtl8723b_set_FwWoWlanRelated_cmd.\n Rename rtl8723b_set_FwWoWlanRelated_cmd to rtl8723b_set_wowlan_cmd to\n maintain compatibility with call site.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190725052645.2372-4-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 9dd2f1d9a1580717605a8f2ce882105710966b7a\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Jul 25 10:56:39 2019 +0530\n\n staging: rtl8723bs: hal: Replace function PHY_GetTxPowerIndex()\n \n Remove function PHY_GetTxPowerIndex as it does nothing except call\n PHY_GetTxPowerIndex_8723B.\n Rename PHY_GetTxPowerIndex_8723B to PHY_GetTxPowerIndex to maintain\n compatibility with call site.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190725052645.2372-3-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit fe2ffeb5fdd297ad90f6252defbd635ca3037367\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Jul 25 10:56:38 2019 +0530\n\n staging: rtl8723bs: hal: Replace function PHY_SetTxPowerIndex()\n \n Remove function PHY_SetTxPowerIndex as it does nothing except call\n PHY_SetTxPowerIndex_8723B.\n Rename PHY_SetTxPowerIndex_8723B to PHY_SetTxPowerIndex to maintain\n compatibility with call site.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190725052645.2372-2-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 2e20a5ac67f8a589ad2ba56717b8419298064354\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Jul 25 10:56:37 2019 +0530\n\n staging: rtl8723bs: rtw_pwrctrl: Change true/false checks\n \n Change comparisons to true to the variable itself.\n Change comparisons to false to the negation of the variable.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190725052645.2372-1-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 4b2c85cacde3f92c2214af60ff89dd6bd93289ca\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Tue Jul 23 15:50:22 2019 +0100\n\n drivers/staging/rtl8192u: fix indentation issue, remove extra tab\n \n A statement is indented one level too deeply; clean this up by\n removing a tab.\n \n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190723145022.11608-1-colin.king@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7f9b583358a4fe31cdbde61e5f50d568be4cc69e\nAuthor: Sergio Paracuellos \u003csergio.paracuellos@gmail.com\u003e\nDate: Sat Jul 20 09:29:08 2019 +0200\n\n staging: mt7621-pci: avoid use \u0027err\u0027 local variable\n \n Function \u0027mt7621_pcie_request_resources\u0027 is using \u0027err\u0027\n local variable to get value returned from \u0027devm_request_pci_bus_resources\u0027\n and returning after that. This is not needed at all. Instead of that\n just directly return returned value by the function which is also\n returning zero on success.\n \n Signed-off-by: Sergio Paracuellos \u003csergio.paracuellos@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190720072908.16795-1-sergio.paracuellos@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 756e4624ddcfdc2792fe389262ba7447f22d8913\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Jul 17 22:35:51 2019 +0800\n\n staging: rtl8192e: remove set but not used variable \u0027payload \u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/staging/rtl8192e/rtllib_rx.c: In function rtllib_rx_InfraAdhoc:\n drivers/staging/rtl8192e/rtllib_rx.c:1303:6: warning:\n variable payload set but not used [-Wunused-but-set-variable]\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190717143551.29200-1-yuehaibing@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 3774fb6bd2b99858f37d9559bf0e3f9a9649f3f3\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Tue Jul 23 15:59:05 2019 +0100\n\n staging: rtl8723bs: hal: remove redundant assignment to variable n\n \n The variable n is being assigned a value that is never read, the\n assignment is redundant and can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190723145905.13514-1-colin.king@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 8638eeed74817358280d8d393503d50f60788527\nAuthor: Michael Straube \u003cstraube.linux@gmail.com\u003e\nDate: Sun Jul 21 19:57:35 2019 +0200\n\n staging: rtl8723bs: remove unused file hal_phy.c\n \n Remove the unused file hal_phy.c. No function from this file is used\n in the driver code and it is not listed in the Makefile.\n \n Signed-off-by: Michael Straube \u003cstraube.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190721175735.24173-1-straube.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 1c4bfacf1d70c3daba4733eb1a6450d27422e087\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Wed Jul 17 23:36:35 2019 +0530\n\n staging: rtl8723bs: core: Remove Unneeded variable ret\n \n Remove Unneeded variable ret . Return _FAIL .\n \n We cannot change return type of on_action_spct as its callback function.\n \n Issue identified with coccicheck.\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190717180635.GA11412@hari-Inspiron-1545\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit fee0cc591ae4e887c1b4304b7d69a7c209b2f24a\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Wed Jul 17 23:26:43 2019 +0530\n\n staging: rtl8723bs: os_dep: change return type of rtw_suspend_ap_wow\n \n Change return type of rtw_suspend_ap_wow as its always return SUCCCESS.\n \n Issue found with coccicheck\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190717175642.GA10582@hari-Inspiron-1545\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 8143be90aaab5848b570aa6b706727f8ffe68960\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Jul 17 22:20:14 2019 +0800\n\n staging: rtl8723bs: remove set but not used variable \u0027cck_highpwr\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/staging/rtl8723bs/hal/odm_HWConfig.c:\n In function odm_RxPhyStatus92CSeries_Parsing:\n drivers/staging/rtl8723bs/hal/odm_HWConfig.c:92:5: warning:\n variable cck_highpwr set but not used [-Wunused-but-set-variable]\n \n It is never used and can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190717142014.43216-1-yuehaibing@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit caecc46bf4e78bada4e3f6df3156c711c5ef6b78\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Jul 22 14:53:41 2019 +0530\n\n staging: rtl8712: r8712_free_recvframe(): Change return type\n \n Change return type of r8712_free_recvframe from int to void as it always\n returns _SUCCESS and its return value is never used.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190722092341.21030-8-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 606c5cf07b7907294e74eb5344fcf489b93983a5\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Jul 22 14:53:40 2019 +0530\n\n staging: rtl8712: r8712_init_recvbuf(): Change return type\n \n Change return type of r8712_init_recvbuf from int to void as it always\n returned _SUCCESS and this return value was never used at any call site.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190722092341.21030-7-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 34ef8dbe02f403b03999ccefd489ca3a7a5e08f1\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Jul 22 14:53:39 2019 +0530\n\n staging: rtl8712: r8712_set_802_11_add_wep(): Change return values\n \n Change call sites of r8712_set_802_11_add_wep to check for 0 and\n non-zero values (e.g -EINVAL, -ENOMEM) instead of for _SUCCESS and\n _FAIL.\n Change return values of r8712_set_802_11_add_wep from true and false and\n _SUCCESS to -EINVAL and the return value of r8712_set_key.\n Change return type from u8 to int to accommodate these new return\n values.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190722092341.21030-6-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 42ceb6731d781e6a9134ee73bbe85988395e0895\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Jul 22 14:53:38 2019 +0530\n\n staging: rtl8712: init_mp_priv(): Change return values\n \n Change return values of init_mp_priv from _SUCCESS/_FAIL to 0/-ENOMEM\n respectively.\n Remove label that returns the required value and return the values\n directly as they are set.\n Consequently remove now-unnecessary return variable.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190722092341.21030-5-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7c93fdf090dfed579bbb40aae0cef0853286848c\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Jul 22 14:53:37 2019 +0530\n\n staging: rtl8712: r8712_register_cmd_alive(): Change return values\n \n Change return values of r8712_register_cmd_alive from _SUCCESS and _FAIL\n to 0 and -EINVAL.\n Change the return type of this function from sint to int.\n Modify its call site to check for 0 instead of _SUCCESS.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190722092341.21030-4-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 8a73a8c4657013e02bd9e6db975e6c3584c158f8\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Jul 22 14:53:36 2019 +0530\n\n staging: rtl8712: r8712_set_key(): Change return values\n \n Change return values of r8712_set_key from _SUCCESS and _FAIL to 0 and\n -ENOMEM or -EINVAL, as the case may require.\n Modify return statements and return variable accordingly.\n Change return type from sint to int.\n As there is only one site where the return value is used, update that\n call site according to the change in the return values.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190722092341.21030-3-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 33972d48c2de0a77a5f482bb0728e62cb8420ca3\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Jul 22 14:53:35 2019 +0530\n\n staging: rtl8712: r8712_set_auth(): Change return values\n \n Change return values of r8712_set_auth from _SUCCESS/_FAIL to 0/-ENOMEM\n respectively. Modify call site accordingly. Also change return type of\n the function from sint to int.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190722092341.21030-2-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 3e67274a41b29dc93915afe857fe27eb1a851d61\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Jul 22 14:53:34 2019 +0530\n\n staging: rtl8712: r8712_init_mlme_priv(): Change return values\n \n Change return values of r8712_init_mlme_priv from _SUCCESS/_FAIL to\n 0/-ENOMEM respectively. Modify call site accordingly.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190722092341.21030-1-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit ec2b26f712610a85aab933c9cb4dfa3250aa4f58\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Fri Jul 19 15:09:22 2019 +0800\n\n staging: rtl8712: Merge memcpy + be16_to_cpus to get_unaligned_be16\n \n Merge the combo of memcpy and be16_to_cpus.\n Use get_unaligned_be16 instead.\n This simplifies the code.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190719070921.27749-1-hslester96@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 814b75b4ed9c1a7633bfec7bab33c96db87a39db\nAuthor: Christian L Moreno \u003cchristianluciano.m@gmail.com\u003e\nDate: Tue Jul 16 16:09:15 2019 -0300\n\n staging: rtl8712: recv_linux.c: Align * on block comment\n \n Block comments should align the * on each line.\n This warning was reported by checkpatch.pl\n \n Signed-off-by: Christian L Moreno \u003cchristianluciano.m@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190716190915.30869-1-christianluciano.m@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7a047abbb6ef8ac0050f8f0aa6a668985fb75a2c\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Thu Jul 25 10:52:03 2019 +0800\n\n staging: wilc1000: Merge memcpy + le32_to_cpus to get_unaligned_le32\n \n Merge the combo use of memcpy and le32_to_cpus.\n Use get_unaligned_le32 instead.\n This simplifies the code.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190725025203.8772-1-hslester96@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 28771088953f548a637c01cd59b62f9ee3018824\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Jul 25 10:53:59 2019 +0530\n\n staging: comedi: daqboard2000: Remove function db2k_initialize_dac()\n \n Remove function db2k_initialize_dac as all it does is call\n db2k_dac_disarm.\n Modify call site accordingly.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190725052359.2308-1-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 3e6998b17f36ab4f8f5608b3599df7b3c17d97a5\nAuthor: Lecopzer Chen \u003clecopzer.chen@mediatek.com\u003e\nDate: Fri Jul 12 16:47:17 2019 +0800\n\n staging: android: ion: Remove unused rbtree for ion_buffer\n \n ion_buffer_add() insert ion_buffer into rbtree every time creating\n an ion_buffer but never use it after ION reworking.\n Also, buffer_lock protects only rbtree operation, remove it together.\n \n Signed-off-by: Lecopzer Chen \u003clecopzer.chen@mediatek.com\u003e\n Cc: YJ Chiang \u003cyj.chiang@mediatek.com\u003e\n Cc: Lecopzer Chen \u003clecopzer.chen@mediatek.com\u003e\n Acked-by: Laura Abbott \u003clabbott@redhat.com\u003e\n Link: https://lore.kernel.org/r/20190712084717.12441-1-lecopzer.chen@mediatek.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 8d5682f66252c421b96f720fda7f94431e444593\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Wed Jul 24 17:18:13 2019 -0700\n\n drm/i915/uc: Unify uC firmware upload\n \n The way we load the firmwares is the same for both GuC and HuC, the only\n difference is in the wopcm destination address and the dma flags, so we\n easily can move the logic to a common function and pass in offset and\n flags. The only other difference in the uplaod path are some the extra\n steps that guc does before and after the xfer, but those don\u0027t require\n the guc fw to be pinned in ggtt and can safely be performed before\n calling the uc_upload function.\n \n Note that this patch re-introduces the dma xfer wait for guc loading that\n was removed with \"drm/i915/guc: Propagate the fw xfer timeout\". This is\n not going to slow us down on a successful load (the dma has to complete\n before fw init can start), but could slightly increase the timeout in case\n of a fw init error.\n \n v2: use _fw variants for uncore accesses (Chris), fix guc_fw status on\n failed wait.\n \n v3: use dev_err and print DMA_CTRL (Chris)\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725001813.4740-9-daniele.ceraolospurio@intel.com\n\ncommit 4ca8d2ef8dab4ad61347e52174b870c44f1508a3\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Wed Jul 24 17:18:12 2019 -0700\n\n drm/i915/uc: Plumb the gt through fw_upload\n \n The gt is our new central structure for uc-related code, so we can use\n that instead of jumping back to i915 via the fw object. Since we have it\n in the upload function it is easy to pass it through the lower levels of\n the xfer process instead of continuosly jumping via uc_fw-\u003euc-\u003egt, which\n will also make things a bit cleaner for the next patch.\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725001813.4740-8-daniele.ceraolospurio@intel.com\n\ncommit 9cb27945570567b75af7a8751803d41d0a6ce1a7\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Wed Jul 24 17:18:11 2019 -0700\n\n drm/i915/huc: Copy huc rsa only once\n \n The binary is perma-pinned and the rsa is not going to change, so copy\n it only once and not on every load.\n \n v2: onion unwind (Chris)\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Fernando Pacheco \u003cfernando.pacheco@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e #v1\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725001813.4740-7-daniele.ceraolospurio@intel.com\n\ncommit 90dd992260cc44cec8f6e41ab5e64938921912c0\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Wed Jul 24 17:18:10 2019 -0700\n\n drm/i915/uc: Move xfer rsa logic to common function\n \n The way we copy the RSA is the same for GuC and HuC, so we can move the\n logic in a common function. this will also make any update needed for\n local memory easier.\n \n v2: return the number of copied bytes and check it (Chris)\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e #v1\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725001813.4740-6-daniele.ceraolospurio@intel.com\n\ncommit 91e55e545ab8bbedd3a518dbe6124c10388ff92f\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Wed Jul 24 17:18:09 2019 -0700\n\n drm/i915/uc: Unify uc_fw status tracking\n \n We currently track fetch and load status separately, but the 2 are\n actually sequential in the uc lifetime (fetch must complete before we\n can attempt the load!). Unifying the 2 variables we can better follow\n the sequential states and improve our trackng of the uC state.\n \n Also, sprinkle some GEM_BUG_ON to make sure we transition correctly\n between states.\n \n v2: rename states, add the running state (Michal), drop some logs in\n the fetch path (Michal, Chris)\n \n v3: re-rename states, extend early status check to all helpers (Michal)\n \n Suggested-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725001813.4740-5-daniele.ceraolospurio@intel.com\n\ncommit 21a27d1cdd41a0df8ea5b3ecc61ed7973bd10368\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Wed Jul 24 17:18:08 2019 -0700\n\n drm/i915/uc: Unify uC FW selection\n \n Instead of having 2 identical functions for GuC and HuC firmware\n selection, we can unify the selection logic and just use different lists\n based on FW type.\n \n Note that the revid is not relevant for current blobs, but the upcoming\n CML will be identified as CFL rev 5, so by considering the revid we\u0027re\n ready for that.\n \n v2: rework blob list defs (Michal), add order check (Chris), fuse GuC\n and HuC lists into one.\n \n v3: remove difference between no uC HW and no uC FW, simplify related\n selection code, check the whole fw list (Michal)\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e #v2\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725001813.4740-4-daniele.ceraolospurio@intel.com\n\ncommit 305ceebd5284a3a6cbeb0f67f1f839cd92be1847\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Wed Jul 24 17:18:07 2019 -0700\n\n drm/i915: Fix handling of non-supported uC\n \n There are 2 issues around handling of missing uC support:\n \n - We treat lack of uC HW and lack of uC FW definition as 2 different\n cases, but both of them mean that we don\u0027t support the uC on the\n platform we\u0027re running on.\n \n - We rely on the modparam to decide if we can take uC paths or not, but\n we don\u0027t sanitize it if it is set incorrectly on platform with no uC\n support.\n \n To fix both of them, unify the 2 cases in a single one and sanitize the\n modparam on invalid configuration (after printing an error message).\n The log has been adapted as well, since the user doesn\u0027t care why we\n don\u0027t support GuC/HuC (no HW or no FW), just that we do not. Developers\n can easily find the answer based on the platform, so we can simplify the\n log.\n \n Correcting the modparam has been preferred over failing the load since\n this is what we usually do for non-supported feature (e.g. the now gone\n enable_ppgtt would fall back to the highest supported PPGTT mode if the\n selected one was not available).\n \n Note that this patch purposely doesn\u0027t change the behavior for platforms\n that do have uC support, in which case we will still fail if enable_guc\n is set and the firmware is not available on the system.\n \n Suggested-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725001813.4740-3-daniele.ceraolospurio@intel.com\n\ncommit 702668e606c0618fd62c7a5d051c9faee256c049\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Wed Jul 24 17:18:06 2019 -0700\n\n drm/i915/uc: Unify uC platform check\n \n We have several HAS_* checks for GuC and HuC but we mostly use HAS_GUC\n and HAS_HUC, with only 1 exception. Since our HW always has either\n both uC or neither of them, just replace all the checks with a unified\n HAS_UC.\n \n v2: use HAS_GT_UC (Michal)\n v3: fix comment (Michal)\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Acked-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190725001813.4740-2-daniele.ceraolospurio@intel.com\n\ncommit 66da768644b5ee13d2b62d96e9514c1bb6f82398\nAuthor: Mike Rapoport \u003crppt@linux.ibm.com\u003e\nDate: Wed Jul 24 17:10:32 2019 +0300\n\n xtensa: remove free_initrd_mem\n \n The xtensa free_initrd_mem() verifies that initrd is mapped and then\n frees its memory using free_reserved_area().\n \n The initrd is considered mapped when its memory was successfully reserved\n with mem_reserve().\n \n Resetting initrd_start to 0 in case of mem_reserve() failure allows to\n switch to generic free_initrd_mem() implementation.\n \n Signed-off-by: Mike Rapoport \u003crppt@linux.ibm.com\u003e\n Message-Id: \u003c1563977432-8376-1-git-send-email-rppt@linux.ibm.com\u003e\n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit ff09297ec9964b3fe4bade77c92c75fde34fa8e9\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Wed Jul 24 23:01:55 2019 -0400\n\n autofs: simplify get_next_positive_...(), get rid of trylocks\n \n * new helper: positive_after(parent, child); parent-\u003ed_lock is\n held by caller, grabs and returns the first thing after child\n in the list of children that has simple_positive() true. NULL\n if nothing\u0027s found; NULL child \u003d\u003d search the entire list.\n \n * get_next_positive_subdir() loses the redundant check for\n d_count and switches to use of that helper. BTW, dput(NULL) is\n a no-op for a good reason...\n \n * get_next_positive_dentry() switched to the same helper. Logics:\n look for positive child in prev; if not found, look for the\n positive child of prev\u0027s parent following prev, etc. That way\n we are guaranteed that we are only moving rootwards through the\n ancestors of prev, which is pinned and thus not going anywhere.\n Since -\u003ed_parent on autofs never changes, the same goes for\n the entire chain of ancestors and we don\u0027t need overlapping\n -\u003ed_lock on them. Which avoids the trylock loops, in addition\n to simplifying the logics in there...\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 1417dba1034d644953238add381e31ada3a3a199\nAuthor: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\nDate: Mon Jul 15 22:03:27 2019 -0700\n\n remoteproc: qcom: Move glink_ssr notification after stop\n \n glink_ssr is used to signal a remote processor \"A\" about the stopping of\n another remote processor \"B\", so that in the event that remote processor\n B is ever booted again the fifos of the glink channel between A and B is\n in a known state.\n \n But if remote processor A receives this notification before B is\n actually stopped the newly reset fifo indices will be interpreted as\n there being \"data\" on the channel and either side of the channel will\n enter a fatal error handler.\n \n Move the glink_ssr notification to the \"unprepare\" state of the\n rproc_subdev to avoid this issue.\n \n This has the side effect of us not notifying the dying remote processor\n itself about its fate, which has been seen to block in certain resource\n constraint scenarios.\n \n Tested-by: Sibi Sankar \u003csibis@codeaurora.org\u003e\n Reviewed-by: Sibi Sankar \u003csibis@codeaurora.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 9eed21c01c7827f029000f77d59b61d1b16cb024\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Mon Jul 22 17:01:55 2019 +0200\n\n mlx4: avoid large stack usage in mlx4_init_hca()\n \n The mlx4_dev_cap and mlx4_init_hca_param are really too large\n to be put on the kernel stack, as shown by this clang warning:\n \n drivers/net/ethernet/mellanox/mlx4/main.c:3304:12: error: stack frame size of 1088 bytes in function \u0027mlx4_load_one\u0027 [-Werror,-Wframe-larger-than\u003d]\n \n With gcc, the problem is the same, but it does not warn because\n it does not inline this function, and therefore stays just below\n the warning limit, while clang is just above it.\n \n Use kzalloc for dynamic allocation instead of putting them\n on stack. This gets the combined stack frame down to 424 bytes.\n \n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Reviewed-by: Saeed Mahameed \u003csaeedm@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7c116e02a4a7575c8c62bfd2106e3e3ec8fb99dc\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Mon Jul 22 17:01:23 2019 +0200\n\n qed: reduce maximum stack frame size\n \n clang warns about an overly large stack frame in one function\n when it decides to inline all __qed_get_vport_*() functions into\n __qed_get_vport_stats():\n \n drivers/net/ethernet/qlogic/qed/qed_l2.c:1889:13: error: stack frame size of 1128 bytes in function \u0027_qed_get_vport_stats\u0027 [-Werror,-Wframe-larger-than\u003d]\n \n Use a noinline_for_stack annotation to prevent clang from inlining\n these, which keeps the maximum stack usage at around half of that\n in the worst case, similar to what we get with gcc.\n \n Fixes: 86622ee75312 (\"qed: Move statistics to L2 code\")\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 81cd17a4121d7dc7cad28e51251f31ff12b1de2b\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Wed Jul 24 23:34:45 2019 +0200\n\n r8169: improve rtl_set_rx_mode\n \n This patch improves and simplifies rtl_set_rx_mode a little.\n No functional change intended.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d601be97126e9736db4f3528c66792178fcf1796\nAuthor: Qian Cai \u003ccai@lca.pw\u003e\nDate: Wed Jul 24 12:17:59 2019 -0400\n\n net/ixgbevf: fix a compilation error of skb_frag_t\n \n The linux-next commit \"net: Rename skb_frag_t size to bv_len\" [1]\n introduced a compilation error on powerpc as it forgot to deal with the\n renaming from \"size\" to \"bv_len\" for ixgbevf.\n \n [1] https://lore.kernel.org/netdev/20190723030831.11879-1-willy@infradead.org/T/#md052f1c7de965ccd1bdcb6f92e1990a52298eac5\n \n In file included from ./include/linux/cache.h:5,\n from ./include/linux/printk.h:9,\n from ./include/linux/kernel.h:15,\n from ./include/linux/list.h:9,\n from ./include/linux/module.h:9,\n from\n drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c:12:\n drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c: In function\n \u0027ixgbevf_xmit_frame_ring\u0027:\n drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c:4138:51: error:\n \u0027skb_frag_t\u0027 {aka \u0027struct bio_vec\u0027} has no member named \u0027size\u0027\n count +\u003d TXD_USE_COUNT(skb_shinfo(skb)-\u003efrags[f].size);\n ^\n ./include/uapi/linux/kernel.h:13:40: note: in definition of macro\n \u0027__KERNEL_DIV_ROUND_UP\u0027\n #define __KERNEL_DIV_ROUND_UP(n, d) (((n) + (d) - 1) / (d))\n ^\n drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c:4138:12: note: in\n expansion of macro \u0027TXD_USE_COUNT\u0027\n count +\u003d TXD_USE_COUNT(skb_shinfo(skb)-\u003efrags[f].size);\n \n Signed-off-by: Qian Cai \u003ccai@lca.pw\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit e62088ea6a20c3517c490d94bcc437ab4c24c9d1\nAuthor: Masanari Iida \u003cstandby24x7@gmail.com\u003e\nDate: Thu Jul 25 00:29:51 2019 +0900\n\n selftests: mlxsw: Fix typo in qos_mc_aware.sh\n \n This patch fix some spelling typo in qos_mc_aware.sh\n \n Signed-off-by: Masanari Iida \u003cstandby24x7@gmail.com\u003e\n Acked-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\n Reviewed-by: Ido Schimmel \u003cidosch@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 2949dc443116a66fd1a92d9ef107be16cdd197cd\nAuthor: Moritz Fischer \u003cmdf@kernel.org\u003e\nDate: Wed Feb 21 09:33:06 2018 -0800\n\n dt-bindings: fpga: Consolidate bridge properties\n \n Consolidate bridge properties in a single file, instead of duplicating\n the same optional property over and over again.\n \n Acked-by: Alan Tull \u003catull@kernel.org\u003e\n Acked-by: Michal Simek \u003cmichal.simek@xilinx.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Moritz Fischer \u003cmdf@kernel.org\u003e\n\ncommit 2ec0a8909189cb801ded3428d66cdc1ea7e4578b\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Wed Jul 24 21:01:26 2019 +0800\n\n qlge: Fix build error without CONFIG_ETHERNET\n \n Now if CONFIG_ETHERNET is not set, QLGE driver\n building fails:\n \n drivers/staging/qlge/qlge_main.o: In function `qlge_remove\u0027:\n drivers/staging/qlge/qlge_main.c:4831: undefined reference to `unregister_netdev\u0027\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Fixes: 955315b0dc8c (\"qlge: Move drivers/net/ethernet/qlogic/qlge/ to drivers/staging/qlge/\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 71d8e94dabee7fceac473d87445a03e848469a71\nAuthor: Moritz Fischer \u003cmdf@kernel.org\u003e\nDate: Wed Jun 26 17:33:09 2019 -0700\n\n fpga: altera-pr-ip: Make alt_pr_unregister function void\n \n Make alt_pr_unregister function void, since it always returns 0,\n and nothing would act on the value anyways.\n \n Signed-off-by: Moritz Fischer \u003cmdf@kernel.org\u003e\n\ncommit ed72a9bb9af06c9eb06ff2e8640faac44f8da94a\nAuthor: Corentin Musard \u003ccorentinmusard@gmail.com\u003e\nDate: Wed Jul 24 14:34:43 2019 +0200\n\n r8169: fix a typo in a comment\n \n Replace \"additonal\" by \"additional\" in a comment.\n Typo found by checkpatch.pl.\n \n Signed-off-by: Corentin Musard \u003ccorentinmusard@gmail.com\u003e\n Reviewed-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 998c1de56dac7ff6bad4f810259cc25c7d6d2843\nAuthor: Carlos A Petry \u003ccapetry.dev@gmail.com\u003e\nDate: Wed Nov 7 23:13:12 2018 -0200\n\n fpga: altera-cvp: Fix function definition argument\n \n Fix the following checkpatch warning:\n WARNING: function definition argument \u0027struct altera_cvp_conf *\u0027\n \n Signed-off-by: Carlos A Petry \u003ccapetry.dev@gmail.com\u003e\n Signed-off-by: Moritz Fischer \u003cmdf@kernel.org\u003e\n\ncommit e5e9a2ecfe780975820e157b922edee715710b66\nAuthor: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\nDate: Mon Jul 8 12:55:45 2019 +0800\n\n e1000e: add workaround for possible stalled packet\n \n This works around a possible stalled packet issue, which may occur due to\n clock recovery from the PCH being too slow, when the LAN is transitioning\n from K1 at 1G link speed.\n \n Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id\u003d204057\n \n Signed-off-by: Kai-Heng Feng \u003ckai.heng.feng@canonical.com\u003e\n Tested-by: Aaron Brown \u003caaron.f.brown@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 6d37a38243bc4bb3336b9771961e2904500fb3a9\nAuthor: Sasha Neftin \u003csasha.neftin@intel.com\u003e\nDate: Tue Jul 2 14:39:55 2019 +0300\n\n igc: Add more SKUs for i225 device\n \n Add support for more SKUs.\n \n Signed-off-by: Sasha Neftin \u003csasha.neftin@intel.com\u003e\n Tested-by: Aaron Brown \u003caaron.f.brown@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit bb4265ec24c13aacf3a613e64c701294b122beb7\nAuthor: Sasha Neftin \u003csasha.neftin@intel.com\u003e\nDate: Wed Jun 26 08:53:47 2019 +0300\n\n igc: Update the MAC reset flow\n \n Use Device Reset flow instead of Port Reset flow.\n This flow performs a reset of the entire controller device,\n resulting in a state nearly approximating the state\n following a power-up reset or internal PCIe reset,\n except for system PCI configuration.\n \n Signed-off-by: Sasha Neftin \u003csasha.neftin@intel.com\u003e\n Tested-by: Aaron Brown \u003caaron.f.brown@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 6145787d5e2eb973fa9f73d228d7a6a13650b1d1\nAuthor: Sasha Neftin \u003csasha.neftin@intel.com\u003e\nDate: Sun Jun 23 13:42:32 2019 +0300\n\n igc: Remove the unused field from a device specification structure\n \n This patch comes to clean up the device specification structure.\n \n Signed-off-by: Sasha Neftin \u003csasha.neftin@intel.com\u003e\n Tested-by: Aaron Brown \u003caaron.f.brown@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 2b69286dbd999516b67147bd48360808d0abdfc1\nAuthor: Sasha Neftin \u003csasha.neftin@intel.com\u003e\nDate: Sun Jun 23 13:42:31 2019 +0300\n\n igc: Remove the polarity field from a PHY information structure\n \n Polarity and cable length fields is not applicable for the i225 device.\n This patch comes to clean up PHY information structure.\n \n Signed-off-by: Sasha Neftin \u003csasha.neftin@intel.com\u003e\n Tested-by: Aaron Brown \u003caaron.f.brown@intel.com\u003e\n Signed-off-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\n\ncommit 939b482a6411bd5a798662389efb9b8eb814ee5a\nAuthor: Stefan Wahren \u003cwahrenst@gmx.net\u003e\nDate: Sun May 19 12:20:00 2019 +0200\n\n ARM: bcm283x: Reduce register ranges for UART, SPI and I2C\n \n The assigned register ranges for UART, SPI and I2C were too wasteful.\n In order to avoid overlapping with the new functions on BCM2711\n reduce the ranges.\n \n Signed-off-by: Stefan Wahren \u003cwahrenst@gmx.net\u003e\n Acked-by: Eric Anholt \u003ceric@anholt.net\u003e\n\ncommit a511f82218fb14b3d6fbbce3c35759aa4920a734\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri May 31 10:21:00 2019 +0100\n\n RDMA/hns: Fix comparison of unsigned long variable \u0027end\u0027 with less than zero\n \n Currently the comparison of end with less than zero is always false\n because end is an unsigned long. Also, replace checks of end with\n non-zero with end \u003e 0 as it is possible that the #defined decrement may be\n changed in the future causing end to step over zero and go negative.\n \n The initialization of end with 0 is also redundant as this value is never\n read and is later set to HW_SYNC_TIMEOUT_MSECS, so fix this by\n initializing it with this value to begin with.\n \n Link: https://lore.kernel.org/r/20190531092101.28772-1-colin.king@canonical.com\n Addresses-Coverity: (\"Unsigned compared against 0\")\n Fixes: 669cefb654cb (\"RDMA/hns: Remove jiffies operation in disable interrupt context\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit ecc53f8a3c097d13f4b6793855e0bd30a63a5c3a\nAuthor: Andrey Konovalov \u003candreyknvl@google.com\u003e\nDate: Tue Jul 23 19:58:48 2019 +0200\n\n RDMA/mlx4: Untag user pointers in mlx4_get_umem_mr\n \n This patch is a part of a series that extends kernel ABI to allow to pass\n tagged user pointers (with the top byte set to something else other than\n 0x00) as syscall arguments.\n \n mlx4_get_umem_mr() uses provided user pointers for vma lookups, which can\n only by done with untagged pointers.\n \n Untag user pointers in this function.\n \n Link: https://lore.kernel.org/r/7969018013a67ddbbf784ac7afeea5a57b1e2bcb.1563904656.git.andreyknvl@google.com\n Reviewed-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n Acked-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Signed-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit befa45fb5bdd514a40a19e659491193f7fd022f2\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Mon Jul 8 20:33:54 2019 +0800\n\n PCI: Use devm_add_action_or_reset()\n \n devm_add_action_or_reset() is a helper function which internally calls\n devm_add_action(). If the devm_add_action() fails, it will execute the\n action mentioned and return the error code.\n \n Use devm_add_action_or_reset() to reduce source code size (avoid writing\n the action twice) and reduce the likelihood of bugs.\n \n Link: https://lore.kernel.org/r/20190708123354.12127-1-huangfq.daxian@gmail.com\n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit 92493a2f8a8d5a5bc1188fc71ef02df859ebd932\nAuthor: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\nDate: Wed Jul 24 04:36:15 2019 -0700\n\n Build fixes for skb_frag_size conversion\n \n I missed a few places. One is in some ifdeffed code which will probably\n never be re-enabled; the others are in drivers which can\u0027t currently be\n compiled on x86.\n \n Signed-off-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 8f75ec1a220e0a4b0f1abbc564987fe754cdc6f0\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Wed Jul 24 19:27:06 2019 +0800\n\n sfc-falcon: Use dev_get_drvdata where possible\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 3e03a8ba8e1d6c9940a032992f4da2bfba391fdf\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Wed Jul 24 19:26:58 2019 +0800\n\n sfc: Use dev_get_drvdata where possible\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Acked-by: Edward Cree \u003cecree@solarflare.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit dfa56f83153f08c73e7adfacefd93750da2266ad\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Wed Jul 24 19:26:48 2019 +0800\n\n forcedeth: Use dev_get_drvdata where possible\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7bdb9234952b08b1f86927eab4adbf6609d4444a\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Wed Jul 24 19:26:34 2019 +0800\n\n net: marvell: Use dev_get_drvdata where possible\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit bdc873feec900005d52f9afd3a63290531c0c1f4\nAuthor: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\nDate: Tue Jul 23 16:58:54 2019 +0200\n\n ASoC: Intel: Skylake: Cleanup skl_module_cfg declaration\n \n With removal of MCPS, CPS and CPC ambiguity, ibs and obs params for\n struct skl_module_cfg have been left unused. Update struct declaration\n by removing these two.\n \n Signed-off-by: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190723145854.8527-8-cezary.rojewski@intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 84b71067ea840fadee32588aa3967d0d8c4e0b9a\nAuthor: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\nDate: Tue Jul 23 16:58:53 2019 +0200\n\n ASoC: Intel: Skylake: Make MCPS and CPS params obsolete\n \n As per FW Interface Modules Configuration, init instance IPC request\n requires base initial module configuration. This configuration structure\n is made of:\n - cpc (chunks per cycle)\n - ibs (input buffer size)\n - obs (output buffer size)\n - is_pages (memory pages required)\n - audio_fmt (self explanatory)\n \n Skylake topology accepts following tokens: MCPS, CPS and CPC. All of\n these are directly connected. Moreover, assigning one of these allows\n to calculate the remaining two. In simplest scenario and assuming 1ms\n scheduling, following is true:\n \n CPS \u003d CPC times 1000\n MCPS \u003d CPS times 1000 000\n Note: these calculations vary depending on scenario and scheduling\n requirements.\n \n Given the current implementation, userspace is allowed to provide\n different values for all three causing informational chaos. On top of\n that, struct skl_base_cfg which represents base module configuration,\n incorrectly takes CPS param instead of CPC.\n \n This ambiguity may lead to user unintentionally providing improper\n values to DSP firmware and thus impacting module scheduling in\n unexpected fashion. Fix by making MCPS and CPS topology params obsolete\n and relying solely on CPC value.\n \n Signed-off-by: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190723145854.8527-7-cezary.rojewski@intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit d4e23b7c87e3439f6d0ec0c65ad7f919d6f4c124\nAuthor: Amadeusz Sławiński \u003camadeuszx.slawinski@intel.com\u003e\nDate: Tue Jul 23 16:58:52 2019 +0200\n\n ASoC: Intel: Skylake: Do not disable FW notifications\n \n As per FW team recommendation we should not disable notifications.\n By default, all notifications are enabled in DSP firmware. These\n notifications provide a vital information whenever an error occurs.\n \n Currently, driver disables them during boot sequences. By doing so,\n Skylake may silently ignore severe stream errors.\n Correct that by removing permissive code.\n \n Signed-off-by: Amadeusz Sławiński \u003camadeuszx.slawinski@intel.com\u003e\n Signed-off-by: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190723145854.8527-6-cezary.rojewski@intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit e5c29dc5221c328b6e54c5b2aba1feca4003c051\nAuthor: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\nDate: Tue Jul 23 16:58:51 2019 +0200\n\n ASoC: Intel: Skylake: Remove memory available check\n \n Current memory availability check is a stub, while actual memory\n management takes place in firmware. Leave this task to firmware entirely\n and remove redundant code.\n \n Signed-off-by: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190723145854.8527-5-cezary.rojewski@intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit f2a167ca681d21aa46f127032cc005879c328823\nAuthor: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\nDate: Tue Jul 23 16:58:50 2019 +0200\n\n ASoC: Intel: Skylake: Remove MCPS available check\n \n The entire logic for MCPS calculation and DSP scheduling is found\n within DSP firmware. Currently driver implements simplistic, inaccurate\n logic itself which may prevent pipeline creation despite firmware being\n completely fine its parameters.\n \n Remove that logic and leave the MCPS calculation to DSP alone.\n \n Signed-off-by: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190723145854.8527-4-cezary.rojewski@intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 88abcc90cb04cbabcc3e384e8d3431a49f56a2b3\nAuthor: Amadeusz Sławiński \u003camadeuszx.slawinski@intel.com\u003e\nDate: Tue Jul 23 16:58:49 2019 +0200\n\n ASoC: Intel: Skylake: Combine snd_soc_skl_ipc and snd_soc_skl\n \n As both modules are core part of Skylake driver and none can live\n without the other, combine snd_soc_skl_ipc and snd_soc_skl.\n \n It\u0027s highly probable IPC module was to be treated as an interface for\n platform specific code implementations e.g.: possibility of existence of\n BXT specific code without SKL one. However, most funtionalities are\n being inherited from one DSP firmware to another, and thus this\n assumption fails.\n \n skl-sst, bxt-sst and cnl-sst are not individuals pointing respectively\n to SKL (cAVS 1.5), BXT (cAVS 1.5+) \u0026 CNL (cAVS 1.8) standalone\n implementations. Code found within these is shared among all platforms\n whenever necessary to avoid code duplication and reduce development\n burden.\n \n Merge also helps in cleaning up internal code in future changes.\n \n Signed-off-by: Amadeusz Sławiński \u003camadeuszx.slawinski@intel.com\u003e\n Signed-off-by: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190723145854.8527-3-cezary.rojewski@intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit bcc2a2dc3ba8c3a7aed856f840afa6a47e3cb8e0\nAuthor: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\nDate: Tue Jul 23 16:58:48 2019 +0200\n\n ASoC: Intel: Skylake: Merge skl_sst and skl into skl_dev struct\n \n Skylake driver is divided into two modules:\n - snd_soc_skl\n - snd_soc_skl_ipc\n \n and nothing would be wrong if not for the fact that both cannot exist\n without one another. IPC module is not some kind of extension, as it is\n the case for snd_hda_ext_core which is separated from snd_hda_core -\n legacy hda interface. It\u0027s as much core Skylake module as snd_soc_skl\n is.\n \n Statement backed up by existence of circular dependency between this\n two. To eliminate said problem, struct skl_sst has been created. From\n that very momment, Skylake has been plagued by header errors (incomplete\n structs, unknown references etc.) whenever something new is to be added\n or code is cleaned up.\n \n As this design is being corrected, struct skl_sst is no longer needed,\n so combine it with struct skl. To avoid ambiguity when searching for skl\n stuff (struct skl *skl) it has also been renamed to skl_dev.\n \n No functional changes.\n \n Signed-off-by: Piotr Maziarz \u003cpiotrx.maziarz@intel.com\u003e\n Signed-off-by: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190723145854.8527-2-cezary.rojewski@intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 570d785ba46b947e0ded0c38955c43d1d7867c12\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Wed Jul 24 00:06:59 2019 -0600\n\n drivers: net: xgene: Remove acpi_has_method() calls\n \n acpi_evaluate_object will already return an error if the needed method\n does not exist. Remove unnecessary acpi_has_method() calls and check the\n returned acpi_status for failure instead.\n \n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a18670f4617d41829ce88ab3e47bbc406e4dc5e8\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Sun Jul 21 14:55:47 2019 +0200\n\n watchdog: ath79_wdt: fix a typo in the name of a function\n \n It is likely that \u0027ath97_wdt_shutdown()\u0027 should be \u0027ath79_wdt_shutdown()\u0027\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Reviewed-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n Signed-off-by: Wim Van Sebroeck \u003cwim@linux-watchdog.org\u003e\n\ncommit 2a3b6f7b02cd141c990c6f5aed41070b5b7f4a7d\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Wed Jul 24 20:23:31 2019 +0800\n\n spi: dw-pci: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190724122331.21856-1-hslester96@gmail.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit e8758a5ed2783c417be1f5aab5af9fe4be60956f\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Wed Jul 24 07:26:32 2019 +0200\n\n ASoC: wm8955: Fix a typo in \u0027wm8995_pll_factors()\u0027 function name\n \n This should be \u0027wm8955_pll_factors()\u0027 instead.\n Fix it and use it.\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Acked-by: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n Link: https://lore.kernel.org/r/20190724052632.30476-1-christophe.jaillet@wanadoo.fr\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 511885d7061eda3eb1faf3f57dcc936ff75863f1\nAuthor: Davidlohr Bueso \u003cdave@stgolabs.net\u003e\nDate: Wed Jul 24 08:23:23 2019 -0700\n\n lib/timerqueue: Rely on rbtree semantics for next timer\n \n Simplify the timerqueue code by using cached rbtrees and rely on the tree\n leftmost node semantics to get the timer with earliest expiration time.\n This is a drop in conversion, and therefore semantics remain untouched.\n \n The runtime overhead of cached rbtrees is be pretty much the same as the\n current head-\u003enext method, noting that when removing the leftmost node,\n a common operation for the timerqueue, the rb_next(leftmost) is O(1) as\n well, so the next timer will either be the right node or its parent.\n Therefore no extra pointer chasing. Finally, the size of the struct\n timerqueue_head remains the same.\n \n Passes several hours of rcutorture.\n \n Signed-off-by: Davidlohr Bueso \u003cdbueso@suse.de\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190724152323.bojciei3muvfxalm@linux-r8p5\n\ncommit 4599c6671b8119ed5455e4ed6b967b461c27a9f7\nAuthor: Nikolas Nyby \u003cnikolas@gnu.org\u003e\nDate: Wed Jul 24 00:13:37 2019 -0400\n\n x86/crash: Remove unnecessary comparison\n \n The ret comparison and return are unnecessary as of commit f296f2634920\n (\"x86/kexec: Remove walk_iomem_res() call with GART type\")\n \n elf_header_exclude_ranges() returns ret in any case, with or without this\n comparison.\n \n [ tglx: Use a proper commit reference instead of full SHA ]\n \n Signed-off-by: Nikolas Nyby \u003cnikolas@gnu.org\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190724041337.8346-1-nikolas@gnu.org\n\ncommit a30f1743e4f57b359903fd679476b2903e7e5dac\nAuthor: Pavel Machek \u003cpavel@denx.de\u003e\nDate: Wed Jul 24 11:52:54 2019 +0200\n\n ALSA: line6: sizeof (byte) is always 1, use that fact.\n \n sizeof (byte) is always 1, use that fact and make interesting code explicit.\n \n Signed-off-by: Pavel Machek \u003cpavel@denx.de\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 21e450d21ccad4cb7c7984c29ff145012b47736d\nAuthor: Jan Kiszka \u003cjan.kiszka@siemens.com\u003e\nDate: Tue Jun 18 09:32:11 2019 +0200\n\n x86/mm: Avoid redundant interrupt disable in load_mm_cr4()\n \n load_mm_cr4() is always called with interrupts disabled from:\n \n - switch_mm_irqs_off()\n - refresh_pce(), which is a on_each_cpu() callback\n \n Thus, disabling interrupts in cr4_set/clear_bits() is redundant.\n \n Implement cr4_set/clear_bits_irqsoff() helpers, rename load_mm_cr4() to\n load_mm_cr4_irqsoff() and use the new helpers. The new helpers do not need\n a lockdep assert as __cr4_set() has one already.\n \n The renaming in combination with the checks in __cr4_set() ensure that any\n changes in the boundary conditions at the call sites will be detected.\n \n [ tglx: Massaged change log ]\n \n Signed-off-by: Jan Kiszka \u003cjan.kiszka@siemens.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/0fbbcb64-5f26-4ffb-1bb9-4f5f48426893@siemens.com\n\ncommit 4fcf8544fc677fc8af135f1d86b3ba69c4ad429d\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Jul 2 16:43:57 2019 +0100\n\n iommu: Introduce iommu_iotlb_gather_add_page()\n \n Introduce a helper function for drivers to use when updating an\n iommu_iotlb_gather structure in response to an -\u003eunmap() call, rather\n than having to open-code the logic in every page-table implementation.\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit a7d20dc19d9ea7012227be5144353012ffa3ddc4\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Jul 2 16:43:48 2019 +0100\n\n iommu: Introduce struct iommu_iotlb_gather for batching TLB flushes\n \n To permit batching of TLB flushes across multiple calls to the IOMMU\n driver\u0027s -\u003eunmap() implementation, introduce a new structure for\n tracking the address range to be flushed and the granularity at which\n the flushing is required.\n \n This is hooked into the IOMMU API and its caller are updated to make use\n of the new structure. Subsequent patches will plumb this into the IOMMU\n drivers as well, but for now the gathering information is ignored.\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 298f78895b081911e0b3605f07d79ebd3d4cf7b0\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Jul 2 16:43:34 2019 +0100\n\n iommu/io-pgtable: Rename iommu_gather_ops to iommu_flush_ops\n \n In preparation for TLB flush gathering in the IOMMU API, rename the\n iommu_gather_ops structure in io-pgtable to iommu_flush_ops, which\n better describes its purpose and avoids the potential for confusion\n between different levels of the API.\n \n $ find linux/ -type f -name \u0027*.[ch]\u0027 | xargs sed -i \u0027s/gather_ops/flush_ops/g\u0027\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit f71da46719460acd5afa411e52dc8cdf1cb9b0ce\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Jul 2 16:43:24 2019 +0100\n\n iommu/io-pgtable-arm: Remove redundant call to io_pgtable_tlb_sync()\n \n Commit b6b65ca20bc9 (\"iommu/io-pgtable-arm: Add support for non-strict\n mode\") added an unconditional call to io_pgtable_tlb_sync() immediately\n after the case where we replace a block entry with a table entry during\n an unmap() call. This is redundant, since the IOMMU API will call\n iommu_tlb_sync() on this path and the patch in question mentions this:\n \n | To save having to reason about it too much, make sure the invalidation\n | in arm_lpae_split_blk_unmap() just performs its own unconditional sync\n | to minimise the window in which we\u0027re technically violating the break-\n | before-make requirement on a live mapping. This might work out redundant\n | with an outer-level sync for strict unmaps, but we\u0027ll never be splitting\n | blocks on a DMA fastpath anyway.\n \n However, this sync gets in the way of deferred TLB invalidation for leaf\n entries and is at best a questionable, unproven hack. Remove it.\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 6d1bcb957be2850e0776f24c289e1f87c256baeb\nAuthor: Will Deacon \u003cwill@kernel.org\u003e\nDate: Tue Jul 2 16:43:07 2019 +0100\n\n iommu: Remove empty iommu_tlb_range_add() callback from iommu_ops\n \n Commit add02cfdc9bc (\"iommu: Introduce Interface for IOMMU TLB Flushing\")\n added three new TLB flushing operations to the IOMMU API so that the\n underlying driver operations can be batched when unmapping large regions\n of IO virtual address space.\n \n However, the -\u003eiotlb_range_add() callback has not been implemented by\n any IOMMU drivers (amd_iommu.c implements it as an empty function, which\n incurs the overhead of an indirect branch). Instead, drivers either flush\n the entire IOTLB in the -\u003eiotlb_sync() callback or perform the necessary\n invalidation during -\u003eunmap().\n \n Attempting to implement -\u003eiotlb_range_add() for arm-smmu-v3.c revealed\n two major issues:\n \n 1. The page size used to map the region in the page-table is not known,\n and so it is not generally possible to issue TLB flushes in the most\n efficient manner.\n \n 2. The only mutable state passed to the callback is a pointer to the\n iommu_domain, which can be accessed concurrently and therefore\n requires expensive synchronisation to keep track of the outstanding\n flushes.\n \n Remove the callback entirely in preparation for extending -\u003eunmap() and\n -\u003eiotlb_sync() to update a token on the caller\u0027s stack.\n \n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit 15e830e90fde81b32c96490fb9ff1a26097609eb\nAuthor: Mao Wenan \u003cmaowenan@huawei.com\u003e\nDate: Wed Jul 24 14:25:45 2019 +0800\n\n mwifiex: use eth_broadcast_addr() to assign broadcast address\n \n This patch is to use eth_broadcast_addr() to assign broadcast address\n insetad of memcpy().\n \n Signed-off-by: Mao Wenan \u003cmaowenan@huawei.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit b43d6c8e8d7f19b44d4b40666fc38c8444b8143a\nAuthor: Ping-Ke Shih \u003cpkshih@realtek.com\u003e\nDate: Tue Jul 23 11:10:23 2019 +0800\n\n rtlwifi: remove assignment to itself\n \n Module parameters of \u0027sw_crypto\u0027 and \u0027disable_watchdog\u0027 are false by\n default. If new value is desired, we can do it during inserting module,\n assignment existing in source code is not reasonable.\n \n Reported-by: David Binderman \u003cdcb314@hotmail.com\u003e\n CC: Larry Finger \u003cLarry.Finger@lwfinger.net\u003e\n Signed-off-by: Ping-Ke Shih \u003cpkshih@realtek.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit e3b1d879ccda9ffd5332777bb1beeb2cc913faa8\nAuthor: Rafał Miłecki \u003crafal@milecki.pl\u003e\nDate: Sun Jul 21 21:52:17 2019 +0200\n\n brcmfmac: don\u0027t net_ratelimit() CONSOLE messages on firmware crash\n \n Firmware crash is a pretty rare event and can\u0027t happen too frequently as\n it has to be followed by a hardware reinitialization and config reload.\n It should be safe to don\u0027t use net_ratelimit() when it happens.\n \n For reporting \u0026 debugging purposes it\u0027s important to provide a complete\n log as the last lines are actually the most important. This change\n modifies brcmfmac to print all messages in an unlimited way in that\n specific case. With this change there should be finally a backtrace of\n firmware finally visible after a crash.\n \n Signed-off-by: Rafał Miłecki \u003crafal@milecki.pl\u003e\n Acked-by: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 2ec4ad49b98e4a14147d04f914717135eca7c8b1\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Tue Jul 16 22:42:18 2019 +0800\n\n libertas_tf: Use correct channel range in lbtf_geo_init\n \n It seems we should use \u0027range\u0027 instead of \u0027priv-\u003erange\u0027\n in lbtf_geo_init(), because \u0027range\u0027 is the corret one\n related to current regioncode.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Fixes: 691cdb49388b (\"libertas_tf: command helper functions for libertas_tf\")\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 8812022cb2fdaa1c6d7e0c3e028b45ca039650ea\nAuthor: Zong-Zhe Yang \u003ckevin_yang@realtek.com\u003e\nDate: Tue Jul 16 13:28:20 2019 +0800\n\n rtw88: debug: dump tx power indexes in use\n \n Add a read entry in debugfs to dump current tx power\n indexes in use for each path and each rate section.\n The corresponding power bases, power by rate, and\n power limit are also included.\n \n Also this patch fixes unused function warning.\n \n Fixes: b741422218ef (\"rtw88: refine flow to get tx power index\")\n Signed-off-by: Zong-Zhe Yang \u003ckevin_yang@realtek.com\u003e\n Signed-off-by: Yan-Hsuan Chuang \u003cyhchuang@realtek.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 9a29f7d8476c143ddbb9ecca614b66d2842c25d8\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Sat Jul 13 00:45:35 2019 +0530\n\n rtlwifi: btcoex: fix issue possible condition with no effect (if \u003d\u003d else)\n \n fix below issue reported by coccicheck\n drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.c:514:1-3:\n WARNING: possible condition with no effect (if \u003d\u003d else)\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Acked-by: Ping-Ke Shih \u003cpkshih@realtek.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit f0248ec49bdeeba29ec2ec238ba54c4c287033a8\nAuthor: Stanislaw Gruszka \u003csgruszka@redhat.com\u003e\nDate: Fri Jul 12 14:09:50 2019 +0200\n\n mt7601u: use params-\u003essn value directly\n \n There is no point to use pointer to params-\u003essn.\n \n Signed-off-by: Stanislaw Gruszka \u003csgruszka@redhat.com\u003e\n Acked-by: Jakub Kicinski \u003ckubakici@wp.pl\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 4b11c915f00caeef3292ed0429acc579b9da762a\nAuthor: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\nDate: Thu Jul 11 11:05:12 2019 +0200\n\n brcmfmac: remove unnecessary strlcpy() upon obtaining \"ver\" iovar\n \n Recently a strcpy() was replaced by strlcpy(). However, the strcpy()\n was not needed in the first place. So removing that line of code.\n \n Reviewed-by: Hante Meuleman \u003chante.meuleman@broadcom.com\u003e\n Reviewed-by: Pieter-Paul Giesberts \u003cpieter-paul.giesberts@broadcom.com\u003e\n Reviewed-by: Franky Lin \u003cfranky.lin@broadcom.com\u003e\n Signed-off-by: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit e0bfb9601d4812719167cc4124a0d6db1e2f55e4\nAuthor: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\nDate: Thu Jul 11 11:05:11 2019 +0200\n\n brcmfmac: simply remove flowring if bus is down\n \n When the bus is down, eg. due to rmmod, there is no need to\n attempt to inform firmware about it.\n \n Reviewed-by: Hante Meuleman \u003chante.meuleman@broadcom.com\u003e\n Reviewed-by: Pieter-Paul Giesberts \u003cpieter-paul.giesberts@broadcom.com\u003e\n Reviewed-by: Franky Lin \u003cfranky.lin@broadcom.com\u003e\n Signed-off-by: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 1ac11ae949dd883854f4523ef8e3a32aabfd6256\nAuthor: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\nDate: Thu Jul 11 11:05:10 2019 +0200\n\n brcmfmac: avoid firmware commands when bus is down\n \n Upon rmmod a few attempts are made to inform firmware, but there is\n no point as the bus is down and these will fail. Avoid them to keep\n the logs clean.\n \n Reported-by: Stefan Wahren \u003cstefan.wahren@i2se.com\u003e\n Reviewed-by: Hante Meuleman \u003chante.meuleman@broadcom.com\u003e\n Reviewed-by: Pieter-Paul Giesberts \u003cpieter-paul.giesberts@broadcom.com\u003e\n Reviewed-by: Franky Lin \u003cfranky.lin@broadcom.com\u003e\n Signed-off-by: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit c33330ac06fe863289643e7a13ecdb6a2502dad7\nAuthor: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\nDate: Thu Jul 11 11:05:09 2019 +0200\n\n brcmfmac: clear events in brcmf_fweh_detach() will always fail\n \n Clearing firmware events in brcmf_fweh_detach() is always failing\n because it is called only upon driver remove and communication\n with firmware is no longer possible.\n \n Reviewed-by: Hante Meuleman \u003chante.meuleman@broadcom.com\u003e\n Reviewed-by: Pieter-Paul Giesberts \u003cpieter-paul.giesberts@broadcom.com\u003e\n Reviewed-by: Franky Lin \u003cfranky.lin@broadcom.com\u003e\n Signed-off-by: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit c613085b74941024194e41b200601b9aa6ee388f\nAuthor: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\nDate: Thu Jul 11 11:05:08 2019 +0200\n\n brcmfmac: avoid firmware command in brcmf_netdev_open() when bus is down\n \n No point in sending a firmware command when bus is down so make it\n conditional checking the state.\n \n Reviewed-by: Hante Meuleman \u003chante.meuleman@broadcom.com\u003e\n Reviewed-by: Pieter-Paul Giesberts \u003cpieter-paul.giesberts@broadcom.com\u003e\n Reviewed-by: Franky Lin \u003cfranky.lin@broadcom.com\u003e\n Signed-off-by: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 14fcfd1cc0c05ea58f47dd693fdd13f25dfe995e\nAuthor: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\nDate: Thu Jul 11 11:05:07 2019 +0200\n\n brcmfmac: change the order of things in brcmf_detach()\n \n When brcmf_detach() from the bus layer upon rmmod we can no longer\n communicate. Hence we will set the bus state to DOWN and cleanup\n the event and protocol layer. The network interfaces need to be\n deleted before brcmf_cfg80211_detach() because the latter does the\n wiphy_unregister() which issues a warning if there are still network\n devices linked to the wiphy instance.\n \n Reviewed-by: Hante Meuleman \u003chante.meuleman@broadcom.com\u003e\n Reviewed-by: Pieter-Paul Giesberts \u003cpieter-paul.giesberts@broadcom.com\u003e\n Reviewed-by: Franky Lin \u003cfranky.lin@broadcom.com\u003e\n Signed-off-by: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\n Tested-by: Rafał Miłecki \u003crafal@milecki.pl\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit a84a60ccdd65278485fb495f468a5ab91a75c649\nAuthor: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\nDate: Thu Jul 11 11:05:06 2019 +0200\n\n Revert \"brcmfmac: fix NULL pointer derefence during USB disconnect\"\n \n This reverts commit 5cdb0ef6144f47440850553579aa923c20a63f23. Subsequent\n changes make rework the driver code fixing the issue differently.\n \n Signed-off-by: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit fa9050927fa885410055ee03c948c2252693d296\nAuthor: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\nDate: Thu Jul 11 10:45:32 2019 +0200\n\n brcmfmac: allow 160MHz in custom regulatory rules\n \n The driver has custom regulatory rules which had maximum bandwidth\n for 5GHz channels set to 80MHz. As a consequence the driver can\n not use 160MHz in AP mode even when the device supports it. So\n relax the rules allowing 160MHz. After wiphy_register() the channel\n flags are updated according what the device actually supports.\n \n Reviewed-by: Hante Meuleman \u003chante.meuleman@broadcom.com\u003e\n Reviewed-by: Pieter-Paul Giesberts \u003cpieter-paul.giesberts@broadcom.com\u003e\n Reviewed-by: Franky Lin \u003cfranky.lin@broadcom.com\u003e\n Signed-off-by: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 011a56a3336a5de9c3152c169cd52ff79b8c3f89\nAuthor: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\nDate: Thu Jul 11 10:45:31 2019 +0200\n\n brcmfmac: enable DFS_OFFLOAD extended feature if supported\n \n If the firmware supports 802.11h and the device can operate in 5GHz\n band we can enable DFS_OFFLOAD extended feature.\n \n Reviewed-by: Hante Meuleman \u003chante.meuleman@broadcom.com\u003e\n Reviewed-by: Pieter-Paul Giesberts \u003cpieter-paul.giesberts@broadcom.com\u003e\n Reviewed-by: Franky Lin \u003cfranky.lin@broadcom.com\u003e\n Signed-off-by: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit f491645f039420fb7e14283e21b90772571c807c\nAuthor: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\nDate: Thu Jul 11 10:45:30 2019 +0200\n\n brcmfmac: add 160MHz in chandef_to_chanspec()\n \n The function chandef_to_chanspec() was not handling 160MHz bandwidth\n resulting in wrong encoding of the channel. That resulting in firmware\n rejecting the provided channel specification.\n \n Reviewed-by: Hante Meuleman \u003chante.meuleman@broadcom.com\u003e\n Reviewed-by: Pieter-Paul Giesberts \u003cpieter-paul.giesberts@broadcom.com\u003e\n Reviewed-by: Franky Lin \u003cfranky.lin@broadcom.com\u003e\n Signed-off-by: Arend van Spriel \u003carend.vanspriel@broadcom.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 29b68a920f6abb7b5ba21ab4b779f62d536bac9b\nAuthor: Jian-Hong Pan \u003cjian-hong@endlessm.com\u003e\nDate: Thu Jul 11 13:24:27 2019 +0800\n\n rtw88: pci: Use DMA sync instead of remapping in RX ISR\n \n Since each skb in RX ring is reused instead of new allocation, we can\n treat the DMA in a more efficient way by DMA synchronization.\n \n Signed-off-by: Jian-Hong Pan \u003cjian-hong@endlessm.com\u003e\n Cc: \u003cstable@vger.kernel.org\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit ee6db78f5db9bfe426c57a1ec9713827ebccd2d4\nAuthor: Jian-Hong Pan \u003cjian-hong@endlessm.com\u003e\nDate: Thu Jul 11 13:24:26 2019 +0800\n\n rtw88: pci: Rearrange the memory usage for skb in RX ISR\n \n Testing with RTL8822BE hardware, when available memory is low, we\n frequently see a kernel panic and system freeze.\n \n First, rtw_pci_rx_isr encounters a memory allocation failure (trimmed):\n \n rx routine starvation\n WARNING: CPU: 7 PID: 9871 at drivers/net/wireless/realtek/rtw88/pci.c:822 rtw_pci_rx_isr.constprop.25+0x35a/0x370 [rtwpci]\n [ 2356.580313] RIP: 0010:rtw_pci_rx_isr.constprop.25+0x35a/0x370 [rtwpci]\n \n Then we see a variety of different error conditions and kernel panics,\n such as this one (trimmed):\n \n rtw_pci 0000:02:00.0: pci bus timeout, check dma status\n skbuff: skb_over_panic: text:00000000091b6e66 len:415 put:415 head:00000000d2880c6f data:000000007a02b1ea tail:0x1df end:0xc0 dev:\u003cNULL\u003e\n ------------[ cut here ]------------\n kernel BUG at net/core/skbuff.c:105!\n invalid opcode: 0000 [#1] SMP NOPTI\n RIP: 0010:skb_panic+0x43/0x45\n \n When skb allocation fails and the \"rx routine starvation\" is hit, the\n function returns immediately without updating the RX ring. At this\n point, the RX ring may continue referencing an old skb which was already\n handed off to ieee80211_rx_irqsafe(). When it comes to be used again,\n bad things happen.\n \n This patch allocates a new, data-sized skb first in RX ISR. After\n copying the data in, we pass it to the upper layers. However, if skb\n allocation fails, we effectively drop the frame. In both cases, the\n original, full size ring skb is reused.\n \n In addition, to fixing the kernel crash, the RX routine should now\n generally behave better under low memory conditions.\n \n Buglink: https://bugzilla.kernel.org/show_bug.cgi?id\u003d204053\n Signed-off-by: Jian-Hong Pan \u003cjian-hong@endlessm.com\u003e\n Cc: \u003cstable@vger.kernel.org\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 764f3f1ecffc434096e0a2b02f1a6cc964a89df6\nAuthor: Kevin Easton \u003ckevin@guarana.org\u003e\nDate: Wed Jul 10 13:31:38 2019 +0000\n\n libertas: Add missing sentinel at end of if_usb.c fw_table\n \n This sentinel tells the firmware loading process when to stop.\n \n Reported-and-tested-by: syzbot+98156c174c5a2cad9f8f@syzkaller.appspotmail.com\n Signed-off-by: Kevin Easton \u003ckevin@guarana.org\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 5ff29d836d1beb347080bd96e6321c811a8e3f62\nAuthor: Joe Perches \u003cjoe@perches.com\u003e\nDate: Tue Jul 9 22:04:22 2019 -0700\n\n rtw88: Fix misuse of GENMASK macro\n \n Arguments are supposed to be ordered high then low.\n \n Signed-off-by: Joe Perches \u003cjoe@perches.com\u003e\n Acked-by: Yan-Hsuan Chuang \u003cyhchuang@realtek.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit c032461936de82a3f4108cacc7c3d1b204c42eef\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Jul 5 11:37:32 2019 +0100\n\n wl3501_cs: remove redundant variable rc\n \n The variable rc is being initialized with a value that is never\n read and it is being updated later with a new value that is returned.\n The variable is redundant and can be replaced with a return 0 as\n there are no other return points in this function.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 4c8a468510196ff0a2c5ea28e8c27cd543a95861\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Jul 5 09:17:34 2019 +0100\n\n libertas: remove redundant assignment to variable ret\n \n The variable ret is being initialized with a value that is never\n read and it is being updated later with a new value. The\n initialization is redundant and can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 18e714687bea7f9f5155625aec04e335c0022128\nAuthor: Chris Chiu \u003cchiu@endlessm.com\u003e\nDate: Thu Jul 4 18:55:28 2019 +0800\n\n rtl8xxxu: Fix wifi low signal strength issue of RTL8723BU\n \n The WiFi tx power of RTL8723BU is extremely low after booting. So\n the WiFi scan gives very limited AP list and it always fails to\n connect to the selected AP. This module only supports 1x1 antenna\n and the antenna is switched to bluetooth due to some incorrect\n register settings.\n \n Compare with the vendor driver https://github.com/lwfinger/rtl8723bu,\n we realized that the 8723bu\u0027s enable_rf() does the same thing as\n rtw_btcoex_HAL_Initialize() in vendor driver. And it by default\n sets the antenna path to BTC_ANT_PATH_BT which we verified it\u0027s\n the cause of the wifi weak tx power. The vendor driver will set\n the antenna path to BTC_ANT_PATH_PTA in the consequent btcoexist\n mechanism, by the function halbtc8723b1ant_PsTdma.\n \n This commit hand over the antenna control to PTA(Packet Traffic\n Arbitration), which compares the weight of bluetooth/wifi traffic\n then determine whether to continue current wifi traffic or not.\n After PTA take control, The wifi signal will be back to normal and\n the bluetooth scan can also work at the same time. However, the\n btcoexist still needs to be handled under different circumstances.\n If there\u0027s a BT connection established, the wifi still fails to\n connect until BT disconnected.\n \n Signed-off-by: Chris Chiu \u003cchiu@endlessm.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 1dc244064c47d6df7925ca0895f8365e68d3abd1\nAuthor: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nDate: Wed Jul 3 13:39:56 2019 +0200\n\n rt2x00: no need to check return value of debugfs_create functions\n \n When calling debugfs functions, there is no need to ever check the\n return value. The function can work or not, but the code logic should\n never do something different based on this.\n \n Because we don\u0027t need to save the individual debugfs files and\n directories, remove the local storage of them and just remove the entire\n debugfs directory in a single call, making things a lot simpler.\n \n Cc: Stanislaw Gruszka \u003csgruszka@redhat.com\u003e\n Cc: Helmut Schaa \u003chelmut.schaa@googlemail.com\u003e\n Cc: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n Cc: \"David S. Miller\" \u003cdavem@davemloft.net\u003e\n Cc: linux-wireless@vger.kernel.org\n Cc: netdev@vger.kernel.org\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Acked-by: Stanislaw Gruszka \u003csgruszka@redhat.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 3b902fa811cf6bf7f9ad0ffb77d0a133e0b3bd61\nAuthor: Soeren Moch \u003csmoch@web.de\u003e\nDate: Mon Jul 1 12:53:14 2019 +0200\n\n rt2x00usb: remove unnecessary rx flag checks\n \n In contrast to the TX path, there is no need to separately read the transfer\n status from the device after receiving RX data. Consequently, there is no\n real STATUS_PENDING RX processing queue entry state.\n Remove the unnecessary ENTRY_DATA_STATUS_PENDING flag checks from the RX path.\n Also remove the misleading comment about reading RX status from device.\n \n Suggested-by: Stanislaw Gruszka \u003csgruszka@redhat.com\u003e\n Signed-off-by: Soeren Moch \u003csmoch@web.de\u003e\n Acked-by: Stanislaw Gruszka \u003csgruszka@redhat.com\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 231e83fdcd03bce4f5c71fc318cbdbe65bd2b80b\nAuthor: Enrico Weigelt \u003cinfo@metux.net\u003e\nDate: Thu Jun 27 16:30:02 2019 +0200\n\n rsi: return explicit error values\n \n Explicitly return constants instead of variable (and rely on\n it to be explicitly initialized), if the value is supposed\n to be fixed anyways. Align it with the rest of the driver,\n which does it the same way.\n \n Signed-off-by: Enrico Weigelt \u003cinfo@metux.net\u003e\n Signed-off-by: Kalle Valo \u003ckvalo@codeaurora.org\u003e\n\ncommit 005ad18727b489eb9fd6182f8941042e274387dc\nAuthor: Vasyl Gomonovych \u003cgomonovych@gmail.com\u003e\nDate: Fri Jul 19 18:28:06 2019 +0200\n\n extcon: sm5502: Add IRQ_ONESHOT\n \n Do not fire irq again until thread done\n \n Signed-off-by: Vasyl Gomonovych \u003cgomonovych@gmail.com\u003e\n Signed-off-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n\ncommit 8bc4810b435a3338346efd70e94c4d6fd0863635\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Thu May 30 20:39:32 2019 +0200\n\n extcon: gpio: Request reasonable interrupts\n \n The only thing that makes sense is to request a falling edge interrupt\n if the line is active low and a rising edge interrupt if the line is\n active high, so just do that and get rid of the assignment from\n platform data. The GPIO descriptor knows if the line is active high\n or low.\n \n Also make irq a local variable in probe(), it\u0027s not used anywhere else.\n \n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n [cw00.choi: Fix build error of data-\u003eirq_flags]\n Signed-off-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n\ncommit 7d2b02275e89d564ddb0ace09ef4aa9e2ac2973d\nAuthor: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\nDate: Wed Jul 24 10:49:14 2019 +0100\n\n extcon: arizona: Update binding example to use available defines\n \n Signed-off-by: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n Signed-off-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n\ncommit 6527c6856faca69b3abcc2dbed0a00ea40ca8e16\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Tue Jul 23 19:43:01 2019 +0200\n\n extcon: fsa9480: Support the FSA880 variant\n \n The older compatible variant of this chip is called FSA880\n and works the same way, if we need some quirks in the future,\n it is good to let it have its own compatible string.\n \n Cc: devicetree@vger.kernel.org\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n\ncommit 2bf8fb39eb70b6cf1d0f7efbe9fa9fb27d55fdf4\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Wed Jul 24 09:58:49 2019 +0100\n\n drm/i915/guc: Set GuC init params only once\n \n All the GuC objects are perma-pinned, so their offset can\u0027t change at\n runtime. We can therefore set (and log!) the parameters only once during\n boot.\n \n Suggested-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190724085849.18047-1-chris@chris-wilson.co.uk\n\ncommit a6d9cfcb853f32cda972841dace0060e9508ac7a\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:01 2019 +0200\n\n extcon: extcon-max77843: convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n\ncommit c2433827c1a149b72f1413f0151155f6fa4b3214\nAuthor: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nDate: Sat Jul 13 01:23:18 2019 -0700\n\n Input: cyapa - switch to using devm_device_add_group()\n \n Instead of installing custom devm cleanup action to remove attribute\n groups on failure, let\u0027s use the dedicated devm API.\n \n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 2e75cfaa32c113568994eda716a0bdddc92b376e\nAuthor: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nDate: Sat Jul 13 01:18:14 2019 -0700\n\n Input: elan_i2c - switch to using devm_add_action_or_reset()\n \n Instead of manually disabling regulators when devm_add_action() fails we can\n use devm_add_action_or_reset() which does it for us.\n \n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 9609b904c7eb411020d367d1f0904d0691819592\nAuthor: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\nDate: Sat Jul 13 01:13:16 2019 -0700\n\n Input: elan_i2c - switch to using devm_device_add_groups()\n \n Instead of installing custom devm cleanup action to remove attribute\n groups on failure, let\u0027s use the dedicated devm API.\n \n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 3ccd63d369c80d7e73e19ca996d29b6e5d7225bb\nAuthor: Denis Efremov \u003cefremov@linux.com\u003e\nDate: Sun Jul 14 11:26:34 2019 -0700\n\n Input: synaptics-rmi4 - remove the exporting of rmi_2d_sensor_set_input_params\n \n The function rmi_2d_sensor_set_input_params is declared static and marked\n EXPORT_SYMBOL_GPL, which is at best an odd combination. Because the\n function is not used outside of the drivers/input/rmi4/rmi_2d_sensor.c\n file it is defined in, this commit removes the EXPORT_SYMBOL_GPL() marking.\n \n Signed-off-by: Denis Efremov \u003cefremov@linux.com\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit a104dbc527551bda0d97db493a2d3d31e3076d4f\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 20:40:21 2019 +0300\n\n extcon: arizona: Switch to use device_property_count_u32()\n \n Use use device_property_count_u32() directly, that makes code neater.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n Signed-off-by: Chanwoo Choi \u003ccw00.choi@samsung.com\u003e\n\ncommit f8890bd2875bf9d6aca4cd7fbf246364dbeb8a0f\nAuthor: Mukesh Ojha \u003cmojha@codeaurora.org\u003e\nDate: Tue Jul 16 09:34:39 2019 +0200\n\n Input: ts4800-ts - use devm_platform_ioremap_resource()\n \n devm_platform_ioremap_resource() internally have platform_get_resource()\n and devm_ioremap_resource() in it. So instead of calling them separately\n use devm_platform_ioremap_resource() directly.\n \n Signed-off-by: Mukesh Ojha \u003cmojha@codeaurora.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 9601fa8fa9feb90afdddab345d942e33aa505835\nAuthor: Mukesh Ojha \u003cmojha@codeaurora.org\u003e\nDate: Tue Jul 16 09:34:13 2019 +0200\n\n Input: sun4i-ts - use devm_platform_ioremap_resource()\n \n devm_platform_ioremap_resource() internally have platform_get_resource()\n and devm_ioremap_resource() in it. So instead of calling them separately\n use devm_platform_ioremap_resource() directly.\n \n Signed-off-by: Mukesh Ojha \u003cmojha@codeaurora.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 46bec7a918715d3b0e8d22e5e99d3245c1097c65\nAuthor: Mukesh Ojha \u003cmojha@codeaurora.org\u003e\nDate: Tue Jul 16 09:27:22 2019 +0200\n\n Input: mxs-lradc-ts - use devm_platform_ioremap_resource()\n \n devm_platform_ioremap_resource() internally have platform_get_resource()\n and devm_ioremap_resource() in it. So instead of calling them separately\n use devm_platform_ioremap_resource() directly.\n \n Signed-off-by: Mukesh Ojha \u003cmojha@codeaurora.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 9d41cbe23db0a3c51c0b28db987a6898f5d1255c\nAuthor: Mukesh Ojha \u003cmojha@codeaurora.org\u003e\nDate: Tue Jul 16 09:26:56 2019 +0200\n\n Input: fsl-imx25-tcq - use devm_platform_ioremap_resource()\n \n devm_platform_ioremap_resource() internally have platform_get_resource()\n and devm_ioremap_resource() in it. So instead of calling them separately\n use devm_platform_ioremap_resource() directly.\n \n Signed-off-by: Mukesh Ojha \u003cmojha@codeaurora.org\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit fba76a58452694b9b13c07e48839fa84c75f57af\nAuthor: Oliver Hartkopp \u003csocketcan@hartkopp.net\u003e\nDate: Tue Jul 23 15:17:55 2019 +0200\n\n can: Add SPDX license identifiers for CAN subsystem\n \n Add missing SPDX identifiers for the CAN network layer and correct the SPDX\n license for two of its include files to make sure the BSD-3-Clause applies\n for the entire subsystem.\n \n Signed-off-by: Oliver Hartkopp \u003csocketcan@hartkopp.net\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 60649d4e0af6c26b6c423dea9c57f39e823fc0c5\nAuthor: Oliver Hartkopp \u003csocketcan@hartkopp.net\u003e\nDate: Tue Jul 23 14:08:47 2019 +0200\n\n can: remove obsolete empty ioctl() handler\n \n With commit c7cbdbf29f488a (\"net: rework SIOCGSTAMP ioctl handling\") the only\n ioctl function in can_ioctl() has been removed.\n \n As this SIOCGSTAMP ioctl command is now handled in net/socket.c we can entirely\n remove the CAN specific ioctl functions.\n \n Signed-off-by: Oliver Hartkopp \u003csocketcan@hartkopp.net\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 4c7f715485159ab8a55b9471fc6fd75db51fd623\nAuthor: Jeroen Hofstee \u003cjhofstee@victronenergy.com\u003e\nDate: Mon Apr 29 12:03:32 2019 +0000\n\n can: ti_hecc: use timestamp based rx-offloading\n \n As already mentioned in [1] and included in [2], there is an off by one\n issue since the high bank is already enabled when the _next_ mailbox to\n be read has index 12, so the mailbox being read was 13. The message can\n therefore go into mailbox 31 and the driver will be repolled until the\n mailbox 12 eventually receives a msg. Or the message might end up in the\n 12th mailbox, but then it would become disabled after reading it and only\n be enabled again in the next \"round\" after mailbox 13 was read, which can\n cause out of order messages, since the lower priority mailboxes can\n accept messages in the meantime.\n \n As mentioned in [3] there is a hardware race condition when changing the\n CANME register while messages are being received. Even when including a\n busy poll on reception, like in [2] there are still overflows and out of\n order messages at times, but less then without the busy loop polling.\n Unlike what the patch suggests, the polling time is not in the microsecond\n range, but takes as long as a current CAN bus reception needs to finish,\n so typically more in the fraction of millisecond range. Since the timeout\n is in jiffies it won\u0027t timeout.\n \n Even with these additional fixes the driver is still not able to provide a\n proper FIFO which doesn\u0027t drop packages. So change the driver to use\n rx-offload and base order on timestamp instead of message box numbers. As\n a side affect, this also fixes [4] and [5].\n \n Before this change messages with a single byte counter were dropped /\n received out of order at a bitrate of 250kbit/s on an am3517. With this\n patch that no longer occurs up to and including 1Mbit/s.\n \n [1] https://linux-can.vger.kernel.narkive.com/zgO9inVi/patch-can-ti-hecc-fix-rx-wrong-sequence-issue#post6\n [2] http://arago-project.org/git/projects/?p\u003dlinux-omap3.git;a\u003dcommit;h\u003d02346892777f07245de4d5af692513ebd852dcb2\n [3] https://linux-can.vger.kernel.narkive.com/zgO9inVi/patch-can-ti-hecc-fix-rx-wrong-sequence-issue#post5\n [4] https://patchwork.ozlabs.org/patch/895956/\n [5] https://www.spinics.net/lists/netdev/msg494971.html\n \n Cc: Anant Gole \u003canantgole@ti.com\u003e\n Cc: AnilKumar Ch \u003canilkumar@ti.com\u003e\n Signed-off-by: Jeroen Hofstee \u003cjhofstee@victronenergy.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 915f9666421cf65cc30668fd42760b6f78f9744d\nAuthor: Sean Nyekjaer \u003csean@geanix.com\u003e\nDate: Tue Apr 9 10:39:48 2019 +0200\n\n can: flexcan: add support for DT property \u0027wakeup-source\u0027\n \n The flexcan controller can be forced as a wakeup source by\n stating that explicitly in the device\u0027s .dts file using the\n \"wakeup-source\" boolean property.\n \n Signed-off-by: Sean Nyekjaer \u003csean@geanix.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 10e0c525fc41c1a80d33e3075d6f7f5902e74bbd\nAuthor: Sean Nyekjaer \u003csean@geanix.com\u003e\nDate: Tue Apr 9 10:39:49 2019 +0200\n\n dt-bindings: can: flexcan: add can wakeup property\n \n This patch adds the wakeup-source boolean property.\n \n Signed-off-by: Sean Nyekjaer \u003csean@geanix.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 8c306bec2d09c8436eea989b1be2a381cae93418\nAuthor: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\nDate: Thu Dec 13 07:08:00 2018 +0000\n\n can: flexcan: add support for PE clock source select\n \n Add support to select the clock source for CAN Protocol Engine (PE).\n It\u0027s SoC Implementation dependent. Refer to RM for detailed definition\n of each SoC. We select clock source 1 (peripheral clock) by default in\n driver now, this patch adds support to parse the clock source from the DT.\n \n Signed-off-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Signed-off-by: Joakim Zhang \u003cqiangqing.zhang@nxp.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 9d733992772dc04ce5e8839f867b4ff83ee75c34\nAuthor: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\nDate: Thu Dec 13 07:07:57 2018 +0000\n\n dt-bindings: can: flexcan: add PE clock source property to device tree\n \n The FlexCAN controller can parse clock source property from DTS file to\n select PE clock source.\n \n Signed-off-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Signed-off-by: Joakim Zhang \u003cqiangqing.zhang@nxp.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit ca10989632d8820749fad37e13843750198e450a\nAuthor: Aisheng Dong \u003caisheng.dong@nxp.com\u003e\nDate: Fri Nov 30 08:53:26 2018 +0000\n\n can: flexcan: implement can Runtime PM\n \n Flexcan will be disabled during suspend if no wakeup function required and\n enabled after resume accordingly. During this period, we could explicitly\n disable clocks.\n Since PM is optional, the clock is enabled at probe to guarante the\n clock is running when PM is not enabled in the kernel.\n \n Implement Runtime PM which will:\n 1) Without CONFIG_PM, clock is running whether Flexcan up or down.\n 2) With CONFIG_PM, clock enabled while Flexcan up and disabled when\n Flexcan down.\n 3) Disable clock when do system suspend and enable clock while system\n resume.\n 4) Make Power Domain framework be able to shutdown the corresponding\n power domain of this device.\n \n Signed-off-by: Aisheng Dong \u003caisheng.dong@nxp.com\u003e\n Signed-off-by: Joakim Zhang \u003cqiangqing.zhang@nxp.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 26bca9fe5f1dcd9093f5ed17fcb78d1f6fb32e5e\nAuthor: Robert P. J. Day \u003crpjday@crashcourse.ca\u003e\nDate: Wed Jun 12 11:48:56 2019 -0400\n\n can: Kconfig: correct history of the CAN protocol\n \n Current history of CAN protocol is wrong, fix it in the Kconfig file.\n \n Signed-off-by: Robert P. J. Day \u003crpjday@crashcourse.ca\u003e\n Acked-by: Oliver Hartkopp \u003csocketcan@hartkopp.net\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 64d6ce8fe48e4768d3bcba7873a1a368162af334\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Wed Jun 19 17:45:17 2019 +0100\n\n can: xilinx_can: clean up indentation issue\n \n A statement is indented one level too deep, fix this.\n \n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Reviewed-by: Michal Simek \u003cmichal.simek@xilinx.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit c223da689324f2ec669a3d10ac312a316410614b\nAuthor: Appana Durga Kedareswara rao \u003cappana.durga.rao@xilinx.com\u003e\nDate: Mon Mar 18 17:02:44 2019 +0530\n\n can: xilinx_can: Add support for CANFD FD frames\n \n CANFD IP supports both CAN and CAN FD frames,\n Existing driver supports only CAN frames, This patch\n adds support for CAN FD frames.\n \n Signed-off-by: Naga Sureshkumar Relli \u003cnaga.sureshkumar.relli@xilinx.com\u003e\n Signed-off-by: Appana Durga Kedareswara rao \u003cappana.durga.rao@xilinx.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 8dce7ea41a1e313cb1ba512b1098c7181ee0fd21\nAuthor: Appana Durga Kedareswara rao \u003cappana.durga.rao@xilinx.com\u003e\nDate: Mon Mar 18 17:02:43 2019 +0530\n\n can: xilinx_can: Add cantype parameter in xcan_devtype_data struct\n \n To differentiate between different CAN IP\u0027s this patch adds\n cantype enum variable in the xcan_devtype_data structure\n \n Signed-off-by: Appana Durga Kedareswara rao \u003cappana.durga.rao@xilinx.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 3281b380ec9f8a9d89f58d5f5353c7e3b2046f93\nAuthor: Appana Durga Kedareswara rao \u003cappana.durga.rao@xilinx.com\u003e\nDate: Mon Mar 18 17:02:42 2019 +0530\n\n can: xilinx_can: Fix flags field initialization for axi can and canps\n \n AXI CAN IP and CANPS IP supports tx fifo empty feature, this patch updates\n the flags field for the same.\n \n Signed-off-by: Appana Durga Kedareswara rao \u003cappana.durga.rao@xilinx.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit c942a575b95f762dd8721487619d24068471af60\nAuthor: Appana Durga Kedareswara rao \u003cappana.durga.rao@xilinx.com\u003e\nDate: Mon Mar 18 17:02:46 2019 +0530\n\n can: xilinx_can: Fix kernel doc warnings\n \n This patch fixes below kernel doc warnings\n warning: Function parameter or member \u0027priv\u0027 not described in\n \u0027xcan_write_frame\u0027\n warning: Function parameter or member \u0027skb\u0027 not described in\n \u0027xcan_start_xmit_fifo\u0027\n warning: Function parameter or member \u0027ndev\u0027 not described in\n \u0027xcan_start_xmit_fifo\u0027\n warning: Function parameter or member \u0027skb\u0027 not described in\n \u0027xcan_start_xmit_mailbox\u0027\n warning: Function parameter or member \u0027ndev\u0027 not described in\n \u0027xcan_start_xmit_mailbox\u0027\n warning: Function parameter or member \u0027priv\u0027 not described in\n \u0027xcan_rx_fifo_get_next_frame\u0027\n \n Signed-off-by: Appana Durga Kedareswara rao \u003cappana.durga.rao@xilinx.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit d38f9180da881f130801f964339a98015f7e53da\nAuthor: Appana Durga Kedareswara rao \u003cappana.durga.rao@xilinx.com\u003e\nDate: Mon Mar 18 17:02:41 2019 +0530\n\n can: xilinx_can: Fix style issues\n \n This patch fixes below checkpatch warnings and checks in the driver.\n \n CHECK: Alignment should match open parenthesis\n + void (*write_reg)(const struct xcan_priv *priv, enum xcan_reg reg,\n + u32 val);\n \n CHECK: Alignment should match open parenthesis\n +static void xcan_write_reg_le(const struct xcan_priv *priv, enum xcan_reg reg,\n + u32 val)\n \n CHECK: Alignment should match open parenthesis\n +static void xcan_write_reg_be(const struct xcan_priv *priv, enum xcan_reg reg,\n + u32 val)\n \n CHECK: Alignment should match open parenthesis\n + netdev_dbg(ndev, \"BRPR\u003d0x%08x, BTR\u003d0x%08x\\n\",\n + priv-\u003eread_reg(priv, XCAN_BRPR_OFFSET),\n \n CHECK: Alignment should match open parenthesis\n + netdev_warn(ndev,\n + \"timed out for correct mode\\n\");\n \n CHECK: Alignment should match open parenthesis\n + netdev_dbg(ndev, \"status:#x%08x\\n\",\n + priv-\u003eread_reg(priv, XCAN_SR_OFFSET));\n \n CHECK: spaces preferred around that \u0027-\u0027 (ctx:VxV)\n + (CAN_EFF_ID_BITS-CAN_SFF_ID_BITS)) \u003c\u003c\n ^\n \n CHECK: Alignment should match open parenthesis\n + netdev_dbg(ndev, \"%s: error status register:0x%x\\n\",\n + __func__, priv-\u003eread_reg(priv, XCAN_ESR_OFFSET));\n \n WARNING: line over 80 characters\n + offset \u003d XCAN_RXMSG_2_FRAME_OFFSET(fsr \u0026 XCAN_FSR_RI_MASK);\n \n WARNING: line over 80 characters\n + offset \u003d XCAN_RXMSG_FRAME_OFFSET(fsr \u0026 XCAN_FSR_RI_MASK);\n \n WARNING: line over 80 characters\n + while ((isr \u0026 XCAN_IXR_TXOK_MASK) \u0026\u0026 !WARN_ON(++retries \u003d\u003d 100)) {\n \n WARNING: line over 80 characters\n + priv-\u003ewrite_reg(priv, XCAN_ICR_OFFSET, XCAN_IXR_TXOK_MASK);\n \n CHECK: Alignment should match open parenthesis\n + netdev_err(ndev, \"%s: pm_runtime_get failed(%d)\\n\",\n + __func__, ret);\n \n CHECK: Alignment should match open parenthesis\n + ret \u003d request_irq(ndev-\u003eirq, xcan_interrupt, priv-\u003eirq_flags,\n + ndev-\u003ename, ndev);\n \n CHECK: Alignment should match open parenthesis\n +static int xcan_get_berr_counter(const struct net_device *ndev,\n + struct can_berr_counter *bec)\n \n CHECK: Alignment should match open parenthesis\n + netdev_err(ndev, \"%s: pm_runtime_get failed(%d)\\n\",\n + __func__, ret);\n \n CHECK: Please don\u0027t use multiple blank lines\n +\n +\n \n CHECK: Alignment should match open parenthesis\n + netdev_err(ndev, \"%s: pm_runtime_get failed(%d)\\n\",\n + __func__, ret);`\n \n Signed-off-by: Appana Durga Kedareswara rao \u003cappana.durga.rao@xilinx.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 5443c226ba9166521360ed3148175d63cd5ab263\nAuthor: Dan Murphy \u003cdmurphy@ti.com\u003e\nDate: Thu May 9 11:11:08 2019 -0500\n\n can: tcan4x5x: Add tcan4x5x driver to the kernel\n \n Add the TCAN4x5x SPI CAN driver.\n \n This device uses the Bosch MCAN IP core along with a SPI interface map.\n Register to the MCAN common core code to manage the MCAN IP.\n \n This device has a special method to indicate a write/read operation on\n the data payload.\n \n Acked-by: Wolfgang Grandegger \u003cwg@grandegger.com\u003e\n Signed-off-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 4edd396a1911222da7a5d4b2bc58ab546de02bd5\nAuthor: Dan Murphy \u003cdmurphy@ti.com\u003e\nDate: Thu May 9 11:11:07 2019 -0500\n\n dt-bindings: can: tcan4x5x: Add DT bindings for TCAN4x5X driver\n \n DT binding documentation for TI TCAN4x5x driver.\n \n Signed-off-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 441ac340169b792bf3df274f062e4939abb93ce8\nAuthor: Dan Murphy \u003cdmurphy@ti.com\u003e\nDate: Thu May 9 11:11:06 2019 -0500\n\n can: m_can: Rename m_can_priv to m_can_classdev\n \n Rename the common m_can_priv class structure to m_can_classdev as this\n is more descriptive.\n \n Acked-by: Wolfgang Grandegger \u003cwg@grandegger.com\u003e\n Signed-off-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n Acked-by: Faiz Abbas \u003cfaiz_abbas@ti.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit f524f829b75a7d934f56f63f2ed4d42f4e1d06d9\nAuthor: Dan Murphy \u003cdmurphy@ti.com\u003e\nDate: Thu May 9 11:11:05 2019 -0500\n\n can: m_can: Create a m_can platform framework\n \n Create a m_can platform framework that peripheral\n devices can register to and use common code and register sets.\n The peripheral devices may provide read/write and configuration\n support of the IP.\n \n Acked-by: Wolfgang Grandegger \u003cwg@grandegger.com\u003e\n Signed-off-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n Acked-by: Faiz Abbas \u003cfaiz_abbas@ti.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 69652195b6e071d9adcd7c8129943d95813c42ea\nAuthor: Dan Murphy \u003cdmurphy@ti.com\u003e\nDate: Thu May 9 11:11:09 2019 -0500\n\n can: m_can: Fix checkpatch issues on existing code\n \n Fix checkpatch issues found during the m_can framework creation, before\n framework creation in the following patches.\n \n Fix these 4 check issues:\n CHECK: Unnecessary parentheses around \u0027cdev-\u003ecan.state !\u003d CAN_STATE_ERROR_WARNING\u0027\n if (psr \u0026 PSR_EW \u0026\u0026\n (cdev-\u003ecan.state !\u003d CAN_STATE_ERROR_WARNING)) {\n \n CHECK: Unnecessary parentheses around \u0027cdev-\u003ecan.state !\u003d CAN_STATE_ERROR_PASSIVE\u0027\n if ((psr \u0026 PSR_EP) \u0026\u0026\n (cdev-\u003ecan.state !\u003d CAN_STATE_ERROR_PASSIVE)) {\n \n CHECK: Unnecessary parentheses around \u0027cdev-\u003ecan.state !\u003d CAN_STATE_BUS_OFF\u0027\n if ((psr \u0026 PSR_BO) \u0026\u0026\n (cdev-\u003ecan.state !\u003d CAN_STATE_BUS_OFF)) {\n \n CHECK: Unnecessary parentheses around \u0027priv-\u003eversion \u003c\u003d 31\u0027\n if ((priv-\u003eversion \u003c\u003d 31) \u0026\u0026 (irqstatus \u0026 IR_MRAF) \u0026\u0026\n (m_can_read(priv, M_CAN_ECR) \u0026 ECR_RP)) {\n \n Signed-off-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n Acked-by: Faiz Abbas \u003cfaiz_abbas@ti.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit b07fbf239738f8ab380af15e1b0e1d27552695a9\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Tue Jan 29 12:06:12 2019 -0600\n\n can: mark expected switch fall-throughs\n \n In preparation to enabling -Wimplicit-fallthrough, mark switch cases\n where we are expecting to fall through.\n \n This patch fixes the following warnings:\n \n drivers/net/can/peak_canfd/peak_pciefd_main.c:668:3: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n drivers/net/can/spi/mcp251x.c:875:7: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n drivers/net/can/usb/peak_usb/pcan_usb.c:422:6: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n drivers/net/can/at91_can.c:895:6: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n drivers/net/can/at91_can.c:953:15: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n drivers/net/can/usb/peak_usb/pcan_usb.c: In function ‘pcan_usb_decode_error’:\n drivers/net/can/usb/peak_usb/pcan_usb.c:422:6: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n if (n \u0026 PCAN_USB_ERROR_BUS_LIGHT) {\n ^\n drivers/net/can/usb/peak_usb/pcan_usb.c:428:2: note: here\n case CAN_STATE_ERROR_WARNING:\n ^~~~\n \n Warning level 3 was used: -Wimplicit-fallthrough\u003d3\n \n This patch is part of the ongoing efforts to enabling\n -Wimplicit-fallthrough.\n \n Notice that in some cases spelling mistakes were fixed.\n In other cases, the /* fall through */ comment is placed\n at the bottom of the case statement, which is what GCC\n is expecting to find.\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Acked-by: Nicolas Ferre \u003cnicolas.ferre@microchip.com\u003e # for the at91_can.c\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 26ad340e582d3d5958ed8456a1911d79cfb567b4\nAuthor: Henning Colliander \u003chenning.colliander@evidente.se\u003e\nDate: Tue May 28 14:48:21 2019 +0200\n\n can: kvaser_pciefd: Add driver for Kvaser PCIEcan devices\n \n This patch adds support for Kvaser PCIEcan devices. This includes\n support for up to 4 CAN channels on a single card, depending on device.\n \n Signed-off-by: Henning Colliander \u003chenning.colliander@evidente.se\u003e\n Signed-off-by: Jimmy Assarsson \u003cextja@kvaser.com\u003e\n Signed-off-by: Christer Beskow \u003cchbe@kvaser.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit d9e5d174ad8b23ef3e9a2b76cce98548595f6e2b\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Thu Feb 7 21:10:35 2019 -0600\n\n can: kvaser_usb: Use struct_size() in alloc_candev()\n \n One of the more common cases of allocation size calculations is finding\n the size of a structure that has a zero-sized array at the end, along\n with memory for some number of elements for that array. For example:\n \n struct foo {\n int stuff;\n void *entry[];\n };\n \n instance \u003d alloc(sizeof(struct foo) + count * sizeof(void *));\n \n Instead of leaving these open-coded and prone to type mistakes, we can\n now use the new struct_size() helper:\n \n instance \u003d alloc(struct_size(instance, entry, count));\n \n This code was detected with the help of Coccinelle.\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 2d91fdccad4d3bd890e03775bcd7e8beed07c610\nAuthor: Ji-Ze Hong (Peter Hong) \u003chpeter@gmail.com\u003e\nDate: Tue Jul 23 17:03:06 2019 +0800\n\n can: sja1000: f81601: add Fintek F81601 support\n \n This patch add support for Fintek PCIE to 2 CAN controller support\n \n Signed-off-by: Ji-Ze Hong (Peter Hong) \u003chpeter+linux_kernel@gmail.com\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 86be740845e3811c4517de1a8a36121190155e22\nAuthor: Clément Péron \u003cperon.clem@gmail.com\u003e\nDate: Sat Jun 8 01:10:59 2019 +0200\n\n arm64: dts: allwinner: h6: Enable IR on H6 boards\n \n Beelink GS1, OrangePi H6 boards and Pine H64 have an IR receiver.\n \n Enable it in their device-tree.\n \n Signed-off-by: Clément Péron \u003cperon.clem@gmail.com\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 9267811aad3524c857cf2e16bbadd8c569e15ab9\nAuthor: Clément Péron \u003cperon.clem@gmail.com\u003e\nDate: Sat Jun 8 01:10:58 2019 +0200\n\n arm64: dts: allwinner: h6: Add IR receiver node\n \n Allwinner H6 IR is similar to A31 and can use same driver.\n \n Add support for it.\n \n Signed-off-by: Clément Péron \u003cperon.clem@gmail.com\u003e\n Acked-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 63eb1e149576294717e3e5de48e902ca9d2f080d\nAuthor: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\nDate: Sat Jun 8 01:10:56 2019 +0200\n\n arm64: dts: allwinner: a64: Enable IR on Orange Pi Win\n \n OrangePi Win board contains IR receiver. Enable it.\n \n Signed-off-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Signed-off-by: Clément Péron \u003cperon.clem@gmail.com\u003e\n Acked-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 44a4f416c8388449fc5f9263788857d449e2a65f\nAuthor: Igors Makejevs \u003cgit_bb@bwzone.com\u003e\nDate: Sat Jun 8 01:10:55 2019 +0200\n\n arm64: dts: allwinner: a64: Add IR node\n \n IR peripheral is completely compatible with A31 one.\n \n Signed-off-by: Igors Makejevs \u003cgit_bb@bwzone.com\u003e\n Signed-off-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Signed-off-by: Clément Péron \u003cperon.clem@gmail.com\u003e\n Acked-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 8fa345e711bfdb69a18f548b717d5eb502b9892a\nAuthor: Clément Péron \u003cperon.clem@gmail.com\u003e\nDate: Sat Jun 8 01:10:53 2019 +0200\n\n ARM: dts: sunxi: Prefer A31 bindings for IR\n \n Since A31, memory mapping of the IR driver has changed.\n \n Prefer the A31 bindings instead of A13.\n \n Signed-off-by: Clément Péron \u003cperon.clem@gmail.com\u003e\n Acked-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 342d23a7dacf9c254c6b98b9b211e566820b7bad\nAuthor: Clément Péron \u003cperon.clem@gmail.com\u003e\nDate: Sat Jun 8 01:10:52 2019 +0200\n\n ARM: dts: sunxi: Prefer A31 bindings for IR\n \n Since A31, memory mapping of the IR driver has changed.\n \n Prefer the A31 bindings instead of A13.\n \n Signed-off-by: Clément Péron \u003cperon.clem@gmail.com\u003e\n Acked-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 2a0dd1d7415a5aa61802a24555aeb41b947c9433\nAuthor: Clément Péron \u003cperon.clem@gmail.com\u003e\nDate: Sat Jun 8 01:11:00 2019 +0200\n\n arm64: defconfig: Enable IR SUNXI option\n \n Enable CONFIG_IR_SUNXI option for ARM64, so that Allwinner A64/H6 SoCs\n can use their IR receiver controller.\n \n Signed-off-by: Clément Péron \u003cperon.clem@gmail.com\u003e\n Acked-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit a9c342ab77ab6e8ab662105335ae1e44e01271d4\nAuthor: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\nDate: Tue Jul 23 09:54:25 2019 +0200\n\n drm/mgag200: Don\u0027t unpin the current cursor image\u0027s buffer.\n \n Currently the displayed cursor buffer might be evicted from video memory.\n Not unpinning the BO fixes this problem. At this point, pixels_current\n also references the BO and it will be unpinned during the next cursor\n update.\n \n Signed-off-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Fixes: 94dc57b10399 (\"drm/mgag200: Rewrite cursor handling\")\n Cc: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Cc: Dave Airlie \u003cairlied@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190723075425.24028-4-tzimmermann@suse.de\n\ncommit cf578c8c8e0b1fcd7c7de649effaf2c7625433c5\nAuthor: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\nDate: Tue Jul 23 09:54:24 2019 +0200\n\n drm/mgag200: Set cursor scanout address to correct BO\n \n The hardware requires the correct memory address of the buffer. Currently\n the same BO\u0027s address is programmed unconditionally, so only every second\n cursor update actually becomes visible.\n \n Signed-off-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Fixes: 94dc57b10399 (\"drm/mgag200: Rewrite cursor handling\")\n Cc: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Cc: Dave Airlie \u003cairlied@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190723075425.24028-3-tzimmermann@suse.de\n\ncommit a0fd72d2c95b56c78feec0867f2077b5ff5991e6\nAuthor: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\nDate: Tue Jul 23 09:54:23 2019 +0200\n\n drm/mgag200: Pin displayed cursor BO to video memory\n \n The cursor BO has to be pinned to video ram while it\u0027s being displayed.\n With the current code, the BO might be pinned to system memory instead.\n The patch fixes this problem.\n \n Signed-off-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Fixes: 94dc57b10399 (\"drm/mgag200: Rewrite cursor handling\")\n Cc: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Cc: Dave Airlie \u003cairlied@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190723075425.24028-2-tzimmermann@suse.de\n\ncommit 4dfc39e1b639c7220ac7275dc8684cd8e82f188a\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Mon Jul 22 11:36:08 2019 +0200\n\n can: sja1000: Makefile/Kconfig: sort alphabetically\n \n This patch sorts the drivers in the Makefile alphabetically and arranges\n the Kconfig file accordingly.\n \n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 61670d62bd9eb3fa3d423b4f8f1d37bca853910d\nAuthor: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\nDate: Mon Jul 1 15:42:46 2019 +0200\n\n MAINTAINERS: can: add missing files to CAN NETWORK DRIVERS and CAN NETWORK LAYER\n \n This patch adds missing files to the CAN NETWORK DRIVERS and CAN NETWORK\n LAYER entry.\n \n Reported-by: Robert P. J. Day \u003crpjday@crashcourse.ca\u003e\n Signed-off-by: Marc Kleine-Budde \u003cmkl@pengutronix.de\u003e\n\ncommit 70a658073726ae79cf9747b651f577d1fdf45c3a\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Wed Jul 17 21:29:52 2019 -0600\n\n PCI: Unexport pci_bus_sem\n \n pci_bus_sem is not used by a loadable kernel module and does not need to be\n exported.\n \n It was exported by ce29ca3ea407 (\"PCI: acpiphp: remove all functions in\n slot, even without ACPI _EJx\"), which added a use of pci_bus_sem in\n acpiphp, which could be built as a module at that time. But since\n 6037a803b05e (\"PCI: acpiphp: Convert acpiphp to be builtin only, not\n modular\"), it can no longer be built as a module.\n \n Link: https://lore.kernel.org/r/20190718032951.40188-1-skunberg.kelsey@gmail.com\n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n\ncommit fae6b93b19b42fcae0ef71e669b643f3366f6824\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Wed Jul 17 12:23:53 2019 -0600\n\n PCI: Unexport pci_bus_get() and pci_bus_put()\n \n pci_bus_get() and pci_bus_put() are not used by a loadable kernel module\n and do not need to be exported.\n \n These were exported by fe830ef62ac6 (\"PCI: Introduce pci_bus_{get|put}() to\n manage PCI bus reference count\"), but there are no loadable modules in the\n tree that use them.\n \n Link: https://lore.kernel.org/r/20190717182353.45557-1-skunberg.kelsey@gmail.com\n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Acked-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 4f77e0956bd99901af2bfc6641437ff8bff8d4a4\nAuthor: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\nDate: Mon Jul 15 14:34:16 2019 -0600\n\n PCI: Remove pci_block_cfg_access() et al (unused)\n \n Remove the following unused functions from include/linux/pci.h:\n \n pci_block_cfg_access()\n pci_block_cfg_access_in_atomic()\n pci_unblock_cfg_access()\n \n These were added by fb51ccbf217c (\"PCI: Rework config space blocking\n services\"), though no callers were added.\n \n Link: https://lore.kernel.org/r/20190715203416.37547-1-skunberg.kelsey@gmail.com\n Signed-off-by: Kelsey Skunberg \u003cskunberg.kelsey@gmail.com\u003e\n Signed-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\n Reviewed-by: Lukas Bulwahn \u003clukas.bulwahn@gmail.com\u003e\n\ncommit 47da6e4dc3d37fefd913233c71575666c96491b5\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Tue Jul 23 14:34:45 2019 -0700\n\n selftests/bpf: remove perf buffer helpers\n \n libbpf\u0027s perf_buffer API supersedes trace_helper.h\u0027s helpers.\n Remove those helpers after all existing users were already moved to\n perf_buffer API.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit c17bec549c9dc969b4e725c56aa9ebb125378397\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Tue Jul 23 14:34:44 2019 -0700\n\n samples/bpf: switch trace_output sample to perf_buffer API\n \n Convert trace_output sample to libbpf\u0027s perf_buffer API.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit f58a4d51d8da7b248d8796e9981feb3d5a43d3d2\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Tue Jul 23 14:34:43 2019 -0700\n\n samples/bpf: convert xdp_sample_pkts_user to perf_buffer API\n \n Convert xdp_sample_pkts_user to libbpf\u0027s perf_buffer API.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Acked-by: Toke Høiland-Jørgensen \u003ctoke@redhat.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 898ca681cd78e57cec1f5d18df0829d35c464fe8\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Tue Jul 23 14:34:42 2019 -0700\n\n selftests/bpf: switch test_tcpnotify to perf_buffer API\n \n Switch test_tcpnotify test to use libbpf\u0027s perf_buffer API instead of\n re-implementing portion of it.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit 58b80815362ef18f528e17711f0abec842f56d59\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Tue Jul 23 14:34:41 2019 -0700\n\n selftests/bpf: convert test_get_stack_raw_tp to perf_buffer API\n \n Convert test_get_stack_raw_tp test to new perf_buffer API.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit f2a3e4e95f408314938d37fa3146a9f7b304ce74\nAuthor: Andrii Nakryiko \u003candriin@fb.com\u003e\nDate: Tue Jul 23 14:11:33 2019 -0700\n\n libbpf: provide more helpful message on uninitialized global var\n \n When BPF program defines uninitialized global variable, it\u0027s put into\n a special COMMON section. Libbpf will reject such programs, but will\n provide very unhelpful message with garbage-looking section index.\n \n This patch detects special section cases and gives more explicit error\n message.\n \n Signed-off-by: Andrii Nakryiko \u003candriin@fb.com\u003e\n Acked-by: Song Liu \u003csongliubraving@fb.com\u003e\n Signed-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\n\ncommit a581563f1bef035e4c8d634a1df26dae9140b115\nAuthor: Peng Wang \u003crocking@whu.edu.cn\u003e\nDate: Wed Jul 3 10:07:49 2019 +0800\n\n cgroup: minor tweak for logic to get cgroup css\n \n We could only handle the case that css exists\n and css_try_get_online() fails.\n \n Signed-off-by: Peng Wang \u003crocking@whu.edu.cn\u003e\n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n\ncommit 85db0023376f529c477c6110043e069ccee16d9c\nAuthor: Markus Elfring \u003celfring@users.sourceforge.net\u003e\nDate: Tue Jul 2 19:26:59 2019 +0200\n\n cgroup: Replace a seq_printf() call by seq_puts() in cgroup_print_ss_mask()\n \n A string which did not contain a data format specification should be put\n into a sequence. Thus use the corresponding function “seq_puts”.\n \n This issue was detected by using the Coccinelle software.\n \n Signed-off-by: Markus Elfring \u003celfring@users.sourceforge.net\u003e\n Signed-off-by: Tejun Heo \u003ctj@kernel.org\u003e\n\ncommit c2aeaaea175652af6610f97a0de6d7cd07311e18\nAuthor: Paul Burton \u003cpaul.burton@mips.com\u003e\nDate: Mon Jul 22 22:00:03 2019 +0000\n\n MIPS: Remove unused R8000 CPU support\n \n Our R8000 CPU support can only be included if a system selects\n CONFIG_SYS_HAS_CPU_R8000. No system does, making all R8000-related CPU\n support dead code. Remove it.\n \n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-mips@vger.kernel.org\n\ncommit ccd51b9fc3bf264482dab86875754c40cbe13045\nAuthor: Paul Burton \u003cpaul.burton@mips.com\u003e\nDate: Mon Jul 22 22:00:00 2019 +0000\n\n MIPS: Remove unused R5432_CP0_INTERRUPT_WAR\n \n R5432_CP0_INTERRUPT_WAR is defined as 0 for every system we support, and\n so the workaround is never used. Remove the dead code.\n \n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-mips@vger.kernel.org\n\ncommit 8e96b08472e6698011d3105912031e90e7ef553f\nAuthor: Paul Burton \u003cpaul.burton@mips.com\u003e\nDate: Mon Jul 22 21:59:50 2019 +0000\n\n MIPS: Remove unused R5432 CPU support\n \n Our R5432 CPU support can only be included if a system selects\n CONFIG_SYS_HAS_CPU_R5432. No system does, making all R5432-related CPU\n support dead code. Remove it.\n \n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-mips@vger.kernel.org\n\ncommit f9065b54d437c4660e3d974ad9ce5188c068cd76\nAuthor: Paul Burton \u003cpaul.burton@mips.com\u003e\nDate: Mon Jul 22 21:59:43 2019 +0000\n\n MIPS: Remove unused R4300 CPU support\n \n Our R4300 CPU support can only be included if a system selects\n CONFIG_SYS_HAS_CPU_R4300. No system does, making all R4300-related CPU\n support dead code. Remove it.\n \n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-mips@vger.kernel.org\n\ncommit e2dd73ac4440f7143e990e76bad9a46dc63a5951\nAuthor: Nicolas Saenz Julienne \u003cnsaenzjulienne@suse.de\u003e\nDate: Wed Jun 12 20:24:59 2019 +0200\n\n arm64: defconfig: enable cpufreq support for RPi3\n \n This enables both the new firmware clock driver and cpufreq driver\n available for the RPi3 family of boards.\n \n Signed-off-by: Nicolas Saenz Julienne \u003cnsaenzjulienne@suse.de\u003e\n Acked-by: Stefan Wahren \u003cstefan.wahren@i2se.com\u003e\n Signed-off-by: Stefan Wahren \u003cwahrenst@gmx.net\u003e\n\ncommit 3e3bb69589e482e0783f28d4cd1d8e56fda0bcbb\nAuthor: Roman Mashak \u003cmrv@mojatatu.com\u003e\nDate: Tue Jul 23 15:01:59 2019 -0400\n\n tc-testing: added tdc tests for [b|p]fifo qdisc\n \n Signed-off-by: Roman Mashak \u003cmrv@mojatatu.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ce103204cbe61f8d5d995f7a09a6b18e6b6ac3c4\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Tue Jul 23 19:39:43 2019 +0300\n\n hv_sock: Use consistent types for UUIDs\n \n The rest of Hyper-V code is using new types for UUID handling.\n Convert hv_sock as well.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Dexuan Cui \u003cdecui@microsoft.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 4c6f5d4038af2c7332630bdd75cfdc0309e97242\nAuthor: Nicolas Saenz Julienne \u003cnsaenzjulienne@suse.de\u003e\nDate: Wed Jun 12 20:24:58 2019 +0200\n\n ARM: defconfig: enable cpufreq driver for RPi\n \n This enables on both multi_v7_defconfig and bcm2835_defconfig the new\n firmware based clock and cpufreq drivers for the Raspberry Pi platform.\n \n In the case of bcm2835_defconfig, as the cpufreq subsystem was disabled,\n the conservative governor was selected as default since it better\n handles the high frequency transition latency.\n \n Signed-off-by: Nicolas Saenz Julienne \u003cnsaenzjulienne@suse.de\u003e\n Acked-by: Stefan Wahren \u003cstefan.wahren@i2se.com\u003e\n Signed-off-by: Stefan Wahren \u003cwahrenst@gmx.net\u003e\n\ncommit e03e47a3dcecfa56420328b339f143a17499b414\nAuthor: John Hurley \u003cjohn.hurley@netronome.com\u003e\nDate: Tue Jul 23 15:34:02 2019 +0100\n\n nfp: flower: offload MPLS set action\n \n Recent additions to the kernel include a TC action module to manipulate\n MPLS headers on packets. Such actions are available to offload via the\n flow_offload intermediate representation API.\n \n Modify the NFP driver to allow the offload of MPLS set actions to\n firmware. Set actions update the outermost MPLS header. The offload\n includes a mask to specify which fields should be set.\n \n Signed-off-by: John Hurley \u003cjohn.hurley@netronome.com\u003e\n Reviewed-by: Simon Horman \u003csimon.horman@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 35b7c70cc3b82c159c181f297a54d4bea6f4fd47\nAuthor: John Hurley \u003cjohn.hurley@netronome.com\u003e\nDate: Tue Jul 23 15:34:01 2019 +0100\n\n nfp: flower: offload MPLS pop action\n \n Recent additions to the kernel include a TC action module to manipulate\n MPLS headers on packets. Such actions are available to offload via the\n flow_offload intermediate representation API.\n \n Modify the NFP driver to allow the offload of MPLS pop actions to\n firmware. The act_mpls TC module enforces that the next protocol is\n supplied along with the pop action. Passing this to firmware allows it\n to properly rebuild the underlying packet after the pop.\n \n Signed-off-by: John Hurley \u003cjohn.hurley@netronome.com\u003e\n Reviewed-by: Simon Horman \u003csimon.horman@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit a6eb1817fb9860cbd630ceb05dae0491993bfd85\nAuthor: John Hurley \u003cjohn.hurley@netronome.com\u003e\nDate: Tue Jul 23 15:34:00 2019 +0100\n\n nfp: flower: offload MPLS push action\n \n Recent additions to the kernel include a TC action module to manipulate\n MPLS headers on packets. Such actions are available to offload via the\n flow_offload intermediate representation API.\n \n Modify the NFP driver to allow the offload of MPLS push actions to\n firmware.\n \n Signed-off-by: John Hurley \u003cjohn.hurley@netronome.com\u003e\n Reviewed-by: Simon Horman \u003csimon.horman@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 6749d59016981bca6d7000e40bdb08eed78dfa6f\nAuthor: John Hurley \u003cjohn.hurley@netronome.com\u003e\nDate: Tue Jul 23 15:33:59 2019 +0100\n\n net: sched: include mpls actions in hardware intermediate representation\n \n A recent addition to TC actions is the ability to manipulate the MPLS\n headers on packets.\n \n In preparation to offload such actions to hardware, update the IR code to\n accept and prepare the new actions.\n \n Note that no driver currently impliments the MPLS dec_ttl action so this\n is not included.\n \n Signed-off-by: John Hurley \u003cjohn.hurley@netronome.com\u003e\n Reviewed-by: Jakub Kicinski \u003cjakub.kicinski@netronome.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 658688ce6c936254c34ea1f31549ec62439574aa\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Tue Jul 23 12:02:26 2019 +0000\n\n net/mlx5e: xsk: dynamically allocate mlx5e_channel_param\n \n The structure is too large to put on the stack, resulting in a\n warning on 32-bit ARM:\n \n drivers/net/ethernet/mellanox/mlx5/core/en/xsk/setup.c:59:5: error: stack frame size of 1344 bytes in function\n \u0027mlx5e_open_xsk\u0027 [-Werror,-Wframe-larger-than\u003d]\n \n Use kvzalloc() instead.\n \n Fixes: a038e9794541 (\"net/mlx5e: Add XSK zero-copy support\")\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Signed-off-by: Maxim Mikityanskiy \u003cmaximmi@mellanox.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit baf6722aa0cbc5381f11520eb661be71b3ac379e\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 19:49:52 2019 +0800\n\n ide: Use dev_get_drvdata where possible\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c9b6c56d52bea8adbf10ae0b9e4a7b88017c6f27\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 22:16:42 2019 +0800\n\n net: jme: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 5daab287c67d8008a01e4a0db8bcb92b386b8adc\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 22:16:24 2019 +0800\n\n igb: Use dev_get_drvdata where possible\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1c8aa7b1f15bff91a9abef3f083770793ea2f773\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 22:15:51 2019 +0800\n\n i40e: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7f53be6f6b8d339bfce34514e23bb8eb7057648c\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 22:15:33 2019 +0800\n\n fm10k: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit ee2e80c194628a0ba5cd0a5c37b46fa136e11004\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 22:15:13 2019 +0800\n\n e1000e: Use dev_get_drvdata where possible\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f521eaa9d2ef6d85bc6c318148f019e9f40fc344\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 21:19:29 2019 +0800\n\n net: broadcom: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f54b0fc86ce69fd923153b19604afbea1455c1b4\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 21:18:56 2019 +0800\n\n net: atheros: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 9f293c9af856bc6b404759eedb9bc6ad8d9cf631\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 21:18:44 2019 +0800\n\n net: 3com: 3c59x: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit c349c0a28326c98ca145b59e0f7ba69e4eef3b80\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 16:13:14 2019 +0800\n\n atm: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 038029c03e21be3bab44ff9d8005b51b597a5bf2\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Tue Jul 23 20:31:49 2019 +0900\n\n sparc: remove unneeded uapi/asm/statfs.h\n \n statfs.h is listed in include/uapi/asm-generic/Kbuild, so Kbuild will\n automatically generate it.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7d9d60bdfa45181e439b50bb067e97168b99f89e\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 20:14:16 2019 +0200\n\n leds: is31fl319x: simplify getting the adapter of a client\n \n We have a dedicated pointer for that, so use it. Much easier to read and\n less computation involved.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 084323f62b0b976c9fd931d86c5d2553af5eb9f7\nAuthor: David S. Miller \u003cdavem@davemloft.net\u003e\nDate: Tue Jul 23 11:45:44 2019 -0700\n\n ftgmac100: Fix build.\n \n drivers/net/ethernet/faraday/ftgmac100.c:777:13: error: \u0027skb_frag_t {aka struct bio_vec}\u0027 has no member named \u0027size\u0027\n \n Fallout from the skb_frag_t conversion to bio_vec, simply\n use skb_frag_size().\n \n Fixes: b8b576a16f79 (\"net: Rename skb_frag_t size to bv_len\")\n Reported-by: René van Dorst \u003copensource@vdorst.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 69ab285b682328ef7622f0567f8afebb00b404f3\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Sun Jul 21 11:56:33 2019 +0200\n\n percpu: fix typo in pcpu_setup_first_chunk() comment\n \n s/perpcu/percpu/\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Signed-off-by: Dennis Zhou \u003cdennis@kernel.org\u003e\n [Dennis: updated title]\n\ncommit aae364d2a88897e1b3ac0f53414dab39726b80e9\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:18 2019 +0200\n\n rtc: s5m: convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Reviewed-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Tested-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190722172618.4061-5-wsa+renesas@sang-engineering.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit ca83542cdb5c14dd369de2539bdca2670d89feea\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:17 2019 +0200\n\n rtc: s35390a: convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Link: https://lore.kernel.org/r/20190722172618.4061-4-wsa+renesas@sang-engineering.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 7150710f3084de8d35ce3221eeae2caee8813f92\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:16 2019 +0200\n\n rtc: max77686: convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Reviewed-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Tested-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190722172618.4061-3-wsa+renesas@sang-engineering.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 955315b0dc8c8641311430f40fbe53990ba40e33\nAuthor: Benjamin Poirier \u003cbpoirier@suse.com\u003e\nDate: Tue Jul 23 15:14:13 2019 +0900\n\n qlge: Move drivers/net/ethernet/qlogic/qlge/ to drivers/staging/qlge/\n \n The hardware has been declared EOL by the vendor more than 5 years ago.\n What\u0027s more relevant to the Linux kernel is that the quality of this driver\n is not on par with many other mainline drivers.\n \n Cc: Manish Chopra \u003cmanishc@marvell.com\u003e\n Message-id: \u003c20190617074858.32467-1-bpoirier@suse.com\u003e\n Signed-off-by: Benjamin Poirier \u003cbpoirier@suse.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 467fece8fbc6774a3a3bd0981e1a342fb5022706\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Mon Jul 22 10:36:16 2019 +0900\n\n ASoC: soc-dai: move snd_soc_dai_stream_valid() to soc-dai.c\n \n snd_soc_dai_stream_valid() is function to check stream validity.\n But, some code is using it, some code are checking stream-\u003echannels_min\n directly. Doing samethings by different method is confusable.\n This patch uses same funcntion for same purpose.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87ftmyhmzz.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 88fdffa2eef45fc9f6009ee590741c44a57c49c7\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Mon Jul 22 10:36:27 2019 +0900\n\n ASoC: soc-core: use existing snd_soc_dai_digital_mute()\n \n ALSA SoC already has snd_soc_dai_digital_mute() for digital_mute,\n let\u0027s use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87ef2ihmzo.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit b423c4202135f7794e0a9c55a884f5933d8e7156\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Mon Jul 22 10:35:29 2019 +0900\n\n ASoC: soc-dai: add snd_soc_dai_compress_new()\n \n Current ALSA SoC is directly using dai-\u003edriver-\u003exxx,\n thus, it has deep nested bracket, and it makes code unreadable.\n This patch adds new snd_soc_dai_compress_new() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87h87ehn1a.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit dcdab5820edd6123911dbd767ee1e389008b6a83\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Mon Jul 22 10:35:05 2019 +0900\n\n ASoC: soc-dai: add snd_soc_dai_remove()\n \n Current ALSA SoC is directly using dai-\u003edriver-\u003exxx,\n thus, it has deep nested bracket, and it makes code unreadable.\n This patch adds new snd_soc_dai_remvoe() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87imruhn1x.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit cfd9b5fbfe1e8763018aea2600aa0d6ff015ebfc\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Mon Jul 22 10:34:56 2019 +0900\n\n ASoC: soc-dai: add snd_soc_dai_probe()\n \n Current ALSA SoC is directly using dai-\u003edriver-\u003exxx,\n thus, it has deep nested bracket, and it makes code unreadable.\n This patch adds new snd_soc_dai_probe() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87k1cahn26.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 24b09d051164680f0a1d1910efe21ce36ad5c1ca\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Mon Jul 22 10:34:43 2019 +0900\n\n ASoC: soc-dai: add snd_soc_dai_resume()\n \n Current ALSA SoC is directly using dai-\u003edriver-\u003exxx,\n thus, it has deep nested bracket, and it makes code unreadable.\n This patch adds new snd_soc_dai_resume() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87lfwqhn2j.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit e0f2262292d0c8160cfd9a8c40425107fb65ab29\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Mon Jul 22 10:34:29 2019 +0900\n\n ASoC: soc-dai: add snd_soc_dai_suspend()\n \n Current ALSA SoC is directly using dai-\u003edriver-\u003exxx,\n thus, it has deep nested bracket, and it makes code unreadable.\n This patch adds new snd_soc_dai_suspend() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87muh6hn2x.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 1dea80d4b2bd3b53c58f008ca2bcd73182583711\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Mon Jul 22 10:34:09 2019 +0900\n\n ASoC: soc-dai: add snd_soc_dai_delay()\n \n Current ALSA SoC is directly using dai-\u003edriver-\u003eops-\u003exxx,\n thus, it has deep nested bracket, and it makes code unreadable.\n This patch adds new snd_soc_dai_delay() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87o91mhn3i.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 5c0769af4caf8fbdad2e9c0051ab0081b8e22b0a\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Mon Jul 22 10:33:56 2019 +0900\n\n ASoC: soc-dai: add snd_soc_dai_bespoke_trigger()\n \n Current ALSA SoC is directly using dai-\u003edriver-\u003eops-\u003exxx,\n thus, it has deep nested bracket, and it makes code unreadable.\n This patch adds new snd_soc_dai_bespoke_trigger() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87r26ihn3u.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 95aef35533844f35544851b0cdc1fc154b603307\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Mon Jul 22 10:33:51 2019 +0900\n\n ASoC: soc-dai: add snd_soc_dai_trigger()\n \n Current ALSA SoC is directly using dai-\u003edriver-\u003eops-\u003exxx,\n thus, it has deep nested bracket, and it makes code unreadable.\n This patch adds new snd_soc_dai_trigger() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87sgqyhn40.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 4beb8e109d30d339d44308a767dd6f5614492f3e\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Mon Jul 22 10:33:45 2019 +0900\n\n ASoC: soc-dai: add snd_soc_dai_prepare()\n \n Current ALSA SoC is directly using dai-\u003edriver-\u003eops-\u003exxx,\n thus, it has deep nested bracket, and it makes code unreadable.\n This patch adds new snd_soc_dai_prepare() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87tvbehn46.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 330fcb5135e0588b1ea3b0bbab587d1317c1cf7b\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Mon Jul 22 10:33:39 2019 +0900\n\n ASoC: soc-dai: add snd_soc_dai_shutdown()\n \n Current ALSA SoC is directly using dai-\u003edriver-\u003eops-\u003exxx,\n thus, it has deep nested bracket, and it makes code unreadable.\n This patch adds new snd_soc_dai_shutdown() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87v9vuhn4b.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 5a52a04531486e2ab069b7882432c8b266db36e6\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Mon Jul 22 10:33:32 2019 +0900\n\n ASoC: soc-dai: add snd_soc_dai_startup()\n \n Current ALSA SoC is directly using dai-\u003edriver-\u003eops-\u003exxx,\n thus, it has deep nested bracket, and it makes code unreadable.\n This patch adds new snd_soc_dai_startup() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87wogahn4i.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 846faaed9df7899e74311db3aec0a41a2f6bc345\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Mon Jul 22 10:33:19 2019 +0900\n\n ASoC: soc-dai: add snd_soc_dai_hw_free()\n \n Current ALSA SoC is directly using dai-\u003edriver-\u003eops-\u003exxx,\n thus, it has deep nested bracket, and it makes code unreadable.\n This patch adds new snd_soc_dai_hw_free() and use it.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87y30qhn4w.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit aa6166c2ac28392d64f2d8b3acfb56c8fe657147\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Mon Jul 22 10:33:04 2019 +0900\n\n ASoC: soc-dai: mv soc_dai_hw_params() to soc-dai\n \n Sometimes ALSA SoC naming is very random.\n Current soc_dai_hw_params() should use snd_soc_dai_xxx() style.\n And then, 1st parameter should be dai. Otherwise it is confusable.\n - soc_dai_hw_params(..., dai);\n + snd_soc_dai_hw_params(dai, ...);\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/87zhl6hn5b.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 06f6e1d41427f394ad3f67ecf06efcd28a46932c\nAuthor: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\nDate: Mon Jul 22 10:32:12 2019 +0900\n\n ASoC: add soc-dai.c\n \n Current ALSA SoC has many snd_soc_dai_xxx() function which is\n using dai-\u003edriver-\u003eops-\u003exxx.\n But, some of them are implemented as snd_soc_dai_xxx(),\n but others are directly using dai-\u003edriver-\u003eops-\u003exxx.\n Because of it, the code is not easy to read.\n \n This patch creats new soc-dai.c and moves snd_soc_dai_xxx()\n functions into it.\n One exception is snd_soc_dai_is_dummy() which is based on\n soc-utils local variable. We need to keep it as-is there.\n \n Others which is directly using dai-\u003edriver-\u003eops-\u003exxx will be\n implemented at soc-dai.c by incremental patches.\n \n Signed-off-by: Kuninori Morimoto \u003ckuninori.morimoto.gx@renesas.com\u003e\n Link: https://lore.kernel.org/r/871ryij1r6.wl-kuninori.morimoto.gx@renesas.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 859fd6cbf1fb32b5428c26f837215c085b8a822e\nAuthor: Timo Wischer \u003ctwischer@de.adit-jv.com\u003e\nDate: Mon Jul 22 16:24:01 2019 +0900\n\n ASoC: rsnd: Support hw_free() callback at DAI level\n \n This patch provides the needed infrastructure to support calling hw_free()\n at the DAI level. This is for example required to free resources allocated\n in hw_params() callback.\n \n The modification of __rsnd_mod_add_hw_params does not have any side\n effects because rsnd_mod_ops::hw_params callback is not used by anyone\n until now.\n \n Signed-off-by: Timo Wischer \u003ctwischer@de.adit-jv.com\u003e\n Signed-off-by: Jiada Wang \u003cjiada_wang@mentor.com\u003e\n Link: https://lore.kernel.org/r/20190722072403.11008-2-jiada_wang@mentor.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit a5f888c53db575e3b04e7b4597699da67b9cd2eb\nAuthor: Michał Mirosław \u003cmirq-linux@rere.qmqm.pl\u003e\nDate: Tue Jul 23 18:32:07 2019 +0200\n\n regulator: act8865: support regulator-pull-down property\n \n AC8865 has internal 1.5k pull-down resistor that can be enabled when LDO\n is shut down.\n \n Signed-off-by: Michał Mirosław \u003cmirq-linux@rere.qmqm.pl\u003e\n Link: https://lore.kernel.org/r/d7338f0dfcac63eb958a6b5e42e2d540b3d3f54a.1563898936.git.mirq-linux@rere.qmqm.pl\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit cd81a91d04b8b869bbbf62ae80bd5e8c0e7a2b93\nAuthor: Michał Mirosław \u003cmirq-linux@rere.qmqm.pl\u003e\nDate: Tue Jul 23 18:32:06 2019 +0200\n\n regulator: act8865: rename fixed LDO ops\n \n Rename act8865_ldo_ops to act8865_fixed_ldo_ops to make room for\n variable-output LDO ops change.\n \n Signed-off-by: Michał Mirosław \u003cmirq-linux@rere.qmqm.pl\u003e\n Link: https://lore.kernel.org/r/12b1fe419e93dfe663990009bf1b2fbf630e9934.1563898936.git.mirq-linux@rere.qmqm.pl\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 906339a95e5a35f01cdf0430fca6dd526870a5ae\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 18:39:16 2019 +0800\n\n drm/i915: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190723103915.3964-1-hslester96@gmail.com\n\ncommit 44c1ee17644c822842782c5e33bf2df19b67df03\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Tue Jul 23 08:37:33 2019 -0700\n\n drm/i915/huc: fix status check\n \n Fix botched refactoring of the code that uncorrectly split a check on a\n bool, treating it as a u32.\n \n Reported-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Fixes: 84b1ca2f0e68 (\"drm/i915/uc: prefer intel_gt over i915 in GuC/HuC paths\")\n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190723153733.19401-1-daniele.ceraolospurio@intel.com\n\ncommit 14f8a0eb02531bdeca723af123be3e20c4aa1045\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jul 23 10:12:18 2019 +0100\n\n drm/i915: Squelch nop wait-for-idle trace\n \n If the system is already idle, omit the GEM_TRACE saying we are about to\n wait for idle. It looks confusing in the logs to see a continual stream\n of wait-for-idle, as one immediately assumes it is stuck in a loop.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190723091218.5886-1-chris@chris-wilson.co.uk\n\ncommit ca786b8db751c0dd980fccf2d65acb77a296f629\nAuthor: Shaokun Zhang \u003czhangshaokun@hisilicon.com\u003e\nDate: Tue Jul 2 15:35:53 2019 +0800\n\n arm64: perf: Remove unused macro\n \n ARMV8_EVENT_ATTR_RESOLVE became unused after commit \u003c4b1a9e6934ec\u003e\n (\"arm64/perf: Filter common events based on PMCEIDn_EL0\").\n \n Remove it.\n \n Cc: Will Deacon \u003cwill@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Signed-off-by: Shaokun Zhang \u003czhangshaokun@hisilicon.com\u003e\n Signed-off-by: Will Deacon \u003cwill@kernel.org\u003e\n\ncommit a7db285664dda63e066b943e6ec3c5f69473bf77\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Tue Jul 23 16:39:24 2019 +0800\n\n gpio: pch: Use dev_get_drvdata\n \n Instead of using to_pci_dev + pci_get_drvdata,\n use dev_get_drvdata to make code simpler.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Acked-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n\ncommit e2727ec58c113c2c95cbb75f6a129bd1fa0bf761\nAuthor: André Almeida \u003candrealmeid@collabora.com\u003e\nDate: Mon Jun 24 08:37:29 2019 -0400\n\n media: vimc.rst: add vimc-streamer source documentation\n \n Since vimc-streamer.{c, h} are fully documented and conforming with the\n kernel-doc syntax, add those files to vimc.rst\n \n Signed-off-by: André Almeida \u003candrealmeid@collabora.com\u003e\n Suggested-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit edb0b4e24baa2cd34ea142f8ffd479853e7e6710\nAuthor: André Almeida \u003candrealmeid@collabora.com\u003e\nDate: Mon Jun 24 08:37:28 2019 -0400\n\n media: vimc.rst: Add a proper alt attribute to vimc.dot\n \n According to W3C, \"the content of the alt attribute is: use text that\n fulfills the same function as the image\". While it\u0027s hard to describe\n the whole content of this image, replace the actual alt to something\n more useful to people with slow connection or that uses screen readers.\n \n Signed-off-by: André Almeida \u003candrealmeid@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit ed391879dd73ced1b3e41a721154deb6ed79e147\nAuthor: André Almeida \u003candrealmeid@collabora.com\u003e\nDate: Mon Jun 24 08:37:27 2019 -0400\n\n media: vimc: stream: format comments as kernel-doc\n \n Format the current existing comments as kernel-doc comments, to be\n reused at kernel documention. Add opening marks (/**) and return values.\n \n Signed-off-by: André Almeida \u003candrealmeid@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 959f9c841b6b7935f5168e7ade7387bc6ec324a4\nAuthor: André Almeida \u003candrealmeid@collabora.com\u003e\nDate: Mon Jun 24 08:37:26 2019 -0400\n\n media: vimc: stream: fix style of argument description\n \n As in \"Function parameters\" at doc-guide/kernel-doc.rst, \"the\n continuation of the description should start at the same column as the\n previous line\". Make the @producer_pixfmt comply with that.\n \n Signed-off-by: André Almeida \u003candrealmeid@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 7770f89b26ceb635e9e1b99e73c0d9e54dbda95c\nAuthor: André Almeida \u003candrealmeid@collabora.com\u003e\nDate: Mon Jun 24 08:37:25 2019 -0400\n\n media: vimc: stream: remove obsolete function doc\n \n As a more complete version of vimc_streamer_s_streamer comment was added\n at \"media: vimc: stream: add missing function documentation\" commit in\n .c file, remove the old documentation from .h file.\n \n Signed-off-by: André Almeida \u003candrealmeid@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 65953018b30ca64a34a7c0141d8a415c74a9b5bf\nAuthor: Hans Verkuil \u003chverkuil@xs4all.nl\u003e\nDate: Wed Jul 17 09:43:51 2019 -0400\n\n media: mx2_emmaprp: use struct v4l2_fh\n \n Convert this driver to use struct v4l2_fh and as a result switch to\n using vb2/v4l2_mem2mem helper functions as well.\n \n Only compile-tested due to lack of hardware. This driver is however\n very similar to the m2m-deinterlace driver in this respect, and that\n v4l2_fh conversion has been properly tested.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Reviewed-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit cfc1c34585492179e9e74a0b64d1be285646be67\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Thu Jul 4 16:02:39 2019 +0300\n\n pinctrl: intel: Drop double check for data in intel_pinctrl_probe_by_uid()\n \n There is no need to duplicate the check which is done in the common\n intel_pinctrl_probe().\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n\ncommit 854aca6d3ad14e041c19ecaa98aa708261189ade\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jul 17 09:29:09 2019 -0400\n\n media: omap_vout: use struct v4l2_fh\n \n This driver is one of the few that is still not using struct\n v4l2_fh. Convert it.\n \n Tested on a Pandaboard.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Reviewed-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit b4133ad3a5c8fee195079bde0ed9fd470c49d3d8\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jul 17 05:36:10 2019 -0400\n\n media: m2m-deinterlace: use struct v4l2_fh\n \n Convert this driver to use struct v4l2_fh and as a result switch to\n using vb2/v4l2_mem2mem helper functions as well.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Reviewed-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 2dbcb6fb96f5faca146279d5fc8a3c7b803b25cb\nAuthor: Hans Verkuil \u003chverkuil@xs4all.nl\u003e\nDate: Wed Jul 17 05:24:13 2019 -0400\n\n media: media/i2c: don\u0027t return ENOTTY if SUBDEV_API is not set\n \n If CONFIG_VIDEO_V4L2_SUBDEV_API is not set, then it is still possible\n to call set_fmt for V4L2_SUBDEV_FORMAT_TRY, the result is just not\n stored. So return 0 instead of -ENOTTY.\n \n Calling get_fmt with V4L2_SUBDEV_FORMAT_TRY should return -EINVAL\n instead of -ENOTTY, after all the get_fmt functionality is still\n present, just not supported for TRY.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Acked-by: Lad, Prabhakar \u003cprabhakar.csengg@gmail.com\u003e\n Acked-by: Rui Miguel Silva \u003crmfrfs@gmail.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 596a5a58e89a4bfe0d8a275ab552de990b18f0ac\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Mon Jul 8 09:07:42 2019 -0400\n\n media: vb2-dc: skip CPU sync in map/unmap dma_buf\n \n This is rougly equivalent to ca0e68e21aae (drm/prime: skip CPU sync\n in map/unmap dma_buf). The contig memory allocated is already device\n coherent memory, so there is no point in doing a CPU sync when\n mapping it to another device. Also most importers currently cache\n the mapping so the CPU sync would only happen on the first import,\n so we are better off with not pretending to do a cache synchronization\n at all.\n \n This gets rid of a lot of CPU overhead in uses where those dma-bufs\n are regularily imported and detached again, like Weston is currently\n doing in the DRM compositor.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Acked-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n [hverkuil-cisco@xs4all.nl: fix checkpatch warnings]\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 81f96ebc8a663869948c197c60f0356f247ac91c\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Jul 5 04:32:46 2019 -0400\n\n media: saa7134: remove redundant assignment to variable value\n \n The variable value is being assigned with a value that is never\n read and it is being updated later with a new value. The\n assignment is redundant and can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n \n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 322b169476aaeeb96515d69baaee330e47ac10a0\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Sat Jun 29 15:31:27 2019 -0400\n\n media: staging/media/davinci_vpfe: Add null check post kmalloc\n \n Add NULL check post memory operations\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 094efbe748c204fb2e10ebf6f100da926e10fc2f\nAuthor: Wen Yang \u003cwen.yang99@zte.com.cn\u003e\nDate: Thu Jun 27 23:01:16 2019 -0400\n\n media: ti-vpe: fix leaked of_node references\n \n The call to of_get_parent returns a node pointer with refcount\n incremented thus it must be explicitly decremented after the last\n usage.\n \n Detected by coccinelle with the following warnings:\n drivers/media/platform/ti-vpe/cal.c:1621:1-7: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 1607, but without a corresponding object release within this function.\n \n Signed-off-by: Wen Yang \u003cwen.yang99@zte.com.cn\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit da79bf41a4d170ca93cc8f3881a70d734a071c37\nAuthor: Wen Yang \u003cwen.yang99@zte.com.cn\u003e\nDate: Thu Jun 27 23:01:15 2019 -0400\n\n media: exynos4-is: fix leaked of_node references\n \n The call to of_get_child_by_name returns a node pointer with refcount\n incremented thus it must be explicitly decremented after the last\n usage.\n \n Detected by coccinelle with the following warnings:\n drivers/media/platform/exynos4-is/fimc-is.c:813:2-8: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 807, but without a corresponding object release within this function.\n drivers/media/platform/exynos4-is/fimc-is.c:870:1-7: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 807, but without a corresponding object release within this function.\n drivers/media/platform/exynos4-is/fimc-is.c:885:1-7: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 807, but without a corresponding object release within this function.\n drivers/media/platform/exynos4-is/media-dev.c:545:1-7: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 541, but without a corresponding object release within this function.\n drivers/media/platform/exynos4-is/media-dev.c:528:1-7: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 499, but without a corresponding object release within this function.\n drivers/media/platform/exynos4-is/media-dev.c:534:1-7: ERROR: missing of_node_put; acquired a node pointer with refcount incremented on line 499, but without a corresponding object release within this function.\n \n Signed-off-by: Wen Yang \u003cwen.yang99@zte.com.cn\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 02c25c003405776084d440c30041a33836b74710\nAuthor: Steve Longerbeam \u003cslongerbeam@gmail.com\u003e\nDate: Wed Jun 26 14:52:25 2019 -0400\n\n media: staging/imx: Fix NULL deref in find_pipeline_entity()\n \n Fix a cut\u0026paste error in find_pipeline_entity(). The start entity must be\n passed to media_entity_to_video_device() in find_pipeline_entity(), not\n pad-\u003eentity. The pad is only put to use later, after determining the start\n entity is not the entity being searched for.\n \n Fixes: 3ef46bc97ca2 (\"media: staging/imx: Improve pipeline searching\")\n Reported-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Steve Longerbeam \u003cslongerbeam@gmail.com\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 1a03f91c2c2419c3709c4554952c66695575e91c\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Thu Jul 18 10:16:43 2019 -0400\n\n media: vivid: work around high stack usage with clang\n \n Building a KASAN-enabled kernel with clang ends up in a case where too\n much is inlined into vivid_thread_vid_cap() and the stack usage grows\n a lot, possibly when the register allocation fails to produce efficient\n code and spills a lot of temporaries to the stack. This uses more\n than twice the amount of stack than the sum of the individual functions\n when they are not inlined:\n \n drivers/media/platform/vivid/vivid-kthread-cap.c:766:12: error: stack frame size of 2208 bytes in function \u0027vivid_thread_vid_cap\u0027 [-Werror,-Wframe-larger-than\u003d]\n \n Marking two of the key functions in here as \u0027noinline_for_stack\u0027 avoids\n the pathological case in clang without any apparent downside for gcc.\n \n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Acked-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 6732ecb335f5c8957ae319f50949b15f78d54621\nAuthor: Frederick Lawler \u003cfred@fredlawl.com\u003e\nDate: Wed Jul 17 22:07:40 2019 -0400\n\n media: cobalt: Prefer pcie_capability_read_word()\n \n Commit 8c0d3a02c130 (\"PCI: Add accessors for PCI Express Capability\")\n added accessors for the PCI Express Capability so that drivers didn\u0027t\n need to be aware of differences between v1 and v2 of the PCI\n Express Capability.\n \n Replace pci_read_config_word() and pci_write_config_word() calls with\n pcie_capability_read_word() and pcie_capability_write_word().\n \n Signed-off-by: Frederick Lawler \u003cfred@fredlawl.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit ec8b22b4bd8e9d797bbafc5b2a941e215587c8b5\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jul 17 05:03:45 2019 -0400\n\n media: via-camera: fix v4l2-compliance fails\n \n The patch fixes various v4l2-compliance failures:\n \n - missing support for control events\n - support of s/g_std even though this doesn\u0027t apply to a webcam\n - missing colorspace reporting\n - bus_info wasn\u0027t filled in by VIDIOC_QUERYCAP\n - parm.capture.readbuffers was overridden with wrong value\n - viacam_enum_framesizes/intervals didn\u0027t check the arguments\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit de6786bee48e6f29846b7b220ecfea32364117d4\nAuthor: Marko Kohtala \u003cmarko.kohtala@okoko.fi\u003e\nDate: Tue Jun 18 10:41:11 2019 +0300\n\n video: ssd1307fb: Add devicetree configuration of display setup\n \n Various displays have differences that only mean initializing the display\n driver IC with different fixed register values. Defining these in\n devicetree offers easier way to adapt the driver to new displays than\n requiring a patch to the kernel.\n \n This adds devicetree properties needed to make the initialization match\n the example setup as offered by Densitron for their 128x36 display.\n \n It also makes some old one bit parameter handling a little cleaner.\n \n Signed-off-by: Marko Kohtala \u003cmarko.kohtala@okoko.fi\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Michal Vokáč \u003cmichal.vokac@ysoft.com\u003e\n [b.zolnierkie: fix parenthesis alignment]\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190618074111.9309-7-marko.kohtala@okoko.fi\n\ncommit ca6216fce8db1a22d20517a945f7e3551cafc98e\nAuthor: Marko Kohtala \u003cmarko.kohtala@okoko.fi\u003e\nDate: Tue Jun 18 10:41:10 2019 +0300\n\n dt-bindings: display: ssd1307fb: Add initialization properties\n \n Document new bindings for adapting ssd1307fb driver to new displays.\n \n Signed-off-by: Marko Kohtala \u003cmarko.kohtala@okoko.fi\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Michal Vokáč \u003cmichal.vokac@ysoft.com\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190618074111.9309-6-marko.kohtala@okoko.fi\n\ncommit b0020d8af991a904de57dbd9a7bc7a61982a6a8c\nAuthor: Marko Kohtala \u003cmarko.kohtala@okoko.fi\u003e\nDate: Tue Jun 18 10:41:09 2019 +0300\n\n video: ssd1307fb: Handle width and height that are not multiple of 8\n \n Some displays have dimensions that are not multiple of eight, for example\n height of 36, but the driver divided the dimensions by 8. Defining display\n to the next multiple of 8 is not good as then the display registers get\n configured to dimensions that do not match. This contradicts intructions\n by some display manufacturers.\n \n Use DIV_ROUND_UP to multiple of 8 when needed so correct values can be\n used.\n \n The ssd1307fb_update_display bit reordering receives a simplification in\n the process.\n \n Signed-off-by: Marko Kohtala \u003cmarko.kohtala@okoko.fi\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Michal Vokáč \u003cmichal.vokac@ysoft.com\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190618074111.9309-5-marko.kohtala@okoko.fi\n\ncommit dd9782834dd9dde3624ff1acea8859f3d3e792d4\nAuthor: Marko Kohtala \u003cmarko.kohtala@okoko.fi\u003e\nDate: Tue Jun 18 10:41:08 2019 +0300\n\n video: ssd1307fb: Start page range at page_offset\n \n The page_offset was only applied to the end of the page range. This caused\n the display updates to cause a scrolling effect on the display because the\n amount of data written to the display did not match the range display\n expected.\n \n Fixes: 301bc0675b67 (\"video: ssd1307fb: Make use of horizontal addressing mode\")\n Signed-off-by: Marko Kohtala \u003cmarko.kohtala@okoko.fi\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Michal Vokáč \u003cmichal.vokac@ysoft.com\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190618074111.9309-4-marko.kohtala@okoko.fi\n\ncommit 6bc9431111be1d4c23ceb0b7c10c0ce09d1c9e9d\nAuthor: Marko Kohtala \u003cmarko.kohtala@okoko.fi\u003e\nDate: Tue Jun 18 10:41:07 2019 +0300\n\n video: ssd1307fb: Remove unneeded semicolons\n \n coccicheck reported unneeded semicolons. Remove them.\n \n Signed-off-by: Marko Kohtala \u003cmarko.kohtala@okoko.fi\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Michal Vokáč \u003cmichal.vokac@ysoft.com\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190618074111.9309-3-marko.kohtala@okoko.fi\n\ncommit ed1dc7d565788eb2b378360c79d7060ad2d41770\nAuthor: Marko Kohtala \u003cmarko.kohtala@okoko.fi\u003e\nDate: Tue Jun 18 10:41:06 2019 +0300\n\n video: ssd1307fb: Use screen_buffer instead of screen_base\n \n sparse reported incorrect type due to different address spaces.\n The screen_base is __iomem, but the memory is not from a device so we can\n use screen_buffer instead and avoid some type casts.\n \n Signed-off-by: Marko Kohtala \u003cmarko.kohtala@okoko.fi\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Michal Vokáč \u003cmichal.vokac@ysoft.com\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190618074111.9309-2-marko.kohtala@okoko.fi\n\ncommit 8cbff03bb090d537063c935058ab6fceb8ee30f5\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Mon Jul 15 11:20:17 2019 +0800\n\n video: fbdev-MMP: Remove call to memset after dma_alloc_coherent\n \n In commit 518a2f1925c3\n (\"dma-mapping: zero memory returned from dma_alloc_*\"),\n dma_alloc_coherent has already zeroed the memory.\n So memset is not needed.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190715032017.7311-1-huangfq.daxian@gmail.com\n\ncommit 9b9f35e7c6594b109ee4d70c3fd19a82ff88e8a6\nAuthor: Souptick Joarder \u003cjrdr.linux@gmail.com\u003e\nDate: Wed Jul 10 23:46:26 2019 +0530\n\n video: fbdev: nvidia: Remove dead code\n \n This is dead code since 3.15. If there is no plan to use it\n further, this can be removed forever.\n \n Signed-off-by: Souptick Joarder \u003cjrdr.linux@gmail.com\u003e\n Cc: Antonino Daplas \u003cadaplas@gmail.com\u003e\n Cc: Sabyasachi Gupta \u003csabyasachi.linux@gmail.com\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/1562782586-3994-1-git-send-email-jrdr.linux@gmail.com\n\ncommit b519294ee13aee030d6defd009998c84060ab205\nAuthor: Souptick Joarder \u003cjrdr.linux@gmail.com\u003e\nDate: Wed Jul 10 23:33:15 2019 +0530\n\n video: fbdev: nvidia: Remove extra return\n \n Minor cleanup to remove extra return statement.\n \n Signed-off-by: Souptick Joarder \u003cjrdr.linux@gmail.com\u003e\n Cc: Antonino Daplas \u003cadaplas@gmail.com\u003e\n Cc: Sabyasachi Gupta \u003csabyasachi.linux@gmail.com\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/1562781795-3494-1-git-send-email-jrdr.linux@gmail.com\n\ncommit 5ae01cba59c63e042080ebccf2bd3406288beaa6\nAuthor: Christoph Hellwig \u003chch@lst.de\u003e\nDate: Tue Jun 25 14:13:21 2019 +0200\n\n au1200fb: don\u0027t use DMA_ATTR_NON_CONSISTENT\n \n au1200fb allocates DMA memory using DMA_ATTR_NON_CONSISTENT, but never\n calls dma_cache_sync to synchronize the memory between the CPU and the\n device. If it was use on a not cache coherent bus that would be fatal,\n but as far as I can tell from the naming and the mips platform\n implementation it always is used in cache coherent systems. Remove\n the DMA_ATTR_NON_CONSISTENT flag, which is a no-op in that case.\n \n Signed-off-by: Christoph Hellwig \u003chch@lst.de\u003e\n Acked-by: Manuel Lauss \u003cmanuel.lauss@gmail.com\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190625121321.10197-1-hch@lst.de\n\ncommit a74cefd915124a16a2956773c30addc58e344047\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Mon Jun 24 23:37:24 2019 +0100\n\n fbmem: remove redundant assignment to err\n \n Variable err is initialized to a value that is never read and it\n is re-assigned later. The initialization is redundant and can\n be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190624223724.13629-1-colin.king@canonical.com\n\ncommit 710ae47dc6cb11f11a3007f7ade7d937804e160d\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Fri Jul 19 17:59:16 2019 +0200\n\n drm/tinydrm: Move tinydrm_display_pipe_init() to mipi-dbi\n \n tinydrm_display_pipe_init() has only one user now, so move it to mipi-dbi.\n \n Changes:\n - Remove drm_connector_helper_funcs.detect, it\u0027s always connected.\n - Store the connector and mode in mipi_dbi instead of it\u0027s own struct.\n \n Otherwise remove some leftover tinydrm-helpers.h inclusions.\n \n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: David Lechner \u003cdavid@lechnology.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190719155916.62465-12-noralf@tronnes.org\n\ncommit cc431212710860143b13877a0aa9d11a42598f49\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Fri Jul 19 17:59:15 2019 +0200\n\n drm/tinydrm/mipi-dbi: Add mipi_dbi_init_with_formats()\n \n The MIPI DBI standard support more pixel formats than what this helper\n supports. Add an init function that lets the driver use different\n format(s). This avoids open coding mipi_dbi_init() in st7586.\n \n st7586 sets preferred_depth but this is not necessary since it only\n supports one format.\n \n v2: Forgot to remove the mipi-\u003erotation assignment in st7586,\n mipi_dbi_init_with_formats() handles it.\n \n Cc: David Lechner \u003cdavid@lechnology.com\u003e\n Acked-by: David Lechner \u003cdavid@lechnology.com\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Tested-by: David Lechner \u003cdavid@lechnology.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190719155916.62465-11-noralf@tronnes.org\n\ncommit 1321db837549a0ff9dc2c95ff76c46770f7f02a1\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Fri Jul 19 17:59:14 2019 +0200\n\n drm/tinydrm/repaper: Don\u0027t use tinydrm_display_pipe_init()\n \n tinydrm.ko is going away so let\u0027s implement a connector.\n \n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190719155916.62465-10-noralf@tronnes.org\n\ncommit f729d8d9628d4093675abfd62d8d3a06d3d11732\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Fri Jul 19 17:59:13 2019 +0200\n\n drm/tinydrm: Move tinydrm_machine_little_endian()\n \n The tinydrm helper is going away so move it into the only user mipi-dbi.\n \n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190719155916.62465-9-noralf@tronnes.org\n\ncommit d23d4d4dac0119c7bfba63342a86a68b38c01779\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Fri Jul 19 17:59:12 2019 +0200\n\n drm/tinydrm: Move tinydrm_spi_transfer()\n \n This is only used by mipi-dbi drivers so move it there.\n \n The reason this isn\u0027t moved to the SPI subsystem is that it will in a\n later patch pass a dummy rx buffer for SPI controllers that need this.\n Low memory boards (64MB) can run into a problem allocating such a \"large\"\n contiguous buffer on every transfer after a long up time.\n This leaves a very specific use case, so we\u0027ll keep the function here.\n mipi-dbi will first go through a refactoring though, before this will\n be done.\n \n Remove SPI todo entry now that we\u0027re done with the tinydrm.ko SPI code.\n \n v2: Drop moving the mipi_dbi_spi_init() declaration (Sam)\n \n Cc: David Lechner \u003cdavid@lechnology.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: : David Lechner \u003cdavid@lechnology.com\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190719155916.62465-8-noralf@tronnes.org\n\ncommit 083a6c23b9ecf478265bb27ccacb8d587732058e\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Fri Jul 19 17:59:11 2019 +0200\n\n drm/tinydrm: Clean up tinydrm_spi_transfer()\n \n Prep work before moving the function to mipi-dbi.\n \n tinydrm_spi_transfer() was made to support one class of drivers in\n drivers/staging/fbtft that has not been converted to DRM yet, so strip\n away the unused functionality:\n - Start byte (header) is not used.\n - No driver relies on the automatic 16-bit byte swapping on little endian\n machines with SPI controllers only supporting 8 bits per word.\n \n Other changes:\n - No need to initialize ret\n - No need for the WARN since mipi-dbi only uses 8 and 16 bpw.\n - Use spi_message_init_with_transfers()\n \n Cc: David Lechner \u003cdavid@lechnology.com\u003e\n Acked-by: : David Lechner \u003cdavid@lechnology.com\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190719155916.62465-7-noralf@tronnes.org\n\ncommit e143364b4c1774f68e923a5a0bb0fca28ac25888\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Fri Jul 19 17:59:10 2019 +0200\n\n drm/tinydrm: Remove tinydrm_spi_max_transfer_size()\n \n spi-bcm2835 can handle \u003e64kB buffers now so there is no need to check\n -\u003emax_dma_len. The tinydrm_spi_max_transfer_size() max_len argument is\n not used by any callers, so not needed.\n \n Then we have the spi_max module parameter. It was added because\n staging/fbtft has support for it and there was a report that someone used\n it to set a small buffer size to avoid popping on a USB soundcard on a\n Raspberry Pi. In hindsight it shouldn\u0027t have been added, I should have\n waited for it to become a problem first. I don\u0027t know it anyone is\n actually using it, but since tinydrm_spi_transfer() is being moved to\n mipi-dbi, I\u0027m taking the opportunity to remove it. I\u0027ll add it back to\n mipi-dbi if someone complains.\n \n With that out of the way, spi_max_transfer_size() can be used instead.\n \n The chosen 16kB buffer size for Type C Option 1 (9-bit) interface is\n somewhat arbitrary, but a bigger buffer will have a miniscule impact on\n transfer speed, so it\u0027s probably fine.\n \n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190719155916.62465-6-noralf@tronnes.org\n\ncommit 8a18ac3a7de53f6792feed2cecc8cdaed927c667\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Fri Jul 19 17:59:09 2019 +0200\n\n drm/tinydrm: Remove spi debug buffer dumping\n \n The SPI event tracing can dump the buffer now so no need for this.\n Remove the debug print from tinydrm_spi_transfer() since this info can be\n gleaned from the trace event.\n \n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190719155916.62465-5-noralf@tronnes.org\n\ncommit cfcc8fcb117091bcd95b10b6b75deed30037a1ff\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Fri Jul 19 17:59:08 2019 +0200\n\n drm/tinydrm: Use spi_is_bpw_supported()\n \n This means that tinydrm_spi_bpw_supported() can be removed.\n \n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190719155916.62465-4-noralf@tronnes.org\n\ncommit e6652156a31498c63533a6353bcaeec8d0904d26\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Fri Jul 19 17:59:07 2019 +0200\n\n drm/tinydrm: Use DRM_MODE_CONNECTOR_SPI\n \n tinydrm drivers announce DRM_MODE_CONNECTOR_VIRTUAL for its SPI drivers.\n Use the new SPI connector type instead.\n \n X server will now list the connector as Unknown instead of Virtual:\n \n X.Org X Server 1.19.2\n Release Date: 2017-03-02\n \u003c...\u003e\n [ 53523.905] (II) modeset(0): Output Unknown19-1 has no monitor section\n [ 53523.908] (II) modeset(0): EDID for output Unknown19-1\n [ 53523.910] (II) modeset(0): Printing probed modes for output Unknown19-1\n [ 53523.911] (II) modeset(0): Modeline \"320x240\"x0.0 0.00 320 320 320 320 240 240 240 240 (0.0 kHz eP)\n [ 53523.911] (II) modeset(0): Output Unknown19-1 connected\n [ 53523.912] (II) modeset(0): Using exact sizes for initial modes\n [ 53523.912] (II) modeset(0): Output Unknown19-1 using initial mode 320x240 +0+0\n \n I won\u0027t chase down and fix userspace, but the new connector type will\n trickle out to userspace eventually.\n \n v2: Split patch in core and driver changes, expand commit message (Daniel)\n \n Cc: David Lechner \u003cdavid@lechnology.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: David Lechner \u003cdavid@lechnology.com\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190719155916.62465-3-noralf@tronnes.org\n\ncommit fc06bf1d76d67ac455879107c4fbe281aaf93c14\nAuthor: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\nDate: Fri Jul 19 17:59:06 2019 +0200\n\n drm: Add SPI connector type\n \n tinydrm drivers announce DRM_MODE_CONNECTOR_VIRTUAL for its SPI drivers.\n Add a SPI connector type to match the actual connector.\n \n X will list the connector as Unknown:\n \n X.Org X Server 1.19.2\n Release Date: 2017-03-02\n \u003c...\u003e\n [ 53523.905] (II) modeset(0): Output Unknown19-1 has no monitor section\n [ 53523.908] (II) modeset(0): EDID for output Unknown19-1\n [ 53523.910] (II) modeset(0): Printing probed modes for output Unknown19-1\n [ 53523.911] (II) modeset(0): Modeline \"320x240\"x0.0 0.00 320 320 320 320 240 240 240 240 (0.0 kHz eP)\n [ 53523.911] (II) modeset(0): Output Unknown19-1 connected\n [ 53523.912] (II) modeset(0): Using exact sizes for initial modes\n [ 53523.912] (II) modeset(0): Output Unknown19-1 using initial mode 320x240 +0+0\n \n The weston source shows that it will be listed as UNNAMED.\n \n v2: Split patch in core and driver changes, expand commit message (Daniel)\n \n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: David Lechner \u003cdavid@lechnology.com\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190719155916.62465-2-noralf@tronnes.org\n\ncommit fa01aff97637a961041ef7e6279cbd32514622b6\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Wed Jul 3 17:44:20 2019 +0300\n\n pinctrl: merrifield: Use devm_platform_ioremap_resource()\n \n Use the new helper that wraps the calls to platform_get_resource()\n and devm_ioremap_resource() together.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n\ncommit 9d5b6a9520ad854c271832b634b9016443a140d6\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Wed Jul 3 17:44:20 2019 +0300\n\n pinctrl: intel: Use devm_platform_ioremap_resource()\n \n Use the new helper that wraps the calls to platform_get_resource()\n and devm_ioremap_resource() together.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n\ncommit a86f12b2770a5bc25ea721f976a6cc5c55d6f481\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Wed Jul 3 17:44:20 2019 +0300\n\n pinctrl: cherryview: Use devm_platform_ioremap_resource()\n \n Use the new helper that wraps the calls to platform_get_resource()\n and devm_ioremap_resource() together.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n\ncommit 24025010d1b9d49bda11727f068a9dae381db6f8\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Wed Jul 3 17:44:20 2019 +0300\n\n pinctrl: baytrail: Use devm_platform_ioremap_resource()\n \n Use the new helper that wraps the calls to platform_get_resource()\n and devm_ioremap_resource() together.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Acked-by: Mika Westerberg \u003cmika.westerberg@linux.intel.com\u003e\n\ncommit fa564e90257e5aa5b69a0c70fe59a91da3547897\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jul 17 05:03:44 2019 -0400\n\n media: ov7670: don\u0027t return ENOTTY if SUBDEV_API is not set\n \n If CONFIG_VIDEO_V4L2_SUBDEV_API is not set, then it is still possible\n to call set_fmt for V4L2_SUBDEV_FORMAT_TRY, the result is just not\n stored. So return 0 instead of -ENOTTY.\n \n Calling get_fmt with V4L2_SUBDEV_FORMAT_TRY should return -EINVAL\n instead of -ENOTTY, after all the get_fmt functionality is still\n present, just not supported for TRY.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 4965bcd3c1f3d3fcade75bb8693fc1a97c7b2773\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jul 17 05:03:43 2019 -0400\n\n media: via-camera: use struct v4l2_fh\n \n Modern V4L2 drivers should use struct v4l2_fh to represent a filehandle.\n This driver was one of the few that didn\u0027t use it.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Reviewed-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 9230a4cc316b241e641565c377a415cf5d911425\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jul 17 05:03:42 2019 -0400\n\n media: via-camera: call viafb_pm_unregister in remove()\n \n The power management hooks were never unregistered, which caused a\n crash when unloading the module.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit ba78170ef15371276c22255a88453c2f15529130\nAuthor: Joe Perches \u003cjoe@perches.com\u003e\nDate: Thu Jul 4 19:57:44 2019 -0400\n\n media: go7007: Fix misuse of strscpy\n \n Probable cut\u0026paste typo - use the correct field size.\n \n Signed-off-by: Joe Perches \u003cjoe@perches.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit f2180adfd7a3d0715726474411bdf925010d4c29\nAuthor: Joe Perches \u003cjoe@perches.com\u003e\nDate: Thu Jul 4 19:57:43 2019 -0400\n\n media: m2m-deinterlace: Fix misuse of strscpy\n \n Probable cut\u0026paste typo - use the correct field size.\n \n Signed-off-by: Joe Perches \u003cjoe@perches.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit ed2cea4f95bf2d091061e2b89b22912fff575a14\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Wed Jul 3 04:00:35 2019 -0400\n\n media: cobalt: remove redundant assignment to variable data\n \n The variable data is being initialized with a value that is never\n read and it is being updated later with a new value. The\n initialization is redundant and can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n \n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit aef5f47c9907ed6bfe9245124279b9c049077fd7\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Mon Jul 1 06:47:02 2019 -0400\n\n media: platform: meson-ao-cec-g12a: add support for SM1\n \n Add support for the Amlogic SM1 SoC Family to the G12A AO-CECB\n derivative.\n \n It only adds a single init register.\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n [hverkuil-cisco@xs4all.nl: dropped spurious newline]\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 9bef0d1d053de8db30bc07aa132c0ee94a05609c\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Mon Jul 1 06:47:01 2019 -0400\n\n media: dt-bindings: media: meson-ao-cec: add SM1 compatible\n \n Add AO-CEC compatible string for the Amlogic SM1 SoC family,\n a derivate of the G12A AO-CECB controller.\n \n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 31bd0e5e61dce5cf4b95dd9f463d7f88f2df08b4\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Jun 28 06:51:05 2019 -0400\n\n media: davinci-vpbe: remove obsolete includes\n \n The driver builds fine without these, and they cause build\n problems once davinci multiplatform support is enabled.\n \n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Acked-by: Lad, Prabhakar \u003cprabhakar.csengg@gmail.com\u003e\n Acked-by: Sekhar Nori \u003cnsekhar@ti.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit aa9c2182c45421d54ed27c2a1765f7adedce291b\nAuthor: Vandana BN \u003cbnvandana@gmail.com\u003e\nDate: Thu Jun 27 04:26:43 2019 -0400\n\n media: vivid:add sanity check to avoid divide error and set value to 1 if 0.\n \n Syzbot reported divide error in vivid_thread_vid_cap, which has been\n seen only once and does not have a reproducer.\n This patch adds sanity checks for the\n denominator value with WARN_ON if it is 0 and replaces it with 1.\n \n divide error: 0000 [#1] PREEMPT SMP KASAN\n kobject: \u0027tx-0\u0027 (0000000017161f7f): kobject_uevent_env\n CPU: 0 PID: 23689 Comm: vivid-003-vid-c Not tainted 5.0.0-rc4+ #58\n Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS\n Google 01/01/2011\n RIP: 0010:vivid_cap_update_frame_period\n drivers/media/platform/vivid/vivid-kthread-cap.c:661 [inline]\n RIP: 0010:vivid_thread_vid_cap+0x221/0x284\n drivers/media/platform/vivid/vivid-kthread-cap.c:789\n Code: 48 c1 e9 03 0f b6 0c 11 48 89 f2 48 69 c0 00 ca 9a 3b 83 c2 03 38\n ca\n 7c 08 84 c9 0f 85 f0 1e 00 00 41 8b 8f 24 64 00 00 31 d2 \u003c48\u003e f7 f1 49\n 89\n c4 48 89 c3 49 8d 87 28 64 00 00 48 89 c2 48 89 45\n RSP: 0018:ffff88808b4afd68 EFLAGS: 00010246\n kobject: \u0027tx-0\u0027 (0000000017161f7f): fill_kobj_path: path\n \u003d \u0027/devices/virtual/net/gre0/queues/tx-0\u0027\n RAX: 000000de5a6f8e00 RBX: 0000000100047b22 RCX: 0000000000000000\n RDX: 0000000000000000 RSI: 0000000000000004 RDI: 0000000000000004\n RBP: ffff88808b4aff00 R08: ffff88804862e1c0 R09: ffffffff89997008\n R10: ffffffff89997010 R11: 0000000000000001 R12: 00000000fffffffc\n R13: ffff8880a17e0500 R14: ffff88803e40f760 R15: ffff8882182b0140\n FS: 0000000000000000(0000) GS:ffff8880ae800000(0000)\n knlGS:0000000000000000\n CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n CR2: 00000000004cdc90 CR3: 000000005d827000 CR4: 00000000001426f0\n Call Trace:\n kobject: \u0027gretap0\u0027 (00000000d7549098): kobject_add_internal: parent:\n \u0027net\u0027,\n set: \u0027devices\u0027\n kobject: \u0027loop2\u0027 (0000000094ed4ee4): kobject_uevent_env\n kobject: \u0027loop2\u0027 (0000000094ed4ee4): fill_kobj_path: path\n \u003d \u0027/devices/virtual/block/loop2\u0027\n kthread+0x357/0x430 kernel/kthread.c:246\n kobject: \u0027gretap0\u0027 (00000000d7549098): kobject_uevent_env\n ret_from_fork+0x3a/0x50 arch/x86/entry/entry_64.S:352\n Modules linked in:\n kobject: \u0027gretap0\u0027 (00000000d7549098): fill_kobj_path: path\n \u003d \u0027/devices/virtual/net/gretap0\u0027\n ---[ end trace bc5c8b25b64d768f ]---\n kobject: \u0027loop1\u0027 (0000000032036b86): kobject_uevent_env\n RIP: 0010:vivid_cap_update_frame_period\n drivers/media/platform/vivid/vivid-kthread-cap.c:661 [inline]\n RIP: 0010:vivid_thread_vid_cap+0x221/0x2840\n drivers/media/platform/vivid/vivid-kthread-cap.c:789\n kobject: \u0027loop1\u0027 (0000000032036b86): fill_kobj_path: path\n \u003d \u0027/devices/virtual/block/loop1\u0027\n Code: 48 c1 e9 03 0f b6 0c 11 48 89 f2 48 69 c0 00 ca 9a 3b 83 c2 03 38\n ca\n 7c 08 84 c9 0f 85 f0 1e 00 00 41 8b 8f 24 64 00 00 31 d2 \u003c48\u003e f7 f1 49\n 89\n c4 48 89 c3 49 8d 87 28 64 00 00 48 89 c2 48 89 45\n kobject: \u0027loop0\u0027 (00000000dd9927c3): kobject_uevent_env\n RSP: 0018:ffff88808b4afd68 EFLAGS: 00010246\n RAX: 000000de5a6f8e00 RBX: 0000000100047b22 RCX: 0000000000000000\n kobject: \u0027queues\u0027 (000000007ed20666): kobject_add_internal:\n parent: \u0027gretap0\u0027, set: \u0027\u003cNULL\u003e\u0027\n RDX: 0000000000000000 RSI: 0000000000000004 RDI: 0000000000000004\n RBP: ffff88808b4aff00 R08: ffff88804862e1c0 R09: ffffffff89997008\n kobject: \u0027loop0\u0027 (00000000dd9927c3): fill_kobj_path: path\n \u003d \u0027/devices/virtual/block/loop0\u0027\n R10: ffffffff89997010 R11: 0000000000000001 R12: 00000000fffffffc\n kobject: \u0027queues\u0027 (000000007ed20666): kobject_uevent_env\n R13: ffff8880a17e0500 R14: ffff88803e40f760 R15: ffff8882182b0140\n FS: 0000000000000000(0000) GS:ffff8880ae800000(0000)\n knlGS:0000000000000000\n kobject: \u0027loop5\u0027 (00000000a41f9e79): kobject_uevent_env\n CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n kobject: \u0027queues\u0027 (000000007ed20666): kobject_uevent_env: filter\n function\n caused the event to drop!\n CR2: 00000000004cdc90 CR3: 000000005d827000 CR4: 00000000001426f0\n kobject: \u0027loop5\u0027 (00000000a41f9e79): fill_kobj_path: path\n \u003d \u0027/devices/virtual/block/loop5\u0027\n \n Reported-by: syz...@syzkaller.appspotmail.com\n Signed-off-by: Vandana BN \u003cbnvandana@gmail.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit b98fd3cb6db8dbba6377c41a66c6b673f9d93da3\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Thu Jun 27 04:05:41 2019 -0400\n\n media: vivid: fix potential integer overflow on left shift\n \n There is a potential integer overflow when int 2 is left shifted\n as this is evaluated using 32 bit arithmetic but is being used in\n a context that expects an expression of type s64. Fix this by\n generating a mask using GENMASK to avoid a 32 bit overflow.\n \n Addresses-Coverity: (\"Unintentional integer overflow\")\n \n Fixes: 8a99e9faa131 (\"media: vivid: add HDMI (dis)connect RX emulation\")\n Fixes: 79a792dafac6 (\"media: vivid: add HDMI (dis)connect TX emulation\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit b61789f53d49db8ca1f6b8fed411b317b4e8f044\nAuthor: Hans Verkuil \u003chverkuil@xs4all.nl\u003e\nDate: Wed Jun 26 05:58:02 2019 -0400\n\n media: drivers/media: don\u0027t set pix-\u003epriv \u003d 0\n \n The priv field of struct v4l2_pix_format shouldn\u0027t be set by drivers,\n it\u0027s set by the v4l2 core instead to V4L2_PIX_FMT_PRIV_MAGIC.\n \n Drop this from the few media drivers that still do this.\n \n Note that the gspca patch is slightly more involved since some of the\n sub-gspca drivers use the priv field internally.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n [hverkuil-cisco@xs4all.nl: fix clash in gspca between priv arg and priv variable]\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 7708065f5d51a9ae80df29d1257d91fc6a6e50a0\nAuthor: Michael Tretter \u003cm.tretter@pengutronix.de\u003e\nDate: Thu Jun 27 08:44:33 2019 -0400\n\n media: v4l2-mem2mem: reorder checks in v4l2_m2m_poll()\n \n When reaching the end of stream, V4L2 m2m clients may expect the\n V4L2_EOS_EVENT. Although the V4L2_EOS_EVENT is deprecated behavior,\n drivers must signal that event before dequeuing the buffer that has the\n V4L2_BUF_FLAG_LAST flag set.\n \n If a driver queues the V4L2_EOS_EVENT event and returns the buffer after\n the check for events but before the check for buffers, vb2_m2m_poll()\n will signal that the buffer with V4L2_BUF_FLAG_LAST can be read but not\n that the V4L2_EOS_EVENT is available.\n \n Split the check for buffers into a separate function and check for\n available buffers before checking for events. This ensures that if\n vb2_m2m_poll() signals POLLIN | POLLRDNORM for the V4L2_BUF_FLAG_LAST\n buffer, it signals POLLPRI for the V4L2_EOS_EVENT, too.\n \n Signed-off-by: Michael Tretter \u003cm.tretter@pengutronix.de\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n [hverkuil-cisco@xs4all.nl: fix checkpatch alignment warning]\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 8d86a15649957c182e90fa2b1267c16699bc12f1\nAuthor: Michael Tretter \u003cm.tretter@pengutronix.de\u003e\nDate: Thu Jun 27 08:44:32 2019 -0400\n\n media: vb2: reorder checks in vb2_poll()\n \n When reaching the end of stream, V4L2 clients may expect the\n V4L2_EOS_EVENT before being able to dequeue the last buffer, which has\n the V4L2_BUF_FLAG_LAST flag set.\n \n If the vb2_poll() function first checks for events and afterwards if\n buffers are available, a driver can queue the V4L2_EOS_EVENT event and\n return the buffer after the check for events but before the check for\n buffers. This causes vb2_poll() to signal that the buffer with\n V4L2_BUF_FLAG_LAST can be read without the V4L2_EOS_EVENT being\n available.\n \n First, check for available buffers and afterwards for events to ensure\n that if vb2_poll() signals POLLIN | POLLRDNORM for the\n V4L2_BUF_FLAG_LAST buffer, it also signals POLLPRI for the\n V4L2_EOS_EVENT.\n \n Signed-off-by: Michael Tretter \u003cm.tretter@pengutronix.de\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 43266ad2b47db87b91f3198afa5ad61a4206c253\nAuthor: Hans Verkuil \u003chverkuil@xs4all.nl\u003e\nDate: Wed Jun 26 05:52:16 2019 -0400\n\n media: input/touchscreen/sur40: use COLORSPACE_RAW\n \n This driver set the colorspace to SRGB, but that makes no sense for\n a touchscreen. Use RAW instead. This also ensures consistency with the\n v4l_pix_format_touch() call that\u0027s used in v4l2-ioctl.c.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Acked-by: Florian Echtler \u003cfloe@butterbrot.org\u003e\n Acked-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 999a4312e871d9674f2ac8734820b8a242a0783f\nAuthor: Hans Verkuil \u003chverkuil@xs4all.nl\u003e\nDate: Wed Jun 26 05:48:18 2019 -0400\n\n media: v4l2-ioctl: call v4l_pix_format_touch() for TRY_FMT\n \n The function v4l_pix_format_touch() is called for S_FMT to set\n v4l2_pix_format fields to default values for a v4l-touch device,\n but it wasn\u0027t called for TRY_FMT. Add this.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit ebe15c7679680308268b99d911b1db15d514c7b8\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Mon Jul 15 06:09:06 2019 -0400\n\n media: tegra-cec: use cec_notifier_cec_adap_(un)register\n \n Use cec_notifier_cec_adap_(un)register instead of\n cec_notifier_get, cec_notifier_put and cec_register_cec_notifier.\n \n Also enable the CEC_CAP_CONNECTOR_INFO capability.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 72a171d72aa66b3d0b52889a08effd73fb3e5c5f\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Mon Jul 15 06:08:53 2019 -0400\n\n media: stih-cec: use cec_notifier_cec_adap_(un)register\n \n Use cec_notifier_cec_adap_(un)register instead of\n cec_notifier_get, cec_notifier_put and cec_register_cec_notifier.\n \n Also enable the CEC_CAP_CONNECTOR_INFO capability.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Cc: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit ca3cc03b7138c4162895318e6ae915cf26eea763\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Mon Jul 15 06:08:35 2019 -0400\n\n media: s5p-cec: use cec_notifier_cec_adap_(un)register\n \n Use cec_notifier_cec_adap_(un)register instead of\n cec_notifier_get, cec_notifier_put and cec_register_cec_notifier.\n \n Also enable the CEC_CAP_CONNECTOR_INFO capability.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Cc: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n Cc: Sylwester Nawrocki \u003csnawrocki@kernel.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 276c821f77317211fbf15c0ebf24e26ebf6bf985\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Mon Jul 15 06:07:07 2019 -0400\n\n media: seco-cec: use cec_notifier_cec_adap_(un)register\n \n Use cec_notifier_cec_adap_(un)register instead of\n cec_notifier_get_conn, cec_notifier_put and cec_register_cec_notifier.\n \n Also enable the CEC_CAP_CONNECTOR_INFO capability.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Cc: Ettore Chimenti \u003cek5.chimenti@gmail.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit e6777689406545611d1bda4f716a56a3961bc7f8\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Mon Jul 15 06:04:30 2019 -0400\n\n media: cros-ec-cec: use cec_notifier_cec_adap_(un)register\n \n Use cec_notifier_cec_adap_(un)register instead of\n cec_notifier_get_conn, cec_notifier_put and cec_register_cec_notifier.\n \n Also enable the CEC_CAP_CONNECTOR_INFO capability.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Cc: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 20958f9d5c48bf1606cdd7f00d0c1c6cd653b758\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Mon Jun 24 11:01:20 2019 -0400\n\n media: meson/ao-cec: use cec_notifier_cec_adap_(un)register\n \n Use the new cec_notifier_cec_adap_(un)register() functions to\n (un)register the notifier for the CEC adapter.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Cc: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 967b109096b2c2b7ef45a3fce44908efe05779a9\nAuthor: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nDate: Tue Jul 23 10:07:40 2019 +1000\n\n media: uapi: new file needs types.h\n \n Today\u0027s linux-next build (x86_64 allmodconfig) failed like this:\n \n include/media/vp8-ctrls.h:25:2: error: unknown type name \u0027__s8\u0027\n __s8 quant_update[4];\n ^~~~\n ...\n include/media/vp8-ctrls.h:107:2: error: unknown type name \u0027__u64\u0027\n __u64 flags;\n ^~~~~\n \n Caused by commit a57d6acaf352 (\"media: uapi: Add VP8 stateless decoder API\")\n \n Fixes: a57d6acaf352 (\"media: uapi: Add VP8 stateless decoder API\")\n Signed-off-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 3b8720e63f4a1fc6f422a49ecbaa3b59c86d5aaf\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Sun Jul 21 22:19:56 2019 +0200\n\n fbdev: Ditch fb_edid_add_monspecs\n \n It\u0027s dead code ever since\n \n commit 34280340b1dc74c521e636f45cd728f9abf56ee2\n Author: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n Date: Fri Dec 4 17:01:43 2015 +0100\n \n fbdev: Remove unused SH-Mobile HDMI driver\n \n Also with this gone we can remove the cea_modes db. This entire thing\n is massively incomplete anyway, compared to the CEA parsing that\n drm_edid.c does.\n \n Acked-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n Cc: Tavis Ormandy \u003ctaviso@gmail.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Signed-off-by: Bartlomiej Zolnierkiewicz \u003cb.zolnierkie@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190721201956.941-1-daniel.vetter@ffwll.ch\n\ncommit bdd50d7421b2f8fd99f953e1f747e0cb3f3bed64\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Tue Jul 23 16:44:15 2019 +0900\n\n x86/bitops: Use __builtin_constant_p() directly instead of IS_IMMEDIATE()\n \n __builtin_constant_p(nr) is used everywhere now. It does not make much\n sense to define IS_IMMEDIATE() as its alias.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190723074415.26811-1-yamada.masahiro@socionext.com\n\ncommit 701010532164eaacd415ec5683717da03f4b822d\nAuthor: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nDate: Tue Jul 23 20:26:45 2019 +0900\n\n x86/build: Remove unneeded uapi asm-generic wrappers\n \n These are listed in include/uapi/asm-generic/Kbuild, so Kbuild will\n automatically generate them.\n \n Signed-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190723112646.14046-1-yamada.masahiro@socionext.com\n\ncommit 03b77f0b8587a9a0f9d2f1503da3d120aa6fe730\nAuthor: Axel Lin \u003caxel.lin@ingics.com\u003e\nDate: Tue Jul 23 09:41:02 2019 +0800\n\n regulator: stm32-booster: Remove .min_uV and .list_voltage for fixed regulator\n \n Setting .n_voltages \u003d 1 and .fixed_uV is enough for fixed regulator,\n remove the redundant .min_uV and .list_voltage settings.\n \n Signed-off-by: Axel Lin \u003caxel.lin@ingics.com\u003e\n Acked-by: Fabrice Gasnier \u003cfabrice.gasnier@st.com\u003e\n Link: https://lore.kernel.org/r/20190723014102.25103-1-axel.lin@ingics.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 2d09a79bf637f91d1bbfcfd4520e3639dd15897c\nAuthor: Maarten ter Huurne \u003cmaarten@treewalker.org\u003e\nDate: Mon Jul 22 21:14:18 2019 -0400\n\n regulator: act8865: Add support for act8600 charger\n \n This provides a way to monitor battery charge status via the power\n supply subsystem.\n \n Signed-off-by: Maarten ter Huurne \u003cmaarten@treewalker.org\u003e\n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Link: https://lore.kernel.org/r/20190723011418.29143-1-paul@crapouillou.net\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit a789adae592e3d28ae87902c0012a8b11c96eb79\nAuthor: Shuming Fan \u003cshumingf@realtek.com\u003e\nDate: Tue Jul 23 18:12:47 2019 +0800\n\n ASoC: rt1308: Convert headers to SPDX\n \n Convert include and source files to SPDX.\n \n Signed-off-by: Shuming Fan \u003cshumingf@realtek.com\u003e\n Link: https://lore.kernel.org/r/20190723101247.19741-1-shumingf@realtek.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit f90aa354be7bffaec2b440eb1831c429ecb1a5e2\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Mon Jul 22 23:26:39 2019 +0200\n\n ASoC: rt5665: Fix a typo in the name of a function\n \n All function names start with rt5665_, except \u0027rt5655_set_verf()\u0027.\n It is likely a typo.\n \n Fix it to be consistent.\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Link: https://lore.kernel.org/r/20190722212639.26954-1-christophe.jaillet@wanadoo.fr\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 810dbea3656912d6ad8db691a9a4b2ad97d4b736\nAuthor: Rander Wang \u003crander.wang@linux.intel.com\u003e\nDate: Mon Jul 22 09:14:00 2019 -0500\n\n ASoC: SOF: Intel: hda: fix stream id setting\n \n snd_hdac_ext_link_clear_stream_id maps stream id to\n link output, which is for playback, not capture.\n \n Tested on Whiskey Lake platform.\n \n Signed-off-by: Rander Wang \u003crander.wang@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190722141402.7194-20-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 934bf82203d679a8f2ff22817e341bc059f5f7fc\nAuthor: Rander Wang \u003crander.wang@linux.intel.com\u003e\nDate: Mon Jul 22 09:13:59 2019 -0500\n\n ASoC: SOF: Intel: hda: fix link DMA config\n \n For this bug, there are two capture pcm streams active, with one\n stream and its related stream tag released before suspend. Later\n when system suspend is done, the stream tag for the remaining\n active stream is released by SOF driver. After system resume, hda\n codec driver restores the stream tag for the active pcm stream,\n but SOF goes to assign a new one, which now doesn\u0027t match with the\n stream tag used by codec driver, and this causes DMA to fail\n receiving data, leading to unrecoverable XRUN condition in FW.\n \n For stream tag is stored in both hda codec and SOF driver, it\n shouldn\u0027t be released only in SOF driver. This patch just keeps the\n stream information in dma data and checks whether there is a stored\n DMA data for stream resuming from S3 and restores it. And it also\n removes DMA data when the stream is released.\n \n Tested on Whiskey Lake platform.\n \n GitHub issue: https://github.com/thesofproject/sof/issues/1594\n Signed-off-by: Rander Wang \u003crander.wang@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190722141402.7194-19-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 7ccafa2b3879612ab9ed88bb6664bd4085957186\nAuthor: Shengjiu Wang \u003cshengjiu.wang@nxp.com\u003e\nDate: Thu Jul 11 18:49:46 2019 +0800\n\n ASoC: fsl_esai: recover the channel swap after xrun\n \n There is chip errata ERR008000, the reference doc is\n (https://www.nxp.com/docs/en/errata/IMX6DQCE.pdf),\n \n The issue is \"While using ESAI transmit or receive and\n an underrun/overrun happens, channel swap may occur.\n The only recovery mechanism is to reset the ESAI.\"\n \n This issue exist in imx3/imx5/imx6(partial) series.\n \n In this commit add a tasklet to handle reset of ESAI\n after xrun happens to recover the channel swap.\n \n Signed-off-by: Shengjiu Wang \u003cshengjiu.wang@nxp.com\u003e\n Acked-by: Nicolin Chen \u003cnicoleotsuka@gmail.com\u003e\n Link: https://lore.kernel.org/r/326035cb99975361699d9ed748054b08bc06a341.1562842206.git.shengjiu.wang@nxp.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit d8bf0e7627e6e887bd0ed3707216e0e69ec95710\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jul 23 10:58:00 2019 +0100\n\n drm/i915/selftests: Let igt_vma_partial et al breathe\n \n Give the scheduler a chance to breathe by calling cond_resched() as some\n of the loops may take some time on slower machines, and so catch the\n attention of the watchdogs.\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111196\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.william.auld@gmail.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.william.auld@gmail.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190723095800.2820-1-chris@chris-wilson.co.uk\n\ncommit 6298b78742be6593d372ed1b5bfa5397e1393595\nAuthor: Janusz Jankowski \u003cjanusz.jankowski@linux.intel.com\u003e\nDate: Mon Jul 22 09:14:02 2019 -0500\n\n ASoC: SOF: Intel: ssp: BCLK delay parameter\n \n Some codecs require BCLK to be on for some time, before sending\n any data. SOF can enable BCLK and then wait for guaranteed time,\n before starting DMA on SSP start.\n \n Signed-off-by: Janusz Jankowski \u003cjanusz.jankowski@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190722141402.7194-22-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit a3ebccb52efdfb3459fa51516238cc464ee310c9\nAuthor: Kai Vehmanen \u003ckai.vehmanen@linux.intel.com\u003e\nDate: Mon Jul 22 09:13:58 2019 -0500\n\n ASoC: SOF: Intel: hda: reset link DMA state in prepare\n \n When application goes through SUSPEND/STOP-\u003ePREPARE-\u003eSTART\n cycle, we should always reprogram the DAI link DMA to ensure\n it is in sync with the host PCM DMA.\n \n Use same state tracking logic to handle both restart and\n system resume flows. Use link_prepared field of\n \u0027struct hdac_ext_stream\u0027 to store the state, instead of\n adding redundant fields to SOF specific structs.\n \n Signed-off-by: Kai Vehmanen \u003ckai.vehmanen@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190722141402.7194-18-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 672ff5e3596ee27b64edcc73251f4ae1c8ab12ac\nAuthor: Guennadi Liakhovetski \u003cguennadi.liakhovetski@linux.intel.com\u003e\nDate: Mon Jul 22 09:13:57 2019 -0500\n\n ASoC: SOF: Intel: hda: add a parameter to disable MSI\n \n Enabling MSI on HDA can fail, in which case the legacy PCI IRQ mode\n will be used. To make testing this mode easier add an \"enable_msi\"\n module parameter, which is only enabled if debugging is enabled too.\n \n Signed-off-by: Guennadi Liakhovetski \u003cguennadi.liakhovetski@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190722141402.7194-17-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit dc7a36f178a94604d29c5dd15c77187905d8e882\nAuthor: Zhu Yingjiang \u003cyingjiang.zhu@linux.intel.com\u003e\nDate: Mon Jul 22 09:13:56 2019 -0500\n\n ASoC: SOF: Intel: hda: remove duplicated clear WAKESTS\n \n Remove the first clear WAKESTS, only one clear is needed during init\n chip.\n \n Signed-off-by: Zhu Yingjiang \u003cyingjiang.zhu@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190722141402.7194-16-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit d06973515f95ac74d68fb3e1f1fb4b055531296c\nAuthor: Zhu Yingjiang \u003cyingjiang.zhu@linux.intel.com\u003e\nDate: Mon Jul 22 09:13:55 2019 -0500\n\n ASoC: SOF: Intel: hda: use SOF defined init chip in resume\n \n Unify resume code by using SOF common function hda_dsp_ctrl_init_chip()\n which can handle both HDA and non-HDA cases. Move code to reset\n stream-to-link mapping into hda_dsp_ctrl_init_chip().\n \n Signed-off-by: Zhu Yingjiang \u003cyingjiang.zhu@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190722141402.7194-15-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 061edb232505797aee53a955225af23fa30d45b8\nAuthor: Zhu Yingjiang \u003cyingjiang.zhu@linux.intel.com\u003e\nDate: Mon Jul 22 09:13:54 2019 -0500\n\n ASoC: SOF: Intel: hda: set position buffer in init chip\n \n Set the HDA stream position buffer during init chip. The position buffer\n needs to be set in both HDA codec and nocodec cases. Using SOF defined\n function and move it to common code.\n \n Signed-off-by: Zhu Yingjiang \u003cyingjiang.zhu@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190722141402.7194-14-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit fd15f2f5e272145269bcbf834e0e0b560a575891\nAuthor: Rander Wang \u003crander.wang@linux.intel.com\u003e\nDate: Mon Jul 22 09:13:53 2019 -0500\n\n ASoC: SOF: Intel: hda: Enable jack detection\n \n In commit 7d4f606c50ff (\"ALSA: hda - WAKEEN feature enabling for\n runtime pm\"), legacy HD-A driver sets hda controller in reset mode after\n entering runtime-suspend. And when resuming from suspend mode, it checks\n hda controller \u0026 codec status to detect headphone hotplug event. Now\n this patch does the same job in SOF runtime pm functions.\n \n And we need to check all the non-hdmi codecs for some cases like playback\n with HDMI or capture with DMIC connected to dsp. In these cases, only\n controller is active and codecs are suspended, so codecs can\u0027t send\n unsolicited event to controller. The jack polling operation will activate\n codecs and unsolicited event can work even codecs become suspended later.\n \n Tested on whiskylake with hda codecs.\n \n Signed-off-by: Rander Wang \u003crander.wang@linux.intel.com\u003e\n Signed-off-by: Guennadi Liakhovetski \u003cguennadi.liakhovetski@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190722141402.7194-13-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 6aa232e1ccfcd2b55306b86803dc547c36f8ad15\nAuthor: Rander Wang \u003crander.wang@linux.intel.com\u003e\nDate: Mon Jul 22 09:13:52 2019 -0500\n\n ASoC: SOF: Intel: hda: reduce ifdef usage for hda\n \n Move the code for hda to one point\n \n Signed-off-by: Rander Wang \u003crander.wang@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190722141402.7194-12-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 184fdfca4ba63d57a668564639a4dcf188fc9822\nAuthor: Keyon Jie \u003cyang.jie@linux.intel.com\u003e\nDate: Mon Jul 22 09:13:51 2019 -0500\n\n ASoC: SOF: Intel: hda: correct ROM state mask\n \n The ROM state is represented by the 24 LSB bits in the ROM status\n register, so the mask should be 0xffffff instead of 0xf.\n \n Signed-off-by: Keyon Jie \u003cyang.jie@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190722141402.7194-11-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 1c38c9223da3af619d35f052ad6f4bbaa0f08ac2\nAuthor: Fred Oh \u003cfred.oh@linux.intel.com\u003e\nDate: Mon Jul 22 09:13:50 2019 -0500\n\n ASoC: SOF: remove unused state variable in suspend function\n \n Remove unused and no plan to use variable from suspend function.\n \n Signed-off-by: Fred Oh \u003cfred.oh@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190722141402.7194-10-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 441c58cf8e5e0c594b3d1b489e7810ec7248820d\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Mon Jul 22 09:13:49 2019 -0500\n\n ASoC: SOF: loader: Don\u0027t ignore SRAM block types\n \n On i.MX8 data/heap/stack is kept in System RAM so\n do not ignore SRAM block types received from FW.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190722141402.7194-9-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 7198879ef57775df3b1a75c62bb9e313c56876d4\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Mon Jul 22 09:13:48 2019 -0500\n\n ASoC: SOF: loader: Use the BAR provided by FW\n \n Make sure to use the newly introduced function snd_sof_dsp_get_bar_index\n that converts the section type to appropriate BAR index.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190722141402.7194-8-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit ce8234a6357ec8f81b977d8536ff941dbd4162f3\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Mon Jul 22 09:13:47 2019 -0500\n\n ASoC: SOF: Introduce snd_sof_dsp_get_bar_index ops\n \n FW encapsulates information about section types (e.g DRAM, IRAM)\n inside module block header. This information can be used in order\n to correctly load the section to the appropriate place in memory.\n \n SOF Linux driver needs to know for each platform how to map the\n section type with the corresponding memory BAR. So, this patch\n introduces get_bar_index, a new operation inside snd_sof_dsp_ops.\n \n Intel platforms, usually load all the section in a contiguous memory\n area (usually denoted by sdev-\u003emmio_bar) so things are relatively\n simple there. Anyhow, on i.MX8 IRAM and DRAM for example are mapped\n to distinct BARs.\n \n By default, if no get_bar function is provided the core implementation\n will always return sdev-\u003emmio_bar so that there will be no need for\n a change to existing Intel code.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190722141402.7194-7-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c9b54a3360166ca2fec11d55c86b5e685e56bab8\nAuthor: Kai Vehmanen \u003ckai.vehmanen@linux.intel.com\u003e\nDate: Mon Jul 22 09:13:46 2019 -0500\n\n ASoC: SOF: core: increase default IPC timeouts\n \n Increase the default timeout values for boot (100ms to 2sec) and\n IPC message sending (5ms to 500ms). The values should be overridden\n with values from platform data.\n \n There is no functional need to have such short timeouts as both boot\n and IPC send errors are considered fatal errors. More relaxed timeouts\n are convenient when running the driver on top of emulation such as QEMU.\n \n Signed-off-by: Kai Vehmanen \u003ckai.vehmanen@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190722141402.7194-6-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 6dd78b310b8d4ab5d0b851e369b5b3d46c6a32a9\nAuthor: Kai Vehmanen \u003ckai.vehmanen@linux.intel.com\u003e\nDate: Mon Jul 22 09:13:45 2019 -0500\n\n ASoC: SOF: ipc: use timeout configured at probe\n \n Do not hardcode IPC timeout value in ipc.c, but rather use the timeout\n value configured during device probe. For platforms that do not override\n the IPC timeout, default value TIMEOUT_DEFAULT_IPC_MS has already been\n defined in core.c.\n \n Signed-off-by: Kai Vehmanen \u003ckai.vehmanen@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190722141402.7194-5-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit a49b687192752bd373f33551520cc98389460ad0\nAuthor: Kai Vehmanen \u003ckai.vehmanen@linux.intel.com\u003e\nDate: Mon Jul 22 09:13:44 2019 -0500\n\n ASoC: SOF: use common code to send PCM_FREE IPC\n \n Remove duplicated code by using a common helper function\n to send the PCM_FREE IPC message to FW.\n \n Signed-off-by: Kai Vehmanen \u003ckai.vehmanen@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190722141402.7194-4-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 04c8027764bc82a325d3abc6f39a6a4642a937cb\nAuthor: Kai Vehmanen \u003ckai.vehmanen@linux.intel.com\u003e\nDate: Mon Jul 22 09:13:43 2019 -0500\n\n ASoC: SOF: reset DMA state in prepare\n \n When application goes through SUSPEND/STOP-\u003ePREPARE-\u003eSTART\n cycle, we should always reprogram the SOF device to start\n DMA from a known state so that hw_ptr/appl_ptrs remain valid.\n This is expected by ALSA core as it resets the buffer\n state as part of prepare (see snd_pcm_do_prepare()).\n \n Fix the issue by forcing reconfiguration of the FW with\n STREAM_PCM_PARAMS in prepare(). Use combined logic to handle\n prepare and the existing flow to reprogram hw-params after\n system suspend.\n \n Without the fix, first call to pcm pointer() will return\n an invalid hw_ptr and application may immediately observe XRUN\n status, unless \"start_threshold\" SW parameter is set to maximum\n value by the application.\n \n Signed-off-by: Kai Vehmanen \u003ckai.vehmanen@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190722141402.7194-3-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit f1b1b9b136827915624136624ff54aba5890a15b\nAuthor: Pan Xiuli \u003cxiuli.pan@linux.intel.com\u003e\nDate: Mon Jul 22 09:13:42 2019 -0500\n\n ASoC: SOF: pci: mark last_busy value at runtime PM init\n \n If last_busy value is not set at runtime PM enable, the device will be\n suspend immediately after usage counter is 0. Set the last_busy value to\n make sure delay is working at first boot up.\n \n Signed-off-by: Pan Xiuli \u003cxiuli.pan@linux.intel.com\u003e\n Signed-off-by: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Signed-off-by: Pierre-Louis Bossart \u003cpierre-louis.bossart@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190722141402.7194-2-pierre-louis.bossart@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit f8f85216f8d309daadb37aba8a4b0826783d8747\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Mon Jul 22 23:15:28 2019 +0200\n\n ASoC: pcm3168a: Fix a typo in the name of a constant\n \n There is a typo in PCM1368A_MAX_SYSCLK, it should be PCM3168A_MAX_SYSCLK\n (1 and 3 switched in 3168)\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Link: https://lore.kernel.org/r/20190722211528.26600-1-christophe.jaillet@wanadoo.fr\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit de6a263400f232a3787bdb112d9c974ebd29c4b4\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Tue Jul 23 10:14:04 2019 +0100\n\n drm/i915/uc: Sanitize uC when GT is sanitized\n \n The microcontrollers are part of GT so it makes logical sense to have\n them sanitized at the same time. This also fixed an issue with our\n status tracking where the FW load status is not reset around\n hibernation.\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190723091404.6449-2-chris@chris-wilson.co.uk\n\ncommit 3fcba88188a85a398c2c9245d148c952e2d53889\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Tue Jul 23 10:14:03 2019 +0100\n\n drm/i915/uc: Gt-fy uc reset\n \n This was the last place in gt/uc that was still using I915_READ\n with the global dev_priv.\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190723091404.6449-1-chris@chris-wilson.co.uk\n\ncommit aed7450b153666074d1e1d6c38f6a1ff14c7f80c\nAuthor: Anshuman Gupta \u003canshuman.gupta@intel.com\u003e\nDate: Fri Jul 19 11:25:13 2019 +0530\n\n drm/i915: Add HDCP capability info to i915_display_info.\n \n To identify the HDCP capability of the display connected to CI\n systems, we need to add the hdcp capability probing in i915_display_info.\n \n This will also help to populate the HDCP capability of the CI systems\n to CI H/W logs maintained at https://intel-gfx-ci.01.org/hardware/.\n It will facilitate to determine the kms_content_protection behavior on\n a particular CI system.\n \n v2: Reused the intel_hdcp_info() in i915_hdcp_sink_capability_show(). [Ram]\n Shifted intel_hdcp_info() to the end of intel_dp_info. [Ram]\n v3: used seq_puts() instead of seq_pritnf(). [Ram]\n \n Cc: daniel.vetter@intel.com\n Cc: ramalingam.c@intel.com\n Signed-off-by: Anshuman Gupta \u003canshuman.gupta@intel.com\u003e\n Reviewed-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190719055513.2089-1-anshuman.gupta@intel.com\n\ncommit d40113fb5f7e632232546a74aae6e6d8ca2e70e2\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Tue Jul 23 10:44:07 2019 +0200\n\n ARM: dts: sunxi: Fix the HDMI PHY name\n \n Even though the binding mentions that the PHY name must be \"phy\", it turns\n out that all our DTs had \"hdmi-phy\" instead.\n \n The code doesn\u0027t care about the phy-names property, so we can just change\n our DTs to match the binding, without any side effect.\n \n Acked-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 6f21a03b07ce9480087138e588081ed3c3c0fa1e\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Tue Jul 23 10:44:06 2019 +0200\n\n ARM: dts: sunxi: Remove simple-panel compatible\n \n simple-panel based bindings need only the display compatible, and\n simple-panel isn\u0027t documented anywhere. Remove it.\n \n Acked-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 163ce24a435061f9ff7e1673fec5be0868bdb7d3\nAuthor: Nicholas Mc Guire \u003chofrat@osadl.org\u003e\nDate: Tue Apr 16 06:22:42 2019 +0200\n\n MAINTAINERS: add soc/ux500\n \n get_maintainer.pl only returns linux-kernel@vger.kernel.org for the\n drivers/soc/ux500/ files - fix it by adding drivers/soc/ux500/ entry\n \n Signed-off-by: Nicholas Mc Guire \u003chofrat@osadl.org\u003e\n Suggested-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 428b504e84bc19aa14108186714d1f5928dcae16\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Tue Jun 9 12:05:46 2015 +0200\n\n ARM: ux500: simplify and move debug UART\n \n The debug UART code defined three different virtual addresses\n while only one is ever used. Get rid of this, and move the\n UART remapping to 0xfffe8000 where DTCM reside on some platforms\n but not on Ux500, so it can be reused moving the UART out of the\n vmalloc area.\n \n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 1a47dc0240bf177878251fae10aabccbaa5a4f20\nAuthor: Andra Danciu \u003candradanciu1997@gmail.com\u003e\nDate: Mon Jul 22 13:27:30 2019 +0300\n\n dt-bindings: arm: fsl: Add the pico-pi-imx8m board\n \n Add an entry for TechNexion PICO-PI-IMX8M board based on i.MX8MQ SoC\n Datasheet can be found at:\n https://s3.us-east-2.amazonaws.com/technexion/datasheets/picopiimx8m.pdf\n \n Cc: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Signed-off-by: Andra Danciu \u003candradanciu1997@gmail.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit c39eff747b845aadf0c75a3abde39d57205371b4\nAuthor: Manivannan Sadhasivam \u003cmanivannan.sadhasivam@linaro.org\u003e\nDate: Fri Jul 19 12:39:25 2019 +0530\n\n dt-bindings: arm: Document i.MX8QXP AI_ML board binding\n \n Document devicetree binding of i.MX8QXP AI_ML board from Einfochips.\n \n Signed-off-by: Manivannan Sadhasivam \u003cmanivannan.sadhasivam@linaro.org\u003e\n Reviewed-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 417393a5072f81cc009b00755f4ac23a9cf83987\nAuthor: Manivannan Sadhasivam \u003cmanivannan.sadhasivam@linaro.org\u003e\nDate: Fri Jul 19 12:39:24 2019 +0530\n\n dt-bindings: Add Vendor prefix for Einfochips\n \n Add devicetree vendor prefix for Einfochips.\n https://www.einfochips.com/\n \n Signed-off-by: Manivannan Sadhasivam \u003cmanivannan.sadhasivam@linaro.org\u003e\n Reviewed-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 88574a24ea12b7b8d7399e3a6b6c165f1f375323\nAuthor: Pramod Kumar \u003cpramod.kumar_1@nxp.com\u003e\nDate: Tue Jul 16 13:43:27 2019 +0000\n\n dt-bindings: arm: nxp: Add device tree binding for ls1046a-frwy board\n \n Add \"fsl,ls1046a-frwy\" bindings for ls1046afrwy board based on ls1046a SoC\n \n Signed-off-by: Vabhav Sharma \u003cvabhav.sharma@nxp.com\u003e\n Signed-off-by: Pramod Kumar \u003cpramod.kumar_1@nxp.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit d43dc52274d40a1543ca962008feb9ff784e3a49\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Thu Jul 18 13:25:19 2019 +0300\n\n firmware: imx: scu-pd: Add IRQSTR_DSP PD range\n \n The DSP interrupt steer gathers interrupts from the system\n and can be used to steer them to DSP.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Reviewed-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 590b346b58badfe87cb13a64abdc8af25e907846\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Thu Jul 18 13:25:18 2019 +0300\n\n firmware: imx: scu-pd: Add mu13 b side PD range\n \n LSIO subsystem contains 14 MU instances.\n \n 5 MUs to communicate between AP \u003c-\u003e SCU\n - side-A PD range managed by AP\n - side-B PD range managed by SCU\n \n 9 MUs to communicate between all cores (AP/M4/DSP).\n - side-A PD range managed by core-A (AP/M4/DSP)\n - side-B PD range managed by core-B (AP/M4/DSP).\n \n Communication between AP \u003c-\u003e DSP is done through the\n assigned MU number 13.\n \n So, we power up side-A by the AP and we decide to\n power up side-B also from AP. This is because powering\n it up from DSP would be painful.\n \n Powering up side B from DSP would require the DSP to\n communicate with SCU and to keep things simple we don\u0027t\n want that now.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Reviewed-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit dbc3c6295195267ea7bc48d46030c7b244f8b11e\nAuthor: Nicholas Mc Guire \u003chofrat@osadl.org\u003e\nDate: Sat Apr 13 09:14:21 2019 +0200\n\n ARM: ux500: add missing of_node_put()\n \n of_find_compatible_node() returns a pointer with refcount incremented\n so both in the error path as well as after usage in soc_info_populate()\n respectively actually b8500_read_soc_id() an explicit of_node_put is\n needed to release backupram.\n \n Signed-off-by: Nicholas Mc Guire \u003chofrat@osadl.org\u003e\n Fixes: commit 18a992787896 (\"ARM: ux500: move soc_id driver to drivers/soc\")\n Acked-by: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 6d9d21711b38c9cd5f70e04623151f498f3129ac\nAuthor: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\nDate: Thu Jul 18 13:25:17 2019 +0300\n\n firmware: imx: scu-pd: Rename mu PD range to mu_a\n \n The Messaging Unit module enables two processors within the SoC to\n communicate and coordinate by passing messages through the MU interface.\n \n MUs have 2 “sides” with independent programming interfaces. Rename\n mu PD range to mu_a because it\u0027s actually side A of MUs.\n \n Signed-off-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Reviewed-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit df7126cc42105e9df9f0182794482dc137479727\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Thu Jul 18 17:15:08 2019 +0800\n\n ARM: dts: imx6sll: move GIC to right location in DT\n \n GIC is inside of SoC from architecture perspective, it should\n be located inside of soc node in DT.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 8c1a1f4879b608f6000f98d1ea3859d9a59a2b7e\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Thu Jul 18 17:15:07 2019 +0800\n\n ARM: dts: imx6ul: move GIC to right location in DT\n \n GIC is inside of SoC from architecture perspective, it should\n be located inside of soc node in DT.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit b051589c69bf34371ebe04370685002b93ca00d1\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Thu Jul 18 17:15:06 2019 +0800\n\n ARM: dts: imx6sl: move GIC to right location in DT\n \n GIC is inside of SoC from architecture perspective, it should\n be located inside of soc node in DT.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 211ded7861466dd443e61ad29ec34ce93e5a3e38\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Thu Jul 18 17:15:05 2019 +0800\n\n ARM: dts: imx6sx: move GIC to right location in DT\n \n GIC is inside of SoC from architecture perspective, it should\n be located inside of soc node in DT.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit b605c44c30b59990e806f930c37bd288b9d901a5\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Mon Jul 15 23:52:18 2019 +0200\n\n PM: sleep: Drop dpm_noirq_begin() and dpm_noirq_end()\n \n Note that after previous changes dpm_noirq_begin() and\n dpm_noirq_end() each have only one caller, so move the code from\n them to their respective callers and drop them.\n \n Also note that dpm_noirq_resume_devices() and\n dpm_noirq_suspend_devices() need not be exported any more, so make\n them both static.\n \n This change is not expected to alter functionality.\n \n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Acked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n\ncommit 8eb0fd3b55f084320ae511cd5a64d356cf497c83\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Mon Jul 15 23:52:12 2019 +0200\n\n PM: sleep: Integrate suspend-to-idle with generig suspend flow\n \n After previous changes the suspend-to-idle code flow can be\n integrated more tightly with the generic system suspend code flow\n by making suspend_enter() call s2idle_loop() later and removing\n the direct invocations of dpm_noirq_begin(),\n dpm_noirq_suspend_devices(), dpm_noirq_end(), and\n dpm_noirq_resume_devices() from the latter, so do that.\n \n This change is not expected to alter functionality.\n \n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Acked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n\ncommit 56b991849009f5def0443bfb2f48c8321d888e15\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Mon Jul 15 23:52:03 2019 +0200\n\n PM: sleep: Simplify suspend-to-idle control flow\n \n After commit 33e4f80ee69b (\"ACPI / PM: Ignore spurious SCI wakeups\n from suspend-to-idle\") the \"noirq\" phases of device suspend and\n resume may run for multiple times during suspend-to-idle, if there\n are spurious system wakeup events while suspended. However, this\n is complicated and fragile and actually unnecessary.\n \n The main reason for doing this is that on some systems the EC may\n signal system wakeup events (power button events, for example) as\n well as events that should not cause the system to resume (spurious\n system wakeup events). Thus, in order to determine whether or not\n a given event signaled by the EC while suspended is a proper system\n wakeup one, the EC GPE needs to be dispatched and to start with that\n was achieved by allowing the ACPI SCI action handler to run, which\n was only possible after calling resume_device_irqs().\n \n However, dispatching the EC GPE this way turned out to take too much\n time in some cases and some EC events might be missed due to that, so\n commit 68e22011856f (\"ACPI: EC: Dispatch the EC GPE directly on\n s2idle wake\") started to dispatch the EC GPE right after a wakeup\n event has been detected, so in fact the full ACPI SCI action handler\n doesn\u0027t need to run any more to deal with the wakeups coming from the\n EC.\n \n Use this observation to simplify the suspend-to-idle control flow\n so that the \"noirq\" phases of device suspend and resume are each\n run only once in every suspend-to-idle cycle, which is reported to\n significantly reduce power drawn by some systems when suspended to\n idle (by allowing them to reach a deep platform-wide low-power state\n through the suspend-to-idle flow). [What appears to happen is that\n the \"noirq\" resume of devices after a spurious EC wakeup brings some\n devices into a state in which they prevent the platform from reaching\n the deep low-power state going forward, even after a subsequent\n \"noirq\" suspend phase, and on some systems the EC triggers such\n wakeups already when the \"noirq\" suspend of devices is running for\n the first time in the given suspend/resume cycle, so the platform\n cannot reach the deep low-power state at all.]\n \n First, make acpi_s2idle_wake() use the acpi_ec_dispatch_gpe() return\n value to determine whether or not the wakeup may have been triggered\n by the EC (in which case the system wakeup is canceled and ACPI\n events are processed in order to determine whether or not the event\n is a proper system wakeup one) and use rearm_wake_irq() (introduced\n by a previous change) in it to rearm the ACPI SCI for system wakeup\n detection in case the system will remain suspended.\n \n Second, drop acpi_s2idle_sync(), which is not needed any more, and\n the corresponding global platform suspend-to-idle callback.\n \n Next, drop the pm_wakeup_pending() check (which is an optimization\n only) from __device_suspend_noirq() to prevent it from returning\n errors on system wakeups occurring before the \"noirq\" phase of\n device suspend is complete (as in the case of suspend-to-idle it is\n not known whether or not these wakeups are suprious at that point),\n in order to avoid having to carry out a \"noirq\" resume of devices\n on a spurious system wakeup.\n \n Finally, change the code flow in s2idle_loop() to (1) run the\n \"noirq\" suspend of devices once before starting the loop, (2) check\n for spurious EC wakeups (via the platform -\u003ewake callback) for the\n first time before calling s2idle_enter(), and (3) run the \"noirq\"\n resume of devices once after leaving the loop.\n \n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Acked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n\ncommit 41275eb5c7181febdfaa63c3a0ad9b7acdadcd52\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Mon Jul 15 23:51:19 2019 +0200\n\n ACPI: PM: Set s2idle_wakeup earlier and clear it later\n \n The role of the s2idle_wakeup variable is to cause\n acpi_pm_wakeup_event() and acpi_pm_notify_handler() to\n increment pm_abort_suspend and trigger a wakeup from\n suspend-to-idle in case the ACPI SCI wakeup was canceled\n by acpi_s2idle_wake().\n \n However, for this purpose it need not be set in acpi_s2idle_wake()\n and cleared in acpi_s2idle_sync(), respectively. In fact, it\n may be set as early as in acpi_s2idle_prepare() and cleared as\n late as in acpi_s2idle_restore(), so do that to allow subsequent\n changes to be simpler.\n \n This change is not expected to alter functionality.\n \n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Acked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n\ncommit 2933954b71f10d392764f95eec0f0aa2d103054b\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Mon Jul 15 13:03:48 2019 +0200\n\n PM: sleep: Fix possible overflow in pm_system_cancel_wakeup()\n \n It is not actually guaranteed that pm_abort_suspend will be\n nonzero when pm_system_cancel_wakeup() is called which may lead to\n subtle issues, so make it use atomic_dec_if_positive() instead of\n atomic_dec() for the safety sake.\n \n Fixes: 33e4f80ee69b (\"ACPI / PM: Ignore spurious SCI wakeups from suspend-to-idle\")\n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Acked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n\ncommit 9089f16e053afc5e18feaeb9f64cc7c90d6bd687\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Mon Jul 15 13:03:39 2019 +0200\n\n ACPI: EC: Return bool from acpi_ec_dispatch_gpe()\n \n On some systems, if suspend-to-idle is used, the EC may signal system\n wakeup events (power button events, for example) as well as events\n that should not cause the system to resume and acpi_ec_dispatch_gpe()\n needs to be called to determine whether or not the system should\n resume then. In particular, if acpi_ec_dispatch_gpe() doesn\u0027t detect\n any EC events at all, the system should remain suspended, so it is\n useful to know when that is the case.\n \n For this reason, make acpi_ec_dispatch_gpe() return a bool value\n indicating whether or not any EC events have been detected by it.\n \n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Acked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n\ncommit 6921de898ba8f2ec91cfea70e7160b89c477382e\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Mon Jul 15 13:03:28 2019 +0200\n\n ACPICA: Return u32 from acpi_dispatch_gpe()\n \n In some cases it is useful to know whether or not the\n acpi_ev_detect_gpe() called by acpi_dispatch_gpe() has found\n the GPE to be active, so return the return value of it (whose\n data type is u32) from latter.\n \n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Acked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n\ncommit 3a79bc63d90750f737ab9d7219bd3091d2fd6d84\nAuthor: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nDate: Mon Jul 15 13:03:20 2019 +0200\n\n PCI: irq: Introduce rearm_wake_irq()\n \n Introduce a new function, rearm_wake_irq(), allowing a wakeup IRQ\n to be armed for systen wakeup detection again without running any\n action handlers associated with it after it has been armed for\n wakeup detection and triggered.\n \n That is useful for IRQs, like ACPI SCI, that may deliver wakeup\n as well as non-wakeup interrupts when armed for systen wakeup\n detection. In those cases, it may be possible to determine whether\n or not the delivered interrupt is a systen wakeup one without\n running the entire action handler (or handlers, if the IRQ is\n shared) for the IRQ, and if the interrupt turns out to be a\n non-wakeup one, the IRQ can be rearmed with the help of the\n new function.\n \n Signed-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n Acked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n\ncommit 9461e35f84184662e395ee4c0222687a8f121586\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Wed Jul 17 08:02:53 2019 -0700\n\n ARM: dts: vf610-zii-scu4-aib: Drop unused pinctrl_i2c3 pinmux config\n \n Pinctrl_i2c3 pinmux config is not used anywhere. Drop it.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit aee2a02bcf55160f9ae891b703b0c3f81298c7b2\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Wed Jul 17 08:02:52 2019 -0700\n\n ARM: dts: vf610-zii-scu4-aib: Fix pinctrl_i2c1\u0027s identation\n \n Fix pinctrl_i2c1\u0027s inconsistent identation.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 6d234bc6a4b4dc818b7f9a77131a4994cc4d74b5\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Wed Jul 17 08:02:51 2019 -0700\n\n ARM: dts: vf610-zii-spb4: Drop unused pinctrl_i2c1 pinmux config\n \n Pinctrl_i2c1 pinmux config is not used anywhere. Drop it.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 07523a6cb6f868dfb3202a7fd5c7db2a43194f24\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Mon Aug 25 14:46:10 2014 +0200\n\n ARM: dts: ux500: set pull-up on STUIB STMPE IRQ line\n \n Set up the STMPE IRQ line to be in pull-up mode.\n \n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit aef41a4d88ef11a04b29d9f2ffdfd46eb317d7d8\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Mon Jul 15 10:05:02 2019 +0200\n\n ARM: dts: ux500: Fix up the thermal nodes\n \n The thermal driver for the DB8500 was never properly converted\n to device tree, the node should definitely be activated for\n all board variants so move this down into the main SoC\n DTSI, and default on.\n \n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 46269abf9720f1670b27158e4023b8e7f150cb84\nAuthor: Leo Yan \u003cleo.yan@linaro.org\u003e\nDate: Wed May 8 10:18:55 2019 +0800\n\n ARM: dts: ste: Update coresight DT bindings\n \n CoreSight DT bindings have been updated, thus the old compatible strings\n are obsolete and the drivers will report warning if DTS uses these\n obsolete strings.\n \n This patch switches to the new bindings for CoreSight dynamic funnel and\n static replicator, so can dismiss warning during initialisation.\n \n Cc: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Cc: Lee Jones \u003clee.jones@linaro.org\u003e\n Cc: Mathieu Poirier \u003cmathieu.poirier@linaro.org\u003e\n Cc: Suzuki K Poulose \u003csuzuki.poulose@arm.com\u003e\n Signed-off-by: Leo Yan \u003cleo.yan@linaro.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 7a8689df21f425e77207b4245b0b89d3185866e7\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Mon Jul 15 10:06:35 2019 +0200\n\n ARM: ux500: Stop populating the PRCMU devices early\n \n The PRCMU devices are getting populated from the MFD\n driver and that is enough for them to come up. We already\n have an early quirk for the things that need to happen\n really early and the PRCMU timer that need to be registered\n really early will find its own node just fine as the\n timer/clocksource subsystem looks through the whole device\n tree for compatible nodes.\n \n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n\ncommit 513be3334431ec5c645e9e8c3ee7fdc7c03e99a1\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Mon Jul 22 13:56:27 2019 +0800\n\n drm/komeda: remove set but not used variable \u0027old\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/gpu/drm/arm/display/komeda/komeda_plane.c:\n In function komeda_plane_atomic_duplicate_state:\n drivers/gpu/drm/arm/display/komeda/komeda_plane.c:161:35:\n warning: variable old set but not used [-Wunused-but-set-variable\n \n It is not used since commit 990dee3aa456 (\"drm/komeda:\n Computing image enhancer internally\")\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190722055627.38008-1-yuehaibing@huawei.com\n\ncommit f4411786b39beafd7c52e2d7b3f2d5052901035a\nAuthor: Stefan Riedmueller \u003cs.riedmueller@phytec.de\u003e\nDate: Tue Jul 9 09:19:27 2019 +0200\n\n ARM: dts: imx6ull: Add support for PHYTEC phyBOARD-Segin with i.MX 6ULL\n \n In addition to the PHYTEC phyCORE-i.MX 6UL the PHYTEC phyBOARD-Segin is\n also available with the PHYTEC phyCORE-i.MX 6ULL. So this adds support\n for this SOM and its baseboards.\n \n It comes in a full featured option with either NAND flash or eMMC and in\n a low cost option only available with NAND flash.\n \n The hardware specs are:\n \n - Full featured with NAND or eMMC:\n * i.MX 6ULL Y2\n * 512MB DDR3 memory\n * 512MB NAND flash or 4GB/8GB eMMC\n * Dual 10/100 Ethernet\n * USB Host and USB OTG\n * RS232\n * MicroSD external storage\n * Audio, RS232, I2C, SPI, CAN headers\n * Further I/O options via A/V and Expansion headers\n \n - Low cost with NAND:\n * i.MX 6ULL Y0\n * 256MB DDR3 memory\n * 128MB NAND flash\n * Single 10/100 Ethernet\n * USB OTG\n * RS232\n * MicroSD external storage\n * I2C\n * Further I/O options via Expansion headers\n \n Signed-off-by: Stefan Riedmueller \u003cs.riedmueller@phytec.de\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 9ae6390aa1079293a05e11b12a97ef6742993127\nAuthor: Stefan Riedmueller \u003cs.riedmueller@phytec.de\u003e\nDate: Tue Jul 9 09:19:26 2019 +0200\n\n ARM: dts: imx6ul: segin: Move machine include to dts files\n \n Move the imx6ul.dtsi include to the dts files so it is easier to reuse\n the SOM dtsi for e.g. an i.MX 6ULL SOM.\n \n Signed-off-by: Stefan Riedmueller \u003cs.riedmueller@phytec.de\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit f638e7fdbfdfd4411a917beff761720f97feff7f\nAuthor: Stefan Riedmueller \u003cs.riedmueller@phytec.de\u003e\nDate: Tue Jul 9 09:19:25 2019 +0200\n\n ARM: dts: imx6ul: segin: Move ECSPI interface to board include file\n \n The ECSPI interface is available on the expansion connector of every\n PHYTEC phyBOARD-Segin. Move its definition to the board include file\n for better reuse.\n \n Signed-off-by: Stefan Riedmueller \u003cs.riedmueller@phytec.de\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit b349580a4caef6f1fc23c0bd5393d3557ccf467a\nAuthor: Stefan Riedmueller \u003cs.riedmueller@phytec.de\u003e\nDate: Tue Jul 9 09:19:24 2019 +0200\n\n ARM: dts: imx6ul: phycore: Add eMMC at usdhc2\n \n The phyCORE-i.MX 6UL/ULL now can have eMMC instead of the NAND flash\n memory. Add the eMMC node and disable it by default so it can be enabled\n in case it is populated.\n \n Signed-off-by: Stefan Riedmueller \u003cs.riedmueller@phytec.de\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 87dc2cd47989892b2374b43eda0a5f28c679ff2b\nAuthor: Stefan Riedmueller \u003cs.riedmueller@phytec.de\u003e\nDate: Tue Jul 9 09:19:23 2019 +0200\n\n ARM: dts: imx6ul: segin: Only enable NAND if it is populated\n \n The phyCORE-i.MX 6UL/ULL now comes either with NAND flash or eMMC. We\n have to configure the populated memory type in the device tree files. So\n the GPMI node gets disabled by default and only enabled if populated.\n \n Signed-off-by: Stefan Riedmueller \u003cs.riedmueller@phytec.de\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit f1da57d8ea3a9eeb0986fea22389402505d132e7\nAuthor: Stefan Riedmueller \u003cs.riedmueller@phytec.de\u003e\nDate: Tue Jul 9 09:19:22 2019 +0200\n\n ARM: dts: imx6ul: segin: Make FEC and ethphy configurable in dts\n \n To disable Ethernet interfaces in case they are not populated\n make the FEC and Ethernet PHY status configurable in the dts files.\n \n Also change the Ethernet PHYs labels to make them correspond to\n the MDIO address.\n \n Signed-off-by: Stefan Riedmueller \u003cs.riedmueller@phytec.de\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 783c10aa740f70e22833baf33b7a87f6f7d94696\nAuthor: Stefan Riedmueller \u003cs.riedmueller@phytec.de\u003e\nDate: Tue Jul 9 09:19:21 2019 +0200\n\n ARM: dts: imx6ul: segin: Fix LED naming for phyCORE and PEB-EVAL-01\n \n Fix node name for the user LEDs to prevent the pinctrl of the phyCORE\u0027s\n leds node from being overwritten. Also use more generic names for user\n LEDs.\n \n Signed-off-by: Stefan Riedmueller \u003cs.riedmueller@phytec.de\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 458264157b1044d242afa163ac407dc55fa8134e\nAuthor: Stefan Riedmueller \u003cs.riedmueller@phytec.de\u003e\nDate: Tue Jul 9 09:19:20 2019 +0200\n\n ARM: dts: imx6ul: segin: Reduce eth drive strength\n \n Reduce the drive strength for the MDC, MDIO and TX pins of FEC1 and FEC2\n on the phyBOARD-Segin to improve signal quality and EMC. Also disable\n internal pull-ups on the MDC and MDIO pins.\n \n Signed-off-by: Stefan Riedmueller \u003cs.riedmueller@phytec.de\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit e37816bf9114ee1604a1511f7b4155b753ab36b9\nAuthor: Stefan Riedmueller \u003cs.riedmueller@phytec.de\u003e\nDate: Tue Jul 9 09:19:19 2019 +0200\n\n ARM: dts: imx6ul: segin: Add boot media to dts filename\n \n There is now a PHYTEC phyCORE-i.MX 6UL with eMMC instead of NAND flash\n available. The dts filename needs to reflect that to differentiate both.\n \n Signed-off-by: Stefan Riedmueller \u003cs.riedmueller@phytec.de\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 591c175076d71e58b9337385afe36a5ab0cf9e06\nAuthor: Stefan Riedmueller \u003cs.riedmueller@phytec.de\u003e\nDate: Tue Jul 9 09:19:18 2019 +0200\n\n ARM: dts: imx6ul: phyboard-segin: Rename dts to PHYTEC name scheme\n \n Use the same name scheme for the phyBOARD-Segin and the phyCORE-i.MX\n 6UL as is used for the PHYTEC phyBOARD-Mira and phyCORE-i.MX 6.\n \n This is only a cosmetic change and there is no functional change\n intended.\n \n Signed-off-by: Stefan Riedmueller \u003cs.riedmueller@phytec.de\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 8842d285bafa9ff7719f4107b6545a11dcd41995\nAuthor: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\nDate: Mon Jul 22 20:08:31 2019 -0700\n\n net: Convert skb_frag_t to bio_vec\n \n There are a lot of users of frag-\u003epage_offset, so use a union\n to avoid converting those users today.\n \n Signed-off-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b8b576a16f79efbdde49348147f491b176537d88\nAuthor: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\nDate: Mon Jul 22 20:08:30 2019 -0700\n\n net: Rename skb_frag_t size to bv_len\n \n Improved compatibility with bvec\n \n Signed-off-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 1dfa5bd38545c6f6a8b6c496e58db93f80da1076\nAuthor: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\nDate: Mon Jul 22 20:08:29 2019 -0700\n\n net: Rename skb_frag page to bv_page\n \n One step closer to turning the skb_frag_t into a bio_vec.\n \n Signed-off-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit f58ecf1b7d58911921014ffd12c77a4ad33ade71\nAuthor: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\nDate: Mon Jul 22 20:08:28 2019 -0700\n\n net: Reorder the contents of skb_frag_t\n \n Match the layout of bio_vec.\n \n Signed-off-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit b656722906efe434f0befe1d4ae4bb7a66fdc124\nAuthor: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\nDate: Mon Jul 22 20:08:27 2019 -0700\n\n net: Increase the size of skb_frag_t\n \n To increase commonality between block and net, we are going to replace\n the skb_frag_t with the bio_vec. This patch increases the size of\n skb_frag_t on 32-bit machines from 8 bytes to 12 bytes. The size is\n unchanged on 64-bit machines.\n \n Signed-off-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d8e18a516f8f67404c0d21af8c93d0474fba0876\nAuthor: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\nDate: Mon Jul 22 20:08:26 2019 -0700\n\n net: Use skb accessors in network core\n \n In preparation for unifying the skb_frag and bio_vec, use the fine\n accessors which already exist and use skb_frag_t instead of\n struct skb_frag_struct.\n \n Signed-off-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit d7840976e3915669382c62ddd1700960f348328e\nAuthor: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\nDate: Mon Jul 22 20:08:25 2019 -0700\n\n net: Use skb accessors in network drivers\n \n In preparation for unifying the skb_frag and bio_vec, use the fine\n accessors which already exist and use skb_frag_t instead of\n struct skb_frag_struct.\n \n Signed-off-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 7e24b4ed5ac4321e41415b0c6f0f8a8ac14852b2\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Mon Jul 22 15:41:34 2019 +0800\n\n net: usb: Merge cpu_to_le32s + memcpy to put_unaligned_le32\n \n Merge the combo uses of cpu_to_le32s and memcpy.\n Use put_unaligned_le32 instead.\n This simplifies the code.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 24183825c3c7a9378e5a3a5555c6504edd0a302b\nAuthor: Joel Stanley \u003cjoel@jms.id.au\u003e\nDate: Tue Jul 23 12:49:16 2019 +0930\n\n ARM: configs: aspeed: Refresh defconfigs\n \n A noop change to git rid of cruft.\n \n Signed-off-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n\ncommit 2f90ec143af83146a4958d78b47d5220024c7713\nAuthor: William A. Kennington III \u003cwak@google.com\u003e\nDate: Wed Jan 16 12:54:59 2019 +1100\n\n ARM: configs: aspeed: Enable commonly used network functionality\n \n This addresses some assumptions made by systemd about having multiple\n routing table support in the kernel. systemd-networkd will try and\n provision mutliple routing tables + policies and will silently break\n neighbor advertisement responses due to policy configurations on a\n single table.\n \n It also adds support for SLAAC based router settings and faster\n duplicate address detection.\n \n Signed-off-by: William A. Kennington III \u003cwak@google.com\u003e\n Signed-off-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n\ncommit bb1dba7be1ee721b83a9cca24fb6d83f762b147b\nAuthor: Hongwei Zhang \u003chongweiz@ami.com\u003e\nDate: Tue Jun 4 17:42:04 2019 -0400\n\n ARM: dts: aspeed: Add SGPM pinmux\n \n Add SGPM pinmux to ast2500-pinctrl function and group, to prepare for\n supporting SGPIO in AST2500 SoC.\n \n Signed-off-by: Hongwei Zhang \u003chongweiz@ami.com\u003e\n Acked-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n\ncommit b3b06766315e06e0655d334e3323da7eb21a7e39\nAuthor: Vijay Khemka \u003cvijaykhemka@fb.com\u003e\nDate: Mon Jul 22 17:32:16 2019 -0700\n\n ARM: dts: aspeed: tiogapass: Add VR devices\n \n Adds voltage regulators Infineon pxe1610 devices to Facebook\n tiogapass platform.\n \n Signed-off-by: Vijay Khemka \u003cvijaykhemka@fb.com\u003e\n Acked-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Signed-off-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n\ncommit dc4bea0b989cfa3e3dfd8fe79a18abcfaca19767\nAuthor: Andrew Peng \u003cpengms1@lenovo.com\u003e\nDate: Tue Jun 25 14:38:16 2019 +0800\n\n ARM: dts: aspeed: Add Lenovo Hr855xg2 BMC\n \n Initial introduction of Lenovo Hr855xg2 family equipped with\n Aspeed 2500 BMC SoC. Hr855xg2 is a x86 server development kit\n with a ASPEED ast2500 BMC manufactured by Lenovo.\n \n This adds the Hr855xg2 platform device tree file used by the Hr855xg2\n BMC machines.\n \n Signed-off-by: Andrew Peng \u003cpengms1@lenovo.com\u003e\n Signed-off-by: Yonghui Liu \u003cliuyh21@lenovo.com\u003e\n Signed-off-by: Lisa Liu \u003cliuyj19@lenovo.com\u003e\n Signed-off-by: Harry Sung \u003chsung1@lenovo.com\u003e\n Signed-off-by: Derek Lin \u003cdlin23@lenovo.com\u003e\n Reviewed-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Signed-off-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n\ncommit da335d2bcb426bd016dd0be5263a25b60903c3d2\nAuthor: Tao Ren \u003ctaoren@fb.com\u003e\nDate: Mon Jun 17 21:24:21 2019 -0700\n\n ARM: dts: aspeed: Add Facebook Minipack BMC\n \n Add initial version of device tree for Facebook Minipack ast2500 BMC.\n \n Signed-off-by: Tao Ren \u003ctaoren@fb.com\u003e\n Reviewed-by: Andrew Jeffery \u003candrew@aj.id.au\u003e\n Signed-off-by: Joel Stanley \u003cjoel@jms.id.au\u003e\n\ncommit 1843f225625bb7c15109136ac870ff39e304fc42\nAuthor: Fugang Duan \u003cfugang.duan@nxp.com\u003e\nDate: Thu Jul 4 21:43:55 2019 +0800\n\n dt-bindings: serial: lpuart: add the clock requirement for imx8qxp\n \n Add the baud clock requirement for imx8qxp.\n \n Signed-off-by: Fugang Duan \u003cfugang.duan@nxp.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit fcd4e60885af969d190999a6c42454c5578d52f1\nAuthor: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\nDate: Mon Jul 22 22:01:15 2019 +0200\n\n r8169: improve rtl_rx\n \n This patch improves few aspects of rtl_rx, no functional change intended.\n \n 1. inline rtl8169_try_rx_copy\n 2. make pkt_size unsigned\n 3. use constant ETH_FCS_LEN instead of value 4\n 4. We just created the skb, so we don\u0027t need the checks in skb_put.\n Also we don\u0027t need the return value of skb_put.\n Set skb-\u003etail and skb-\u003elen directly.\n \n Signed-off-by: Heiner Kallweit \u003chkallweit1@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 722eb2f65acc4cebeb710fc7cc98f51513e90f1f\nAuthor: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\nDate: Fri Jun 21 07:59:51 2019 -0700\n\n arm64: dts: qcom: Add Asus NovaGo TP370QL\n \n This adds the initial DT for the Asus NovaGo TP370QL laptop. Supported\n functionality includes USB (host), microSD-card, keyboard, and trackpad.\n \n Signed-off-by: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 3f527d311932791fde67ffec32536d22d5dd3030\nAuthor: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\nDate: Fri Jun 21 07:57:21 2019 -0700\n\n arm64: dts: qcom: Add HP Envy x2\n \n This adds the initial DT for the HP Envy x2 laptop. Supported\n functionality includes USB (host), microSD-card, keyboard, and trackpad.\n \n Signed-off-by: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 2c6d2d3a580a852fe0a694e13af502a862293e0e\nAuthor: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\nDate: Fri Jun 21 07:54:50 2019 -0700\n\n arm64: dts: qcom: Add Lenovo Miix 630\n \n This adds the initial DT for the Lenovo Miix 630 laptop. Supported\n functionality includes USB (host), microSD-card, keyboard, and trackpad.\n \n Signed-off-by: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit af311ff9a69189a03548efd5a47d4bb44644fd45\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Fri May 17 14:09:22 2019 -0700\n\n firmware: qcom_scm: Cleanup code in qcom_scm_assign_mem()\n \n There are some questionable coding styles in this function. It looks\n quite odd to deref a pointer with array indexing that only uses the\n first element. Also, destroying an input/output variable halfway through\n the function and then overwriting it on success is not clear. It\u0027s\n better to use a local variable and the kernel macros to step through\n each bit set in a bitmask and clearly show where outputs are set.\n \n Cc: Ian Jackson \u003cian.jackson@citrix.com\u003e\n Cc: Julien Grall \u003cjulien.grall@arm.com\u003e\n Cc: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Cc: Avaneesh Kumar Dwivedi \u003cakdwived@codeaurora.org\u003e\n Tested-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n [bjorn: Changed for_each_set_bit() size to BITS_PER_LONG]\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 3bdd4f8485310687affe117f1b1a6e9012e897f6\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Jul 22 23:28:47 2019 +0100\n\n drm/i915: Rely on spinlock protection for GPU error capture\n \n Trust that we now have adequate protection over the low level structures\n via the engine-\u003eactive.lock to allow ourselves to capture the GPU error\n state without the heavy hammer of stop_machine(). Sadly this does mean\n that we have to forgo some of the lesser used information (not derived\n from the active state) that is not controlled by the active locks. This\n includes the list of buffers in the ppGTT and pinned globally in the\n GGTT. Originally this was used to manually verify relocations, but\n hasn\u0027t been required for sometime and modern mesa now has the habit of\n ensuring that all interesting buffers within a batch are captured in their\n entirety (that are the auxiliary state buffers, but not the textures).\n \n A useful side-effect is that this allows us to restore error capturing\n for Braswell and Broxton.\n \n v2: Use pagevec for a typical arbitrary number of preallocated pages\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190722222847.24178-1-chris@chris-wilson.co.uk\n\ncommit df8cf31e749788d8de8c3028e530e6fd9ef8bc79\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Jul 18 08:00:06 2019 +0100\n\n drm/i915/gt: Hook up intel_context_fini()\n \n Prior to freeing the struct, call the fini function to cleanup the\n common members. Currently this only calls the debug functions to mark\n the structs as destroyed, but may be extended to real work in future.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190718070024.21781-2-chris@chris-wilson.co.uk\n\ncommit f3bcb0cccd36c6eb6e3afbd61f83eeca32d6cc50\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Jul 18 08:00:10 2019 +0100\n\n drm/i915: Remove obsolete engine cleanup\n \n Remove the outer layer cleanup of engine stubs; as i915_drv itself no\n longer tries to preallocate and so is not responsible for either the\n allocation or free. By the time we call the cleanup function, we already\n have cleaned up the engines.\n \n v2: Lack of symmetry between mmio_probe and mmio_release for handling\n the error cleanup. engine-\u003edestroy() is a compound function that is\n called earlier in the normal release as it ties together other bits of\n state.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190718070024.21781-6-chris@chris-wilson.co.uk\n\ncommit d808097627e51d53cf9b1aa13239b5c4a6adaefb\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Mon Jul 22 23:18:16 2019 +0200\n\n drm/i915: Fix up broken merge\n \n Maxime didn\u0027t really compile-test this :-/\n \n We need to re-apply\n \n commit e4fa8457b2197118538a1400b75c898f9faaf164\n Author: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Date: Fri Jun 14 22:35:25 2019 +0200\n \n drm/prime: Align gem_prime_export with obj_funcs.export\n \n plus make sure i915_gem_dma_buf.c doesn\u0027t get zombie-resurrect. It\n moved in\n \n commit 10be98a77c558f8cfb823cd2777171fbb35040f6\n Author: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Date: Tue May 28 10:29:49 2019 +0100\n \n drm/i915: Move more GEM objects under gem/\n \n v2: Remember the selftests (Chris).\n \n Fixes: 03b0f2ce735e (\"Merge v5.3-rc1 into drm-misc-next\")\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Acked-by: Dave Airlie \u003cairlied@gmail.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190722213759.26612-1-daniel.vetter@ffwll.ch\n\ncommit f23478f6ea2ed6c8dccc6f7b902c2362ebc197ef\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Fri Jan 25 17:09:27 2019 -0300\n\n MIPS: qi_lb60: Move MMC configuration to devicetree\n \n Move the MMC configuration from the board C file to devicetree.\n \n The \u0027power\u0027 GPIO was removed and instead the vmmc regulator is used,\n to follow the changes introduced in the jz4740-mmc driver.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n\ncommit 838c1efc6a0adbf44d16e4714f43443db598cdb5\nAuthor: Bean Huo \u003cbeanhuo@micron.com\u003e\nDate: Mon Jul 15 11:21:10 2019 +0000\n\n scsi: ufs: change msleep to usleep_range\n \n This patch is to change msleep() to usleep_range() based on\n Documentation/timers/timers-howto.txt. It suggests using usleep_range()\n for small msec(1ms - 20ms) since msleep() will often sleep longer than\n desired value.\n \n After changing, booting time will be 5ms-10ms faster than before. I\n tested this change on two different platforms, one has 5ms faster, another\n one is about 10ms. I think this is different on different platform.\n \n Actually, from UFS host side, 1ms-5ms delay is already sufficient for its\n initialization of the local UIC layer.\n \n Fixes: 7a3e97b0dc4b ([SCSI] ufshcd: UFS Host controller driver)\n Signed-off-by: Bean Huo \u003cbeanhuo@micron.com\u003e\n Acked-by: Stanley Chu \u003cstanley.chu@mediatek.com\u003e\n Reviewed-by: Avri Altman \u003cavri.altman@wdc.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 70a51d8c53176080e82b2aef050f820ccdd4b3f8\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Jul 11 22:10:37 2019 +0800\n\n scsi: lpfc: Remove unnecessary null check before kfree\n \n A null check before a kfree is redundant, so remove it. This is detected\n by coccinelle.\n \n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Reviewed-by: James Smart \u003cjames.smart@broadcom.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 9e5470fe2d61bd87c27c7722099bafab277f81ec\nAuthor: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\nDate: Thu May 30 13:28:11 2019 +0200\n\n scsi: virtio_scsi: implement request batching\n \n Adding the command and kicking the virtqueue so far was done one after\n another. Make the kick optional, so that we can take into account\n SCMD_LAST. We also need a commit_rqs callback to kick the device if\n blk-mq aborts the submission before the last request is reached.\n \n Suggested-by: Stefan Hajnoczi \u003cstefanha@redhat.com\u003e\n Signed-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n Reviewed-by: Hannes Reinecke \u003chare@suse.com\u003e\n Reviewed-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 8930a6c207918d5a5675eedab06a71096b1a3d47\nAuthor: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\nDate: Thu May 30 13:28:10 2019 +0200\n\n scsi: core: add support for request batching\n \n This allows a list of requests to be issued, with the LLD only writing the\n hardware doorbell when necessary, after the last request was prepared.\n This is more efficient if we have lists of requests to issue, particularly\n on virtualized hardware, where writing the doorbell is more expensive than\n on real hardware.\n \n The use case for this is plugged IO, where blk-mq flushes a batch of\n requests all at once.\n \n The API is the same as for blk-mq, just with blk-mq concepts tweaked to\n fit the SCSI subsystem API: the \"last\" flag in blk_mq_queue_data becomes a\n flag in scsi_cmnd, while the queue_num in the commit_rqs callback is\n extracted from the hctx and passed as a parameter.\n \n The only complication is that blk-mq uses different plugging heuristics\n depending on whether commit_rqs is present or not. So we have two\n different sets of blk_mq_ops and pick one depending on whether the\n scsi_host template uses commit_rqs or not.\n \n Signed-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n Reviewed-by: Bart Van Assche \u003cbvanassche@acm.org\u003e\n Reviewed-by: Ming Lei \u003cming.lei@redhat.com\u003e\n Reviewed-by: Hannes Reinecke \u003chare@suse.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit cf949bbe22bee8749078e0b810ee2dc60a983746\nAuthor: Avri Altman \u003cavri.altman@wdc.com\u003e\nDate: Wed Jun 12 16:34:37 2019 +0300\n\n scsi: ufs: uapi: Fix SPDX license identifier\n \n Added \u0027WITH Linux-syscall-note\u0027 exception which is the officially assigned\n exception identifier for the kernel syscall exception. This exception\n makes it possible to include GPL headers into non GPL code without\n confusing license compliance tools.\n \n Fixes: a851b2bd3632 (scsi: uapi: ufs: Make utp_upiu_req visible to user space)\n Signed-off-by: Avri Altman \u003cavri.altman@wdc.com\u003e\n Reviewed-by: Pedro Sousa \u003cpedrom.sousa@synopsys.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 8808b4e9dcdc908e1d48b1809416c5f80af9c87e\nAuthor: Stanley Chu \u003cstanley.chu@mediatek.com\u003e\nDate: Wed Jul 10 21:38:21 2019 +0800\n\n scsi: ufs: Add history of fatal events\n \n Currently only \"interrupt-based\" errors have their own history, however\n there are some \"non-interrupt-based\" errors or events which need history\n to improve debugging or help know the health status of UFS devices.\n \n Example of fatal errors:\n \n - Link startup error\n \n - Suspend error\n \n - Resume error\n \n Example of abnormal events:\n \n - Task or request abort\n \n - Device reset (now equals to Logical Unit Reset)\n \n - Host reset\n \n This patch tries to track above errors and events by existed UFS error\n history mechanism.\n \n Signed-off-by: Stanley Chu \u003cstanley.chu@mediatek.com\u003e\n Reviewed-by: Avri Altman \u003cavri.altman@wdc.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit c5397f13fc744b15fc9ad5df27172885e2303c39\nAuthor: Stanley Chu \u003cstanley.chu@mediatek.com\u003e\nDate: Wed Jul 10 21:38:20 2019 +0800\n\n scsi: ufs: Do not reset error history during host reset\n \n Currently UFS error history will be reset and lost during host reset flow\n by ufschd_probe_hba().\n \n Avoid resetting it so error history can be kept as complete as possible to\n improve debugging.\n \n In addition, fix a minor display error in ufshcd_print_err_hist().\n \n [mkp: commit desc clarification]\n \n Signed-off-by: Stanley Chu \u003cstanley.chu@mediatek.com\u003e\n Reviewed-by: Avri Altman \u003cavri.altman@wdc.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit d3c615bf586f2796f24097bfa13386effaa4bf3e\nAuthor: Stanley Chu \u003cstanley.chu@mediatek.com\u003e\nDate: Wed Jul 10 21:38:19 2019 +0800\n\n scsi: ufs: Add fatal and auto-hibern8 error history\n \n Provide more information about fatal and auto-hibern8 errors to improve\n debugging by extending existing UFS error history framework.\n \n [mkp: typos]\n \n Signed-off-by: Stanley Chu \u003cstanley.chu@mediatek.com\u003e\n Reviewed-by: Avri Altman \u003cavri.altman@wdc.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 48d5b9732580bca18fed9ef20da8aa210c646755\nAuthor: Stanley Chu \u003cstanley.chu@mediatek.com\u003e\nDate: Wed Jul 10 21:38:18 2019 +0800\n\n scsi: ufs: Change names related to error history\n \n Remove \"uic\" term in below error history functions and structures for more\n general usages,\n \n struct ufs_uic_err_reg_hist;\n void ufshcd_update_uic_reg_hist(struct ufs_uic_err_reg_hist *reg_hist,\n u32 reg);\n void ufshcd_print_uic_err_hist(struct ufs_hba *hba,\n struct ufs_uic)err_reg_hist *err_hist, char *err_name);\n \n Signed-off-by: Stanley Chu \u003cstanley.chu@mediatek.com\u003e\n Reviewed-by: Avri Altman \u003cavri.altman@wdc.com\u003e\n Signed-off-by: Martin K. Petersen \u003cmartin.petersen@oracle.com\u003e\n\ncommit 04364138ce9bd812e8eeda84febf7259e27ff57d\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 19 14:07:37 2019 +0100\n\n drm/i915/gtt: Fix rounding for 36b\n \n The top-level page directory for 36b is a single entry, not multiple\n like 32b. Fix up the rounding on the calculation of the size of the top\n level so that we populate the 4th level correctly for 36b.\n \n Reported-by: Jose Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Fixes: 1eda701eace2 (\"drm/i915/gtt: Recursive cleanup for gen8\")\n Cc: Abdiel Janulgue \u003cabdiel.janulgue@linux.intel.com\u003e\n Cc: Jose Souza \u003cjose.souza@intel.com\u003e\n Tested-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Reviewed-by: Abdiel Janulgue \u003cabdiel.janulgue@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190719130737.5835-1-chris@chris-wilson.co.uk\n\ncommit 46eabee1f6e6495ea465d7ad8bb971c40ba4dc9e\nAuthor: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\nDate: Mon Jul 22 19:26:15 2019 +0200\n\n rtc: isl12026: convert to i2c_new_dummy_device\n \n Move from i2c_new_dummy() to i2c_new_dummy_device(), so we now get an\n ERRPTR which we use in error handling.\n \n Signed-off-by: Wolfram Sang \u003cwsa+renesas@sang-engineering.com\u003e\n Link: https://lore.kernel.org/r/20190722172618.4061-2-wsa+renesas@sang-engineering.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 564225415e7791f809122d8d16f089d276686bda\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jul 22 16:09:21 2019 +0200\n\n dt-bindings: rtc: Remove the PCF8563 from the trivial RTCs\n \n The PCF8563 has a binding of its own, with some, clocks related, additional\n properties.\n \n Remove it from the trivial RTC bindings.\n \n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://lore.kernel.org/r/20190722140921.22681-1-maxime.ripard@bootlin.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit f7234a9813b764d3d0ffa67c9f983316bab7a58d\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jul 17 16:14:11 2019 +0800\n\n rtc: imxdi: use devm_platform_ioremap_resource() to simplify code\n \n Use the new helper devm_platform_ioremap_resource() which wraps the\n platform_get_resource() and devm_ioremap_resource() together, to\n simplify the code.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190717081411.30622-1-Anson.Huang@nxp.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 874532cdeefefa4c531189ceaf57d05d2daafdbb\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jul 17 16:12:17 2019 +0800\n\n rtc: mxc_v2: use devm_platform_ioremap_resource() to simplify code\n \n Use the new helper devm_platform_ioremap_resource() which wraps the\n platform_get_resource() and devm_ioremap_resource() together, to\n simplify the code.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190717081217.30518-1-Anson.Huang@nxp.com\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit e858c5951f69d4eb15f51340d622ab1164c71ae8\nAuthor: Fabio Estevam \u003cfestevam@gmail.com\u003e\nDate: Mon Jul 22 16:40:49 2019 -0300\n\n drm/bridge: Improve the help text for DRM_ANALOGIX_ANX78XX\n \n Improve the help text for DRM_ANALOGIX_ANX78XX by adding the missing\n \"power\" word.\n \n After this change the help text matches with the ANX7814\n product description from the Analogix website:\n \n https://www.analogix.com/en/products/convertersbridges/anx7814\n \n Signed-off-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190722194049.20761-1-festevam@gmail.com\n\ncommit 95ccc155081be9cea95202d290fe06637f47c8de\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Thu Jul 18 15:25:04 2019 -0500\n\n drm/amdgpu/smu: move fan rpm query into the asic specific code\n \n On vega20, there is an SMU message to query it. On navi, it\u0027s fetched\n from the metrics table.\n \n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 60439711b69b893e84d5bab56de66f9433e68a30\nAuthor: Enrico Weigelt \u003cinfo@metux.net\u003e\nDate: Mon Jul 15 16:57:33 2019 +0200\n\n leds: apu: add pr_fmt prefix for better log output\n \n Define pr_fmt() macro for prefixing the module name to all\n pr_*() outputs, in order to increase log message clarity.\n \n Signed-off-by: Enrico Weigelt \u003cinfo@metux.net\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit d52d6de2809a31b58a62f9ecba58bfd8fbf1bc63\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Fri Jul 19 19:09:38 2019 +0800\n\n drm/amdgpu: set sdma irq src num according to sdma instances\n \n Otherwise, it will cause driver access non-existing sdma registers\n in gpu reset code path\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Le Ma \u003cLe.Ma@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 93562049a62c5519b3787a428cdd3cfa7fc9e7d4\nAuthor: Enrico Weigelt \u003cinfo@metux.net\u003e\nDate: Mon Jul 22 14:08:51 2019 +0200\n\n leds: apu: fix error message on probing failure\n \n The current error message on failed probing tends to be a bit\n misleading. Fix it to tell exactly that an APU v1 was not found.\n \n Signed-off-by: Enrico Weigelt \u003cinfo@metux.net\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit b629167d686787507f35a8bfb46b8b12f4f26eb7\nAuthor: Kevin Wang \u003ckevin1.wang@amd.com\u003e\nDate: Fri Jul 19 16:06:29 2019 +0800\n\n drm/amd/powerplay: remove redundancy debug log in smu\n \n remove redundacy debug log in smu.\n eg:\n [ 6897.969447] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled\n [ 6897.969448] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled\n [ 6897.969448] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled\n [ 6899.024114] amdgpu: [powerplay] Unsupported SMU message: 38\n [ 6899.024151] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled\n [ 6899.024151] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled\n [ 6899.024152] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled\n [ 6900.078296] amdgpu: [powerplay] Unsupported SMU message: 38\n [ 6900.078332] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled\n [ 6900.078332] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled\n [ 6900.078333] amdgpu: [powerplay] smu 11 clk dpm feature 1 is not enabled\n [ 6901.133230] amdgpu: [powerplay] Unsupported SMU message: 38\n \n Signed-off-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit cc92e3cce5db4efd6b555102dbf267c8c76bb811\nAuthor: Enrico Weigelt \u003cinfo@metux.net\u003e\nDate: Mon Jul 15 16:57:31 2019 +0200\n\n leds: apu: drop profile field from priv data\n \n As this driver now only supports the APU1 boards, we don\u0027t need\n to handle profiles anymore and just can use the only one global\n array directly.\n \n Signed-off-by: Enrico Weigelt \u003cinfo@metux.net\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 08e83826a5ebcaea099a999d4e9915148b78f00a\nAuthor: Enrico Weigelt \u003cinfo@metux.net\u003e\nDate: Mon Jul 15 16:57:30 2019 +0200\n\n leds: apu: drop iosize field from priv data\n \n As this driver now only supports the APU1 boards, we don\u0027t need\n to handle different io sizes anymore.\n \n Signed-off-by: Enrico Weigelt \u003cinfo@metux.net\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit da97735cbdb44253b549d6acf8204124de8d72f0\nAuthor: Enrico Weigelt \u003cinfo@metux.net\u003e\nDate: Mon Jul 15 16:57:29 2019 +0200\n\n leds: apu: drop enum_apu_led_platform_types\n \n As this driver now only supports the APU1 boards, we don\u0027t need\n to differenciate between board types anymore. Therefore optimize\n away the now obsolete code.\n \n Signed-off-by: Enrico Weigelt \u003cinfo@metux.net\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit ebf8fc31cbcedc9d6a81642082661c82eae284fb\nAuthor: Kevin Wang \u003ckevin1.wang@amd.com\u003e\nDate: Fri Jul 19 11:39:21 2019 +0800\n\n drm/amd/powerplay: custom peak clock freq for navi10\n \n v2:\n add function smu_default_set_performance_level as default dpm level handler.\n change function name smu_set_performance_level to smu_asic_set_performance_level\n \n v1:\n 1.NAVI10_PEAK_SCLK_XTX 1830 Mhz\n 2.NAVI10_PEAK_SCLK_XT 1755 Mhz\n 3.NAVI10_PEAK_SCLK_XL 1625 Mhz\n \n Signed-off-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Jack Gui \u003cJack.Gui@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 0344e6166bbf90ad3fe05497ffe754bd65f9e7ac\nAuthor: Enrico Weigelt \u003cinfo@metux.net\u003e\nDate: Mon Jul 15 16:57:28 2019 +0200\n\n leds: apu: drop superseeded apu2/3 led support\n \n This driver only supports gpio-attached LEDs on apu1/2/3 boards,\n but neither gpio\u0027s themselves, nor other gpio-attached devices\n (eg. front button).\n \n For apu2+ a newer, more complete driver exists, based on a generic\n driver for the AMD SOCs gpio-controller, supporting LEDs as well\n other devices. Therefore these boards don\u0027t need legacy support\n from this driver anymore.\n \n Both drivers claim the same device, so only one driver may exist\n in a system, putting distros in the position to decide between\n either apu2+ keys or apu1 led support.\n \n Therefore drop the apu2+ led support from the old driver, solve\n the conflict and reduce unnecessary code.\n \n Signed-off-by: Enrico Weigelt \u003cinfo@metux.net\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 967ea03cc345d86c8e134d7610e61a3012046407\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Sun Jul 21 23:05:39 2019 +0200\n\n leds: ktd2692: Fix a typo in the name of a constant\n \n There is a typo in KTD2962_MM_MIN_CURR_THRESHOLD_SCALE. 6 and 9 are\n switched in 2962.\n \n Define and use KTD2692_MM_MIN_CURR_THRESHOLD_SCALE instead.\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Reviewed-by: Enrico Weigelt \u003cinfo@metux.net\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 529a8f939a5fdbfa27d10bba728d9764212ab26f\nAuthor: Sven Eckelmann \u003csven@narfation.org\u003e\nDate: Sat Jul 6 14:56:13 2019 +0200\n\n batman-adv: Replace usage of strlcpy with strscpy\n \n The strscpy was introduced to fix some API problems around strlcpy. And\n checkpatch started to report recently that strlcpy is deprecated and\n strscpy is preferred.\n \n The functionality introduced in commit 30035e45753b (\"string: provide\n strscpy()\") improves following points compared to strlcpy:\n \n * it doesn\u0027t read from memory beyond (src + size)\n * provides an easy way to check for destination buffer overflow\n * robust against asynchronous source buffer changes\n \n Since batman-adv doesn\u0027t depend on any of the previously mentioned behavior\n changes, the usage of strlcpy can simply be replaced by strscpy to silence\n checkpatch.\n \n Signed-off-by: Sven Eckelmann \u003csven@narfation.org\u003e\n Signed-off-by: Simon Wunderlich \u003csw@simonwunderlich.de\u003e\n\ncommit bc986391c80decf6b5a16756f29400377833b1e9\nAuthor: Simon Wunderlich \u003csw@simonwunderlich.de\u003e\nDate: Sat Jul 6 15:07:50 2019 +0200\n\n batman-adv: Start new development cycle\n \n Signed-off-by: Simon Wunderlich \u003csw@simonwunderlich.de\u003e\n\ncommit f3b357c280738781df3b25c1cbb8166e11982eeb\nAuthor: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\nDate: Sun Jul 21 22:59:55 2019 +0200\n\n leds: an30259a: Fix typo\n \n All this file is about an30259a, including the reference to the datasheet\n at the top of the file.\n \n So change the 2 places where an32059a is used instead.\n \n Signed-off-by: Christophe JAILLET \u003cchristophe.jaillet@wanadoo.fr\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 66193b24514c91aeda88da744554b2665471aeae\nAuthor: Jan Pieter van Woerkom \u003cjp@jpvw.nl\u003e\nDate: Wed Jul 17 14:09:10 2019 -0400\n\n media: dvbsky: add support for Mygica T230C v2\n \n Adds support for the \"Mygica T230C v2\" to the dvbsky driver.\n \n Signed-off-by: Jan Pieter van Woerkom \u003cjp@jpvw.nl\u003e\n Tested-by: Frank Rysanek \u003cFrantisek.Rysanek@post.cz\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit bc28d36bf11acc5b7bbd1a6ffbe315d15e7c0d0d\nAuthor: Jan Pieter van Woerkom \u003cjp@jpvw.nl\u003e\nDate: Wed Jul 17 20:36:54 2019 -0400\n\n media: si2168: add support for Mygica T230C v2\n \n The T230C v2 hardware needs a mode of the si2168 chip to be\n set for which the si2168 driver previously had no support.\n This patch uses a specific measure to configure this on the\n T230C v2 hardware only - see the flag passed via the ts_mode\n attribute and its dependency on USB_PID_MYGICA_T230C2.\n \n Signed-off-by: Jan Pieter van Woerkom \u003cjp@jpvw.nl\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 19d41a2899dcca7862af453d62a8e61933f3252d\nAuthor: A Sun \u003cas1033x@comcast.net\u003e\nDate: Sun Jul 14 22:51:26 2019 -0400\n\n media: mceusb: USB reset device following USB clear halt error\n \n This patch schedules a USB reset device call following a USB clear\n halt error. The issues solved, and patch implementation,\n are similar to those found in\n drivers/hid/usbhid/hid-core.c.\n \n As seen on very rare occasions approximately one time per month\n (mceusb device 2304:0225 in this sample)\n Jul 27 2018 15:09:39\n [59388.696941] mceusb 1-1.1.2:1.0: Error: urb status \u003d -32 (RX HALT)\n [59388.698838] mceusb 1-1.1.2:1.0: rx clear halt error -32\n the device can get into RX or TX HALT state where usb_clear_halt()\n also fails and also returns -EPIPE (HALT/STALL). After which, all\n further mceusb device control and data I/O always fail with HALT/STALL.\n Subsequently, the entire mceusb device no longer functions.\n Cause and problem replication conditions remain unknown.\n \n Further troubleshooting reveals usb_reset_device()\n restores mceusb device operation.\n \n Patch test 1:\n \n Hot unplugging the mceusb device triggers USB RX HALT and\n USB clear halt errors. A mceusb_dev_disconnect() call follows unplug.\n This patch\u0027s reset device call invokes an extra\n mceusb_dev_probe()\n mceusb_dev_disconnect()\n cycle, before the mceusb driver detaches.\n The additional probe/disconnect verifies the patch\u0027s device reset\n code executed.\n \n But note this patch is for USB clear halt error cases not caused by\n unplugging the mceusb device.\n \n Patch test 2:\n \n Simulate a RX HALT and a clear halt error with instrumented code in\n the driver.\n Jul 12 2019 19:41:18\n [522745.263104] mceusb 1-1.3:1.0: set rx halt retval, 0\n [522745.263943] mceusb 1-1.3:1.0: Error: rx urb status \u003d -32 (RX HALT)\n [522745.263970] mceusb 1-1.3:1.0: kevent 1 scheduled\n [522745.264016] mceusb 1-1.3:1.0: kevent handler called (flags 0x2)\n [522745.272883] mceusb 1-1.3:1.0: rx clear halt status \u003d 0\n [522745.272917] mceusb 1-1.3:1.0: stuck RX HALT state requires USB Reset Device to clear\n [522745.273005] mceusb 1-1.3:1.0: mceusb_dev_disconnect called\n [522745.702815] usb 1-1.3: reset full-speed USB device number 14 using dwc_otg\n [522745.836812] mceusb 1-1.3:1.0: mceusb_dev_probe called\n [522745.836823] mceusb 1-1.3:1.0: acceptable bulk inbound endpoint found\n [522745.836832] mceusb 1-1.3:1.0: acceptable bulk outbound endpoint found\n ...\n The result matches what is expected when the device gets into\n a real rx clear halt error case by itself.\n This is the same sequence of messages when manually invoking\n the ./usbreset command line utility with an unpatched mceusb driver.\n \n Signed-off-by: A Sun \u003cas1033x@comcast.net\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 765bb8610d305ee488b35d07e2a04ae52fb2df9c\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Fri Jun 28 08:14:53 2019 -0400\n\n media: dib0700: fix link error for dibx000_i2c_set_speed\n \n When CONFIG_DVB_DIB9000 is disabled, we can still compile code that\n now fails to link against dibx000_i2c_set_speed:\n \n drivers/media/usb/dvb-usb/dib0700_devices.o: In function `dib01x0_pmu_update.constprop.7\u0027:\n dib0700_devices.c:(.text.unlikely+0x1c9c): undefined reference to `dibx000_i2c_set_speed\u0027\n \n The call sites are both through dib01x0_pmu_update(), which gets passed\n an \u0027i2c\u0027 pointer from dib9000_get_i2c_master(), which has returned\n NULL. Checking this pointer seems to be a good idea anyway, and it avoids\n the link failure in most cases.\n \n Sean Young found another case that is not fixed by that, where certain\n gcc versions leave an unused function in place that causes the link error,\n but adding an explict IS_ENABLED() check also solves this.\n \n Fixes: b7f54910ce01 (\"V4L/DVB (4647): Added module for DiB0700 based devices\")\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 0c4df39e504bf925ab666132ac3c98d6cbbe380b\nAuthor: Sean Young \u003csean@mess.org\u003e\nDate: Wed Jul 3 10:52:39 2019 -0400\n\n media: technisat-usb2: break out of loop at end of buffer\n \n Ensure we do not access the buffer beyond the end if no 0xff byte\n is encountered.\n \n Reported-by: syzbot+eaaaf38a95427be88f4b@syzkaller.appspotmail.com\n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Reviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 5dd4b89dc098bf22cd13e82a308f42a02c102b2b\nAuthor: Sean Young \u003csean@mess.org\u003e\nDate: Fri Jul 12 18:47:00 2019 -0400\n\n media: mtk-cir: lower de-glitch counter for rc-mm protocol\n \n The rc-mm protocol can\u0027t be decoded by the mtk-cir since the de-glitch\n filter removes pulses/spaces shorter than 294 microseconds.\n \n Tested on a BananaPi R2.\n \n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Acked-by: Sean Wang \u003csean.wang@kernel.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit a81431e7d1077a84695fc7be487fc2b4930f8c3d\nAuthor: Sean Young \u003csean@mess.org\u003e\nDate: Fri Jul 12 18:46:59 2019 -0400\n\n media: rc: remove unused #define RC_PROTO_BIT_ALL\n \n This lists all the protocols that the kernel knows about, however there\n are no users.\n \n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit eab865203f405501f799e1d7b7b020746e6ea26e\nAuthor: Sean Young \u003csean@mess.org\u003e\nDate: Fri Jul 12 18:46:58 2019 -0400\n\n media: mtk-cir: only allow protocols that have software decoders\n \n RC_PROTO_BIT_ALL includes protocols like unknown and other that do not\n have IR decoders by definition. If these protocols are set in the\n allowed_protocols, they will show in the protocols sysfs file but cannot\n be enabled.\n \n Signed-off-by: Sean Young \u003csean@mess.org\u003e\n Acked-by: Sean Wang \u003csean.wang@kernel.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit d3bfed33e2d80ce5a25549c2f935c91775b2b3f3\nAuthor: Jonas Karlman \u003cjonas@kwiboo.se\u003e\nDate: Wed Jun 12 05:42:08 2019 -0400\n\n media: hantro: Add support for MPEG-2 decoding on RK3328\n \n Add necessary bits to support MPEG2 decoding on RK3328.\n \n Signed-off-by: Jonas Karlman \u003cjonas@kwiboo.se\u003e\n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 56613e7153a53d6cc7eb8d1d6abf00e9fe71c6f1\nAuthor: Jonas Karlman \u003cjonas@kwiboo.se\u003e\nDate: Wed May 29 06:54:27 2019 -0400\n\n media: dt-bindings: rockchip: Document RK3328 VPU binding\n \n Update devicetree binding documentation for Rockchip VPU on RK3328.\n \n Signed-off-by: Jonas Karlman \u003cjonas@kwiboo.se\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit bdd034050a46be32a9a68cc2cd1db8046efe5324\nAuthor: ZhiChao Yu \u003czhichao.yu@rock-chips.com\u003e\nDate: Thu Jul 11 16:26:44 2019 -0400\n\n media: hantro: Add support for VP8 decoding on rk3288\n \n Introduce VP8 decoding support in RK3288.\n \n Signed-off-by: ZhiChao Yu \u003czhichao.yu@rock-chips.com\u003e\n Signed-off-by: Tomasz Figa \u003ctfiga@chromium.org\u003e\n Signed-off-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 298c62d3856111e6ef41c9c00a233aecf2d19651\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Thu Jul 11 16:26:43 2019 -0400\n\n media: v4l2-ctrl: Validate VP8 stateless decoder controls\n \n Only one field needs to be validated: \u0027num_dct_parts\u0027.\n This field is used to iterate over the user-provided array\n \u0027dct_part_sizes\u0027.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n [hverkuil-cisco@xs4all.nl: s -\u003e (s) in zero_padding macro]\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit a57d6acaf352d91e52271704f45c72e14cd2d98a\nAuthor: Pawel Osciak \u003cposciak@chromium.org\u003e\nDate: Thu Jul 11 16:26:42 2019 -0400\n\n media: uapi: Add VP8 stateless decoder API\n \n Add the parsed VP8 frame pixel format and controls, to be used\n with the new stateless decoder API for VP8 to provide parameters\n for accelerator (aka stateless) codecs.\n \n Reviewed-by: Tomasz Figa \u003ctfiga@chromium.org\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Reviewed-by: Nicolas Dufresne \u003cnicolas.dufresne@collabora.com\u003e\n Signed-off-by: Pawel Osciak \u003cposciak@chromium.org\u003e\n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit d1854d509d61d36af44f2130423bff8836e1592e\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Fri Jul 19 17:07:15 2019 +0800\n\n ax88179_178a: Merge memcpy + le32_to_cpus to get_unaligned_le32\n \n Merge the combo use of memcpy and le32_to_cpus.\n Use get_unaligned_le32 instead.\n This simplifies the code.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 5864118b6a442ac947207489193319489911e7b3\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Fri Jul 19 16:27:31 2019 +0800\n\n usbnet: smsc75xx: Merge memcpy + le32_to_cpus to get_unaligned_le32\n \n Merge the combo use of memcpy and le32_to_cpus.\n Use get_unaligned_le32 instead.\n This simplifies the code.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit bb448f8a60ea93722edb28418448e0008d148b0c\nAuthor: Chuhong Yuan \u003chslester96@gmail.com\u003e\nDate: Fri Jul 19 15:36:15 2019 +0800\n\n net: lan78xx: Merge memcpy + lexx_to_cpus to get_unaligned_lexx\n \n Merge the combo use of memcpy and lexx_to_cpus.\n Use get_unaligned_lexx instead.\n This simplifies the code.\n \n Signed-off-by: Chuhong Yuan \u003chslester96@gmail.com\u003e\n Acked-by: Woojung Huh \u003cwoojung.huh@microchip.com\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit de8145452eebe5510bd2b142b31560db548d6abb\nAuthor: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\nDate: Thu Jul 11 16:26:41 2019 -0400\n\n media: v4l2-ctrl: Move compound control validation\n \n Rework std_validate moving the compound controls to\n its own validation function.\n \n While here, fix the pointer math to account the index parameter.\n \n Signed-off-by: Ezequiel Garcia \u003cezequiel@collabora.com\u003e\n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 66b5f1c439843bcbab01cc7f3854ae2742f3d1e3\nAuthor: Maciej Żenczykowski \u003cmaze@google.com\u003e\nDate: Thu Jul 18 23:30:03 2019 -0700\n\n net-ipv6-ndisc: add support for RFC7710 RA Captive Portal Identifier\n \n This is trivial since we already have support for the entirely\n identical (from the kernel\u0027s point of view) RDNSS and DNSSL that\n also contain opaque data that needs to be passed down to userspace.\n \n As specified in RFC7710, Captive Portal option contains a URL.\n 8-bit identifier of the option type as assigned by the IANA is 37.\n This option should also be treated as userland.\n \n Hence, treat ND option 37 as userland (Captive Portal support)\n \n See:\n https://tools.ietf.org/html/rfc7710\n https://www.iana.org/assignments/icmpv6-parameters/icmpv6-parameters.xhtml\n \n Fixes: e35f30c131a56\n Signed-off-by: Maciej Żenczykowski \u003cmaze@google.com\u003e\n Cc: Lorenzo Colitti \u003clorenzo@google.com\u003e\n Cc: Remin Nguyen Van \u003creminv@google.com\u003e\n Cc: Alexey I. Froloff \u003craorn@raorn.name\u003e\n Signed-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n\ncommit 795130b31986e7dcb9682a37f962f49df5a25e82\nAuthor: Mike Marciniszyn \u003cmike.marciniszyn@intel.com\u003e\nDate: Mon Jul 15 12:45:52 2019 -0400\n\n IB/hfi1: Remove unused define\n \n The patch noted in Fixes missed deleting the define it obsoleted.\n \n Fixes: da9de5f8527f (\"IB/hfi1: Close PSM sdma_progress sleep window\")\n Link: https://lore.kernel.org/r/20190715164552.74174.99396.stgit@awfm-01.aw.intel.com\n Reviewed-by: Dennis Dalessandro \u003cdennis.dalessandro@intel.com\u003e\n Reviewed-by: Kaike Wan \u003ckaike.wan@intel.com\u003e\n Signed-off-by: Mike Marciniszyn \u003cmike.marciniszyn@intel.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit b2590bdd0b1dfb91737e6cb07ebb47bd74957f7e\nAuthor: Kaike Wan \u003ckaike.wan@intel.com\u003e\nDate: Mon Jul 15 12:45:46 2019 -0400\n\n IB/hfi1: Do not update hcrc for a KDETH packet during fault injection\n \n When a KDETH packet is subject to fault injection during transmission,\n HCRC is supposed to be omitted from the packet so that the hardware on the\n receiver side would drop the packet. When creating pbc, the PbcInsertHcrc\n field is set to be PBC_IHCRC_NONE if the KDETH packet is subject to fault\n injection, but overwritten with PBC_IHCRC_LKDETH when update_hcrc() is\n called later.\n \n This problem is fixed by not calling update_hcrc() when the packet is\n subject to fault injection.\n \n Fixes: 6b6cf9357f78 (\"IB/hfi1: Set PbcInsertHcrc for TID RDMA packets\")\n Cc: \u003cstable@vger.kernel.org\u003e\n Link: https://lore.kernel.org/r/20190715164546.74174.99296.stgit@awfm-01.aw.intel.com\n Reviewed-by: Mike Marciniszyn \u003cmike.marciniszyn@intel.com\u003e\n Signed-off-by: Kaike Wan \u003ckaike.wan@intel.com\u003e\n Signed-off-by: Mike Marciniszyn \u003cmike.marciniszyn@intel.com\u003e\n Signed-off-by: Jason Gunthorpe \u003cjgg@mellanox.com\u003e\n\ncommit 616cfb69b9848ad8dcb67f00c1914e69552325a2\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jun 26 03:17:47 2019 -0400\n\n media: omap3isp: set device_caps in struct video_device\n \n Instead of filling in the struct v4l2_capability device_caps\n field, fill in the struct video_device device_caps field.\n \n That way the V4L2 core knows what the capabilities of the\n video device are.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit eefe332c15967f40df7cc0cbc054523e7d6b1fc9\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jun 26 03:17:17 2019 -0400\n\n media: ti-vpe: set device_caps in struct video_device\n \n Instead of filling in the struct v4l2_capability device_caps\n field, fill in the struct video_device device_caps field.\n \n That way the V4L2 core knows what the capabilities of the\n video device are.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Acked-by: Benoit Parrot \u003cbparrot@ti.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit c7711aba06696c48343cb4616813f6de295a5573\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jun 26 03:16:54 2019 -0400\n\n media: omap_vout: set device_caps in struct video_device\n \n Instead of filling in the struct v4l2_capability device_caps\n field, fill in the struct video_device device_caps field.\n \n That way the V4L2 core knows what the capabilities of the\n video device are.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 4038b214b88e6962406bcc32e1056f3fb6eb11d7\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jun 26 03:16:36 2019 -0400\n\n media: vsp1: set device_caps in struct video_device\n \n Instead of filling in the struct v4l2_capability device_caps\n field, fill in the struct video_device device_caps field.\n \n That way the V4L2 core knows what the capabilities of the\n video device are.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 8defd70a9b2ffb49bd070dc2211d79f6cfb29d7e\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jun 26 03:16:07 2019 -0400\n\n media: sh_veu/sh_vou: set device_caps in struct video_device\n \n Instead of filling in the struct v4l2_capability device_caps\n field, fill in the struct video_device device_caps field.\n \n That way the V4L2 core knows what the capabilities of the\n video device are.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Reviewed-by: Jacopo Mondi \u003cjacopo+renesas@jmondi.org\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit b1df3129b71979b1196935a1e29bfe9ac7f28ab6\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jun 26 03:15:44 2019 -0400\n\n media: xilinx: set device_caps in struct video_device\n \n Instead of filling in the struct v4l2_capability device_caps\n field, fill in the struct video_device device_caps field.\n \n That way the V4L2 core knows what the capabilities of the\n video device are.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 0fe3331e07f0a09c10fd81e48976255b50b2e607\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jun 26 03:15:23 2019 -0400\n\n media: via-camera: set device_caps in struct video_device\n \n Instead of filling in the struct v4l2_capability device_caps\n field, fill in the struct video_device device_caps field.\n \n That way the V4L2 core knows what the capabilities of the\n video device are.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 3c7812e7ef30d177fd6234d03d1c1612447fd9f8\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jun 26 03:15:02 2019 -0400\n\n media: bdisp: set device_caps in struct video_device\n \n Instead of filling in the struct v4l2_capability device_caps\n field, fill in the struct video_device device_caps field.\n \n That way the V4L2 core knows what the capabilities of the\n video device are.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Reviewed-by: Fabien Dessenne \u003cfabien.dessenne@st.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit de5b9b71975927ca36441944e5f76126bf3e9d7c\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jun 26 03:14:39 2019 -0400\n\n media: pxa_camera: don\u0027t set caps in querycap\n \n The device_caps in struct video_device are already set correctly,\n so there is no need to set them in querycap again.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 34828fad5ecc709bd76194a1ec89424973f70c48\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jun 26 03:14:15 2019 -0400\n\n media: marvell-ccic: set device_caps in struct video_device\n \n Instead of filling in the struct v4l2_capability device_caps\n field, fill in the struct video_device device_caps field.\n \n That way the V4L2 core knows what the capabilities of the\n video device are.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Cc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit bc6be26ed825ef7a9a131637627ab4719a586f4c\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jun 26 03:13:50 2019 -0400\n\n media: mx2_emmaprp: set device_caps in struct video_device\n \n Instead of filling in the struct v4l2_capability device_caps\n field, fill in the struct video_device device_caps field.\n \n That way the V4L2 core knows what the capabilities of the\n video device are.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit f456ccc457cb2a141a5269e7fcfdaa8cbf641229\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jun 26 03:13:23 2019 -0400\n\n media: m2m-deinterlace: set device_caps in struct video_device\n \n Instead of filling in the struct v4l2_capability device_caps\n field, fill in the struct video_device device_caps field.\n \n That way the V4L2 core knows what the capabilities of the\n video device are.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit b3322e3e1aae4705f0d3ca3478a5cf484255c0a7\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jun 26 03:12:55 2019 -0400\n\n media: fsl-viu: set device_caps in struct video_device\n \n Instead of filling in the struct v4l2_capability device_caps\n field, fill in the struct video_device device_caps field.\n \n That way the V4L2 core knows what the capabilities of the\n video device are.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 994587c3b27615fd0309ce61fe2cc5695a64b934\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jun 26 02:51:41 2019 -0400\n\n media: s3c-camif/s5p-g2d/s5p-jpeg: set device_caps in struct video_device\n \n Instead of filling in the struct v4l2_capability device_caps\n field, fill in the struct video_device device_caps field.\n \n That way the V4L2 core knows what the capabilities of the\n video device are.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit e7999846b9448a84083a913f0e882a93b4d9d123\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jun 26 02:48:20 2019 -0400\n\n media: coda: set device_caps in struct video_device\n \n Instead of filling in the struct v4l2_capability device_caps\n field, fill in the struct video_device device_caps field.\n \n That way the V4L2 core knows what the capabilities of the\n video device are.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit b2778a30c73b25bfd276bc9e7fad7585f2e80dbd\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jun 26 02:46:37 2019 -0400\n\n media: am437x/davinci: set device_caps in struct video_device\n \n Instead of filling in the struct v4l2_capability device_caps\n field, fill in the struct video_device device_caps field.\n \n That way the V4L2 core knows what the capabilities of the\n video device are.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Acked-by: Lad, Prabhakar \u003cprabhakar.csengg@gmail.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 59fe916c84f891aab35019adc45377a10f5690b1\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Tue Jun 11 10:25:15 2019 -0400\n\n media: media/platform: don\u0027t set description in ENUM_FMT\n \n The V4L2 core sets the format description and flags for the driver in order\n to ensure consistent naming.\n \n So drop the strscpy of the description in drivers. Also remove any\n description strings in driver-internal structures since those are\n no longer needed.\n \n And in am437x-vpfe.c drop an unnecessary f-\u003etype assignment in\n vpfe_enum_fmt().\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Acked-by: Benoit Parrot \u003cbparrot@ti.com\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n [hverkuil-cisco@xs4all.nl: addressed some small suggestions from Laurent]\n Acked-by: Lad, Prabhakar \u003cprabhakar.csengg@gmail.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 642ac63d166d07e43396a4d8c48ab24cb072cb18\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Tue Jun 11 10:02:43 2019 -0400\n\n media: drivers/staging/media: don\u0027t set description for ENUM_FMT\n \n The V4L2 core sets the format description and flags for the driver in order\n to ensure consistent naming.\n \n So drop the strscpy of the description in drivers. Also remove any\n description strings in driver-internal structures since those are\n no longer needed.\n \n Note that bcm2835-camera.c: the formats array still stores the flags\n field for compressed formats since that information is used elsewhere\n in the driver. But enum_fmt doesn\u0027t use it anymore, since the core\n will set the COMPRESSED flag correctly.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 2169e6daa1ffa6e9869fcc56ff7df23c9287f1ec\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Tue Jun 11 09:49:53 2019 -0400\n\n media: media/pci: don\u0027t set description for ENUM_FMT\n \n The V4L2 core sets the description for the driver in order to ensure\n consistent naming.\n \n So drop the strscpy of the description in drivers. Also remove any\n description strings in driver-internal structures since those are\n no longer needed.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit ed2abfebb041473092b41527903f93390d38afa7\nAuthor: Nick Stoughton \u003cnstoughton@logitech.com\u003e\nDate: Wed Jul 17 14:56:06 2019 -0700\n\n leds: leds-lp5562 allow firmware files up to the maximum length\n \n Firmware files are in ASCII, using 2 hex characters per byte. The\n maximum length of a firmware string is therefore\n \n 16 (commands) * 2 (bytes per command) * 2 (characters per byte) \u003d 64\n \n Fixes: ff45262a85db (\"leds: add new LP5562 LED driver\")\n Signed-off-by: Nick Stoughton \u003cnstoughton@logitech.com\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 79937a4beeaeb3e2584db5dfe3a7c9ee3d5f97f6\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Jul 16 12:54:24 2019 +0530\n\n leds: ns2: Add of_node_put() before return\n \n Each iteration of for_each_child_of_node puts the previous node, but in\n the case of a return from the middle of the loop, there is no put, thus\n causing a memory leak. Hence create a new label, err_node_put, that puts\n the previous node before returning the required value. Edit the mid-loop\n return sites to instead go to this new label.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 730f693d79da8130c8ac18d1083cdb333fd8d7c8\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Thu Jul 11 13:24:05 2019 +0530\n\n leds: max77650: Add of_node_put() before return\n \n Each iteration of for_each_child_of_node puts the previous node, but in\n the case of a return from the middle of the loop, there is no put, thus\n causing a memory leak.\n Hence create a new label, err_node_put, which puts the previous node and\n returns variable rv. Modify the mid-loop return statements to instead\n store the return value in rv and jump to err_node_put.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Reviewed-by: Dan Murphy \u003cdmurphy@ti.com\u003e\n Acked-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 533016c870a8d0cc01478ed9e780a46e830b3b22\nAuthor: Joe Perches \u003cjoe@perches.com\u003e\nDate: Thu Jul 4 16:57:42 2019 -0700\n\n leds: as3645a: Fix misuse of strlcpy\n \n Probable cut\u0026paste typo - use the correct field size.\n \n Signed-off-by: Joe Perches \u003cjoe@perches.com\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 51251542c9688c399e14d852258a675b331e761b\nAuthor: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\nDate: Wed Jul 3 10:47:38 2019 +0200\n\n leds: max77650: add MODULE_ALIAS()\n \n Define a MODULE_ALIAS() in the LED sub-driver for max77650 so that\n the appropriate module gets loaded together with the core mfd driver.\n \n Signed-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n Acked-by: Pavel Machek \u003cpavel@ucw.cz\u003e\n Signed-off-by: Jacek Anaszewski \u003cjacek.anaszewski@gmail.com\u003e\n\ncommit 009cb7d5794aaf40b037857510c59847298747dd\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Tue Jun 11 09:48:54 2019 -0400\n\n media: media/usb: don\u0027t set description in ENUM_FMT\n \n The V4L2 core sets the description for the driver in order to ensure\n consistent naming.\n \n So drop the strscpy of the description in drivers. Also remove any\n description strings in driver-internal structures since those are\n no longer needed.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit 4747bd0f4179858d00498becc3b26c92c8e9724a\nAuthor: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\nDate: Wed Jun 26 07:07:45 2019 -0400\n\n media: v4l2-ioctl: add missing pixelformats\n \n The v4l_fill_fmtdesc() is supposed to be updated whenever a new pixelformat\n is added, but a bunch of recently added pixelformats were forgotten.\n \n Update the list.\n \n Also change a few lower case words to upper case to keep the same style.\n \n Signed-off-by: Hans Verkuil \u003chverkuil-cisco@xs4all.nl\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n [hverkuil-cisco@xs4all.nl: changed 1-5-5-5 to 5-5-5-1 as per Laurent\u0027s suggestion]\n Signed-off-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n\ncommit f51edcec5288556069d27ab7040a4ce8ca55b74e\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Thu Jun 27 12:53:02 2019 -0700\n\n MAINTAINERS: Add an entry for generic architecture topology\n \n arm and arm64 shared lot of CPU topology related code. This was\n consolidated under driver/base/arch_topology.c by Juri. Now RISC-V\n is also started sharing the same code pulling more code from arm64\n into arch_topology.c\n \n Since I was involved in the review from the beginning, I would like\n to assume maintenance for the same.\n \n Cc: Will Deacon \u003cwill.deacon@arm.com\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Acked-by: Juri Lelli \u003cjuri.lelli@redhat.com\u003e\n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Acked-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Paul Walmsley \u003cpaul.walmsley@sifive.com\u003e\n\ncommit c181831f161c8da460b1ffa26632448323c3bfcc\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Thu Jun 27 12:53:01 2019 -0700\n\n base: arch_topology: update Kconfig help description\n \n Commit 5d777b185f6d (\"arch_topology: Make cpu_capacity sysfs node as read-only\")\n made cpu_capacity sysfs node read-only. Update the GENERIC_ARCH_TOPOLOGY\n Kconfig help section to reflect the same.\n \n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Signed-off-by: Paul Walmsley \u003cpaul.walmsley@sifive.com\u003e\n\ncommit 03f11f03dbfe37c0bff2768b8f2e277a29ac62b9\nAuthor: Atish Patra \u003catish.patra@wdc.com\u003e\nDate: Thu Jun 27 12:53:00 2019 -0700\n\n RISC-V: Parse cpu topology during boot.\n \n Currently, there are no topology defined for RISC-V.\n Parse the cpu-map node from device tree and setup the\n cpu topology.\n \n CPU topology after applying the patch.\n $cat /sys/devices/system/cpu/cpu2/topology/core_siblings_list\n 0-3\n $cat /sys/devices/system/cpu/cpu3/topology/core_siblings_list\n 0-3\n $cat /sys/devices/system/cpu/cpu3/topology/physical_package_id\n 0\n $cat /sys/devices/system/cpu/cpu3/topology/core_id\n 3\n \n Signed-off-by: Atish Patra \u003catish.patra@wdc.com\u003e\n Acked-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Signed-off-by: Paul Walmsley \u003cpaul.walmsley@sifive.com\u003e\n\ncommit ca74b316df96d7c40ee3e8301065607c11c60c27\nAuthor: Atish Patra \u003catish.patra@wdc.com\u003e\nDate: Thu Jun 27 12:52:59 2019 -0700\n\n arm: Use common cpu_topology structure and functions.\n \n Currently, ARM32 and ARM64 uses different data structures to represent\n their cpu topologies. Since, we are moving the ARM64 topology to common\n code to be used by other architectures, we can reuse that for ARM32 as\n well.\n \n Take this opprtunity to remove the redundant functions from ARM32 and\n reuse the common code instead.\n \n To: Russell King \u003clinux@armlinux.org.uk\u003e\n Signed-off-by: Atish Patra \u003catish.patra@wdc.com\u003e\n Tested-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e (on TC2)\n Reviewed-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Signed-off-by: Paul Walmsley \u003cpaul.walmsley@sifive.com\u003e\n\ncommit 60c1b220d8bc6baeaf837cd60f94a331b25c26bc\nAuthor: Atish Patra \u003catish.patra@wdc.com\u003e\nDate: Thu Jun 27 12:52:58 2019 -0700\n\n cpu-topology: Move cpu topology code to common code.\n \n Both RISC-V \u0026 ARM64 are using cpu-map device tree to describe\n their cpu topology. It\u0027s better to move the relevant code to\n a common place instead of duplicate code.\n \n To: Will Deacon \u003cwill.deacon@arm.com\u003e\n To: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n Signed-off-by: Atish Patra \u003catish.patra@wdc.com\u003e\n [Tested on QDF2400]\n Tested-by: Jeffrey Hugo \u003cjhugo@codeaurora.org\u003e\n [Tested on Juno and other embedded platforms.]\n Tested-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Reviewed-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Acked-by: Will Deacon \u003cwill.deacon@arm.com\u003e\n Acked-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Paul Walmsley \u003cpaul.walmsley@sifive.com\u003e\n\ncommit 124e46a86580c71e0eee8459c5da7649318118db\nAuthor: Atish Patra \u003catish.patra@wdc.com\u003e\nDate: Thu Jun 27 12:52:57 2019 -0700\n\n dt-binding: cpu-topology: Move cpu-map to a common binding.\n \n cpu-map binding can be used to described cpu topology for both\n RISC-V \u0026 ARM. It makes more sense to move the binding to document\n to a common place.\n \n The relevant discussion can be found here.\n https://lkml.org/lkml/2018/11/6/19\n \n Signed-off-by: Atish Patra \u003catish.patra@wdc.com\u003e\n Reviewed-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Paul Walmsley \u003cpaul.walmsley@sifive.com\u003e\n\ncommit 748fd07e2b9ca4132e3d2aae25395aedc4d1aee8\nAuthor: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\nDate: Mon Jul 22 14:52:09 2019 +0100\n\n ASoC: madera: Read device tree configuration\n \n Read the configuration of the Madera ASoC driver from device tree.\n \n Signed-off-by: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n Link: https://lore.kernel.org/r/20190722135209.30302-1-ckeepax@opensource.cirrus.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit c8b08fc0d6f83427973b94694b7ec8855d2d1e37\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Fri May 17 14:09:23 2019 -0700\n\n firmware: qcom_scm: Fix some typos in docs and printks\n \n Some words are misspelled and we put a full stop after a return value\n integer. Fix these things up so it doesn\u0027t look so odd.\n \n Cc: Ian Jackson \u003cian.jackson@citrix.com\u003e\n Cc: Julien Grall \u003cjulien.grall@arm.com\u003e\n Cc: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Cc: Avaneesh Kumar Dwivedi \u003cakdwived@codeaurora.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 6e37ccf78a53296c6c7bf426065762c27829eb84\nAuthor: Stephen Boyd \u003cswboyd@chromium.org\u003e\nDate: Fri May 17 14:09:21 2019 -0700\n\n firmware: qcom_scm: Use proper types for dma mappings\n \n We need to use the proper types and convert between physical addresses\n and dma addresses here to avoid mismatch warnings. This is especially\n important on systems with a different size for dma addresses and\n physical addresses. Otherwise, we get the following warning:\n \n drivers/firmware/qcom_scm.c: In function \"qcom_scm_assign_mem\":\n drivers/firmware/qcom_scm.c:469:47: error: passing argument 3 of \"dma_alloc_coherent\" from incompatible pointer type [-Werror\u003dincompatible-pointer-types]\n \n We also fix the size argument to dma_free_coherent() because that size\n doesn\u0027t need to be aligned after it\u0027s already aligned on the allocation\n size. In fact, dma debugging expects the same arguments to be passed to\n both the allocation and freeing sides of the functions so changing the\n size is incorrect regardless.\n \n Reported-by: Ian Jackson \u003cian.jackson@citrix.com\u003e\n Cc: Ian Jackson \u003cian.jackson@citrix.com\u003e\n Cc: Julien Grall \u003cjulien.grall@arm.com\u003e\n Cc: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Cc: Avaneesh Kumar Dwivedi \u003cakdwived@codeaurora.org\u003e\n Tested-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Signed-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\n Signed-off-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n\ncommit 849b384f92bcadf2bd967f81ceeff815c9cd6af9\nAuthor: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nDate: Thu Jun 27 12:52:56 2019 -0700\n\n Documentation: DT: arm: add support for sockets defining package boundaries\n \n The current ARM DT topology description provides the operating system\n with a topological view of the system that is based on leaf nodes\n representing either cores or threads (in an SMT system) and a\n hierarchical set of cluster nodes that creates a hierarchical topology\n view of how those cores and threads are grouped.\n \n However this hierarchical representation of clusters does not allow to\n describe what topology level actually represents the physical package or\n the socket boundary, which is a key piece of information to be used by\n an operating system to optimize resource allocation and scheduling.\n \n Lets add a new \"socket\" node type in the cpu-map node to describe the\n same.\n \n Signed-off-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Paul Walmsley \u003cpaul.walmsley@sifive.com\u003e\n\ncommit 72503b25ee363827aafffc3e8d872e6a92a7e422\nAuthor: Stefan Wahren \u003cwahrenst@gmx.net\u003e\nDate: Tue Jul 16 19:15:18 2019 +0200\n\n dmaengine: bcm2835: Print error in case setting DMA mask fails\n \n During enabling of the RPi 4, we found out that the driver doesn\u0027t provide\n a helpful error message in case setting DMA mask fails. So add one.\n \n Signed-off-by: Stefan Wahren \u003cwahrenst@gmx.net\u003e\n Link: https://lore.kernel.org/r/1563297318-4900-1-git-send-email-wahrenst@gmx.net\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit d2bfe7b5d182dad54aedb9dcdb736e8816ead1c3\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Mon Jul 22 10:16:45 2019 +0200\n\n dmaengine: edma: make edma_filter_fn private\n \n With the audio driver no longer referring to this function, it\n can be made private to the dmaengine driver itself, and the\n header file removed.\n \n Acked-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Link: https://lore.kernel.org/r/20190722081705.2084961-2-arnd@arndb.de\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 9c71b9eb3cb25856e29f0486eae9ee1ba864ba51\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Mon Jul 22 10:16:44 2019 +0200\n\n dmaengine: omap-dma: make omap_dma_filter_fn private\n \n With the audio driver no longer referring to this function, it\n can be made private to the dmaengine driver itself, and the\n header file removed.\n \n Acked-by: Peter Ujfalusi \u003cpeter.ujfalusi@ti.com\u003e\n Link: https://lore.kernel.org/lkml/20190307151646.1016966-1-arnd@arndb.de/\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Link: https://lore.kernel.org/r/20190722081705.2084961-1-arnd@arndb.de\n Signed-off-by: Vinod Koul \u003cvkoul@kernel.org\u003e\n\ncommit 042c805545d04f8d358475c83a54178da14f60ed\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jul 22 16:08:17 2019 +0200\n\n arm64: dts: allwinner: h6: Fix SID node name\n \n The SID node one the H6 doesn\u0027t have a standard node name. Switch to the\n one we use for the other SoCs.\n \n Acked-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 5ea40f7106aaa4e2736d18075ca635389797fc16\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jul 22 16:08:16 2019 +0200\n\n ARM: dts: sunxi: Unify the DE2 bus clocks order\n \n The DE2 bus takes two clocks, named bus and mod according to the binding.\n \n However, the order of these clocks change from one SoC to another. Even\n though it might not be an issue in most cases, having consistency will help\n if we ever need to have some code to deal with deprecated bindings, and in\n general it\u0027s just better.\n \n Acked-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit d2d5a6b0253d30ef04d513447cdd00df4fa627c3\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Jul 22 12:37:10 2019 +0900\n\n ALSA: firewire-motu: more code refactoring for MOTU data block processing layer\n \n MOTU data block processing layer has some tracepoints events. This\n commit is code refactoring to split probing the events from processing\n data blocks.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 9a738ad1b1a64ef6d8271d8d42d715b75db6eb0d\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Jul 22 12:37:09 2019 +0900\n\n ALSA: firewire-lib: process payload of isoc context according to packet descriptors\n \n This commit changes each of data block processing layer so that it\n receives list of packet descriptor.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 4731c672c930e490515f1747fb34d56d61ae57ac\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Jul 22 12:37:08 2019 +0900\n\n ALSA: fireface: code refactoring for FF data block processing layer\n \n This is code refactoring for FF data block processing layer so that\n it can receive list of packet descriptor.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 00d004db6493c62b8bbd7073a9b3b7290cbe0cea\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Jul 22 12:37:07 2019 +0900\n\n ALSA: firewire-motu: code refactoring for MOTU data block processing layer\n \n This is code refactoring for MOTU data block processing layer so that\n it can receive list of packet descriptor.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 806653b70ee21b374d2844d144609b2e0070a86b\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Jul 22 12:37:06 2019 +0900\n\n ALSA: firewire-tascam: code refactoring for TASCAM data block processing layer\n \n This is code refactoring for TASCAM data block processing layer so that\n it can receive list of packet descriptor.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit bb473966774a79ae9a826f721059496d9b393acc\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Jul 22 12:37:05 2019 +0900\n\n ALSA: firewire-digi00x: code refactoring for DOT data block processing layer\n \n This is code refactoring for DOT data block processing layer so that\n it can receive list of packet descriptor.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 9fc90644cae9c0abfcce0576ed0a14cca67f69b9\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Jul 22 12:37:04 2019 +0900\n\n ALSA: firewire-lib: code refactoring for AM824 data block processing layer\n \n This is code refactoring for AM824 data block processing layer so that\n it can receive list of packet descriptor.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit d2c104a3426be9991b35c65f0f260a107c4b2942\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Jul 22 12:37:03 2019 +0900\n\n ALSA: firewire-lib: pass packet descriptor to data block processing layer\n \n This commit changes signature of callback function to call data block\n processing layer with packet descriptor. At present, the layer is called\n per packet.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 0f5cfcb24d9c931e4571e9995bbfc08cc76d5d4a\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Jul 22 12:37:02 2019 +0900\n\n ALSA: firewire-lib: code refactoring to process context payloads\n \n This is code refactoring for common processing for payloads of\n ishocornous context.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 5e2ece0fdceb02f6c47f53bef309395701a65393\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Jul 22 12:37:01 2019 +0900\n\n ALSA: firewire-lib: code refactoring to process PCM substream\n \n This is code refactoring to separate PCM substream processing from packet\n queueing.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 753e717986c24b60962367708479555212730b9f\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Jul 22 12:37:00 2019 +0900\n\n ALSA: firewire-lib: use packet descriptor for IR context\n \n This commit uses packet descriptor to parse headers of IR context and\n parse timing information for sequence transferred by target device.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit f4f6ae7b7c1fdbaaaecafd183fabcf5314df2b59\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Jul 22 12:36:59 2019 +0900\n\n ALSA: firewire-lib: use packet descriptor for IT context\n \n This commit uses packet descriptor to parse headers of IT context and\n generate timing information for ideal sequence.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 04130cf8e296bfbd65d2673a5975bd0f200b941d\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Jul 22 12:36:58 2019 +0900\n\n ALSA: firewire-lib: add list of packet descriptor\n \n In current implementation of ALSA IEC 61883-1/6 packet streaming engine,\n 16 packets are handled in one interrupt of isochronous context of OHCI\n 1394.\n \n Overall packet processing runs for each. However, this is not better in\n a point to split the processing into several parts.\n \n This commit is an attempt to add intermediate representation for\n parameters required for the parts.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 600c8018df605a91575aef6811cf927e2d933d30\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Jul 22 12:36:57 2019 +0900\n\n ALSA: firewire-lib: pass no syt information to data block processing layer\n \n In a previous commit, the variable passed from packet streaming layer\n for syt variable is useless. This commit obsoletes it.\n \n In my future work, the syt information is passed to data block processing\n layer by another way.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 3baf30532f94030244a31a275b546d042656c535\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Jul 22 12:36:56 2019 +0900\n\n ALSA: firewire-lib: add syt_override member for some protocols\n \n Some protocols don\u0027t use syt field of CIP header to represent\n presentation timestamp. For such protocol, ALSA IEC 61883-1/6\n packet streaming engine uses a pointer into local variable for\n the value of syt to call data block processing layer. However,\n it can decide the value when initializing packet streaming\n layer.\n \n This commit adds \u0027syt_override\u0027 member for packet streaming\n layer.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit a35463d1f77d9e5244d6f4953d3c7f3a27c7625e\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Jul 22 12:36:55 2019 +0900\n\n ALSA: firewire-lib: operate data block counter in top level of processing for IR context\n \n In ALSA IEC 61883-1/6 packet streaming engine, two types of data block\n counter are supported. This commit applies code refactoring to make it\n explicitly for IR context.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 860d798c370409efbedf66bae2d41b3cbdf14a79\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Jul 22 12:36:54 2019 +0900\n\n ALSA: firewire-lib: operate data block counter in top level of processing for IT context\n \n In ALSA IEC 61883-1/6 packet streaming engine, two types of data block\n counter are supported. This commit applies code refactoring to make it\n explicitly for IT context.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit ab75481202d2925a017e4cb5407b0bb49c356de8\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Jul 22 12:36:53 2019 +0900\n\n ALSA: firewire-lib: pass data block counter to data block processing layer\n \n This is a preparation for future commit that \u0027struct\n amdtp_stream.data_block_count\u0027 does not represent the value of\n data block count for current data block.\n \n However, data block count is required for calculation of sequence index\n in sequence-multiplied data channel. Some of data block processing layer\n require it; e.g. for AM824 data block.\n \n This commit passes data block count to the processing layer.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit cdefaa44d535a9d555ac4f1925d59ae4d0ef98e8\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Jul 22 12:36:52 2019 +0900\n\n ALSA: firewire-lib: pass data block count as an argument to tracepoints event\n \n This is a preparation for future commit that \u0027struct\n amdtp_stream.data_block_count\u0027 does not represent the value of\n data block count for current data block.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 588f2e2caf6795ca29d50a45ea1e1438274e75e0\nAuthor: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\nDate: Mon Jul 22 12:36:51 2019 +0900\n\n ALSA: firewire-lib: obsolete ctx_data.tx.first_dbc with CIP_UNALIGHED_DBC flag\n \n Recent firmware for Fireworks board module have a quirk to start\n transmission of CIP with non-zero value for its data block counter.\n In current implementation of ALSA firewire stack, the quirk is handled\n by \u0027struct amdtp_stream.ctx_data.tx.first_dbc\u0027 with value 0x02. However,\n the value comes from reverse engineering. It\u0027s better to handle this\n quirk without the explicit value.\n \n In a process to parse CIP header, the quirk of data block counter\n affects decision of sequence index in sequence-multiplexed data channel;\n i.e. MIDI conformant data channel. In Fireworks, the index is decided\n by the number of data blocks from top of the same CIP, thus the value\n of data block counter is useless.\n \n This commit adds CIP_UNALIGHED_DBC flag and obsoletes the explicit\n value for this quirk.\n \n Signed-off-by: Takashi Sakamoto \u003co-takashi@sakamocchi.jp\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 66ad66293edeb2bcc27a329f35018b18601da452\nAuthor: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\nDate: Thu Jul 11 10:29:36 2019 +0200\n\n gpio: em: use a helper variable for \u0026pdev-\u003edev\n \n Instead of always dereferencing \u0026pdev-\u003edev, just assign a helper local\n variable of type struct device * and use it where applicable.\n \n Signed-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n Reviewed-by: Geert Uytterhoeven \u003cgeert+renesas@glider.be\u003e\n\ncommit 10742fee98eb3b3e8453ef27a33dee314b15f7bd\nAuthor: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\nDate: Wed Jul 3 14:07:54 2019 +0200\n\n eeprom: at24: remove unneeded include\n \n We used to have a call to ilog2() in AT24_DEVICE_MAGIC(). That\u0027s long\n gone so this header is no longer needed.\n \n Signed-off-by: Bartosz Golaszewski \u003cbgolaszewski@baylibre.com\u003e\n Acked-by: Wolfram Sang \u003cwsa@the-dreams.de\u003e\n\ncommit b6319b061ba279577fd7030a9848fbd6a17151e3\nAuthor: Oleksandr Suvorov \u003coleksandr.suvorov@toradex.com\u003e\nDate: Fri Jul 19 10:05:37 2019 +0000\n\n ASoC: sgtl5000: Fix charge pump source assignment\n \n If VDDA !\u003d VDDIO and any of them is greater than 3.1V, charge pump\n source can be assigned automatically [1].\n \n [1] https://www.nxp.com/docs/en/data-sheet/SGTL5000.pdf\n \n Signed-off-by: Oleksandr Suvorov \u003coleksandr.suvorov@toradex.com\u003e\n Reviewed-by: Marcel Ziswiler \u003cmarcel.ziswiler@toradex.com\u003e\n Reviewed-by: Igor Opaniuk \u003cigor.opaniuk@toradex.com\u003e\n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190719100524.23300-7-oleksandr.suvorov@toradex.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 631bc8f0134ae9620d86a96b8c5f9445d91a2dca\nAuthor: Oleksandr Suvorov \u003coleksandr.suvorov@toradex.com\u003e\nDate: Fri Jul 19 10:05:35 2019 +0000\n\n ASoC: sgtl5000: Fix of unmute outputs on probe\n \n To enable \"zero cross detect\" for ADC/HP, change\n HP_ZCD_EN/ADC_ZCD_EN bits only instead of writing the whole\n CHIP_ANA_CTRL register.\n \n Signed-off-by: Oleksandr Suvorov \u003coleksandr.suvorov@toradex.com\u003e\n Reviewed-by: Marcel Ziswiler \u003cmarcel.ziswiler@toradex.com\u003e\n Reviewed-by: Igor Opaniuk \u003cigor.opaniuk@toradex.com\u003e\n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190719100524.23300-6-oleksandr.suvorov@toradex.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 694b14554d75f2a1ae111202e71860d58b434a21\nAuthor: Oleksandr Suvorov \u003coleksandr.suvorov@toradex.com\u003e\nDate: Fri Jul 19 10:05:34 2019 +0000\n\n ASoC: sgtl5000: add ADC mute control\n \n This control mute/unmute the ADC input of SGTL5000\n using its CHIP_ANA_CTRL register.\n \n Signed-off-by: Oleksandr Suvorov \u003coleksandr.suvorov@toradex.com\u003e\n Reviewed-by: Marcel Ziswiler \u003cmarcel.ziswiler@toradex.com\u003e\n Reviewed-by: Igor Opaniuk \u003cigor.opaniuk@toradex.com\u003e\n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190719100524.23300-5-oleksandr.suvorov@toradex.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit cadee10c983145c16d6c75af88f6a943b22ebeb2\nAuthor: Oleksandr Suvorov \u003coleksandr.suvorov@toradex.com\u003e\nDate: Fri Jul 19 10:05:33 2019 +0000\n\n ASoC: sgtl5000: Fix definition of VAG Ramp Control\n \n SGTL5000_SMALL_POP is a bit mask, not a value. Usage of\n correct definition makes device probing code more clear.\n \n Signed-off-by: Oleksandr Suvorov \u003coleksandr.suvorov@toradex.com\u003e\n Reviewed-by: Marcel Ziswiler \u003cmarcel.ziswiler@toradex.com\u003e\n Reviewed-by: Igor Opaniuk \u003cigor.opaniuk@toradex.com\u003e\n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190719100524.23300-4-oleksandr.suvorov@toradex.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit b1f373a11d25fc9a5f7679c9b85799fe09b0dc4a\nAuthor: Oleksandr Suvorov \u003coleksandr.suvorov@toradex.com\u003e\nDate: Fri Jul 19 10:05:31 2019 +0000\n\n ASoC: sgtl5000: Improve VAG power and mute control\n \n VAG power control is improved to fit the manual [1]. This patch fixes as\n minimum one bug: if customer muxes Headphone to Line-In right after boot,\n the VAG power remains off that leads to poor sound quality from line-in.\n \n I.e. after boot:\n - Connect sound source to Line-In jack;\n - Connect headphone to HP jack;\n - Run following commands:\n $ amixer set \u0027Headphone\u0027 80%\n $ amixer set \u0027Headphone Mux\u0027 LINE_IN\n \n Change VAG power on/off control according to the following algorithm:\n - turn VAG power ON on the 1st incoming event.\n - keep it ON if there is any active VAG consumer (ADC/DAC/HP/Line-In).\n - turn VAG power OFF when there is the latest consumer\u0027s pre-down event\n come.\n - always delay after VAG power OFF to avoid pop.\n - delay after VAG power ON if the initiative consumer is Line-In, this\n prevents pop during line-in muxing.\n \n According to the data sheet [1], to avoid any pops/clicks,\n the outputs should be muted during input/output\n routing changes.\n \n [1] https://www.nxp.com/docs/en/data-sheet/SGTL5000.pdf\n \n Cc: stable@vger.kernel.org\n Fixes: 9b34e6cc3bc2 (\"ASoC: Add Freescale SGTL5000 codec support\")\n Signed-off-by: Oleksandr Suvorov \u003coleksandr.suvorov@toradex.com\u003e\n Reviewed-by: Marcel Ziswiler \u003cmarcel.ziswiler@toradex.com\u003e\n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Reviewed-by: Cezary Rojewski \u003ccezary.rojewski@intel.com\u003e\n Link: https://lore.kernel.org/r/20190719100524.23300-3-oleksandr.suvorov@toradex.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit cfc8f568aada98f9608a0a62511ca18d647613e2\nAuthor: Oleksandr Suvorov \u003coleksandr.suvorov@toradex.com\u003e\nDate: Fri Jul 19 10:05:30 2019 +0000\n\n ASoC: Define a set of DAPM pre/post-up events\n \n Prepare to use SND_SOC_DAPM_PRE_POST_PMU definition to\n reduce coming code size and make it more readable.\n \n Cc: stable@vger.kernel.org\n Signed-off-by: Oleksandr Suvorov \u003coleksandr.suvorov@toradex.com\u003e\n Reviewed-by: Marcel Ziswiler \u003cmarcel.ziswiler@toradex.com\u003e\n Reviewed-by: Igor Opaniuk \u003cigor.opaniuk@toradex.com\u003e\n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190719100524.23300-2-oleksandr.suvorov@toradex.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 4aefb7387198740318c5e7bf7211701f4e0d849e\nAuthor: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\nDate: Fri Jul 19 20:05:58 2019 +0200\n\n ASoC: rockchip-max98090: Remove MICBIAS as supply of input pin IN34\n \n Commit ec0d23b295b9 (\"ASoC: rockchip-max98090: Fix the Headset Mic\n route.\") moved the MICBIAS widget to supply Headset Mic but forget to\n remove the MICBIAS widget to supply IN34 which is not really needed, so\n remove that path so we have:\n \n IN34 \u003c\u003d\u003d\u003d\u003d Headset MIC \u003c\u003d\u003d\u003d\u003d MICBIAS\n \n Signed-off-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Link: https://lore.kernel.org/r/20190719180558.11459-1-enric.balletbo@collabora.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 03b0deb6ba5f47d346fbb0783018c0d415203142\nAuthor: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\nDate: Mon Jul 22 10:35:16 2019 +0100\n\n ASoC: cs47l35: Use define for DSPCLK enable bit\n \n Signed-off-by: Charles Keepax \u003cckeepax@opensource.cirrus.com\u003e\n Link: https://lore.kernel.org/r/20190722093516.12906-1-ckeepax@opensource.cirrus.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 730e051c78b05f0ce87df06e0fc1ea002d722fe8\nAuthor: Shuming Fan \u003cshumingf@realtek.com\u003e\nDate: Fri Jul 19 14:32:49 2019 +0800\n\n ASoC: rt1308: add silence detection and manual PDB control\n \n We enable the silence detection function in initial settings.\n PDB control changes to manual mode, hence the driver could\n fully control the AMP output on/off.\n \n Signed-off-by: Shuming Fan \u003cshumingf@realtek.com\u003e\n Link: https://lore.kernel.org/r/20190719063249.18806-1-shumingf@realtek.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 980c10cda954770640d9d7cd7b793efa12cae0d9\nAuthor: Shuming Fan \u003cshumingf@realtek.com\u003e\nDate: Fri Jul 19 14:32:35 2019 +0800\n\n ASoC: rt1308: simplify the EFUSE read function\n \n The rt1308_efuse function contains many redundant settings.\n We remove the redundant settings and the function still works.\n \n Signed-off-by: Shuming Fan \u003cshumingf@realtek.com\u003e\n Link: https://lore.kernel.org/r/20190719063235.18757-1-shumingf@realtek.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit b7e814deae33eb30f8f8c6528e8e69b107978d88\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Fri Jul 19 16:36:37 2019 +0200\n\n ASoC: tlv320aic31xx: suppress error message for EPROBE_DEFER\n \n Both the supplies and reset GPIO might need a probe deferral for the\n resource to be available. Don\u0027t print a error message in that case, as\n it is a normal operating condition.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Acked-by: Andrew F. Davis \u003cafd@ti.com\u003e\n Link: https://lore.kernel.org/r/20190719143637.2018-1-l.stach@pengutronix.de\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 96cd3b97bf32557ba722c2e96095cb6361b5dc42\nAuthor: fengchunguo \u003cchunguo.feng@amlogic.com\u003e\nDate: Mon Jul 15 17:55:56 2019 +0800\n\n ASoC: max98383: fix i2c probe failure\n \n Added reset_gpio configuration for i2c probe successfully.\n If not,i2c address can\u0027t be found rightly.\n \n Error information:\n max98373 3-0031: Failed to read: 0x21FF\n \n Fixed:\n [3.761299@3] max98373 3-0031: MAX98373 revisionID: 0x43\n [3.828911@3] asoc-aml-card auge_sound: max98373-aif1 \u003c-\u003e TDM-B mapping ok\n \n Signed-off-by: fengchunguo \u003cchunguo.feng@amlogic.com\u003e\n Link: https://lore.kernel.org/r/20190715095556.1614-1-chunguo.feng@amlogic.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 6d19d8a3cec74a9680947ecb6abdeda38583110e\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Wed Jul 17 12:51:56 2019 +0200\n\n ASoC: fsl_sai: mark regmap as fast_io\n \n The regmap is only ever used to access MMIO registers, so it\u0027s fair\n to say that register access is fast.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Link: https://lore.kernel.org/r/20190717105156.15721-1-l.stach@pengutronix.de\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit bd517707d85f19a7339ea8b882fcbf0fd9976bd6\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Wed Jul 17 12:56:34 2019 +0200\n\n ASoC: fsl_sai: derive TX FIFO watermark from FIFO depth\n \n The DMA request schould be triggered as soon as the FIFO has space\n for another burst. As different versions of the SAI block have\n different FIFO sizes, the watrmark level needs to be derived from\n version specific data.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Tested-by: Angus Ainslie \u003cangus@akkea.ca\u003e\n Reviewed-by: Angus Ainslie \u003cangus@akkea.ca\u003e\n Link: https://lore.kernel.org/r/20190717105635.18514-3-l.stach@pengutronix.de\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 89c9679f699d88986ce552738dc7c5c500c8fc67\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Wed Jul 17 12:56:33 2019 +0200\n\n ASoC: fsl_sai: add of_match data\n \n New revisions of the SAI IP block have even more differences that need\n be taken into account by the driver. To avoid sprinking compatible\n checks all over the driver move the current differences into of_match_data.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Tested-by: Angus Ainslie \u003cangus@akkea.ca\u003e\n Reviewed-by: Angus Ainslie \u003cangus@akkea.ca\u003e\n Reviewed-by: Daniel Baluta \u003cdaniel.baluta@nxp.com\u003e\n Link: https://lore.kernel.org/r/20190717105635.18514-2-l.stach@pengutronix.de\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit a962a809e5e45715081fa247726a562b8dbd082a\nAuthor: Tzung-Bi Shih \u003ctzungbi@google.com\u003e\nDate: Tue Jul 16 11:24:17 2019 +0800\n\n ASoC: mediatek: mt8183: make headset codec optional\n \n Make headset codec optional because some variant machines may not\n have an audio jack.\n \n Signed-off-by: Tzung-Bi Shih \u003ctzungbi@google.com\u003e\n Link: https://lore.kernel.org/r/20190716032417.19015-1-tzungbi@google.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit af62a3c8a872631e63833e2a99a8f43f81532fae\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Wed Jul 10 07:46:27 2019 +0530\n\n ASoC: wcd9335: add irqflag IRQF_ONESHOT flag\n \n Add IRQF_ONESHOT to ensure \"Interrupt is not reenabled after the hardirq\n handler finished\".\n \n fixes below issue reported by coccicheck\n \n sound/soc/codecs/wcd9335.c:4068:8-33: ERROR: Threaded IRQ with no\n primary handler requested without IRQF_ONESHOT\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190710021627.GA13396@hari-Inspiron-1545\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit e1fb714bddf452ef0ecf67c420957438977f107d\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Wed Jul 10 00:12:37 2019 +0530\n\n ASoC: bcm: cygnus-pcm: Unneeded variable: \"ret\".\n \n This patch fixes below issues reported by coccicheck\n \n sound/soc/bcm/cygnus-pcm.c:642:5-8: Unneeded variable: \"ret\". Return \"0\"\n on line 650\n sound/soc/bcm/cygnus-pcm.c:671:5-8: Unneeded variable: \"ret\". Return \"0\"\n on line 696\n \n We cannot change return type of these functions as they are callback\n functions of snd_pcm_ops\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190709184236.GA7873@hari-Inspiron-1545\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 3da9834d9381dd99273f2ad4e6d096c9187dc4f2\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Wed Jul 10 14:42:43 2019 +0300\n\n spi: dw-mmio: Clock should be shut when error occurs\n \n When optional clock requesting fails, the main clock is still up and running,\n we should shut it down in such caee.\n \n Fixes: 560ee7e91009 (\"spi: dw: Add support for an optional interface clock\")\n Cc: Phil Edworthy \u003cphil.edworthy@renesas.com\u003e\n Cc: Gareth Williams \u003cgareth.williams.jx@renesas.com\u003e\n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Reviewed-by: Gareth Williams \u003cgareth.williams.jx@renesas.com\u003e\n Link: https://lore.kernel.org/r/20190710114243.30101-1-andriy.shevchenko@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 0521050215d0daae5d84b212fbe9fc16462f58f6\nAuthor: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nDate: Wed Jul 10 14:42:30 2019 +0300\n\n spi: dw-mmio: Use devm_platform_ioremap_resource()\n \n Use the new helper that wraps the calls to platform_get_resource()\n and devm_ioremap_resource() together.\n \n Signed-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Link: https://lore.kernel.org/r/20190710114230.30047-1-andriy.shevchenko@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit d850c6f5fad60b6edec08300977303aae855ffff\nAuthor: Axel Lin \u003caxel.lin@ingics.com\u003e\nDate: Thu Jul 11 20:21:38 2019 +0800\n\n regulator: rk808: Return REGULATOR_MODE_INVALID for invalid mode\n \n -EINVAL is not a valid return value for .of_map_mode, return\n REGULATOR_MODE_INVALID instead.\n \n Signed-off-by: Axel Lin \u003caxel.lin@ingics.com\u003e\n Link: https://lore.kernel.org/r/20190711122138.5221-1-axel.lin@ingics.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 962f170d9344e5d9edb3903971c591f42d55e226\nAuthor: Axel Lin \u003caxel.lin@ingics.com\u003e\nDate: Wed Jun 26 21:26:32 2019 +0800\n\n regulator: lm363x: Fix n_voltages setting for lm36274\n \n According to the datasheet http://www.ti.com/lit/ds/symlink/lm36274.pdf:\n Table 23. VPOS Bias Register Field Descriptions VPOS[5:0]:\n VPOS voltage (50-mV steps): VPOS \u003d 4 V + (Code × 50 mV), 6.5 V max\n 000000 \u003d 4 V\n 000001 \u003d 4.05 V\n :\n 011110 \u003d 5.5 V (Default)\n :\n 110010 \u003d 6.5 V\n 110011 to 111111 map to 6.5 V\n \n So the LM36274_LDO_VSEL_MAX should be 0b110010 (0x32).\n The valid selectors are 0 ... LM36274_LDO_VSEL_MAX, n_voltages should be\n LM36274_LDO_VSEL_MAX + 1. Similarly, the n_voltages should be\n LM36274_BOOST_VSEL_MAX + 1 for LM36274_BOOST.\n \n Fixes: bff5e8071533 (\"regulator: lm363x: Add support for LM36274\")\n Signed-off-by: Axel Lin \u003caxel.lin@ingics.com\u003e\n Link: https://lore.kernel.org/r/20190626132632.32629-2-axel.lin@ingics.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 1e2cc8c5e0745b545d4974788dc606d678b6e564\nAuthor: Axel Lin \u003caxel.lin@ingics.com\u003e\nDate: Wed Jun 26 21:26:31 2019 +0800\n\n regulator: lm363x: Fix off-by-one n_voltages for lm3632 ldo_vpos/ldo_vneg\n \n According to the datasheet https://www.ti.com/lit/ds/symlink/lm3632a.pdf\n Table 20. VPOS Bias Register Field Descriptions VPOS[5:0]\n Sets the Positive Display Bias (LDO) Voltage (50 mV per step)\n 000000: 4 V\n 000001: 4.05 V\n 000010: 4.1 V\n ....................\n 011101: 5.45 V\n 011110: 5.5 V (Default)\n 011111: 5.55 V\n ....................\n 100111: 5.95 V\n 101000: 6 V\n Note: Codes 101001 to 111111 map to 6 V\n \n The LM3632_LDO_VSEL_MAX should be 0b101000 (0x28), so the maximum voltage\n can match the datasheet.\n \n Fixes: 3a8d1a73a037 (\"regulator: add LM363X driver\")\n Signed-off-by: Axel Lin \u003caxel.lin@ingics.com\u003e\n Link: https://lore.kernel.org/r/20190626132632.32629-1-axel.lin@ingics.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 1d3dd532883be6167da5df6117efd6d4e8790456\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Tue Jul 9 23:55:43 2019 +0530\n\n sound: soc: codecs: mt6358: change return type of mt6358_codec_init_reg\n \n As mt6358_codec_init_reg function always returns 0 , change return type\n from int to void.\n \n fixes below issue reported by coccicheck\n sound/soc/codecs/mt6358.c:2260:5-8: Unneeded variable: \"ret\". Return \"0\"\n on line 2289\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Acked-by: Tzung-Bi Shih \u003ctzungbi@google.com\u003e\n Link: https://lore.kernel.org/r/20190709182543.GA6611@hari-Inspiron-1545\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 5be6155b50bbf7083b4bfa219e4ce6d1491f42f0\nAuthor: Shengjiu Wang \u003cshengjiu.wang@nxp.com\u003e\nDate: Thu Jul 11 18:49:45 2019 +0800\n\n ASoC: fsl_esai: Wrap some operations to be functions\n \n Extract the operation to be functions, to improve the\n readability.\n \n In this patch, fsl_esai_hw_init, fsl_esai_register_restore,\n fsl_esai_trigger_start and fsl_esai_trigger_stop are\n extracted.\n \n Signed-off-by: Shengjiu Wang \u003cshengjiu.wang@nxp.com\u003e\n Acked-by: Nicolin Chen \u003cnicoleotsuka@gmail.com\u003e\n Link: https://lore.kernel.org/r/804d7e75ae7e06a913479912b578b3538ca7cd3f.1562842206.git.shengjiu.wang@nxp.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 4dc057a786dc23fd1ed70e6bcab691c0f26daa1e\nAuthor: Curtis Malainey \u003ccujomalainey@chromium.org\u003e\nDate: Wed Jul 10 17:28:38 2019 -0700\n\n ASoC: rt5677: remove superfluous set\n \n We define this variable with the same value, no need to set it twice\n \n Signed-off-by: Curtis Malainey \u003ccujomalainey@chromium.org\u003e\n Link: https://lore.kernel.org/r/20190711002838.35363-1-cujomalainey@chromium.org\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit eaba5585944e6d692828bca701c9362f60172b8a\nAuthor: Codrin Ciubotariu \u003ccodrin.ciubotariu@microchip.com\u003e\nDate: Wed Jul 10 13:51:19 2019 +0300\n\n ASoC: codecs: ad193x: Use regmap_multi_reg_write() when initializing\n \n Using regmap_multi_reg_write() when we set the default values for our\n registers makes the code smaller and easier to read.\n \n Suggested-by: Tzung-Bi Shih \u003ctzungbi@google.com\u003e\n Signed-off-by: Codrin Ciubotariu \u003ccodrin.ciubotariu@microchip.com\u003e\n Reviewed-by: Tzung-Bi Shih \u003ctzungbi@google.com\u003e\n Link: https://lore.kernel.org/r/20190710105119.22987-1-codrin.ciubotariu@microchip.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit f7408a3d5b5fd10571a653d1a81ce9167c62727f\nAuthor: Joe Perches \u003cjoe@perches.com\u003e\nDate: Tue Jul 9 22:04:25 2019 -0700\n\n ASoC: wcd9335: Fix misuse of GENMASK macro\n \n Arguments are supposed to be ordered high then low.\n \n Signed-off-by: Joe Perches \u003cjoe@perches.com\u003e\n Link: https://lore.kernel.org/r/92e31a9f321fe731d428ec3ec9d4654ea8a16d1b.1562734889.git.joe@perches.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit ef9bec27485fefb6b93168fea73fda0dc9638046\nAuthor: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\nDate: Wed Jun 26 00:04:50 2019 -0700\n\n ASoC: SOF: Intel: hda: Make hdac_device device-managed\n \n snd_hdac_ext_bus_device_exit() has been recently modified\n to no longer free the hdac device. SOF allocates memory for\n hdac_device and hda_hda_priv with kzalloc. Make them\n device-managed instead so that they will be freed when the\n SOF driver is unloaded.\n \n Because of the above change, hda_codec is device-managed and\n it will be freed when the ASoC device is removed. Freeing\n the codec in snd_hda_codec_dev_release() leads to kernel\n panic while unloading and reloading the ASoC driver. So,\n avoid freeing the hda_codec for ASoC driver. This is done in\n the same patch to avoid bisect failure.\n \n Signed-off-by: Libin Yang \u003clibin.yang@intel.com\u003e\n Signed-off-by: Ranjani Sridharan \u003cranjani.sridharan@linux.intel.com\u003e\n Reviewed-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n Link: https://lore.kernel.org/r/20190626070450.7229-1-ranjani.sridharan@linux.intel.com\n Signed-off-by: Mark Brown \u003cbroonie@kernel.org\u003e\n\ncommit 6b0e42771795334bd24d089402d04d93bac048d3\nAuthor: Jeff Layton \u003cjlayton@kernel.org\u003e\nDate: Sun Jul 21 19:01:34 2019 -0400\n\n locks: revise generic_add_lease tracepoint\n \n Now that check_conflicting_open uses inode-\u003ei_readcount instead of\n the dentry-\u003ed_count to detect opens for read, revise the tracepoint\n to display that value instead.\n \n Also, fl is never NULL, so no need to check for that in the fast\n assign section.\n \n Cc: Amir Goldstein \u003camir73il@gmail.com\u003e\n Reported-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n Signed-off-by: Jeff Layton \u003cjlayton@kernel.org\u003e\n\ncommit be261ffce6f13229dad50f59c5e491f933d3167f\nAuthor: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\nDate: Thu Jul 4 10:46:37 2019 -0500\n\n x86: Remove X86_FEATURE_MFENCE_RDTSC\n \n AMD and Intel both have serializing lfence (X86_FEATURE_LFENCE_RDTSC).\n They\u0027ve both had it for a long time, and AMD has had it enabled in Linux\n since Spectre v1 was announced.\n \n Back then, there was a proposal to remove the serializing mfence feature\n bit (X86_FEATURE_MFENCE_RDTSC), since both AMD and Intel have\n serializing lfence. At the time, it was (ahem) speculated that some\n hypervisors might not yet support its removal, so it remained for the\n time being.\n \n Now a year-and-a-half later, it should be safe to remove.\n \n I asked Andrew Cooper about whether it\u0027s still needed:\n \n So if you\u0027re virtualised, you\u0027ve got no choice in the matter.  lfence\n is either dispatch-serialising or not on AMD, and you won\u0027t be able to\n change it.\n \n Furthermore, you can\u0027t accurately tell what state the bit is in, because\n the MSR might not be virtualised at all, or may not reflect the true\n state in hardware.  Worse still, attempting to set the bit may not be\n successful even if there isn\u0027t a fault for doing so.\n \n Xen sets the DE_CFG bit unconditionally, as does Linux by the looks of\n things (see MSR_F10H_DECFG_LFENCE_SERIALIZE_BIT).  ISTR other hypervisor\n vendors saying the same, but I don\u0027t have any information to hand.\n \n If you are running under a hypervisor which has been updated, then\n lfence will almost certainly be dispatch-serialising in practice, and\n you\u0027ll almost certainly see the bit already set in DE_CFG.  If you\u0027re\n running under a hypervisor which hasn\u0027t been patched since Spectre,\n you\u0027ve already lost in many more ways.\n \n I\u0027d argue that X86_FEATURE_MFENCE_RDTSC is not worth keeping.\n \n So remove it. This will reduce some code rot, and also make it easier\n to hook barrier_nospec() up to a cmdline disable for performance\n raisins, without having to need an alternative_3() macro.\n \n Signed-off-by: Josh Poimboeuf \u003cjpoimboe@redhat.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/d990aa51e40063acb9888e8c1b688e41355a9588.1562255067.git.jpoimboe@redhat.com\n\ncommit f240652b6032b48ad7fa35c5e701cc4c8d697c0b\nAuthor: Dave Hansen \u003cdave.hansen@linux.intel.com\u003e\nDate: Fri Jul 5 10:53:21 2019 -0700\n\n x86/mpx: Remove MPX APIs\n \n MPX is being removed from the kernel due to a lack of support in the\n toolchain going forward (gcc).\n \n The first step is to remove the userspace-visible ABIs so that applications\n will stop using it. The most visible one are the enable/disable prctl()s.\n Remove them first.\n \n This is the most minimal and least invasive change needed to ensure that\n apps stop using MPX with new kernels.\n \n Signed-off-by: Dave Hansen \u003cdave.hansen@linux.intel.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190705175321.DB42F0AD@viggo.jf.intel.com\n\ncommit e28df79ae2dfebf18e08dc66c0948b7950e4368a\nAuthor: Dave Hansen \u003cdave.hansen@linux.intel.com\u003e\nDate: Fri Jul 5 10:53:20 2019 -0700\n\n x86/mpx: Remove selftests themselves\n \n MPX is being removed from the kernel due to a lack of support in the\n toolchain going forward (gcc).\n \n Remove the x86 selftests since they have been causing some issues because\n of their propensity to do some debug-aiding tracepoint mucking.\n \n Signed-off-by: Dave Hansen \u003cdave.hansen@linux.intel.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190705175320.6542F8AE@viggo.jf.intel.com\n\ncommit 48febc03e6c239d96f46d8b38d91863769fc18c8\nAuthor: Dave Hansen \u003cdave.hansen@linux.intel.com\u003e\nDate: Fri Jul 5 10:53:18 2019 -0700\n\n x86/mpx: Remove selftests Makefile entry\n \n MPX is being removed from the kernel due to a lack of support in the\n toolchain going forward (gcc).\n \n This is the smallest possible patch to fix some issues that have been\n reported around running the MPX selftests. It it would also have been part\n of any removal series, it is offered first.\n \n Signed-off-by: Dave Hansen \u003cdave.hansen@linux.intel.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190705175318.784C233E@viggo.jf.intel.com\n\ncommit 69732102426b1c55a257386841fb80ec1f425d32\nAuthor: Pingfan Liu \u003ckernelfans@gmail.com\u003e\nDate: Tue Jul 16 16:40:24 2019 +0800\n\n x86/realmode: Remove trampoline_status\n \n There is no reader of trampoline_status, it\u0027s only written.\n \n It turns out that after commit ce4b1b16502b (\"x86/smpboot: Initialize\n secondary CPU only if master CPU will wait for it\"), trampoline_status is\n not needed any more.\n \n Signed-off-by: Pingfan Liu \u003ckernelfans@gmail.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/1563266424-3472-1-git-send-email-kernelfans@gmail.com\n\ncommit 83527ef7abf7c02c33a90b00f0954db35415adbd\nAuthor: Maya Nakamura \u003cm.maya.nakamura@gmail.com\u003e\nDate: Fri Jul 12 08:25:18 2019 +0000\n\n drivers: hv: vmbus: Replace page definition with Hyper-V specific one\n \n Replace PAGE_SIZE with HV_HYP_PAGE_SIZE because the guest page size may not\n be 4096 on all architectures and Hyper-V always runs with a page size of\n 4096.\n \n Signed-off-by: Maya Nakamura \u003cm.maya.nakamura@gmail.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Michael Kelley \u003cmikelley@microsoft.com\u003e\n Reviewed-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\n Acked-by: Sasha Levin \u003csashal@kernel.org\u003e\n Link: https://lkml.kernel.org/r/0d9e80ecabcc950dc279fdd2e39bea4060123ba4.1562916939.git.m.maya.nakamura@gmail.com\n\ncommit 8c3e44bde7fd1b8291515f046008225711ac7beb\nAuthor: Maya Nakamura \u003cm.maya.nakamura@gmail.com\u003e\nDate: Fri Jul 12 08:21:25 2019 +0000\n\n x86/hyperv: Add functions to allocate/deallocate page for Hyper-V\n \n Introduce two new functions, hv_alloc_hyperv_page() and\n hv_free_hyperv_page(), to allocate/deallocate memory with the size and\n alignment that Hyper-V expects as a page. Although currently they are not\n used, they are ready to be used to allocate/deallocate memory on x86 when\n their ARM64 counterparts are implemented, keeping symmetry between\n architectures with potentially different guest page sizes.\n \n Signed-off-by: Maya Nakamura \u003cm.maya.nakamura@gmail.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Michael Kelley \u003cmikelley@microsoft.com\u003e\n Reviewed-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\n Link: https://lore.kernel.org/lkml/alpine.DEB.2.21.1906272334560.32342@nanos.tec.linutronix.de/\n Link: https://lore.kernel.org/lkml/87muindr9c.fsf@vitty.brq.redhat.com/\n Link: https://lkml.kernel.org/r/706b2e71eb3e587b5f8801e50f090fae2a00e35d.1562916939.git.m.maya.nakamura@gmail.com\n\ncommit fcd3f6222a4ece735d0b3ffb93f646eff693aa69\nAuthor: Maya Nakamura \u003cm.maya.nakamura@gmail.com\u003e\nDate: Fri Jul 12 08:14:47 2019 +0000\n\n x86/hyperv: Create and use Hyper-V page definitions\n \n Define HV_HYP_PAGE_SHIFT, HV_HYP_PAGE_SIZE, and HV_HYP_PAGE_MASK because\n the Linux guest page size and hypervisor page size concepts are different,\n even though they happen to be the same value on x86.\n \n Also, replace PAGE_SIZE with HV_HYP_PAGE_SIZE.\n \n Signed-off-by: Maya Nakamura \u003cm.maya.nakamura@gmail.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Michael Kelley \u003cmikelley@microsoft.com\u003e\n Reviewed-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\n Link: https://lkml.kernel.org/r/e95111629abf65d016e983f72494cbf110ce605f.1562916939.git.m.maya.nakamura@gmail.com\n\ncommit 018ebca8bd704f18d56f8fff38e2c3d76d7d39fb\nAuthor: Gayatri Kammela \u003cgayatri.kammela@intel.com\u003e\nDate: Wed Jul 17 16:46:32 2019 -0700\n\n x86/cpufeatures: Enable a new AVX512 CPU feature\n \n Add a new AVX512 instruction group/feature for enumeration in\n /proc/cpuinfo: AVX512_VP2INTERSECT.\n \n CPUID.(EAX\u003d7,ECX\u003d0):EDX[bit 8] AVX512_VP2INTERSECT\n \n Detailed information of CPUID bits for this feature can be found in\n the Intel Architecture Intsruction Set Extensions Programming Reference\n document (refer to Table 1-2). A copy of this document is available at\n https://bugzilla.kernel.org/show_bug.cgi?id\u003d204215.\n \n Signed-off-by: Gayatri Kammela \u003cgayatri.kammela@intel.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190717234632.32673-3-gayatri.kammela@intel.com\n\ncommit 1e0c08e3034de0659367393bfa825188462f22e6\nAuthor: Gayatri Kammela \u003cgayatri.kammela@intel.com\u003e\nDate: Wed Jul 17 16:46:31 2019 -0700\n\n cpu/cpuid-deps: Add a tab to cpuid dependent features\n \n Improve code readability by adding a tab between the elements of each\n structure in an array of cpuid-dep struct so longer feature names will fit.\n \n Signed-off-by: Gayatri Kammela \u003cgayatri.kammela@intel.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/20190717234632.32673-2-gayatri.kammela@intel.com\n\ncommit 6365b842aae4490ebfafadfc6bb27a6d3cc54757\nAuthor: Andy Lutomirski \u003cluto@kernel.org\u003e\nDate: Wed Jul 3 13:34:04 2019 -0700\n\n x86/syscalls: Split the x32 syscalls into their own table\n \n For unfortunate historical reasons, the x32 syscalls and the x86_64\n syscalls are not all numbered the same. As an example, ioctl() is nr 16 on\n x86_64 but 514 on x32.\n \n This has potentially nasty consequences, since it means that there are two\n valid RAX values to do ioctl(2) and two invalid RAX values. The valid\n values are 16 (i.e. ioctl(2) using the x86_64 ABI) and (514 | 0x40000000)\n (i.e. ioctl(2) using the x32 ABI).\n \n The invalid values are 514 and (16 | 0x40000000). 514 will enter the\n \"COMPAT_SYSCALL_DEFINE3(ioctl, ...)\" entry point with in_compat_syscall()\n and in_x32_syscall() returning false, whereas (16 | 0x40000000) will enter\n the native entry point with in_compat_syscall() and in_x32_syscall()\n returning true. Both are bogus, and both will exercise code paths in the\n kernel and in any running seccomp filters that really ought to be\n unreachable.\n \n Splitting out the x32 syscalls into their own tables, allows both bogus\n invocations to return -ENOSYS. I\u0027ve checked glibc, musl, and Bionic, and\n all of them appear to call syscalls with their correct numbers, so this\n change should have no effect on them.\n \n There is an added benefit going forward: new syscalls that need special\n handling on x32 can share the same number on x32 and x86_64. This means\n that the special syscall range 512-547 can be treated as a legacy wart\n instead of something that may need to be extended in the future.\n \n Also add a selftest to verify the new behavior.\n \n Signed-off-by: Andy Lutomirski \u003cluto@kernel.org\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/208024256b764312598f014ebfb0a42472c19354.1562185330.git.luto@kernel.org\n\ncommit f85a8573ceb225e606fcf38a9320782316f47c71\nAuthor: Andy Lutomirski \u003cluto@kernel.org\u003e\nDate: Wed Jul 3 13:34:03 2019 -0700\n\n x86/syscalls: Disallow compat entries for all types of 64-bit syscalls\n \n A \"compat\" entry in the syscall tables means to use a different entry on\n 32-bit and 64-bit builds.\n \n This only makes sense for syscalls that exist in the first place in 32-bit\n builds, so disallow it for anything other than i386.\n \n Signed-off-by: Andy Lutomirski \u003cluto@kernel.org\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/4b7565954c5a06530ac01d98cb1592538fd8ae51.1562185330.git.luto@kernel.org\n\ncommit a8d03c3f300eefff3b5c14798409e4b43e37dd9b\nAuthor: Andy Lutomirski \u003cluto@kernel.org\u003e\nDate: Wed Jul 3 13:34:02 2019 -0700\n\n x86/syscalls: Use the compat versions of rt_sigsuspend() and rt_sigprocmask()\n \n I\u0027m working on some code that detects at build time if there\u0027s a\n COMPAT_SYSCALL_DEFINE() that is not referenced in the x86 syscall tables.\n It catches three offenders: rt_sigsuspend(), rt_sigprocmask(), and\n sendfile64().\n \n For rt_sigsuspend() and rt_sigprocmask(), the only potential difference\n between the native and compat versions is that the compat version converts\n the sigset_t, but, on little endian architectures, the conversion is a\n no-op. This is why they both currently work on x86.\n \n To make the code more consistent, and to make the upcoming patches work,\n rewire x86 to use the compat vesions.\n \n sendfile64() is more complicated, and will be addressed separately.\n \n Signed-off-by: Andy Lutomirski \u003cluto@kernel.org\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/51643ac3157b5921eae0e172a8a0b1d953e68ebb.1562185330.git.luto@kernel.org\n\ncommit 45e29d119e9923ff14dfb840e3482bef1667bbfb\nAuthor: Andy Lutomirski \u003cluto@kernel.org\u003e\nDate: Wed Jul 3 13:34:05 2019 -0700\n\n x86/syscalls: Make __X32_SYSCALL_BIT be unsigned long\n \n Currently, it\u0027s an int. This is bizarre. Fortunately, the code using it\n still works: ~__X32_SYSCALL_BIT is also int, so, if nr is unsigned long,\n then C kindly sign-extends the ~__X32_SYSCALL_BIT part, and it actually\n results in the desired value.\n \n This is far more subtle than it deserves to be. Syscall numbers are, for\n all practical purposes, unsigned long, so make __X32_SYSCALL_BIT be\n unsigned long.\n \n Signed-off-by: Andy Lutomirski \u003cluto@kernel.org\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/99b0d83ad891c67105470a1a6b63243fd63a5061.1562185330.git.luto@kernel.org\n\ncommit 83b584d9c6a1494170abd3a8b24f41939b23d625\nAuthor: Andrew Cooper \u003candrew.cooper3@citrix.com\u003e\nDate: Mon Jul 15 16:16:41 2019 +0100\n\n x86/paravirt: Drop {read,write}_cr8() hooks\n \n There is a lot of infrastructure for functionality which is used\n exclusively in __{save,restore}_processor_state() on the suspend/resume\n path.\n \n cr8 is an alias of APIC_TASKPRI, and APIC_TASKPRI is saved/restored by\n lapic_{suspend,resume}(). Saving and restoring cr8 independently of the\n rest of the Local APIC state isn\u0027t a clever thing to be doing.\n \n Delete the suspend/resume cr8 handling, which shrinks the size of struct\n saved_context, and allows for the removal of both PVOPS.\n \n Signed-off-by: Andrew Cooper \u003candrew.cooper3@citrix.com\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Reviewed-by: Juergen Gross \u003cjgross@suse.com\u003e\n Link: https://lkml.kernel.org/r/20190715151641.29210-1-andrew.cooper3@citrix.com\n\ncommit 229b969b3d38bc28bcd55841ee7ca9a9afb922f3\nAuthor: Andy Lutomirski \u003cluto@kernel.org\u003e\nDate: Sun Jul 14 08:23:14 2019 -0700\n\n x86/apic: Initialize TPR to block interrupts 16-31\n \n The APIC, per spec, is fundamentally confused and thinks that interrupt\n vectors 16-31 are valid. This makes no sense -- the CPU reserves vectors\n 0-31 for exceptions (faults, traps, etc). Obviously, no device should\n actually produce an interrupt with vector 16-31, but robustness can be\n improved by setting the APIC TPR class to 1, which will prevent delivery of\n an interrupt with a vector below 32.\n \n Note: This is *not* intended as a security measure against attackers who\n control malicious hardware. Any PCI or similar hardware that can be\n controlled by an attacker MUST be behind a functional IOMMU that remaps\n interrupts. The purpose of this change is to reduce the chance that a\n certain class of device malfunctions crashes the kernel in hard-to-debug\n ways.\n \n Suggested-by: Andrew Cooper \u003candrew.cooper3@citrix.com\u003e\n Signed-off-by: Andy Lutomirski \u003cluto@kernel.org\u003e\n Signed-off-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://lkml.kernel.org/r/dc04a9f8b234d7b0956a8d2560b8945bcd9c4bf7.1563117760.git.luto@kernel.org\n\ncommit de332de26d19a6102c6e82c8774f983b32a74488\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Wed Jun 19 11:03:17 2019 +0200\n\n ARM: dts: sunxi: Switch from phy to phy-handle\n \n The phy device tree property has been deprecated in favor of phy-handle,\n let\u0027s replace it.\n \n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 2a40d1b291138a7dce787b60f017d888b7d34349\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Thu Jun 6 21:21:09 2019 +0200\n\n ARM: dts: sunxi: Switch to the generic PHY properties\n \n The DWMAC specific properties to manage the PHY have been superseeded by\n the generic PHY properties. Let\u0027s move to it.\n \n Reviewed-by: Andrew Lunn \u003candrew@lunn.ch\u003e\n Tested-by: Chen-Yu Tsai \u003cwens@csie.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit caa8422d01e983782548648e125fd617cadcec3f\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Sat Jul 20 12:33:37 2019 +0100\n\n ALSA: hda: Flush interrupts on disabling\n \n I was looking at\n \n \u003c4\u003e [241.835158] general protection fault: 0000 [#1] PREEMPT SMP PTI\n \u003c4\u003e [241.835181] CPU: 1 PID: 214 Comm: kworker/1:3 Tainted: G U 5.2.0-CI-CI_DRM_6509+ #1\n \u003c4\u003e [241.835199] Hardware name: Dell Inc. OptiPlex 745 /0GW726, BIOS 2.3.1 05/21/2007\n \u003c4\u003e [241.835234] Workqueue: events snd_hdac_bus_process_unsol_events [snd_hda_core]\n \u003c4\u003e [241.835256] RIP: 0010:input_handle_event+0x16d/0x5e0\n \u003c4\u003e [241.835270] Code: 48 8b 93 58 01 00 00 8b 52 08 89 50 04 8b 83 f8 06 00 00 48 8b 93 00 07 00 00 8d 70 01 48 8d 04 c2 83 e1 08 89 b3 f8 06 00 00 \u003c66\u003e 89 28 66 44 89 60 02 44 89 68 04 8b 93 f8 06 00 00 0f 84 fd fe\n \u003c4\u003e [241.835304] RSP: 0018:ffffc9000019fda0 EFLAGS: 00010046\n \u003c4\u003e [241.835317] RAX: 6b6b6b6ec6c6c6c3 RBX: ffff8880290fefc8 RCX: 0000000000000000\n \u003c4\u003e [241.835332] RDX: 000000006b6b6b6b RSI: 000000006b6b6b6c RDI: 0000000000000046\n \u003c4\u003e [241.835347] RBP: 0000000000000005 R08: 0000000000000000 R09: 0000000000000001\n \u003c4\u003e [241.835362] R10: ffffc9000019faa0 R11: 0000000000000000 R12: 0000000000000004\n \u003c4\u003e [241.835377] R13: 0000000000000000 R14: ffff8880290ff1d0 R15: 0000000000000293\n \u003c4\u003e [241.835392] FS: 0000000000000000(0000) GS:ffff88803de80000(0000) knlGS:0000000000000000\n \u003c4\u003e [241.835409] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n \u003c4\u003e [241.835422] CR2: 00007ffe9a99e9b7 CR3: 000000002f588000 CR4: 00000000000006e0\n \u003c4\u003e [241.835436] Call Trace:\n \u003c4\u003e [241.835449] input_event+0x45/0x70\n \u003c4\u003e [241.835464] snd_jack_report+0xdc/0x100\n \u003c4\u003e [241.835490] snd_hda_jack_report_sync+0x83/0xc0 [snd_hda_codec]\n \u003c4\u003e [241.835512] snd_hdac_bus_process_unsol_events+0x5a/0x70 [snd_hda_core]\n \u003c4\u003e [241.835530] process_one_work+0x245/0x610\n \n which has the hallmarks of a worker queued from interrupt after it was\n supposedly cancelled (note the POISON_FREE), and I could not see where\n the interrupt would be flushed on shutdown so added the likely suspects.\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111174\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit ade49db337a9d44ac5835cfce1ee873549011b27\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Wed Jul 11 18:05:52 2018 +0200\n\n ALSA: hda/hdmi - Allow audio component for AMD/ATI and Nvidia HDMI\n \n AMD/ATI and Nvidia HDMI codec drivers didn\u0027t have the audio component\n binding like i915, but it worked only with the traditional HD-audio\n unsolicited event for the HDMI hotplug detection and the ELD read-up\n thereafter. This has been a problem in many ways: first of all, it\n goes through the hardware event transition (from GPU register write,\n HD-audio controller trigger, and finally to HD-audio unsolicited event\n handling), which is often unreliable and may miss some opportunities.\n Second, each unsol event handling and ELD read-up need the explicit\n power up / down when the codec is in the runtime suspend. Last but\n not least, which is the most important, the hotplug wakeup may be\n missed when the HD-audio controller is in runtime suspend. Especially\n the last point is a big problem due to the recent change relevant with\n vga_switcheroo that forcibly enables the runtime PM for AMD HDMI\n controllers.\n \n These issues are solved by introducing the audio component; the\n hotplug notification is done by a direct function callback, which is\n more accurate and reliable, and it can be processed without the actual\n hardware access, i.e. no runtime PM trigger is needed, and the\n HD-audio gets the event even if it\u0027s in runtime suspend. The same for\n ELD query, as it\u0027s read directly from the cached ELD bytes stored in\n the DRM driver, hence the whole hardware access can be skipped.\n \n So here it is: this patch implements the audio component binding with\n AMD/ATI and Nouveau DRM drivers. The biggest difference from i915\n implementation is that this binding is fully optional and it can be\n enabled asynchronously on the fly. That is, the driver will switch\n from the HD-audio unsolicited event to the notify callback once when\n the DRM component gets bound. Similarly, when DRM driver gets\n unloaded, the HDMI event handling returns to the legacy mode, too.\n \n Also, another difference from i915 is that the new code registers the\n component in the codec driver, while i915 HDMI codec assumes the\n component binding was already done in the HD-audio controller driver.\n Hence the new code does need to de-register the component binding at\n the codec exit, too.\n \n Some other details:\n - The match component ops assumes that both VGA and HD-audio\n controller PCI entries belong to the same PCI bus, and only accepts\n such an entry.\n \n - The pin2port audio_ops is implemented with assumption of the fixed\n widget layout. For AMD, it\u0027s starting from 3, with step 2 (3, 5, 7,\n ...), while for Nvidia, it\u0027s starting from 4, with step 1 (4, 5, 6,\n ...)\n \n As of this patch, the corresponding component isn\u0027t implemented in DRM\n side, so this change alone won\u0027t give any benefit. By the following\n changes in DRM sides, the mission will be completed.\n \n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit 551626ec0ad28dc43cae3094c35be7088cc625ab\nAuthor: Takashi Iwai \u003ctiwai@suse.de\u003e\nDate: Wed Jul 17 14:30:23 2019 +0200\n\n ALSA: hda/hdmi - Don\u0027t report spurious jack state changes\n \n The HDMI jack handling reports the state change always via\n snd_jack_report() whenever hdmi_present_sense() is called, even if the\n state itself doesn\u0027t change from the previous time. This is mostly\n harmless but still a bit confusing to user-space.\n \n This patch reduces such spurious jack state changes and reports only\n when the state really changed. Also, as a minor optimization, avoid\n overwriting the pin ELD data when the state is identical.\n \n Reviewed-by: Kai Vehmanen \u003ckai.vehmanen@linux.intel.com\u003e\n Signed-off-by: Takashi Iwai \u003ctiwai@suse.de\u003e\n\ncommit d9aa4d4fca67823838fe9861456201430c545e69\nAuthor: Sébastien Szymanski \u003csebastien.szymanski@armadeus.com\u003e\nDate: Thu Jul 4 16:23:23 2019 +0200\n\n ARM: dts: opos6uldev: use OF graph to describe the display\n \n To make use of the new eLCDIF DRM driver OF graph description is\n required. Describe the display using OF graph nodes.\n \n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Sébastien Szymanski \u003csebastien.szymanski@armadeus.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit bf58de429387b928268a8567e34cb4d3c109b727\nAuthor: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\nDate: Thu Jul 4 11:53:22 2019 +0300\n\n arm64: defconfig: Build imx8 ddr pmu as module\n \n This is available on all imx8 but is not \"boot critical\" in any way so\n build as a module.\n \n Signed-off-by: Leonard Crestez \u003cleonard.crestez@nxp.com\u003e\n Acked-by: Frank Li \u003cfrank.li@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit f9429c1fba87ee820cd31e8eaa2e4c64e27d6439\nAuthor: Icenowy Zheng \u003cicenowy@aosc.io\u003e\nDate: Sat Jul 13 11:46:29 2019 +0800\n\n dt-bindings: clk: sunxi-ccu: add compatible string for V3 CCU\n \n Despite Allwinner V3 and V3s shares the same die, one peripheral (I2S)\n is only available on V3, and thus the clocks is not declared for V3s\n CCU.\n \n Add a V3 CCU compatible string to the binding to prepare for a CCU\n driver that provide I2S clock on V3, but not on V3s.\n \n Signed-off-by: Icenowy Zheng \u003cicenowy@aosc.io\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit c5ed9475c22c89d5409402055142372e35d26a3f\nAuthor: Icenowy Zheng \u003cicenowy@aosc.io\u003e\nDate: Sat Jul 13 11:46:28 2019 +0800\n\n clk: sunxi-ng: v3s: add the missing PLL_DDR1\n \n The user manual of V3/V3s/S3 declares a PLL_DDR1, however it\u0027s forgot\n when developing the V3s CCU driver.\n \n Add back the missing PLL_DDR1.\n \n Fixes: d0f11d14b0bc (\"clk: sunxi-ng: add support for V3s CCU\")\n Signed-off-by: Icenowy Zheng \u003cicenowy@aosc.io\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n\ncommit 3355c91b79394593ebbb197c8e930a91826f4ff3\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Sun Jul 21 23:38:15 2019 +0530\n\n cpufreq: ap806: Add NULL check after kcalloc\n \n Add NULL check after kcalloc.\n \n Fix below issue reported by coccicheck\n ./drivers/cpufreq/armada-8k-cpufreq.c:138:1-12: alloc with no test,\n possible model on line 151\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit b880c18336252df4866d07f850ea7872b6e83f0a\nAuthor: Yangtao Li \u003ctiny.windzz@gmail.com\u003e\nDate: Wed Jun 12 12:28:16 2019 -0400\n\n dt-bindings: cpufreq: Document allwinner,sun50i-h6-operating-points\n \n Allwinner Process Voltage Scaling Tables defines the voltage and\n frequency value based on the speedbin blown in the efuse combination.\n The sunxi-cpufreq-nvmem driver reads the efuse value from the SoC to\n provide the OPP framework with required information.\n This is used to determine the voltage and frequency value for each\n OPP of operating-points-v2 table when it is parsed by the OPP framework.\n \n The \"allwinner,sun50i-h6-operating-points\" DT extends the\n \"operating-points-v2\"\n with following parameters:\n - nvmem-cells (NVMEM area containig the speedbin information)\n - opp-microvolt-\u003cname\u003e: voltage in micro Volts.\n At runtime, the platform can pick a \u003cname\u003e and matching\n opp-microvolt-\u003cname\u003e property.\n HW: \u003cname\u003e:\n sun50i-h6 speed0 speed1 speed2\n \n Signed-off-by: Yangtao Li \u003ctiny.windzz@gmail.com\u003e\n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit f328584f7bff86858249bb358f1adcecde48388a\nAuthor: Yangtao Li \u003ctiny.windzz@gmail.com\u003e\nDate: Wed Jun 12 12:28:15 2019 -0400\n\n cpufreq: Add sun50i nvmem based CPU scaling driver\n \n For some SoCs, the CPU frequency subset and voltage value of each OPP\n varies based on the silicon variant in use. The sun50i-cpufreq-nvmem\n driver reads the efuse value from the SoC to provide the OPP framework\n with required information.\n \n Signed-off-by: Yangtao Li \u003ctiny.windzz@gmail.com\u003e\n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\n\ncommit c4011dde241d371972788a735cd1b5d547991d46\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Jul 16 11:29:08 2019 +0530\n\n staging: mt7621-pinctrl: Add of_node_put() before return\n \n Each iteration of for_each_compatible_node puts the previous node, but\n in the case of a return from the middle of the loop, there is no put,\n thus causing a memory leak. Add an of_node_put before the return.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190716055908.3183-1-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 2125a44149cdf6be61fdabce898d26e9ba37dcd0\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Tue Jul 16 11:29:44 2019 +0530\n\n staging: mt7621-pci: Add of_node_put() before return\n \n Each iteration of for_each_available_child_of_node puts the previous\n node, but in the case of a return from the middle of the loop, there is\n no put, thus causing a memory leak. Hence add an of_node_put before the\n return in two places.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190716055944.3237-1-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 9da681e017a36a411f70ca1ff68c499a30d2039d\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Tue Jul 16 17:32:56 2019 +0800\n\n staging: erofs: support bmap\n \n Add erofs_bmap() to support FIBMAP ioctl on flatmode inode.\n \n Reviewed-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190716093256.108791-1-yuchao0@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 441dfcc8861ed9cbc778c655f85e5ce5db08035e\nAuthor: Chao Yu \u003cyuchao0@huawei.com\u003e\nDate: Tue Jul 16 17:44:22 2019 +0800\n\n staging: erofs: avoid opened loop codes\n \n Use __GFP_NOFAIL to avoid opened loop codes in z_erofs_vle_unzip().\n \n Signed-off-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Reviewed-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190716094422.110805-1-yuchao0@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 44a22dc3dd890ff476e708e39895c2536ad99cf2\nAuthor: Benjamin Sherman \u003cbenjamin@bensherman.io\u003e\nDate: Tue Jul 16 00:57:55 2019 -0400\n\n staging: vt6656: change alignment to match parenthesis\n \n Change indentation to match parentheses. This complies with the Linux\n kernel coding style and improves readability.\n \n Signed-off-by: Benjamin Sherman \u003cbenjamin@bensherman.io\u003e\n Link: https://lore.kernel.org/r/20190716045754.fivh5n44bybe2uce@valkyrie-mobile.localdomain\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 384329ddc42ad91a9f1f0eab3c53ce0b0402bc60\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 12 12:47:45 2019 +0530\n\n staging: rtl8188eu: Replace function rtl88eu_phy_rf_config()\n \n Remove function rtl88eu_phy_rf_config as all it does is call\n rtl88e_phy_rf6052_config.\n Rename rtl88e_phy_rf6052_config to rtl88eu_phy_rf_config and change its\n type from static to non-static to maintain compatibility with call\n sites.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190712071746.2474-8-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 792d89ee4bd6ca4ba803a9f522cf9d0a35dd763f\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 12 12:47:44 2019 +0530\n\n staging: rtl8188eu: Replace function rtl88e_phy_rf6052_config()\n \n Remove function rtl88e_phy_rf6052_config as all it does is call\n rf6052_conf_para.\n Rename rf6052_conf_para to rtl88e_phy_rf6052_config to maintain\n compatibility with call site.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190712071746.2474-7-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 5cace672b9a4da8586ade62d3a7f204a4adced2e\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 12 12:47:43 2019 +0530\n\n staging: rtl8188eu: Replace function rtw_alloc_network()\n \n Change external call site of _rtw_alloc_network to rtw_alloc_network.\n Remove function rtw_alloc_network as all it does is call\n _rtw_alloc_network.\n Rename _rtw_alloc_network to rtw_alloc_network to maintain compatibility\n with call sites. Keep its type as non-static (even though the old\n rtw_alloc_network was static) as this functionality is used in other\n files as well.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190712071746.2474-6-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 04c8c1985618b45e59433d6c46e3f787e146017e\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 12 12:47:42 2019 +0530\n\n staging: rtl8188eu: Remove function rtw_modular64()\n \n Remove function rtw_modular64 as all it does is call do_div.\n Replace call to rtw_modular64 with call to do_div.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190712071746.2474-5-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit c6255d2d2432623ee50f48970c51050e979befc3\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 12 12:47:41 2019 +0530\n\n staging: rtl8188eu: Remove function rtw_os_indicate_scan_done()\n \n In function rtw_indicate_scan_done, replace call to\n rtw_os_indicate_scan_done with call to indicate_wx_scan_complete_event\n as all that rtw_os_indicate_scan_done does is call\n indicate_wx_scan_complete_event.\n Remove now-unused function rtw_os_indicate_scan_done.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190712071746.2474-4-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 01625b0bbb523e6ce2818d4691812f66fe9c42b4\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 12 12:47:40 2019 +0530\n\n staging: rtl8188eu: Replace function beacon_timing_control()\n \n Remove function beacon_timing_control as all it does is call\n rtw_hal_bcn_related_reg_setting.\n Rename rtw_hal_bcn_related_reg_setting to beacon_timing_control for\n compatibility with call sites.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190712071746.2474-3-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d6a496f937523048aab1fef0df74786b0ae6fe17\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 12 12:47:39 2019 +0530\n\n staging: rtl8188eu: core: Remove function enable_rate_adaptive()\n \n Remove function enable_rate_adaptive as it does nothing except call\n Update_RA_Entry.\n Modify call site of enable_rate_adaptive to call Update_RA_Entry\n instead.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190712071746.2474-2-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 6290da3e439d5aa4a692196b61f34a3b6e8e9b1c\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 12 12:47:38 2019 +0530\n\n staging: rtl8188eu: core: Replace function Set_MSR()\n \n Change reference to Set_NETYPE0_MSR to Set_MSR.\n Replace the contents of Set_MSR with the contents of Set_NETYPE0_MSR as\n Set_MSR does nothing but call Set_NETYPE0_MSR.\n Delete Set_NETYPE0_MSR.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190712071746.2474-1-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 067756acdac8e0dd91edf4800e844fcb6e0cc72a\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Mon Jul 15 23:16:18 2019 +0530\n\n staging: rtl8723bs: core: Change return type of init_mlme_ext_priv\n \n As init_mlme_ext_priv function always returns SUCCESS , We can change\n return type from int to void.\n \n Fixes below issue identified by coccicheck\n drivers/staging/rtl8723bs/core/rtw_mlme_ext.c:464:5-8: Unneeded\n variable: \"res\". Return \"_SUCCESS\" on line 492\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190715174618.GA8947@hari-Inspiron-1545\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b21f9b8352aa189e708c4ccfb5216b56880a1851\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Sun Jul 14 23:01:34 2019 +0530\n\n staging: rtl8723bs: os_dep: Remove code valid only for 5GHz\n \n As per TODO ,remove code valid only for 5 GHz(channel \u003e 14).\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190714173134.GA7111@hari-Inspiron-1545\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 2615f966f9f27b30de7d9c10316bf482d1d2964e\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Sun Jul 14 22:58:26 2019 +0530\n\n staging: rtl8723bs: core: Remove code valid only for 5GHz\n \n As per TODO ,remove code valid only for 5 GHz(channel \u003e 14).\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190714172826.GA6950@hari-Inspiron-1545\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 2e7a3f1158eecb0675199d7b030f3ef9db82eb25\nAuthor: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nDate: Sun Jul 14 22:54:51 2019 +0530\n\n staging: rtl8723bs: hal: Remove code valid only for 5GHz\n \n As per TODO ,remove code valid only for 5 GHz(channel \u003e 14).\n \n Signed-off-by: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190714172451.GA6779@hari-Inspiron-1545\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d59381d639af22da6eb49c5d851faf15feb6b462\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Jul 11 22:07:26 2019 +0800\n\n staging: kpc2000: kpc_spi: Remove unnecessary null check before kfree\n \n A null check before a kfree is redundant, so remove it.\n This is detected by coccinelle.\n \n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190711140726.46732-1-yuehaibing@huawei.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 6c90bade52bbabdbdf3604d7f9279c878ebafed2\nAuthor: Michael Straube \u003cstraube.linux@gmail.com\u003e\nDate: Sun Jul 7 18:56:49 2019 +0200\n\n staging: rtl8188eu: remove unused definitions from hal8188e_phy_reg.h\n \n Remove unused definitions from the file hal8188e_phy_reg.h.\n \n Signed-off-by: Michael Straube \u003cstraube.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190707165649.1558-1-straube.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit c5bd72f8713e13085267a8ebccfa5b8b3f2c674d\nAuthor: Razvan Stefanescu \u003crazvan.stefanescu@nxp.com\u003e\nDate: Fri Jul 5 17:27:16 2019 +0300\n\n staging: fsl-dpaa2/ethsw: Add comments to ETHSW_VLAN flags\n \n Document each ETHSW_VLAN flag with the appropriate comment.\n \n Signed-off-by: Razvan Stefanescu \u003crazvan.stefanescu@nxp.com\u003e\n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Link: https://lore.kernel.org/r/1562336836-17119-7-git-send-email-ioana.ciornei@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit de7ee3aea2e8255b28d047a0d8588f96bf522f4b\nAuthor: Razvan Stefanescu \u003crazvan.stefanescu@nxp.com\u003e\nDate: Fri Jul 5 17:27:14 2019 +0300\n\n staging: fsl-dpaa2/ethsw: Add ndo_get_phys_port_name\n \n Add the ndo_get_phys_port_name callback to the ethsw driver.\n \n Signed-off-by: Razvan Stefanescu \u003crazvan.stefanescu@nxp.com\u003e\n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Link: https://lore.kernel.org/r/1562336836-17119-5-git-send-email-ioana.ciornei@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 7dbac0c9a1c4a86be928e2b13398346ca5b57233\nAuthor: Razvan Stefanescu \u003crazvan.stefanescu@nxp.com\u003e\nDate: Fri Jul 5 17:27:13 2019 +0300\n\n staging: fsl-dpaa2/ethsw: Remove netdevice on port probing error\n \n If the ethsw_port_init() call failed, the netdevice remains registered in\n the system.\n \n Use labels to ensure that netdevice is unregistered and freed in this case.\n \n Signed-off-by: Razvan Stefanescu \u003crazvan.stefanescu@nxp.com\u003e\n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Link: https://lore.kernel.org/r/1562336836-17119-4-git-send-email-ioana.ciornei@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit b2fdbfa91b8c54c8884323b8914b0a7b81e67fc2\nAuthor: Razvan Stefanescu \u003crazvan.stefanescu@nxp.com\u003e\nDate: Fri Jul 5 17:27:12 2019 +0300\n\n staging: fsl-dpaa2/ethsw: Add network interface statistics\n \n Allocate MC portal with atomic context for I/O and enable network interface\n statistics for hardware counters.\n \n Signed-off-by: Razvan Stefanescu \u003crazvan.stefanescu@nxp.com\u003e\n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Link: https://lore.kernel.org/r/1562336836-17119-3-git-send-email-ioana.ciornei@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 543ba01d99cc6b4007c1f9a475c1d572c46a5543\nAuthor: Razvan Stefanescu \u003crazvan.stefanescu@nxp.com\u003e\nDate: Fri Jul 5 17:27:11 2019 +0300\n\n staging: fsl-dpaa2/ethsw: Fix setting port learning/flooding flags\n \n ethsw_set_learning()/ethsw_set_flood() use flags parameter as an\n enable/disable (1/0) indicator. Previous usage sent incorrect values.\n \n Signed-off-by: Razvan Stefanescu \u003crazvan.stefanescu@nxp.com\u003e\n Signed-off-by: Ioana Ciornei \u003cioana.ciornei@nxp.com\u003e\n Link: https://lore.kernel.org/r/1562336836-17119-2-git-send-email-ioana.ciornei@nxp.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 26a7ae2949a86e1b74e2485d725efb528e2a81bb\nAuthor: J. Bruce Fields \u003cbfields@redhat.com\u003e\nDate: Wed Jul 10 19:00:56 2019 -0400\n\n staging: wlan-ng: use \"%*pE\" for serial number\n \n Almost every user of \"%*pE\" in the kernel uses just bare \"%*pE\". This\n is the only user of \"%pEhp\". I can\u0027t see why it\u0027s needed.\n \n Signed-off-by: J. Bruce Fields \u003cbfields@redhat.com\u003e\n Link: https://lore.kernel.org/r/1562799656-13401-1-git-send-email-bfields@redhat.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 4ad366015d0560abf749d62b525177d38b02a1a9\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 12 12:50:09 2019 +0530\n\n staging: wilc1000: Replace function wilc_wlan_cfg_get_val()\n \n Include wilc_wlan_cfg.h in wilc_netdev.c to enable it to call functions\n in wilc_wlan_cfg.c.\n Remove function wilc_wlan_cfg_get_val as all it does is call\n wilc_wlan_cfg_get_wid_value in file wilc_wlan_cfg.c.\n Rename function wilc_wlan_cfg_get_wid_value to wilc_wlan_cfg_get_val to\n maintain compatibility with call sites (including in wilc_netdev.c).\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190712072009.2550-1-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 68336404aa5123401133c9433a8c91e4b96322e6\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 12 11:58:07 2019 +0530\n\n staging: wlan-ng: Replace function hfa384x_dowmem()\n \n Remove parameters mode, cmdcb, usercb, and usercb_data from\n hfa384x_dowmem as these parameters are only assigned the same constant\n values (DOWAIT, NULL, NULL, NULL respectively).\n Modify hfa384x_dowmem to use these constants directly. Remove check for\n value of mode (as it will always be DOWAIT).\n Remove function hfa384x_dowmem_wait as it does nothing except call\n hfa384x_dowmem with these extra arguments.\n Modify call sites of hfa384x_dowmem_wait to call hfa384x_dowmem instead.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190712062807.9361-10-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 28c03f82607730901830170259a4a17b33d5152e\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 12 11:58:06 2019 +0530\n\n staging: wlan-ng: Replace function hfa384x_dormem()\n \n Remove parameters mode, cmdcb, usercb, and usercb_data from\n hfa384x_dormem as these parameters are only assigned the same constant\n values (DOWAIT, NULL, NULL, NULL respectively).\n Modify hfa384x_dormem to use these constants directly. Remove check for\n value of mode (as it will always be DOWAIT).\n Remove function hfa384x_dormem_wait as it does nothing except call\n hfa384x_dormem with these extra arguments.\n Modify call sites of hfa384x_dormem_wait to call hfa384x_dormem instead.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190712062807.9361-9-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 53055d6a915cb1753334dc18d6249f8b4ac62d88\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 12 11:58:05 2019 +0530\n\n staging: wlan-ng: Remove function hfa384x_docmd_wait()\n \n As hfa384x_docmd is only called by hfa384x_docmd_wait, and\n hfa384x_docmd_wait always passes 4 constant arguments (DOWAIT,\n NULL, NULL, NULL) to hfa384x_docmd, these constant parameters may be\n removed from hfa384x_docmd and their values used directly instead.\n Remove check for one of these constant parameters of hfa384x_docmd as\n it is no longer necessary.\n Remove hfa384x_docmd_wait as it does nothing except call hfa384x_docmd.\n Modify call sites of hfa384x_docmd_wait to call hfa384x_docmd instead.\n Change type of hfa384x_docmd to inline to match type of former\n hfa384x_docmd_wait.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190712062807.9361-8-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit aef29ca497f005fe3cd38aefc77639dc551bbd18\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 12 11:58:04 2019 +0530\n\n staging: wlan-ng: Remove unused function hfa384x_docmd_async()\n \n Remove unused function hfa384x_docmd_async.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190712062807.9361-7-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 71b289cc5aa6b3bbe7e1a87612037b2c5bae9649\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 12 11:58:03 2019 +0530\n\n staging: wlan-ng: Remove function hfa384x_dowmem_async()\n \n Remove unused function hfa384x_dowmem_async.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190712062807.9361-6-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit f2a82b0941ea0075fa402792ea8bc9c6c5db7bfe\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 12 11:58:02 2019 +0530\n\n staging: wlan-ng: Remove unused function hfa384x_dormem_async()\n \n Remove unused function hfa384x_dormem_async.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190712062807.9361-5-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit bb75c586cabd0d192b9e713143f49b44112e64cb\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 12 11:58:01 2019 +0530\n\n staging: wlan-ng: Remove unused function hfa384x_dorrid_async()\n \n Remove unused function hfa384x_dorrid_async.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190712062807.9361-4-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 57d100b8aa686ce7bccc74e702aa1f3e8051d518\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 12 11:58:00 2019 +0530\n\n staging: wlan-ng: Remove function hfa384x_dowrid_async()\n \n Remove function hfa384x_dowrid_async as it does nothing except call\n hfa384x_dowrid, and itself is called only once.\n Move the contents of hfa384x_dowrid_async (i.e the call to\n hfa384x_dowrid) to its call site.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190712062807.9361-3-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 83c23c8e79ee85c0e7f71844a50aab0c1156a788\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 12 11:57:59 2019 +0530\n\n staging: wlan-ng: Remove function hfa384x_dowrid_wait()\n \n Remove inline function hfa384x_dowrid_wait as it is only called once (in\n hfa384x_drvr_setconfig) and its contents are only a single line (a call\n to hfa384x_dowrid).\n Replace the call to hfa384x_dowrid_wait with its contents i.e the call\n to hfa384x_dowrid.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190712062807.9361-2-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 72da91bc590a3a789b2130718c9dc810851273fe\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 12 11:57:58 2019 +0530\n\n staging: wlan-ng: Remove function hfa384x_dorrid_wait()\n \n Remove function hfa384x_dorrid_wait as it is only called once and it\n does nothing except call hfa384x_dorrid.\n Move contents of hfa384x_dorrid_wait to its only call site to maintain\n functionality.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190712062807.9361-1-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 6fa4e8eb3f1906decdd9d426708bf5134d77f5ad\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Jul 8 12:11:45 2019 +0530\n\n staging: most: dim2: Remove function dimcb_io_write()\n \n Remove function dimcb_io_write as all it does is call writel.\n Modify calls to dimcb_io_write to writel, flipping the order of the\n arguments as required.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190708064145.3250-3-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 41e359e6daf47cd47c9fde77220653886b8f5969\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Jul 8 12:11:44 2019 +0530\n\n staging: most: dim2: Remove function dimcb_io_read()\n \n Remove function dimcb_io_read as it does nothing except call inbuilt\n function readl.\n Modify call sites accordingly.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190708064145.3250-2-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 8826a1985f96d402a66e25fe6eab09144b4d8003\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Mon Jul 8 12:11:43 2019 +0530\n\n staging: most: dim2: Replace function dim_norm_ctrl_async_buffer_size()\n \n Remove function dim_norm_ctrl_async_buffer_size as it does nothing\n except call norm_ctrl_async_buffer_size.\n Rename norm_ctrl_async_buffer_size to dim_norm_ctrl_async_buffer_size to\n maintain compatibility with call sites of the latter.\n Change type of new dim_norm_ctrl_async_buffer_size from static inline to\n non-static to match the old version.\n Modify only remaining call site of norm_ctrl_async_buffer_size to call\n dim_norm_ctrl_async_buffer_size instead.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190708064145.3250-1-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit d327ace0f301c6c203c0ed1aa5da823ab6fef5f6\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Jul 5 09:25:54 2019 +0100\n\n staging: rtl8712: remove redundant assignment to variable res\n \n The variable res is being initialized with a value that is never\n read and it is being updated later with a new value. The\n initialization is redundant and can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Link: https://lore.kernel.org/r/20190705082554.15588-1-colin.king@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 2ac3b368f40f31fccad2377939a66e6986b54976\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 12 12:42:47 2019 +0530\n\n staging: erofs: Remove function erofs_kill_sb()\n \n Remove function erofs_kill_sb as all it does is call kill_block_super.\n Modify references to the former to point to the latter.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Reviewed-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Reviewed-by: Chao Yu \u003cyuchao0@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190712071247.2357-1-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit e82a9a17d49c52a5ea96589a0ee738d3fef33323\nAuthor: Pratik Shinde \u003cpratikshinde320@gmail.com\u003e\nDate: Mon Jul 15 17:51:27 2019 +0530\n\n staging: erofs:converting all \u0027unsigned\u0027 to \u0027unsigned int\u0027\n \n Fixed checkpatch warnings: converting all \u0027unsigned\u0027 to \u0027unsigned int\u0027\n \n Signed-off-by: Pratik Shinde \u003cpratikshinde320@gmail.com\u003e\n Reviewed-by: Gao Xiang \u003cgaoxiang25@huawei.com\u003e\n Link: https://lore.kernel.org/r/20190715122127.13733-1-pratikshinde320@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit e1928327353d5ddd00b7d4d46f46961ea5c2dc34\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Thu Jul 4 14:30:31 2019 +0100\n\n staging: greybus: remove redundant assignment to variable is_empty\n \n The variable is_empty is being initialized with a value that is never\n read and it is being updated later with a new value. The\n initialization is redundant and can be removed.\n \n Addresses-Coverity: (\"Unused value\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Reviewed-by: Vaibhav Agarwal \u003cvaibhav.sr@gmail.com\u003e\n Reviewed-by: Mark Greer \u003cmgreer@animalcreek.com\u003e\n Link: https://lore.kernel.org/r/20190704133031.28809-1-colin.king@canonical.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit baacd53a3b72d1dc74a340d16b8cf560dfe24717\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Fri Jul 5 18:41:31 2019 +0530\n\n staging: greybus: Replace function gb_i2c_device_setup()\n \n Remove function gb_i2c_device_setup as all it does is call\n gb_i2c_functionality_operation.\n Rename gb_i2c_functionality_operation to gb_i2c_device_setup to maintain\n compatibility with call sites.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190705131131.27354-1-nishkadg.linux@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit ebf256e36754fc5c4a259bb28f3c0a081eb70467\nAuthor: Keyur Patel \u003ciamkeyur96@gmail.com\u003e\nDate: Sun Jul 14 13:27:06 2019 -0400\n\n staging: most: Delete an error message for a failed memory allocation\n \n The kfifo_alloc() failure generates enough information and doesn\u0027t need\n to be accompanied by another error statement.\n \n Signed-off-by: Keyur Patel \u003ciamkeyur96@gmail.com\u003e\n Link: https://lore.kernel.org/r/20190714172708.5067-1-iamkeyur96@gmail.com\n Signed-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n\ncommit 5d4b45a1dd7b00feab57624035dcdbc1bab2e0f8\nAuthor: Markus Koch \u003cmarkus@notsyncing.net\u003e\nDate: Sun Jul 21 20:20:28 2019 +0300\n\n Input: add support for the FlySky FS-iA6B RC receiver\n \n This patch adds support for the FlySky FS-iA6B RC receiver (serial IBUS).\n \n It allows the usage of the FlySky FS-i6 and other AFHDS compliant remote\n controls as a joystick input device.\n \n To use it, a patch to inputattach which adds the FS-iA6B as a 115200 baud\n serial device is required. I will upstream it after this patch is merged.\n \n More information about the hardware can be found here:\n \n https://notsyncing.net/?p\u003dblog\u0026b\u003d2018.linux-fsia6b\n \n Signed-off-by: Markus Koch \u003cmarkus@notsyncing.net\u003e\n Signed-off-by: Dmitry Torokhov \u003cdmitry.torokhov@gmail.com\u003e\n\ncommit 1991529f0fc91e0931ef534906397138e21f56d3\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Fri Jun 21 15:07:17 2019 +0800\n\n arm64: Enable TIMER_IMX_SYS_CTR for ARCH_MXC platforms\n \n ARCH_MXC platforms needs system counter as broadcast timer\n to support cpuidle, enable it by default.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 87ff1f88552a7d3b3f865d89ce4ed002777fe2ab\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jun 19 13:52:47 2019 +0800\n\n arm64: defconfig: Select CONFIG_CLK_IMX8MN by default\n \n Enable CONFIG_CLK_IMX8MN to support i.MX8MN clock driver.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 9d9005a5a23fc92832cc1b042ac6de76891caccc\nAuthor: Guido Günther \u003cagx@sigxcpu.org\u003e\nDate: Tue Jun 25 10:27:21 2019 +0200\n\n arm64: dts: imx8mq-librem5: Enable MIPI D-PHY\n \n This enables the Mixel MIPI D-PHY on the Librem 5 devkit\n \n Signed-off-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n Acked-by: Angus Ainslie (Purism) \u003cangus@akkea.ca\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit a99b26b14bea50624db9e971f7b9e422ee2bb6c1\nAuthor: Guido Günther \u003cagx@sigxcpu.org\u003e\nDate: Tue Jun 25 10:27:20 2019 +0200\n\n arm64: dts: imx8mq: Add MIPI D-PHY\n \n Add a node for the Mixel MIPI D-PHY, \"disabled\" by default.\n \n Signed-off-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n Acked-by: Angus Ainslie (Purism) \u003cangus@akkea.ca\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 0be9af7770493955a27b704bf502b146f8167b6e\nAuthor: Robin Gong \u003cyibin.gong@nxp.com\u003e\nDate: Thu Jul 18 10:48:18 2019 +0800\n\n ARM: dts: imx7ulp: add edma device node\n \n Add edma device node in dts.\n \n Signed-off-by: Robin Gong \u003cyibin.gong@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 5f7a6cd1345716195c27876b1b2a100291da6bab\nAuthor: Fabio Estevam \u003cfestevam@gmail.com\u003e\nDate: Mon Jun 24 17:54:32 2019 -0300\n\n ARM: dts: imx7-colibri: Remove unneeded snvs_pwrkey node\n \n Since commit 4664179fe679 (\"ARM: dts: imx7s: Enable SNVS power key\n according to board design\") snvs_pwrkey is disabled by default, so there\n is no need for disabling it explicitly in the board dts anymore.\n \n Signed-off-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Reviewed-by: Chris Healy \u003ccphealy@gmail.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 4931b5f160164807c33f0d935c76781695ab7757\nAuthor: Fabio Estevam \u003cfestevam@gmail.com\u003e\nDate: Mon Jun 24 17:54:31 2019 -0300\n\n ARM: dts: imx7d-zii-rpu2: Remove unneeded snvs_pwrkey node\n \n Since commit 4664179fe679 (\"ARM: dts: imx7s: Enable SNVS power key\n according to board design\") snvs_pwrkey is disabled by default, so there\n is no need for disabling it explicitly in the board dts anymore.\n \n Signed-off-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Reviewed-by: Chris Healy \u003ccphealy@gmail.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit f496e6750083af4200e2f8fb9914db9c1bf93d32\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Jun 24 11:30:43 2019 -0700\n\n ARM: dts: Add ZII support for ZII i.MX7 RMU2 board\n \n Add support for ZII\u0027s i.MX7 based Remote Modem Unit 2 (RMU2) board.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Cc: Shawn Guo \u003cshawnguo@kernel.org\u003e\n Cc: Rob Herring \u003crobh@kernel.org\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Cc: Bob Langer \u003cBob.Langer@zii.aero\u003e\n Cc: Liang Pan \u003cLiang.Pan@zii.aero\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: linux-kernel@vger.kernel.org\n Cc: devicetree@vger.kernel.org\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 779988c5e61847774d81ec43d77b244f43ea6d79\nAuthor: Robin Gong \u003cyibin.gong@nxp.com\u003e\nDate: Mon Jun 10 16:17:51 2019 +0800\n\n ARM: dts: imx6sll: correct sdma compatible\n \n Correct sdma compatible since ecspi errata ERR009165 has been fixed\n on i.mx6sll as i.mx6ul.\n \n Signed-off-by: Robin Gong \u003cyibin.gong@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit c6c0ad740916425669d5c4a64cf1436f76e40065\nAuthor: Robin Gong \u003cyibin.gong@nxp.com\u003e\nDate: Mon Jun 10 16:17:50 2019 +0800\n\n ARM: dts: imx6ul: add dma support on ecspi\n \n Add dma support on ecspi.\n \n Signed-off-by: Robin Gong \u003cyibin.gong@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 8dfe397431e0e05d9e66a9f20511833bda91f978\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Wed Jun 26 15:44:15 2019 +0800\n\n soc: imx8: Add i.MX8MM UID(unique identifier) support\n \n Add i.MX8MM SoC UID(unique identifier) support, user\n can read it from sysfs:\n \n root@imx8mmevk:~# cat /sys/devices/soc0/soc_uid\n B365FA0A5C85D6EE\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 3026d0b7a4972fd5fb56f5a454d1efe7534e84fc\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Thu Jul 18 15:13:11 2019 +0800\n\n soc: imx8: Add i.MX8MQ UID(unique identifier) support\n \n Add i.MX8MQ SoC UID(unique identifier) support, user\n can read it from sysfs:\n \n root@imx8mqevk:~# cat /sys/devices/soc0/soc_uid\n D56911D6F060954B\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 4d53ea811b212b58b5696ef501b6822430253ba9\nAuthor: Anson Huang \u003cAnson.Huang@nxp.com\u003e\nDate: Tue Jun 11 20:25:35 2019 +0800\n\n arm64: defconfig: Select CONFIG_PINCTRL_IMX8MN by default\n \n Enable CONFIG_PINCTRL_IMX8MN by default to support i.MX8MN\n pinctrl driver.\n \n Signed-off-by: Anson Huang \u003cAnson.Huang@nxp.com\u003e\n Reviewed-by: Dong Aisheng \u003caisheng.dong@nxp.com\u003e\n Acked-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 7f4e4afa140cd3bccd0f0b408f7c4a211eaef2ac\nAuthor: Robin Gong \u003cyibin.gong@nxp.com\u003e\nDate: Wed Jul 17 14:47:10 2019 +0800\n\n arm64: defconfig: Enable SDMA on i.mx8mq/8mm\n \n Enable SDMA support on i.mx8mq/8mm chips, including enabling\n CONFIG_FW_LOADER_USER_HELPER/CONFIG_FW_LOADER_USER_HELPER_FALLBACK\n for firmware loaded by udev.\n \n Signed-off-by: Robin Gong \u003cyibin.gong@nxp.com\u003e\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 849b7e3bb2029cfd6fb353509449b5fac914e682\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Mon Jun 24 11:30:44 2019 -0700\n\n dt-bindings: arm: fsl: Add support for ZII i.MX7 RMU2 board\n \n Add support for ZII i.MX7 RMU2 board.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Cc: Shawn Guo \u003cshawnguo@kernel.org\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Cc: Bob Langer \u003cBob.Langer@zii.aero\u003e\n Cc: Liang Pan \u003cLiang.Pan@zii.aero\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n\ncommit 84ebd2da6d046781a67430cf3e6a138afe2ce15e\nAuthor: Sean Paul \u003cseanpaul@chromium.org\u003e\nDate: Mon Apr 1 10:17:20 2019 -0700\n\n arm64: dts: rockchip: Specify override mode for kevin panel\n \n This patch adds an override mode for kevin devices. The mode increases\n both back porches to allow a pixel clock of 26666kHz as opposed to the\n \u0027typical\u0027 value of 252750kHz. This is needed to avoid interference with\n the touch digitizer on these laptops.\n \n Cc: Doug Anderson \u003cdianders@chromium.org\u003e\n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Cc: Jeffy Chen \u003cjeffy.chen@rock-chips.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Stéphane Marchesin \u003cmarcheu@chromium.org\u003e\n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: devicetree@vger.kernel.org\n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-rockchip@lists.infradead.org\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Tested-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Signed-off-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit 123643e5c40a9c7d77649e306ccb5b0354938d49\nAuthor: Douglas Anderson \u003cdianders@chromium.org\u003e\nDate: Mon Apr 1 10:17:24 2019 -0700\n\n ARM: dts: rockchip: Specify rk3288-veyron-minnie\u0027s display timings\n \n Just like we did for rk3288-veyron-chromebook, we want to be able to\n use one of the fixed PLLs in the system to make the pixel clock for\n minnie.\n \n Specifying these timings matches us with how the display is used on\n the downstream Chrome OS kernel. See https://crrev.com/c/323211.\n \n Unlike what we did for rk3288-veyron-chromebook, this CL actually\n changes the timings (though not the pixel clock) that is used when\n using the upstream kernel. Booting up a minnie shows that it ended up\n with a 66.67 MHz pixel clock but it was still using the\n porches/blankings it would have wanted for a 72.5 MHz pixel clock.\n \n NOTE: compared to the downstream kernel, this seems to cause a\n slightly different result reported in the \u0027modetest\u0027 command on a\n Chromebook. The downstream kernel shows:\n 1280x800 60 1280 1298 1330 1351 800 804 822 830 66667\n \n With this patch we have:\n 1280x800 59 1280 1298 1330 1351 800 804 822 830 66666\n \n Specifically modetest was reporting 60 Hz on the downstream kernel but\n the upstream kernel does the math and comesup with 59 (because we\n actually achieve 59.45 Hz). Also upstream doesn\u0027t round the Hz up\n when converting to kHz--it seems to truncate.\n \n ALSO NOTE: when I look at the EDID from the datasheet, I see:\n -hsync -vsync\n ...but it seems like we\u0027ve never actually run with that so I\u0027ve\n continued leaving that out.\n \n Signed-off-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit 95671ec23696d7351b47d159159c6bdcb64fafe4\nAuthor: Douglas Anderson \u003cdianders@chromium.org\u003e\nDate: Mon Apr 1 10:17:23 2019 -0700\n\n ARM: dts: rockchip: Specify rk3288-veyron-chromebook\u0027s display timings\n \n Let\u0027s document the display timings that most veyron chromebooks (like\n jaq, jerry, mighty, speedy) have been using out in the field. This\n uses the standard blankings but a slightly slower clock rate, thus\n getting a refresh rate 58.3 Hz.\n \n NOTE: this won\u0027t really do anything except cause DRM to properly\n report the refresh rate since vop_crtc_mode_fixup() was rounding the\n pixel clock to 74.25 MHz anyway. Apparently the adjusted rate isn\u0027t\n exposed to userspace so it\u0027s important that the rate we\u0027re trying to\n achieve is mostly right.\n \n For the downstream kernel change related to this see See\n https://crrev.com/c/324558.\n \n NOTE: minnie uses a different panel will be fixed up in a future\n patch, so for now we\u0027ll just delete the panel timings there.\n \n Signed-off-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n\ncommit c9b07eab0c8760bdd4cf8624c482ee145a322a3b\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Sun Jul 14 13:22:27 2019 -0400\n\n audit_inode(): switch to passing AUDIT_INODE_...\n \n don\u0027t bother with remapping LOOKUP_... values - all callers pass\n constants and we can just as well pass the right ones from the\n very beginning.\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 39145f5f0cc924b494ad55a2bc9c1b4969b5a038\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Sun Jul 14 12:19:57 2019 -0400\n\n filename_mountpoint(): make LOOKUP_NO_EVAL unconditional there\n \n user_path_mountpoint_at() always gets it and the reasons to have it\n there (i.e. in umount(2)) apply to kern_path_mountpoint() callers\n as well.\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit ff0ebee239ce3461cab68ebb8e1a359fc34329de\nAuthor: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nDate: Sun Jul 14 12:18:05 2019 -0400\n\n filename_lookup(): audit_inode() argument is always 0\n \n We hadn\u0027t been passing LOOKUP_PARENT in flags to that thing\n since filename_parentat() had been split off back in 2015.\n \n Signed-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n\ncommit 61cbfff4b1a7c15a7e403473ca5a290fd13d5656\nAuthor: Dmitry Korotin \u003cdkorotin@wavecomp.com\u003e\nDate: Tue Jun 18 21:25:43 2019 +0000\n\n MIPS: pte_special()/pte_mkspecial() support\n \n Add support for pte_special() \u0026 pte_mkspecial(), replacing our previous\n stubs with functional implementations.\n \n Signed-off-by: Dmitry Korotin \u003cdkorotin@wavecomp.com\u003e\n [paul.burton@mips.com:\n - Fix for CONFIG_PHYS_ADDR_T_64BIT \u0026\u0026 CONFIG_CPU_MIPS32.\n - Rewrite commit message.]\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-mips@vger.kernel.org\n\ncommit a23c4134955e87d9e16149daedefd3a602fb019b\nAuthor: Jonas Gorski \u003cjonas.gorski@gmail.com\u003e\nDate: Thu May 2 14:26:57 2019 +0200\n\n MIPS: BMIPS: add clock controller nodes\n \n Now that we have a driver for the clock controller, add nodes to allow\n devices to make use of it.\n \n Signed-off-by: Jonas Gorski \u003cjonas.gorski@gmail.com\u003e\n Reviewed-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Reviewed-by: Philippe Mathieu-Daudé \u003cf4bug@amsat.org\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: linux-clk@vger.kernel.org\n Cc: devicetree@vger.kernel.org\n Cc: linux-mips@vger.kernel.org\n Cc: Michael Turquette \u003cmturquette@baylibre.com\u003e\n Cc: Stephen Boyd \u003csboyd@kernel.org\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: bcm-kernel-feedback-list@broadcom.com\n Cc: Kevin Cernekee \u003ccernekee@gmail.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n\ncommit 5dad549d94c8dac1a99bbc72d481d64dc4974ea3\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Fri Jan 25 17:09:26 2019 -0300\n\n MIPS: DTS: jz4740: Add node for the MMC driver\n \n Add a devicetree node for the jz4740-mmc driver.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Mark Rutland \u003cmark.rutland@arm.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: Ulf Hansson \u003culf.hansson@linaro.org\u003e\n Cc: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Cc: devicetree@vger.kernel.org\n Cc: linux-mips@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n Cc: linux-mmc@vger.kernel.org\n\ncommit 368fb26c1e55a187131a794f95420d96d31d0e5e\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Wed May 8 00:43:57 2019 +0200\n\n MIPS: Decode config3 register on Ingenic SoCs\n \n XBurst misses a config2 register, so config3 decode was skipped in\n decode_configs().\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: od@zcrc.me\n Cc: linux-mips@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n\ncommit 3b25b763116482596227225bea7c03fcde11c9ed\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Wed May 8 00:43:56 2019 +0200\n\n MIPS: Rename JZRISC to XBURST\n \n The real name of the CPU present in the JZ line of SoCs from Ingenic is\n XBurst, not JZRISC.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n [paul.burton@mips.com: Leave /proc/cpuinfo string as-is.]\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: od@zcrc.me\n Cc: linux-mips@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n\ncommit 8a6836ee384909ae01e3bd0731d082e0e6667e54\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Thu May 16 17:56:14 2019 -0500\n\n ARM: dts: at91: Avoid colliding \u0027display\u0027 node and property names\n \n While properties and child nodes with the same name are valid DT, the\n practice is not encouraged. Furthermore, the collision is problematic for\n YAML encoded DT. Let\u0027s just avoid the issue and rename the nodes.\n \n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://lore.kernel.org/r/20190516225614.1458-1-robh@kernel.org\n Acked-by: Nicolas Ferre \u003cnicolas.ferre@microchip.com\u003e\n Signed-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n\ncommit 7176b6ac18469f63b3aa1f6e2997776ad81273b4\nAuthor: Daniel Silsby \u003cdansilsby@gmail.com\u003e\nDate: Mon Jul 15 17:40:03 2019 -0400\n\n MIPS: Undefine PMD_ORDER for 32-bit builds\n \n During an update long ago to conform to 4-level page code, PMD_ORDER was\n changed from 0 to 1, despite the fact that a PMD table is not used at\n all in a 32-bit MIPS build. PMD_ORDER does not seem to be used in these\n builds. Now, it matches PUD_ORDER, a nonsense #define to give a build\n failure with informative error.\n \n The older commit that had redefined PMD_ORDER was\n commit c6e8b587718c (\"Update MIPS to use the 4-level pagetable code\n thereby getting rid of the compacrapability headers.\")\n \n Signed-off-by: Daniel Silsby \u003cdansilsby@gmail.com\u003e\n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: od@zcrc.me\n Cc: linux-mips@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n\ncommit b35d265301ecdc51ca90741739f71b1abfedaea1\nAuthor: Daniel Silsby \u003cdansilsby@gmail.com\u003e\nDate: Mon Jul 15 17:40:02 2019 -0400\n\n MIPS: ingenic: Add support for huge pages\n \n The Ingenic jz47xx SoC series of 32-bit MIPS CPUs support huge pages.\n \n Signed-off-by: Daniel Silsby \u003cdansilsby@gmail.com\u003e\n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: od@zcrc.me\n Cc: linux-mips@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n\ncommit 45e03e62724f636e2b353a7eda5f7be9bfdae2c3\nAuthor: Daniel Silsby \u003cdansilsby@gmail.com\u003e\nDate: Mon Jul 15 17:40:01 2019 -0400\n\n MIPS: Decouple CPU_SUPPORTS_HUGEPAGES from 64BIT\n \n We now have partial 32-bit MIPS huge page support, so there\u0027s no need\n to restrict these config options only to 64-bit systems.\n \n Signed-off-by: Daniel Silsby \u003cdansilsby@gmail.com\u003e\n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: od@zcrc.me\n Cc: linux-mips@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n\ncommit 35476311e5292ae0ef00dde5c2cccbe8514660c2\nAuthor: Daniel Silsby \u003cdansilsby@gmail.com\u003e\nDate: Mon Jul 15 17:40:00 2019 -0400\n\n MIPS: Add partial 32-bit huge page support\n \n This adds initial support for huge pages to 32-bit MIPS systems.\n Systems with extended addressing enabled (EVA,XPA,Alchemy/Netlogic)\n are not yet supported.\n With huge pages enabled, this implementation will increase page table\n memory overhead to match that of a 64-bit MIPS system. However, the\n cache-friendliness of page table walks is not affected significantly.\n \n Signed-off-by: Daniel Silsby \u003cdansilsby@gmail.com\u003e\n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: od@zcrc.me\n Cc: linux-mips@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n\ncommit 171543e752723191281994e51278050526d1c017\nAuthor: Daniel Silsby \u003cdansilsby@gmail.com\u003e\nDate: Mon Jul 15 17:39:59 2019 -0400\n\n MIPS: Disallow CPU_SUPPORTS_HUGEPAGES for XPA,EVA\n \n In preparation for 32-bit MIPS huge page support.\n \n EVA,XPA are extended-addressing modes for 32-bit MIPS systems. Because\n huge pages aren\u0027t currently supported in 32-bit MIPS, this doesn\u0027t take\n any features away from EVA,XPA-enabled systems. However, the soon-to-\n come 32-bit MIPS huge page support doesn\u0027t yet support them.\n \n This also disables CPU_SUPPORTS_HUGEPAGES for the small number of 32-bit\n MIPS CPUs from Alchemy/Netlogic that support a custom 36-bit extended\n addressing. It\u0027s unknown if they even support huge pages in hardware.\n \n Signed-off-by: Daniel Silsby \u003cdansilsby@gmail.com\u003e\n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Signed-off-by: Paul Burton \u003cpaul.burton@mips.com\u003e\n Cc: Ralf Baechle \u003cralf@linux-mips.org\u003e\n Cc: James Hogan \u003cjhogan@kernel.org\u003e\n Cc: od@zcrc.me\n Cc: linux-mips@vger.kernel.org\n Cc: linux-kernel@vger.kernel.org\n\ncommit e4f86e43716443e934d705952902d40de0fa9a05\nAuthor: Hans de Goede \u003chdegoede@redhat.com\u003e\nDate: Sun Jul 21 15:25:25 2019 +0200\n\n drm: Add Grain Media GM12U320 driver v2\n \n Add a modesetting driver for Grain Media GM12U320 based devices\n (primarily Acer C120 projector, but there may be compatible devices).\n \n This is based on the fb driver from Viacheslav Nurmekhamitov:\n https://github.com/slavrn/gm12u320\n \n This driver uses drm_simple_display_pipe to deal with all the atomic\n stuff, gem_shmem_helper functions for buffer management and\n drm_fbdev_generic_setup for fbdev emulation, so that leaves the driver\n itself with only the actual code for talking to the gm12u320 chip,\n leading to a nice simple and clean driver.\n \n Changes in v2:\n -Add drm-misc tree to MAINTAINERS\n -Drop mode_config.preferred_depth \u003d 24 / fix fbdev support\n \n Reviewed-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Signed-off-by: Hans de Goede \u003chdegoede@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190721132525.10396-1-hdegoede@redhat.com\n\ncommit 73415b4ea5b5a257db6b5bd135b305cd945e63de\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jul 21 16:06:10 2019 +0200\n\n drm/fb: remove unused function: drm_gem_fbdev_fb_create()\n \n After migrating several drivers to the generic fbdev\n emulation there are no users left of drm_gem_fbdev_fb_create.\n Delete the function.\n \n Noticed that there was no callers while browsing\n around in the drm_fb* code.\n \n The code that referenced the function was removed by:\n commit 13aff184ed9f (\"drm/qxl: remove dead qxl fbdev emulation code\")\n \n The actual use was removed by:\n commit 26d4707d445d (\"drm/qxl: use generic fbdev emulation\")\n \n v2:\n - Updated changelog based on feedback from Noralf\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Cc: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190721140610.GA20842@ravnborg.org\n\ncommit b72cb0dc4c46fe75a52a9adb4a13ba1213e50000\nAuthor: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\nDate: Sat Jul 13 14:03:46 2019 +0200\n\n drm/sun4i: sun8i-csc: Add support for color encoding and range\n \n Conversion from YUV to RGB depends on range (limited or full) and\n encoding (BT.601 or BT.709). Current code doesn\u0027t consider this and\n always uses BT.601 encoding and limited range.\n \n Fix this by introducing new CSC matrices, which are selected based on\n range and encoding parameters.\n \n Signed-off-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190713120346.30349-4-jernej.skrabec@siol.net\n\ncommit ab38c3b7c3472a6626aceb12fc2a23828152daf3\nAuthor: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\nDate: Sat Jul 13 14:03:45 2019 +0200\n\n drm/sun4i: sun8i_csc: Simplify register writes\n \n It turns out addition of 0x200 to constant parts (+0.5) is not really\n necessary. Besides, we can consider that before and fix value in CSC\n matrix.\n \n This simplifies register writes quiet a bit.\n \n Signed-off-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190713120346.30349-3-jernej.skrabec@siol.net\n\ncommit 5917e0bdbc90cae408561f713eaa7d60e512e599\nAuthor: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\nDate: Sat Jul 13 14:03:44 2019 +0200\n\n drm/sun4i: Introduce color encoding and range properties\n \n In order to correctly convert YUV color space to RGB, we have to know\n color encoding and range.\n \n Introduce these two properties using helper method.\n \n Signed-off-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190713120346.30349-2-jernej.skrabec@siol.net\n\ncommit 1eef08cb7dc7f240905071ea18eebd1232f302f6\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Thu Jul 18 18:15:06 2019 +0200\n\n drm/mediatek: direct include of drm.h in mtk_drm_gem.c\n \n Do not rely on including drm.h from drm_file.h,\n as the include in drm_file.h will be dropped.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Sean Paul \u003csean@poorly.run\u003e\n Acked-by: CK Hu \u003cck.hu@mediatek.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: linux-mediatek@lists.infradead.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190718161507.2047-11-sam@ravnborg.org\n\ncommit d89281c5edadc646a0445f2967513d15ea5876df\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Thu Jul 18 18:15:05 2019 +0200\n\n drm: direct include of drm.h in drm_syncobj.c\n \n Do not rely on including drm.h from drm_file.h,\n as the include in drm_file.h will be dropped.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Sean Paul \u003csean@poorly.run\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Cc: Chunming Zhou \u003cdavid1.zhou@amd.com\u003e\n Cc: Christian König \u003cckoenig.leichtzumerken@gmail.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190718161507.2047-10-sam@ravnborg.org\n\ncommit 39716270d88c157b722e1f11adf9dada6bec3f11\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Thu Jul 18 18:15:04 2019 +0200\n\n drm: direct include of drm.h in drm_prime.c\n \n Do not rely on including drm.h from drm_file.h,\n as the include in drm_file.h will be dropped.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Sean Paul \u003csean@poorly.run\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Christian König \u003cchristian.koenig@amd.com\u003e\n Cc: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190718161507.2047-9-sam@ravnborg.org\n\ncommit d3ea256aa4c7db9f7d366eb196c0edf5cca2f1f2\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Thu Jul 18 18:15:03 2019 +0200\n\n drm: direct include of drm.h in drm_gem_shmem_helper.c\n \n Do not rely on including drm.h from drm_file.h,\n as the include in drm_file.h will be dropped.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Sean Paul \u003csean@poorly.run\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Cc: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190718161507.2047-8-sam@ravnborg.org\n\ncommit 1c53587629f62d71f2ef22fb64c045ecd57a0e2e\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Thu Jul 18 18:15:02 2019 +0200\n\n drm: direct include of drm.h in drm_gem.c\n \n Do not rely on including drm.h from drm_file.h,\n as the include in drm_file.h will be dropped.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Sean Paul \u003csean@poorly.run\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Cc: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190718161507.2047-7-sam@ravnborg.org\n\ncommit 03fda35d109b0a175573f2fd9145d622328ae8a0\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Thu Jul 18 18:15:00 2019 +0200\n\n drm/ati_pcigart: drop dependency on drm_os_linux.h\n \n The drm_os_linux.h header is deprecated.\n Just opencode the sole DRM_WRITE32().\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Sean Paul \u003csean@poorly.run\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190718161507.2047-5-sam@ravnborg.org\n\ncommit 3e37c7156e94442025e88de58de5ed966f2b64a0\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Thu Jul 18 18:14:59 2019 +0200\n\n drm: drop uapi dependency from drm_vblank.h\n \n drm_vblank.h included uapi/drm/drm.h.\n It turns out this include was not required - delete it.\n \n Note: uapi/drm/drm.h is included indirect via drm_file.h,\n but there are no dependencies in drm_vblank.h so the removal\n is legit.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Reviewed-by: Sean Paul \u003csean@poorly.run\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190718161507.2047-4-sam@ravnborg.org\n\ncommit bf3f5e98559360661a3d2af340d46522512c0b00\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Thu Jul 18 18:14:57 2019 +0200\n\n drm/panel: make drm_panel.h self-contained\n \n Fix build warning if drm_panel.h is built with CONFIG_OF\u003dn or\n CONFIG_DRM_PANEL\u003dn and included without the prerequisite err.h:\n \n ./include/drm/drm_panel.h: In function ‘of_drm_find_panel’:\n ./include/drm/drm_panel.h:203:9: error: implicit declaration of function ‘ERR_PTR’ [-Werror\u003dimplicit-function-declaration]\n return ERR_PTR(-ENODEV);\n ^~~~~~~\n ./include/drm/drm_panel.h:203:9: error: returning ‘int’ from a function with return type ‘struct drm_panel *’ makes pointer from integer without a cast [-Werror\u003dint-conversion]\n return ERR_PTR(-ENODEV);\n ^~~~~~~~~~~~~~~~\n \n Fixes: 5fa8e4a22182 (\"drm/panel: Make of_drm_find_panel() return an ERR_PTR() instead of NULL\")\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Sean Paul \u003csean@poorly.run\u003e\n Cc: Boris Brezillon \u003cbbrezillon@kernel.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190718161507.2047-2-sam@ravnborg.org\n\ncommit 0cc35a9c8279da03c34547830d5140cf38f7ce02\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Jul 19 01:51:36 2019 +0000\n\n drm/i915/dsi: remove set but not used variable \u0027hfront_porch\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/gpu/drm/i915/display/icl_dsi.c: In function \u0027gen11_dsi_set_transcoder_timings\u0027:\n drivers/gpu/drm/i915/display/icl_dsi.c:768:6: warning:\n variable \u0027hfront_porch\u0027 set but not used [-Wunused-but-set-variable]\n \n It is never used and can be removed.\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190719015136.103988-1-yuehaibing@huawei.com\n\ncommit 5cad0ddf4b78559221a123c627df291cbe4d3ccd\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Fri Jul 19 15:33:22 2019 +0000\n\n drm/i915/gtt: Don\u0027t try to clear failed empty pd allocation\n \n When __gen8_ppgtt_alloc fails without allocating anything\n we should not try to call __gen8_ppgtt_clear as there is\n nothing to clear and underlying code will complain with:\n \n [ 157.861645] gen8_pd_range:881 GEM_BUG_ON(start \u003e\u003d end)\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190719153322.10464-1-michal.wajdeczko@intel.com\n\ncommit 6b5f3cb1aa20e196d35aba21fbf363d3736ff4d5\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 19 14:15:24 2019 +0100\n\n drm/i915/gtt: Correct unshifted \u0027from\u0027 for gen8_ppgtt_alloc errors\n \n Since the underlying __gen8_ppgtt_clear takes the shifted address, we\n must remember to provide it with the shifted original start address.\n \n Reported-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Abdiel Janulgue \u003cabdiel.janulgue@linux.intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Tested-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190719131524.827-1-chris@chris-wilson.co.uk\n\ncommit a0301020c568ff6b94da0b988ab984b90ea702a1\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jul 19 10:48:45 2019 +0100\n\n Revert \"drm/i915: Update description of i915.enable_guc modparam\"\n \n This reverts commit 0629d4da1f159778063767fb0ac1c951034c5477.\n \n If GuC firmware is not present on the filesystem driver crashes the\n machine on boot.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Acked-by: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Fixes: 0629d4da1f15 (\"drm/i915: Update description of i915.enable_guc modparam\")\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@linux.intel.com\u003e\n Cc: intel-gfx@lists.freedesktop.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190719094845.6242-3-tvrtko.ursulin@linux.intel.com\n\ncommit 92508da7e3daa1cd7ba18685d067a41c89a0f3ec\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jul 19 10:48:44 2019 +0100\n\n Revert \"drm/i915/guc: Turn on GuC/HuC auto mode\"\n \n This reverts commit f774f09649192f326fa030564afd3f8f5d82c1e4.\n \n If GuC firmware is not present on the filesystem driver crashes the\n machine on boot.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Acked-by: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Fixes: f774f0964919 (\"drm/i915/guc: Turn on GuC/HuC auto mode\")\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@linux.intel.com\u003e\n Cc: intel-gfx@lists.freedesktop.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190719094845.6242-2-tvrtko.ursulin@linux.intel.com\n\ncommit b83a309a9889225ca5acf75897ec35a9bc33cdaf\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Wed Jul 17 19:06:24 2019 +0100\n\n drm/i915/icl: Add Wa_1409178092\n \n We were missing this workaround which can cause hangs if fine grained\n coherency was used.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190717180624.20354-7-tvrtko.ursulin@linux.intel.com\n\ncommit 935ba6f3bf0fad42aac2b10d19498cf27c33cc98\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Wed Jul 17 19:06:23 2019 +0100\n\n drm/i915/icl: Verify engine workarounds in GEN8_L3SQCREG4\n \n Having fixed the incorect MCR programming in an earlier patch, we can now\n stop ignoring read back of GEN8_L3SQCREG4 during engine workaround\n verification.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190717180624.20354-6-tvrtko.ursulin@linux.intel.com\n\ncommit fa380486d5f995b6914b4d4149743d330125414e\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Wed Jul 17 19:06:22 2019 +0100\n\n drm/i915: Skip CS verification of L3 bank registers\n \n Access to 0xb100 - 0xb3ff mmio range is controlled by the MCR selector\n which only affects CPU MMIO. Therefore these registers cannot be realiably\n read with MI_SRM from the command streamer so skip their verification.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190717180624.20354-5-tvrtko.ursulin@linux.intel.com\n\ncommit 6c2b0103ad92b4238c26b0b8f224f4581d98b2fe\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Wed Jul 17 19:06:21 2019 +0100\n\n drm/i915: Fix and improve MCR selection logic\n \n A couple issues were present in this code:\n \n 1.\n fls() usage was incorrect causing off by one in subslice mask lookup,\n which in other words means subslice mask of all zeroes is always used\n (subslice mask of a slice which is not present, or even out of bounds\n array access), rendering the checks in wa_init_mcr either futile or\n random.\n \n 2.\n Condition in WARN_ON was not correct. It is doing a bitwise and operation\n between a positive (present subslices) and negative mask (disabled L3\n banks).\n \n This means that with corrected fls() usage the assert would always\n incorrectly fail.\n \n We could fix this by inverting the fuse bits in the check, but instead do\n one better and improve the code so it not only asserts, but finds the\n first common index between the two masks and only warns if no such index\n can be found.\n \n v2:\n * Simplify check for logic and redability.\n * Improve commentary explaining what is really happening ie. what the\n assert is really trying to check and why.\n \n v3:\n * Find first common index instead of just asserting.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Fixes: fe864b76c2ab (\"drm/i915: Implement WaProgramMgsrForL3BankSpecificMmioReads\")\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e # v1\n Cc: Michał Winiarski \u003cmichal.winiarski@intel.com\u003e\n Cc: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190717180624.20354-4-tvrtko.ursulin@linux.intel.com\n\ncommit 7405cb77e008aa83908c9824ff7897c403929d57\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Wed Jul 17 19:06:20 2019 +0100\n\n drm/i915: Trust programmed MCR in read_subslice_reg\n \n Instead of re-calculating the MCR selector in read_subslice_reg do the\n rwm on its existing value and restore it when done.\n \n This consolidates MCR programming to one place for cnl+, and avoids\n re-calculating its default value on older platforms during hangcheck.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190717180624.20354-3-tvrtko.ursulin@linux.intel.com\n\ncommit 15160879d47213c32f357bc67b6014d9aaf14ed7\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Wed Jul 17 19:06:19 2019 +0100\n\n drm/i915: Fix GEN8_MCR_SELECTOR programming\n \n fls returns bit positions starting from one for the lsb and the MCR\n register expects zero based (sub)slice addressing.\n \n Incorrent MCR programming can have the effect of directing MMIO reads of\n registers in the 0xb100-0xb3ff range to invalid subslice returning zeroes\n instead of actual content.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Fixes: 1e40d4aea57b (\"drm/i915/cnl: Implement WaProgramMgsrForCorrectSliceSpecificMmioReads\")\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190717180624.20354-2-tvrtko.ursulin@linux.intel.com\n\ncommit 1e9907362453b9ddf9e73a144bec91f15ca0d536\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Thu Jul 4 16:50:54 2019 +0200\n\n drm/doc: Document kapi doc expectations\n \n We\u0027ve had this already for anything new. With my drm_prime.c cleanup I\n also think documentation for everything already existing is complete,\n and we can bake this in as a requirements subsystem wide.\n \n v2: Improve wording a bit (Laurent), fix typo in commit message (Sam).\n \n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Acked-by: Sean Paul \u003csean@poorly.run\u003e\n Acked-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Acked-by: Alex Deucher \u003calexdeucher@gmail.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@linux.intel.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190704145054.5701-1-daniel.vetter@ffwll.ch\n\ncommit 0d392cb9eb58099496a4d2981047df7ff7d22949\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Jul 19 02:41:00 2019 +0000\n\n drm/i915: Remove set but not used variable \u0027src_y\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/gpu/drm/i915/display/intel_sprite.c: In function \u0027g4x_sprite_check_scaling\u0027:\n drivers/gpu/drm/i915/display/intel_sprite.c:1494:13: warning:\n variable \u0027src_y\u0027 set but not used [-Wunused-but-set-variable]\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190719024100.64738-1-yuehaibing@huawei.com\n\ncommit 7d6b60dbc6a015dbdc444e4d39549600f7156690\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jul 16 13:49:30 2019 +0100\n\n drm/i915/execlists: Cancel breadcrumb on preempting the virtual engine\n \n As we unwind the requests for a preemption event, we return a virtual\n request back to its original virtual engine (so that it is available for\n execution on any of its siblings). In the process, this means that its\n breadcrumb should no longer be associated with the original physical\n engine, and so we are forced to decouple it. Previously, as the request\n could not complete without our awareness, we would move it to the next\n real engine without any danger. However, preempt-to-busy allowed for\n requests to continue on the HW and complete in the background as we\n unwound, which meant that we could end up retiring the request before\n fixing up the breadcrumb link.\n \n [51679.517943] INFO: trying to register non-static key.\n [51679.517956] the code is fine but needs lockdep annotation.\n [51679.517960] turning off the locking correctness validator.\n [51679.517966] CPU: 0 PID: 3270 Comm: kworker/u8:0 Tainted: G U 5.2.0+ #717\n [51679.517971] Hardware name: Intel Corporation NUC7i5BNK/NUC7i5BNB, BIOS BNKBL357.86A.0052.2017.0918.1346 09/18/2017\n [51679.518012] Workqueue: i915 retire_work_handler [i915]\n [51679.518017] Call Trace:\n [51679.518026] dump_stack+0x67/0x90\n [51679.518031] register_lock_class+0x52c/0x540\n [51679.518038] ? find_held_lock+0x2d/0x90\n [51679.518042] __lock_acquire+0x68/0x1800\n [51679.518047] ? find_held_lock+0x2d/0x90\n [51679.518073] ? __i915_sw_fence_complete+0xff/0x1c0 [i915]\n [51679.518079] lock_acquire+0x90/0x170\n [51679.518105] ? i915_request_cancel_breadcrumb+0x29/0x160 [i915]\n [51679.518112] _raw_spin_lock+0x27/0x40\n [51679.518138] ? i915_request_cancel_breadcrumb+0x29/0x160 [i915]\n [51679.518165] i915_request_cancel_breadcrumb+0x29/0x160 [i915]\n [51679.518199] i915_request_retire+0x43f/0x530 [i915]\n [51679.518232] retire_requests+0x4d/0x60 [i915]\n [51679.518263] i915_retire_requests+0xdf/0x1f0 [i915]\n [51679.518294] retire_work_handler+0x4c/0x60 [i915]\n [51679.518301] process_one_work+0x22c/0x5c0\n [51679.518307] worker_thread+0x37/0x390\n [51679.518311] ? process_one_work+0x5c0/0x5c0\n [51679.518316] kthread+0x116/0x130\n [51679.518320] ? kthread_create_on_node+0x40/0x40\n [51679.518325] ret_from_fork+0x24/0x30\n [51679.520177] ------------[ cut here ]------------\n [51679.520189] list_del corruption, ffff88883675e2f0-\u003enext is LIST_POISON1 (dead000000000100)\n \n Fixes: 22b7a426bbe1 (\"drm/i915/execlists: Preempt-to-busy\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716124931.5870-4-chris@chris-wilson.co.uk\n\ncommit 7a83645ac0cc2258de101a6a491d31499eb27cc5\nAuthor: Dale Zhao \u003cdale.zhao@amd.com\u003e\nDate: Wed Jul 10 17:36:53 2019 +0800\n\n drm/amd/display: handle active dongle port type is DP++ or DP case\n \n [Why]:\n Some active dongles have DP++ port and DP port at the same time. Current\n code doesn\u0027t cover DP++ case and processes as default DVI case, in which\n audio is disabled. Because of dual mode, DP case is also treat as DVI case\n for the other port.\n \n [How]:\n According DP 1.4 spec, add DP++ procedure similar with HDMI case. Also\n add None dongle type for DP case.\n \n Signed-off-by: Dale Zhao \u003cdale.zhao@amd.com\u003e\n Reviewed-by: Wenjing Liu \u003cwenjing.liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 21ffcc94d5b3dc024fedac700f1e7f9dacf4ab4f\nAuthor: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\nDate: Thu Jul 11 12:32:43 2019 -0400\n\n drm/amd/display: Copy GSL groups when committing a new context\n \n [Why]\n DC configures the GSL group for the pipe when pipe_split is enabled\n and we\u0027re switching flip types (buffered \u003c-\u003e immediate flip) on DCN2.\n \n In order to record what GSL group the pipe is using DC stores it in\n the pipe\u0027s stream_res. DM is not aware of this internal grouping, nor\n is DC resource.\n \n So when DM creates a dc_state context and passes it to DC the current\n GSL group is lost - DM never knew about it in the first place.\n \n After 3 immediate flips we run out of GSL groups and we\u0027re no longer\n able to correctly perform *any* flip for multi-pipe scenarios.\n \n [How]\n The gsl_group needs to be copied to the new context.\n \n DM has no insight into GSL grouping and could even potentially create\n a brand new context without referencing current hardware state. So this\n makes the most sense to have happen in DC.\n \n There are two places where DC can apply a new context:\n - dc_commit_state\n - dc_commit_updates_for_stream\n \n But what\u0027s shared between both of these is apply_ctx_for_surface.\n \n This logic only matters for DCN2, so it can be placed in\n dcn20_apply_ctx_for_surface. Before doing any locking (where the GSL\n group is setup) we can copy over the GSL groups before committing the\n new context.\n \n Signed-off-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Reviewed-by: Hersen Wu \u003chersen.wu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 291ac8fb0cd934104b1d154220695ec049592a25\nAuthor: Lewis Huang \u003cLewis.Huang@amd.com\u003e\nDate: Tue Jul 2 22:33:34 2019 -0500\n\n drm/amd/display: Add debug entry to destroy disconnected edp link\n \n Add a flag to dc_debug_options to determine if a disconnected edp link\n should be destroyed.\n \n Signed-off-by: Lewis Huang \u003cLewis.Huang@amd.com\u003e\n Reviewed-by: Yongqiang Sun \u003cyongqiang.sun@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 46825fcfbe162609393976e18f106dafddca6f67\nAuthor: Tony Cheng \u003ctony.cheng@amd.com\u003e\nDate: Sun Jun 23 12:07:02 2019 -0500\n\n drm/amd/display: avoid power gate domains that doesn\u0027t exist\n \n Signed-off-by: Tony Cheng \u003ctony.cheng@amd.com\u003e\n Reviewed-by: Yongqiang Sun \u003cyongqiang.sun@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9295f1113c74baab85a18628c8b2ef9db04dac14\nAuthor: Fatemeh Darbehani \u003cfatemeh.darbehani@amd.com\u003e\nDate: Wed Jul 10 10:34:31 2019 -0400\n\n drm/amd/display: Add SMU version field to clk_mgr_internal\n \n For some platforms, we need to know SMU version for driver/SMU\n compatibility. This change adds that field.\n \n Signed-off-by: Fatemeh Darbehani \u003cfatemeh.darbehani@amd.com\u003e\n Reviewed-by: Yongqiang Sun \u003cyongqiang.sun@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 057fc695e934a77bae0c6c7f3be01251774b61cf\nAuthor: Jun Lei \u003cJun.Lei@amd.com\u003e\nDate: Mon Jul 8 15:15:42 2019 -0400\n\n drm/amd/display: support \"dummy pstate\"\n \n [why]\n Existing support in DC for pstate only accounts for a single latency. This is sufficient when the\n variance of latency is small, or that pstate support isn\u0027t necessary for correct ASIC functionality.\n \n Newer ASICs violate both existing assumptions. PState support is mandatory of correct ASIC\n functionality, but not all latencies have to be supported. Existing code supports a \"full p state\" which\n allows memory clock to change, but is hard for DCN to support (as it requires very large buffers).\n New code will now fall back to a \"dummy p state\" support when \"full p state\" cannot be support.\n This easy p state support should always be allowed.\n \n [how]\n Define a new latency in socBB. Add fallback logic to support it. Note DML is also updated to ensure\n that fallback will always work.\n \n Signed-off-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Reviewed-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 0488a56465877a48596a874bc0f4f31d6fbf334e\nAuthor: Eric Yang \u003cEric.Yang2@amd.com\u003e\nDate: Sat Jun 29 16:02:37 2019 -0400\n\n drm/amd/display: fix mpcc assert condition\n \n [Why]\n In DCN2x asic, the MPCC status register definition changed, and\n our logic for assert is incorrect. disabled is valid state,\n where we should see idle and not busy, where as in not\n disabled state, we should see not idle.\n \n [How]\n Change assert condition to be more sensible.\n \n Signed-off-by: Eric Yang \u003cEric.Yang2@amd.com\u003e\n Reviewed-by: Yongqiang Sun \u003cyongqiang.sun@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit dcbb45b6eeedfbc3a087f4fa375c706adabd1ce0\nAuthor: Eric Yang \u003cEric.Yang2@amd.com\u003e\nDate: Sat Jun 29 14:38:04 2019 -0400\n\n drm/amd/display: do not read link setting if edp not connected\n \n [Why]\n Previously assume eDP sink present if connector present. Do not\n need to enforce this restriction. Fix issue where driver attempt\n to read link setting even though no edp connected.\n \n {How]\n Only read link setting after reading connection status.\n \n Signed-off-by: Eric Yang \u003cEric.Yang2@amd.com\u003e\n Reviewed-by: Yongqiang Sun \u003cyongqiang.sun@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1b1ff454bd66f9b3117aafe3c39df0b3efecd629\nAuthor: Tai Man \u003ctaiman.wong@amd.com\u003e\nDate: Fri Jun 28 11:40:38 2019 -0400\n\n drm/amd/display: Increase size of audios array\n \n [Why]\n The audios array defined in \"struct resource_pool\" is only 6 (MAX_PIPES)\n but the max number of audio devices (num_audio) is 7. In some projects,\n it will run out of audios array.\n \n [How]\n Incraese the audios array size to 7.\n \n Signed-off-by: Tai Man \u003ctaiman.wong@amd.com\u003e\n Reviewed-by: Joshua Aberback \u003cJoshua.Aberback@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f53fc02b9ac38ad096842bea7880e6c066e01814\nAuthor: Aric Cyr \u003caric.cyr@amd.com\u003e\nDate: Sun Jul 7 21:25:35 2019 -0400\n\n drm/amd/display: 3.2.42\n \n Signed-off-by: Aric Cyr \u003caric.cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ba94b08d0ba008c9f162e1df3ed0dc52194d8ccb\nAuthor: Zhan Liu \u003czhan.liu@amd.com\u003e\nDate: Tue Jul 2 15:17:07 2019 -0400\n\n drm/amd/display: drop ASSERT() if eDP panel is not connected\n \n [Why]\n For boards that support eDP but do not have a physical eDP\n display connected an ASSERT will be thrown. This is not a\n critical failure and shouldn\u0027t be treated as such.\n \n [How]\n Drop the assertion.\n \n Signed-off-by: Zhan Liu \u003czhan.liu@amd.com\u003e\n Reviewed-by: Nicholas Kazlauskas \u003cNicholas.Kazlauskas@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 3f52aa9f9f6f611d19035961fba0f8343bcb3e6c\nAuthor: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\nDate: Wed Jul 3 10:02:39 2019 -0400\n\n drm/amd/display: Set enabled to false at start of audio disable\n \n [Why]\n In an effort to stop redundant calls to dce110_disable_audio_stream\n the audio-\u003eenabled flag was added to the audio resource struct. While\n this state probably shouldn\u0027t have been tracked on the audio struct\n itself it still works fine for some sequences.\n \n However, it does not work for cases where we\u0027re freeing the audio\n resource (such as hotplugs) or when dynamic audio is enabled.\n \n In these cases the pipe_ctx-\u003estream_res.audio \u003d NULL before we can\n set audio-\u003eenabled \u003d false. The next time we acquire the audio resource\n such as on hotplug the audio will not be enabled for the stream since\n DC thinks it\u0027s still enabled.\n \n Audio state tracking should cover this sequence.\n \n [How]\n Set audio-\u003eenabled \u003d false at the start as long as we have\n pipe_ctx-\u003estream_res.audio.\n \n Signed-off-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Reviewed-by: Zhan Liu \u003cZhan.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f591344e89dca887afeb5bbe21fa199687fd4a5f\nAuthor: Julian Parkin \u003cjulian.parkin@amd.com\u003e\nDate: Wed Jul 3 13:59:26 2019 -0400\n\n drm/amd/display: Clean up dynamic metadata logic\n \n [Why]\n Code to enable DCN20 dynamic metadata feature is duplicated in two places\n and was added to DCE110 enable stream.\n \n [How]\n Create DCN20 specific enable stream function for clarity, and add a hardware\n sequencer function to program dynamic metadata to avoid the duplicate\n code.\n \n Signed-off-by: Julian Parkin \u003cjulian.parkin@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ce08aad3896f342d6c99e8e2dc1f71a0926efafc\nAuthor: Alvin Lee \u003calvin.lee2@amd.com\u003e\nDate: Thu Jul 4 15:17:42 2019 -0400\n\n drm/amd/display: Only enable audio if speaker allocation exists\n \n [Why]\n \n In dm_helpers_parse_edid_caps, there is a corner case where no speakers\n can be allocated even though the audio mode count is greater than 0.\n Enabling audio when no speaker allocations exists can cause issues in\n the video stream.\n \n [How]\n \n Add a check to not enable audio unless one or more speaker allocations\n exist (since doing this can cause issues in the video stream).\n \n Signed-off-by: Alvin Lee \u003calvin.lee2@amd.com\u003e\n Reviewed-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8897810adfa768b0c127c1e204bedc4449e3f945\nAuthor: Julian Parkin \u003cjulian.parkin@amd.com\u003e\nDate: Tue Jun 25 14:55:53 2019 -0400\n\n drm/amd/display: Fix dc_create failure handling and 666 color depths\n \n [Why]\n It is possible (but very unlikely) that constructing dc fails\n before current_state is created.\n \n We support 666 color depth in some scenarios, but this\n isn\u0027t handled in get_norm_pix_clk. It uses exactly the\n same pixel clock as the 888 case.\n \n [How]\n Check for non null current_state before destructing.\n \n Add case for 666 color depth to get_norm_pix_clk to\n avoid assertion.\n \n Signed-off-by: Julian Parkin \u003cjulian.parkin@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6da16270ee026a0397a67f5e2b4661b6dfa45afb\nAuthor: Jun Lei \u003cJun.Lei@amd.com\u003e\nDate: Wed Jul 3 16:52:38 2019 -0400\n\n drm/amd/display: populate last calculated bb state with max clocks\n \n [why]\n update_bounding_box calculates intermediate bb states based on clock relationship\n however, the last state doesn\u0027t need to maintain a minimum relationship, but should\n actually contain maximum of every clock. otherwise maximum clocks are not usable\n \n [how]\n once the calculated bb is built, override the last state with max values\n \n Signed-off-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Reviewed-by: Aric Cyr \u003cAric.Cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 12e2b2d4c65f6164830e25fcd9624519a424b182\nAuthor: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\nDate: Wed Jul 3 16:20:42 2019 -0400\n\n drm/amd/display: add dcc programming for dual plane\n \n Add dual plane dcc programming support for\n surfaces.\n \n Removes unions from plane size and dcc params as they\n serve no practical purpose only making our code\n more convoluted. This results in easy dual plane\n dcc and surface size programming.\n \n Temporary diags_dm code is used to handle the interface\n change without breaking functionality as a diags change\n needs to be applied after this one.\n \n Signed-off-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f7938bc0f08b3d748d20dc2c88914f03331dce94\nAuthor: Reza Amini \u003cReza.Amini@amd.com\u003e\nDate: Mon Jun 10 16:45:50 2019 -0400\n\n drm/amd/display: Implement DAL3 GPU Integer Scaling\n \n [WHY]\n Users want to not have filtering when scaling by integer\n multiples to native timing.\n \n [HOW]\n If timing is a multiple integer of view, we set number of taps\n to 1 (effectivly closest neighbour).\n \n Signed-off-by: Reza Amini \u003cReza.Amini@amd.com\u003e\n Reviewed-by: Aric Cyr \u003cAric.Cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Acked-by: Tony Cheng \u003cTony.Cheng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f1cdc98fd9d9bb0eb4b6599825b52ce25757eb9f\nAuthor: Dingchen Zhang \u003cdingchen.zhang@amd.com\u003e\nDate: Fri Jun 28 17:23:24 2019 -0400\n\n drm/amd/display: add pipe CRC sources without disabling dithering.\n \n [Why]\n need to verify the impact of spatial dithering on 8bpc bypass mode.\n \n [How]\n added CRC sources and configure dihter option from dc stream.\n \n Signed-off-by: Dingchen Zhang \u003cdingchen.zhang@amd.com\u003e\n Reviewed-by: Hanghong Ma \u003cHanghong.Ma@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e9bcc1e03048b3872df46cc6ee28d5c43850e379\nAuthor: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\nDate: Tue Jul 2 14:51:01 2019 -0400\n\n drm/amd/display: use min disp and dpp clk debug option for dcn2\n \n This allows to set a minimum display and dpp clock on dcn2+ HW\n by adjusting clocks used for dml calculations.\n \n Signed-off-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Reviewed-by: Tony Cheng \u003cTony.Cheng@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 39bca3599aab8edbc4a6c6d10fc76b1a932e5dfa\nAuthor: Charlene Liu \u003ccharlene.liu@amd.com\u003e\nDate: Tue Jul 2 14:04:35 2019 -0400\n\n drm/amd/display: add a option to force the clock at every mode change.\n \n [Description]\n This is for HW negative stress testing use.\n force reset the dispclk and dppclk even the same clock already set in HW.\n \n Signed-off-by: Charlene Liu \u003ccharlene.liu@amd.com\u003e\n Reviewed-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 54eef8a411699779fdbb1a2cd333f1b914bd5699\nAuthor: Murton Liu \u003cmurton.liu@amd.com\u003e\nDate: Tue Jul 2 11:32:19 2019 -0400\n\n drm/amd/display: Change offset_to_id to reflect what id_to_offset returns\n \n id_to_offset does not point to the same reg offset that offset_to_id checks for,\n causing unintended asserts\n \n Signed-off-by: Murton Liu \u003cmurton.liu@amd.com\u003e\n Reviewed-by: Aric Cyr \u003cAric.Cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 925f566cb7aedbcf26005035cf894ec824e8ca2f\nAuthor: Charlene Liu \u003ccharlene.liu@amd.com\u003e\nDate: Thu Jun 27 18:16:21 2019 -0400\n\n drm/amd/display: add set and get clock for testing purposes\n \n add dc_set_clock\n add dc_get_clock\n \n this is for testing and diagnostics to get/set DPPCLK and DISPCLK.\n \n Signed-off-by: Charlene Liu \u003ccharlene.liu@amd.com\u003e\n Reviewed-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 30db43b62c16cb0a0ff3eb89d9b25bf455859ef8\nAuthor: Derek Lai \u003cDerek.Lai@amd.com\u003e\nDate: Tue Jul 2 17:50:41 2019 +0800\n\n drm/amd/display: allocate 4 ddc engines for RV2\n \n [Why]\n Driver will create 0, 1, and 2 ddc engines for RV2,\n but some platforms used 0, 1, and 3.\n \n [How]\n Still allocate 4 ddc engines for RV2.\n \n Signed-off-by: Derek Lai \u003cDerek.Lai@amd.com\u003e\n Reviewed-by: Aric Cyr \u003cAric.Cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c2cd9d04ecf0f26fb472ffbd8274a55c05df0ffb\nAuthor: Murton Liu \u003cmurton.liu@amd.com\u003e\nDate: Tue Jun 25 11:15:09 2019 -0400\n\n drm/amd/display: Hook up calls to do stereo mux and dig programming to stereo control interface\n \n [Why]\n Implementation of stereo mux register is complete, but unused. Need to\n call functions to write relevant configs.\n \n [How]\n Add function to write stereo config for enable/disable case and call in\n stereo control interface.\n \n Signed-off-by: Murton Liu \u003cmurton.liu@amd.com\u003e\n Reviewed-by: Aric Cyr \u003cAric.Cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 692626fc4dfc214cecd6125b47884920990feb85\nAuthor: Aric Cyr \u003caric.cyr@amd.com\u003e\nDate: Mon Jul 1 23:15:42 2019 -0400\n\n drm/amd/display: 3.2.41\n \n Signed-off-by: Aric Cyr \u003caric.cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 473e0ecbb4f62e0a971b47b27ec778371653f5ac\nAuthor: Ilya Bakoulin \u003cIlya.Bakoulin@amd.com\u003e\nDate: Wed Jun 26 14:52:46 2019 -0400\n\n drm/amd/display: Fixes for some MPO cases\n \n [Why]\n Alpha could be improperly applied (only affecting half the\n frame) for some source pixel formats.\n \n [How]\n Change how alpha is enabled in MPC/DPP LB and change the\n bottom plane blend mode in MPC.\n \n Signed-off-by: Ilya Bakoulin \u003cIlya.Bakoulin@amd.com\u003e\n Reviewed-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8bee5c5c7e9ed478272bcd1288a425b5a101b376\nAuthor: Ilya Bakoulin \u003cIlya.Bakoulin@amd.com\u003e\nDate: Wed Jun 19 17:33:02 2019 -0400\n\n drm/amd/display: Cache the use_pitch_c conditional\n \n For clarity, save the use_pitch_c logic to a bool\n \n Signed-off-by: Ilya Bakoulin \u003cIlya.Bakoulin@amd.com\u003e\n Reviewed-by: Eric Bernstein \u003ceric.bernstein@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 69866d132355c83f115070c79537c72f9e2d6965\nAuthor: Nevenko Stupar \u003cNevenko.Stupar@amd.com\u003e\nDate: Fri Jun 28 12:12:13 2019 -0400\n\n drm/amd/display: Add DIG_CLOCK_PATTERN register\n \n Add this register for future use\n \n Signed-off-by: Nevenko Stupar \u003cNevenko.Stupar@amd.com\u003e\n Reviewed-by: Vitaly Prosyak \u003cvitaly.prosyak@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 290129c256179d5bbf822016e1729b8e9ed4875b\nAuthor: Qingqing Zhuo \u003cqingqing.zhuo@amd.com\u003e\nDate: Thu Jun 27 19:17:23 2019 -0400\n\n drm/amd/display: Add CM_BYPASS via debug option\n \n [Why]\n bypass CM block and MPC ogam for debug or triage use.\n \n [How]\n create a new flag cm_bypass_mode, which will set both CM_CONTROL\n and MPCC_OGAM_MODE to bypass when set to 1.\n \n Signed-off-by: Qingqing Zhuo \u003cqingqing.zhuo@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a2080098692173cf27282f686e341b184f9d68e3\nAuthor: Vitaly Prosyak \u003cvitaly.prosyak@amd.com\u003e\nDate: Fri Jun 21 10:13:16 2019 -0500\n\n drm/amd/display: Add MPC 3DLUT resource management\n \n [Why \u0026 How]\n Number of 3DLUT\u0027s in MPC are not equal to number of pipes.\n Resource management is required.\n Activate on FPGA entire tm solution which includes\n the following :hdr multiplier, shaper, 3dlut.\n \n Signed-off-by: Vitaly Prosyak \u003cvitaly.prosyak@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Gary Kattan \u003cGary.Kattan@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 70f1476a7eedf85078f5b5a3b2647591f6df4d17\nAuthor: Joshua Aberback \u003cjoshua.aberback@amd.com\u003e\nDate: Fri May 17 14:43:10 2019 -0400\n\n drm/amd/display: Add debug option to disable timing sync\n \n [Why]\n We want a debug option to disable timing sync for testing.\n \n [How]\n New dc debug option that must be false to call program_timing_sync\n \n Signed-off-by: Joshua Aberback \u003cjoshua.aberback@amd.com\u003e\n Reviewed-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d68a74541735e030dea56f72746cd26d19986f41\nAuthor: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\nDate: Tue Jun 25 17:19:25 2019 -0400\n\n drm/amd/display: Clear FEC_READY shadow register if DPCD write fails\n \n [why]\n As a fail-safe, in case \u0027set FEC_READY\u0027 DPCD write fails, a HW shadow\n register should be cleared and the internal FEC stat should be set to\n \u0027not ready\u0027. This is to make sure HW settings will be consistent with\n FEC_READY state on the RX.\n \n Signed-off-by: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\n Reviewed-by: Joshua Aberback \u003cJoshua.Aberback@amd.com\u003e\n Acked-by: Chris Park \u003cChris.Park@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e752058b8671c6c87e484cff144c5c6309a37253\nAuthor: Krunoslav Kovac \u003cKrunoslav.Kovac@amd.com\u003e\nDate: Tue Jun 18 17:38:43 2019 -0400\n\n drm/amd/display: Optimize gamma calculations\n \n [Why\u0026How]\n \n 1. Stack usage is pretty high as fixed31_32 struct is 8 bytes and we\n have functions with \u003e30 vars on the stack.\n \n 2. Optimize gamma calculation by reducing number of calls to\n dc_fixpt_pow Our X points are divided into 32 regions wth 16 pts each.\n Each region is 2x the previous, meaning x[i] \u003d 2*x[i-16] for i\u003e\u003d16.\n Using (2x)^gamma \u003d 2^gamma * x^gamma, we can recursively compute powers\n of gamma, we just need first 16 pts to start it up. dc_fixpt_pow() is\n expensive, it computes x^y by doing exp(y*logx) Exp is done by Taylor\n series approximation, and log by Newton-like approximation that also\n uses exp internally. In short, it\u0027s significantly heavier than\n run-of-the-mill addition/subtraction/multiply.\n \n Signed-off-by: Krunoslav Kovac \u003cKrunoslav.Kovac@amd.com\u003e\n Reviewed-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Acked-by: Aric Cyr \u003cAric.Cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c43f89f81cc0c54c7c460f1d6365445939069c83\nAuthor: Eric Yang \u003cEric.Yang2@amd.com\u003e\nDate: Mon Jun 24 18:18:58 2019 -0400\n\n drm/amd/display: put back front end initialization sequence\n \n [Why]\n Seamless boot optimization removed proper front end power off sequence.\n In driver disable enable case, this causes driver to power gate hubp\n and dpp while there is still memory fetching going on, this can cause\n invalid memory requests to be generated which will hang data fabric.\n \n [How]\n Put back proper front end power off sequence\n \n Signed-off-by: Eric Yang \u003cEric.Yang2@amd.com\u003e\n Reviewed-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Acked-by: Tony Cheng \u003cTony.Cheng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 008a4016c5cf922d33456916ec3fad9ac4c98962\nAuthor: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\nDate: Mon Jun 24 15:44:42 2019 -0400\n\n drm/amd/display: Set FEC_READY always before link training\n \n [why]\n Right now we FEC_READY is set only before the final link training,\n i.e. at mode set time. This means FEC_READY won\u0027t be set when doing\n link training as a response to HPD. It also fails UCD400 FEC test in\n DP compliance.\n \n [how]\n Move FEC_READY setup to link training.\n \n Signed-off-by: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\n Reviewed-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Acked-by: Abdoulaye Berthe \u003cAbdoulaye.Berthe@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a280a71f5e4a7c4ed44ac2cacc9718bccbb2cf65\nAuthor: Eric Bernstein \u003ceric.bernstein@amd.com\u003e\nDate: Mon Jun 24 14:11:55 2019 -0400\n\n drm/amd/display: Use helper for determining HDMI signal\n \n Use helper to determine if HDMI signal when processing avmute.\n \n Signed-off-by: Eric Bernstein \u003ceric.bernstein@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d40605b6d088b20827e442903022c65f0f165c84\nAuthor: Murton Liu \u003cmurton.liu@amd.com\u003e\nDate: Mon Jun 24 11:28:06 2019 -0400\n\n drm/amd/display: Implement generic MUX registers (v2)\n \n [Why]\n Logic \u0026 structures for generic regs does not exist in DC currently.\n \n [How]\n Implement register masks/shifts and relevant functions for generic mux,\n similar to existing HPD and DDC objects.\n \n V2: fix includes for kalloc/free (Alex)\n \n Signed-off-by: Murton Liu \u003cmurton.liu@amd.com\u003e\n Reviewed-by: Aric Cyr \u003cAric.Cyr@amd.com\u003e\n Acked-by: Joshua Aberback \u003cJoshua.Aberback@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4c6a9618076c2690a361e8bc32549720f2f42305\nAuthor: Alvin Lee \u003calvin.lee2@amd.com\u003e\nDate: Mon Jun 24 09:49:44 2019 -0400\n\n drm/amd/display: Wait for flip to complete\n \n [why]\n In pipe split issue occurs when we program immediate flip while vsync flip is pending\n \n [how]\n Don\u0027t program immediate flip until flip is no longer pending\n \n Signed-off-by: Alvin Lee \u003calvin.lee2@amd.com\u003e\n Reviewed-by: Jaehyun Chung \u003cJaehyun.Chung@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 351960b09850deaa6e74f4e5f39d53deb1114833\nAuthor: Aric Cyr \u003caric.cyr@amd.com\u003e\nDate: Sun Jun 23 17:27:43 2019 -0400\n\n drm/amd/display: 3.2.40\n \n Signed-off-by: Aric Cyr \u003caric.cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7e6472966e8341bf72f558babfdafac8aeff3257\nAuthor: Fatemeh Darbehani \u003cfatemeh.darbehani@amd.com\u003e\nDate: Fri Jun 21 17:44:50 2019 -0400\n\n drm/amd/display: Change min_h_sync_width from 8 to 4\n \n [Why]\n Some display\u0027s hsync width is lower than the minimum dcn20 is set\n to support right now. This will cause optc1_validate_timing to fail which\n eventually will result in wrong set mode. This was set to 8 as per\n HW team\u0027s request for no valid reason.\n \n [How]\n Changing min_h_sync_width to 4 will let us validate timing for\n preffered mode and light up the headset. This change was made\n to Vega 10 before for a similar issue.\n \n Signed-off-by: Fatemeh Darbehani \u003cfatemeh.darbehani@amd.com\u003e\n Reviewed-by: Joshua Aberback \u003cJoshua.Aberback@amd.com\u003e\n Acked-by: Aric Cyr \u003cAric.Cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1b3c61036e9fb643d35b09aa8172420c3da21b2d\nAuthor: Alvin Lee \u003calvin.lee2@amd.com\u003e\nDate: Fri Jun 21 17:58:41 2019 -0400\n\n drm/amd/display: Remove second initialization of pp_smu\n \n [why]\n We initialize pp_smu twice\n \n [how]\n Remove second initialization of pp_smu\n \n Signed-off-by: Alvin Lee \u003calvin.lee2@amd.com\u003e\n Reviewed-by: Jaehyun Chung \u003cJaehyun.Chung@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 18b401874aee10c80b5745c9b93280dae5a59809\nAuthor: Su Sung Chung \u003cSu.Chung@amd.com\u003e\nDate: Fri Jun 21 16:14:36 2019 -0400\n\n drm/amd/display: fix not calling ppsmu to trigger PME\n \n [why]\n dcn20_clk_mgr_construct was not initializing pp_smu, and PME call gets\n filtered out by the null check\n \n [how]\n initialize pp_smu dcn20_clk_mgr_construct\n \n Signed-off-by: Su Sung Chung \u003cSu.Chung@amd.com\u003e\n Reviewed-by: Eric Yang \u003ceric.yang2@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5feb9f07a2035394ed808e266d294a9df3a8d544\nAuthor: Tai Man \u003ctaiman.wong@amd.com\u003e\nDate: Fri Jun 7 17:32:27 2019 -0400\n\n drm/amd/display: use encoder\u0027s engine id to find matched free audio device\n \n [Why]\n On some platforms, the encoder id 3 is not populated. So the encoders\n are not stored in right order as index (id: 0, 1, 2, 4, 5) at pool. This\n would cause encoders id 4 \u0026 id 5 to fail when finding corresponding\n audio device, defaulting to the first available audio device. As result,\n we cannot stream audio into two DP ports with encoders id 4 \u0026 id 5.\n \n [How]\n It need to create enough audio device objects (0 - 5) to perform matching.\n Then use encoder engine id to find matched audio device.\n \n Signed-off-by: Tai Man \u003ctaiman.wong@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 252f3d950d2e7c45869096aa1efc883464149b31\nAuthor: Harry Wentland \u003charry.wentland@amd.com\u003e\nDate: Fri May 31 14:17:43 2019 -0400\n\n drm/amd/display: Remove unnecessary NULL check in set_preferred_link_settings\n \n [Why]\n link_stream is never NULL here as we\u0027ve dereferenced it a couple lines before\n and have done so for a couple months now.\n \n [How]\n - Drop the NULL check.\n - Initialize where we know link_stream is non-NULL\n \n Signed-off-by: Harry Wentland \u003charry.wentland@amd.com\u003e\n Reviewed-by: Nicholas Kazlauskas \u003cNicholas.Kazlauskas@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 03f3e40c8b5c227e1435bee0afa2cbc657c15cf4\nAuthor: Charlene Liu \u003ccharlene.liu@amd.com\u003e\nDate: Wed Jun 19 21:35:35 2019 -0400\n\n drm/amd/display: Do not fill Null packet in the blank period\n \n [Description]\n Do not fill Null packet in the blank period for new packet gen\n This is based on HW IP team recommended default setting change.\n \n Signed-off-by: Charlene Liu \u003ccharlene.liu@amd.com\u003e\n Reviewed-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5a42dc2be09a550777619a29187867d0423c353a\nAuthor: Zi Yu Liao \u003cziyu.liao@amd.com\u003e\nDate: Thu Jun 20 10:55:26 2019 -0400\n\n drm/amd/display: fix DMCU hang when going into Modern Standby\n \n [why]\n When the system is going into suspend, set_backlight gets called\n after the eDP got blanked. Since smooth brightness is enabled,\n the driver will make a call into the DMCU to ramp the brightness.\n The DMCU would try to enable ABM to do so. But since the display is\n blanked, this ends up causing ABM1_ACE_DBUF_REG_UPDATE_PENDING to\n get stuck at 1, which results in a dead lock in the DMCU firmware.\n \n [how]\n Disable brightness ramping when the eDP display is blanked.\n \n Signed-off-by: Zi Yu Liao \u003cziyu.liao@amd.com\u003e\n Reviewed-by: Eric Yang \u003ceric.yang2@amd.com\u003e\n Acked-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1d0610bc26d8e2fe82de77d8be3e1ed0958feddc\nAuthor: Alvin Lee \u003calvin.lee2@amd.com\u003e\nDate: Thu Jun 20 13:03:25 2019 -0400\n\n drm/amd/display: Disable Audio on reinitialize hardware\n \n [Why]\n When we recover from hang, we do not want to skip the audio enable call.\n \n [How]\n Disable audio in dc_reinitialize_hardware\n \n Signed-off-by: Alvin Lee \u003calvin.lee2@amd.com\u003e\n Reviewed-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 75c35000235f3662f2810e9a59b0c8eed045432e\nAuthor: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\nDate: Wed Jun 19 14:30:52 2019 -0400\n\n drm/amd/display: Power-gate all DSCs at driver init time\n \n [why]\n DSC should be powered-on only on as-needed basis, i.e. if the mode\n requires it\n \n [how]\n Loop over all the DSCs at driver init time and power-gate each\n \n Signed-off-by: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\n Reviewed-by: Nevenko Stupar \u003cNevenko.Stupar@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8fca3d94157bfd401801b18f0fa8f1d0c693f2d7\nAuthor: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\nDate: Tue Jun 18 15:57:03 2019 -0400\n\n drm/amd/display: Remove dsc disable_ich flag programming.\n \n Current default is sufficient for a flag that does not change.\n \n Signed-off-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 701c75cea29b725a5447c7a7d3b8b4c8472ab79a\nAuthor: Derek Lai \u003cDerek.Lai@amd.com\u003e\nDate: Tue Jun 18 14:55:57 2019 +0800\n\n drm/amd/display: Read max down spread\n \n [Why]\n When launch D10.2, driver will write DPCD 0x107 with 0x00\n \n [How]\n Read MAX_DOWNSPREAD (0x0003h) then keep in current\n link settings\n \n Signed-off-by: Derek Lai \u003cDerek.Lai@amd.com\u003e\n Reviewed-by: Tony Cheng \u003cTony.Cheng@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit eced4bceed22114697a17295a768cebea1dbc2b0\nAuthor: Ilya Bakoulin \u003cIlya.Bakoulin@amd.com\u003e\nDate: Wed Jun 12 12:40:42 2019 -0400\n\n drm/amd/display: Fix some HUBP programming issues\n \n [Why]\n A hubp pointer was being passed to DCN1 functions, which\n expect the enclosing structure (for the purpose of container_of macros)\n to be dcn10_hubp, but the actual type was dcn20_hubp.\n \n [How]\n Copy existing DCN1 functions and alter them slightly for use with\n dcn20_hubp.\n \n Signed-off-by: Ilya Bakoulin \u003cIlya.Bakoulin@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 24f1d1cee2bc9cb591145e21823d7865208a7991\nAuthor: Ilya Bakoulin \u003cIlya.Bakoulin@amd.com\u003e\nDate: Wed May 29 18:52:17 2019 -0400\n\n drm/amd/display: Check for valid stream_encode\n \n Before accessing it\u0027s vtable, check that stream_encoder is non-null.\n \n Signed-off-by: Ilya Bakoulin \u003cIlya.Bakoulin@amd.com\u003e\n Reviewed-by: Eric Bernstein \u003cEric.Bernstein@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4fb6ec466dd50fdd94a1288b27b692da67b6c23d\nAuthor: Ilya Bakoulin \u003cIlya.Bakoulin@amd.com\u003e\nDate: Wed May 29 18:52:17 2019 -0400\n\n drm/amd/display: Expose enc2_set_dynamic_metadata\n \n [Why]\n Need to implement register programming for HDR dynamic\n metadata transmission and tests.\n \n Signed-off-by: Ilya Bakoulin \u003cIlya.Bakoulin@amd.com\u003e\n Reviewed-by: Eric Bernstein \u003cEric.Bernstein@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6d58f99444e29d68ff6d5d52ff42c77692fe306f\nAuthor: Aric Cyr \u003caric.cyr@amd.com\u003e\nDate: Mon Jun 17 10:58:57 2019 -0400\n\n drm/amd/display: 3.2.39\n \n Signed-off-by: Aric Cyr \u003caric.cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d35d77a6a512f38a89aee9328c67d0c913c7fbea\nAuthor: SivapiriyanKumarasamy \u003csivapiriyan.kumarasamy@amd.com\u003e\nDate: Fri Jun 14 15:04:00 2019 -0400\n\n drm/amd/display: Wait for backlight programming completion in set backlight level\n \n [WHY]\n Currently we don\u0027t wait for blacklight programming completion in DMCU\n when setting backlight level. Some sequences such as PSR static screen\n event trigger reprogramming requires it to be complete.\n \n [How]\n Add generic wait for dmcu command completion in set backlight level.\n \n Signed-off-by: SivapiriyanKumarasamy \u003csivapiriyan.kumarasamy@amd.com\u003e\n Reviewed-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8fb843d179a6fff1bfe037601b06980ba9a56167\nAuthor: Dingchen Zhang \u003cdingchen.zhang@amd.com\u003e\nDate: Wed May 29 18:52:52 2019 -0400\n\n drm/amd/display: add functionality to get pipe CRC source.\n \n [Why]\n We need to check the pipe crc source through debugfs for bypass mode test.\n \n [How]\n add implementation of amdgpu_dm_crtc_get_crc_sources and hook into drm_crtc\n callback get_crc_sources.\n \n Signed-off-by: Dingchen Zhang \u003cdingchen.zhang@amd.com\u003e\n Reviewed-by: Nicholas Kazlauskas \u003cNicholas.Kazlauskas@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 14b2584636c66efbb9a8dd1c702d5da73ecf620d\nAuthor: Dingchen Zhang \u003cdingchen.zhang@amd.com\u003e\nDate: Wed May 15 17:15:05 2019 -0400\n\n drm/amd/display: add functionality to grab DPRX CRC entries.\n \n [Why]\n We need to compare DPRX CRCs with framebuffer CRCs for digital bypass mode.\n \n [How]\n Hook into DRM to grab DP receiver CRCs through drm_dp_start_crc.\n \n Signed-off-by: Dingchen Zhang \u003cdingchen.zhang@amd.com\u003e\n Reviewed-by: Harry Wentland \u003cHarry.Wentland@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 24253476977af5456af962bbaaf53e9f1dae5cc1\nAuthor: Julian Parkin \u003cjulian.parkin@amd.com\u003e\nDate: Thu Jun 13 12:49:37 2019 -0400\n\n drm/amd/display: Poll for GPUVM context ready (v2)\n \n [Why]\n Hardware docs state that we must wait until the GPUVM context is ready\n after programming it.\n \n [How]\n Poll until the valid bit of PAGE_TABLE_BASE_ADDR_LO32 is set to 1 after\n programming it.\n \n v2: fix include for udelay (Alex)\n \n Signed-off-by: Julian Parkin \u003cjulian.parkin@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 88eac241a1fc500ce5274a09ddc4bd5fc2b5adb6\nAuthor: Anthony Koo \u003canthony.koo@amd.com\u003e\nDate: Wed Jun 12 16:30:15 2019 -0400\n\n drm/amd/display: add monitor patch to add T7 delay\n \n [Why]\n Specifically to one panel,\n TCON is able to accept active video signal quickly, but\n the Source Driver requires 2-3 frames of extra time.\n \n It is a Panel issue since TCON needs to take care of\n all Sink requirements including Source Driver. But in\n this case it does not.\n \n Customer is asking to add fixed T7 delay as panel\n workaround.\n \n [How]\n Add monitor specific patch to add T7 delay\n \n Signed-off-by: Anthony Koo \u003canthony.koo@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 606b355170b56549890e8202a2b62f97d28b395e\nAuthor: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\nDate: Wed Jun 5 16:35:08 2019 -0400\n\n drm/amd/display: add hdmi2.1 dsc pps packet programming\n \n This change adds EMP packet programming for enabling dsc with\n hdmi. The packets are structured according to VESA HDMI 2.1x\n r2 spec, section 10.10.2.2.\n \n Signed-off-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c1f2e0154065963f95ce22e03ff05a420cb3aac7\nAuthor: Su Sung Chung \u003cSu.Chung@amd.com\u003e\nDate: Wed May 22 14:28:52 2019 -0400\n\n drm/amd/display: refactor dump_clk_registers\n \n [why]\n for 2 purposes:\n \n 1. get raw register value dumped on the log, which will make it easier\n to talk to other team who only knows about the register\n \n 2. enable other HW to be able to use the same interface as raven to log\n clock register data\n \n Signed-off-by: Su Sung Chung \u003cSu.Chung@amd.com\u003e\n Reviewed-by: Tony Cheng \u003cTony.Cheng@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 29344d152d0a7a54945606d9065c5de79161f100\nAuthor: Wenjing Liu \u003cWenjing.Liu@amd.com\u003e\nDate: Tue Jun 11 18:18:36 2019 -0400\n\n drm/amd/display: wait for the whole frame after global unlock\n \n [why]\n The current code will not wait for the entire frame\n after global unlock.\n This causes dsc dynamic target bpp update corruption when\n there is a surface update immediately happens after this.\n \n [how]\n Wait for the entire whole frame after unlock before continuing\n the rest of stream and surface update.\n \n Signed-off-by: Wenjing Liu \u003cWenjing.Liu@amd.com\u003e\n Reviewed-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 40fd9090aec688be730d54a00fd2fdcb37e16701\nAuthor: Nevenko Stupar \u003cNevenko.Stupar@amd.com\u003e\nDate: Tue Jun 11 17:35:16 2019 -0400\n\n drm/amd/display:Use Pixel clock in 100Hz units for HDMI Audio wall clock DTO\n \n [Why]\n -Pass and use pixel clock in 100 Hz to Audio for HDMI\n audio DTO for Audio wall clock programming so audio DTO gets\n increased precision for timings with /1001 factor.\n -For HDMI TMDS for N and CTS ACR tables are based on 10 KHz\n units, these does not need to be modified as N and CTS values\n are still valid using current tables.\n \n Signed-off-by: Nevenko Stupar \u003cNevenko.Stupar@amd.com\u003e\n Reviewed-by: Charlene Liu \u003cCharlene.Liu@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7cecfe9d1976f6da82ba2836918f13526c1cf3d1\nAuthor: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\nDate: Wed Jun 5 15:02:04 2019 -0400\n\n drm/amd/display: Copy max_clks_by_state after dce_clk_mgr_construct\n \n [Why]\n For DCE110, DCE112 and DCE120 the max_clks_by_state for the clk_mgr are\n copied from their respective table before the call to\n dce_clk_mgr_construct, but then dce_clk_mgr_construct overwrites\n these with the dce80_max_clks_by_state.\n \n [How]\n Copy these after we call dce_clk_mgr_construct so we\u0027re using the\n right tables.\n \n Signed-off-by: Nicholas Kazlauskas \u003cnicholas.kazlauskas@amd.com\u003e\n Reviewed-by: David Francis \u003cDavid.Francis@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 08d238e68989b2389afb7d0cb489386ffc43e13c\nAuthor: Murton Liu \u003cmurton.liu@amd.com\u003e\nDate: Mon Jun 10 17:55:28 2019 -0400\n\n drm/amd/display: Clock does not lower in Updateplanes\n \n [why]\n We reset the optimized_required in atomic_plane_disable\n flag immediately after it is set in atomic_plane_disconnect, causing us to\n never have flag set during next flip in UpdatePlanes.\n \n [how]\n Optimize directly after each time plane is removed.\n \n Signed-off-by: Murton Liu \u003cmurton.liu@amd.com\u003e\n Reviewed-by: Tony Cheng \u003cTony.Cheng@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 0430017149c53f20493ebeee856315c669d18f4d\nAuthor: David Galiffi \u003cdavid.galiffi@amd.com\u003e\nDate: Fri Jun 7 21:32:34 2019 -0400\n\n drm/amd/display: Incorrect Read Interval Time For CR Sequence\n \n [WHY]\n TRAINING_AUX_RD_INTERVAL (DPCD 000Eh) modifies the read interval\n for the EQ training sequence. CR read interval should remain 100 us.\n Currently, the CR interval is also being modified.\n \n [HOW]\n lt_settings-\u003ecr_pattern_time should always be 100 us.\n \n Signed-off-by: David Galiffi \u003cdavid.galiffi@amd.com\u003e\n Reviewed-by: Tony Cheng \u003cTony.Cheng@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 0b6cbbd5da8f3ab99442bdc0df6b6bb619f87d5a\nAuthor: Aric Cyr \u003caric.cyr@amd.com\u003e\nDate: Mon Jun 10 08:49:36 2019 -0400\n\n drm/amd/display: 3.2.38\n \n Signed-off-by: Aric Cyr \u003caric.cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c70b4016306a10b2c6e5d5da96a1f04a6248900f\nAuthor: Charlene Liu \u003ccharlene.liu@amd.com\u003e\nDate: Wed Jun 5 15:21:03 2019 -0400\n\n drm/amd/display: Split out common HUBP registers and code\n \n There are shared regs and code across DCN generations. Pull them out\n into a shared common location.\n \n Also, expose some dcn20 init functions.\n \n Signed-off-by: Charlene Liu \u003ccharlene.liu@amd.com\u003e\n Reviewed-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 08900ab73225584e4a260223a0848e2825e226fe\nAuthor: Samson Tam \u003cSamson.Tam@amd.com\u003e\nDate: Tue Jun 4 15:52:59 2019 -0400\n\n drm/amd/display: skip retrain in dc_link_set_preferred_link_settings() if using passive dongle\n \n [Why]\n Fixes issue when we have a display connected using a passive\n dongle and then emulate over it using a DP connection at 1 x 1.62 Ghz.\n System hangs because register bus returns back 0xFFFFFFFF for all\n register reads after setting register DIG_BE_CNTL in\n dcn10_link_encoder_connect_dig_be_to_fe(). Hang occurs later\n when trying to do a register read.\n \n [How]\n At the start of the emulation, dc_link_set_preferred_link_settings()\n and dp_retrain_link_dp_test() is called, even though it is connected\n using a passive dongle.\n \n Add an extra condition in dp_retrain_link_dp_test() to check for\n link-\u003edongle_max_pix_clk \u003e 0. This is the only way we know if the\n connection is using passive dongle so we don\u0027t retrain DP.\n \n Signed-off-by: Samson Tam \u003cSamson.Tam@amd.com\u003e\n Reviewed-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6d988a557ee8bd0970c46aea94396ff4815724cd\nAuthor: Jun Lei \u003cJun.Lei@amd.com\u003e\nDate: Wed Jun 5 10:53:40 2019 -0400\n\n drm/amd/display: swap system aperture high/low\n \n [why]\n Currently logical values are swapped in HW, causing\n system aperture to be undefined, so VA and PA cannot co-exist\n \n [how]\n program values correctly\n \n Signed-off-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Reviewed-by: Yongqiang Sun \u003cyongqiang.sun@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 771b286379282ee9676c4233093e4e3c33c78346\nAuthor: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\nDate: Wed Jun 5 14:29:47 2019 -0400\n\n drm/amd/display: Set one 4:2:0-related PPS field as recommended by DSC spec\n \n [why]\n \u0027second_line_offset_adj\u0027 was mistakenly left at zero, even though DSC spec\n v1.2a recommends setting this field to 512 for 4:2:0.\n \n [how]\n Set \u0027second_line_offset_adj\u0027 to 512 for 4:2:0 and leave at zero otherwise\n \n Signed-off-by: Nikola Cornij \u003cnikola.cornij@amd.com\u003e\n Reviewed-by: Eric Bernstein \u003cEric.Bernstein@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a6f30079b8562b659e1d06f7cb1bc30951869bbc\nAuthor: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\nDate: Tue Jun 4 14:48:33 2019 -0400\n\n drm/amd/display: Set default block_size, even in unexpected cases\n \n We\u0027re not expected to enter the default case, but not returning a\n default value here is incorrect.\n \n Signed-off-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Reviewed-by: Eric Bernstein \u003cEric.Bernstein@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5e335add0bccaa8285f26ac4dd8366082a46ac33\nAuthor: Eric Yang \u003cEric.Yang2@amd.com\u003e\nDate: Tue Jun 4 18:14:43 2019 -0400\n\n drm/amd/display: early return when pipe_cnt is 0 in bw validation\n \n [Why]\n Unintentionally introduced behaviour change from previous refactor,\n which causes clks to be 0 in no stream cases, which will cause\n divide by 0.\n \n [How]\n Skip calculation of clocks when no stream. Which is the same as old\n behaviour.\n \n Signed-off-by: Eric Yang \u003cEric.Yang2@amd.com\u003e\n Reviewed-by: Tony Cheng \u003ctong.cheng@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 64af142a8caf8887471d8d79079806875792b5dd\nAuthor: David Francis \u003cDavid.Francis@amd.com\u003e\nDate: Tue Jun 4 16:05:14 2019 -0400\n\n drm/amd/display: Update drm_dsc to reflect native 4.2.0 DSC spec\n \n [Why]\n Some parts of the DSC spec relating to 4.2.0 were not reflected in\n drm_dsc_compute_rc_parameters, causing unexpected config failures\n \n [How]\n Add nsl_bpg_offset and rbs_min computation\n \n Signed-off-by: David Francis \u003cDavid.Francis@amd.com\u003e\n Reviewed-by: Nikola Cornij \u003cNikola.Cornij@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 63b9caff9874202ec7f6aa53abb6a4be0a197dc1\nAuthor: Chiawen Huang \u003cchiawen.huang@amd.com\u003e\nDate: Tue May 14 16:16:11 2019 +0800\n\n drm/amd/display: Add aux tracing log in dce\n \n [Why]\n dce was re-arch\u0027d, therefore adding aux tracing log into new dce\n \n [How]\n The porting from submit_channel_request/process_channel_reply of aux_engine_dce110.c\n \n Signed-off-by: Chiawen Huang \u003cchiawen.huang@amd.com\u003e\n Reviewed-by: Tony Cheng \u003ctong.cheng@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ac3d76e0665cd66b06ee13f5ed61cdbeb8180229\nAuthor: Harmanprit Tatla \u003charmanprit.tatla@amd.com\u003e\nDate: Tue Jun 4 14:12:21 2019 -0400\n\n drm/amd/display: No audio endpoint for Dell MST display\n \n [Why]\n There are certain MST displays (i.e. Dell P2715Q)\n that although have the MST feature set to off may still\n report it is a branch device and a non-zero\n value for downstream port present.\n This can lead to us incorrectly classifying a\n dp dongle connection as being active and\n disabling the audio endpoint for the display.\n \n [How]\n Modified the placement and\n condition used to assign\n the is_branch_dev bit.\n \n Signed-off-by: Harmanprit Tatla \u003charmanprit.tatla@amd.com\u003e\n Reviewed-by: Aric Cyr \u003caric.cyr@amd.com\u003e\n Acked-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ff4258d59031f783039127b90225320bcaa2625c\nAuthor: Jun Lei \u003cJun.Lei@amd.com\u003e\nDate: Mon Jun 3 11:37:44 2019 -0400\n\n drm/amd/display: cap DCFCLK hardmin to 507 for NV10\n \n [why]\n Due to limitation in SMU/PPLIB, it is not possible to know Fmax @ Vmin for DCFCLK.\n This causes issues at high display configurations where extra headroom of DCFCLK\n can enable P-state switching\n \n [how]\n Use existing override logic. If override not defined, then force\n min \u003d 507\n \n Signed-off-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Reviewed-by: Eric Yang \u003ceric.yang2@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6de202373bf6e0cea0f47c2f81274f713c22d84d\nAuthor: Eric Yang \u003cEric.Yang2@amd.com\u003e\nDate: Wed May 22 14:24:40 2019 -0400\n\n drm/amd/display: move bw calc code into helpers\n \n [Why]\n For better readability and reusability\n \n [How]\n Move snippets of BW calculation code into helpers.\n \n Signed-off-by: Eric Yang \u003cEric.Yang2@amd.com\u003e\n Reviewed-by: Fatemeh Darbehani \u003cFatemeh.Darbehani@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4bc846908be51fcd66fed18aba9ba531a38dc725\nAuthor: Aric Cyr \u003caric.cyr@amd.com\u003e\nDate: Mon Jun 3 11:30:43 2019 -0400\n\n drm/amd/display: 3.2.37\n \n Signed-off-by: Aric Cyr \u003caric.cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ec16ac6b4264dffefe3cd137038e138e79e07848\nAuthor: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\nDate: Thu May 30 15:47:51 2019 -0400\n\n drm/amd/display: fix dsc disable\n \n A regression caused dsc to never get disabled in certain situations.\n \n Signed-off-by: Dmytro Laktyushkin \u003cDmytro.Laktyushkin@amd.com\u003e\n Reviewed-by: Nikola Cornij \u003cNikola.Cornij@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b48935b3bfc1350737e759fef5e92db14a2e2fbb\nAuthor: Jun Lei \u003cjun.lei@amd.com\u003e\nDate: Mon Jun 3 08:13:12 2019 -0400\n\n drm/amd/display: fix up HUBBUB hw programming for VM\n \n [why]\n Some values were not being converted or bit-shifted properly for\n HW registers, causing black screen\n \n [how]\n Fix up the values before programming HW\n \n Signed-off-by: Jun Lei \u003cjun.lei@amd.com\u003e\n Reviewed-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 125cfef3e4d82d9c6bda9c0b5868cc95d173d86d\nAuthor: Aric Cyr \u003caric.cyr@amd.com\u003e\nDate: Mon Jun 3 09:12:55 2019 -0400\n\n drm/amd/display: 3.2.36\n \n Signed-off-by: Aric Cyr \u003caric.cyr@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e0a6440a2961b1da3ea895b0bef082fc1a78e190\nAuthor: David Galiffi \u003cDavid.Galiffi@amd.com\u003e\nDate: Thu May 30 11:56:39 2019 -0400\n\n drm/amd/display: Add ability to set preferred link training parameters.\n \n [WHY]\n To add support for OS requirement to set preferred link training\n parameters.\n \n [HOW]\n Create new structure of dp link training overrides. During link training\n processes, these values should be used instead of the default training\n parameters.\n \n Signed-off-by: David Galiffi \u003cDavid.Galiffi@amd.com\u003e\n Reviewed-by: Tony Cheng \u003cTony.Cheng@amd.com\u003e\n Acked-by: Anthony Koo \u003cAnthony.Koo@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 93c25fbdc30a6058115e7a9f3fa161f5a26f7043\nAuthor: Jun Lei \u003cJun.Lei@amd.com\u003e\nDate: Fri May 31 15:14:13 2019 -0400\n\n drm/amd/display: initialize p_state to proper value\n \n [why]\n On some modes SMU will be in infinite loop state at boot, this is\n because driver assumes p_state_support is false, but this is the\n opposite of the assumed boot state by SMU. we optimize away\n notifying SMU about no pstate, and so they will get stuck\n \n [how]\n when we init clk manager, init pstate to true, so it matches driver load\n assumption\n \n Signed-off-by: Jun Lei \u003cJun.Lei@amd.com\u003e\n Reviewed-by: Tony Cheng \u003cTony.Cheng@amd.com\u003e\n Acked-by: Leo Li \u003csunpeng.li@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 53ef3969ddb88a2ee0b69d37f60bd2b1f02f23c9\nAuthor: Leo Liu \u003cleo.liu@amd.com\u003e\nDate: Thu Jul 18 11:38:46 2019 -0400\n\n drm/amdgpu: use VCN firmware offset for cache window\n \n Since we are using the signed FW now, and also using PSP firmware loading,\n but it\u0027s still potential to break driver when loading FW directly\n instead of PSP, so we should add offset.\n \n Signed-off-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b64d98656f0284cb88fa203541d4d0404b0e6c1a\nAuthor: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\nDate: Thu Jul 11 11:10:21 2019 +0800\n\n drm/ttm: use the same attributes when freeing d_page-\u003evaddr\n \n In function __ttm_dma_alloc_page(), d_page-\u003eaddr is allocated\n by dma_alloc_attrs() but freed with use dma_free_coherent() in\n __ttm_dma_free_page().\n Use the correct dma_free_attrs() to free d_page-\u003evaddr.\n \n Signed-off-by: Fuqian Huang \u003chuangfq.daxian@gmail.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c49b1b59a2fa3fa6972afd7e216799f696e67fe5\nAuthor: Kevin Wang \u003ckevin1.wang@amd.com\u003e\nDate: Thu Jul 18 15:46:55 2019 +0800\n\n drm/amd/powerplay: change sysfs pp_dpm_xxx format for navi10\n \n v2:\n set average clock value on level 1 when current clock equal\n min or max clock (fine grained dpm support).\n \n the navi10 gfxclk (sclk) support fine grained DPM,\n so use level 1 to show current dpm freq in sysfs pp_dpm_xxx\n \n Signed-off-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Reviewed-by: Kenneth Feng \u003ckenneth.feng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 33c976c9612a8b414d56e1f591bc39bbacd18b9c\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Thu Jul 18 16:03:46 2019 +0800\n\n drm/amdgpu: drop ras self test\n \n this function is not needed any more. error injection is\n the only way to validate ras but it can\u0027t be executed in\n amdgpu_ras_init, where gpu is even not initialized\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a5dd40ca81bb2465a754709d999a93c5343d1a97\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Thu Jul 18 13:59:38 2019 +0800\n\n drm/amdgpu: only allow error injection to UMC IP block\n \n error injection to other IP blocks (except UMC) will be enabled\n until RAS feature stablize on those IP blocks\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4d249d3abd424e4de4a24ef95306a612e6820f6f\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Thu Jul 18 12:52:56 2019 +0800\n\n drm/amdgpu: disable GFX RAS by default\n \n GFX RAS has not been stablized yet. disable GFX ras until\n it is fully funcitonal.\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit fb2a36075a8650c2fc04f8fbef6901bacf0c834b\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Thu Jul 18 12:49:15 2019 +0800\n\n drm/amdgpu: do not create ras debugfs/sysfs node for ASICs that don\u0027t have ras ability\n \n driver shouldn\u0027t init any ras debugfs/sysfs node for ASICs that don\u0027t have ras\n hardware ability\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 3a48c10d7fb055ff655096eb50e96f36b22b8021\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Wed Jul 17 16:32:27 2019 +0800\n\n drm/amd/powerplay: report bootup clock as max supported on dpm disabled\n \n With gfxclk or uclk dpm disabled, it\u0027s reasonable to report bootup clock\n as the max supported.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit fbdc5d8d84cc521573f1265d5e6ceb04a16075c9\nAuthor: Joseph Greathouse \u003cJoseph.Greathouse@amd.com\u003e\nDate: Wed Jul 17 11:55:22 2019 -0500\n\n drm/amdgpu: Default disable GDS for compute VMIDs\n \n The GDS and GWS blocks default to allowing all VMIDs to\n access all entries. Graphics VMIDs can handle setting\n these limits when the driver launches work. However,\n compute workloads under HWS control don\u0027t go through the\n kernel driver. Instead, HWS firmware should set these\n limits when a process is put into a VMID slot.\n \n Disable access to these devices by default by turning off\n all mask bits (for OA) and setting BASE\u003dSIZE\u003d0 (for GDS\n and GWS) for all compute VMIDs. If a process wants to use\n these resources, they can request this from the HWS\n firmware (when such capabilities are enabled). HWS will\n then handle setting the base and limit for the process when\n it is assigned to a VMID.\n \n This will also prevent user kernels from getting \u0027stuck\u0027 in\n GWS by accident if they write GWS-using code but HWS\n firmware is not set up to handle GWS reset. Until HWS is\n enabled to handle GWS properly, all GWS accesses will\n MEM_VIOL fault the kernel.\n \n v2: Move initialization outside of SRBM mutex\n \n Signed-off-by: Joseph Greathouse \u003cJoseph.Greathouse@amd.com\u003e\n Reviewed-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c52e7ebce72bc9d269c6025da5a4d41601e5f6ca\nAuthor: Jonathan Kim \u003cjonathan.kim@amd.com\u003e\nDate: Thu Jul 11 12:19:44 2019 -0400\n\n drm/amdgpu: exposing fica registers to df offsets\n \n exposing fica registers to poll df pie data for xgmi error counters for\n vega20.\n \n Signed-off-by: Jonathan Kim \u003cJonathan.Kim@amd.com\u003e\n Reviewed-by: Alexander Deucher \u003cAlexander.Deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c06403045aad6ae2edd935f6309e0c72e118c6de\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Thu Jul 11 14:36:44 2019 +0800\n\n drm/amd/powerplay: input check for unsupported message/clock index\n \n This can avoid them to be handled in a wrong way without notice.\n Since not all SMU messages/clocks are supported on every SMU11 ASIC.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7e01a2ec96bf8a149c5e83d0352cf6ea286275cf\nAuthor: Evan Quan \u003cevan.quan@amd.com\u003e\nDate: Thu Jul 11 10:23:17 2019 +0800\n\n drm/amd/powerplay: correct SW SMU valid mapping check\n \n Current implementation is not actually able to detect\n invalid message/table/workload mapping.\n \n Signed-off-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a08a4dae7a11d6665c8d2a49a2c01e8a71a18955\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Mon Jul 15 16:16:00 2019 -0500\n\n drm/amdgpu: flag arcturus as experimental for now\n \n Current support will only work in internal engineering\n boards.\n \n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ad91b134a2e6e18f4c7ea32043f96e46a569ae43\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Thu Jul 11 22:10:31 2019 -0500\n\n drm/amdgpu: drop unused function definitions\n \n These were dropped and the headers never got cleaned up.\n \n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1da418ba65a735bd90ceb8e72afb735f6103a354\nAuthor: James Zhu \u003cJames.Zhu@amd.com\u003e\nDate: Mon Jul 1 19:12:14 2019 -0400\n\n drm/amdgpu:add all VCN rings into schedule request queue\n \n Add all VCN instances\u0027 decode/encode/jpeg decode rings into\n drm_sched_rq list.\n \n Signed-off-by: James Zhu \u003cJames.Zhu@amd.com\u003e\n Reviewed-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Reviewed-by: Boyuan Zhang \u003cboyuan.zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 47a7fe53167cf7b00f6ee9b33992b2faf1e3213e\nAuthor: Oak Zeng \u003cOak.Zeng@amd.com\u003e\nDate: Tue Jul 2 22:01:59 2019 -0500\n\n drm/amdkfd: Increase vcrat size for GPU\n \n GPU cache info (part of virtual CRAT) size depends on CU number.\n For arcturus, CU number has been increased. So the required memory\n for vcrat also increases.\n \n Signed-off-by: Oak Zeng \u003cOak.Zeng@amd.com\u003e\n Reviewed-by: Yong Zhao \u003cYong.Zhao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 69d4de94f8d35111b3cb54821d3d933b93d87343\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Jul 2 11:15:37 2019 +0800\n\n drm/amdgpu: enable all 8 sdma instances for Arcturus silicon\n \n The more 6 sdma instances work fine now with DF fix in vbios:\n * mmDF_PIE_AON_MiscClientsEnable(0x1c728)\u003d0x3fe(DF_ALL_INSTANCE)\n [9:4]MmhubsEnable\u003d3f (change from 0)\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 37f86a9b3617d55ad8189e1b7e6468b85dba4b88\nAuthor: Jay Cornwall \u003cJay.Cornwall@amd.com\u003e\nDate: Mon Jul 1 15:46:56 2019 -0500\n\n drm/amdkfd: Merge gfx9/arcturus trap handlers, add ACC VGPR save\n \n ACC VGPRs are a secondary VGPR set of same size as the primary VGPRs.\n Save them as a block immediately following VGPRs.\n \n Signed-off-by: Jay Cornwall \u003cJay.Cornwall@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5ddd4a9a7c25a6a23a79f973e7a87b1403503719\nAuthor: Yong Zhao \u003cYong.Zhao@amd.com\u003e\nDate: Mon Jul 1 00:48:40 2019 -0400\n\n drm/amdgpu: Add more detail to the VM fault printing\n \n With the printing, we don\u0027t need to parse the value on our own any more.\n \n Signed-off-by: Yong Zhao \u003cYong.Zhao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e30d90fca3fc1aa24e8b9f182a857c32f6186be2\nAuthor: Oak Zeng \u003cOak.Zeng@amd.com\u003e\nDate: Tue Jul 9 13:16:37 2019 -0500\n\n drm/amdkfd: Add device id for real asics\n \n Add pci device ids.\n \n Signed-off-by: Oak Zeng \u003cOak.Zeng@amd.com\u003e\n Reviewed-by: Yong Zhao \u003cYong.Zhao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit fc1e272e8d21ebe5c992db534b855c604b060cdb\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Sun Jun 30 11:35:32 2019 +0800\n\n drm/amdgpu: limit sdma instances to 2 for Arcturus in BU phase\n \n Another 6 sdma instances do not work at present. Disable them to unblock KFD\n for silicon bringup as a workaround\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f9cf36fcafee8e12ffc6026dc8410c17aa21c819\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Sat Jun 29 22:22:13 2019 +0800\n\n drm/amdgpu: skip gfx 9 common golden settings for arct\n \n They are not needed by arct\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 3baa24f0fc11d826313c312c4154a43a47474367\nAuthor: Oak Zeng \u003cOak.Zeng@amd.com\u003e\nDate: Tue Jul 9 13:13:33 2019 -0500\n\n drm/amdkfd: Add arcturus CWSR trap handler\n \n CWSR (compute wave save/restore) is used for\n preempting compute queues.\n \n Signed-off-by: Oak Zeng \u003cOak.Zeng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 54bd77f3d08d4d044b13041286817d9205ecf688\nAuthor: Yong Zhao \u003cYong.Zhao@amd.com\u003e\nDate: Fri Jun 28 05:45:39 2019 -0400\n\n amd/powerplay: No SW XGMI dpm for Arcturus rev 2\n \n xgmi dpm is handled by the SMU.\n \n Signed-off-by: Yong Zhao \u003cYong.Zhao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a80955176ddd89ced22181d738ce8d77b90d0802\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Fri Jun 28 15:08:04 2019 +0800\n\n drm/amdgpu: clean up nonexistent firmware declaration for Arcturus\n \n CPG firmwares are not used.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Evan Quan \u003cevan.quan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 22f5ea4ca04cb68910bccdc8c6c0f1a44237570f\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Fri Jun 28 13:22:32 2019 +0800\n\n drm/amdgpu: init gds config for arct\n \n arct has 4KB gds (4 banks inside) so the max_wave_id\n should be 0xfff\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Le Ma \u003cLe.Ma@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit bfa3a9bb9859849ef2e8c0ee4628e8ed5aa13024\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Fri Jun 28 11:07:53 2019 +0800\n\n drm/amdgpu: keep stolen memory for arct\n \n Any dce register read back from arct is invalid. use hard code\n stolen memory for arct until we validate the s3.\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Le Ma \u003cLe.Ma@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d57c3d5634a305ab1575e573ef0fb0af73ac2957\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Thu Jun 27 18:05:30 2019 +0800\n\n drm/amdgpu: init arct external rev id\n \n Properly set the external silicon revision id.\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Le Ma \u003cLe.Ma@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 582870de568d0a275f747c4ec84c870c3b950902\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Thu Jun 27 15:08:48 2019 +0800\n\n drm/amdgpu: add arct gc golden settings\n \n Golden GC register settings from the hw team.\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Le Ma \u003cLe.Ma@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ca1961a2f587a2e967060425f1d4d99f33e83d9f\nAuthor: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\nDate: Thu Jun 27 14:47:42 2019 +0800\n\n drm/amdgpu: add arct sdma golden settings\n \n Golden SDMA register settings from the hw team.\n \n Signed-off-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Reviewed-by: Le Ma \u003cLe.Ma@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 48c69cda452f4817d040908d743a818187767f17\nAuthor: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\nDate: Tue Jul 9 13:10:53 2019 -0500\n\n drm/amdgpu: add pci DID for Arcturus GL-XL.\n \n Add device ids for Arcturus.\n \n Signed-off-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Reviewed-by: Le Ma \u003cLe.Ma@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6c54afc7e83e801cccfcf773c92f7b4f614face8\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Mon Apr 22 17:14:59 2019 +0800\n\n drm/amdgpu: assign fb_start/end in mmhub v9.4 interface\n \n Align with mmhub v1.0.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit cd1fd7b3818a5724374ff80d9a13a4243b7b9df6\nAuthor: James Zhu \u003cJames.Zhu@amd.com\u003e\nDate: Wed Jul 10 12:07:29 2019 -0500\n\n drm/amdgpu: add harvest support for Arcturus\n \n Add VCN harvest support for Arcturus\n \n Signed-off-by: James Zhu \u003cJames.Zhu@amd.com\u003e\n Reviewed-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit fa739f4b06864d3530e8e461eed223d3566c3633\nAuthor: James Zhu \u003cJames.Zhu@amd.com\u003e\nDate: Wed Jul 10 11:06:37 2019 -0500\n\n drm/amdgpu: add multiple instances support for Arcturus\n \n Arcturus has dual-VCN. Need add multiple instances support for Arcturus.\n \n Signed-off-by: James Zhu \u003cJames.Zhu@amd.com\u003e\n Reviewed-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c01b6a1d38675652199d12b898c1c23b96b5055f\nAuthor: James Zhu \u003cJames.Zhu@amd.com\u003e\nDate: Wed Jul 10 10:53:34 2019 -0500\n\n drm/amdgpu: modify amdgpu_vcn to support multiple instances\n \n Arcturus has dual-VCN. Need Restruct amdgpu_device::vcn to support\n multiple vcns. There are no any logical changes here\n \n Signed-off-by: James Zhu \u003cJames.Zhu@amd.com\u003e\n Reviewed-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 989b6a0549977faf0b5b8d7e1c2634e880c579a2\nAuthor: James Zhu \u003cJames.Zhu@amd.com\u003e\nDate: Wed Jul 10 10:50:24 2019 -0500\n\n drm/amdgpu: add vcn nbio doorbell range setting for 2nd vcn instance\n \n add vcn nbio doorbell range setting for 2nd vcn instance\n \n Signed-off-by: James Zhu \u003cJames.Zhu@amd.com\u003e\n Reviewed-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8b75a521c05fbb16c944ef30d0503942c219886d\nAuthor: James Zhu \u003cJames.Zhu@amd.com\u003e\nDate: Tue Jul 9 11:07:51 2019 -0500\n\n drm/amdgpu/: increase AMDGPU_MAX_RINGS to add 2nd vcn instance\n \n increase AMDGPU_MAX_RINGS to add 2nd vcn instance\n \n Signed-off-by: James Zhu \u003cJames.Zhu@amd.com\u003e\n Reviewed-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6da061dca995ea5a784762cd1f9b38b4afd78fc0\nAuthor: James Zhu \u003cJames.Zhu@amd.com\u003e\nDate: Tue Jun 4 14:47:10 2019 -0400\n\n drm/amdgpu/: add doorbell assignment for 2nd vcn instance\n \n add doorbell assignment for 2nd vcn instance\n \n Signed-off-by: James Zhu \u003cJames.Zhu@amd.com\u003e\n Reviewed-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a07d163c90bf7d649998fb43ac9eb9a01764d662\nAuthor: James Zhu \u003cJames.Zhu@amd.com\u003e\nDate: Tue Jun 4 14:44:33 2019 -0400\n\n drm/amdgpu/: add ucodeID for 2nd vcn instance\n \n add ucodeID for 2nd vcn instance\n \n Signed-off-by: James Zhu \u003cJames.Zhu@amd.com\u003e\n Reviewed-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8a6fcd35326e95dea685fbccce9ed20172080484\nAuthor: James Zhu \u003cJames.Zhu@amd.com\u003e\nDate: Tue Jun 4 14:41:48 2019 -0400\n\n drm/amdgpu/: add clientID for 2nd vcn instance\n \n add clientID for 2nd vcn instance, remove unused SOC15_IH_CLIENTID_SYSHUB.\n \n Signed-off-by: James Zhu \u003cJames.Zhu@amd.com\u003e\n Reviewed-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b6689cf7b9cd2600ebd6981e19fb5f697819a60b\nAuthor: Oak Zeng \u003cOak.Zeng@amd.com\u003e\nDate: Wed Jun 5 15:45:26 2019 -0500\n\n drm/amdkfd: Set number of xgmi optimized SDMA engines for arcturus\n \n some sdma engines are optimized for xgmi on arcturus.\n \n Signed-off-by: Oak Zeng \u003cOak.Zeng@amd.com\u003e\n Reviewed-by: Yong Zhao \u003cYong.Zhao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5fb7c66508223d62238b5c3574f0ca36effa52af\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Mon May 20 17:04:05 2019 +0800\n\n drm/amdgpu: correct ip for mmHDP_READ_CACHE_INVALIDATE register access\n \n Use the proper IP index for HDP registers.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7d0670f44199f8f38227726843952b7eb3e407ab\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Fri Apr 26 16:36:44 2019 +0800\n\n drm/amdgpu: set system aperture to cover whole FB region in mmhub v9.4\n \n In XGMI configuration, the FB region covers vram region from peer\n device, adjust system aperture to cover all of them\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 75b2fce2d89e7025a88d39efc194413f9a3386ea\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Jun 4 14:58:49 2019 +0800\n\n drm/amdgpu: skip get/update xgmi topology info when no psp exists\n \n We don\u0027t currently have psp support for arcturus so provide a alternative\n mechanism in the meantime.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e7773c1c1133c1fabeaa5442e03d3cef22431392\nAuthor: Chengming Gui \u003cJack.Gui@amd.com\u003e\nDate: Tue Jul 9 11:04:17 2019 -0500\n\n drm/amd/powerplay: get smc firmware and pptable\n \n get smc firmware and pptable for arcturus\n \n Signed-off-by: Chengming Gui \u003cJack.Gui@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1fb8cfc2a3a5fa499bce091ad7fb0a7082394183\nAuthor: Chengming Gui \u003cJack.Gui@amd.com\u003e\nDate: Mon May 27 19:01:39 2019 +0800\n\n drm/amdgpu/powerplay: add smu11 driver interface for arcturus. (v2)\n \n add smu11_driver_if_arcturus.h file.\n \n v2: add license, fix header guard (Alex)\n \n Signed-off-by: Chengming Gui \u003cJack.Gui@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6fba59060c6c87cdb80cd2b3de0b16fc09ba4b03\nAuthor: Chengming Gui \u003cJack.Gui@amd.com\u003e\nDate: Tue Jul 9 10:52:20 2019 -0500\n\n drm/amdgpu/powerplay: add arcturus ppt functions\n \n add arcturus_ppsmc.h arcturus_ppt.c and arcturus_ppt.h files.\n \n This is the initial power management support for Arcturus.\n \n Signed-off-by: Chengming Gui \u003cJack.Gui@amd.com\u003e\n Reviewed-by: Kevin Wang \u003ckevin1.wang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 530e30fc32d3834288980777825a8ec0303fbccc\nAuthor: Leo Liu \u003cleo.liu@amd.com\u003e\nDate: Fri Apr 26 13:46:21 2019 -0400\n\n drm/amdgpu: enable the Doorbell support for VCN2.5\n \n Including decode, encode, and JPEG decode rings\n \n Signed-off-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 39a5053fb2231d26f296734390bae7d7d36a9130\nAuthor: Leo Liu \u003cleo.liu@amd.com\u003e\nDate: Tue Jul 9 10:18:36 2019 -0500\n\n drm/amdgpu: add vcn doorbell range function to nbio7.4 (v2)\n \n To setup the aperture for VCN2.5\n \n v2: setup vcn doorbells in vcn2.5 hw_init (Alex)\n \n Signed-off-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 08249a3a322b8950db31996cc5b00bcd292e91f4\nAuthor: Leo Liu \u003cleo.liu@amd.com\u003e\nDate: Tue Apr 16 11:42:56 2019 -0400\n\n drm/amdgpu: enable VCN2.5 on Arcturus\n \n VCN is the video decode and encode engine on Arcturus\n \n Signed-off-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8001073708108e27112d3c63cedc40b9f66c3bac\nAuthor: Leo Liu \u003cleo.liu@amd.com\u003e\nDate: Mon Apr 22 12:21:16 2019 -0400\n\n drm/amdgpu/VCN2.5: set JPEG decode ring functions\n \n Also reuse most of the JPEG2.0 decode ring functions\n \n Signed-off-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e87d5a7a23c7020cf0770726ae34e50ac66f7ced\nAuthor: Leo Liu \u003cleo.liu@amd.com\u003e\nDate: Tue Apr 16 11:32:22 2019 -0400\n\n drm/amdgpu: add JPEG2.5 HW start and stop\n \n JPEG engine initialization and suspend sequences\n \n Signed-off-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a4767886e5cb687ca04f5f1aaf8cbf42b3f88404\nAuthor: Leo Liu \u003cleo.liu@amd.com\u003e\nDate: Mon Apr 22 12:17:38 2019 -0400\n\n drm/amdgpu/VCN2.5: set encode ring functions\n \n Also reuse most of the VCN2.0 encode ring functions\n \n Signed-off-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 185a579700ec9c24e0a10e49db431c3eca9aba72\nAuthor: Leo Liu \u003cleo.liu@amd.com\u003e\nDate: Tue Apr 16 11:17:46 2019 -0400\n\n drm/amdgpu/VCN2.5: set decode ring functions\n \n Also reuse most of the VCN2.0 decode ring functions\n \n Signed-off-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a7c0e4019f7473055713fa011079dd19a8a16d5f\nAuthor: Leo Liu \u003cleo.liu@amd.com\u003e\nDate: Tue Jul 9 10:09:06 2019 -0500\n\n drm/amdgpu: add Arcturus to the VCN family\n \n including firmware support etc.\n \n Signed-off-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit cbead2bdfcf1ddaa3e65de22a3f88034736a71fd\nAuthor: Leo Liu \u003cleo.liu@amd.com\u003e\nDate: Mon Apr 15 12:41:09 2019 -0400\n\n drm/amdgpu: add VCN2.5 VCPU start and stop\n \n HW engine initialization and suspend sequences.\n \n Signed-off-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 28c17d72072b7ce5cc13a527dcbfc045cc995e7d\nAuthor: Leo Liu \u003cleo.liu@amd.com\u003e\nDate: Mon Apr 15 12:21:42 2019 -0400\n\n drm/amdgpu: add VCN2.5 basic supports\n \n i.e. basic VCN IP SW structures\n \n VCN is the video codec block on the GPU.\n \n Signed-off-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit cdbd115eaf1d17cff0d84c99e53774acba96eca7\nAuthor: Leo Liu \u003cleo.liu@amd.com\u003e\nDate: Tue Jul 9 10:04:39 2019 -0500\n\n drm/amdgpu/VCN2: expose rings functions\n \n They can be reused by VCN2.x family\n \n Signed-off-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 22a8f442866bf539c7a659923155d9afa03d77bb\nAuthor: Leo Liu \u003cleo.liu@amd.com\u003e\nDate: Mon Apr 15 09:39:06 2019 -0400\n\n drm/amdgpu/VCN2: put IB internal registers offset to structure\n \n So the ring functions can be shared with different VCN versions\n with different internal registers offsets\n \n Signed-off-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit eec28ef03c903f3e404c57ad0d204e0a9d7d9701\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue May 21 19:34:41 2019 +0800\n\n drm/amdgpu: declare sdma firmware binary files for Arcturus\n \n So that they are properly picked up as a driver dependency.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit db6a49d958db4725a1003d208d6890c55a8a811c\nAuthor: James Zhu \u003cJames.Zhu@amd.com\u003e\nDate: Tue Jul 9 10:01:35 2019 -0500\n\n drm/amdgpu: Clear build undefined warning\n \n Add amdgpu_amdkfd_arcturus_get_functions stub when\n CONFIG_HSA_AMD is undefinded.\n \n Signed-off-by: James Zhu \u003cJames.Zhu@amd.com\u003e\n Reviewed-by: Yong Zhao \u003cYong.Zhao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit eb39aff7e0e33870319dd7c70a293c80e2df26c7\nAuthor: Oak Zeng \u003cOak.Zeng@amd.com\u003e\nDate: Wed Mar 20 16:04:10 2019 -0500\n\n drm/amdgpu: Enable xgmi support for Arcturus\n \n xgmi is a high performance cross-GPU communication channel.\n \n Signed-off-by: Oak Zeng \u003cOak.Zeng@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2f2eab3acc3c8b37d1cdbf00a441373a973690a9\nAuthor: Oak Zeng \u003cOak.Zeng@amd.com\u003e\nDate: Thu Mar 14 11:44:24 2019 -0500\n\n drm/amdgpu: Hack xgmi topology info when there is no psp fw\n \n This is only needed on emulation platform where psp fw might\n not be available, to hack xgmi topology info such as hive id and\n node id.\n \n v2: Add offset to hacked hive/node id\n v3: Don\u0027t use introduce new module parameter.\n \n Signed-off-by: Oak Zeng \u003cOak.Zeng@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 0ad8c5e296fc9ee04e9e918e05f05936ac275623\nAuthor: Yong Zhao \u003cYong.Zhao@amd.com\u003e\nDate: Fri Feb 8 23:06:55 2019 -0500\n\n drm/amdkfd: Support MMHUB1 in kfd interrupt path\n \n Handle interrupts for second mmhub.\n \n Signed-off-by: Yong Zhao \u003cYong.Zhao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c9ffdf5acd4119fbfeaf31cdbeee70e021fd46a5\nAuthor: Yong Zhao \u003cYong.Zhao@amd.com\u003e\nDate: Thu Mar 14 20:28:50 2019 -0400\n\n drm/amdgpu: Set VM_L2_CNTL.PDE_FAULT_CLASSIFICATION to 0 for MMHUB 9.4\n \n Should be set to 0 for mmhub 9.4.\n \n Signed-off-by: Yong Zhao \u003cYong.Zhao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 3e205a0849a760166578b4d95b17e904f23d962e\nAuthor: Oak Zeng \u003cOak.Zeng@amd.com\u003e\nDate: Tue Jul 9 09:59:30 2019 -0500\n\n drm/amdkfd: Implement kfd2kgd_calls for Arcturus\n \n Arcturus shares most of the kfd2kgd_calls with gfx9. But due to\n SDMA register address change, it can\u0027t share SDMA related functions.\n Export gfx9 kfd2kgd_calls and implement SDMA related functions\n for Arcturus.\n \n Signed-off-by: Oak Zeng \u003cOak.Zeng@amd.com\u003e\n Reviewed-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 35cdc81bfa94d10373ecae279f3c48ca858ac4fd\nAuthor: Oak Zeng \u003cOak.Zeng@amd.com\u003e\nDate: Tue Jul 9 09:40:15 2019 -0500\n\n drm/amdkfd: Fix sdma_bitmap overflow issue\n \n In the original formula, when sdma queue number is 64,\n the left shift overflows. Use an equivalence that won\u0027t\n overflow.\n \n Signed-off-by: Oak Zeng \u003cOak.Zeng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 3a68a638a96f1468e267681f833928fd66185a6a\nAuthor: Oak Zeng \u003cozeng@amd.com\u003e\nDate: Wed Dec 12 16:33:56 2018 -0600\n\n drm/amdkfd: Change arcturus sdma engines number\n \n Arcturus has 8 sdma engines\n \n Signed-off-by: Oak Zeng \u003cOak.Zeng@amd.com\u003e\n Reviewed-by: Yong Zhao \u003cyong.zhao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a23e72d7e9dbbc661fc434491ed99ef6efab85a7\nAuthor: Yong Zhao \u003cYong.Zhao@amd.com\u003e\nDate: Fri Oct 12 16:16:15 2018 -0400\n\n drm/amdkfd: Support two MMHUBs when setting up page table base in KFD\n \n 2 mmhubs on arcturus.\n \n Signed-off-by: Yong Zhao \u003cYong.Zhao@amd.com\u003e\n Signed-off-by: Oak Zeng \u003cOak.Zeng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6d5311ab2c2a5e7b5412466be1dc63df1052f6e7\nAuthor: Yong Zhao \u003cYong.Zhao@amd.com\u003e\nDate: Thu Nov 15 12:59:34 2018 -0500\n\n drm/amdkfd: Expose function mmhub_v9_4_setup_vm_pt_regs() for kfd to use\n \n Signed-off-by: Yong Zhao \u003cYong.Zhao@amd.com\u003e\n Signed-off-by: Oak Zeng \u003cOak.Zeng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 49adcf8a6f951450417c14afa6a404b7caea25ef\nAuthor: Yong Zhao \u003cYong.Zhao@amd.com\u003e\nDate: Tue Jul 9 09:37:04 2019 -0500\n\n amd/amdkfd: Add ASIC ARCTURUS to kfd\n \n Add initial support for ARCTURUS to kfd.\n \n Signed-off-by: Yong Zhao \u003cYong.Zhao@amd.com\u003e\n Signed-off-by: Oak Zeng \u003cOak.Zeng@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2fb1e49fda46e78db38eb0bed5ebda3940564458\nAuthor: Oak Zeng \u003cOak.Zeng@amd.com\u003e\nDate: Tue Feb 19 14:59:51 2019 -0600\n\n drm/amdkfd: Support bigger gds size\n \n Extend map_process and set_resources pm4 packet to support\n bigger gds size for arcturus.\n \n v2: Only make the change for v9\n \n Signed-off-by: Oak Zeng \u003cOak.Zeng@amd.com\u003e\n Reviewed-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 3a65d14d259547794db679ce5741da182d6f0b2d\nAuthor: Oak Zeng \u003cOak.Zeng@amd.com\u003e\nDate: Thu Feb 14 14:53:15 2019 -0600\n\n drm/amdkfd: Extend PM4 packets to support 8 SDMA\n \n Extend map_queue and unmap_queue PM4 packets to support 8\n SDMA engines. The new format is backward compatible.\n \n Signed-off-by: Oak Zeng \u003cOak.Zeng@amd.com\u003e\n Reviewed-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7f40581c2e468f7180de7b95e53fbbc9f64682aa\nAuthor: Oak Zeng \u003cOak.Zeng@amd.com\u003e\nDate: Wed Dec 19 08:44:38 2018 -0600\n\n drm/amdgpu: Initialize asic functions for Arcturus\n \n After cherry-picking doorbell rework changes from drm-next\n branch, Arcturus asic functions pointer need to be initialized\n to init doorbell index for Arcturus.\n \n Signed-off-by: Oak Zeng \u003cOak.Zeng@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f864e3e65556968c63f7fc3b129b040e34620400\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Jul 16 13:39:42 2019 -0500\n\n drm/amdgpu: add paging queue support for 8 SDMA instances on Arcturus\n \n Properly enable all 8 instances for paging queue.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f9c84ae582f675c58ff37728c708085278583935\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Feb 26 20:37:17 2019 +0800\n\n drm/amdgpu: correct programming of ih_chicken for Arcturus\n \n ih_chicken is a register that is not allowed to access by driver\n in the L0 security policy.\n psp bl need to enable field to allow driver to use physical\n bus address for ih ring on secure part.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Snow Zhang \u003csnow.zhang@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5ce40fd86cf155e0eefc73509343bb3eeaafa4bc\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Thu Nov 15 18:54:35 2018 +0800\n\n drm/amdgpu: add Arcturus chip_name for init sdma microcode\n \n So we load the proper firmware for arcturus.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Acked-by: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 121d859918a3220db5fb49bcf07803dc5675679e\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Nov 20 15:15:31 2018 +0800\n\n drm/amdgpu: enable 8 SDMA instances for Arcturus\n \n All the 8 SDMA instances work fine on the latest Gopher build model.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Snow Zhang \u003cSnow.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5cd54ab85d6148b666f8ca1e5eb94fae5e4d392a\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Thu Nov 15 18:56:17 2018 +0800\n\n drm/amdgpu: correct Arcturus SDMA address space base index\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 3d81f67a1b1bfc789651e02dfe9705bb8483f133\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Wed Sep 19 14:17:37 2018 +0800\n\n drm/amdgpu: support sdma 2~7 doorbell range register offset\n \n Update the doorbell range registers to support additional\n SDMA rings.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f5cdc2da460267fe234f8b5246896bea62c0db4a\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Sep 18 17:39:59 2018 +0800\n\n drm/amdgpu: skip all gfx ring settings for Arcturus\n \n Not needed on Arcturus.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f8b733b9d4a316122378861526aeb90d3144c83e\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Sep 18 17:04:42 2018 +0800\n\n drm/amdgpu: skip load cp gfx firmware for Arcturus\n \n Arcturus has no CPG component any more.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 24c44c891782088a2d98a3883838186add8e2511\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Sep 18 16:11:44 2018 +0800\n\n drm/amdgpu: optimize gfx9 init_microcode function\n \n Split each type of firmware into single function for easy to maintain.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 65e60f6e06e55ea214d9f56a68f93a6f7c4b8455\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Jul 9 09:30:19 2019 -0500\n\n drm/amdgpu: add Arcturus gpu info firmware\n \n Add GPU info firmware for Arcturus.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f2d6657111ed5c065b1f934b548a279205c3d370\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Sep 11 13:11:28 2018 +0800\n\n drm/amdgpu: skip pasid mapping for second mmhub on Arcturus\n \n There\u0027s no LUT register for second mmhub to convert pasid since it has no ATC.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2065aa5494e465c0efb20fffd607db6cbfa74067\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Sep 11 12:33:11 2018 +0800\n\n drm/amdgpu: skip to get 3D engine clockgating state for Arcturus\n \n It\u0027s because Arcturus has not 3D engine.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2564444f8cbb29e0dabadd067e48c5c68540880d\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Sep 11 12:14:37 2018 +0800\n\n drm/amdgpu: add to set rlc funcs for Arcturus\n \n Shared with other gfx9 parts so use the same functions.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 763bee0fb3a9d1e51fe843b504fd659101c14101\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Sep 11 12:13:41 2018 +0800\n\n drm/amdgpu: add number of mec for Arcturus\n \n MEC is the CP compute microcontroller.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6155e98ac018b886b31f8468380f2acd0fceb390\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Sep 11 12:11:36 2018 +0800\n\n drm/amdgpu: add gfx config for Arcturus\n \n Add Arcturus GFX config.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 84519350a73ec6d98f3d5f43c8a90f95e88b7f67\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Sep 11 11:58:48 2018 +0800\n\n drm/amdgpu: add support for Arcturus firmware\n \n Add support for Arcturus gfx firmwares.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9f6ef81620f7e82c748ee0c48271f220d3f108fb\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Jul 9 09:23:11 2019 -0500\n\n drm/amdgpu/dce_virtual: add Arcturus virtual display support\n \n Virtual dce is a sw only display driver for emulation and\n virtualization and cases where we want to use a virtual\n display subsystem.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7fafd613c03b044883b1b538c57a4149260215a8\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Sep 11 11:35:34 2018 +0800\n\n drm/amdgpu: set Arcturus fw load type as direct\n \n We currently only support direct firmware loading.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 61cf44c1dbb66c4f653de76a3740f732168658cc\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Sep 11 11:20:18 2018 +0800\n\n drm/amdgpu: add to set Arcturus ip blocks\n \n Add IP blocks for Arcturus.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 0e54df05724e8ba7476681405f38e23f2277edd1\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Sep 11 11:07:09 2018 +0800\n\n drm/amdgpu/soc15: add Arcturus common ip blocks\n \n Add common IP blocks for Arcturus.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 0fe6a7b49f61c4ae692c790dd7afc47841314f0e\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Mon Sep 10 19:51:07 2018 +0800\n\n drm/amdgpu: support hdp flush for more sdma instances\n \n The bit RSVD_ENG0 to RSVD_ENG5 in GPU_HDP_FLUSH_REQ/GPU_HDP_FLUSH_DONE\n can be leveraged for sdma instance 2~7 to poll register/memory.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Acked-by: Snow Zhang \u003c Snow.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b482a134ade1a3a69eaf77d3c792706d90596b35\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Mon Sep 10 14:26:44 2018 +0800\n\n drm/amdgpu: specify sdma instance 5~7 with second mmhub type\n \n On Arcturus, sdma instance 5~7 is connected to the second mmhub. The vmhub type\n in amdgpu_ring_funcs is constant, so we create an individual amdgpu_ring_funcs\n with different vmhub type(AMDGPU_MMHUB_1) for these sdma instances.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Acked-by: Snow Zhang \u003c Snow.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 667a48226edfbf4b0b8d1b61255105918fea4af7\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Wed Sep 5 16:21:20 2018 +0800\n\n drm/amdgpu: reorganize sdma v4 code to support more instances\n \n This change is needed for Arcturus which has 8 sdma instances.\n The CG/PG part is not covered for now.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Acked-by: Snow Zhang \u003c Snow.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 51c608984fd5eed89bb9a8bd49133cde405892da\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Thu Sep 6 19:37:51 2018 +0800\n\n drm/amdgpu: update vmc interrupt routine to support 3 vmhubs\n \n There is one more vmc interrupt and mmhub on Arcturus.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Acked-by: Snow Zhang \u003c Snow.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7d19b15f7013c9a9ebdcf253eb016ff37752ecc9\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Thu Sep 6 17:34:06 2018 +0800\n\n drm/amdgpu: add VMC1 interrupt client id for Arcturus\n \n New IH client id for VMC1.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Acked-by: Snow Zhang \u003c Snow.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e78705ec5a7f607a3340c71a104b0d1c327958a4\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Jul 9 09:21:53 2019 -0500\n\n drm/amdgpu: dynamically initialize IP offset for Arcturus\n \n Add support for the IP offsets on Arcturus.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Acked-by: Snow Zhang \u003c Snow.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 113b47e78067db00cc449b6666916715d02ed01c\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Jul 9 09:20:24 2019 -0500\n\n drm/amdgpu: increase max number of ip base instances to 8\n \n For Arcturus, the number of IP base instances is 8.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Acked-by: Snow Zhang \u003c Snow.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit fa5d2e6f0ab273e55786991699127334bf8ef034\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Sep 4 13:36:22 2018 +0800\n\n drm/amdgpu: add SDMA 2~7 ip block type\n \n Add IP block type.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Acked-by: Snow Zhang \u003c Snow.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8024f1d5e180e2a41121e7e6ac1b79a400430a8c\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Mon Sep 3 19:27:52 2018 +0800\n\n drm/amdgpu: add SDMA 2~7 interrupt client id for Arcturus\n \n Add new client ids.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Acked-by: Snow Zhang \u003c Snow.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 51cce480fda970118dd661ce6e0715f3269cbc31\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Sep 4 15:29:52 2018 +0800\n\n drm/amdgpu: use new mmhub interfaces for Arcturus\n \n Arcturus has two MMHUBs.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Acked-by: Snow Zhang \u003c Snow.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2cb2ea1e073fb8d8c1850383b70b78e69ef8b3cc\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Jul 9 09:18:03 2019 -0500\n\n drm/amdgpu: add mmhub v9.4.1 block for Arcturus (v2)\n \n Arcturus as an updated mmhub block. mmhub is the\n memory controller hub used for sdma and multimedia.\n \n v2: squash in AGP BAR programming (Alex)\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c8a6e2a3170064c1f476407139c3dd97d9a9087c\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Fri Aug 31 14:46:47 2018 +0800\n\n drm/amdgpu: add one more mmhub instance for Arcturus (v2)\n \n v2: set mmhub num under CHIP_ARCTURUS switch case and add one more mmhub id_mgr\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 1daa2bfa17aa5ace20bb02509cd3d2c2c0afc642\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Fri Aug 31 14:17:28 2018 +0800\n\n drm/amdgpu: add new member in amdgpu_device for vmhub counts per asic chip\n \n It aims to replace AMDGPU_MAX_VMHUBS in for loop to initialize registers.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a2d15ed733365ba2e0e3eb55a83c7a493eaaaa2c\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Jul 16 13:29:19 2019 -0500\n\n drm/amdgpu: rename AMDGPU_GFXHUB/MMHUB macro with hub number\n \n The number of GFXHUB/MMHUB may be expanded in later ASICs.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 3de2ff5d60d0695568dc8333947b0446a35d49d8\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Sep 4 14:52:25 2018 +0800\n\n drm/amdgpu: add gmc basic support for Arcturus\n \n Add initial GMC support for Arcturus\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d6c3b24ea28ddae06c9ff9fe9ae58664144d7ae8\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Tue Jul 9 09:16:01 2019 -0500\n\n drm/amdgpu: add Arcturus asic type\n \n Add asic type for Arcturus.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit f1cf876931afa1d48bbefa0f3095bfd3cf9fb04c\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Wed Aug 29 16:28:38 2018 +0800\n\n drm/amdgpu: add Arcturus ip_offset header (v3)\n \n Provides the absolute offsets of the IP register\n blocks.\n \n v2: update chip name in source code\n v3: squash in MP offset updates (Alex)\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c54a60db0d722413628b8d6c4c480c7ac1238428\nAuthor: Leo Liu \u003cleo.liu@amd.com\u003e\nDate: Thu Mar 21 09:08:19 2019 -0400\n\n drm/amdgpu: add VCN2.5 headers\n \n VCN is the multi-media block.\n \n Signed-off-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4f727ecefefbd180de10e25b3e74c03dce3f1e75\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Wed Aug 29 16:50:20 2018 +0800\n\n drm/amdgpu: add sdma 4.2.2 header files for Arcturus\n \n SDMA is the system DMA block.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 0e96cf7f676972b1154c4c2b68e878266a405a84\nAuthor: Le Ma \u003cle.ma@amd.com\u003e\nDate: Wed Aug 29 16:49:19 2018 +0800\n\n drm/amdgpu: add mmhub 9.4.1 header files for Acrturus\n \n mmhub is the GPU memory hub used by SDMA and VCN.\n \n Signed-off-by: Le Ma \u003cle.ma@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d1daf8502e890dc4fc3fc17b57ae106397ad316c\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Tue Jul 2 14:42:25 2019 -0500\n\n drm/amdgpu: consolidate navi14 IP init\n \n It\u0027s the same as navi10.\n \n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2665ec41718a8514c85f2becf83c58b8c5041ed6\nAuthor: Alex Deucher \u003calexander.deucher@amd.com\u003e\nDate: Tue Jul 2 14:35:36 2019 -0500\n\n drm/amdgpu: disable concurrent flushes on Navi14\n \n Same thing applies to navi14 as navi10.\n \n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ba02636de54e7c2f8d549401ce9c9f508a05ef7a\nAuthor: Jack Xiao \u003cJack.Xiao@amd.com\u003e\nDate: Fri Jul 5 16:00:08 2019 -0500\n\n drm/amdgpu: enable gfxoff code path for navi14\n \n Based on navi10 gfxoff logic, enable the related code\n path for navi14.\n \n Signed-off-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 0377b08823418e418bcd50d950d2baaeb8b7a1aa\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Jul 2 12:52:52 2019 -0500\n\n drm/amdgpu/vcn: enable indirect DPG SRAM mode for navi14\n \n Enable VCN dynamic powergating for navi14.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e017bb8035ce890ff5a4cded2e8b91b814fb7b21\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Mon May 13 17:02:12 2019 +0800\n\n drm/amd/powerplay: disable gfxoff for navi14\n \n gfxoff doesn\u0027t work on navi14 yet, so disable it for now\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 03917df7e510b4987c0b5bf7e05edf356c03e939\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Jul 2 12:49:41 2019 -0500\n\n drm/amdgpu/nv: set vcn pg flag for navi14\n \n Enable VCN power gating by default.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c5cc14e34d62cac69d39a556c1b4761099404dfe\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu Apr 18 17:46:17 2019 +0800\n\n drm/amd/display: disable display writeback for navi14\n \n not used.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Huang Rui \u003cray.huang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5e0f378d8afb8d2f62a01933b43a296aa117adea\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu Mar 28 16:43:16 2019 +0800\n\n drm/amdgpu: enable async gfx ring for navi14\n \n Same as navi10.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d0c39f8cbfaa4bd599328920d7d13541fd9bbb10\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Wed Mar 20 16:12:54 2019 +0800\n\n drm/amdgpu: enable clock gatings for navi14\n \n Set appropriate CG flags for navi14.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 29e6fd7c86bf13840be40ca507eae25ab1a860d5\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Fri Mar 22 13:10:03 2019 +0800\n\n drm/amdgpu/athub2: set clock gating for navi14\n \n same as navi10.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 408c49de9b8858a6bad725b6d123b0ea22f38df3\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Fri Mar 22 13:03:01 2019 +0800\n\n drm/amdgpu/mmhub2: set clock gating for navi14\n \n same as navi10.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8687b47e3ae0fade2b37ab2653a0eb66a5db9e59\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Wed Mar 20 12:37:45 2019 +0800\n\n drm/amdgpu: declare asd firmware for navi14\n \n So the dependency gets properly tracked.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Snow Zhang \u003csnow.zhang@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e149a2f6ed2c84d10087e4a4f9bb1e7d8893a965\nAuthor: James Zhu \u003cJames.Zhu@amd.com\u003e\nDate: Fri Mar 1 16:23:55 2019 -0500\n\n drm/amdgpu: Enable VCN on navi14\n \n Add navi14 vcn firmware, and enable VCN on navi14.\n \n Signed-off-by: James Zhu \u003cJames.Zhu@amd.com\u003e\n Reviewed-by: Leo Liu \u003cleo.liu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit baebcf2e911267b2f5e4cdff3b822d0648ce33c5\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Fri Mar 15 19:15:21 2019 +0800\n\n drm/amd/display: skip to load dmcu firmware for navi14\n \n not needed for navi14 at the moment.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e470d287c3f10d03eef25efecad11791691b737c\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Fri Mar 15 19:10:47 2019 +0800\n\n drm/amdgpu: skip to load ta firmware for navi14\n \n Not relevant on navi14.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c6e764b40a78b6ffdd70431b03cacc999c069e9d\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Tue Jul 2 10:43:55 2019 -0500\n\n drm/amd/display: add ASICREV defines v2\n \n Add revs for navi10 and 14.\n \n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Reviewed-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 8fceceb69e5370af575d0496f2e25bd035ad69b6\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Tue Feb 26 13:38:17 2019 -0500\n\n drm/amd/display: add dm block\n \n enable DC for navi14.\n \n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 2ebe177344d79bc61294d4365376978610d316c3\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Tue Feb 26 14:38:59 2019 -0500\n\n drm/amd/display: add NAVI14 in resource construct\n \n Change the pipes to 5 if the asic is nv14\n \n This is a temp patch, there was some refactor in the dml part of the code.\n which is not in this branch. for now this is good, we can implement this\n properly once we have an updated branch.\n \n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit fce651e393a3d130dfb2e34de6afc9afc565b2f2\nAuthor: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\nDate: Tue Jul 2 10:41:40 2019 -0500\n\n drm/amd/display: add nv14 cases to amdgpu_dm\n \n Mostly shared with navi10.\n \n Signed-off-by: Bhawanpreet Lakha \u003cBhawanpreet.Lakha@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ab5e51211cbe08843c0f0e796601903c094f6db4\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Wed Feb 13 03:34:54 2019 +0800\n\n drm/amdgpu: enable sw smu ip for navi14\n \n same as navi10.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b02ff126898939a685c9a066b5d80dde3ff1c66f\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Wed Feb 13 03:44:10 2019 +0800\n\n drm/amdgpu/smu11: add support for navi14\n \n Same as navi10.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7840d8babe71d5bd447465779fd73ff59855d2fa\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Feb 12 03:12:31 2019 +0000\n\n drm/amdgpu/psp: start rlc autoload after psp received rlcg for navi14\n \n Update for navi14.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4adc0732fceaa2166c3bbd6ff5d8f40b74783cf0\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Sun Feb 10 21:45:32 2019 +0000\n\n drm/amdgpu: enable psp ip block for navi14\n \n Same as navi10.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Snow Zhang \u003cSnow.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 82522b2d7f54a80daecf46a5931ac9bebcb90307\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Mon Jul 8 14:03:15 2019 -0500\n\n drm/amdgpu/psp: add psp support for navi14 (v3)\n \n Same as navi10.\n \n v2: squash in logic fix (Colin Ian King)\n v3: squash in logic simplification (Alex)\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Snow Zhang \u003cSnow.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 691f69b6a6b283e488ac91581640fd4e8fdb8c4a\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Wed Jan 16 10:23:17 2019 +0800\n\n drm/amdgpu: enable virtual display for navi14\n \n Virtual display is a sw based kms interface for virtualization\n and emulation.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a29bfd1283abddae6a58686a934acc7102cf6778\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Wed Dec 19 20:39:37 2018 +0800\n\n drm/amdgpu: add ip blocks for navi14\n \n Add the initial IP blocks for navi14\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 5e71e011ff84cd467a6c255e8ea8dcd77ec442ff\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Mon Dec 17 18:23:27 2018 +0800\n\n drm/amdgpu/soc15: add support for navi14\n \n same as navi10\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b8cb98cd3e2b600794585958dbea53d433d86736\nAuthor: Jack Xiao \u003cJack.Xiao@amd.com\u003e\nDate: Tue May 28 13:27:11 2019 +0800\n\n drm/amdgpu/gfx10: fix programming of SC_HIZ_TILE_FIFO_SIZE field\n \n max fifo size is 128 and PA_SC_FIFO_SIZE[20:15]\u003dSC_HIZ_TILE_FIFO_SIZE\n field is programmed in units of two entries, but 6 bits is insufficient\n to hold value 128/2 \u003d 64, so set this field as 0 which is interpreted by\n the hardware as maximum physical fifo size(128).\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Signed-off-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 3ddec51511fc6ba84595565b6f48badabd0e47dc\nAuthor: Tao Zhou \u003ctao.zhou1@amd.com\u003e\nDate: Tue Jul 2 14:20:04 2019 -0500\n\n drm/amdgpu/gfx10: update gfx golden settings for navi14\n \n Updated settings from hw team.\n \n Signed-off-by: Tao Zhou \u003ctao.zhou1@amd.com\u003e\n Acked-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4904ede127925df6225b1d96621c0b9e83a74f59\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Fri Mar 29 19:46:46 2019 +0800\n\n drm/amdgpu/gfx10: update gfx golden settings for navi14\n \n Add updated settings from hw team.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit be184b4ccc92be23de8ca8bce94d41c3f706121a\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu Feb 14 07:12:00 2019 +0800\n\n drm/amdgpu/gfx: update gc_v10_1_1 golden setting\n \n Updated settings for hw team.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7417846725e9bd4801c5434600d233fa93e2bea4\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Feb 12 02:56:57 2019 +0000\n\n drm/amdgpu/gfx10: add gfx v10_1_1 golden settings for navi14\n \n Add golden settings for navi14.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Snow Zhang \u003cSnow.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 58acab6629e114733dd1e2664b1be24647f5f536\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Thu Feb 14 09:06:10 2019 +0800\n\n drm/amdgpu/gfx: add definition of mmCGTT_GS_NGG_CLK_CTRL\n \n Needed for clockgating.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit d55c193dbd43c8e548c93b69cda7f5a60e70c596\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Jan 29 22:36:15 2019 +0800\n\n drm/amdgpu/gfx10: set tcp harvest for navi14\n \n Update settings for navi14.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e938ded64810151b99717b635219dbbc201970f1\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Dec 25 14:45:57 2018 +0800\n\n drm/amdgpu: set rlc funcs for navi14\n \n Same as navi10.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 0c090023c639577e1a00438e21a112cffb3843d9\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Dec 25 14:45:21 2018 +0800\n\n drm/amdgpu: add me/mec configurations for navi14\n \n Add navi14 to appropriate cases.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 41957a8ea9e0fb2d57b9d3b429f197ad96d42ac1\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Mon Dec 17 18:22:16 2018 +0800\n\n drm/amdgpu/gfx10: add clockgating support for navi14\n \n Same as navi10.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 4bd80a4663c2b20cc92752ac73794458a86d1dab\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Mon Dec 17 18:21:35 2018 +0800\n\n drm/amdgpu/gfx10: add gfx config for navi14\n \n Add gfx config details for navi14.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 47b67bd7d426f67d595ec2af358fd8ecad91ab61\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Jul 16 13:22:04 2019 -0500\n\n drm/amdgpu/gfx10: add placeholder for navi14 golden settings\n \n To be filled in once available.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 654bcee0e4ea837ce661b4c305ade229f757093e\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Mon Dec 17 18:08:28 2018 +0800\n\n drm/amdgpu/gfx10: add support for navi14 firmware\n \n Add support for navi14 CP firmware files.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 9571710f0c4e4ad7ef98b1ae491fe3009e827b1a\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Fri Mar 22 13:14:25 2019 +0800\n\n drm/amdgpu/sdma5: set clock gating for navi14\n \n same as navi10.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a994b742b89df02f15193f4074f6a43e026262cb\nAuthor: tiancyin \u003ctianci.yin@amd.com\u003e\nDate: Tue May 21 14:43:48 2019 +0800\n\n drm/amdgpu/sdma5: update sdma5 golden settings for navi14\n \n add new registers:\n mmSDMA0_RLC3_RB_WPTR_POLL_CNTL,\n mmSDMA1_RLC3_RB_WPTR_POLL_CNTL\n \n Reviewed-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Signed-off-by: tiancyin \u003ctianci.yin@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c049af3e850be62d3add8c902a9f5c45dcae8373\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Feb 12 02:58:06 2019 +0000\n\n drm/amdgpu/sdma5: add sdma5_0 golden settings for navi14\n \n Add settings for navi14.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Snow Zhang \u003cSnow.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 06823925ad6b4fe14e652e3a386ac1472f371fa9\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Mon Dec 17 18:07:22 2018 +0800\n\n drm/amdgpu/sdma5: add placeholder for navi14 golden settings\n \n To be filled in once they are available.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 6041f2a281e50728390da740ae9ef78bf579911e\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Mon Dec 17 18:05:32 2018 +0800\n\n drm/amdgpu/sdma5: add support for navi14 firmware\n \n Add support for navi14 sdma firmware files.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 05d72b8d36bceed5142f676b592d1a35fc23f584\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Mon Dec 17 18:19:42 2018 +0800\n\n drm/amdgpu/gmc10: add navi14 support\n \n same as navi10\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c8ff09bf41f851e6e9bb2a9f8353f6c78f80f3c1\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Tue Dec 25 14:44:23 2018 +0800\n\n drm/amdgpu: increase max instance number for hw ip\n \n max instance number is 6 for navi10 and 7 for navi14, and we increase the\n reg_offset array size to avoid out-of-bound access\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit c20697674d3264f1cb2094eeec41c36d8f3f6320\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Wed Jun 5 17:58:57 2019 +0800\n\n drm/amdgpu/discovery: init reg base offset via ip discovery for navi14\n \n Add IP discovery for navi14.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit a0f6d926f139d6fe20ca39f4a27ba0c51458bfd1\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Mon Dec 17 18:24:03 2018 +0800\n\n drm/amdgpu/soc15: initialize reg base for navi14 (v2)\n \n Initialize the IP register base offsets for navi14.\n \n v2: squash in MP, CLK, THM updates\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit e2d2607f90732e59153c5cb74a31b86259d83bc2\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Mon Dec 17 18:04:19 2018 +0800\n\n drm/amdgpu: add navi14 ucode loading method\n \n Same as navi10.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 7ecb5cd451b7bce8196bf40c6e10c5f266b3994d\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Mon Dec 17 18:02:43 2018 +0800\n\n drm/amdgpu: set asic family and ip blocks for navi14\n \n same with navi10\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit ed42cfe1ac905ebca2f59e4d9bed230e8e1599f5\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Mon Dec 17 18:01:38 2018 +0800\n\n drm/amdgpu: add gpu_info firmware for navi14\n \n Add navi14 to case statement to load the GPU info firmware.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 87dbad02d2254b741c71ce859c451fb1ae6f5340\nAuthor: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\nDate: Mon Dec 17 18:00:26 2018 +0800\n\n drm/amdgpu: add navi14 asic type\n \n Add CHIP_NAVI14 to the list of asic types.\n \n Signed-off-by: Xiaojie Yuan \u003cxiaojie.yuan@amd.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Jack Xiao \u003cJack.Xiao@amd.com\u003e\n Signed-off-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit b5ea9c9337007d6e700280c8a60b4e10d070fb53\nAuthor: Dhinakaran Pandiyan \u003cdhinakaran.pandiyan@intel.com\u003e\nDate: Wed Jul 17 15:34:51 2019 -0700\n\n drm/i915/vbt: Fix VBT parsing for the PSR section\n \n A single 32-bit PSR2 training pattern field follows the sixteen element\n array of PSR table entries in the VBT spec. But, we incorrectly define\n this PSR2 field for each of the PSR table entries. As a result, the PSR1\n training pattern duration for any panel_type !\u003d 0 will be parsed\n incorrectly. Secondly, PSR2 training pattern durations for VBTs with bdb\n version \u003e\u003d 226 will also be wrong.\n \n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: stable@vger.kernel.org\n Cc: stable@vger.kernel.org #v5.2\n Fixes: 88a0d9606aff (\"drm/i915/vbt: Parse and use the new field with PSR2 TP2/3 wakeup time\")\n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111088\n Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id\u003d204183\n Signed-off-by: Dhinakaran Pandiyan \u003cdhinakaran.pandiyan@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Acked-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Tested-by: François Guerraz \u003ckubrick@fgv6.net\u003e\n Signed-off-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190717223451.2595-1-dhinakaran.pandiyan@intel.com\n\ncommit bdae33b8b82bb379a5b11040b0b37df25c7871c9\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Jul 18 15:54:05 2019 +0100\n\n drm/i915: Use maximum write flush for pwrite_gtt\n \n As recently disovered by forcing big-core (!llc) machines to use the GTT\n paths, we need our full GTT write flush before manipulating the GTT PTE\n or else the writes may be directed to the wrong page.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Matthew Auld \u003cmatthew.william.auld@gmail.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: stable@vger.kernel.org\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190718145407.21352-2-chris@chris-wilson.co.uk\n\ncommit d45a4dd527068f0d7ee6cdd9e13c03b3f9e59dba\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Jul 18 15:54:04 2019 +0100\n\n drm/i915: Drop wmb() inside pread_gtt\n \n Inside pread, we only ever read from the GTT so the serialising wmb()\n instructions around the GGTT PTE updates are pointless.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190718145407.21352-1-chris@chris-wilson.co.uk\n\ncommit bffb31f73b29a60ef693842d8744950c2819851d\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Jul 17 14:45:36 2019 +0300\n\n drm/i915: Make sure cdclk is high enough for DP audio on VLV/CHV\n \n On VLV/CHV there is some kind of linkage between the cdclk frequency\n and the DP link frequency. The spec says:\n \"For DP audio configuration, cdclk frequency shall be set to\n meet the following requirements:\n DP Link Frequency(MHz) | Cdclk frequency(MHz)\n 270 | 320 or higher\n 162 | 200 or higher\"\n \n I suspect that would more accurately be expressed as\n \"cdclk \u003e\u003d DP link clock\", and in any case we can express it like\n that in the code because of the limited set of cdclk (200, 266,\n 320, 400 MHz) and link frequencies (162 and 270 MHz) we support.\n \n Without this we can end up in a situation where the cdclk\n is too low and enabling DP audio will kill the pipe. Happens\n eg. with 2560x1440 modes where the 266MHz cdclk is sufficient\n to pump the pixels (241.5 MHz dotclock) but is too low for\n the DP audio due to the link frequency being 270 MHz.\n \n v2: Spell out the cdclk and link frequencies we actually support\n \n Cc: stable@vger.kernel.org\n Tested-by: Stefan Gottwald \u003cgottwald@igel.com\u003e\n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111149\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190717114536.22937-1-ville.syrjala@linux.intel.com\n Acked-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n\ncommit 9a36a6517d5cc8bf7d9c1fde9058269701802e31\nAuthor: Vivek Kasireddy \u003cvivek.kasireddy@intel.com\u003e\nDate: Tue Jul 16 19:13:16 2019 -0700\n\n drm/i915/ehl: Use an id of 4 while accessing DPLL4\u0027s CR0 and CR1\n \n Although, DPLL4 enable and disable is associated with MGPLL1_ENABLE\n register, we can use ICL_DPLL_CFGCR0/CR1 macros to access this dpll\u0027s\n CR0 and CR1 registers by passing an id of 4 to these macros.\n \n Reported-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Cc: Imre Deak \u003cimre.deak@intel.com\u003e\n Signed-off-by: Vivek Kasireddy \u003cvivek.kasireddy@intel.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190717021316.18610-1-vivek.kasireddy@intel.com\n\ncommit 5270130db8c8694f7d97cc4fdc9440fb30bc2192\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Jun 26 21:03:40 2019 +0300\n\n drm/i915: Add gen8_de_pipe_fault_mask()\n \n Reduce the clutter a bit by introducing gen8_de_pipe_fault_mask().\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626180344.26314-2-ville.syrjala@linux.intel.com\n\ncommit 6b2436aeb945961f86e2b6b74be1a4b1a372a1a4\nAuthor: Michel Thierry \u003cmichel.thierry@intel.com\u003e\nDate: Fri Jul 12 14:02:39 2019 -0700\n\n x86/gpu: add TGL stolen memory support\n \n Reuse Gen11 stolen memory changes since Tiger Lake uses the same BSM\n register (and format).\n \n Cc: Ingo Molnar \u003cmingo@redhat.com\u003e\n Cc: Borislav Petkov \u003cbp@alien8.de\u003e\n Cc: \"H. Peter Anvin\" \u003chpa@zytor.com\u003e\n Cc: x86@kernel.org\n Signed-off-by: Michel Thierry \u003cmichel.thierry@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Acked-by: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712210238.5622-1-lucas.demarchi@intel.com\n\ncommit c30d5dc653cbc78f9b634b7b72e25057a68c527c\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jul 16 13:49:28 2019 +0100\n\n drm/i915/gt: Push engine stopping into reset-prepare\n \n Push the engine stop into the back reset_prepare (where it already was!)\n This allows us to avoid dangerously setting the RING registers to 0 for\n logical contexts. If we clear the register on a live context, those\n invalid register values are recorded in the logical context state and\n replayed (with hilarious results).\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716124931.5870-2-chris@chris-wilson.co.uk\n\ncommit e362cc6a247551ead1b3121521ac7a11d397db81\nAuthor: Douglas Anderson \u003cdianders@chromium.org\u003e\nDate: Fri Jul 12 09:33:33 2019 -0700\n\n drm/panel: simple: document panel_desc; rename a few functions\n \n This attempts to address outstanding review feedback from\n commit b8a2948fa2b3 (\"drm/panel: simple: Add ability to override typical timing\")\n \n Specifically:\n \n * It was requested that I document (in the structure definition) that\n the device tree override had no effect if \u0027struct drm_display_mode\u0027\n was used in the panel description. I have provided full Doxygen\n comments for \u0027struct panel_desc\u0027 to accomplish that.\n * panel_simple_get_fixed_modes() was thought to be a confusing name,\n so it has been renamed to panel_simple_get_display_modes().\n * panel_simple_parse_override_mode() was thought to be better named as\n panel_simple_parse_panel_timing_node().\n \n Suggested-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712163333.231884-1-dianders@chromium.org\n\ncommit 57a1b0893782090738a879293efeb93885e0519c\nAuthor: Sean Paul \u003cseanpaul@chromium.org\u003e\nDate: Wed Jul 17 12:01:48 2019 -0400\n\n drm: Make the bw/link rate calculations more forgiving\n \n Although the DisplayPort spec explicitly calls out the 1.62/2.7/5.4/8.1\n link rates, the value of LINK_BW_SET is calculated. The DisplayPort\n spec says \"Main-Link Bandwidth Setting \u003d Value x 0.27Gbps/lane\".\n \n A bridge that we\u0027re looking to upstream uses 6.75Gbps rate (value 0x19)\n [1], and that precludes it from using these functions.\n \n This 6.75Gbps rate is defined in the spec as (credit to Ville for posting this):\n A MyDP Source device, upon reading the MAX_LINK_RATE register of the\n downstream DPRX programmed to 19h (which can be the case only for a\n MyDP-to-Legacy or MyDP-to-DP lane count converter) can program the\n LINK_BW_SET register (DPCD Address 00100h) to 19h to enable 6.75Gbps/lane.\"\n \n So to avoid failing on legitimate rates in the future, this patch calculates thevalues according to spec instead of restricting these values to one of the\n DP_LINK_BW_* #defines.\n \n No functional change for the well-defined values, but we lose the\n warning (and return the correct value) for ill-defined bw values.\n \n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n \n [1] https://chromium-review.googlesource.com/c/chromiumos/third_party/kernel/+/1689251/2/drivers/gpu/drm/bridge/analogix/anx7625.c#636\n \n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190717160148.256826-1-sean@poorly.run\n\ncommit f293d4d4649cc20068a312cfc5752ce6aab0126b\nAuthor: Yoshinori Sato \u003cysato@users.sourceforge.jp\u003e\nDate: Wed Jul 17 22:55:00 2019 +0900\n\n h8300: Add missing symbol \"BOOT_LINK_OFFSET\"\n \n Signed-off-by: Yoshinori Sato \u003cysato@users.sourceforge.jp\u003e\n\ncommit b4bff4be5503c1e0e7542809fe89f97e3b6dc98a\nAuthor: Yoshinori Sato \u003cysato@users.sourceforge.jp\u003e\nDate: Wed Jul 17 22:49:08 2019 +0900\n\n h8300: fix memset return value.\n \n The return address is not as specified.\n \n Signed-off-by: Yoshinori Sato \u003cysato@users.sourceforge.jp\u003e\n\ncommit fff8102aaed59014cb2d8034bdca231185496b16\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jul 16 13:49:29 2019 +0100\n\n drm/i915/execlists: Process interrupted context on reset\n \n By stopping the rings, we may trigger an arbitration point resulting in\n a premature context-switch (i.e. a completion event before the request\n is actually complete). This clears the active context before the reset,\n but we must remember to rewind the incomplete context for replay upon\n resume.\n \n Fixes: 1863e3020ab5 (\"drm/i915/execlists: Always reset the context\u0027s RING registers\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716124931.5870-3-chris@chris-wilson.co.uk\n\ncommit 0629d4da1f159778063767fb0ac1c951034c5477\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Wed Jul 17 11:44:18 2019 +0100\n\n drm/i915: Update description of i915.enable_guc modparam\n \n Commit f774f0964919 (\"drm/i915/guc: Turn on GuC/HuC auto mode\") changed\n the default from 0 to -1 but forgot to update the description.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Fixes: f774f0964919 (\"drm/i915/guc: Turn on GuC/HuC auto mode\")\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@linux.intel.com\u003e\n Cc: intel-gfx@lists.freedesktop.org\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190717104418.23809-1-tvrtko.ursulin@linux.intel.com\n\ncommit 9aef5867c86c7bfad92f4208f2ad673de359f51b\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 16 08:42:20 2019 +0200\n\n drm/mediatek: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n \n While touching the include files divide them up in blocks\n in the typical order:\n \n \\#include \u003clinux/*\u003e\n \n \\#include \u003cvideo/*\u003e\n \n \\#include \u003cdrm/*\u003e\n \n \\#include \"\"\n \n And sort the includes in the blocks\n Add the necessary includes to fix build after removal of drmP.h\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: CK Hu \u003cck.hu@mediatek.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: linux-mediatek@lists.infradead.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716064220.18157-20-sam@ravnborg.org\n\ncommit c2156ccd9657c9962d2b9f2322976785a070f02f\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 16 08:42:19 2019 +0200\n\n drm/rockchip: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n \n While touching the list of include files move the\n blocks so they follow the common pattern:\n \n \\#include \u003clinux/*\u003e\n \n \\#include \u003cvideo/*\u003e\n \n \\#include \u003cdrm/*\u003e\n \n \\#include \"\"\n \n Within each block sort the include files.\n Add the includes needed to fix build after the removal of drmP.h.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Sandy Huang \u003chjc@rock-chips.com\u003e\n Cc: \"Heiko Stübner\" \u003cheiko@sntech.de\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: linux-rockchip@lists.infradead.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716064220.18157-19-sam@ravnborg.org\n\ncommit 05f0940b4de0a3e3ac1c9c0fec30b650fb73be63\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 16 08:42:18 2019 +0200\n\n drm/imx: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n While touching the include files divide them in blocks and sort the\n include files within each block.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Shawn Guo \u003cshawnguo@kernel.org\u003e\n Cc: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\n Cc: Pengutronix Kernel Team \u003ckernel@pengutronix.de\u003e\n Cc: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Cc: NXP Linux Team \u003clinux-imx@nxp.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716064220.18157-18-sam@ravnborg.org\n\ncommit 71466ffd7322cb24901a817c563c6e2d2c20fb2d\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 16 08:42:17 2019 +0200\n\n drm/i2c/sil164: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n Fix fallout.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Francisco Jerez \u003ccurrojerez@riseup.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716064220.18157-17-sam@ravnborg.org\n\ncommit 758380bd5b6dee884fa5a096cb74c24c501dc709\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 16 08:42:16 2019 +0200\n\n drm/i2c/ch7006: drop use of drmP.h\n \n Drop using the deprecated drmP.h header file.\n Fix fallout.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716064220.18157-16-sam@ravnborg.org\n\ncommit fcb57664172edb25b2fdabb775b13662275abc5b\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 16 08:42:15 2019 +0200\n\n drm/tilcdc: drop use of drmP.h\n \n Dropped drmP.h and all other header files not used by tilcdc_drv.h.\n Added the minimal includes and forwards to make the header file\n self-contained.\n \n Then dropped the remaining uses of drmP.h and fixed all fall-out.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Jyri Sarha \u003cjsarha@ti.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716064220.18157-15-sam@ravnborg.org\n\ncommit 2175749b3671ac4171177cba69c0c94b92fa8e81\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 16 08:42:14 2019 +0200\n\n drm/i810: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header.\n Replace with necessary includes.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716064220.18157-14-sam@ravnborg.org\n\ncommit 7b42a2f4e4b850ba9069a3927c50754da8aceb51\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 16 08:42:13 2019 +0200\n\n drm/vgem: drop use of drmP.h\n \n Drop use of the deprecated header drmP.h.\n Replace with necessary includes in the individual .c files.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Deepak Sharma \u003cdeepak.sharma@amd.com\u003e\n Cc: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716064220.18157-13-sam@ravnborg.org\n\ncommit 53fa84b2fabfeb67bbc2f8f21b137ded0ae0bc43\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 16 08:42:12 2019 +0200\n\n drm/tdfx: drop use of drmP.h\n \n Replace drmP.h include with necessary include files.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716064220.18157-12-sam@ravnborg.org\n\ncommit 54ccae0c08929f4a2da0a1a2b0a86e744579ef8b\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 16 08:42:11 2019 +0200\n\n drm/selftests: drop use of drmP.h\n \n Drop the single user of drmP.h - replace it with relevant includes.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Alexandru-Cosmin Gheorghe \u003cAlexandru-Cosmin.Gheorghe@arm.com\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716064220.18157-11-sam@ravnborg.org\n\ncommit 81f6156ca5aea59640cc3a11565e10b0c4cac60b\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 16 08:42:10 2019 +0200\n\n drm/omapdrm: drop use of drmP.h\n \n Drop use of the deprecated header drmP.h.\n Rearranged list of include files to match rest of\n DRM too.\n The drmP.h file was deleted from the header file, and the necessary\n includes was added to the .c files to fix build.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716064220.18157-10-sam@ravnborg.org\n\ncommit a9dcf3804ac832453b32ab99750a207020816ad4\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 16 08:42:09 2019 +0200\n\n drm/udl: drop use of drmP.h\n \n The drmP.h header file is deprecated.\n Drop it from all files in the udl driver.\n \n Made the header files self contained, which then\n made it simpler to update the .c files.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Cc: Robert Tarasov \u003ctutankhamen@chromium.org\u003e\n Cc: Oleksandr Andrushchenko \u003coleksandr_andrushchenko@epam.com\u003e\n Cc: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Cc: \"Noralf Trønnes\" \u003cnoralf@tronnes.org\u003e\n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: Mikulas Patocka \u003cmpatocka@redhat.com\u003e\n Cc: Emil Lundmark \u003clndmrk@chromium.org\u003e\n Cc: Eugeniy Paltsev \u003cEugeniy.Paltsev@synopsys.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716064220.18157-9-sam@ravnborg.org\n\ncommit 9e28e0ce2d41525c8200619cd7a842ebd95cc1ac\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 16 08:42:08 2019 +0200\n\n drm/r128: drop use of drmP.h\n \n Replace use of the deprecated drmP.h header file\n with relevant includes.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716064220.18157-8-sam@ravnborg.org\n\ncommit fd6d6d80cb5f24a7fec65cc24633659db79708bd\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 16 08:42:07 2019 +0200\n\n drm/vc4: drop use of drmP.h\n \n Drop use of the deprecated header drmP.h.\n Fix so vc4_drv.h is now self-contained, and fixed fall-out in remaining\n files.\n \n Divided include files in blocks.\n Sorted include files within their blocks.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716064220.18157-7-sam@ravnborg.org\n\ncommit 9c25a297664518b36933dae0887cbea3008fb840\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 16 08:42:06 2019 +0200\n\n drm/sun4i: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n \n While touching the list of include file, use the typical order of the\n blocks:\n \\#include \u003clinux/*\u003e\n \n \\#include \u003cvideo/*\u003e\n \n \\#include \u003cdrm/*\u003e\n \n \\#include \"\"\n \n Within each block, sort the files.\n Include necessary files to fix build after the drmP.h removal.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Chen-Yu Tsai \u003cwens@csie.org\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716064220.18157-6-sam@ravnborg.org\n\ncommit 647d295b70480cf9f3e2c2dfe87547c8d58498f1\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 16 08:42:05 2019 +0200\n\n drm/zte: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n Fix fallout.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Cc: Shawn Guo \u003cshawnguo@kernel.org\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716064220.18157-5-sam@ravnborg.org\n\ncommit 09a56fbe8a06cb5ac184f4bbc8833b5f5ae90e05\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 16 08:42:04 2019 +0200\n\n drm/pl111: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header.\n Sort includes in blocks while touching the files.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716064220.18157-4-sam@ravnborg.org\n\ncommit 220989e7097a5cc083624dc1c925c1c255247574\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 16 08:42:03 2019 +0200\n\n drm/v3d: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n Made v3d_drv.h self-contained with only sufficient\n include files.\n Fixed fallout in remaining files.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716064220.18157-3-sam@ravnborg.org\n\ncommit 66620f48cb59440fd265b77d30b9960e2cf1c345\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Tue Jul 16 08:42:02 2019 +0200\n\n drm/meson: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header.\n While doing so used the opportunity\n to clean up a little so includes are now\n sorted and removed unused include files.\n In a few cases added some forwards to allow header\n files to built in different include order.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Cc: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Cc: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Cc: Maxime Jourdan \u003cmjourdan@baylibre.com\u003e\n Cc: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Cc: linux-amlogic@lists.infradead.org\n Cc: linux-arm-kernel@lists.infradead.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716064220.18157-2-sam@ravnborg.org\n\ncommit 22d6552f827ef76ade3edf6bbb3f05048a0a7d8b\nAuthor: Nicolas Dichtel \u003cnicolas.dichtel@6wind.com\u003e\nDate: Mon Jul 15 12:00:23 2019 +0200\n\n xfrm interface: fix management of phydev\n \n With the current implementation, phydev cannot be removed:\n \n $ ip link add dummy type dummy\n $ ip link add xfrm1 type xfrm dev dummy if_id 1\n $ ip l d dummy\n kernel:[77938.465445] unregister_netdevice: waiting for dummy to become free. Usage count \u003d 1\n \n Manage it like in ip tunnels, ie just keep the ifindex. Not that the side\n effect, is that the phydev is now optional.\n \n Fixes: f203b76d7809 (\"xfrm: Add virtual xfrm interfaces\")\n Signed-off-by: Nicolas Dichtel \u003cnicolas.dichtel@6wind.com\u003e\n Tested-by: Julien Floret \u003cjulien.floret@6wind.com\u003e\n Signed-off-by: Steffen Klassert \u003csteffen.klassert@secunet.com\u003e\n\ncommit c5d1030f23002430c2a336b2b629b9d6f72b3564\nAuthor: Nicolas Dichtel \u003cnicolas.dichtel@6wind.com\u003e\nDate: Mon Jul 15 12:00:22 2019 +0200\n\n xfrm interface: fix list corruption for x-netns\n \n dev_net(dev) is the netns of the device and xi-\u003enet is the link netns,\n where the device has been linked.\n changelink() must operate in the link netns to avoid a corruption of\n the xfrm lists.\n \n Note that xi-\u003enet and dev_net(xi-\u003ephysdev) are always the same.\n \n Before the patch, the xfrmi lists may be corrupted and can later trigger a\n kernel panic.\n \n Fixes: f203b76d7809 (\"xfrm: Add virtual xfrm interfaces\")\n Reported-by: Julien Floret \u003cjulien.floret@6wind.com\u003e\n Signed-off-by: Nicolas Dichtel \u003cnicolas.dichtel@6wind.com\u003e\n Tested-by: Julien Floret \u003cjulien.floret@6wind.com\u003e\n Signed-off-by: Steffen Klassert \u003csteffen.klassert@secunet.com\u003e\n\ncommit e0aaa332e6a97dae57ad59cdb19e21f83c3d081c\nAuthor: Nicolas Dichtel \u003cnicolas.dichtel@6wind.com\u003e\nDate: Mon Jul 15 12:00:21 2019 +0200\n\n xfrm interface: ifname may be wrong in logs\n \n The ifname is copied when the interface is created, but is never updated\n later. In fact, this property is used only in one error message, where the\n netdevice pointer is available, thus let\u0027s use it.\n \n Fixes: f203b76d7809 (\"xfrm: Add virtual xfrm interfaces\")\n Signed-off-by: Nicolas Dichtel \u003cnicolas.dichtel@6wind.com\u003e\n Signed-off-by: Steffen Klassert \u003csteffen.klassert@secunet.com\u003e\n\ncommit e9e7e85d75f3731079ffd77c1a66f037aef04fe7\nAuthor: Nicolas Dichtel \u003cnicolas.dichtel@6wind.com\u003e\nDate: Mon Jul 15 12:00:20 2019 +0200\n\n xfrm interface: avoid corruption on changelink\n \n The new parameters must not be stored in the netdev_priv() before\n validation, it may corrupt the interface. Note also that if data is NULL,\n only a memset() is done.\n \n $ ip link add xfrm1 type xfrm dev lo if_id 1\n $ ip link add xfrm2 type xfrm dev lo if_id 2\n $ ip link set xfrm1 type xfrm dev lo if_id 2\n RTNETLINK answers: File exists\n $ ip -d link list dev xfrm1\n 5: xfrm1@lo: \u003cNOARP\u003e mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000\n link/none 00:00:00:00:00:00 brd 00:00:00:00:00:00 promiscuity 0 minmtu 68 maxmtu 1500\n xfrm if_id 0x2 addrgenmode eui64 numtxqueues 1 numrxqueues 1 gso_max_size 65536 gso_max_segs 65535\n \n \u003d\u003e \"if_id 0x2\"\n \n Fixes: f203b76d7809 (\"xfrm: Add virtual xfrm interfaces\")\n Signed-off-by: Nicolas Dichtel \u003cnicolas.dichtel@6wind.com\u003e\n Tested-by: Julien Floret \u003cjulien.floret@6wind.com\u003e\n Signed-off-by: Steffen Klassert \u003csteffen.klassert@secunet.com\u003e\n\ncommit a9877da2d62958972548e98ecad0413d482b5eac\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jul 16 22:34:43 2019 +0100\n\n drm/i915/oa: Reconfigure contexts on the fly\n \n Avoid a global idle barrier by reconfiguring each context by rewriting\n them with MI_STORE_DWORD from the kernel context.\n \n v2: We only need to determine the desired register values once, they are\n the same for all contexts.\n v3: Don\u0027t remove the kernel context from the list of known GEM contexts;\n the world is not ready for that yet.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190716213443.9874-1-chris@chris-wilson.co.uk\n\ncommit 30fe7b07f6cb7f391181eb9d2ea2687cd9bcf395\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 12 09:03:14 2019 +0100\n\n dma-buf: Relax the write-seqlock for reallocating the shared fence list\n \n As the set of shared fences is not being changed during reallocation of\n the reservation list, we can skip updating the write_seqlock.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712080314.21018-2-chris@chris-wilson.co.uk\n\ncommit bea5faf7be325bba63d82b5d37e31b43140ddabf\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 12 12:27:25 2019 +0100\n\n drm/i915/gtt: Tidy up ppgtt insertion for gen8\n \n Apply the new radix shift helpers to extract the multi-level indices\n cleanly when inserting pte into the gtt tree.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Abdiel Janulgue \u003cabdiel.janulgue@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712112725.2892-5-chris@chris-wilson.co.uk\n\ncommit 8a98e839601f83fc1b31963224d9e1279e5d122f\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 12 12:27:24 2019 +0100\n\n drm/i915/gtt: Recursive ppgtt alloc for gen8\n \n Refactor the separate allocation routines into a single recursive\n function.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Abdiel Janulgue \u003cabdiel.janulgue@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712112725.2892-4-chris@chris-wilson.co.uk\n\ncommit 09975b861aa0c5cc0f1a6f9543807f344c4c12ae\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jul 9 10:12:33 2019 +0100\n\n drm/i915/execlists: Disable preemption under GVT\n \n Preempt-to-busy uses a GPU semaphore to enforce an idle-barrier across\n preemption, but mediated gvt does not fully support semaphores.\n \n v2: Fiddle around with the flags and settle on using has-semaphores for\n the core bits so that we retain the ability to preempt our own\n semaphores.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n Cc: Xiaolin Zhang \u003cxiaolin.zhang@intel.com\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Acked-by: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190709091233.8573-1-chris@chris-wilson.co.uk\n\ncommit cfe7288c276e359eebf057699fe86c2f8af14224\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Jul 15 09:09:28 2019 +0100\n\n drm/i915: Lock the engine while dumping the active request\n \n We cannot let the request be retired and freed while we are trying to\n dump it during error capture. It is not sufficient just to grab a\n reference to the request, as during retirement we may free the ring\n which we are also dumping. So take the engine lock to prevent retiring\n and freeing of the request.\n \n Reported-by: Alex Shumsky \u003calexthreed@gmail.com\u003e\n Fixes: 83c317832eb1 (\"drm/i915: Dump the ringbuffer of the active request for debugging\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Alex Shumsky \u003calexthreed@gmail.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190715080946.15593-6-chris@chris-wilson.co.uk\n\ncommit 0be0bfd2de9dfdd2098a9c5b14bdd8f739c9165d\nAuthor: Amir Goldstein \u003camir73il@gmail.com\u003e\nDate: Fri Jul 12 15:24:34 2019 +0300\n\n ovl: fix regression caused by overlapping layers detection\n \n Once upon a time, commit 2cac0c00a6cd (\"ovl: get exclusive ownership on\n upper/work dirs\") in v4.13 added some sanity checks on overlayfs layers.\n This change caused a docker regression. The root cause was mount leaks\n by docker, which as far as I know, still exist.\n \n To mitigate the regression, commit 85fdee1eef1a (\"ovl: fix regression\n caused by exclusive upper/work dir protection\") in v4.14 turned the\n mount errors into warnings for the default index\u003doff configuration.\n \n Recently, commit 146d62e5a586 (\"ovl: detect overlapping layers\") in\n v5.2, re-introduced exclusive upper/work dir checks regardless of\n index\u003doff configuration.\n \n This changes the status quo and mount leak related bug reports have\n started to re-surface. Restore the status quo to fix the regressions.\n To clarify, index\u003doff does NOT relax overlapping layers check for this\n ovelayfs mount. index\u003doff only relaxes exclusive upper/work dir checks\n with another overlayfs mount.\n \n To cover the part of overlapping layers detection that used the\n exclusive upper/work dir checks to detect overlap with self upper/work\n dir, add a trap also on the work base dir.\n \n Link: https://github.com/moby/moby/issues/34672\n Link: https://lore.kernel.org/linux-fsdevel/20171006121405.GA32700@veci.piliscsaba.szeredi.hu/\n Link: https://github.com/containers/libpod/issues/3540\n Fixes: 146d62e5a586 (\"ovl: detect overlapping layers\")\n Cc: \u003cstable@vger.kernel.org\u003e # v4.19+\n Signed-off-by: Amir Goldstein \u003camir73il@gmail.com\u003e\n Tested-by: Colin Walters \u003cwalters@verbum.org\u003e\n Signed-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n\ncommit 45b2fda3d82d686cc014e4c485332f85e4bd64de\nAuthor: Qian Cai \u003ccai@lca.pw\u003e\nDate: Mon Jul 15 09:42:53 2019 -0400\n\n gpu/drm: fix a few kernel-doc \"/**\" mark warnings\n \n The opening comment mark \"/**\" is reserved for kernel-doc comments, so\n it will generate warnings for comments that are not kernel-doc with\n \"make W\u003d1\". For example,\n \n drivers/gpu/drm/drm_memory.c:2: warning: Cannot understand * \\file\n drm_memory.c\n \n Signed-off-by: Qian Cai \u003ccai@lca.pw\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/1563198173-7317-1-git-send-email-cai@lca.pw\n\ncommit 66a639cbe36c830d2bd2dd89fd10b60f67ffd40e\nAuthor: Rob Clark \u003crobdclark@chromium.org\u003e\nDate: Sat Jul 6 13:31:02 2019 -0700\n\n drm/bridge: ti-sn65dsi86: use dev name for debugfs\n \n This should be more future-proof if we ever encounter a device with two\n of these bridges.\n \n Suggested-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n Reviewed-by: Bjorn Andersson \u003cbjorn.andersson@linaro.org\u003e\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190706203105.7810-1-robdclark@gmail.com\n\ncommit 0a869e8d82efa43aaa09ff3c145cc98b01cd8a7c\nAuthor: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\nDate: Sat Jul 6 18:57:42 2019 +0530\n\n drm/pl111: pl111_vexpress.c: Add of_node_put() before return\n \n Each iteration of for_each_available_child_of_node puts the previous\n node, but in the case of a break from the middle of the loop there is\n no put, thus causing a memory leak. Hence add an of_node_put before the\n break.\n Issue found with Coccinelle.\n \n Signed-off-by: Nishka Dasgupta \u003cnishkadg.linux@gmail.com\u003e\n Signed-off-by: Eric Anholt \u003ceric@anholt.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190706132742.3250-1-nishkadg.linux@gmail.com\n\ncommit bb80c9255770fa1ed54e889a6bee628bdd0f6762\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Thu Jul 11 17:53:43 2019 -0700\n\n drm/i915: Enable hotplug retry\n \n Right now we are aware of two cases that needs another hotplug retry:\n - Unpowered type-c dongles\n - HDMI slow unplug\n \n Both have a complete explanation in the code to schedule another run\n of the hotplug handler.\n \n It could have more checks to just trigger the retry in those two\n specific cases but why would sink signal a long pulse if there is\n no change? Also the drawback of running the hotplug handler again\n is really low and that could fix another cases that we are not\n aware.\n \n Also retrying for old DP ports(non-DDI) to make it consistent and not\n cause CI failures if those systems are connected to chamelium boards\n that will be used to simulate the issues reported in here.\n \n v2: Also retrying for old DP ports(non-DDI)(Imre)\n \n v4: Renamed INTEL_HOTPLUG_NOCHANGE to INTEL_HOTPLUG_UNCHANGED to keep\n it consistent(Rodrigo)\n \n Tested-by: Timo Aaltonen \u003ctjaalton@ubuntu.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Imre Deak \u003cimre.deak@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Reviewed-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712005343.24571-2-jose.souza@intel.com\n\ncommit 3944709df8e9298225fc2b29e53ee8e6f4b26618\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Thu Jul 11 17:53:42 2019 -0700\n\n drm/i915: Add support for retrying hotplug\n \n There is some scenarios that we are aware that sink probe can fail,\n so lets add the infrastructure to let hotplug() hook to request\n another probe after some time.\n \n v2: Handle shared HPD pins (Imre)\n v3: Rebased\n v4: Renamed INTEL_HOTPLUG_NOCHANGE to INTEL_HOTPLUG_UNCHANGED to keep\n it consistent(Rodrigo)\n v5: Making the working queue used explicit through all the callers to\n hotplug_work (Ville)\n \n Tested-by: Timo Aaltonen \u003ctjaalton@ubuntu.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712005343.24571-1-jose.souza@intel.com\n\ncommit 48cf0a1ec1701787945008c945aa35bb6cfacb70\nAuthor: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\nDate: Fri Jul 12 15:16:41 2019 -0700\n\n drm/i915/ehl: Map MCC pins based on PHY, not port\n \n Now that we distinguish between phy and port(ddi), mcc_port_to_ddc_pin\n should use the phy, not the DDI, for determining DDC pins.\n \n We\u0027re only converting the MCC function at the moment since EHL is the\n only platform that has configurations where port!\u003dphy.\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712221641.21031-1-matthew.d.roper@intel.com\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n\ncommit 71866a56bc7594068d894bacd9ac957878c6816a\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 30 08:19:03 2019 +0200\n\n drm/atmel_hlcdc: drop use of drmP.h\n \n Drop use of the deprecated header drmP.h.\n Make header file self-contained, with only the required set\n of include files.\n And fixed fallout in remaining files.\n Divide include files in blocks and sort them within each block.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Nicolas Ferre \u003cnicolas.ferre@microchip.com\u003e\n Cc: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Cc: Ludovic Desroches \u003cludovic.desroches@microchip.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190630061922.7254-15-sam@ravnborg.org\n\ncommit a803bf0ade8d266b950b21daab3748e77a334230\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Mon Jul 15 11:02:27 2019 +0200\n\n drm/shmobile: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n Fix fallout.\n \n v2:\n - Add additional include/forward to shmob_drm_crtc.h\n to avoid relying on indirect includes (Laurent)\n - Add additional includes to shmob_drm_regs.h to make\n it self-contained\n - Add additional includes to shmob_drm_plane.h to make\n it self-contained\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: linux-renesas-soc@vger.kernel.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190715090227.GA27652@ravnborg.org\n\ncommit 40b4db43f7e449ef3414642537c9cee2de4b5371\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 30 08:19:22 2019 +0200\n\n drm/hisilicon: drop use of drmP.h\n \n Drop the deprecated drmP.h header file.\n \n Made the header file selfcontained, and dropped unused header files.\n Fixed fallout in remaining files.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: Rongrong Zou \u003czourongrong@gmail.com\u003e\n Cc: Xinwei Kong \u003ckong.kongxinwei@hisilicon.com\u003e\n Cc: Chen Feng \u003cpuck.chen@hisilicon.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Cc: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Allison Randal \u003callison@lohutok.net\u003e\n Cc: Harry Wentland \u003charry.wentland@amd.com\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Souptick Joarder \u003cjrdr.linux@gmail.com\u003e\n Cc: John Garry \u003cjohn.garry@huawei.com\u003e\n Cc: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Cc: \"Christian König\" \u003cchristian.koenig@amd.com\u003e\n Cc: Junwei Zhang \u003cJerry.Zhang@amd.com\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Hans de Goede \u003chdegoede@redhat.com\u003e\n Cc: CK Hu \u003cck.hu@mediatek.com\u003e\n Cc: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: \"Noralf Trønnes\" \u003cnoralf@tronnes.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190630061922.7254-34-sam@ravnborg.org\n\ncommit fbbbd1608b0d23bdc436ea572ace98ec2d663abb\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 30 08:19:21 2019 +0200\n\n drm/ast: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n While touching the files divide include files in blocks\n and sort the include files in the individual blocks.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Dave Airlie \u003cairlied@redhat.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Cc: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: \"Christian König\" \u003cchristian.koenig@amd.com\u003e\n Cc: Sam Bobroff \u003csbobroff@linux.ibm.com\u003e\n Cc: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Cc: Huang Rui \u003cray.huang@amd.com\u003e\n Cc: \"Y.C. Chen\" \u003cyc_chen@aspeedtech.com\u003e\n Cc: Hans de Goede \u003chdegoede@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190630061922.7254-33-sam@ravnborg.org\n\ncommit 016f363ee11ce804ef6106fd737af63cb61fff3e\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 30 08:19:20 2019 +0200\n\n drm/bochs: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n Made bochs.h self-contained and then fixed\n fallout in remaining files.\n Several unused includes was dropped in the process.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: virtualization@lists.linux-foundation.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190630061922.7254-32-sam@ravnborg.org\n\ncommit 322d0c84d30e722fb46c2a85956a57a429673c0d\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 30 08:19:19 2019 +0200\n\n drm: add missing include to drm_vram_mm_helper.h\n \n The macro DRM_VRAM_MM_FILE_OPERATIONS referencs\n functions declared in other header files.\n Include these header files so this header files\n pulls in what it references.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190630061922.7254-31-sam@ravnborg.org\n\ncommit a3d639776e601ddc1ee4e23fc6638c421377ae93\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 30 08:19:16 2019 +0200\n\n drm/virtgpu: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n Fix fallout by adding missing include files.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: virtualization@lists.linux-foundation.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190630061922.7254-28-sam@ravnborg.org\n\ncommit 7c1be93c8e983077b8ec964497f09f2bfb6e7b29\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 30 08:19:14 2019 +0200\n\n drm/scheduler: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n Fix fallout.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Cc: Andrey Grodzovsky \u003candrey.grodzovsky@amd.com\u003e\n Cc: Huang Rui \u003cray.huang@amd.com\u003e\n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: Bas Nieuwenhuizen \u003cbas@basnieuwenhuizen.nl\u003e\n Cc: Sharat Masetty \u003csmasetty@codeaurora.org\u003e\n Cc: Nathan Chancellor \u003cnatechancellor@gmail.com\u003e\n Cc: Nayan Deshmukh \u003cnayan26deshmukh@gmail.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190630061922.7254-26-sam@ravnborg.org\n\ncommit ce672a1b21a8015c5a8de3a656d4c2edc22663b0\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 30 08:19:01 2019 +0200\n\n drm/vkms: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header.\n Replace it with the necessary includes in the individual .c files.\n The header files was self-contained, and extra includes were not added\n there.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Haneen Mohammed \u003chamohammed.sa@gmail.com\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190630061922.7254-13-sam@ravnborg.org\n\ncommit c0f4b75c065ba2c6b9610f4526da9bf432049b9e\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 30 08:18:58 2019 +0200\n\n drm/qxl: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n While touching the files divided includes in blocks,\n and when needed sort the blocks.\n Fix fallout.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Dave Airlie \u003cairlied@redhat.com\u003e\n Cc: virtualization@lists.linux-foundation.org\n Cc: spice-devel@lists.freedesktop.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190630061922.7254-10-sam@ravnborg.org\n\ncommit b4b21c83958fc9880cd6adf03d9eb1e5fc98fa47\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 30 08:18:57 2019 +0200\n\n drm/fsl-dcu: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n Fix fallout.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Stefan Agner \u003cstefan@agner.ch\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Alison Wang \u003calison.wang@nxp.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190630061922.7254-9-sam@ravnborg.org\n\ncommit d5742c6cfa29d521a93f3fb3a7a9ab1e79d323d0\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 30 08:18:54 2019 +0200\n\n drm/mxsfb: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n \n While touching the list of include files divided them\n in blocks and sort them within each block.\n Fixed fallout in the relevant files.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Stefan Agner \u003cstefan@agner.ch\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Marek Vasut \u003cmarex@denx.de\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Shawn Guo \u003cshawnguo@kernel.org\u003e\n Cc: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\n Cc: Pengutronix Kernel Team \u003ckernel@pengutronix.de\u003e\n Cc: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Cc: NXP Linux Team \u003clinux-imx@nxp.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190630061922.7254-6-sam@ravnborg.org\n\ncommit 91328ebe4f6f12eaa7754b4ac01b308cf68538d0\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 30 08:18:53 2019 +0200\n\n drm/tve200: drop use of drmP.h\n \n Drop use of the deprecated header drmP.h.\n \n Fix so header file became self-contained,\n and then fixed fallout in the other files.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190630061922.7254-5-sam@ravnborg.org\n\ncommit 2ea2269e296a0c15387b40362410c1569e77502c\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 30 08:18:51 2019 +0200\n\n drm/xen: drop use of drmP.h\n \n The drmP.h header is deprecated. Drop all uses.\n Added includes/forwards to the header files and\n then fixed fallout in the .c files.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Oleksandr Andrushchenko \u003coleksandr_andrushchenko@epam.com\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: xen-devel@lists.xenproject.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190630061922.7254-3-sam@ravnborg.org\n\ncommit c3bbb48e0c548badb71d8339d95974febb33991d\nAuthor: Maya Rashish \u003ccoypu@sdf.org\u003e\nDate: Wed Jul 10 16:17:25 2019 +0000\n\n drm/agp: Remove unused function drm_agp_bind_pages\n \n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Maya Rashish \u003ccoypu@sdf.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190710161725.GA5042@SDF.ORG\n\ncommit 94eb1e10a34d3c7fc42208faaa4954fe482ac091\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 12 09:03:13 2019 +0100\n\n dma-buf: Expand reservation_list to fill allocation\n \n Since kmalloc() will round up the allocation to the next slab size or\n page, it will normally return a pointer to a memory block bigger than we\n asked for. We can query for the actual size of the allocated block using\n ksize() and expand our variable size reservation_list to take advantage\n of that extra space.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Christian König \u003cchristian.koenig@amd.com\u003e\n Cc: Michel Dänzer \u003cmichel.daenzer@amd.com\u003e\n Reviewed-by: Michel Dänzer \u003cmichel.daenzer@amd.com\u003e\n Reviewed-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712080314.21018-1-chris@chris-wilson.co.uk\n\ncommit 506927ec8bbbdd99261cec4cd28cd0fd54e02218\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 12 09:25:49 2019 +0100\n\n drm/i915/selftests: Ignore self-preemption suppression under gvt\n \n GVT forces single port submission of individual requests. We do not\n enjoy the context amalgamation that the test depends upon for setting up\n the test (where port 0 has a large number of requests with a priority\n change somewhere in the middle). Under single request submission of gvt\n it is quite able for the preemption event to occur while another context\n is active and so there be a real need to act upon that preemption.\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111108\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n Acked-by: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712082549.25053-1-chris@chris-wilson.co.uk\n\ncommit a8120bc23eeae8f09a49167e32fa79ce5336546e\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Sat Jul 13 11:00:16 2019 +0100\n\n drm/i915/uc: kill \u003cg,h\u003euc_to_i915\n \n Get rid of them to avoid more users being added while the guc code\n transitions to use gt more than i915.\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Acked-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190713100016.8026-11-chris@chris-wilson.co.uk\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n\ncommit 2239e6dff2067c23f0afb7fab62ef139dc957d48\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Sat Jul 13 11:00:15 2019 +0100\n\n drm/i915/guc: prefer intel_gt in guc interrupt functions\n \n We can get rid of a few more guc_to_i915 and start compartmentalizing\n interrupt management a bit more. We should be able to move more code in\n the future once the gt_pm code is also moved across to gt.\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190713100016.8026-10-chris@chris-wilson.co.uk\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n\ncommit 84b1ca2f0e68e89186f552ab839dc8da7c3ddc87\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Sat Jul 13 11:00:14 2019 +0100\n\n drm/i915/uc: prefer intel_gt over i915 in GuC/HuC paths\n \n With our HW interface logic moving from i915 to gt and with GuC and HuC\n being part of the gt HW, it makes sense to use the intel_gt structure\n instead of i915 as our reference object in GuC/HuC paths.\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190713100016.8026-9-chris@chris-wilson.co.uk\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n\ncommit ca7b2c1bbede6187342dd0bec8666ab4481fa6ad\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Sat Jul 13 11:00:13 2019 +0100\n\n drm/i915/uc: Move intel functions to intel_uc\n \n All the intel_uc_* can now be moved to work on the intel_uc structure\n for better encapsulation of uc-related actions.\n \n Note: I\u0027ve introduced uc_to_gt instead of uc_to_i915 because the aim is\n to move everything to be gt-focused in the medium term, so we would\u0027ve\n had to replace it soon anyway.\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Acked-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190713100016.8026-8-chris@chris-wilson.co.uk\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n\ncommit 8b5689d7e3ca889a7e55c79bc335b33e3f170a18\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Sat Jul 13 11:00:12 2019 +0100\n\n drm/i915/uc: move GuC/HuC inside intel_gt under a new intel_uc\n \n Being part of the GT HW, it make sense to keep the guc/huc structures\n inside the GT structure. To help with the encapsulation work done by the\n following patches, both structures are placed inside a new intel_uc\n container. Although this results in code with ugly nested dereferences\n (i915-\u003egt.uc.guc...), it saves us the extra work required in moving\n the structures twice (i915 -\u003e gt -\u003e uc). The following patches will\n reduce the number of places where we try to access the guc/huc\n structures directly from i915 and reduce the ugliness.\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190713100016.8026-7-chris@chris-wilson.co.uk\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n\ncommit 0f261b241d9cd01d58dce85a35df26fd38320b10\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Sat Jul 13 11:00:11 2019 +0100\n\n drm/i915/uc: move GuC and HuC files under gt/uc/\n \n Both microcontrollers are part of the GT HW and are closely related to\n GT operations. To keep all the files cleanly together, they\u0027ve been\n placed in their own subdir inside the gt/ folder\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Acked-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190713100016.8026-6-chris@chris-wilson.co.uk\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n\ncommit 633023a4e61885c0c9ceefed0865ec0063940616\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Sat Jul 13 11:00:10 2019 +0100\n\n drm/i915/guc: unify guc irq handling\n \n The 16-bit guc irq vector is unchanged across gens, the only thing that\n moved is its position (from the upper 16 bits of the PM regs to its own\n register). Instead of duplicating all defines and functions to handle\n the 2 different positions, we can work on the vector and shift it as\n appropriate. While at it, update the handler to work on intel_guc.\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190713100016.8026-5-chris@chris-wilson.co.uk\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n\ncommit 9cbd51c2c0edbafdaab7f0fa7569d1f455113a9b\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Sat Jul 13 11:00:09 2019 +0100\n\n drm/i915/guc: move guc irq functions to intel_guc parameter\n \n No functional change, just moving the guc_to_i915 from the caller into\n the irq function. This will help with the upcoming move of guc under\n intel_gt.\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190713100016.8026-4-chris@chris-wilson.co.uk\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n\ncommit bb2881f8bdde127188bb5c3e6382a6157668d579\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Sat Jul 13 11:00:08 2019 +0100\n\n drm/i915/uc: introduce intel_uc_fw_supported\n \n Instead of always checking in the device config is GuC and HuC are\n supported or not, we can save the state in the uc_fw structure and\n avoid going through i915 every time from the low-level uc management\n code. while at it FIRMWARE_NONE has been renamed to better indicate that\n we haven\u0027t started the fetch/load yet, but we might have already selected\n a blob.\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190713100016.8026-3-chris@chris-wilson.co.uk\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n\ncommit e3f503f1c7331a4f895b97426e7d769a89e864db\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Sat Jul 13 11:00:07 2019 +0100\n\n drm/i915/uc: replace uc init/fini misc\n \n The \"misc\" terminology doesn\u0027t clearly explain what we intend to cover\n in this phase. The only thing we used ot do in there apart from FW fetch\n was initializing the log workqueue, with the latter being required only in\n the very rare case where we enable the log relay. As we no longer create\n our own workqueue, piggybacking on the system_highpri_wq instead, we can\n rename the function to clarify that they only fetch/release the blobs.\n \n v2: only create log wq when needed (Michal), reword commit msg\n accordingly\n v3: after rebase the wq is gone, reword commit msg accordingly\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190713100016.8026-2-chris@chris-wilson.co.uk\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n\ncommit a30eff49d08ca32b6c4e573a4b07c13b4881f2d3\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Sat Jul 13 11:00:06 2019 +0100\n\n drm/i915/guc: Use system workqueue for log capture\n \n We only employ a single task for log capture, and created a workqueue\n for the purpose of ensuring we had a high priority queue for low\n latency. We can simply use the system_highpri_wq and avoid the\n complication with creating our own admist the maze of mutexes.\n (Currently we create the wq early before we even know we need it in\n order to avoid trying to create it on demand while we hold the logging\n mutex.)\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Michał Winiarski \u003cmichal.winiarski@intel.com\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190713100016.8026-1-chris@chris-wilson.co.uk\n\ncommit cb823ed9915b0d4064f3f51e936fbe13c089948a\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 12 20:29:53 2019 +0100\n\n drm/i915/gt: Use intel_gt as the primary object for handling resets\n \n Having taken the first step in encapsulating the functionality by moving\n the related files under gt/, the next step is to start encapsulating by\n passing around the relevant structs rather than the global\n drm_i915_private. In this step, we pass intel_gt to intel_reset.c\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712192953.9187-1-chris@chris-wilson.co.uk\n\ncommit 139ab811511ce46d29dd0a2a861d0ac0f04d2fe6\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Thu Jul 11 22:57:06 2019 -0700\n\n drm/i915/tgl: add modular FIA to device info\n \n Tiger Lake has modular FIA bit indicating if we are using it, so add to\n the device info.\n \n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712055706.12143-3-lucas.demarchi@intel.com\n\ncommit 0caf625777300d318565eb29a71d71a8de2ff1d2\nAuthor: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\nDate: Thu Jul 11 22:57:05 2019 -0700\n\n drm/i915: Add modular FIA\n \n Some platforms may have Modular FIA. If Modular FIA is used in the SOC,\n then Display Driver will access the additional instances of\n FIA based on pre-assigned offset in GTTMADDR space.\n \n Each Modular FIA instance has its own IOSF Sideband Port ID\n and it houses only 2 Type-C Port. In SOC that has more than\n two Type-C Ports, there are multiple instances of Modular FIA.\n Gunit will need to use different destination ID when it access\n different pair of Type-C Port.\n \n The DFLEXDPSP register has Modular FIA bit starting on Tiger Lake. If\n Modular FIA is used in the SOC, this register bit exists in all the\n instances of Modular FIA. IOM FW is required to program only the MF bit\n in first FIA instance that houses the Type-C Port 0 and Port 1, for\n Display Driver to read from.\n \n v2 (Lucas):\n - Move all accesses to FIA to be contained in intel_tc.c, along with\n display_fia that is now called tc_phy_fia\n - Save the fia instance number on intel_digital_port, so we don\u0027t have\n to query if modular FIA is used on every access\n v3 (Lucas): Make function static\n v4 (Lucas): Move enum phy_fia to the header and use it in\n intel_digital_port (suggested by Ville)\n v5 (Lucas): Add comment about the mapping between FIA and TC port\n (suggested by Stuart)\n \n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Signed-off-by: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Acked-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Reviewed-by: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712055706.12143-2-lucas.demarchi@intel.com\n\ncommit 4c2be3c5ebfd98fc588f6bc5f53fa2ef516c02ea\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 12 12:27:23 2019 +0100\n\n drm/i915/gtt: Recursive ppgtt clear for gen8\n \n With an explicit level, we can refactor the separate clear functions\n as a simple recursive function. The additional knowledge of the level\n allows us to spot when we can free an entire subtree at once.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Abdiel Janulgue \u003cabdiel.janulgue@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712112725.2892-3-chris@chris-wilson.co.uk\n\ncommit 1eda701eace2ddac0f2e2e7176ff6b504e815453\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 12 12:27:22 2019 +0100\n\n drm/i915/gtt: Recursive cleanup for gen8\n \n With an explicit level, we can refactor the separate cleanup functions\n as a simple recursive function. We take the opportunity to pass down the\n size of each level so that we can deal with the different sizes of\n top-level and avoid over allocating for 32/36-bit vm.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Abdiel Janulgue \u003cabdiel.janulgue@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712112725.2892-2-chris@chris-wilson.co.uk\n\ncommit 801404632c4ba08adce04cb9e81a1d34bfb14b17\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 12 14:42:34 2019 +0100\n\n drm/i915/display: Drop kerneldoc for \u0027intel_atomic_commit\u0027\n \n intel_atomic_commit() is not for use internally, but only as an entry\n point from the core drm atomic helper (drm_atomic_commit).\n \n Squelches the warning for:\n drivers/gpu/drm/i915/display/intel_display.c:14148: warning: Function parameter or member \u0027_state\u0027 not described in \u0027intel_atomic_commit\u0027\n drivers/gpu/drm/i915/display/intel_display.c:14148: warning: Excess function parameter \u0027state\u0027 description in \u0027intel_atomic_commit\u0027\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712134234.29893-1-chris@chris-wilson.co.uk\n\ncommit 913cafbb250f146654ea979186d8615e0c54d6a7\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 12 13:01:47 2019 +0100\n\n drm/vgem: Reclassify buffer creation debug message\n \n A buffer is created in response to the user ioctl, it should therefore\n be a plain DRM_DEBUG() message to reflect it being a user invoked\n response and not a driver construct.\n \n This is just to make the commonplace drm.debug\u003d[26e] quieter when\n running with vgem.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712120147.29830-1-chris@chris-wilson.co.uk\n\ncommit f8c6bfc612b56f02e1b8fae699dff12738aaf889\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Wed Jul 10 15:07:40 2019 +0200\n\n drm/panel: simple: fix AUO g185han01 horizontal blanking\n \n The horizontal blanking periods are too short, as the values are\n specified for a single LVDS channel. Since this panel is dual LVDS\n they need to be doubled. With this change the panel reaches its\n nominal vrefresh rate of 60Fps, instead of the 64Fps with the\n current wrong blanking.\n \n Philipp Zabel added:\n The datasheet specifies 960 active clocks + 40/128/160 clocks blanking\n on each of the two LVDS channels (min/typical/max), so doubled this is\n now correct.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Reviewed-by: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/1562764060.23869.12.camel@pengutronix.de\n\ncommit eaa2b31be152514f638e7c1c482e0ea0920a6aab\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Tue May 28 17:06:50 2019 +0300\n\n drm/i915: Skip SINK_COUNT read on CH7511\n \n CH7511 doesn\u0027t update SINK_COUNT properly so in order to detect\n the device as connected we have to ignore SINK_COUNT.\n \n In order to have access to the quirk list early enough we\n must move the drm_dp_read_desc() call to happen earlier.\n We can also skip re-reading this on eDP since we know it\n won\u0027t change.\n \n Cc: David S. \u003cdavid@majinbuu.com\u003e\n Cc: Peteris Rudzusiks \u003cpeteris.rudzusiks@gmail.com\u003e\n Tested-by: Peteris Rudzusiks \u003cpeteris.rudzusiks@gmail.com\u003e\n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d105406\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190528140650.19230-2-ville.syrjala@linux.intel.com\n Acked-by: Jani Nikula \u003cjani.nikula@intel.com\u003e #irc\n\ncommit e3c91a88e4f265a5ec58f6e3be0880255156f600\nAuthor: Joshua.Henderson@microchip.com \u003cJoshua.Henderson@microchip.com\u003e\nDate: Tue Jul 9 15:35:12 2019 +0000\n\n drm/atmel-hlcdc: set layer REP bit to enable replication logic\n \n This bit enables replication logic to expand an RGB color less than 24\n bits, to 24 bits, which is used internally for all formats. Otherwise,\n the least significant bits are always set to zero and the color may not\n be what is expected.\n \n Signed-off-by: Joshua Henderson \u003cjoshua.henderson@microchip.com\u003e\n Acked-by: Nicolas Ferre \u003cnicolas.ferre@microchip.com\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/1562686509-8747-1-git-send-email-joshua.henderson@microchip.com\n\ncommit f774f09649192f326fa030564afd3f8f5d82c1e4\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Fri Jul 12 11:14:45 2019 +0000\n\n drm/i915/guc: Turn on GuC/HuC auto mode\n \n Using \"enable_guc\" modparam auto mode (-1) will let driver\n decide on which platforms and in which configuration we want\n to use GuC/HuC firmwares.\n \n Today driver will enable HuC firmware authentication by GuC\n only on Gen11+ platforms as HuC firmware is required to unlock\n advanced video codecs in media driver.\n \n Legacy platforms with GuC/HuC are not affected by this change\n as for them driver still defaults to disabled(0) in auto mode.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712111445.21040-3-michal.wajdeczko@intel.com\n\ncommit 87d855e8cfa48613423357587dfbfe29bec23abe\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Fri Jul 12 11:14:44 2019 +0000\n\n drm/i915/guc: Don\u0027t enable GuC/HuC in auto mode on pre-Gen11\n \n We are about to change default setting of \"enable_guc\" modparam\n from 0(disabled) to -1(auto). As we only want to turn on\n GuC/HuC on Gen11+, keep it off for older gens.\n \n Note that it would be still possible to enable GuC/HuC on these\n old platforms using explicit \"enable_guc\u003d2\" modparam.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712111445.21040-2-michal.wajdeczko@intel.com\n\ncommit c9dccacfccc72c32692eedff4a27a4b0833a2afd\nAuthor: Vincent Whitchurch \u003cvincent.whitchurch@axis.com\u003e\nDate: Thu Jul 11 16:29:37 2019 +0200\n\n printk: Do not lose last line in kmsg buffer dump\n \n kmsg_dump_get_buffer() is supposed to select all the youngest log\n messages which fit into the provided buffer. It determines the correct\n start index by using msg_print_text() with a NULL buffer to calculate\n the size of each entry. However, when performing the actual writes,\n msg_print_text() only writes the entry to the buffer if the written len\n is lesser than the size of the buffer. So if the lengths of the\n selected youngest log messages happen to precisely fill up the provided\n buffer, the last log message is not included.\n \n We don\u0027t want to modify msg_print_text() to fill up the buffer and start\n returning a length which is equal to the size of the buffer, since\n callers of its other users, such as kmsg_dump_get_line(), depend upon\n the current behaviour.\n \n Instead, fix kmsg_dump_get_buffer() to compensate for this.\n \n For example, with the following two final prints:\n \n [ 6.427502] AAAAAAAAAAAAA\n [ 6.427769] BBBBBBBB12345\n \n A dump of a 64-byte buffer filled by kmsg_dump_get_buffer(), before this\n patch:\n \n 00000000: 3c 30 3e 5b 20 20 20 20 36 2e 35 32 32 31 39 37 \u003c0\u003e[ 6.522197\n 00000010: 5d 20 41 41 41 41 41 41 41 41 41 41 41 41 41 0a ] AAAAAAAAAAAAA.\n 00000020: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................\n 00000030: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................\n \n After this patch:\n \n 00000000: 3c 30 3e 5b 20 20 20 20 36 2e 34 35 36 36 37 38 \u003c0\u003e[ 6.456678\n 00000010: 5d 20 42 42 42 42 42 42 42 42 31 32 33 34 35 0a ] BBBBBBBB12345.\n 00000020: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................\n 00000030: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................\n \n Link: http://lkml.kernel.org/r/20190711142937.4083-1-vincent.whitchurch@axis.com\n Fixes: e2ae715d66bf4bec (\"kmsg - kmsg_dump() use iterator to receive log buffer content\")\n To: rostedt@goodmis.org\n Cc: linux-kernel@vger.kernel.org\n Cc: \u003cstable@vger.kernel.org\u003e # v3.5+\n Signed-off-by: Vincent Whitchurch \u003cvincent.whitchurch@axis.com\u003e\n Reviewed-by: Sergey Senozhatsky \u003csergey.senozhatsky@gmail.com\u003e\n Signed-off-by: Petr Mladek \u003cpmladek@suse.com\u003e\n\ncommit 0b61b8b07f6bc54b3e255bc40f63c3adfa968594\nAuthor: Janusz Krzysztofik \u003cjanusz.krzysztofik@linux.intel.com\u003e\nDate: Fri Jul 12 13:24:30 2019 +0200\n\n drm/i915: Propagate \"_probe\" function name suffix down\n \n Similar to the \"_release\" and \"_remove\" cases, consequently replace\n \"_init\" components of names of functions called from\n i915_driver_probe() with \"_probe\" suffixes for better code readability.\n \n Signed-off-by: Janusz Krzysztofik \u003cjanusz.krzysztofik@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712112429.740-7-janusz.krzysztofik@linux.intel.com\n\ncommit 78dae1ac35dd0c50d0f118e886b1f466cb9311fe\nAuthor: Janusz Krzysztofik \u003cjanusz.krzysztofik@linux.intel.com\u003e\nDate: Fri Jul 12 13:24:29 2019 +0200\n\n drm/i915: Propagate \"_remove\" function name suffix down\n \n Similar to the \"_release\" case, consistently replace mixed\n \"_cleanup\"/\"_fini\"/\"_fini_hw\" components found in names of functions\n called from i915_driver_remove() with \"_remove\" or \"_driver_remove\"\n suffixes for better code readability.\n \n Signed-off-by: Janusz Krzysztofik \u003cjanusz.krzysztofik@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712112429.740-6-janusz.krzysztofik@linux.intel.com\n\ncommit 3b58a94551368924d8a49baa349bc7694fbd7ddd\nAuthor: Janusz Krzysztofik \u003cjanusz.krzysztofik@linux.intel.com\u003e\nDate: Fri Jul 12 13:24:28 2019 +0200\n\n drm/i915: Propagate \"_release\" function name suffix down\n \n Replace mixed \"_fini\"/\"_cleanup\"/\"_cleanup_hw\" suffixes found in names\n of functions called from i915_driver_release() with \"_release\" suffix\n consistently. This provides better code readability, especially\n helpful when trying to work out which phase the code is in.\n \n Functions names starting with \"i915_driver_\", i.e., those defined in\n drivers/gpu/dri/i915/i915_drv.c, just have their \"cleanup\" or \"fini\"\n parts of their names replaced with the \"_release\" suffix, while names\n of functions coming from other source files have been suffixed with\n \"_driver_release\" to avoid ambiguity with other possible .release entry\n points.\n \n v2: early_probe pairs better with late_release (Chris)\n v3: fix typo in commit message (Joonas)\n \n Suggested-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Janusz Krzysztofik \u003cjanusz.krzysztofik@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712112429.740-5-janusz.krzysztofik@linux.intel.com\n\ncommit f2db53f14d3d1cde1d4c3b5b5aaba556c953f2f5\nAuthor: Janusz Krzysztofik \u003cjanusz.krzysztofik@linux.intel.com\u003e\nDate: Fri Jul 12 13:24:27 2019 +0200\n\n drm/i915: Replace \"_load\" with \"_probe\" consequently\n \n Use the \"_probe\" nomenclature not only in i915_driver_probe() helper\n name but also in other related function / variable names for\n consistency. Only the userspace exposed name of a related module\n parameter is left untouched.\n \n Signed-off-by: Janusz Krzysztofik \u003cjanusz.krzysztofik@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712112429.740-4-janusz.krzysztofik@linux.intel.com\n\ncommit b01558e56f84866dcad5f3f99819ec560d639f09\nAuthor: Janusz Krzysztofik \u003cjanusz.krzysztofik@linux.intel.com\u003e\nDate: Fri Jul 12 13:24:26 2019 +0200\n\n drm/i915: Rename \"_load\"/\"_unload\" to match PCI entry points\n \n Current names of i915_driver_load/unload() functions originate in\n legacy DRM stubs. Reduce nomenclature ambiguity by renaming them to\n match their current use as helpers called from PCI entry points.\n \n Suggested by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Janusz Krzysztofik \u003cjanusz.krzysztofik@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712112429.740-3-janusz.krzysztofik@linux.intel.com\n\ncommit b5893ffc274be966f95aa35f35916fa8725af154\nAuthor: Janusz Krzysztofik \u003cjanusz.krzysztofik@linux.intel.com\u003e\nDate: Fri Jul 12 13:24:25 2019 +0200\n\n drm/i915: Drop extern qualifiers from header function prototypes\n \n Follow dim checkpatch recommendation so it doesn\u0027t complain on that now\n and again on header file modifications.\n \n v2: drop testing leftover (Chris)\n \n Signed-off-by: Janusz Krzysztofik \u003cjanusz.krzysztofik@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712112429.740-2-janusz.krzysztofik@linux.intel.com\n\ncommit 6239901c576d4df87d6463c4f163bd432995d540\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 12 10:43:27 2019 +0100\n\n drm/i915/gtt: Use NULL to encode scratch shadow entries\n \n We can simplify our gtt walking code by comparing against NULL for\n scratch entries as opposed to looking up the distinct per-level scratch\n pointer.\n \n The only caveat is to remember to protect external parties and map the\n NULL to the scratch top pd.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712094327.24437-6-chris@chris-wilson.co.uk\n\ncommit c03cbe4c0a838ce03266a8c579ef8d13181907b6\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 12 10:43:26 2019 +0100\n\n drm/i915/gtt: Convert vm-\u003escratch into an array\n \n Each level has its own scratch. Make the levels more obvious by forgoing\n the fancy similarly names and replace them with a number. 0 is the bottom\n most level, the physical page used for actual data; 1+ are the page\n directories.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712094327.24437-5-chris@chris-wilson.co.uk\n\ncommit 2776326457562f971b7b38f4683e298e55252ccc\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 12 10:43:25 2019 +0100\n\n drm/i915/gtt: Compute the radix for gen8 page table levels\n \n The radix levels of each page directory are easily determined so replace\n the numerous hardcoded constants with precomputed derived constants.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712094327.24437-4-chris@chris-wilson.co.uk\n\ncommit 18c7962b8cf24a5a2a757db16f535ff8c5191966\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 12 10:43:24 2019 +0100\n\n drm/i915/gtt: Markup i915_ppgtt height\n \n This will be useful to consolidate recursive code.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712094327.24437-3-chris@chris-wilson.co.uk\n\ncommit a9abea97856b47bb9a85d60b34461444cb9a2d1e\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 12 10:43:23 2019 +0100\n\n drm/i915/gtt: Reorder gen8 ppgtt free/clear/alloc\n \n In preparation for refactoring the free/clear/alloc, first move the code\n around so that we can avoid forward declarations in the next set of\n patches.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712094327.24437-2-chris@chris-wilson.co.uk\n\ncommit 57a7e30546b2f8de4925406ff55eef11da05615b\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 12 10:43:22 2019 +0100\n\n drm/i915/gtt: Wrap page_table with page_directory\n \n The page directory extends the page table with the shadow entries. Make\n the page directory struct embed the page table for easier code reuse.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712094327.24437-1-chris@chris-wilson.co.uk\n\ncommit 6eebfe8a10a62139d681e2f1af1386252742278b\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 12 08:58:18 2019 +0100\n\n drm/i915/gtt: Use shallow dma pages for scratch\n \n We only use the dma pages for scratch, and so do not need to allocate\n the extra storage for the shadow page directory.\n \n v2: Refrain from reintroducing I915_PDES\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712075818.20616-1-chris@chris-wilson.co.uk\n\ncommit 3e1f0a518d22c361d8e947e94a189ba0c9b3db7a\nAuthor: John Harrison \u003cJohn.C.Harrison@Intel.com\u003e\nDate: Fri Jul 12 00:07:45 2019 -0700\n\n drm/i915: Add engine name to workaround debug print\n \n There is a debug message in the workaround initialisation path that\n reports how many entries were added of each type. However, whitelist\n workarounds exist for multiple engines but the type name is just\n \u0027whitelist\u0027. Tvrtko suggested adding the engine name to make the\n message more useful.\n \n v2: Updated the similar message in the workaround reset selftest.\n \n Signed-off-by: John Harrison \u003cJohn.C.Harrison@Intel.com\u003e\n CC: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712070745.35239-4-John.C.Harrison@Intel.com\n\ncommit aee20aaed8876e90fa6f500ee5637423a3ef9ab9\nAuthor: John Harrison \u003cJohn.C.Harrison@Intel.com\u003e\nDate: Fri Jul 12 00:07:44 2019 -0700\n\n drm/i915: Implement read-only support in whitelist selftest\n \n Newer hardware supports extra feature in the whitelist registers. This\n patch updates the selftest to test that entries marked as read only\n are actually read only.\n \n v2: Removed all use of \u0027rsvd\u0027 for read-only registers to avoid\n ambiguous code or error messages.\n \n Signed-off-by: John Harrison \u003cJohn.C.Harrison@Intel.com\u003e\n CC: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712070745.35239-3-John.C.Harrison@Intel.com\n\ncommit 1e2b7f497c28a47793a95d8f0cc8e135899827f1\nAuthor: John Harrison \u003cJohn.C.Harrison@Intel.com\u003e\nDate: Fri Jul 12 00:07:43 2019 -0700\n\n drm/i915: Add test for invalid flag bits in whitelist entries\n \n As per review feedback by Tvrtko, added a check that no invalid bits\n are being set in the whitelist flags fields.\n \n Also updated the read/write access definitions to make it clearer that\n they are an enum field not a set of single bit flags.\n \n Signed-off-by: John Harrison \u003cJohn.C.Harrison@Intel.com\u003e\n CC: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190712070745.35239-2-John.C.Harrison@Intel.com\n\ncommit 374bf825e7a07b6ae61a8a839c0ea184a7665267\nAuthor: Douglas Anderson \u003cdianders@chromium.org\u003e\nDate: Thu Jul 11 13:34:55 2019 -0700\n\n drm/panel: simple: Use display_timing for AUO b101ean01\n \n Convert the AUO b101ean01 from using a fixed mode to specifying a\n display timing with min/typ/max values.\n \n The AUO b101ean01\u0027s datasheet says:\n * Vertical blanking min is 12\n * Horizontal blanking min is 60\n * Pixel clock is between 65.3 MHz and 75 MHz\n \n The goal here is to be able to specify the proper timing in device\n tree to use on rk3288-veyron-minnie to match what the downstream\n kernel is using so that it can used the fixed PLL.\n \n Changes in v4:\n - display_timing for AUO b101ean01 new for v4.\n Changes in v6:\n - Rebased to drm-misc next\n - Added tags\n \n Signed-off-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n Acked-by: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711203455.125667-4-dianders@chromium.org\n\ncommit d719cbe9a5b39437c442836a6fd5e7b178bcee4d\nAuthor: Douglas Anderson \u003cdianders@chromium.org\u003e\nDate: Thu Jul 11 13:34:54 2019 -0700\n\n drm/panel: simple: Use display_timing for Innolux n116bge\n \n Convert the Innolux n116bge from using a fixed mode to specifying a\n display timing with min/typ/max values.\n \n Note that the n116bge\u0027s datasheet doesn\u0027t fit too well into DRM\u0027s way\n of specifying things. Specifically the panel\u0027s datasheet just\n specifies the vertical blanking period and horizontal blanking period\n and doesn\u0027t break things out. For now we\u0027ll leave everything as a\n fixed value but just allow adjusting the pixel clock. I\u0027ve added a\n comment on what the datasheet claims so someone could later expand\n things to fit their needs if they wanted to test other blanking\n periods.\n \n The goal here is to be able to specify the panel timings in the device\n tree for several rk3288 Chromebooks (like rk3288-veryon-jerry). These\n Chromebooks have all been running in the downstream kernel with the\n standard porches and sync lengths but just with a slightly slower\n pixel clock because the 76.42 MHz clock is not achievable from the\n fixed PLL that was available. These Chromebooks only achieve a\n refresh rate of ~58 Hz. While it\u0027s probable that we could adjust the\n timings to achieve 60 Hz it\u0027s probably wisest to match what\u0027s been\n running on these devices all these years.\n \n I\u0027ll note that though the upstream kernel has always tried to achieve\n 76.42 MHz, it has actually been running at 74.25 MHz also since the\n video processor is parented off the same fixed PLL.\n \n Changes in v4:\n - display_timing for Innolux n116bge new for v4.\n Changes in v5:\n - Added Heiko\u0027s Tested-by\n Changes in v6:\n - Rebased to drm-misc next\n - Added tags\n \n Signed-off-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n Tested-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Acked-by: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711203455.125667-3-dianders@chromium.org\n\ncommit b8a2948fa2b3a5a6da67fd57aa01c7318d795125\nAuthor: Sean Paul \u003cseanpaul@chromium.org\u003e\nDate: Thu Jul 11 13:34:53 2019 -0700\n\n drm/panel: simple: Add ability to override typical timing\n \n This patch adds the ability to override the typical display timing for a\n given panel. This is useful for devices which have timing constraints\n that do not apply across the entire display driver (eg: to avoid\n crosstalk between panel and digitizer on certain laptops). The rules are\n as follows:\n \n - panel must not specify fixed mode (since the override mode will\n either be the same as the fixed mode, or we\u0027ll be unable to\n check the bounds of the overried)\n - panel must specify at least one display_timing range which will be\n used to ensure the override mode fits within its bounds\n \n Changes in v2:\n - Parse the full display-timings node (using the native-mode) (Rob)\n Changes in v3:\n - No longer parse display-timings subnode, use panel-timing (Rob)\n Changes in v4:\n - Don\u0027t add mode from timing if override was specified (Thierry)\n - Add warning if timing and fixed mode was specified (Thierry)\n - Don\u0027t add fixed mode if timing was specified (Thierry)\n - Refactor/rename a bit to avoid extra indentation from \"if\" tests\n - i should be unsigned (Thierry)\n - Add annoying WARN_ONs for some cases (Thierry)\n - Simplify \u0027No display_timing found\u0027 handling (Thierry)\n - Rename to panel_simple_parse_override_mode() (Thierry)\n Changes in v5:\n - Added Heiko\u0027s Tested-by\n Changes in v6:\n - Rebased to drm-misc next\n - Added tags\n \n Cc: Doug Anderson \u003cdianders@chromium.org\u003e\n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Cc: Jeffy Chen \u003cjeffy.chen@rock-chips.com\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Stéphane Marchesin \u003cmarcheu@chromium.org\u003e\n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: devicetree@vger.kernel.org\n Cc: dri-devel@lists.freedesktop.org\n Signed-off-by: Sean Paul \u003cseanpaul@chromium.org\u003e\n Tested-by: Enric Balletbo i Serra \u003cenric.balletbo@collabora.com\u003e\n Signed-off-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n Tested-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@collabora.com\u003e\n Acked-by: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711203455.125667-2-dianders@chromium.org\n\ncommit a4e7e98e90ebd9a801d6a383e1edd10b09d155ba\nAuthor: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\nDate: Tue Jun 25 22:37:05 2019 -0300\n\n drm/vkms: Rename vkms_crc.c into vkms_composer.c\n \n As a preparation work for introducing writeback to vkms, this patch\n renames the file vkms_crc.c into vkms_composer.c. Accordingly, it also\n adjusts the functions and data structures to match the changes.\n \n No functional change.\n \n Signed-off-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Acked-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/dea62063077ebf5cc1dfce8876e56788d15367e6.1561491964.git.rodrigosiqueiramelo@gmail.com\n\ncommit e9d85f731de06a35d2ae6cdcf7d0e037c98ef41a\nAuthor: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\nDate: Tue Jun 25 22:36:18 2019 -0300\n\n drm/vkms: Avoid assigning 0 for possible_crtc\n \n When vkms invoke drm_universal_plane_init(), it sets 0 for\n possible_crtcs parameter which means that planes can\u0027t be attached to\n any CRTC. It currently works due to some safeguard in the drm_crtc file;\n however, it is possible to identify the problem by trying to append a\n second connector. This patch fixes this issue by modifying\n vkms_plane_init() to accept an index parameter which makes the code a\n little bit more flexible and avoid set zero to possible_crtcs.\n \n Signed-off-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/d67849c62a8d8ace1a0af455998b588798a4c45f.1561491964.git.rodrigosiqueiramelo@gmail.com\n\ncommit a1c5f1510b3f39d57a6eaa9d75c70e5beaa952ff\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Thu Jul 11 10:31:15 2019 -0700\n\n drm/i915/tgl: Update DPLL clock reference register\n \n This register definition changed from ICL and has now another meaning.\n Use the right bits on TGL.\n \n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711173115.28296-22-lucas.demarchi@intel.com\n\ncommit 36ca5335f202bd54faf38b37fed1b99078e1839e\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Thu Jul 11 10:31:14 2019 -0700\n\n drm/i915/tgl: Add DPLL registers\n \n On TGL the port programming for combophy is very similar to ICL, so\n adapt the callers to possibly use the different register values.\n \n v2 (Lucas): Add TODO with about DPLL4 (requested by Ville)\n \n Cc: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711173115.28296-21-lucas.demarchi@intel.com\n\ncommit d757535e3116c1a5f024bfef1ab3544a525f03fb\nAuthor: Mahesh Kumar \u003cmahesh1.kumar@intel.com\u003e\nDate: Thu Jul 11 10:31:13 2019 -0700\n\n drm/i915/tgl: Add vbt value mapping for DDC Bus pin\n \n Add VBT-value to DDC bus pin mapping for the same.\n \n Signed-off-by: Mahesh Kumar \u003cmahesh1.kumar@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711173115.28296-20-lucas.demarchi@intel.com\n\ncommit fb81cbe46988cddfd9e6fd2bb9f7ce39cea0c380\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Thu Jul 11 10:31:12 2019 -0700\n\n drm/i915/tgl: port to ddc pin mapping\n \n Make the icl function generic so it is based on phy type and can be\n applied to tgl as well.\n \n I checked if this could not apply to EHL as well, but unfortunately\n there the HPD and DDC/GMBUS pins for DDI C are mapped to TypeC Port 1\n even though it doesn\u0027t have TC phy.\n \n v2: don\u0027t add a separate function for TGL, but rather reuse the ICL one\n (suggested by Rodrigo)\n v3: rebase after the introduction of enum phy and use it for the\n conversions\n \n Cc: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711173115.28296-19-lucas.demarchi@intel.com\n\ncommit 3fd53262f0dd81361efcb5ccba8312525ee6c645\nAuthor: Mahesh Kumar \u003cmahesh1.kumar@intel.com\u003e\nDate: Thu Jul 11 10:31:11 2019 -0700\n\n drm/i915/tgl: Add gmbus gpio pin to port mapping\n \n Add default GPIO pin mapping for all ports. Tiger Lake has 3 combophy\n ports and 6 TC ports, gpio pin1-3 are mapped to combophy \u0026 pin9-14 are\n mapped to TC ports.\n \n Cc: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Mahesh Kumar \u003cmahesh1.kumar@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711173115.28296-18-lucas.demarchi@intel.com\n\ncommit 30fcc338bc66f930dc7f692205474edb7ca6a607\nAuthor: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\nDate: Thu Jul 11 10:31:10 2019 -0700\n\n drm/i915/gen12: MBUS B credit change\n \n Previously, the recommended B credit for all platforms was 24 / number\n of pipes, which would give 6 for newer platforms with 4 pipes. However 6\n is not enough and we need 12 on these cases.\n \n We also need a different BW credit for these platforms.\n \n Cc: Arthur J Runyan \u003carthur.j.runyan@intel.com\u003e\n Signed-off-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711173115.28296-17-lucas.demarchi@intel.com\n\ncommit deea06b47574de81da72a2d2fa1ace4e119ca02a\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Thu Jul 11 14:35:17 2019 -0700\n\n drm/i915/tgl: apply Display WA #1178 to fix type C dongles\n \n Add port C to workaround to cover Tiger Lake.\n \n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190708231629.9296-22-lucas.demarchi@intel.com\n Reviewed-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711213517.13674-1-lucas.demarchi@intel.com\n\ncommit 55cd5048e176318911db97668f562c1599982dfb\nAuthor: Mahesh Kumar \u003cmahesh1.kumar@intel.com\u003e\nDate: Thu Jul 11 10:31:08 2019 -0700\n\n drm/i915/tgl: init ddi port A-C for Tiger Lake\n \n This patch initializes DDI PORT A, B \u0026 C for Tiger lake. Other\n TC ports need to be initialized later once corresponding code is there.\n \n Cc: Madhav Chauhan \u003cmadhav.chauhan@intel.com\u003e\n Signed-off-by: Mahesh Kumar \u003cmahesh1.kumar@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711173115.28296-15-lucas.demarchi@intel.com\n\ncommit 5c71970889d7d58d9fe26f92f454d7fa5db52202\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Thu Jul 11 10:31:07 2019 -0700\n\n drm/i915/tgl: Add additional PHYs for Tiger Lake\n \n Tiger Lake has up to 3 combo phys and 6 TC phys. Extend the helper\n conversion functions from port to phy.\n \n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711173115.28296-14-lucas.demarchi@intel.com\n\ncommit 6c8337dafaa9a328216c62a79c9a03176af3ce70\nAuthor: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\nDate: Thu Jul 11 10:31:06 2019 -0700\n\n drm/i915/tgl: Add additional ports for Tiger Lake\n \n There are 2 new additional typeC ports in Tiger Lake and PORT-C is now a\n combophy port. This results in 6 typeC ports and 3 combophy ports.\n These 6 TC ports can be DP alternate mode, DP over thunderbolt, native\n DP on legacy DP connector or native HDMI on legacy connector.\n \n v2: Rebase on new modular FIA code (Lucas)\n v3: Also add new port in port_identifier(), even though it can\u0027t\n possibly be used there (requested by José)\n v4: Add conversion port-\u003etc_port in helper function after introction of\n phy namespace (Lucas)\n \n Cc: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Signed-off-by: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711173115.28296-13-lucas.demarchi@intel.com\n\ncommit c9014a2c7937e40858b858200a35432075638ede\nAuthor: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\nDate: Thu Jul 11 10:31:05 2019 -0700\n\n drm/i915/tgl: Add pll manager\n \n Add a new pll array for Tiger Lake. The TC pll functions for type C will\n be covered in later patches after its phy is implemented.\n \n Cc: Madhav Chauhan \u003cmadhav.chauhan@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711173115.28296-12-lucas.demarchi@intel.com\n\ncommit 68ff39c3f8c01b1e640d111abdcf814804a6b236\nAuthor: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\nDate: Thu Jul 11 10:31:04 2019 -0700\n\n drm/i915/tgl: Add new pll ids\n \n Add 2 new PLLs for additional TC ports. The names for the PLLs on TGL\n changed, but most registers remained the same, like MGPLL5_ENABLE,\n MGPLL6_ENABLE. So continue to use the name from ICL.\n \n Cc: Madhav Chauhan \u003cmadhav.chauhan@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711173115.28296-11-lucas.demarchi@intel.com\n\ncommit 1db27a7291195057e3a20fb9998e2d365ee897f9\nAuthor: Mika Kahola \u003cmika.kahola@intel.com\u003e\nDate: Thu Jul 11 10:31:03 2019 -0700\n\n drm/i915/tgl: Add power well to support 4th pipe\n \n Add power well 5 to support 4th pipe and transcoder on TGL.\n \n Cc: James Ausmus \u003cjames.ausmus@intel.com\u003e\n Cc: Imre Deak \u003cimre.deak@intel.com\u003e\n Signed-off-by: Mika Kahola \u003cmika.kahola@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711173115.28296-10-lucas.demarchi@intel.com\n\ncommit 656409bbaf8792c015708e567d4ebcd7fb4e7728\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Thu Jul 11 10:31:02 2019 -0700\n\n drm/i915/tgl: Add power well support\n \n The patch adds the new power wells introduced by TGL (GEN 12) and\n maps these to existing/new power domains. The changes for GEN 12 wrt\n to GEN 11 are the following:\n \n - Transcoder#EDP removed from power well#1 (Transcoder#A used in\n low-power mode instead)\n - Transcoder#A is now backed by power well#1 instead of power well#3\n - The DDI#B/C combo PHY ports are now backed by power well#1 instead of\n power well#3\n - New power well#5 added for pipe#D functionality (TODO)\n - 2 additional TC ports (TC#5-6) backed by power well#3, 2 port\n specific IO power wells (only for the non-TBT modes) and 4 port\n specific AUX power wells (2-2 for TBT vs. non-TBT modes)\n - Power well#2 backs now VDSC/joining for pipe#A instead of VDSC for\n eDP and MIPI DSI (TODO)\n \n On TGL Port DDI#C changed to be a combo PHY (native DP/HDMI) and\n BSpec has renamed ports DDI#D-F to TC#4-6 respectively. Thus on ICL we\n have the following naming for ports:\n \n - Combo PHYs (native DP/HDMI):\n DDI#A-B\n - TBT/non-TBT (TC altmode, native DP/HDMI) PHYs:\n DDI#C-F\n \n Starting from GEN 12 we have the following naming for ports:\n - Combo PHYs (native DP/HDMI):\n DDI#A-C\n - TBT/non-TBT (TC altmode, native DP/HDMI) PHYs:\n DDI TC#1-6\n \n To save some space in the power domain enum the power domain naming in\n the driver reflects the above change, that is power domains TC#1-3 are\n added as aliases for DDI#D-F and new power domains are reserved for\n TC#4-6.\n \n v2 (Lucas):\n - Separate out the bits and definitions for TGL from the ICL ones.\n Fix use of TRANSCODER_EDP_VDSC, that is now the correct define since\n we don\u0027t define TRANSCODER_A_VDSC power domain to spare a one bit in\n the bitmask (suggested by Ville)\n v3 (Lucas):\n - Fix missing squashes on v2\n - Rebase on renamed TRANSCODER_EDP_VDSC\n \n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711173115.28296-9-lucas.demarchi@intel.com\n\ncommit 276199e6be6a4058e74828728eb25179b1461978\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Thu Jul 11 10:31:01 2019 -0700\n\n drm/i915/tgl: rename TRANSCODER_EDP_VDSC to use on transcoder A\n \n On TGL the special EDP transcoder is gone and it should be handled by\n transcoder A.\n \n v2 (Lucas):\n - Reuse POWER_DOMAIN_TRANSCODER_EDP_VDSC (suggested by Ville)\n - Use crtc-\u003edev since new_crtc_state-\u003estate may be NULL on atomic\n commit (suggested by Maarten)\n v3 (Lucas):\n - Rename power domain so it\u0027s clear it can also be used for transcoder\n A in TGL (requested by José and Manasi)\n \n Cc: Imre Deak \u003cimre.deak@intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Acked-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Reviewed-by: Manasi Navare \u003cmanasi.d.navare@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711173115.28296-8-lucas.demarchi@intel.com\n\ncommit 7ff0fca4964ff19d8e16669d8d3070018cecdd0f\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Thu Jul 11 10:31:00 2019 -0700\n\n drm/i915/tgl: Check if pipe D is fused\n \n On Tiger Lake there is one more pipe - check if it\u0027s fused.\n \n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Mika Kahola \u003cmika.kahola@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711173115.28296-7-lucas.demarchi@intel.com\n\ncommit 9747f0c2fb9e1a6d7c907d34a373924c84093afa\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Thu Jul 11 10:30:59 2019 -0700\n\n drm/i915/tgl: Add TGL PCI IDs\n \n Current list of PCI IDs for Tiger Lake.\n \n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Reviewed-by: Mika Kahola \u003cmika.kahola@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711173115.28296-6-lucas.demarchi@intel.com\n\ncommit d8df6bec1a30097f65781faf4877e4da7a520002\nAuthor: Mahesh Kumar \u003cmahesh1.kumar@intel.com\u003e\nDate: Thu Jul 11 10:30:58 2019 -0700\n\n drm/i915/tgl: Add TGL PCH detection in virtualized environment\n \n Assume PCH_TGP when platform is TGL.\n \n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Mahesh Kumar \u003cmahesh1.kumar@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711173115.28296-5-lucas.demarchi@intel.com\n\ncommit 7f02889292e2e3fe6390b98d4ebc3dd67f20a812\nAuthor: Radhakrishna Sripada \u003cradhakrishna.sripada@intel.com\u003e\nDate: Thu Jul 11 10:30:57 2019 -0700\n\n drm/i915/tgl: Introduce Tiger Lake PCH\n \n Add the enum additions to TGP.\n \n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: David Weinehall \u003cdavid.weinehall@intel.com\u003e\n Cc: James Ausmus \u003cjames.ausmus@intel.com\u003e\n Signed-off-by: Radhakrishna Sripada \u003cradhakrishna.sripada@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711173115.28296-4-lucas.demarchi@intel.com\n\ncommit abd3a0fe040d51fc2943855169558b4b0b3311c6\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Thu Jul 11 10:30:56 2019 -0700\n\n drm/i915/tgl: add initial Tiger Lake definitions\n \n Tiger Lake is a Intel® Processor containing Intel® HD Graphics.\n \n This is just an initial Tiger Lake definition. PCI IDs, generic support\n and new features coming in following patches.\n \n v2 (Lucas):\n - Remove modular FIA - feature will be re-introduced in future\n \n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711173115.28296-3-lucas.demarchi@intel.com\n\ncommit f1f1d4fa5869c8b0b3e7f9e9a8a2f1983f10123a\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Thu Jul 11 10:30:55 2019 -0700\n\n drm/i915: Add 4th pipe and transcoder\n \n Add pipe D and transcoder D to prepare for platforms having them.\n \n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711173115.28296-2-lucas.demarchi@intel.com\n\ncommit ddb3d12afa92beb67027e84bbc994b0868446022\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Jul 10 15:58:51 2019 +0300\n\n drm/i915: Don\u0027t overestimate 4:2:0 link symbol clock\n \n With 4:2:0 output the LS clock can be half of what it is with 4:4:4.\n Make that happen.\n \n Cc: Gwan-gyeong Mun \u003cgwan-gyeong.mun@intel.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190710125851.3275-1-ville.syrjala@linux.intel.com\n Reviewed-by: Gwan-gyeong Mun \u003cgwan-gyeong.mun@intel.com\u003e\n Tested-by: Gwan-gyeong Mun \u003cgwan-gyeong.mun@intel.com\u003e\n\ncommit 86c9640b3a1ede3215fe8fc1848abb54a36aeac8\nAuthor: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\nDate: Wed Jul 10 20:12:30 2019 +0300\n\n drm/i915: Copy name string into ring buffer for intel_update/disable_plane tracepoints\n \n Currently the intel_update_plane and intel_disable_plane tracepoints record\n the address of plane-\u003ename in the ring buffer, and then when reading the\n ring buffer uses %s to get the name. The issue with this, is that those two\n events can be minutes, hours or even days apart. It is very dangerous to\n dereference a string pointer without knowing if it still exists or not.\n \n The proper way to handle this is to use the __string() macro in the\n tracepoint which will save the string into the ring buffer at the time of\n recording. Then there\u0027s no worries if the original string still exists in\n memory when the ring buffer is read.\n \n Signed-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n [vsyrjala: Rebase on top of drm-tip]\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190710171230.7471-1-ville.syrjala@linux.intel.com\n\ncommit ddafc0f7564051121285e8143948f7a5269b8eb6\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Jul 11 17:24:15 2019 +0100\n\n drm/i915/guc: Drop redundant ctx param from kerneldoc\n \n drivers/gpu/drm/i915/intel_guc_submission.c:799: warning: Excess function parameter \u0027ctx\u0027 description in \u0027guc_client_alloc\u0027\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711162415.2938-1-chris@chris-wilson.co.uk\n\ncommit b12d5944fc285730d75bdd8f7cbc28d22c26f36f\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Jul 10 16:49:37 2019 +0300\n\n drm/i915: Don\u0027t pass stack garbage to pcode in the second data register\n \n Zero initialize val2 so that we don\u0027t pass stack garbage to\n the pcode qgv read command. I suspect in this case pcode\n just ignores the initial value in that registers, but better\n safe than sorry.\n \n Cc: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Reported-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190710134937.25835-1-ville.syrjala@linux.intel.com\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n\ncommit a85fb46777c0e078075e7f4880a1f82178d66c1a\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Mon Jul 1 19:05:50 2019 +0300\n\n drm/i915: Use intel_ types in intel_atomic_commit()\n \n Make life less annoying by favoring the intel_ types over\n the drm_ types in intel_atomic_commit().\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190701160550.24205-6-ville.syrjala@linux.intel.com\n Reviewed-by: Imre Deak \u003cimre.deak@intel.com\u003e\n\ncommit 6a64e985d242921a9b6050bddc5aa716c02aa6ef\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Mon Jul 1 19:05:49 2019 +0300\n\n drm/i915: Use intel_ types in intel_{lock,modeset}_all_pipes()\n \n Streamline the code a bit by using intel_ types instead of the\n drm_ types.\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190701160550.24205-5-ville.syrjala@linux.intel.com\n Reviewed-by: Imre Deak \u003cimre.deak@intel.com\u003e\n\ncommit e3b4089c68ed59fcda9220db02c8ccfe9c49ccd7\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Mon Jul 1 19:05:48 2019 +0300\n\n drm/i915: Polish intel_atomic_track_fbs()\n \n Streamline the code a bit by using intel_ types instead of drm_\n types in intel_atomic_track_fbs().\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190701160550.24205-4-ville.syrjala@linux.intel.com\n Reviewed-by: Imre Deak \u003cimre.deak@intel.com\u003e\n\ncommit 13d723a1177d63f3aa9cea8bc9aae9f7617adc4f\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Mon Jul 1 19:05:47 2019 +0300\n\n drm/i915: Polish intel_shared_dpll_swap_state()\n \n Use swap() instead of hand rolling it in intel_shared_dpll_swap_state(),\n and pass in the intel_atomic_state instead of drm_atomic_state. Makes\n the code less convoluted.\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190701160550.24205-3-ville.syrjala@linux.intel.com\n Reviewed-by: Imre Deak \u003cimre.deak@intel.com\u003e\n\ncommit afe0c21b6228ac241a9e537f71db418771225090\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Mon Jul 1 19:05:46 2019 +0300\n\n drm/i915: Simplify modeset_get_crtc_power_domains() arguments\n \n Pass just the crtc state to modeset_get_crtc_power_domains(). We\n can get the crtc from therein.\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190701160550.24205-2-ville.syrjala@linux.intel.com\n Reviewed-by: Imre Deak \u003cimre.deak@intel.com\u003e\n\ncommit 88016a9fb7417cb99c2e58e96ff943d59fa27099\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Mon Jul 1 19:05:45 2019 +0300\n\n drm/i915: Check crtc_state-\u003ewm.need_postvbl_update before grabbing wm.mutex\n \n wm.mutex does not protect the crtc state so no point in grabbing it\n to check crtc_state-\u003ewm.need_postvbl_update.\n \n Also do a bit of s/intel_crtc/crtc/ while at it.\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190701160550.24205-1-ville.syrjala@linux.intel.com\n Reviewed-by: Imre Deak \u003cimre.deak@intel.com\u003e\n\ncommit 13ce609243c7fbc17d8b272629242bc25a5401bd\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Mon Jul 1 19:15:34 2019 +0300\n\n drm/i915: Use the \"display core\" power domain in vlv/chv set_cdclk()\n \n The PFI credit programming performed during cdclk change on vlv/chv\n requires access to a register in the disp2d power well. So far\n we\u0027ve abused pipe-A power domain for this, but now we have the\n more appropriate \"display core\" domain so let\u0027s make use of it.\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190701161534.6671-1-ville.syrjala@linux.intel.com\n Reviewed-by: Imre Deak \u003cimre.deak@intel.com\u003e\n\ncommit f63dfc1464380833c6a2c8daa1b767c6020565af\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Jul 11 07:51:59 2019 +0100\n\n drm/i915/selftests: Hold the vma manager lock while modifying mmap_offset\n \n Right idea, wrong lock. We already drop struct_mutex before we free the\n mmap_offset when freeing the object, so we need to take the vma manager\n lock when manipulating the mmap_offset address space for our selftests.\n \n Fixes: 8221d21b0664 (\"drm/i915/selftests: Lock the drm_mm while modifying\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190711065215.4004-2-chris@chris-wilson.co.uk\n\ncommit 5a5efbf4ef504a8136963a8ae10a3c0a1d59b479\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Jun 19 21:03:07 2019 +0300\n\n drm/i915/sdvo: Fix handling if zero hbuf size\n \n The spec says:\n \"A value of 0 indicates that this buffer does not exist\"\n So we should not convert a hbuf_size of 0 into 1.\n \n Also pull the relevant code into a helper to avoid making the\n same mistake multiple times.\n \n And while at it fix the debug prints to not say \"hbuf_len\" twice.\n \n v2: s/%i/%u/ in the debug (Imre)\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619180312.31817-1-ville.syrjala@linux.intel.com\n Reviewed-by: Imre Deak \u003cimre.deak@intel.com\u003e\n\ncommit aebf052bb6474f21179f2469b5c3172c4f5e4fd1\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Tue Jul 9 17:54:27 2019 -0700\n\n drm/i915/guc: Simplify guc client\n \n We originally added support, in some cases partial, for different modes\n of operations via guc clients:\n \n - proxy vs direct submission;\n - variable engine mask per-client.\n \n We only ever used one flow (all submissions via a single proxy), so the\n other code paths haven\u0027t been exercised and are most likely\n non-functional. The guc firmware interface is also in the process of\n being updated to better fit the i915 flow and our client abstraction\n will need to change accordingly (or possibly go away entirely), so these\n old unused paths can be considered dead and removed.\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Matthew Brost \u003cmatthew.brost@intel.com\u003e\n Cc: John Harrison \u003cJohn.C.Harrison@Intel.com\u003e\n Acked-by: Matthew Brost \u003cMatthew Brost \u003cmatthew.brost@intel.com\u003e\n Reviewed-by: Michał Winiarski \u003cmichal.winiarski@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190710005437.3496-3-daniele.ceraolospurio@intel.com\n\ncommit 71b0846c17b9c86ffa9a18f09f8cfda4fe41ee86\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jul 9 17:54:26 2019 -0700\n\n drm/i915/guc: Remove preemption support for current fw\n \n Preemption via GuC submission is not being supported with its current\n legacy incarnation. The current FW does support a similar pre-emption\n flow via H2G, but it is class-based instead of being instance-based,\n which doesn\u0027t fit well with the i915 tracking. To fix this, the\n firmware is being updated to better support our needs with a new flow,\n so we can safely remove the old code.\n \n v2 (Daniele): resurrect \u0026 rebase, reword commit message, remove\n preempt_context as well\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Matthew Brost \u003cmatthew.brost@intel.com\u003e\n Cc: John Harrison \u003cJohn.C.Harrison@Intel.com\u003e\n Acked-by: Matthew Brost \u003cmatthew.brost@intel.com\u003e\n Reviewed-by: Michał Winiarski \u003cmichal.winiarski@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190710005437.3496-2-daniele.ceraolospurio@intel.com\n\ncommit bf1315b830a4ea2e3842400982cf66f15b40b7ec\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jul 10 17:14:13 2019 +0100\n\n drm/i915/selftests: Ensure we don\u0027t clamp a random offset to 32b\n \n Specify that we do want a 64b value for sizeof(u32) as we want to\n compute the mask of the upper 62bits.\n \n v2: Use round_down() for automatic type promotion\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190710161413.7115-1-chris@chris-wilson.co.uk\n\ncommit 47c4bdd6d3ea771f31ab2ef48219d713e808970f\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jul 10 21:04:28 2019 +0100\n\n drm/i915/gt: Drop the duplicate icl workaround\n \n The extra w/a was revived in the backmerge that was meant to fix it!\n \n Fixes: 88c90e800675 (\"Merge drm/drm-next into drm-intel-next-queued\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190710200428.3275-1-chris@chris-wilson.co.uk\n\ncommit 719d240026022547e3d908e29f4926bf63584c17\nAuthor: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\nDate: Tue Jul 9 11:39:34 2019 -0700\n\n drm/i915/ehl: Enable DDI-D\n \n EHL has four DDI\u0027s (DDI-A and DDI-D share combo PHY A).\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190709183934.445-6-matthew.d.roper@intel.com\n\ncommit d8fe2ab6bb283dd9b511ea9bea52874fafd3de61\nAuthor: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\nDate: Tue Jul 9 11:39:33 2019 -0700\n\n drm/i915: Transition port type checks to phy checks\n \n Transition the remaining uses of intel_port_is_* over to the equivalent\n intel_phy_is_* functions and drop the port functions.\n \n v5: Fix a call in a debug function that\u0027s only called when\n CONFIG_DRM_I915_DEBUG_RUNTIME_PM is on. (CI)\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Signed-off-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190709183934.445-5-matthew.d.roper@intel.com\n\ncommit dc867bc7d88769910b58748040240906540ea8c1\nAuthor: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\nDate: Tue Jul 9 11:39:32 2019 -0700\n\n drm/i915/gen11: Convert combo PHY logic to use new \u0027enum phy\u0027 namespace\n \n Convert the code that operates directly on gen11 combo PHY\u0027s to use the\n new namespace. Combo PHY registers are those named \"ICL_PORT_*\" plus\n ICL_DPHY_CHKN.\n \n Note that a lot of the PHY programming happens in the MIPI DSI code.\n For clarity I\u0027ve added a for_each_dsi_phy() to loop over the phys used\n by DSI. Since DSI always uses A \u0026 B on gen11, port\u003dphy in all cases so\n it doesn\u0027t actually matter which form we use in the DSI code. I\u0027ve used\n the phy iterator in code that\u0027s explicitly working with the combo PHY,\n but left the rest of the DSI code using the port iterator and namespace\n to minimize patch deltas. We can switch the rest of the DSI code over\n to use phy terminology later if this winds up being too confusing.\n \n v6: Drop an include of drm/i915_drm.h; that was previously included just\n for the definition of \u0027enum port\u0027 which this patch removes the need\n for. (Jose)\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190709183934.445-4-matthew.d.roper@intel.com\n\ncommit befa372b990a3c02fde475be2a457d2c8fceac69\nAuthor: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\nDate: Tue Jul 9 11:39:31 2019 -0700\n\n drm/i915/gen11: Program ICL_DPCLKA_CFGCR0 according to PHY\n \n Although the register name implies that it operates on DDI\u0027s,\n DPCLKA_CFGCR0_ICL actually needs to be programmed according to the PHY\n that\u0027s in use. I.e., when using EHL\u0027s DDI-D on combo PHY A, the bits\n described as \"port A\" in the bspec are what we need to set. The bspec\n clarifies:\n \n \"[For EHL] DDID clock tied to DDIA clock, so DPCLKA_CFGCR0 DDIA\n Clock Select chooses the PLL for both DDIA and DDID and drives\n port A in all cases.\"\n \n Also, since the CNL DPCLKA_CFGCR0 bit defines are still port-based, we\n create separate ICL-specific defines that accept the PHY rather than\n trying to share the same bit definitions between CNL and ICL.\n \n v5: Make icl_dpclka_cfgcr0_clk_off() take phy rather than port. When\n splitting the original patch the hunk to handle this wound up too\n late in the series. (Sparse)\n \n v6: Since we\u0027re already changing this code,\n s/DPCLKA_CFGCR0_ICL/ICL_DPCLKA_CFGCR0/ for consistency. (Jose)\n \n Bspec: 33148\n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190709183934.445-3-matthew.d.roper@intel.com\n\ncommit 358633e71c07475c75476058a289e2ef8aa74be4\nAuthor: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\nDate: Tue Jul 9 11:39:30 2019 -0700\n\n drm/i915/gen11: Start distinguishing \u0027phy\u0027 from \u0027port\u0027\n \n Our past DDI-based Intel platforms have had a fixed DDI\u003c-\u003ePHY mapping.\n Because of this, both the bspec documentation and our i915 code has used\n the term \"port\" when talking about either DDI\u0027s or PHY\u0027s; it was always\n easy to tell what terms like \"Port A\" were referring to from the\n context.\n \n Unfortunately this is starting to break down now that EHL allows PHY-A\n to be driven by either DDI-A or DDI-D. Is a setup with DDI-D driving\n PHY-A considered \"Port A\" or \"Port D?\" The answer depends on which\n register we\u0027re working with, and even the bspec doesn\u0027t do a great job\n of clarifying this.\n \n Let\u0027s try to be more explicit about whether we\u0027re talking about the DDI\n or the PHY on gen11+ by using \u0027port\u0027 to refer to the DDI and creating a\n new \u0027enum phy\u0027 namespace to refer to the PHY in use.\n \n This patch just adds the new PHY namespace, new phy-based versions of\n intel_port_is_*(), and a helper to convert a port to a PHY.\n Transitioning various areas of the code over to using the PHY namespace\n will be done in subsequent patches to make review easier. We\u0027ll remove\n the intel_port_is_*() functions at the end of the series when we\n transition all callers over to using the PHY-based versions.\n \n v2:\n - Convert a few more \u0027port\u0027 uses to \u0027phy.\u0027 (Sparse)\n \n v3:\n - Switch DDI_CLK_SEL() back to \u0027port.\u0027 (Jose)\n - Add a code comment clarifying why DPCLKA_CFGCR0_ICL needs to use PHY\n for its bit definitions, even though the register description is\n given in terms of DDI.\n - To avoid confusion, switch CNL\u0027s DPCLKA_CFGCR0 defines back to using\n port and create separate ICL+ definitions that work in terms of PHY.\n \n v4:\n - Rebase and resolve conflicts with Imre\u0027s TC series.\n - This patch now just adds the namespace and a few convenience\n functions; the important changes are now split out into separate\n patches to make review easier.\n \n Suggested-by: Ville Syrjala \u003cville.syrjala@linux.intel.com\u003e\n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Imre Deak \u003cimre.deak@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Signed-off-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190709183934.445-2-matthew.d.roper@intel.com\n\ncommit 9d44dcb9088aadd11718a7ad5a8567d6b8a1174a\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Mon Jul 8 10:28:14 2019 -0700\n\n drm/i915: move intel_ddi_set_fia_lane_count to intel_tc.c\n \n PORT_TX_DFLEXDPMLE1 is a FIA register so move it to intel_tc.c where we\n access other FIA registers. In Tiger Lake we have multiple/modular FIAs\n so it makes sense to start moving all access to their registers to a\n common place.\n \n While at it, make it clear that we will only ever call this function\n for ports with TC phy. Previously we were relying on tc_mode being\n TC_PORT_TBT_ALT for combo phy ports. However it\u0027s confusing since in\n this same function we have checks for is_tc_port. Also, if we manage to\n make each phy access only their own field, we may in future add them as\n a union inside intel_digital_port.\n \n v2: Fix coding style while moving the code\n \n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190708172815.6814-4-lucas.demarchi@intel.com\n\ncommit c99fd7b3908e287ccd20be4f9eefe597c7d354fd\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Tue Jul 9 08:54:03 2019 -0700\n\n drm/i915: fix include order in intel_tc.*\n \n Separate local includes with a blank line and sort the groups\n alphabetically.\n \n v2: don\u0027t make intel_tc.h be the first include\n v3: don\u0027t make local includes be included first\n \n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190709155403.29370-1-lucas.demarchi@intel.com\n\ncommit b41e434fdc423995b99cb79f86ec7ec428f5f379\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Mon Jul 8 10:28:12 2019 -0700\n\n drm/i915: make new intel_tc.c use uncore accessors\n \n Let\u0027s make the just created intel_tc.c already follow the trend of using\n i915 instead of dev_priv and calling the intel_uncore_*() functions.\n \n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190708172815.6814-2-lucas.demarchi@intel.com\n\ncommit 4f5368b5541a902f6596558b05f5c21a9770dd32\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 08:17:23 2019 +0200\n\n drm/kms: Catch mode_object lifetime errors\n \n Only dynamic mode objects, i.e. those which are refcounted and have a free\n callback, can be added while the overall drm_device is visible to\n userspace. All others must be added before drm_dev_register and\n removed after drm_dev_unregister.\n \n Small issue around drivers still using the load/unload callbacks, we\n need to make sure we set dev-\u003eregistered so that load/unload code in\n these callbacks doesn\u0027t trigger false warnings. Only a small\n adjustement in drm_dev_register was needed.\n \n Motivated by some irc discussions about object ids of dynamic objects\n like blobs become invalid, and me going on a bit an audit spree.\n \n Reviewed-by: Sean Paul \u003csean@poorly.run\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614061723.1173-1-daniel.vetter@ffwll.ch\n\ncommit caceff96d5b7b487da19c0916e0048ff8f1709a9\nAuthor: Uma Shankar \u003cuma.shankar@intel.com\u003e\nDate: Fri Jun 28 13:32:30 2019 +0530\n\n drm/i915/icl: Fixed Input CSC Co-efficients for BT601/709\n \n Input CSC Co-efficients for BT601 and BT709 YCbCR to RGB\n conversion were slightly off. Fixed the same.\n \n v2: Fixed the co-eficients as there was issue with reference\n matrix, spotted by Ville.\n \n v3: Rebase\n \n Signed-off-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628080230.27492-4-uma.shankar@intel.com\n\ncommit 9856bf7f779a6613bad0a9d628b2e82e00fdec66\nAuthor: Uma Shankar \u003cuma.shankar@intel.com\u003e\nDate: Fri Jun 28 13:32:29 2019 +0530\n\n drm/i915/icl: Fix Y pre-offset for Full Range YCbCr\n \n Fixed Y Pre-offset in case of Full Range YCbCr.\n \n v2: Rebase\n \n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Suggested-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628080230.27492-3-uma.shankar@intel.com\n\ncommit 27ee72477c9b77ebf0e2827b8679b547ab556673\nAuthor: Uma Shankar \u003cuma.shankar@intel.com\u003e\nDate: Fri Jun 28 13:32:28 2019 +0530\n\n drm/i915/icl: Handle YCbCr to RGB conversion for BT2020 case\n \n Currently input csc for YCbCR to RGB conversion handles only\n BT601 and Bt709. Extending it to support BT2020 as well.\n \n v2: Fixed the co-efficients for LR to FR conversion,\n as suggested by Ville.\n \n v3: Fixed Y Pre-offset in case of Full Range YCbCr as suggested\n by Ville.\n \n v4: Split the v2 and v3 changes.\n \n v5: Rebase\n \n v6: Fix a rebase fumble.\n \n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Signed-off-by: Shashank Sharma \u003cshashank.sharma@intel.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628080230.27492-2-uma.shankar@intel.com\n\ncommit 14bfcd3e0daeb0f757a02aac85fd03e0933ab37e\nAuthor: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\nDate: Wed Jul 10 11:55:24 2019 +0100\n\n drm/i915/perf: add missing delay for OA muxes configuration\n \n This was dropped from the original patch series, we weren\u0027t sure\n whether it was needed at the time. More recent tests show it\u0027s\n definitely needed to have acurate performance data.\n \n Signed-off-by: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Fixes: 19f81df2859eb1 (\"drm/i915/perf: Add OA unit support for Gen 8+\")\n Acked-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n [ickle: combine duplicate code and comments]\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190710105524.23017-1-chris@chris-wilson.co.uk\n\ncommit 58d1b427148a3bd16e62f16558d2f45527727eb1\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jul 10 07:44:41 2019 +0100\n\n drm/i915/execlists: Record preemption for selftests\n \n Put back the preemption counters lost in commit 22b7a426bbe1\n (\"drm/i915/execlists: Preempt-to-busy\") so that our selftests that\n assert no preemption took place continue to function.\n \n v2: But a timeslice is only a \"soft\" preemption!\n \n Fixes: 22b7a426bbe1 (\"drm/i915/execlists: Preempt-to-busy\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190710064454.682-1-chris@chris-wilson.co.uk\n\ncommit b33c61db7291d376e956f2657c453d966e558974\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Tue Jul 9 10:31:05 2019 +0100\n\n drm/i915: Remove unused i915_gem_context_lookup_engine\n \n There are no known plans to start using it either.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190709093105.24699-1-tvrtko.ursulin@linux.intel.com\n\ncommit 2a98f4e65bba2db83489aaadd3f9db26547cbd35\nAuthor: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\nDate: Tue Jul 9 17:42:27 2019 +0100\n\n drm/i915: add infrastructure to hold off preemption on a request\n \n We want to set this flag in the next commit on requests containing\n perf queries so that the result of the perf query can just be a delta\n of global counters, rather than doing post processing of the OA\n buffer.\n \n Signed-off-by: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n [ickle: add basic selftest for nopreempt]\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190709164227.25859-1-chris@chris-wilson.co.uk\n\ncommit 46c5847e3d97f5afced32f7474e7daea819da7c9\nAuthor: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\nDate: Tue Jul 9 15:33:43 2019 +0300\n\n drm/i915: enumerate scratch fields\n \n We have a bunch of offsets in the scratch buffer. As we\u0027re about to\n add some more, let\u0027s group all of the offsets in a common location.\n \n Signed-off-by: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190709123351.5645-6-lionel.g.landwerlin@intel.com\n\ncommit a5af1df716c123a09341351008fc497bea137b77\nAuthor: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\nDate: Tue Jul 9 15:33:39 2019 +0300\n\n drm/i915/perf: ensure we keep a reference on the driver\n \n The i915 perf stream has its own file descriptor and is tied to\n reference of the driver. We haven\u0027t taken care of keep the driver\n alive.\n \n Signed-off-by: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Suggested-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Fixes: eec688e1420da5 (\"drm/i915: Add i915 perf infrastructure\")\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190709123351.5645-2-lionel.g.landwerlin@intel.com\n\ncommit 681c774d349018c0f47a770133b952f50328d957\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jul 9 09:17:18 2019 +0100\n\n drm/i915/userptr: Don\u0027t mark readonly objects as dirty\n \n If we map an object as readonly into the GTT, we know that the GPU\n cannot have written to it and so the object is not dirty and we don\u0027t\n need to flush the writes back to the system.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190709081718.27843-1-chris@chris-wilson.co.uk\n\ncommit f7ddc80ecb0e19e2c706747d1540b7699b4198d3\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Mon Jul 8 17:07:35 2019 +0300\n\n drm/i915/icl: Clear the shared port PLLs from the new crtc state\n \n For consistency clear the icl_port_dplls from the new crtc state, when\n releasing the DPLLs from the old crtc state. Leaving them set could\n result in releasing the same PLLs multiple times from the same CRTC\n state incorrectly (if the same CRTC was first used for a TypeC port then\n for a combo PHY port).\n \n Leaving the stale pointers behind happens not to cause a problem atm\n (since the incorrect releasing will be a NOP), but we need to fix that\n for consistency.\n \n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190708140735.20198-2-imre.deak@intel.com\n\ncommit 5c28e3a567fed9c22cb79f01294b1aaefee344fd\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Mon Jul 8 17:07:34 2019 +0300\n\n drm/i915: Clear the shared PLL from the put_dplls() hook\n \n For symmetry with the get_dplls() hook which sets the shared_dpll\n pointer clear the same pointer from the put_dplls() hook.\n \n While at it also constify the old crtc state.\n \n v2:\n - Constify the old crtc state. (Ville)\n \n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190708140735.20198-1-imre.deak@intel.com\n\ncommit a46f4e9e5ae99cc6da81d84b6eb4ec7c19a81a2f\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Jun 19 21:03:12 2019 +0300\n\n drm/i915/sdvo: Add helpers to get the cmd/status string\n \n Add sdvo_cmd_name() and sdvo_cmd_status() helpers to avoid bothering\n the callers with the implementation details of the storage for these\n strings.\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619180312.31817-6-ville.syrjala@linux.intel.com\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n\ncommit 7b8062ea60d9e207e42c58ff714c911b01f4c99a\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Jun 19 21:03:10 2019 +0300\n\n drm/i915/sdvo: Shrink sdvo_cmd_names[] strings\n \n Drop the redundant \"SDVO_CMD_\" prefix from the command name\n strings in sdvo_cmd_names[].\n \n While at it throw away the unused struct name, and undef\n SDVO_CMD_NAME_ENTRY() when we\u0027re done.\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619180312.31817-4-ville.syrjala@linux.intel.com\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n\ncommit 8a9c8028399e47ced073e13890ce8912c655ef7e\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Jun 19 21:03:09 2019 +0300\n\n drm/i915/sdvo: Remove duplicate SET_INPUT_TIMINGS_PART1 cmd name string\n \n sdvo_cmd_names[] contains two entries for SET_INPUT_TIMINGS_PART1.\n Remove one of them.\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619180312.31817-3-ville.syrjala@linux.intel.com\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n\ncommit c598a66491cff68237cb3221fa444e4ca9f0d7f7\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Jun 19 21:03:08 2019 +0300\n\n drm/i915/sdvo: Use named initializers for the SDVO command names\n \n Use named initializers to make it easier to associate the SDVO debug\n prints with the SDVO command defines. Also switch to using ARRAY_SIZE()\n instead of assuming that SDVO_CMD_STATUS_SCALING_NOT_SUPP is the last\n command type.\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619180312.31817-2-ville.syrjala@linux.intel.com\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n\ncommit f384e48d7641ef284ae760d150c80920ffc3adbb\nAuthor: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\nDate: Thu Jun 27 20:54:57 2019 +0530\n\n drm/i915: Add icl mipi dsi properties\n \n Add scaling and panel orientation properties for\n icl mipi dsi.\n \n v2: Add platform specific function (Ville)\n v3: Remove redundant check and update scaler call (Jani, Ville)\n \n Signed-off-by: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190627152457.26146-1-vandita.kulkarni@intel.com\n\ncommit 4fb76782ad012d657df51f437a0dfcebce6aea53\nAuthor: Ramalingam C \u003cramalingam.c@intel.com\u003e\nDate: Mon Jul 8 17:03:19 2019 +0530\n\n drm/i915/hdcp: debug logs for sink related failures\n \n Adding few more debug logs to identify the sink specific HDCP failures\n along with a out of mem failure.\n \n v2:\n Capturing the Bug and a-b\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d110991\n Signed-off-by: Ramalingam C \u003cramalingam.c@intel.com\u003e\n Signed-off-by: Uma Shankar \u003cuma.shankar@intel.com\u003e\n cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Acked-by: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190708113319.2517-1-ramalingam.c@intel.com\n\ncommit d006a9b63d4bbb3eb1e2e616712d45acfcda7042\nAuthor: Denis Efremov \u003cefremov@linux.com\u003e\nDate: Wed Jul 3 20:01:50 2019 +0300\n\n drm/client: remove the exporting of drm_client_close\n \n The function drm_client_close is declared as static and marked as\n EXPORT_SYMBOL. It\u0027s a bit confusing for an internal function to be\n exported. The area of visibility for such function is its .c file\n and all other modules. Other *.c files of the same module can\u0027t use it,\n despite all other modules can. Relying on the fact that this is the\n internal function and it\u0027s not a crucial part of the API, the patch\n removes the EXPORT_SYMBOL marking of drm_client_close.\n \n Signed-off-by: Denis Efremov \u003cefremov@linux.com\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703170150.32548-1-efremov@linux.com\n\ncommit 90f479ae51afa45efab97afdde9b94b9660dd3e4\nAuthor: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\nDate: Wed Jul 3 09:58:42 2019 +0200\n\n drm/mgag200: Replace struct mga_fbdev with generic framebuffer emulation\n \n This patch replaces mgag200\u0027s framebuffer console with DRM\u0027s generic\n implememtation. All respective code is being removed from the driver.\n \n The console is set up with a shadow buffer. The actual buffer object is\n not permanently pinned in video ram, but just another buffer object that\n the driver moves in and out of vram as necessary. The driver\u0027s function\n mga_crtc_do_set_base() used to contain special handling for the framebuffer\n console. With the new generic framebuffer, the driver does not need this\n code an longer.\n \n For consistency, this patch also changes the preferred framebuffer depth.\n The original code used 24 bpp by default and 32 bpp for the framebuffer. As\n 24 bpp is not well supported by userspace anyway, setting 32 bpp as default\n makes sense.\n \n v2:\n * rely on fbdev helpers error messages\n * document changes to preferred depth\n * dirty function no longer required\n \n Signed-off-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Acked-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Link: https://patchwork.freedesktop.org/patch/315832/\n\ncommit f1f8555dfb9a70a2f42e7ee578143dcfd5abb02b\nAuthor: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\nDate: Wed Jul 3 09:58:34 2019 +0200\n\n drm/bochs: Use shadow buffer for bochs framebuffer console\n \n The bochs driver (and virtual hardware) requires buffer objects to\n reside in video ram to display them to the screen. So it can not\n display the framebuffer console because the respective buffer object\n is permanently pinned in system memory.\n \n Using a shadow buffer for the console solves this problem. The console\n emulation will pin the buffer object only during updates from the shadow\n buffer. Otherwise, the bochs driver can freely relocated the buffer\n between system memory and video ram.\n \n v2:\n * select shadow FB via struct drm_mode_config.prefer_shadow_fbdev\n \n Signed-off-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Acked-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Link: https://patchwork.freedesktop.org/patch/315833/\n\ncommit 8a99de3dc0dd417e1f46c8e27e63b1363ecfb969\nAuthor: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\nDate: Wed Jul 3 09:58:27 2019 +0200\n\n drm/ast: Replace struct ast_fbdev with generic framebuffer emulation\n \n This patch replaces ast\u0027s framebuffer console with DRM\u0027s generic\n implememtation. All respective code is being removed from the driver.\n \n The console is set up with a shadow buffer. The actual buffer object is\n not permanently pinned in video ram, but just another buffer object that\n the driver moves in and out of vram as necessary. The driver\u0027s function\n ast_crtc_do_set_base() used to contain special handling for the framebuffer\n console. With the new generic framebuffer, the driver does not need this\n code an longer.\n \n v2:\n * use drm_fb_helper_set_suspend_unlocked() in ast_drm_{thaw,freeze}()\n * dirty function no longer required\n \n Signed-off-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Acked-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Link: https://patchwork.freedesktop.org/patch/315835/\n\ncommit f458579ede0b721418f7fc9f0eb624ca058ea0ea\nAuthor: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\nDate: Fri Jul 5 09:31:00 2019 +0200\n\n drm/fb-helper: Instanciate shadow FB if configured in device\u0027s mode_config\n \n Generic framebuffer emulation uses a shadow buffer for framebuffers with\n dirty() function. If drivers want to use the shadow FB without such a\n function, they can now set prefer_shadow or prefer_shadow_fbdev in their\n mode_config structures. The former flag is exported to userspace, the\n latter flag is fbdev-only.\n \n v3:\n * only schedule dirty worker if fbdev uses shadow fb\n * test shadow fb settings with boolean operators\n * use bool for struct drm_mode_config.prefer_shadow_fbdev\n * fix documentation comments\n \n Signed-off-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Reviewed-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Tested-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Link: https://patchwork.freedesktop.org/patch/315834/\n\ncommit cf1ca9aeb930df074bb5bbcde55f935fec04e529\nAuthor: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\nDate: Wed Jul 3 09:58:24 2019 +0200\n\n drm/fb-helper: Map DRM client buffer only when required\n \n This patch changes DRM clients to not map the buffer by default. The\n buffer, like any buffer object, should be mapped and unmapped when\n needed.\n \n An unmapped buffer object can be evicted to system memory and does\n not consume video ram until displayed. This allows to use generic fbdev\n emulation with drivers for low-memory devices, such as ast and mgag200.\n \n This change affects the generic framebuffer console. HW-based consoles\n map their console buffer once and keep it mapped. Userspace can mmap this\n buffer into its address space. The shadow-buffered framebuffer console\n only needs the buffer object to be mapped during updates. While not being\n updated from the shadow buffer, the buffer object can remain unmapped.\n Userspace will always mmap the shadow buffer.\n \n v2:\n * change DRM client to not map buffer by default\n * manually map client buffer for fbdev with HW framebuffer\n \n Signed-off-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Reviewed-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Link: https://patchwork.freedesktop.org/patch/315830/\n\ncommit 15dd0fc833407a513da162451ac3459650beecb6\nAuthor: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\nDate: Wed Jul 3 09:58:18 2019 +0200\n\n drm/client: Support unmapping of DRM client buffers\n \n DRM clients, such as the fbdev emulation, have their buffer objects\n mapped by default. Mapping a buffer implicitly prevents its relocation.\n Hence, the buffer may permanently consume video memory while it\u0027s\n allocated. This is a problem for drivers of low-memory devices, such as\n ast, mgag200 or older framebuffer hardware, which will then not have\n enough memory to display other content (e.g., X11).\n \n This patch introduces drm_client_buffer_vmap() and _vunmap(). Internal\n DRM clients can use these functions to unmap and remap buffer objects\n as needed.\n \n There\u0027s no reference counting for vmap operations. Callers are expected\n to either keep buffers mapped (as it is now), or call vmap and vunmap\n in pairs around code that accesses the mapped memory.\n \n v2:\n * remove several duplicated NULL-pointer checks\n v3:\n * style and typo fixes\n \n Signed-off-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Reviewed-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Link: https://patchwork.freedesktop.org/patch/315831/\n\ncommit 113348d86cebbf9fb07e190e6075562ee431eb15\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Mon Jul 8 19:20:38 2019 +0300\n\n drm/sti: Try to fix up the tvout possible clones\n \n The current possible_clones setup doesn\u0027t look sensible. I\u0027m assuming\n the 0 and 1 are supposed to refer to the indexes of the hdmi and hda\n encoders? So it kinda looks like we want hda+hdmi cloning, but then\n dvo also claims to be cloneable with hdmi, but hdmi won\u0027t recipricate.\n \n Benjamin tells me all encoders should be cloneable with each other,\n so let\u0027s fix up the masks to indicate that.\n \n Cc: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Cc: Vincent Abriou \u003cvincent.abriou@st.com\u003e\n Acked-by: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190708162048.4286-5-ville.syrjala@linux.intel.com\n\ncommit 6234ba980f89b396221f77788a9bb36e79900391\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Mon Jul 8 19:20:37 2019 +0300\n\n drm/sti: Remove pointless casts\n \n There\u0027s no point in the cast for accessing the base class. Just\n take the address of the struct instead.\n \n Cc: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Cc: Vincent Abriou \u003cvincent.abriou@st.com\u003e\n Acked-by: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190708162048.4286-4-ville.syrjala@linux.intel.com\n\ncommit 4a5fdc962fc4ce143791496def78c2acc3e51810\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 5 13:43:25 2019 +0100\n\n drm/i915/gt: Remove presumption of RCS0\n \n We now track features correctly instead of probing i915-\u003eengine[RCS0]\n which is much more flexible and avoids any nasty surprises.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705124325.14270-2-chris@chris-wilson.co.uk\n\ncommit 7c6d6867e9723e9548ed5ae0c6b070652f4d43e3\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 5 13:43:24 2019 +0100\n\n drm/i915/gt: Apply RCS workarounds to the render class\n \n Treat all render engines to the RCS workarounds, simply to avoid using\n engine-\u003eid when we are trying to think in terms of classes.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Stuart Summers \u003cstuart.summers@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705124325.14270-1-chris@chris-wilson.co.uk\n\ncommit cbcec57e9dd75c5bb857ead091e471ecb25afd1d\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Jul 8 12:30:38 2019 +0100\n\n drm/i915/selftests: Fill in a little more of the dummy fence\n \n Initialise the dma_fence innards in preparation for making\n dma_fence_signal() always check the callback list.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190708113038.19251-1-chris@chris-wilson.co.uk\n\ncommit cb6d7c7dc7ff8cace666ddec66334117a6068ce2\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Jul 8 15:03:27 2019 +0100\n\n drm/i915/userptr: Acquire the page lock around set_page_dirty()\n \n set_page_dirty says:\n \n For pages with a mapping this should be done under the page lock\n for the benefit of asynchronous memory errors who prefer a\n consistent dirty state. This rule can be broken in some special\n cases, but should be better not to.\n \n Under those rules, it is only safe for us to use the plain set_page_dirty\n calls for shmemfs/anonymous memory. Userptr may be used with real\n mappings and so needs to use the locked version (set_page_dirty_lock).\n \n Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id\u003d203317\n Fixes: 5cc9ed4b9a7a (\"drm/i915: Introduce mapping of user pages into video memory (userptr) ioctl\")\n References: 6dcc693bc57f (\"ext4: warn when page is dirtied without buffers\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: stable@vger.kernel.org\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190708140327.26825-1-chris@chris-wilson.co.uk\n\ncommit baf08ed50a614f21398f30e11d4b6ba2ffecc36e\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Jul 8 22:55:24 2019 +0100\n\n drm/i915/selftests: Set igt_spinner.gt for early exit\n \n Set up a default gt pointer for an early cleanup of igt_spinnter, before\n a request is created and igt_spinner.gt set to the active engine\u0027s.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190708215524.31639-1-chris@chris-wilson.co.uk\n\ncommit 9f02b8f61f29f4518581770d57bfffe99b1ea599\nAuthor: Rouven Czerwinski \u003cr.czerwinski@pengutronix.de\u003e\nDate: Wed Jul 3 12:03:50 2019 +0200\n\n tee: optee: add might_sleep for RPC requests\n \n If the kernel is compiled with CONFIG_PREEMPT_VOLUNTARY and OP-TEE is\n executing a long running workload, the following errors are raised:\n \n [ 1705.971228] rcu: INFO: rcu_sched detected stalls on CPUs/tasks:\n [ 1705.977195] rcu: (detected by 0, t\u003d2102 jiffies, g\u003d51977, q\u003d3)\n [ 1705.983152] rcu: All QSes seen, last rcu_sched kthread activity 2102 (140596-138494), jiffies_till_next_fqs\u003d1, root -\u003eqsmask 0x0\n [ 1705.994729] optee-xtest R running task 0 169 157 0x00000002\n \n While OP-TEE is returning regularly to the kernel due to timer\n interrupts, the OPTEE_SMC_FUNC_FOREIGN_INTR case does not contain an\n explicit rescheduling point. Add a might_sleep() to the RPC request case\n to ensure that the kernel can reschedule another task if OP-TEE requests\n RPC handling.\n \n Signed-off-by: Rouven Czerwinski \u003cr.czerwinski@pengutronix.de\u003e\n Signed-off-by: Jens Wiklander \u003cjens.wiklander@linaro.org\u003e\n\ncommit a17ce803dffa8e34fda39f8af721a4005c451fef\nAuthor: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\nDate: Mon Jul 8 13:09:06 2019 -0700\n\n drm/i915: Update DRIVER_DATE to 20190708\n \n Signed-off-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n\ncommit 8090858c87ea6e863548cbbafe26b49793bf11a1\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Fri Jul 5 10:42:21 2019 -0600\n\n dt-bindings: display: Convert sgd,gktw70sdae4se panel to DT schema\n \n Convert the sgd,gktw70sdae4se LVDS panel binding to DT schema.\n \n Cc: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705164221.4462-14-robh@kernel.org\n\ncommit 2fb9bf9aeffdaeb6fc18e35970b761417f230c9f\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Fri Jul 5 10:42:20 2019 -0600\n\n dt-bindings: display: Convert mitsubishi,aa121td01 panel to DT schema\n \n Convert the mitsubishi,aa121td01 LVDS panel binding to DT schema.\n \n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705164221.4462-13-robh@kernel.org\n\ncommit ea8e4568d5ff2ece0dd293fda3672d20180c250b\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Fri Jul 5 10:42:19 2019 -0600\n\n dt-bindings: display: Convert mitsubishi,aa104xd12 panel to DT schema\n \n Convert the mitsubishi,aa104xd12 LVDS panel binding to DT schema.\n \n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705164221.4462-12-robh@kernel.org\n\ncommit b4eefa69b63564f2c6c764fedf8c904f2af801da\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Fri Jul 5 10:42:18 2019 -0600\n\n dt-bindings: display: Convert innolux,ee101ia-01 panel to DT schema\n \n Convert the innolux,ee101ia-01 LVDS panel binding to DT schema.\n \n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: dri-devel@lists.freedesktop.org\n Reviewed-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705164221.4462-11-robh@kernel.org\n\ncommit ae18c1176f0d9317c5c2e841d1fdfc218e6b50fb\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Fri Jul 5 10:42:17 2019 -0600\n\n dt-bindings: display: Convert panel-lvds to DT schema\n \n Convert the panel-lvds binding to use DT schema. The panel-lvds schema\n inherits from the panel-common.yaml schema and specific LVDS panel\n bindings should inherit from this schema.\n \n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: dri-devel@lists.freedesktop.org\n Reviewed-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705164221.4462-10-robh@kernel.org\n\ncommit 87548d750c316485a1da903b2cfc69918ebc79bd\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Fri Jul 5 10:42:16 2019 -0600\n\n dt-bindings: display: Convert tfc,s9700rtwv43tr-01b panel to DT schema\n \n Convert the tfc,s9700rtwv43tr-01b panel binding to DT schema.\n \n Cc: Heiko Stuebner \u003cheiko.stuebner@bq.com\u003e\n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: dri-devel@lists.freedesktop.org\n Reviewed-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705164221.4462-9-robh@kernel.org\n\ncommit 78ad29d76229309743e62d017582418e5aaf8487\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Fri Jul 5 10:42:15 2019 -0600\n\n dt-bindings: display: Convert raspberrypi,7inch-touchscreen panel to DT schema\n \n Convert the raspberrypi,7inch-touchscreen panel binding to DT schema.\n \n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: dri-devel@lists.freedesktop.org\n Reviewed-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705164221.4462-8-robh@kernel.org\n\ncommit 746fe0b26796d180537402b6dd16e18e431cd1ce\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Fri Jul 5 10:42:14 2019 -0600\n\n dt-bindings: display: Convert pda,91-00156-a0 panel to DT schema\n \n Convert the pda,91-00156-a0 panel binding to DT schema.\n \n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: dri-devel@lists.freedesktop.org\n Reviewed-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705164221.4462-7-robh@kernel.org\n\ncommit 6c9b89e3cdd473c6f72250d9f386ef1d624113eb\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Fri Jul 5 10:42:13 2019 -0600\n\n dt-bindings: display: Convert dlc,dlc0700yzg-1 panel to DT schema\n \n Convert the dlc,dlc0700yzg-1 panel binding to DT schema.\n \n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: dri-devel@lists.freedesktop.org\n Reviewed-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705164221.4462-6-robh@kernel.org\n\ncommit 56b40739186d944d5fc8973e471fb7dcefc0efb0\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Fri Jul 5 10:42:12 2019 -0600\n\n dt-bindings: display: Convert bananapi,s070wv20-ct16 panel to DT schema\n \n Convert the bananapi,s070wv20-ct16 panel binding to DT schema.\n \n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: dri-devel@lists.freedesktop.org\n Reviewed-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705164221.4462-5-robh@kernel.org\n\ncommit 3cd33c8c54af54528bf509b709432767fb6632b9\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Fri Jul 5 10:42:11 2019 -0600\n\n dt-bindings: display: Convert armadeus,st0700-adapt panel to DT schema\n \n Convert the armadeus,st0700-adapt panel binding to DT schema.\n \n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: dri-devel@lists.freedesktop.org\n Reviewed-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705164221.4462-4-robh@kernel.org\n\ncommit 09ed3b2574f26efacaed0384a8e7a8b80aedc0d2\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Fri Jul 5 10:42:10 2019 -0600\n\n dt-bindings: display: Convert ampire,am-480272h3tmqw-t01h panel to DT schema\n \n Convert the ampire,am-480272h3tmqw-t01h panel binding to DT schema.\n \n Cc: Yannick Fertre \u003cyannick.fertre@st.com\u003e\n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: dri-devel@lists.freedesktop.org\n Reviewed-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705164221.4462-3-robh@kernel.org\n\ncommit 821a1f7171aeea5e7818d30f70d7cf49b007357f\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Fri Jul 5 10:42:09 2019 -0600\n\n dt-bindings: display: Convert common panel bindings to DT schema\n \n Convert the common panel bindings to DT schema consolidating scattered\n definitions to a single schema file.\n \n The \u0027simple-panel\u0027 binding just a collection of properties and not a\n complete binding itself. All of the \u0027simple-panel\u0027 properties are\n covered by the panel-common.txt binding with the exception of the\n \u0027no-hpd\u0027 property, so add that to the schema.\n \n As there are lots of references to simple-panel.txt, just keep the file\n with a reference to common.yaml for now until all the bindings are\n converted.\n \n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: dri-devel@lists.freedesktop.org\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Reviewed-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705164221.4462-2-robh@kernel.org\n\ncommit cf3bd1a0f53eae790a83af445c7795b4ce2e1a9b\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Jul 8 16:23:21 2019 +0100\n\n drm/i915/selftests: Reorder error cleanup for whitelist checking\n \n Reorder the error paths so that we unwind all the locals from any error\n path and so avoid setting off divers alarum in case we find an error in\n case we find an error.\n \n References: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111048\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190708152321.22187-1-chris@chris-wilson.co.uk\n\ncommit 77adbd8fbf3a10e149a76abee6384a4968ed318e\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Jul 8 16:49:14 2019 +0100\n\n drm/i915: Explicitly track active fw_domain timers\n \n Stop guessing over whether we have an extra wakeref held by the delayed\n fw put, and track it explicitly for the sake of debug.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190708154914.26850-1-chris@chris-wilson.co.uk\n\ncommit 15e7facb7b428f5f0dac3ee85ecb519a46caaa34\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Sun Jul 7 16:11:35 2019 +0100\n\n drm/i915: Pull assert_forcewake_active() underneath the lock\n \n Make no assumption that something in the background is not acquiring the\n fw_domain -- but we still do not track owner so assume that any active\n domain is intended by the caller.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190707151135.11700-1-chris@chris-wilson.co.uk\n\ncommit 50b38bc4d39aa6fa5639aabcccfb828c19b89278\nAuthor: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\nDate: Fri Jul 5 22:52:04 2019 +0100\n\n drm/i915/gtt: Introduce release_pd_entry\n \n By encapsulating the locking upper level and used check for entry\n into a helper function, we can use it in all callsites.\n \n v2: get rid of atomic_reads on lower level clears (Chris)\n \n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705215204.4559-4-chris@chris-wilson.co.uk\n\ncommit 73a8fdef5d6f525932a361104c0d46e5e2585c28\nAuthor: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\nDate: Fri Jul 5 22:52:03 2019 +0100\n\n drm/i915/gtt: Setup phys pages for 3lvl pdps\n \n If we setup backing phys page for 3lvl pdps, as they\n are not used, we will lose 5 pages per ppgtt.\n \n Trading this memory on bsw, we gain more common code paths for all\n gen8+ directory manipulation. And those paths are now void of checks\n for page directory type, making the hot paths faster.\n \n v2: don\u0027t shortcut vm (Chris)\n \n Signed-off-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705215204.4559-3-chris@chris-wilson.co.uk\n\ncommit 72230b874a2071593e9f493bee072e41cc734646\nAuthor: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\nDate: Fri Jul 5 22:52:02 2019 +0100\n\n drm/i915/gtt: Tear down setup and cleanup macros for page dma\n \n We don\u0027t use common codepaths to setup and cleanup page\n directories vs page tables. So their setup and cleanup macros\n are of no use and can be removed.\n \n Signed-off-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705215204.4559-2-chris@chris-wilson.co.uk\n\ncommit f20f272f104f224010b955c2c49b0dcb4ca02e4f\nAuthor: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\nDate: Fri Jul 5 22:52:01 2019 +0100\n\n drm/i915/gtt: pde entry encoding is identical\n \n For all page directory entries, the pde encoding is\n identical. Don\u0027t complicate call sites with different\n versions of doing the same thing, so we always check the\n existence of physical page before writing the entry into\n it. This further generalizes the pd so that manipulation in\n callsites will be identical, removing the need to handle\n pdps differently for gen8.\n \n v2: squash\n v3: inc/dec with set/clear (Chris)\n v4: inlines, warn, stray set_pd (Chris)\n \n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Signed-off-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705215204.4559-1-chris@chris-wilson.co.uk\n\ncommit 3e27d70bcc46c68aceb5dafc813db4c4ff675221\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Jul 5 11:31:38 2019 +0000\n\n drm/i915: Remove set but not used variable \u0027intel_dig_port\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/gpu/drm/i915/display/intel_ddi.c: In function \u0027intel_ddi_get_config\u0027:\n drivers/gpu/drm/i915/display/intel_ddi.c:3774:29: warning:\n variable \u0027intel_dig_port\u0027 set but not used [-Wunused-but-set-variable]\n struct intel_digital_port *intel_dig_port;\n \n It is never used, so can be removed.\n \n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705113138.65880-1-yuehaibing@huawei.com\n\ncommit 9d1bc13e7fe1b31d93c58b6811f426b9b7dea59f\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Jul 5 11:31:12 2019 +0000\n\n drm/i915: Remove set but not used variable \u0027encoder\u0027\n \n Fixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n \n drivers/gpu/drm/i915/display/intel_dp.c: In function \u0027intel_dp_set_drrs_state\u0027:\n drivers/gpu/drm/i915/display/intel_dp.c:6623:24: warning:\n variable \u0027encoder\u0027 set but not used [-Wunused-but-set-variable]\n \n It\u0027s never used, so can be removed.Also remove related\n variable \u0027dig_port\u0027\n \n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705113112.64715-1-yuehaibing@huawei.com\n\ncommit badf1f2724d430c8a310be549b49f8333817dbeb\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jul 5 08:45:57 2019 +0100\n\n drm/i915: Order assert forcewake test\n \n Read the current value before computing the expected to ensure that if\n the timer does complete early (against our will), it should not cause a\n false positive.\n \n v2: The local irq disable did not prevent the timer from running.\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111074\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190705074604.16496-1-chris@chris-wilson.co.uk\n\ncommit eef037ea02809afd4d21e204acf27d3bd1525829\nAuthor: Vivek Kasireddy \u003cvivek.kasireddy@intel.com\u003e\nDate: Wed Jul 3 16:03:53 2019 -0700\n\n drm/i915/ehl: Add support for DPLL4 (v10)\n \n This patch adds support for DPLL4 on EHL that include the\n following restrictions:\n \n - DPLL4 cannot be used with DDIA (combo port A internal eDP usage).\n DPLL4 can be used with other DDIs, including DDID\n (combo port A external usage).\n \n - DPLL4 cannot be enabled when DC5 or DC6 are enabled.\n \n - The DPLL4 enable, lock, power enabled, and power state are connected\n to the MGPLL1_ENABLE register.\n \n v2: (suggestions from Bob Paauwe)\n - Rework ehl_get_dpll() function to call intel_find_shared_dpll() and\n iterate twice: once for Combo plls and once for MG plls.\n \n - Use MG pll funcs for DPLL4 instead of creating new ones and modify\n mg_pll_enable to include the restrictions for EHL.\n \n v3: Fix compilation error\n \n v4: (suggestions from Lucas and Ville)\n - Treat DPLL4 as a combo phy PLL and not as MG PLL\n - Disable DC states when this DPLL is being enabled\n - Reuse icl_get_dpll instead of creating a separate one for EHL\n \n v5: (suggestion from Ville)\n - Refcount the DC OFF power domains during the enabling and disabling\n of this DPLL.\n \n v6: rebase\n \n v7: (suggestion from Imre)\n - Add a new power domain instead of iterating over the domains\n assoicated with DC OFF power well.\n \n v8: (Ville and Imre)\n - Rename POWER_DOMAIN_DPLL4 TO POWER_DOMAIN_DPLL_DC_OFF\n - Grab a reference in intel_modeset_setup_hw_state() if this\n DPLL was already enabled perhaps by BIOS.\n - Check for the port type instead of the encoder\n \n v9: (Ville)\n - Move the block of code that grabs a reference to the power domain\n POWER_DOMAIN_DPLL_DC_OFF to intel_modeset_readout_hw_state() to ensure\n that there is a reference present before this DPLL might get disabled.\n \n v10: rebase\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Cc: Imre Deak \u003cimre.deak@intel.com\u003e\n Signed-off-by: Vivek Kasireddy \u003cvivek.kasireddy@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703230353.24059-1-vivek.kasireddy@intel.com\n\ncommit 3e69db291bda338d92f554b55122c7319b612b10\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Jul 3 23:08:24 2019 +0300\n\n drm/i915: Clean up skl vs. icl plane formats\n \n Split the format lists for different planes on skl/icl more cleanly.\n \n On skl+ we have just two types of planes: those can do planar and\n those that can\u0027t.\n \n On icl we have three types of planes: hdr planes, sdr planes that\n can do planar, and sdr planes that can\u0027t do planar. Those latter two\n are the same set of planes we must when choose from when picking the\n UV vs. Y plane for planar scanout. So we shall just designate\n them sdr uv planes and sdr y planes.\n \n Reviewed-by: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703200824.5971-7-ville.syrjala@linux.intel.com\n\ncommit 94e35ce221732497f939713640816ae0a42a71c1\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Jul 3 23:08:23 2019 +0300\n\n drm/i915: Cosmetic fix for skl+ plane switch statement\n \n One of the switch cases has the byte order vs. format bits\n reversed to all the other cases. Appease the ocd and reorder\n them.\n \n Reviewed-by: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703200824.5971-6-ville.syrjala@linux.intel.com\n\ncommit d56e823ac32a793b24c857e8635def13d42b4008\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Jul 3 23:08:22 2019 +0300\n\n drm/i915: Deal with cpp\u003d\u003d8 for g4x watermarks\n \n Docs tell us that on g4x we have to compute the SR watermarks\n using 4 bytes per pixel. I\u0027m going to assume that only applies\n to 1 and 2 byte per pixel formats, and not 8 byte per pixel\n formats. That seems like a recipe for an insufficient watermark\n which could lead to underruns. Use the maximum of the two numbers\n instead.\n \n Reviewed-by: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703200824.5971-5-ville.syrjala@linux.intel.com\n\ncommit 94e15723df815497347408631203b2dabcb0bca3\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Jul 3 23:08:21 2019 +0300\n\n drm/i915: Program plane gamma ramps\n \n All sprite planes have a progammable gamma ramp. Set it up with\n a linear ramp on all platforms. This actually matches the reset\n value but soon we\u0027ll want to reprogram this ramp on some machines,\n so let\u0027s just set it up across the board.\n \n Note that on pre-IVB the hardware bypasses the gamma unit\n unless a YCbCr pixel format is used.\n \n v2: Add parens around \u003c\u003c in ilk_linear_gamma()\n Skip gamma programming for RGB on pre-IVB\n s/DVSGAMC/DVSGAMC_G4X/\n \n Reviewed-by: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703200824.5971-4-ville.syrjala@linux.intel.com\n\ncommit 423ee8e99aa5ee7ee5133616366099b8379870dc\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Jul 3 23:08:20 2019 +0300\n\n drm/i915: Disable sprite gamma on ivb-bdw\n \n We don\u0027t currently have any use for the sprite gamma on ivb-bdw.\n Let\u0027s disable it. We already do that on skl+.\n \n On pre-ivb there is no way to disable the sprite gamma, and it\n only affects YCbCr pixel formats, whereas on ivb+ it also\n affects RGB formats.\n \n Reviewed-by: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703200824.5971-3-ville.syrjala@linux.intel.com\n\ncommit 26443a4bc44850ff28329ae97a3cc9349d7dd87b\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Jul 3 23:08:19 2019 +0300\n\n drm/i915: Add windowing for primary planes on gen2/3 and chv\n \n Plane B and C (note that we don\u0027t actually expose plane C currently)\n on gen2/3 have a window generator, as does the primary plane on CHV\n pipe B. So let\u0027s allow positioning of these planes freely within the\n pipe source area.\n \n Plane A on gen2/3 seems to have some kind of partial window generator\n which would allow you to cut the plane off midway through the scanout,\n but it would still have to start at the top-left corner of the pipe,\n and it would have to be full width. That\u0027s doesn\u0027t sound all that\n useful, so for simplicity let\u0027s just keep to the idea that plane A\n has to be fullscreen.\n \n Gen4 removed the plane A/B windowing support entirely, and it wasn\u0027t\n reintroduced until SKL (apart from the CHV pipe B special case).\n \n v2: s/plane/i9xx_plane/ etc. (James)\n v3: Make it less confusing\n v4: Deal with IS_GEN()\n \n Reviewed-by: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703200824.5971-2-ville.syrjala@linux.intel.com\n\ncommit 58820574f1e937a1cf3eea629f1496e02560a132\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Thu Jul 4 13:17:56 2019 +0100\n\n drm/i915: Move dev_priv-\u003epm_i{m, e}r into intel_gt\n \n PM interrupts belong to the GT so move the variables to be inside\n struct intel_gt.\n \n Signed-off-by: Paulo Zanoni \u003cpaulo.r.zanoni@intel.com\u003e\n Co-developed-by: Paulo Zanoni \u003cpaulo.r.zanoni@intel.com\u003e\n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Acked-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190704121756.27824-3-tvrtko.ursulin@linux.intel.com\n\ncommit f0818984fa5d791583bb078086a8580231341b86\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Thu Jul 4 13:17:55 2019 +0100\n\n drm/i915: Remove some legacy mmio accessors from interrupt handling\n \n Mostly in gen11 interrupt handling and a couple neighbouring functions\n which were easy since uncore local was already available.\n \n Signed-off-by: Paulo Zanoni \u003cpaulo.r.zanoni@intel.com\u003e\n Co-developed-by: Paulo Zanoni \u003cpaulo.r.zanoni@intel.com\u003e\n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Acked-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190704121756.27824-2-tvrtko.ursulin@linux.intel.com\n\ncommit 9b77011e4122d420abbc486bad8ec53aa2a2873d\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Thu Jul 4 13:17:54 2019 +0100\n\n drm/i915: Rework some interrupt handling functions to take intel_gt\n \n Some interrupt handling functions already have gt in their names\n suggesting them as obvious candidates to make them take struct intel_gt\n instead of i915.\n \n Signed-off-by: Paulo Zanoni \u003cpaulo.r.zanoni@intel.com\u003e\n Co-developed-by: Paulo Zanoni \u003cpaulo.r.zanoni@intel.com\u003e\n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Acked-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190704121756.27824-1-tvrtko.ursulin@linux.intel.com\n\ncommit b8cade5959acd712b03b0cecca8299e9292c47a6\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Jul 4 21:04:55 2019 +0100\n\n drm/i915: Show instdone for each engine in debugfs\n \n Although polling each engine quickly is preferable as it should give us\n a sample of each engine at roughly the same time, keep it simple and\n just sample the engine as print out the debug state.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190704200455.14870-3-chris@chris-wilson.co.uk\n\ncommit 8f856c743c770b57308de450b21f70c0d16ab4cf\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Jul 4 22:23:43 2019 +0100\n\n drm/i915/selftests: Be engine agnostic\n \n When using MI operations, we do not care which engine we use, so use\n them all where possible, and where inconvenient double check we have the\n engine we selected at random.\n \n v2: Drop the local copy of engine-\u003esseu to avoid an unchecked deref\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190704212343.6820-1-chris@chris-wilson.co.uk\n\ncommit ec22f256a60ccf0dedd4ee305e616b6f17ed2bb7\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Jul 4 21:04:53 2019 +0100\n\n drm/i915/overlay: Stash the kernel context on initialisation\n \n Simplify runtime request creation by storing the context we need to use\n during initialisation. This allows us to remove one more hardcoded\n engine lookup.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190704200455.14870-1-chris@chris-wilson.co.uk\n\ncommit 1ee2ae896bb48c4b896c85c56214efb44320cd8b\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Wed Jul 3 15:41:16 2019 +0100\n\n drm/i915/hangcheck: Look at instdone for all engines\n \n It seems intel_engine_get_instdone is able to get instdone for all engines\n but intel_hangcheck.c/subunits_stuck decides to ignore it for non render.\n \n We can just drop the check in subunits_stuck since the checks on\n unavailable fields will always return stuck, which when bitwise and with\n the potential unstuck instdone is harmless.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703144116.15593-1-tvrtko.ursulin@linux.intel.com\n\ncommit 6582f4f613e35225c663fc1194b1a7d1f0af2a2f\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Jul 4 17:53:17 2019 +0100\n\n drm/i915/selftests: Drain the freedlists between exec passes\n \n During the context execution tests, we issue a lot of work and discard a\n lot of objects without releasing the lock and allowing the background\n reaper to free those objects. Insert a small break between each pass to\n flush the worker.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190704165317.21060-1-chris@chris-wilson.co.uk\n\ncommit e7539b79f703a6b533385088fc15cb5c9ab3f56f\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Jul 4 21:16:56 2019 +0100\n\n drm/i915/gtt: Mark the freed page table entries with scratch\n \n On unwinding the allocation error path and having freed the page table\n entry, it is imperative that we mark it as scratch.\n \n \u003c4\u003e [416.075569] general protection fault: 0000 [#1] PREEMPT SMP PTI\n \u003c4\u003e [416.075801] CPU: 0 PID: 2385 Comm: kworker/u2:11 Tainted: G U 5.2.0-rc7-CI-Patchwork_13534+ #1\n \u003c4\u003e [416.076162] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.10.1-0-g8891697-prebuilt.qemu-project.org 04/01/2014\n \u003c4\u003e [416.076522] Workqueue: i915 __i915_vm_release [i915]\n \u003c4\u003e [416.076754] RIP: 0010:gen8_ppgtt_cleanup_3lvl+0x58/0xb0 [i915]\n \u003c4\u003e [416.077023] Code: 81 e2 04 fe ff ff 81 c2 ff 01 00 00 4c 8d 74 d6 58 4d 8b 65 00 4d 3b a7 28 02 00 00 74 40 49 8d 5c 24 50 49 81 c4 50 10 00 00 \u003c48\u003e 8b 2b 49 3b af 20 02 00 00 74 13 4c 89 ff 48 89 ee e8 01 fb ff\n \u003c4\u003e [416.077445] RSP: 0018:ffffc9000046bd98 EFLAGS: 00010206\n \u003c4\u003e [416.077625] RAX: 0001000000000000 RBX: 6b6b6b6b6b6b6bbb RCX: 8b4b56d500000000\n \u003c4\u003e [416.077838] RDX: 00000000000001ff RSI: ffff88805a578008 RDI: ffff88805bd0efc8\n \u003c4\u003e [416.078167] RBP: ffff88805bd0efc8 R08: 0000000004e42b93 R09: 0000000000000001\n \u003c4\u003e [416.078381] R10: 0000000000000000 R11: ffff888077a1b0b8 R12: 6b6b6b6b6b6b7bbb\n \u003c4\u003e [416.078594] R13: ffff88805a578058 R14: ffff88805a579058 R15: ffff88805bd0efc8\n \u003c4\u003e [416.078815] FS: 0000000000000000(0000) GS:ffff88807da00000(0000) knlGS:0000000000000000\n \u003c4\u003e [416.079395] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n \u003c4\u003e [416.079851] CR2: 000056160fec2b14 CR3: 0000000071bbc003 CR4: 00000000003606f0\n \u003c4\u003e [416.080388] Call Trace:\n \u003c4\u003e [416.080828] gen8_ppgtt_cleanup+0x64/0x100 [i915]\n \u003c4\u003e [416.081399] __i915_vm_release+0xfc/0x1d0 [i915]\n \n Fixes: 1d1b5490b91c (\"drm/i915/gtt: Replace struct_mutex serialisation for allocation\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190704201656.15775-1-chris@chris-wilson.co.uk\n\ncommit ab9e2f77768211a8ef5263e0c3a939c9b2de01a7\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jul 3 14:58:05 2019 +0100\n\n drm/i915/gt: Pull engine w/a initialisation into common\n \n We need to setup the workarounds on all engines, with the knowledge\n about which platforms each workaround applies to kept together in the\n workaround list. As such, we can pull the w/a initialisation into the\n common setup and try to avoid duplicating knowledge about when to setup\n the workarounds.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703135805.7310-2-chris@chris-wilson.co.uk\n\ncommit 4a54da35102925401b1d670de468127914c67034\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jul 3 14:58:04 2019 +0100\n\n drm/i915: Dump w/a lists on all engines\n \n We store separate wa_list on every engine, so be sure to include all\n when dumping the current set via debugfs.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703135805.7310-1-chris@chris-wilson.co.uk\n\ncommit 2a46fbb25ae8ef98f44003bbcc0b4480b3b9281e\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Wed Jul 3 11:36:39 2019 +0000\n\n drm/i915/guc: Upgrade to GuC 33.0.0\n \n New GuC firmware is available. Let\u0027s use it.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703113640.31100-1-michal.wajdeczko@intel.com\n\ncommit 163fa23435cc9c705a71001d4aa15f3f945554a1\nAuthor: Kefeng Wang \u003cwangkefeng.wang@huawei.com\u003e\nDate: Wed Jul 3 16:25:52 2019 +0800\n\n percpu: Make pcpu_setup_first_chunk() void function\n \n pcpu_setup_first_chunk() will panic or BUG_ON if the are some\n error and doesn\u0027t return any error, hence it can be defined to\n return void.\n \n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Kefeng Wang \u003cwangkefeng.wang@huawei.com\u003e\n Signed-off-by: Dennis Zhou \u003cdennis@kernel.org\u003e\n [Dennis: fixed kbuild warning for pcpu_page_first_chunk()]\n\ncommit ae1c5fd72dfcb66bda5c9002aa341196c7c3e3e1\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Jul 4 11:43:45 2019 +0100\n\n drm/i915/gtt: Handle double alloc failures\n \n Matthew pointed out that we could face a double failure with concurrent\n allocations/frees, and so the assumption that the local var alloc was\n NULL was fraught with danger. Rather than complicate the error paths too\n much to add a second local for a second free, just do the second free\n earlier on the unwind path.\n \n Reported-by: Matthew Auld \u003cmatthew.william.auld@gmail.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.william.auld@gmail.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.william.auld@gmail.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190704104345.6603-1-chris@chris-wilson.co.uk\n\ncommit bf73fc0fa9cf78e37d6ee99e8d12bfa2083594d6\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jul 3 15:37:02 2019 +0100\n\n drm/i915: Show support for accurate sw PMU busyness tracking\n \n Expose whether or not we support the PMU software tracking in our\n scheduler capabilities, so userspace can query at runtime.\n \n v2: Use I915_SCHEDULER_CAP_ENGINE_BUSY_STATS for a less ambiguous\n capability name.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703143702.11339-1-chris@chris-wilson.co.uk\n\ncommit 0c159ffef628fa94d0f4f9128e7f2b6f2b5e86ef\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jul 3 19:06:01 2019 +0100\n\n drm/i915/gem: Defer obj-\u003ebase.resv fini until RCU callback\n \n Since reservation_object_fini() does an immediate free, rather than\n kfree_rcu as normal, we have to delay the release until after the RCU\n grace period has elapsed (i.e. from the rcu cleanup callback) so that we\n can rely on the RCU protected access to the fences while the object is a\n zombie.\n \n i915_gem_busy_ioctl relies on having an RCU barrier to protect the\n reservation in order to avoid having to take a reference and strong\n memory barriers.\n \n v2: Order is important; only release after putting the pages!\n \n Fixes: c03467ba40f7 (\"drm/i915/gem: Free pages before rcu-freeing the object\")\n Testcase: igt/gem_busy/close-race\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703180601.10950-1-chris@chris-wilson.co.uk\n\ncommit 21de5a9e34fb0f3efa31df5712cedc0ae97f7b4d\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jul 3 16:52:25 2019 +0100\n\n drm/i915/gt: Ignore forcewake acquisition for posting_reads\n \n We don\u0027t care about the result of the read, so it may be garbage, we\n only care that the mmio is flushed. As such, we can forgo using an\n individual forcewake and lock around any posting-read for an engine.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703155225.9501-4-chris@chris-wilson.co.uk\n\ncommit 313443b16ac26dd0250462d8513c5b5c0a188e38\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jul 3 16:52:24 2019 +0100\n\n drm/i915/gt: Assume we hold forcewake for execlists resume\n \n We can assume the caller is holding a blanket forcewake for the\n register writes during resume, and so we can skip taking individual\n locks around each write inside execlists resume.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703155225.9501-3-chris@chris-wilson.co.uk\n\ncommit 56e0f78e47e47795873acbabab443029aa30bcb4\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jul 3 16:52:23 2019 +0100\n\n drm/i915/gt: Use caller provided forcewake for intel_mocs_init_engine\n \n During post-reset resume, we call intel_mocs_init_engine to reinitialise\n the MOCS registers. Suprisingly, especially when enhanced by lockdep,\n the acquisition of the forcewake lock around each register write takes a\n substantial portion of the reset time. We don\u0027t need to use the\n individual forcewake here as we can assume that the caller is holding a\n blanket forcewake for the reset\u0026resume and the resume is serialised.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703155225.9501-2-chris@chris-wilson.co.uk\n\ncommit b7dc9395c5db2748aac24704d3f2d96f34f90f07\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Jul 4 11:20:48 2019 +0100\n\n drm/i915: Check caller held wakerefs in assert_forcewakes_active\n \n The intent of the assert is to document that the caller took the\n appropriate wakerefs for the function. However, as Tvrtko pointed out,\n we simply check whether the fw_domains are active and may be confused by\n the auto wakeref which may be dropped between the check and use. Let\u0027s\n be more careful in the assert and check that each fw_domain has an\n explicit caller wakeref above and beyond the automatic wakeref.\n \n v2: Fix spelling for config DRM_I915_DEBUG_RUNTIME_PM\n v3: Timer may still be active after we drop the autowakeref, we need to\n check domain-\u003eactive instead.\n v4: The timer checks domain-\u003eactive, but we still need to check the\n timer. (This is starting to look weird...)\n \n Reported-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190704102048.6436-1-chris@chris-wilson.co.uk\n\ncommit 67b587934c23c6e286486a0a724d486c8e037fd6\nAuthor: Rob Clark \u003crobdclark@chromium.org\u003e\nDate: Tue Jul 2 08:44:18 2019 -0700\n\n drm/bridge: ti-sn65dsi86: correct dsi mode_flags\n \n Noticed while comparing register dump of how bootloader configures DSI\n vs how kernel configures. It seems the bridge still works either way,\n but fixing this clears the \u0027CHA_DATATYPE_ERR\u0027 error status bit.\n \n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n Reviewed-by: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\n Reviewed-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190702154419.20812-4-robdclark@gmail.com\n\ncommit 8efb243a5ce89090df1284c91519ccc3b75758da\nAuthor: Rob Clark \u003crobdclark@chromium.org\u003e\nDate: Tue Jul 2 08:44:17 2019 -0700\n\n drm/bridge: ti-sn65dsi86: add debugfs\n \n Add a debugfs file to show status registers.\n \n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n Reviewed-by: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\n Reviewed-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190702154419.20812-3-robdclark@gmail.com\n\ncommit 5c27d6078cd69a1ecaa84b83680bdc8d25bbd172\nAuthor: Rob Clark \u003crobdclark@chromium.org\u003e\nDate: Tue Jul 2 08:44:16 2019 -0700\n\n drm/bridge: ti-sn65dsi86: add link to datasheet\n \n The bridge has pretty good docs, lets add a link to make them easier to\n find.\n \n Signed-off-by: Rob Clark \u003crobdclark@chromium.org\u003e\n Reviewed-by: Jeffrey Hugo \u003cjeffrey.l.hugo@gmail.com\u003e\n Reviewed-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190702154419.20812-2-robdclark@gmail.com\n\ncommit 4fda44bf16b79a0b78fe36c6b9859e9ce2d09f43\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jul 3 18:19:13 2019 +0100\n\n drm/i915: Flush the workqueue before draining\n \n Trying to drain a workqueue while we may still be adding to it from\n background tasks is, according to kernel/workqueue.c, verboten. So, add\n a flush_workqueue() at the start of our cleanup procedure.\n \n References: https://bugs.freedesktop.org/show_bug.cgi?id\u003d110550\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703171913.16585-4-chris@chris-wilson.co.uk\n\ncommit 2006058e9988421a113e8edc004a8e0eae1a6d3f\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Jul 4 10:19:25 2019 +0100\n\n drm/i915: Move the renderstate setup under gt/\n \n The render state is used to initialise the default RCS context, and only\n used during early setup from within the gt code. As such, it makes a\n good candidate for placing within gt/, even if it is not yet entirely\n clean of our GEM heritage.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190704091925.7391-1-chris@chris-wilson.co.uk\n\ncommit 068610895ebd4bd86f496f01eb7b97e56d7269b2\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jul 3 18:19:12 2019 +0100\n\n drm/i915/gtt: Defer the free for alloc error paths\n \n If we hit an error while allocating the page tables, we have to unwind\n the incomplete updates, and wish to free the unused pd. However, we are\n not allowed to be hoding the spinlock at that point, and so must use the\n later free to defer it until after we drop the lock.\n \n \u003c3\u003e [414.363795] BUG: sleeping function called from invalid context at drivers/gpu/drm/i915/i915_gem_gtt.c:472\n \u003c3\u003e [414.364167] in_atomic(): 1, irqs_disabled(): 0, pid: 3905, name: i915_selftest\n \u003c4\u003e [414.364406] 3 locks held by i915_selftest/3905:\n \u003c4\u003e [414.364408] #0: 0000000034fe8aa8 (\u0026dev-\u003emutex){....}, at: device_driver_attach+0x18/0x50\n \u003c4\u003e [414.364415] #1: 000000006bd8a560 (\u0026dev-\u003estruct_mutex){+.+.}, at: igt_ctx_exec+0xb7/0x410 [i915]\n \u003c4\u003e [414.364476] #2: 000000003dfdc766 (\u0026(\u0026pd-\u003elock)-\u003erlock){+.+.}, at: gen8_ppgtt_alloc_pdp+0x448/0x540 [i915]\n \u003c3\u003e [414.364529] Preemption disabled at:\n \u003c4\u003e [414.364530] [\u003c0000000000000000\u003e] 0x0\n \u003c4\u003e [414.364696] CPU: 0 PID: 3905 Comm: i915_selftest Tainted: G U 5.2.0-rc7-CI-CI_DRM_6403+ #1\n \u003c4\u003e [414.364698] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.10.1-0-g8891697-prebuilt.qemu-project.org 04/01/2014\n \u003c4\u003e [414.364699] Call Trace:\n \u003c4\u003e [414.364704] dump_stack+0x67/0x9b\n \u003c4\u003e [414.364708] ___might_sleep+0x167/0x250\n \u003c4\u003e [414.364777] vm_free_page+0x24/0xc0 [i915]\n \u003c4\u003e [414.364852] free_pd+0xf/0x20 [i915]\n \u003c4\u003e [414.364897] gen8_ppgtt_alloc_pdp+0x489/0x540 [i915]\n \u003c4\u003e [414.364946] gen8_ppgtt_alloc_4lvl+0x8e/0x2e0 [i915]\n \u003c4\u003e [414.364992] ppgtt_bind_vma+0x2e/0x60 [i915]\n \u003c4\u003e [414.365039] i915_vma_bind+0xe8/0x2c0 [i915]\n \u003c4\u003e [414.365088] __i915_vma_do_pin+0xa1/0xd20 [i915]\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111050\n Fixes: 1d1b5490b91c (\"drm/i915/gtt: Replace struct_mutex serialisation for allocation\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703171913.16585-3-chris@chris-wilson.co.uk\n\ncommit f1513d20b9f17c4a9fbc641289e880eb2f879632\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Fri Jun 14 23:36:23 2019 +0800\n\n drm/bridge: sii902x: Make sii902x_audio_digital_mute static\n \n Fix sparse warning:\n \n drivers/gpu/drm/bridge/sii902x.c:665:5: warning:\n symbol \u0027sii902x_audio_digital_mute\u0027 was not declared. Should it be static?\n \n Reported-by: Hulk Robot \u003chulkci@huawei.com\u003e\n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614153623.28708-1-yuehaibing@huawei.com\n\ncommit 415d2e9e07574d3de63b8df77dc686e0ebf64865\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Wed Jul 3 16:38:50 2019 -0600\n\n Revert \"drm/gem: Rename drm_gem_dumb_map_offset() to drm_gem_map_offset()\"\n \n This reverts commit 220df83a5394fbf7c1486ba7848794b7b351d598.\n \n Turns out drm_gem_dumb_map_offset really only worked for the dumb buffer\n case, so revert the name change.\n \n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit be855382bacb5ccfd24f9be6098d87acf4cfbb15\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Wed Jul 3 16:37:12 2019 -0600\n\n Revert \"drm/panfrost: Use drm_gem_map_offset()\"\n \n This reverts commit 583bbf46133c726bae277e8f4e32bfba2a528c7f.\n \n Turns out we need mmap to work on imported BOs even if the current code\n is buggy.\n \n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n\ncommit 2c291417c1646a8651dd9c8db9ad1ce4ff2150f7\nAuthor: Aditya Swarup \u003caditya.swarup@intel.com\u003e\nDate: Thu Jun 27 15:07:08 2019 -0700\n\n drm/i915: Add N \u0026 CTS values for 10/12 bit deep color\n \n Adding N \u0026 CTS values for 10/12 bit deep color from Appendix C\n table in HDMI 2.0 spec. The correct values for N is not chosen\n automatically by hardware for deep color modes.\n \n v2: Remove unnecessary initialization of size\n \n Signed-off-by: Aditya Swarup \u003caditya.swarup@intel.com\u003e\n Cc: Clint Taylor \u003cClinton.A.Taylor@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190627220708.31700-2-aditya.swarup@intel.com\n\ncommit 4a49c2bf9ee45e7475c4c2dec74a3c034345aa5b\nAuthor: Aditya Swarup \u003caditya.swarup@intel.com\u003e\nDate: Thu Jun 27 15:07:07 2019 -0700\n\n drm/i915: Use port clock to set correct N value\n \n Use port_clock to check the clock values in n/cts lookup table instead\n of crtc_clock. As port_clock is already adjusted based on color mode set\n (8 bit or deep color), this will help in checking clock values for deep\n color modes from n/cts lookup table.\n \n Signed-off-by: Aditya Swarup \u003caditya.swarup@intel.com\u003e\n Cc: Clint Taylor \u003cClinton.A.Taylor@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190627220708.31700-1-aditya.swarup@intel.com\n\ncommit 56e9371bc3f3e7d6c1a197a45d550b2ce6af25f6\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Thu Jun 6 15:42:10 2019 +0300\n\n drm/i915: Deal with machines that expose less than three QGV points\n \n When SAGV is forced to disabled/min/med/max in the BIOS pcode will\n only hand us a single QGV point instead of the normal three. Fix\n the code to deal with that instead declaring the bandwidth limit\n to be 0 MB/s (and thus preventing any planes from being enabled).\n \n Also shrink the max_bw sturct a bit while at it, and change the\n deratedbw type to unsigned since the code returns the bw as\n an unsigned int.\n \n Since we now keep track of how many qgv points we got from pcode\n we can drop the earlier check added for the \"pcode doesn\u0027t\n support the memory subsystem query\" case.\n \n Cc: felix.j.degrood@intel.com\n Cc: Mark Janes \u003cmark.a.janes@intel.com\u003e\n Cc: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Cc: Clint Taylor \u003cClinton.A.Taylor@intel.com\u003e\n Fixes: c457d9cf256e (\"drm/i915: Make sure we have enough memory bandwidth on ICL\")\n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d110838\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190606124210.3482-1-ville.syrjala@linux.intel.com\n Reviewed-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n\ncommit 09480072e3ee471fe2735d46f454483f4c26853b\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jul 3 10:17:19 2019 +0100\n\n drm/i915: Mark up vma-\u003eactive as safe for use inside shrinkers\n \n Since a shrinker may be forced to wait on GPU activity,\n i915_active_wait(\u0026vma-\u003eactive) must be safe for use inside a shrinker,\n and so let\u0027s mark up the lock as being acquired by the shrinker to avoid\n any nasty surprises creeping in.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703091726.11690-8-chris@chris-wilson.co.uk\n\ncommit d650d1f5cfd7ca607dbbb2023b76c2afa873246a\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jul 3 10:17:18 2019 +0100\n\n drm/i915: Markup potential lock for i915_active\n \n Make the lockchains more deterministic via i915_active by flagging the\n potential lock.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703091726.11690-7-chris@chris-wilson.co.uk\n\ncommit c03467ba40f783ebe756114bb68e13a6b404c03a\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jul 3 10:17:17 2019 +0100\n\n drm/i915/gem: Free pages before rcu-freeing the object\n \n As we have dropped the final reference to the object, we do not need to\n wait until after the rcu grace period to drop its pages. We still require\n struct_mutex to completely unbind the object to release the pages, so we\n still need a free-worker to manage that from process context. By\n scheduling the release of pages before waiting for the rcu should mean\n that we are not trapping those pages from beyond the reach of the\n shrinker.\n \n v2: Pass along the request to skip if the vma is busy to the underlying\n unbind routine, to avoid checking the reservation underneath the\n i915-\u003emm.obj_lock which may be used from inside irq context.\n \n v3: Flip the bit for unbinding while active, for later convenience.\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111035\n Fixes: a93615f900bd (\"drm/i915: Throw away the active object retirement complexity\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703091726.11690-6-chris@chris-wilson.co.uk\n\ncommit ad9e3792b0c0a63d4416d6d7da8320d11110b61e\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jul 3 10:17:20 2019 +0100\n\n drm/i915/execlists: Hesitate before slicing\n \n Be a little more hesitant before injecting a timeslice, and try to take\n into account any change in priority that is due for the running task\n before switching to another task. This will allow us to arbitrarily\n prevent switching away from a request if we deem it necessarily to\n disable preemption, for instance.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Cc: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Acked-by: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703091726.11690-9-chris@chris-wilson.co.uk\n\ncommit 8221d21b0664d79a4319bad526f4f727d3b15981\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jul 3 10:17:13 2019 +0100\n\n drm/i915/selftests: Lock the drm_mm while modifying\n \n Remember to lock the drm_mm as we modify it, lest it be modified in the\n background by retire/free workers!\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703091726.11690-2-chris@chris-wilson.co.uk\n\ncommit 63251685c141ef8db611cd36609a8369d97b47da\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jul 3 10:17:12 2019 +0100\n\n drm/i915/selftests: Common live setup/teardown\n \n We frequently, but not frequently enough!, remember to flush residual\n operations and objects at the end of a live subtest. The purpose is to\n cleanup after every subtest, leaving a clean slate for the next subtest,\n and perform early detection of leaky state. As this should ideally be\n common for all live subtests, pull the task into a common teardown\n routine.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190703091726.11690-1-chris@chris-wilson.co.uk\n\ncommit 8e9ecb3e1e3366559cd4ed5ab698c391f8236a37\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jul 2 15:09:50 2019 +0100\n\n drm/i915/display: Handle lost primary_port across suspend\n \n icl-dsi is dying on suspend/resume at\n \n RIP: 0010:icl_update_active_dpll+0x2c/0xa0 [i915]\n \n which appears due to the loss of the time primary_port across suspend.\n Protect against the potential NULL dereference by assuming\n ICL_PORT_DPLL_DEFAULT unless the port is actively specified otherwise.\n \n Fixes: 24a7bfe0c2d7 (\"drm/i915: Keep the TypeC port mode fixed when the port is active\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190702140950.7069-1-chris@chris-wilson.co.uk\n\ncommit 315ca4c408ed3bcfe3a204fec9f5c397a98e97a0\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Tue Jul 2 18:17:23 2019 +0300\n\n drm/i915: synchronize_irq() against the actual irq\n \n When eliminating our use of drm_irq_install() I failed to convert\n all our synchronize_irq() calls to consult pdev-\u003eirq instead of\n dev_priv-\u003edrm.irq. As we no longer populate dev_priv-\u003edrm.irq\n we\u0027re no longer synchronizing against anything.\n \n v2: Add intel_syncrhonize_irq() (Chris)\n \n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reported-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Fixes: b318b82455bd (\"drm/i915: Nuke drm_driver irq vfuncs\")\n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d111012\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190702151723.29739-1-ville.syrjala@linux.intel.com\n\ncommit 0ccf52badd40efd3d5c8df91845df4d65e7caab9\nAuthor: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\nDate: Tue Jul 2 13:50:12 2019 +0200\n\n drm/vram: Don\u0027t export driver callback functions for PRIME\n \n PRIME functionality is now provided by GEM object functions. The driver\n callback functions are obsolete. So this patch renames them and turns\n them into static internal functions of the VRAM helper library. The\n implementation of gem_prime_mmap is now unused and the patch removes it.\n \n v3:\n * kept each renamed function at its original location within file\n * kept documentation\n \n Signed-off-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190702115012.4418-6-tzimmermann@suse.de\n\ncommit bf2791eb5cf699a3eec0ca5bda2b17cec6a74bdb\nAuthor: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\nDate: Tue Jul 2 13:50:11 2019 +0200\n\n drm/vbox: Remove empty PRIME functions\n \n The vbox driver uses VRAM helpers for memory management. These helpers\n provide a basic implementation of PRIME functions, so the vbox driver\u0027s\n empty implmentation can be removed. As a side effect of this change,\n vbox can now use generic framebuffer emulation.\n \n Signed-off-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190702115012.4418-5-tzimmermann@suse.de\n\ncommit 80be7eed1d32fb1173837ccc8e163ac3db9f82fb\nAuthor: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\nDate: Tue Jul 2 13:50:10 2019 +0200\n\n drm/hibmc: Update struct drm_driver for GEM object functions\n \n The GEM object\u0027s free function is now called through struct\n drm_gem_object_funcs.free. The function struct drm_driver.gem_prime_mmap\n is now required for mmap\u0027ing GEM objects to userspace.\n \n v2:\n * set drm_driver.gem_prime_mmap to drm_gem_prime_mmap()\n \n Signed-off-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Acked-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190702115012.4418-4-tzimmermann@suse.de\n\ncommit f5f6f36a49c77847b6ad20234815aba5ed9b6c50\nAuthor: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\nDate: Tue Jul 2 13:50:09 2019 +0200\n\n drm/bochs: Remove PRIME helpers from driver structure\n \n VRAM PRIME helpers are now called through GEM object functions. The\n driver callback functions are obsolete.\n \n Signed-off-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Acked-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190702115012.4418-3-tzimmermann@suse.de\n\ncommit 31070a871fdcb16dd209e6bc0e6ca16be7cfb938\nAuthor: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\nDate: Tue Jul 2 13:50:08 2019 +0200\n\n drm/vram: Set GEM object functions for PRIME\n \n PRIME functionality is now provided via the callback functions in\n struct drm_gem_object_funcs. The driver-structure functions are obsolete.\n As a side effect of this patch, VRAM-based drivers get basic PRIME\n support automatically without having to set any flags or additional\n fields.\n \n v2:\n - use existing PRIME functions for object\u0027s table\n v3:\n - move object table to EOF so it can refer to internal interfaces\n \n Signed-off-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190702115012.4418-2-tzimmermann@suse.de\n\ncommit 096a93949b8658b058e8e3d3aee7b6ff36bf4f2e\nAuthor: Michał Winiarski \u003cmichal.winiarski@intel.com\u003e\nDate: Tue Jul 2 13:31:49 2019 +0200\n\n drm/i915/gtt: Don\u0027t check PPGTT presence on PPGTT-only platforms\n \n We missed one place where we check PPGTT-only platform for PPGTT\n presence. Let\u0027s remove it.\n While I\u0027m here let\u0027s assert that this particular code is never called on\n pre-gen8 platforms.\n \n References: 4bdafb9ddfa4 (\"drm/i915: Remove i915.enable_ppgtt override\")\n Signed-off-by: Michał Winiarski \u003cmichal.winiarski@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190702113149.21200-2-michal.winiarski@intel.com\n\ncommit a3389c14ba09898877122cfae8d2c4c9d9e7291e\nAuthor: Michał Winiarski \u003cmichal.winiarski@intel.com\u003e\nDate: Tue Jul 2 13:31:48 2019 +0200\n\n Revert \"drm/i915: Introduce private PAT management\"\n \n This reverts commit 4395890a48551982549d222d1923e2833dac47cf.\n \n It\u0027s been over a year since this was merged, and the actual users of\n intel_ppat_get / intel_ppat_put never materialized.\n \n Time to remove it!\n \n v2: Unbreak suspend (Chris)\n v3: Rebase, drop fixes tag to avoid confusion\n \n Signed-off-by: Michał Winiarski \u003cmichal.winiarski@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Zhi Wang \u003czhi.a.wang@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190702113149.21200-1-michal.winiarski@intel.com\n\ncommit c5f846eed2a1856b78e988eeef08215c70598ecd\nAuthor: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\nDate: Mon Jul 1 13:44:42 2019 +0300\n\n drm/i915: Fix memleak in runtime wakeref tracking\n \n If we untrack wakerefs, the actual count may reach zero.\n However the krealloced owners array is still there and\n needs to be taken care of. Free the owners unconditionally\n to fix the leak.\n \n Fixes: bd780f37a361 (\"drm/i915: Track all held rpm wakerefs\")\n Reported-by: Juha-Pekka Heikkila \u003cjuhapekka.heikkila@gmail.com\u003e\n Cc: Juha-Pekka Heikkila \u003cjuhapekka.heikkila@gmail.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190701104442.9319-1-mika.kuoppala@linux.intel.com\n\ncommit 3fe0107e45ab396342497e06b8924cdd485cde3b\nAuthor: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\nDate: Fri Jun 28 15:07:20 2019 +0300\n\n drm/i915/icl: whitelist PS_(DEPTH|INVOCATION)_COUNT\n \n The same tests failing on CFL+ platforms are also failing on ICL.\n Documentation doesn\u0027t list the\n WaAllowPMDepthAndInvocationCountAccessFromUMD workaround for ICL but\n applying it fixes the same tests as CFL.\n \n v2: Use only one whitelist entry (Lionel)\n \n Signed-off-by: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Tested-by: Anuj Phogat \u003canuj.phogat@gmail.com\u003e\n Cc: stable@vger.kernel.org\n Acked-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628120720.21682-4-lionel.g.landwerlin@intel.com\n\ncommit 2c903da50f5a9522b134e488bd0f92646c46f3c0\nAuthor: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\nDate: Fri Jun 28 15:07:19 2019 +0300\n\n drm/i915: whitelist PS_(DEPTH|INVOCATION)_COUNT\n \n CFL:C0+ changed the status of those registers which are now\n blacklisted by default.\n \n This is breaking a number of CTS tests on GL \u0026 Vulkan :\n \n KHR-GL45.pipeline_statistics_query_tests_ARB.functional_fragment_shader_invocations (GL)\n \n dEQP-VK.query_pool.statistics_query.fragment_shader_invocations.* (Vulkan)\n \n v2: Only use one whitelist entry (Lionel)\n \n Bspec: 14091\n Signed-off-by: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Cc: stable@vger.kernel.org\n Acked-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628120720.21682-3-lionel.g.landwerlin@intel.com\n\ncommit 361b69051326ed0e07553315227678d00d651a9e\nAuthor: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\nDate: Sat Jun 29 14:13:50 2019 +0100\n\n drm/i915: fix whitelist selftests with readonly registers\n \n When a register is readonly there is not much we can tell about its\n value (apart from its default value?). This can be covered by tests\n exercising the value of the register from userspace.\n \n For PS_INVOCATION_COUNT we\u0027ve got the following piglit tests :\n \n KHR-GL45.pipeline_statistics_query_tests_ARB.functional_fragment_shader_invocations\n \n Vulkan CTS tests :\n \n dEQP-VK.query_pool.statistics_query.fragment_shader_invocations.*\n \n v2: Use a local to shrink under 80cols.\n \n Signed-off-by: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Fixes: 86554f48e511 (\"drm/i915/selftests: Verify whitelist of context registers\")\n Tested-by: Anuj Phogat \u003canuj.phogat@gmail.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190629131350.31185-1-chris@chris-wilson.co.uk\n\ncommit afd1bcd4ad776a9892ce6e814111af22dae44956\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jul 2 10:21:17 2019 +0100\n\n drm/i915: Report if i915_active is still busy upon waiting\n \n If we try to wait on an i915_active from within a critical section, it\n will remain busy (such as if we are shrinking from within\n i915_active_ref). Report the failure so that we do not proceed thinking\n it is idle.\n \n Extracted from a future patch \"drm/i915: Coordinate i915_active with its\n own mutex\".\n \n Fixes: 12c255b5dad1 (\"drm/i915: Provide an i915_active.acquire callback\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190702092117.1707-1-chris@chris-wilson.co.uk\n\ncommit 8721215c7ef928293516a9396648592fc7284592\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Thu Jun 27 20:21:14 2019 +0200\n\n DRM: ingenic: Add support for panels with 8-bit serial bus\n \n Add support for the LCD panels with a serial 8-bit bus, where the color\n components of each 24-bit pixel are sent sequentially.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190627182114.27299-3-paul@crapouillou.net\n # *** extracted tags ***\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n\ncommit dba8e987687f141de4e513715e1f1bd6315c2f55\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Thu Jun 27 20:21:13 2019 +0200\n\n DRM: ingenic: Add support for Sharp panels\n \n Add support for the LCD panels that must be driven with the\n Sharp-specific signals SPL, CLS, REV, PS.\n \n An example of such panel is the LS020B1DD01D supported by the\n panel-simple DRM panel driver.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190627182114.27299-2-paul@crapouillou.net\n # *** extracted tags ***\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n\ncommit ffa8aa00e9e589463582b432e40b4522a2c84265\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Thu Jun 27 20:21:12 2019 +0200\n\n DRM: ingenic: Use devm_platform_ioremap_resource\n \n Simplify a bit the probe function by using the newly introduced\n devm_platform_ioremap_resource(), instead of having to call\n platform_get_resource() followed by devm_ioremap_resource().\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190627182114.27299-1-paul@crapouillou.net\n # *** extracted tags ***\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n\ncommit 91abab83839aa2eba073e4a63c729832fdb27ea1\nAuthor: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\nDate: Mon Jul 1 17:03:29 2019 -0400\n\n XArray: Fix xas_next() with a single entry at 0\n \n If there is only a single entry at 0, the first time we call xas_next(),\n we return the entry. Unfortunately, all subsequent times we call\n xas_next(), we also return the entry at 0 instead of noticing that the\n xa_index is now greater than zero. This broke find_get_pages_contig().\n \n Fixes: 64d3e9a9e0cc (\"xarray: Step through an XArray\")\n Reported-by: Kent Overstreet \u003ckent.overstreet@gmail.com\u003e\n Signed-off-by: Matthew Wilcox (Oracle) \u003cwilly@infradead.org\u003e\n\ncommit 8759aa4cc1d36749845b7c99b12ab180e12f2699\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Jul 1 11:04:54 2019 +0100\n\n drm/i915/execlists: Refactor CSB state machine\n \n Daniele pointed out that the CSB status information will change with\n Tigerlake and suggested that we could rearrange our state machine to\n hide the differences in generation. gcc also prefers the explicit state\n machine, so make it so:\n \n process_csb 1980 1967 -13\n \n Suggested-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@linux.intel.com\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190701100502.15639-4-chris@chris-wilson.co.uk\n\ncommit f80c9a9a04d3fa8231d2477f6a2c8b66a84ab41b\nAuthor: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\nDate: Tue Jun 25 17:03:50 2019 -0700\n\n drm/i915/ehl: Don\u0027t program PHY_MISC on EHL PHY C\n \n Although EHL added a third combo PHY, no PHY_MISC register was added for\n PHY C. The bspec indicates that there\u0027s no need to program the \"DE to\n IO Comp Pwr Down\" setting for this PHY that we usually need to set in\n PHY_MISC.\n \n v2:\n - Add IS_ELKHARTLAKE() guards since future platforms that have a PHY C\n are likely to reinstate the PHY_MISC register. (Jose)\n - Use goto\u0027s to skip PHY_MISC programming \u0026 minimize code deltas. (Jose)\n \n Bspec: 33148\n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626000352.31926-4-matthew.d.roper@intel.com\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n\ncommit 0e933162b0420b5258c34da719ac65269e4dbc49\nAuthor: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\nDate: Tue Jun 25 17:03:49 2019 -0700\n\n drm/i915/ehl: Add third combo PHY offset\n \n v2: Rename register to _EHL_COMBOPHY_C. (Jose)\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626000352.31926-3-matthew.d.roper@intel.com\n\ncommit 4a8134d57c6ca7b67acb0812f1f32ef60a2f19c0\nAuthor: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\nDate: Tue Jun 25 17:03:48 2019 -0700\n\n drm/i915/icl: Drop port parameter to icl_get_combo_buf_trans()\n \n The port parameter hasn\u0027t been used since the last bspec phy programming\n update. Drop it to make some upcoming changes simpler.\n \n References: 9659c1af451a (\"drm/i915/icl: combo port vswing programming changes per BSPEC\")\n Cc: Clint Taylor \u003cclinton.a.taylor@intel.com\u003e\n Signed-off-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Reviewed-by: Clint Taylor \u003cClinton.A.Taylor@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626000352.31926-2-matthew.d.roper@intel.com\n\ncommit f18819a3c8da62a85cd6e0c546fc8b6dd4e4c865\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Jul 1 11:04:51 2019 +0100\n\n drm/i915/guc: Avoid reclaim locks during reset\n \n During reset, we must be very selective in which locks we take as most\n are tainted by being held across a wait or reclaim (kmalloc) which\n implicitly waits. Inside the guc reset path, we reset the ADS to sane\n defaults, but must keep it pinned from initialisation to avoid having to\n pin it during reset.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190701100502.15639-1-chris@chris-wilson.co.uk\n\ncommit 3ee2771e47b038624d68aa051d78ff8558c260b9\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:35 2019 +0300\n\n drm/i915: WARN about invalid lane reversal in TBT-alt/DP-alt modes\n \n Lane reversal happens only in the FIA module for TBT-alt/DP-alt mode, so\n WARN if lane reversal is attempted at a different level. See the\n BSpec DDI_BUF_CTL register description.\n \n Cc: Manasi Navare \u003cmanasi.d.navare@intel.com\u003e\n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-24-imre.deak@intel.com\n\ncommit a171f8e7f466345a2a0c7bacdc7992777e1c1f37\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:34 2019 +0300\n\n drm/i915: Remove unneeded disconnect in TypeC legacy port mode\n \n Disconnecting the TypeC PHY when the port is in legacy mode is not\n necessary:\n - BSpec doesn\u0027t specify a disconnect sequence for legacy mode.\n - The use of the PHY is dedicated for the display in legacy mode.\n - We keep the PHY always connected during runtime as well in legacy\n mode.\n \n We disconnect the PHY when needed during a disabling modeset for the\n port, so we can also remove the disconnect call from the destroy hook.\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-23-imre.deak@intel.com\n\ncommit d5ce34da31456a28fb2c35db0d0f57b7e4af477e\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:33 2019 +0300\n\n drm/i915: Add state verification for the TypeC port mode\n \n Add state verification for the TypeC port mode wrt. the port\u0027s AUX power\n well enabling/disabling. Also check the correctness of changing the port\n mode:\n - When enabling/disabling the AUX power well for a TypeC port we must hold\n the TypeC port lock - the case for AUX transfers - or hold a Type C\n port link reference - the case for modeset enabling/disabling.\n - When changing the TypeC port mode the port\u0027s AUX power domain must be\n disabled.\n \n v2: (Ville)\n - Simplify power_well_async_ref_count().\n - Fix the commit log, clarifying what are the valid conditions to\n enable/disable the AUX power wells.\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-22-imre.deak@intel.com\n\ncommit 24a7bfe0c2d7aec06956d48808cdfe2756f618ad\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:32 2019 +0300\n\n drm/i915: Keep the TypeC port mode fixed when the port is active\n \n The TypeC port mode needs to stay fixed whenever the port is active. Do\n that by introducing a tc_link_refcount to account for active ports,\n avoiding changing the port mode if a reference is held.\n \n During the modeset commit phase we also have to reset the port mode and\n update the active PLL reflecting the new port mode. We can do this only\n once the port and its old PLL has been already disabled. Add the new\n encoder update_prepare/complete hooks that are called around the whole\n enabling sequence. The TypeC specific hooks of these will reset the port\n mode, update the active PLL if the port will be active and ensure that\n the port mode will stay fixed for the duration of the whole enabling\n sequence by holding a tc_link_refcount.\n \n During the port enabling, the pre_pll_enable/post_pll_disable hooks will\n take/release a tc_link_refcount to ensure the port mode stays fixed\n while the port is active.\n \n Changing the port mode should also be avoided during connector detection\n and AUX transfers if the port is active, we\u0027ll do that by checking the\n port\u0027s tc_link_refcount.\n \n When resetting the port mode we also have to take into account the\n maximum lanes provided by the FIA. It\u0027s guaranteed to be 4 in TBT-alt\n and legacy modes, but there may be less lanes available in DP-alt mode,\n in which case we have to fall back to TBT-alt mode.\n \n While at it also update icl_tc_phy_connect()\u0027s code comment, reflecting\n the current way of switching the port mode.\n \n v2:\n - Add the update_prepare/complete hooks to the encoder instead of the\n connector. (Ville)\n - Simplify intel_connector_needs_modeset() by removing redundant if.\n (Ville)\n v3:\n - Fix sparse warning, marking static functions as such.\n v4:\n - Rebase on drm-tip.\n \n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-21-imre.deak@intel.com\n\ncommit eea72c4c2161269a9046184607b71c5cfaabe477\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:31 2019 +0300\n\n drm/i915/icl: Reserve all required PLLs for TypeC ports\n \n When enabling a TypeC port we need to reserve all the required PLLs for\n it, the TBT PLL for TBT-alt and the MG PHY PLL for DP-alt/legacy sinks.\n We can select the proper PLL for the current port mode from the reserved\n PLLs only once we selected and locked down the port mode for the whole\n duration of the port\u0027s active state. Resetting and locking down the port\n mode can in turn happen only during the modeset commit phase once we\n disabled the given port and the PLL it used.\n \n To support the above reserve-and-select PLL semantic we store the\n reserved PLLs along with their HW state in the CRTC state and provide a\n way to select the active PLL from these. The selected PLL along with its\n HW state will be pointed at by crtc_state-\u003eshared_dpll/dpll_hw_state as\n in the case of other port types.\n \n Besides reserving all required PLLs no functional changes.\n \n v2:\n - Fix releasing the ICL PLLs, not clearing the PLLs from the old\n crtc_state.\n - Init port_dpll to ICL_PORT_DPLL_DEFAULT closer to where port_dpll is\n used for symmetry with the corresponding ICL_PORT_DPLL_MG_PHY init.\n (Ville)\n v3:\n - Add FIXME: for clearing the ICL port PLLs from the new crtc state.\n \n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-20-imre.deak@intel.com\n\ncommit 726ca99666dba067ac8fad32bed9d3efcfc57f9d\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:30 2019 +0300\n\n drm/i915/icl: Split getting the DPLLs to port type specific functions\n \n For clarity factor out the combo PHY and TypeC PHY specific code from\n icl_get_dplls() into their own functions.\n \n No functional changes.\n \n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-19-imre.deak@intel.com\n\ncommit 01b24f50b67f49e40172c58d5e8034630f776110\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:29 2019 +0300\n\n drm/i915: Sanitize the shared DPLL find/reference interface\n \n Pass the PLL HW state to the PLL find/reference functions making it\n clearer what is their input. Also pass to these the atomic state and the\n CRTC object instead of the CRTC state, since they don\u0027t require the\n latter.\n \n Move setting the PLL in the crtc_state to the get_dpll() hook, which\n is the more logical place for this, where the related PLL HW state was also\n set.\n \n This refactoring is also a preparation for a follow-up patch that will\n have to find/reference multiple PLLs.\n \n No functional changes.\n \n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-18-imre.deak@intel.com\n\ncommit 866955fa452e1f3d2164bd731d71348d4978e8bc\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:28 2019 +0300\n\n drm/i915: Sanitize the shared DPLL reserve/release interface\n \n For consistency s/intel_get_shared_dpll()/intel_reserve_shared_dplls()/\n to better match intel_release_shared_dplls(). Also, pass to the\n reserve/release and get_dplls/put_dplls hooks the intel_atomic_state and\n CRTC object, that way these functions can look up the old or new state\n as needed.\n \n Also release the PLLs from the atomic state via a new\n put_dplls-\u003eintel_unreference_shared_dpll() call chain for better\n symmetry with the reservation via the\n get_dplls-\u003eintel_reference_shared_dpll() call chain.\n \n Since nothing uses the PLL returned by intel_reserve_shared_dplls(),\n make it return only a bool.\n \n While at it also clarify the reserve/release function docbook headers\n making it clear that multiple DPLLs will be reserved/released and\n whether the new or old atomic CRTC state is affected.\n \n This refactoring is also a preparation for a follow-up change that needs\n to reserve multiple DPLLs.\n \n Kudos to Ville for the idea to pass intel_atomic_state around, to make\n things clearer locally where an object\u0027s old/new atomic state is\n required.\n \n No functional changes.\n \n v2:\n - Fix checkpatch issue: typo in code comment.\n v3:\n - Rebase on drm-tip.\n \n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-17-imre.deak@intel.com\n\ncommit 4f36afb26cbeab7658b80350127b54667d31d618\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:27 2019 +0300\n\n drm/i915: Sanitize the TypeC FIA lane configuration decoding\n \n Use hex numbers, since that makes more sense when decoding a bit pattern.\n \n No functional change.\n \n Suggested-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Animesh Manna \u003canimesh.manna@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-16-imre.deak@intel.com\n\ncommit 8c10e2262663951ab4d43d4f74e282566c04e00c\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:26 2019 +0300\n\n drm/i915: Keep the TypeC port mode fixed for detect/AUX transfers\n \n We must keep the TypeC port mode fixed for the duration of the connector\n detection and each AUX transfers. Add a new TypeC lock holding it around\n these two sequences. For consistency also hold the lock during the port\n mode sanitization.\n \n Whenever resetting the port mode (only during the detection for now) the\n port\u0027s AUX power domain must be disabled already. Flush the async power\n domain disabling work to ensure this.\n \n A follow-up patch will make the port mode changing more robust by\n postponing the change for active ports.\n \n v2:\n - Fix checkpatch issue: missing annotation for tc_lock.\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Paulo Zanoni \u003cpaulo.r.zanoni@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-15-imre.deak@intel.com\n\ncommit 32691b58d157584b18faabf25cce755c1e31c370\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:25 2019 +0300\n\n drm/i915: Fix the TypeC port mode sanitization during loading/resume\n \n For using the correct AUX power domains we have to sanitize the TypeC\n port mode early, so move that before encoder sanitization. To do this\n properly read out the actual port mode instead of just relying on the\n VBT legacy port flag (which can be incorrect).\n \n We also verify that the PHY is connected as expected if the port is\n active. In case the port is inactive we connect the PHY in case of a\n legacy port - as we did so far. The PHY will be connected during\n detection for DP-alt mode - as it was done so far. For TBT-alt mode\n nothing needs to be done to connect the PHY.\n \n v2:\n - Use DRM_DEBUG_KMS instead of DRM_DEBUG_DRIVER. (José)\n v3:\n - Detect TCCOLD any time PORT_TX_DFLEXDPCSSS is read. (Ville)\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Paulo Zanoni \u003cpaulo.r.zanoni@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-14-imre.deak@intel.com\n\ncommit 1cd5ef6ee23cdde8afd53dd72b9f2f789c22f58c\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:24 2019 +0300\n\n drm/i915: Sanitize the TypeC connect/detect sequences\n \n Make the order during detection more consistent: first reset the TypeC\n port mode if needed (adding new helpers for this), then detect any\n connected sink.\n \n To check if a port mode reset is needed determine first the target port\n mode based on the live status if a sink is already connected or the\n PHY status complete flag otherwise.\n \n Add a WARN in legacy mode if unexpectedly we can\u0027t set the unsafe mode\n or if the FIA doesn\u0027t provide the 4 lanes required.\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Paulo Zanoni \u003cpaulo.r.zanoni@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-13-imre.deak@intel.com\n\ncommit 424f109f5c5d9c82a16d0c8dc033ff3d513c6e0f\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:23 2019 +0300\n\n drm/i915: Handle the TCCOLD power-down event\n \n Based on a recent BSpec update (Index/21750) we must handle the TCCOLD\n event associated with the DP-alt mode. We can detect this event by\n reading an invalid all-1s value from FIA registers.\n \n After detecting TCCOLD we will:\n - fall back to TBT-alt mode when attempting to switch to DP-alt mode\n - conclude that nothing is connected during live status detection\n - WARN when already in unsafe mode, since then TCCOLD is unexpected\n \n v2:\n - Use DRM_DEBUG_KMS instead of DRM_DEBUG_DRIVER. (José)\n v3:\n - Use 0xffffffff instead of -1 as invalid FIA reg value.\n (José, Ville)\n - Check for TCCOLD in icl_tc_phy_status_complete() too. (Ville)\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-12-imre.deak@intel.com\n\ncommit ddec362724f98a6725aeafda7701a40635c15909\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:22 2019 +0300\n\n drm/i915: Wait for TypeC PHY complete flag to clear in safe mode\n \n The PHY status complete flag normally clears when disconnecting the PHY\n in DP-alt mode (achieved by switching to safe mode), so wait for the\n flag to clear.\n \n v2:\n - Use DRM_DEBUG_KMS instead of DRM_DEBUG_DRIVER. (José)\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-11-imre.deak@intel.com\n\ncommit c905eb28bd3f43e96ef9180971f7809738caaa27\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:21 2019 +0300\n\n drm/i915: Factor out common parts from TypeC port handling functions\n \n Factor out helpers reading/parsing the TypeC specific registers, making\n current users of them clearer and letting us use them later.\n \n While at it also:\n - Simplify icl_tc_phy_connect() with an early return in legacy mode.\n - Simplify the live status check using one bitmask for all HPD bits.\n - Remove a micro-optimisation of the repeated safe-mode clearing.\n - Make sure we fix the legacy port flag in all cases.\n \n Except for the last two, no functional changes.\n \n v2:\n - Don\u0027t do reg reads at variable declarations. (Jani)\n - Prevent constant truncated compiler warning when assigning the\n valid_hpd_mask. (Nick)\n - s/intel_tc_port_get_lane_info/intel_tc_port_get_lane_mask/ (Ville)\n v3:\n - Make valid_hpd_mask init clear. (Ville)\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Paulo Zanoni \u003cpaulo.r.zanoni@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Cc: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-10-imre.deak@intel.com\n\ncommit ab7bc4e1a5508e5cb974985fba325d80086b3f10\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:20 2019 +0300\n\n drm/i915: Unify the TypeC port notation in debug/error messages\n \n Unify the TypeC port notation in log messages, so that it matches the\n spec. For instance the first ICL TypeC port will read as \u0027Port C/TC#1\u0027.\n \n v2:\n - Format print the name only once. (José)\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Paulo Zanoni \u003cpaulo.r.zanoni@intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-9-imre.deak@intel.com\n\ncommit dd7239c545b1715dbe785a93b6722f66becd9248\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:19 2019 +0300\n\n drm/i915: Use the correct AUX power domain in TypeC TBT-alt mode\n \n In the TypeC TBT-alt port mode we must use the TBT AUX power domain,\n fix that.\n \n Cc: Manasi Navare \u003cmanasi.d.navare@intel.com\u003e\n Cc: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-8-imre.deak@intel.com\n\ncommit 29ae36abf08f943b76a2959f5000c44efa335be7\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:18 2019 +0300\n\n drm/i915: Fix the TBT AUX power well enabling\n \n Fix the mapping from a TBT AUX power well index to the DP_AUX_CH_CTL\n register.\n \n Fixes: c7375d9542f1 (\"drm/i915: Configure AUX_CH_CTL when enabling the AUX power domain\")\n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-7-imre.deak@intel.com\n\ncommit 3b2ed431342948646ed1ecf2ad73e4b54e88572b\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:17 2019 +0300\n\n drm/i915: Don\u0027t enable the DDI-IO power in the TypeC TBT-alt mode\n \n According to the spec we should not enable the DDI-IO power domain if\n the TypeC port is in the TBT-alt mode, so do that only in the other\n TypeC modes or for non-TypeC ports. See the internal BSpec Index/22243.\n \n v2:\n - Add the internal BSpec reference to the log message. (José)\n \n Cc: Manasi Navare \u003cmanasi.d.navare@intel.com\u003e\n Cc: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-6-imre.deak@intel.com\n\ncommit e9b7e1422d409bb7d3423b8d1e59865227083669\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:16 2019 +0300\n\n drm/i915: Sanitize the terminology used for TypeC port modes\n \n The TypeC port mode can switch dynamically, to reflect that better call\n the port\u0027s mode as \u0027mode\u0027 rather than \u0027type\u0027.\n \n While at it:\n - s/TC_PORT_TBT/TC_PORT_TBT_ALT/ and s/TC_PORT_TYPEC/TC_PORT_DP_ALT/.\n \u0027TYPEC\u0027 is ambiguous, TBT_ALT and DP_ALT better match the reality.\n \n - Remove the \u0027unknown\u0027 TypeC port mode. The mode is always known, it\u0027s\n the TBT-alt/safe mode after HW reset and after disconnecting the PHY.\n Simplify the tc_port/tc_type checks accordingly.\n \n - Don\u0027t WARN if the port mode changes, that can happen normally.\n \n No functional changes.\n \n Cc: Animesh Manna \u003canimesh.manna@intel.com\u003e\n Cc: Paulo Zanoni \u003cpaulo.r.zanoni@intel.com\u003e\n Cc: Anusha Srivatsa \u003canusha.srivatsa@intel.com\u003e\n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-5-imre.deak@intel.com\n\ncommit bc85328ff431e41bea3fe88e74e738f2d372e6d1\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:15 2019 +0300\n\n drm/i915: Move the TypeC port handling code to a separate file\n \n Move the TypeC port handling functions to a new file for clarity.\n \n While at it:\n - s/icl_tc_port_connected()/intel_tc_port_connected()/\n icl_tc_phy_disconnect(), will be unexported later.\n \n - s/intel_dp_get_fia_supported_lane_count()/\n intel_tc_port_fia_max_lane_count()/\n It\u0027s used for HDMI legacy mode too.\n \n - Simplify function interfaces by passing only dig_port to them.\n \n No functional changes.\n \n v2:\n - Fix checkpatch issues: +1/-1 empty lines in intel_tc.c and add\n missing SPDX to intel_tc.h. (Jani)\n \n Cc: Animesh Manna \u003canimesh.manna@intel.com\u003e\n Cc: Paulo Zanoni \u003cpaulo.r.zanoni@intel.com\u003e\n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Acked-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-4-imre.deak@intel.com\n\ncommit 28212321622d79cbab640d80d0ce0670fd9eaf72\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:14 2019 +0300\n\n drm/i915: Tune down WARNs about TBT AUX power well enabling\n \n The HW completion flag for the TBT AUX power well enabling/disabling\n gets stuck if the firmware tears down the TBT DP tunnel before the\n completion.\n \n We shouldn\u0027t complain about the timeout, since it\u0027s expected to happen\n and doesn\u0027t cause further issues. We suppress the disabling timeout\n already, do the same for enabling.\n \n v2:\n - Make the debug message more precise. (José)\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-3-imre.deak@intel.com\n\ncommit 39a5883f8670360401876505eaaa64f030df7852\nAuthor: Imre Deak \u003cimre.deak@intel.com\u003e\nDate: Fri Jun 28 17:36:13 2019 +0300\n\n drm/i915/icl: Add support to read out the TBT PLL HW state\n \n Add support to read out the TBT PLL HW state.\n \n Cc: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\n Cc: Paulo Zanoni \u003cpaulo.r.zanoni@intel.com\u003e\n Cc: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Signed-off-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628143635.22066-2-imre.deak@intel.com\n\ncommit fdbdcc83ffd7d00265a531e71f1d166566c09d66\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Wed Jun 12 10:51:47 2019 +0200\n\n drm/bridge: dw-hdmi: Use automatic CTS generation mode when using non-AHB audio\n \n When using an I2S source using a different clock source (usually the I2S\n audio HW uses dedicated PLLs, different from the HDMI PHY PLL), fixed\n CTS values will cause some frequent audio drop-out and glitches as\n reported on Amlogic, Allwinner and Rockchip SoCs setups.\n \n Setting the CTS in automatic mode will let the HDMI controller generate\n automatically the CTS value to match the input audio clock.\n \n The DesignWare DW-HDMI User Guide explains:\n For Automatic CTS generation\n Write \"0\" on the bit field \"CTS_manual\", Register 0x3205: AUD_CTS3\n \n The DesignWare DW-HDMI Databook explains :\n If \"CTS_manual\" bit equals 0b this registers contains \"audCTS[19:0]\"\n generated by the Cycle time counter according to specified timing.\n \n Cc: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Jonas Karlman \u003cjonas@kwiboo.se\u003e\n Cc: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Cc: Jerome Brunet \u003cjbrunet@baylibre.com\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Tested-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Reviewed-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Tested-by: Douglas Anderson \u003cdianders@chromium.org\u003e\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190612085147.26971-1-narmstrong@baylibre.com\n\ncommit 52c2197aae1b9809830d59bc8f5fa95f7372b0c7\nAuthor: Lucas Stach \u003cl.stach@pengutronix.de\u003e\nDate: Thu Jun 27 10:59:58 2019 +0200\n\n drm/bridge: tc358767: do a software reset if reset pin isn\u0027t connected\n \n To get the chip into the expected state, even when the hardware reset pin\n isn\u0027t connected, do a software reset in this case. It isn\u0027t as thorough as\n the hardware reset, as the I2C communication block can not be reset for\n obvious reasons, but it\u0027s getting the chip into a defined state.\n \n Signed-off-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190627085958.28331-1-l.stach@pengutronix.de\n\ncommit 83f35bc3a852f1c3892c7474998c5cec707c7ba3\nAuthor: Matt Redfearn \u003cmatt.redfearn@thinci.com\u003e\nDate: Thu Jun 27 15:18:00 2019 +0000\n\n drm/bridge: adv7511: Attach to DSI host at probe time\n \n In contrast to all of the DSI panel drivers in drivers/gpu/drm/panel\n which attach to the DSI host via mipi_dsi_attach() at probe time, the\n ADV7533 bridge device does not. Instead it defers this to the point that\n the upstream device connects to its bridge via drm_bridge_attach().\n The generic Synopsys MIPI DSI host driver does not register it\u0027s own\n drm_bridge until the MIPI DSI has attached. But it does not call\n drm_bridge_attach() on the downstream device until the upstream device\n has attached. This leads to a chicken and the egg failure and the DRM\n pipeline does not complete.\n Since all other mipi_dsi_device drivers call mipi_dsi_attach() in\n probe(), make the adv7533 mipi_dsi_device do the same. This ensures that\n the Synopsys MIPI DSI host registers it\u0027s bridge such that it is\n available for the upstream device to connect to.\n \n Signed-off-by: Matt Redfearn \u003cmatt.redfearn@thinci.com\u003e\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190627151740.2277-1-matt.redfearn@thinci.com\n\ncommit 99a938889048aa8e92d2a54fadb45498122bf87f\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 30 08:18:52 2019 +0200\n\n drm/stm: drop use of drmP.h\n \n Drop use of the deprecated header file drmP.h\n from the sole user in the stm driver.\n Replace with necessary include files.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Yannick Fertre \u003cyannick.fertre@st.com\u003e\n Cc: Philippe Cornu \u003cphilippe.cornu@st.com\u003e\n Cc: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Cc: Vincent Abriou \u003cvincent.abriou@st.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Maxime Coquelin \u003cmcoquelin.stm32@gmail.com\u003e\n Cc: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n Cc: linux-stm32@st-md-mailman.stormreply.com\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: linux-stm32@st-md-mailman.stormreply.com\n Cc: linux-arm-kernel@lists.infradead.org\n Signed-off-by: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190630061922.7254-4-sam@ravnborg.org\n\ncommit ec193640819e014385c0ccc87e66909e44b4705e\nAuthor: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\nDate: Fri Jun 28 10:55:17 2019 +0200\n\n drm/i915: Use intel state as much as possible in wm code\n \n Instead of directly referencing drm_crtc_state, convert to\n intel_ctc_state and use the base struct. This is useful when we\u0027re\n making the split between uapi and hw state, and also makes the\n code slightly more readable.\n \n A lot of places also use cstate, instead of the more common crtc_state.\n Clean those up to use crtc_state. Same for pstate vs plane_state. (Ville)\n \n Signed-off-by: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628085517.31886-7-maarten.lankhorst@linux.intel.com\n\ncommit 4f25720b2c04a386e7698937f8b8468e223a5ce4\nAuthor: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\nDate: Fri Jun 28 10:55:16 2019 +0200\n\n drm/i915: Pass intel state to plane functions as well\n \n Pass along the correct state as much as possible, instead of relying\n on the drm state internally. This is required to rely on hw state\n internally soon.\n \n While at it, clean up intel_plane_atomic_check slightly, by using a\n helper function to get the intel_crtc. (Ville)\n \n Signed-off-by: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628085517.31886-6-maarten.lankhorst@linux.intel.com\n\ncommit 49743e1dfb2a1f345f21ad84104663d92b4ba440\nAuthor: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\nDate: Fri Jun 28 10:55:15 2019 +0200\n\n drm/i915: Use intel_crtc_state in sanitize_watermarks() too\n \n Get rid of all instances of drm_crtc_state, and rename cstate to\n crtc_state for more clarity.\n \n Signed-off-by: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628085517.31886-5-maarten.lankhorst@linux.intel.com\n\ncommit 3b4bf24d27e031078f167afdeb1a31933129c781\nAuthor: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\nDate: Fri Jun 28 10:55:14 2019 +0200\n\n drm/i915: Convert hw state verifier to take more intel state, v2.\n \n Like the rest of the intel atomic functions we should pass along\n intel_crtc_state, and dereference drm_crtc_state only through\n intel_crtc_state-\u003ebase\n \n While at it, rename old/new_state to old/new_crtc_state. (Ville)\n \n Signed-off-by: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628085517.31886-4-maarten.lankhorst@linux.intel.com\n\ncommit 855e0d684a3ef8cd0c434edfa355279957020626\nAuthor: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\nDate: Fri Jun 28 10:55:13 2019 +0200\n\n drm/i915: Convert most of atomic commit to take more intel state\n \n Instead of passing along drm_crtc_state and drm_atomic_state, pass\n along more intel_atomic_state and intel_crtc_state. This will\n make the code more readable by not casting between drm state\n and intel state all the time.\n \n While at it, rename old_state to state, with the get_new/old helpers\n there is no point in distinguishing between state before and after\n swapping state any more. (Ville)\n \n Signed-off-by: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628085517.31886-3-maarten.lankhorst@linux.intel.com\n\ncommit 69f786aea946230b98e2756769169eb6ddb329cb\nAuthor: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\nDate: Fri Jun 28 10:55:12 2019 +0200\n\n drm/i915: Pass intel_crtc_state to needs_modeset()\n \n In i915 we should use intel_crtc_state as much as possible, pass\n intel_crtc_state to needs_modeset, before we clean up all other uses\n of drm_crtc_state.\n \n Signed-off-by: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190628085517.31886-2-maarten.lankhorst@linux.intel.com\n\ncommit 59634388d93248e4f34ae35a4f3e4b49193ca502\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Thu Jun 27 23:16:43 2019 +0200\n\n MAINTAINERS: add Sam Ravnborg for drm/atmel_hlcdc\n \n I have agreed with Boris Brezillon that we will share the\n maintainer role for the drm/atmel_hlcdc driver.\n Nicolas Ferre from Microchip has donated a few boards that\n allows me to test things - thanks!\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Boris Brezillon \u003cbbrezillon@kernel.org\u003e\n Acked-by: Nicolas Ferre \u003cnicolas.ferre@microchip.com\u003e\n Cc: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Cc: Claudiu Beznea \u003cclaudiu.beznea@microchip.com\u003e\n Cc: Peter Rosin \u003cpeda@axentia.se\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190627211643.GA19853@ravnborg.org\n\ncommit 9f397801110476da9d450886d48b666ee22b39b0\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 23 12:35:42 2019 +0200\n\n drm/mgag200: drop use of drmP.h\n \n Drop use of the deprecated drmP.h header file.\n Replace with necessary include files to fix build.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190623103542.30697-5-sam@ravnborg.org\n\ncommit 36999fd02d92e21142471867973d0e3786370724\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 23 12:35:41 2019 +0200\n\n drm/mga: drop use of drmP.h\n \n Drop the use of the deprecated drmP.h header file.\n Clean up list of include files and sort them.\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190623103542.30697-4-sam@ravnborg.org\n\ncommit 968222d9b817e4249306c6095e9653d8cdd94734\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 23 12:35:40 2019 +0200\n\n drm/mga: make header file self contained\n \n This makes migration away from drmP.h simple\n as we do not need to duplicate dependencies required by mga_drv.h\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190623103542.30697-3-sam@ravnborg.org\n\ncommit 91f85e8feb1b4703d8bad63cda603dacf2a825d4\nAuthor: Sam Ravnborg \u003csam@ravnborg.org\u003e\nDate: Sun Jun 23 12:35:39 2019 +0200\n\n drm/mga: drop dependency on drm_os_linux.h\n \n Opencode all macros used from the deprecated drm_os_linux.h header file.\n The DRM_WAIT_ON used 3 * HZ as timeout.\n This was translated to 3000 msec.\n \n The return value of mga_driver_fence_wait() was not\n used, so make it return void to simplify code a bit.\n \n v2:\n - fixed timeout to 3000 msec (original value was 3 * Hz)\n - drop unused return value from mga_driver_fence_wait()\n \n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Acked-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190623103542.30697-2-sam@ravnborg.org\n\ncommit 583bbf46133c726bae277e8f4e32bfba2a528c7f\nAuthor: Steven Price \u003csteven.price@arm.com\u003e\nDate: Thu Jun 27 16:53:18 2019 +0100\n\n drm/panfrost: Use drm_gem_map_offset()\n \n panfrost_ioctl_mmap_bo() contains a reimplementation of\n drm_gem_map_offset() but with a bug - it allows mapping imported\n objects (without going through the exporter). Fix this by switching to\n use the newly renamed drm_gem_map_offset() function instead which has\n the bonus of simplifying the code.\n \n Signed-off-by: Steven Price \u003csteven.price@arm.com\u003e\n Reviewed-by: Alyssa Rosenzweig \u003calyssa@rosenzweig.io\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190627155318.38053-3-steven.price@arm.com\n\ncommit 220df83a5394fbf7c1486ba7848794b7b351d598\nAuthor: Steven Price \u003csteven.price@arm.com\u003e\nDate: Thu Jun 27 16:53:17 2019 +0100\n\n drm/gem: Rename drm_gem_dumb_map_offset() to drm_gem_map_offset()\n \n drm_gem_dumb_map_offset() is a useful helper for non-dumb clients, so\n rename it to remove the _dumb and add a comment that it can be used by\n shmem clients.\n \n Signed-off-by: Steven Price \u003csteven.price@arm.com\u003e\n Acked-by: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190627155318.38053-2-steven.price@arm.com\n\ncommit 501ec325f1ec4338c0b6d32001a6f4cdfc61a6b2\nAuthor: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\nDate: Tue Jun 25 10:54:11 2019 -0700\n\n drm/i915: rework reading pipe disable fuses\n \n This prepares to have possibly more than 3 pipes. I didn\u0027t want to\n continue the previous approach since the check for \"are the disabled\n pipes the last ones\" poses a combinatory explosion. We need that check\n because in several places of the code we have that assumption. If that\n ever becomes false in a new HW, other parts of the code would have to\n change.\n \n Now we start by considering we have info-\u003enum_pipes enabled and disable\n each pipe that is marked as disabled. Then it\u0027s a simple matter of\n checking if we have at least one pipe and that all the enabled ones are\n the first pipes, i.e. there are no holes in the bitmask.\n \n Cc: Jose Souza \u003cjose.souza@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Lucas De Marchi \u003clucas.demarchi@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190625175437.14840-3-lucas.demarchi@intel.com\n\ncommit 52500de68fbb0c11403fd8db13aace99e5ef4f40\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Tue May 21 00:35:00 2019 +0200\n\n drm/vmwgfx: Don\u0027t look at state-\u003eallow_modeset\n \n That\u0027s purely for the uapi layer to implement the ALLOW_MODESET flag.\n \n Drivers should instead look at the state, e.g. through\n drm_atomic_crtc_needs_modeset(), which vmwgfx already does. Also remove\n the confusing comment, since checking allow_modeset is at best a micro\n optimization.\n \n v2: Rebase\n \n Reviewed-by: Deepak Rawat \u003cdrawat@vmware.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: VMware Graphics \u003clinux-graphics-maintainer@vmware.com\u003e\n Cc: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190520223500.6032-1-daniel.vetter@ffwll.ch\n\ncommit 8735f16803f00f5efca7738afe3b9a304b539181\nAuthor: Christian König \u003cchristian.koenig@amd.com\u003e\nDate: Wed Jun 26 16:31:46 2019 +0200\n\n dma-buf: cleanup reservation_object_init/fini\n \n They are not used that often and certainly not in a hot path.\n Make them normal functions instead of an inline.\n \n Signed-off-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/314480/\n\ncommit 5ed7191dd97b212887a9cb4715c2ef09ea50ae47\nAuthor: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\nDate: Thu Jun 27 19:34:10 2019 +0200\n\n drm/ast: Replace struct ast_framebuffer with GEM framebuffer helpers\n \n The ast driver\u0027s struct ast_framebuffer is a buffer object with GEM\n interface. There are already GEM framebuffer helpers that implement\n the same functionality. Convert ast to these.\n \n Signed-off-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Acked-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190627173410.8300-1-tzimmermann@suse.de\n\ncommit 66ab700506e729e49ca13586a7b0201a1c2093af\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Tue Jun 25 22:42:04 2019 +0200\n\n drm/vc4: Use drm_gem_fb_prepare_fb\n \n vc4 has switched to using drm_fb-\u003eobj[], so we can just use the helper\n unchanged.\n \n v2: Make it compile ... oops.\n \n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190625204208.5614-4-daniel.vetter@ffwll.ch\n\ncommit e78ad7654179f51e9ed99fe85358f4f3bbbda7bc\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Tue Jun 25 22:42:03 2019 +0200\n\n drm/msm: Use drm_gem_fb_prepare_fb\n \n msm has switched over to drm_fb-\u003eobj[] a while ago already, so we can\n just use the helper.\n \n v2: Make it compile ... oops.\n \n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Rob Clark \u003crobdclark@gmail.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Rob Clark \u003crobdclark@gmail.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: Jeykumar Sankaran \u003cjsanka@codeaurora.org\u003e\n Cc: Jordan Crouse \u003cjcrouse@codeaurora.org\u003e\n Cc: Bruce Wang \u003cbzwang@chromium.org\u003e\n Cc: Fritz Koenig \u003cfrkoenig@google.com\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: linux-arm-msm@vger.kernel.org\n Cc: freedreno@lists.freedesktop.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190625204208.5614-3-daniel.vetter@ffwll.ch\n\ncommit bd630a86be381992fac99f9ab82c5c5b43a5ee3b\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Tue Jun 25 22:42:02 2019 +0200\n\n drm/fb-helper: use gem_bo.resv, not dma_buf.resv in prepare_fb\n \n With\n \n commit 5f6ed9879a414636405a2bd77f122881695959e4\n Author: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Date: Fri Jun 14 22:35:57 2019 +0200\n \n drm/prime: automatically set gem_obj-\u003eresv on import\n \n we consistently set drm_gem_bo.resv for imported buffers. Which means\n we don\u0027t need to check the dma-buf in the prepare_fb helper, but can\n generalize them so they\u0027re also useful for display+render drivers\n which use gem_bo.resv to track their own rendering for their own\n scanout buffers.\n \n Cc: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: Rob Clark \u003crobdclark@gmail.com\u003e\n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Rob Clark \u003crobdclark@gmail.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190625204208.5614-2-daniel.vetter@ffwll.ch\n\ncommit 5d17718997367c435dbe5341a8e270d9b19478d3\nAuthor: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\nDate: Thu Jun 27 10:09:09 2019 +0200\n\n drm/mgag200: Replace struct mga_framebuffer with GEM framebuffer helpers\n \n The mgag200 driver\u0027s struct mga_framebuffer is a buffer object with GEM\n interface. There are already GEM framebuffer helpers that implement the\n same functionality. Convert mgag200 to these.\n \n Signed-off-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Acked-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190627080909.30471-1-tzimmermann@suse.de\n\ncommit 7218524d3ea00dda75d2f0989ab4dcb631f00b61\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jun 26 23:42:12 2019 +0100\n\n drm/i915: Make i945gm_vblank_work_func static\n \n drivers/gpu/drm/i915/i915_irq.c:3382:6: warning: symbol \u0027i945gm_vblank_work_func\u0027 was not declared. Should it be static?\n CC [M] drivers/gpu/drm/i915/i915_irq.o\n drivers/gpu/drm/i915/i915_irq.c:3382:6: error: no previous prototype for ‘i945gm_vblank_work_func’ [-Werror\u003dmissing-prototypes]\n void i945gm_vblank_work_func(struct work_struct *work)\n \n Jani wrote the idential patch, so for posterity:\n \n The static keyword was apparently accidentally removed in commit\n 08fa8fd0faa5 (\"drm/i915: Switch to per-crtc vblank vfuncs\"), leading to\n sparse warning:\n \n drivers/gpu/drm/i915/i915_irq.c:3382:6: warning: symbol\n \u0027i945gm_vblank_work_func\u0027 was not declared. Should it be static?\n \n Make the function static again.\n \n Meanwhile, the 0-day kbuilder also spotted the mistake.\n \n Fixes: 08fa8fd0faa5 (\"drm/i915: Switch to per-crtc vblank vfuncs\")\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626224212.10141-1-chris@chris-wilson.co.uk\n Link: https://patchwork.freedesktop.org/patch/msgid/20190627091914.30795-1-jani.nikula@intel.com\n\ncommit fdb29b738017672069f95747c48eac947b9beba2\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Jun 18 22:27:16 2019 -0700\n\n drm/bridge: tc358767: Add support for address-only I2C transfers\n \n Transfer size of zero means a request to do an address-only\n transfer. Since the HW support this, we probably shouldn\u0027t be just\n ignoring such requests. While at it allow DP_AUX_I2C_MOT flag to pass\n through, since it is supported by the HW as well.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Reviewed-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Cc: Laurent Pinchart \u003cLaurent.pinchart@ideasonboard.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrey Gusakov \u003candrey.gusakov@cogentembedded.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619052716.16831-16-andrew.smirnov@gmail.com\n\ncommit 32d3621996aa70f4db9bbf862e400471eed81e34\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Jun 18 22:27:15 2019 -0700\n\n drm/bridge: tc358767: Replace magic number in tc_main_link_enable()\n \n We don\u0027t need 8 byte array, DP_LINK_STATUS_SIZE (6) should be\n enough. This also gets rid of a magic number as a bonus.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Reviewed-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Cc: Laurent Pinchart \u003cLaurent.pinchart@ideasonboard.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrey Gusakov \u003candrey.gusakov@cogentembedded.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619052716.16831-15-andrew.smirnov@gmail.com\n\ncommit d174db0788b437d83b7fed4531dbe8ace2c0b2a8\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Jun 18 22:27:14 2019 -0700\n\n drm/bridge: tc358767: Drop unnecessary 8 byte buffer\n \n tc_get_display_props() never reads more than a byte via AUX, so\n there\u0027s no need to reserve 8 for that purpose. No function change\n intended.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Reviewed-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Cc: Laurent Pinchart \u003cLaurent.pinchart@ideasonboard.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrey Gusakov \u003candrey.gusakov@cogentembedded.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619052716.16831-14-andrew.smirnov@gmail.com\n\ncommit 7264892648bac9c1a5f20505a569abf85370b7c4\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Jun 18 22:27:13 2019 -0700\n\n drm/bridge: tc358767: Simplify tc_aux_wait_busy()\n \n We never pass anything but 100 as timeout_ms to tc_aux_wait_busy(), so\n we may as well hardcode that value and simplify function\u0027s signature.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Reviewed-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Cc: Laurent Pinchart \u003cLaurent.pinchart@ideasonboard.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrey Gusakov \u003candrey.gusakov@cogentembedded.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619052716.16831-13-andrew.smirnov@gmail.com\n\ncommit 134fb306b1d8af36d586a08b2f0562b5d30bfab9\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Jun 18 22:27:12 2019 -0700\n\n drm/bridge: tc358767: Introduce tc_pllupdate()\n \n tc_wait_pll_lock() is always called as a follow-up for updating\n PLLUPDATE and PLLEN bit of a given PLL control register. To simplify\n things, merge the two operation into a single helper function\n tc_pllupdate() and convert the rest of the code to use it. No\n functional change intended.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Reviewed-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Cc: Laurent Pinchart \u003cLaurent.pinchart@ideasonboard.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrey Gusakov \u003candrey.gusakov@cogentembedded.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619052716.16831-12-andrew.smirnov@gmail.com\n\ncommit c49f60dfbf91ead366a7650ca7fb2bbf8fe4a4a0\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Jun 18 22:27:11 2019 -0700\n\n drm/bridge: tc358767: Introduce tc_set_syspllparam()\n \n Move common code converting clock rate to an appropriate constant and\n configuring SYS_PLLPARAM register into a separate routine and convert\n the rest of the code to use it. No functional change intended.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Reviewed-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Cc: Laurent Pinchart \u003cLaurent.pinchart@ideasonboard.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrey Gusakov \u003candrey.gusakov@cogentembedded.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619052716.16831-11-andrew.smirnov@gmail.com\n\ncommit 12dfe7c4d9c53442efcb4113cff505f60d654906\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Jun 18 22:27:10 2019 -0700\n\n drm/bridge: tc358767: Use reported AUX transfer size\n \n Don\u0027t assume that requested data transfer size is the same as amount\n of data that was transferred. Change the code to get that information\n from DP0_AUXSTATUS instead.\n \n Since the check for AUX_BUSY in tc_aux_get_status() is pointless (it\n will always called after tc_aux_wait_busy()) and there\u0027s only one user\n of it, inline its code into tc_aux_transfer() instead of trying to\n accommodate the change above.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Reviewed-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Cc: Laurent Pinchart \u003cLaurent.pinchart@ideasonboard.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrey Gusakov \u003candrey.gusakov@cogentembedded.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619052716.16831-10-andrew.smirnov@gmail.com\n\ncommit e0655feaec62d5139b6b13a7b1bbb1ab8f1c2d83\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Jun 18 22:27:09 2019 -0700\n\n drm/bridge: tc358767: Increase AUX transfer length limit\n \n According to the datasheet tc358767 can transfer up to 16 bytes via\n its AUX channel, so the artificial limit of 8 appears to be too\n low. However only up to 15-bytes seem to be actually supported and\n trying to use 16-byte transfers results in transfers failing\n sporadically (with bogus status in case of I2C transfers), so limit it\n to 15.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Reviewed-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Cc: Laurent Pinchart \u003cLaurent.pinchart@ideasonboard.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrey Gusakov \u003candrey.gusakov@cogentembedded.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619052716.16831-9-andrew.smirnov@gmail.com\n\ncommit 792a081a1b8e819898c506ea1b454ce6bcf3ba1b\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Jun 18 22:27:08 2019 -0700\n\n drm/bridge: tc358767: Simplify AUX data write\n \n Simplify AUX data write by dropping index arithmetic and shifting and\n replacing it with a call to a helper function that does two things:\n \n 1. Copies user-provided data into a write buffer\n 2. Transfers contents of the write buffer to up to 4 32-bit\n registers on the chip\n \n Note that separate data endianness fix:\n \n tmp \u003d (tmp \u003c\u003c 8) | buf[i];\n \n that was reserved for DP_AUX_I2C_WRITE looks really strange, since it\n will place data differently depending on the passed user-data\n size. E.g. for a write of 1 byte, data transferred to the chip would\n look like:\n \n [byte0] [dummy1] [dummy2] [dummy3]\n \n whereas for a write of 4 bytes we\u0027d get:\n \n [byte3] [byte2] [byte1] [byte0]\n \n Since there\u0027s no indication in the datasheet that I2C write buffer\n should be treated differently than AUX write buffer and no comment in\n the original code explaining why it was done this way, that special\n I2C write buffer transformation was dropped in this patch.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Reviewed-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Cc: Laurent Pinchart \u003cLaurent.pinchart@ideasonboard.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrey Gusakov \u003candrey.gusakov@cogentembedded.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619052716.16831-8-andrew.smirnov@gmail.com\n\ncommit 53b166dca5feb96859cd235131b96b426ca1c998\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Jun 18 22:27:07 2019 -0700\n\n drm/bridge: tc358767: Simplify AUX data read\n \n Simplify AUX data read by removing index arithmetic and shifting with\n a helper function that does two things:\n \n 1. Fetch data from up to 4 32-bit registers from the chip\n 2. Copy read data into user provided array.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Reviewed-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Cc: Laurent Pinchart \u003cLaurent.pinchart@ideasonboard.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrey Gusakov \u003candrey.gusakov@cogentembedded.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619052716.16831-7-andrew.smirnov@gmail.com\n\ncommit 6d0c38315915d49fbbcc55340c06f811a55a4885\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Jun 18 22:27:06 2019 -0700\n\n drm/bridge: tc358767: Drop custom tc_write()/tc_read() accessors\n \n A very unfortunate aspect of tc_write()/tc_read() macro helpers is\n that they capture quite a bit of context around them and thus require\n the caller to have magic variables \u0027ret\u0027 and \u0027tc\u0027 as well as label\n \u0027err\u0027. That makes a number of code paths rather counter-intuitive and\n somewhat clunky, for example tc_stream_clock_calc() ends up being like\n this:\n \n int ret;\n \n tc_write(DP0_VIDMNGEN1, 32768);\n \n return 0;\n err:\n return ret;\n \n which is rather surprising when you read the code for the first\n time. Since those helpers arguably aren\u0027t really saving that much code\n and there\u0027s no way of fixing them without making them too verbose to\n be worth it change the driver code to not use them at all.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Reviewed-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Cc: Laurent Pinchart \u003cLaurent.pinchart@ideasonboard.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrey Gusakov \u003candrey.gusakov@cogentembedded.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619052716.16831-6-andrew.smirnov@gmail.com\n\ncommit 3f072c304c0a7e1045ef2e5478403c0137746878\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Jun 18 22:27:05 2019 -0700\n\n drm/bridge: tc358767: Simplify tc_set_video_mode()\n \n Simplify tc_set_video_mode() by replacing explicit shifting using\n macros from \u003clinux/bitfield.h\u003e. No functional change intended.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Reviewed-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Cc: Laurent Pinchart \u003cLaurent.pinchart@ideasonboard.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrey Gusakov \u003candrey.gusakov@cogentembedded.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619052716.16831-5-andrew.smirnov@gmail.com\n\ncommit aa92213f388ba88f895cc8eb432b95925b890e3b\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Jun 18 22:27:04 2019 -0700\n\n drm/bridge: tc358767: Simplify polling in tc_link_training()\n \n Replace explicit polling in tc_link_training() with equivalent call to\n tc_poll_timeout() for simplicity. No functional change intended (not\n including slightly altered debug output).\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Reviewed-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Cc: Laurent Pinchart \u003cLaurent.pinchart@ideasonboard.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrey Gusakov \u003candrey.gusakov@cogentembedded.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619052716.16831-4-andrew.smirnov@gmail.com\n\ncommit ebcce4e642474764989f5c213394580ae5700ed7\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Jun 18 22:27:03 2019 -0700\n\n drm/bridge: tc358767: Simplify polling in tc_main_link_setup()\n \n Replace explicit polling loop with equivalent call to\n tc_poll_timeout() for brevity. No functional change intended.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Reviewed-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Cc: Laurent Pinchart \u003cLaurent.pinchart@ideasonboard.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrey Gusakov \u003candrey.gusakov@cogentembedded.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619052716.16831-3-andrew.smirnov@gmail.com\n\ncommit 93a105694f7dbf47ea5e8c60a04502fb72c3fc34\nAuthor: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\nDate: Tue Jun 18 22:27:02 2019 -0700\n\n drm/bridge: tc358767: Simplify tc_poll_timeout()\n \n Implementation of tc_poll_timeout() is almost a 100% copy-and-paste of\n the code for regmap_read_poll_timeout(). Replace copied code with a\n call to the original. While at it change tc_poll_timeout to accept\n \"struct tc_data *\" instead of \"struct regmap *\" for brevity. No\n functional change intended.\n \n Signed-off-by: Andrey Smirnov \u003candrew.smirnov@gmail.com\u003e\n Reviewed-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Reviewed-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Cc: Laurent Pinchart \u003cLaurent.pinchart@ideasonboard.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Andrey Gusakov \u003candrey.gusakov@cogentembedded.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Cory Tusar \u003ccory.tusar@zii.aero\u003e\n Cc: Chris Healy \u003ccphealy@gmail.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-kernel@vger.kernel.org\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619052716.16831-2-andrew.smirnov@gmail.com\n\ncommit 28a902e6d7fbe469b0322da39128a5e9833dbb17\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Mon Jun 24 15:56:35 2019 -0600\n\n dt-bindings: display: rockchip-lvds: Remove panel references\n \n The panel bindings are outside the scope of the Rockchip LVDS interface\n binding. The references are about to change too, so rather than update\n them just drop the section on the panel bindings.\n \n Cc: Sandy Huang \u003chjc@rock-chips.com\u003e\n Cc: \"Heiko Stübner\" \u003cheiko@sntech.de\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: dri-devel@lists.freedesktop.org\n Cc: linux-rockchip@lists.infradead.org\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190624215649.8939-2-robh@kernel.org\n\ncommit e61576c4b9ff59aea2719e3681c4a8809998555e\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Fri Jun 14 15:39:11 2019 +0100\n\n drm/mgag200: add in missing { } around if block\n \n There is an if block that is missing the { } curly brackets. Add\n these in.\n \n Addresses-Coverity: (\"Structurally dead code\")\n Fixes: 94dc57b10399 (\"drm/mgag200: Rewrite cursor handling\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614143911.21806-1-colin.king@canonical.com\n\ncommit b40cf94c76732fbb826207a6456b5c1ac9e14d23\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Wed Jun 26 17:40:19 2019 +0300\n\n drm/i915: make intel_uc_fw.h self-contained\n \n Add the minimal includes/declarations to make the header self-contained,\n and ensure it stays that way.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626144020.2155-13-jani.nikula@intel.com\n\ncommit ba740cfc2f262ff02853c0a76a5b4dd7bc628302\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Wed Jun 26 17:40:18 2019 +0300\n\n drm/i915: make intel_gvt.h self-contained\n \n Add the minimal includes/declarations to make the header self-contained,\n and ensure it stays that way.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626144020.2155-12-jani.nikula@intel.com\n\ncommit 5e0cca98b4d154e235abbb29f7edda5e6b4f4b60\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Wed Jun 26 17:40:17 2019 +0300\n\n drm/i915: make intel_guc_reg.h self-contained\n \n Add the minimal includes/declarations to make the header self-contained,\n and ensure it stays that way.\n \n v2: also include i915_reg.h (Michal)\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626144020.2155-11-jani.nikula@intel.com\n\ncommit 72629d11d56c44ddaae273dd513319a1fb5c03bd\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Wed Jun 26 17:40:16 2019 +0300\n\n drm/i915: make intel_guc_fwif.h self-contained\n \n Add the minimal includes/declarations to make the header self-contained,\n and ensure it stays that way.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626144020.2155-10-jani.nikula@intel.com\n\ncommit 016c1c8938b0402373b92606fecdaca623ab5006\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Wed Jun 26 17:40:15 2019 +0300\n\n drm/i915: make intel_guc_ct.h self-contained\n \n Add the minimal includes/declarations to make the header self-contained,\n and ensure it stays that way.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626144020.2155-9-jani.nikula@intel.com\n\ncommit 961ebc9dd1f56e84dc4ef0aec2d05f63aefc898f\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Wed Jun 26 17:40:14 2019 +0300\n\n drm/i915: make i915_vgpu.h self-contained\n \n Add the minimal includes/declarations to make the header self-contained,\n and ensure it stays that way.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626144020.2155-8-jani.nikula@intel.com\n\ncommit f8daf6418e3041ac967b45fc066539ba1eec5132\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Wed Jun 26 17:40:13 2019 +0300\n\n drm/i915: make i915_pvinfo.h self-contained\n \n Add the minimal includes/declarations to make the header self-contained,\n and ensure it stays that way.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626144020.2155-7-jani.nikula@intel.com\n\ncommit f807d31a2174409c787e53fb6f0609d1b9783e3e\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Wed Jun 26 17:40:12 2019 +0300\n\n drm/i915: make i915_globals.h self-contained\n \n Add the minimal includes/declarations to make the header self-contained,\n and ensure it stays that way.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626144020.2155-6-jani.nikula@intel.com\n\ncommit c75299aea22a92abf5e097830915162655fb46b5\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Wed Jun 26 17:40:10 2019 +0300\n\n drm/i915: make i915_fixed.h self-contained\n \n Add the minimal includes/declarations to make the header self-contained,\n and ensure it stays that way.\n \n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626144020.2155-4-jani.nikula@intel.com\n\ncommit 7fcc7ca549d4889025d38c983681bfada3ff1f3a\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Wed Jun 26 17:36:18 2019 +0300\n\n drm/i915: add header search path to subdir Makefiles\n \n With the subdirectories we lost the ability to build individual files on\n the command line, for example:\n \n $ make drivers/gpu/drm/i915/display/intel_display.o\n \n This was due to the top level directory missing from header search\n path. Add the header search paths to subdir Makefiles.\n \n Note that none of the other options in the top level i915 Makefile are\n taken into account when building individual files. Usually this is not a\n concern.\n \n Reported-by: Imre Deak \u003cimre.deak@intel.com\u003e\n Reported-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626143618.21800-2-jani.nikula@intel.com\n\ncommit 9ef424e58e21b301bc46bbe965deacfac9f72de9\nAuthor: Jani Nikula \u003cjani.nikula@intel.com\u003e\nDate: Wed Jun 26 17:36:17 2019 +0300\n\n drm/i915: prefix header search path with $(srctree)/\n \n Per commit 43068cb7ba1f (\"drm: prefix header search paths with\n $(srctree)/\") this is what we should be doing. Follow suit.\n \n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626143618.21800-1-jani.nikula@intel.com\n\ncommit e5ff5344f4a44cf9f4606c5c2754d913f06aaae1\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 7 00:27:51 2019 +0200\n\n drm/vkms: No need for -\u003epages_lock in crc work anymore\n \n We\u0027re now guaranteed to no longer race against prepare_fb/cleanup_fb,\n which means we can access -\u003evaddr without having to hold a lock.\n \n Before the previous patches it was fairly easy to observe the cursor\n -\u003evaddr being invalid, but that\u0027s now gone, so we can upgrade to a\n full WARN_ON.\n \n Cc: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Cc: Haneen Mohammed \u003chamohammed.sa@gmail.com\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Reviewed-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Tested-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Signed-off-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190606222751.32567-11-daniel.vetter@ffwll.ch\n\ncommit 8b1865873651daeffe412059c2c3c37021b9d92f\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 7 00:27:50 2019 +0200\n\n drm/vkms: totally reworked crc data tracking\n \n The crc computation worker needs to be able to get at some data\n structures and framebuffer mappings, while potentially more atomic\n updates are going on. The solution thus far is to copy relevant bits\n around, but that\u0027s very tedious.\n \n Here\u0027s a new approach, which tries to be more clever, but relies on a\n few not-so-obvious things:\n - crtc_state is always updated when a plane_state changes. Therefore\n we can just stuff plane_state pointers into a crtc_state. That\n solves the problem of easily getting at the needed plane_states.\n - with the flushing changes from previous patches the above also holds\n without races due to the next atomic update being a bit eager with\n cleaning up pending work - we always wait for all crc work items to\n complete before unmapping framebuffers.\n - we also need to make sure that the hrtimer fires off the right\n worker. Keep a new distinct crc_state pointer, under the\n vkms_output-\u003elock protection for this. Note that crtc-\u003estate is\n updated very early in the atomic commit, way before we arm the\n vblank event - the vblank event should always match the buffers we\n use to compute the crc. This also solves an issue in the hrtimer,\n where we\u0027ve accessed drm_crtc-\u003estate without holding the right locks\n (we held none - oops).\n - in the worker itself we can then just access the plane states we\n need, again solving a bunch of ordering and locking issues.\n Accessing plane-\u003estate requires locks, accessing the private\n vkms_crtc_state-\u003eactive_planes pointer only requires that the memory\n doesn\u0027t get freed too early.\n \n The idea behind vkms_crtc_state-\u003eactive_planes is that this would\n contain all visible planes, in z-order, as a first step towards a more\n generic blending implementation.\n \n Note that this patch also fixes races between prepare_fb/cleanup_fb\n and the crc worker accessing -\u003evaddr.\n \n Cc: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Cc: Haneen Mohammed \u003chamohammed.sa@gmail.com\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Reviewed-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Tested-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Signed-off-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190606222751.32567-10-daniel.vetter@ffwll.ch\n\ncommit 1c305e13ecc5579874464842e969f69ae032f4e4\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 7 00:27:49 2019 +0200\n\n drm/vkms: No _irqsave within spin_lock_irq needed\n \n irqs are already off.\n \n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Reviewed-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Tested-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Signed-off-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190606222751.32567-9-daniel.vetter@ffwll.ch\n\ncommit 64cfaa5092715a00a5cdbd3d804d16a38554fe40\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 7 00:27:48 2019 +0200\n\n drm/vkms: Dont flush crc worker when we change crc status\n \n The crc core code can cope with some late crc, the race is kinda\n unavoidable. So no need to flush pending workers, they\u0027ll complete in\n time.\n \n Cc: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Cc: Haneen Mohammed \u003chamohammed.sa@gmail.com\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Reviewed-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Tested-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Signed-off-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190606222751.32567-8-daniel.vetter@ffwll.ch\n\ncommit 5ef8100a3919fb2fabae908554dc0823063e5be4\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 7 00:27:47 2019 +0200\n\n drm/vkms: flush crc workers earlier in commit flow\n \n Currently, we flush pending CRC workers very late in the commit flow,\n when we destroy all the old crtc states. Unfortunately, at that point,\n the framebuffers are already unpinned (and our vaddr possible gone), so\n this isn\u0027t good. Also, the plane_states we need might also already be\n cleaned up, since cleanup order of state structures isn\u0027t well defined.\n \n Fix this by waiting for all CRC workers of the old state to complete\n before we start any of the cleanup work. For correct ordering and\n avoiding races, we can only flush_work after\n drm_atomic_helper_wait_for_vblanks() since we know that all subsequent\n queue_work will be for the new state. Only once that\u0027s done is\n flush_work() useful, before that we might flush the work, and then right\n after the hrtimer that simulates vblank queues it again. Every time you\n have a flush_work before cleaning up the work structure, the following\n sequence must be obeyed, or it can go wrong:\n \n 1. Make sure no one else can re-queue the work anymore (in our case\n that\u0027s done by a combination of first updating output-\u003ecrc_state and\n then waiting for the vblank to pass to make sure the hrtimer has noticed\n that change).\n 2. flush_work()\n 3. Actually clean up stuff (which isn\u0027t done here).\n \n Doing the flush_work before we even completed the output-\u003estate update,\n much less waited for the vblank to make sure that\u0027s happened, missed the\n point.\n \n Note that this is not yet race-free because of the hrtimer and crc\n worker look at the wrong state pointers, but that will be fixed in\n subsequent patches.\n \n Cc: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Cc: Haneen Mohammed \u003chamohammed.sa@gmail.com\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Reviewed-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Tested-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Signed-off-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190606222751.32567-7-daniel.vetter@ffwll.ch\n\ncommit 3d08eb7ddb4d24af2e6ce80197ea3525e01c5eb8\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 7 00:27:46 2019 +0200\n\n drm/vkms: Add our own commit_tail\n \n Just prep work, more will be done here in following patches.\n \n Cc: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Cc: Haneen Mohammed \u003chamohammed.sa@gmail.com\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Reviewed-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Tested-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Signed-off-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190606222751.32567-6-daniel.vetter@ffwll.ch\n\ncommit fb4155fa4c0c0db488a9054502a4aef134a11c0f\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 7 00:27:44 2019 +0200\n\n drm/vkms: Rename vkms_output.state_lock to crc_lock\n \n Plus add a comment about what it actually protects. It\u0027s very little.\n \n Cc: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Cc: Haneen Mohammed \u003chamohammed.sa@gmail.com\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Reviewed-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Tested-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Signed-off-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190606222751.32567-4-daniel.vetter@ffwll.ch\n\ncommit a0e4e5754ccec8c3ae518eaf3830865ae2b39efe\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 7 00:27:43 2019 +0200\n\n drm/vkms: Use spin_lock_irq in process context\n \n The worker is always in process context, no need for the _irqsafe\n version. Same for the set_source callback, that\u0027s only called from the\n debugfs handler in a syscall.\n \n Cc: Shayenne Moura \u003cshayenneluzmoura@gmail.com\u003e\n Cc: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Cc: Haneen Mohammed \u003chamohammed.sa@gmail.com\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Reviewed-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Tested-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Signed-off-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190606222751.32567-3-daniel.vetter@ffwll.ch\n\ncommit 18d0952a838ba559655b0cd9cf85097ad63d9bca\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 7 00:27:42 2019 +0200\n\n drm/vkms: Fix crc worker races\n \n The issue we have is that the crc worker might fall behind. We\u0027ve\n tried to handle this by tracking both the earliest frame for which it\n still needs to compute a crc, and the last one. Plus when the\n crtc_state changes, we have a new work item, which are all run in\n order due to the ordered workqueue we allocate for each vkms crtc.\n \n Trouble is there\u0027s been a few small issues in the current code:\n - we need to capture frame_end in the vblank hrtimer, not in the\n worker. The worker might run much later, and then we generate a lot\n of crc for which there\u0027s already a different worker queued up.\n - frame number might be 0, so create a new crc_pending boolean to\n track this without confusion.\n - we need to atomically grab frame_start/end and clear it, so do that\n all in one go. This is not going to create a new race, because if we\n race with the hrtimer then our work will be re-run.\n - only race that can happen is the following:\n 1. worker starts\n 2. hrtimer runs and updates frame_end\n 3. worker grabs frame_start/end, already reading the new frame_end,\n and clears crc_pending\n 4. hrtimer calls queue_work()\n 5. worker completes\n 6. worker gets re-run, crc_pending is false\n Explain this case a bit better by rewording the comment.\n \n v2: Demote warning level output to debug when we fail to requeue, this\n is expected under high load when the crc worker can\u0027t quite keep up.\n \n Cc: Shayenne Moura \u003cshayenneluzmoura@gmail.com\u003e\n Cc: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Cc: Haneen Mohammed \u003chamohammed.sa@gmail.com\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Reviewed-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Tested-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Signed-off-by: Rodrigo Siqueira \u003crodrigosiqueiramelo@gmail.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190606222751.32567-2-daniel.vetter@ffwll.ch\n\ncommit 5ed7a0cf339463d29163bdb828913d9a3f8c8c9e\nAuthor: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\nDate: Wed Jun 26 12:38:26 2019 +0000\n\n drm/i915: Move OA files to separate folder\n \n OA files look to be auto-generated so we can keep them all in\n dedicated subdirectory.\n \n Signed-off-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Acked-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Acked-by: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Acked-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Reviewed-by: Umesh Nerlige Ramappa \u003cumesh.nerlige.ramappa@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626123826.39760-1-michal.wajdeczko@intel.com\n\ncommit c1a495a558536c5745c6449f7c04ba3cf40be9da\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jun 26 10:43:30 2019 +0100\n\n drm: Allow range of 0 for drm_mm_insert_node_in_range()\n \n We gracefully handle the caller specifying a zero range, so don\u0027t force\n them to special case that condition if it naturally falls out of their\n setup. What we don\u0027t check is if the end \u003c start, so keep that as an\n assert for an illegal call.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626094330.3556-1-chris@chris-wilson.co.uk\n\ncommit 63c9dae71dc53928a82deb32e4651a4a44c7a4fe\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Tue Jun 25 18:40:53 2019 -0700\n\n drm/i915/ehl: Add voltage level requirement table\n \n EHL has it own voltage level requirement depending on cd clock.\n \n BSpec: 21809\n Cc: Clint Taylor \u003cClinton.A.Taylor@intel.com\u003e\n Cc: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Reviewed-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626014053.30541-3-jose.souza@intel.com\n\ncommit 6e63790efdf75529fda4b50e9561d0832da1cdb5\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Tue Jun 25 18:40:52 2019 -0700\n\n drm/i915/ehl: Remove unsupported cd clocks\n \n EHL do not support 648 and 652.8 MHz.\n \n v2:\n - Limiting maximum CD clock by max_cdclk_freq instead of remove it\n from icl_calc_cdclk()(Ville and Jani)\n \n BSpec: 20598\n Cc: Clint Taylor \u003cClinton.A.Taylor@intel.com\u003e\n Cc: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Reviewed-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626014053.30541-2-jose.souza@intel.com\n\ncommit 9c811fce8a443921beaefef812c2caae8c2df427\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Tue Jun 25 18:40:51 2019 -0700\n\n drm/i915/icl: Add new supported CD clocks\n \n Now 180, 172.8 and 192 MHz are supported.\n \n 180 and 172.8 MHz CD clocks will only be used when audio is not\n enabled as state by BSpec and implemented in\n intel_crtc_compute_min_cdclk(), CD clock must be at least twice of\n Azalia BCLK and BCLK by default is 96 MHz, it could be set to 48 MHz\n but we are not reading it.\n \n v3:\n - making icl clock arrays static (Ville)\n \n BSpec: 20598\n BSpec: 15729\n Cc: Clint Taylor \u003cClinton.A.Taylor@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Reviewed-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626014053.30541-1-jose.souza@intel.com\n\ncommit 092be382a2602067766f190a113514d469162456\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jun 26 16:45:49 2019 +0100\n\n drm/i915: Lift intel_engines_resume() to callers\n \n Since the reset path wants to recover the engines itself, it only wants\n to reinitialise the hardware using i915_gem_init_hw(). Pull the call to\n intel_engines_resume() to the module init/resume path so we can avoid it\n during reset.\n \n Fixes: 79ffac8599c4 (\"drm/i915: Invert the GEM wakeref hierarchy\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Imre Deak \u003cimre.deak@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626154549.10066-3-chris@chris-wilson.co.uk\n\ncommit 18398904ca9e3ddd180e2ecd45886e146b1d9d5b\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jun 26 16:45:48 2019 +0100\n\n drm/i915: Only recover active engines\n \n If we issue a reset to a currently idle engine, leave it idle\n afterwards. This is useful to excise a linkage between reset and the\n shrinker. When waking the engine, we need to pin the default context\n image which we use for overwriting a guilty context -- if the engine is\n idle we do not need this pinned image! However, this pinning means that\n waking the engine acquires the FS_RECLAIM, and so may trigger the\n shrinker. The shrinker itself may need to wait upon the GPU to unbind\n and object and so may require services of reset; ergo we should avoid\n the engine wake up path.\n \n The danger in skipping the recovery for idle engines is that we leave the\n engine with no context defined, which may interfere with the operation of\n the power context on some older platforms. In practice, we should only\n be resetting an active GPU but it something to look out for on Ironlake\n (if memory serves).\n \n Fixes: 79ffac8599c4 (\"drm/i915: Invert the GEM wakeref hierarchy\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Cc: Imre Deak \u003cimre.deak@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626154549.10066-2-chris@chris-wilson.co.uk\n\ncommit de5147b8ce6d51f634661d7c531385371485cec6\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jun 26 16:45:47 2019 +0100\n\n drm/i915: Add a wakeref getter for iff the wakeref is already active\n \n For use in the next patch, we want to acquire a wakeref without having\n to wake the device up -- i.e. only acquire the engine wakeref if the\n engine is already active.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626154549.10066-1-chris@chris-wilson.co.uk\n\ncommit 7d23e59376039d08728444cf05b2db6f7b86ff22\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Jun 19 20:08:42 2019 +0300\n\n drm/i915: Initialize drm_driver vblank funcs at compile time\n \n Move the .get_vblank_timestamp() and .get_scanout_position()\n initialization to happen at compile time. No point in delaying\n it since we always assign the same functions.\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619170842.20579-5-ville.syrjala@linux.intel.com\n\ncommit b318b82455bd9b2899a61108a6d84d4a2d4b6df8\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Thu Jun 20 13:33:34 2019 +0300\n\n drm/i915: Nuke drm_driver irq vfuncs\n \n Stop using the irq vfuncs under drm_driver. That\u0027s not going to fly\n in a mixed gen environment since the structure is shared between all\n the devices.\n \n v2: Allow intel_irq_uninstall() to be called twice due to\n intel_modeset_cleanup() calling it as well. Toss in a\n FIXME to remind us that this is not great.\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190620103334.15651-1-ville.syrjala@linux.intel.com\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n\ncommit 08fa8fd0faa5716ed180e109a022e2c0f42c2e4a\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Jun 19 20:08:40 2019 +0300\n\n drm/i915: Switch to per-crtc vblank vfuncs\n \n Switch from the driver-wide vblank vfuncs to the per-crtc ones so that\n we don\u0027t have so many platform specific vfuncs in the driver struct.\n \n We still need to do something about the rest fo the irq vfuncs...\n \n v2: s/INTEL_GEN\u003e\u003d3/IS_GEN3/\n \n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619170842.20579-3-ville.syrjala@linux.intel.com\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n\ncommit 4c888e7bd26f58deb27c2e6ddc90000b89ee9393\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Wed Jun 19 20:08:39 2019 +0300\n\n drm/i915: Fix various tracepoints for gen2\n \n Gen2 doesn\u0027t have a frame counter and apparently we no longer provide\n a fake .get_vblank_counter() hook for it. That means all tracepoints\n calling that hook will oops. Update the tracepoints to use\n intel_crtc_get_vblank_counter() which will gracefully fall back to\n using the software counter. This is actually a better approach since\n we now get (hopefully accurate) frame numbers in the traces.\n \n This also gets rid of the raw driver-\u003eget_vblank_counter() calls, which\n we need to do in order to switch to the per-crtc vblank vfuncs.\n \n v2: Deal with new tracepoints\n v3: Use a distinct variable name for the internal crtc iterator (Chris)\n \n Cc: Shawn Guo \u003cshawn.guo@linaro.org\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Fixes: 967dd4841787 (\"drm: remove drm_vblank_no_hw_counter assignment from driver code\")\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619170842.20579-2-ville.syrjala@linux.intel.com\n\ncommit faaa2902b5a965610e7b9c5e702dd5f6b8970c60\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jun 26 14:44:33 2019 +0100\n\n drm/i915/selftests: Fixup atomic reset checking\n \n We require that the intel_gpu_reset() was atomic, not the whole of\n i915_reset() which is guarded by a mutex. However, we do require that\n i915_reset_engine() is atomic for use from within the submission tasklet.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626134433.6318-3-chris@chris-wilson.co.uk\n\ncommit 1e5deb2632654c16cd4d2b69cbc24626d9463361\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jun 26 14:44:32 2019 +0100\n\n drm/i915/selftests: Drop manual request wakerefs around hangcheck\n \n We no longer need to manually acquire a wakeref for request emission, so\n drop the redundant wakerefs, letting us test our wakeref handling more\n precisely.\n \n References: 79ffac8599c4 (\"drm/i915: Invert the GEM wakeref hierarchy\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626134433.6318-2-chris@chris-wilson.co.uk\n\ncommit d8474795665462dc7888711ae4b8888a4496eda8\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jun 26 14:44:31 2019 +0100\n\n drm/i915/selftests: Serialise nop reset with retirement\n \n In order for the reset count to be accurate across our selftest, we need\n to prevent the background retire worker from modifying our expected\n state. To preserve the intent of symmetry, we apply this to both\n i915_reset and i915_reset_engine, even though it strictly only affects\n i915_reset_engine currently.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190626134433.6318-1-chris@chris-wilson.co.uk\n\ncommit 5ccf2027bb9362f8c4909e980e3856451a55dcb2\nAuthor: Lee Shawn C \u003cshawn.c.lee@intel.com\u003e\nDate: Thu Jun 20 08:44:16 2019 -0700\n\n drm/i915: Check backlight type while doing eDP backlight initializaiton\n \n If LFP backlight type setting from VBT was \"VESA eDP AUX Interface\".\n Driver should check panel capability and try to initialize aux backlight.\n No matter i915_modparams.enable_dpcd_backlight was enabled or not.\n \n v2: access dev_priv-\u003evbt.backlight.type directly and remove unused function.\n v3: 1. Modify i915.enable_dpcd_backlight type from bool to int and give default\n value as 0 (disable).\n 2. Add a judgement to check LFP backlight type was aux interface or not.\n \n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Cc: Jose Roberto de Souza \u003cjose.souza@intel.com\u003e\n Cc: Cooper Chiou \u003ccooper.chiou@intel.com\u003e\n \n Signed-off-by: Lee Shawn C \u003cshawn.c.lee@intel.com\u003e\n Signed-off-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/1561045456-12171-1-git-send-email-shawn.c.lee@intel.com\n\ncommit 848ed7d54246296f684a4b99546e20d0577aec72\nAuthor: Emil Velikov \u003cemil.velikov@collabora.com\u003e\nDate: Mon May 27 09:17:40 2019 +0100\n\n drm/virtio: drop DRM_AUTH usage from the driver\n \n The authentication can be circumvented, by design, by using the render\n node.\n \n From the driver POV there is no distinction between primary and render\n nodes, thus we can drop the token.\n \n Cc: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Cc: virtualization@lists.linux-foundation.org\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Acked-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190527081741.14235-12-emil.l.velikov@gmail.com\n\ncommit e4eee93d25776da998ec2dfaabe7d2206598d26d\nAuthor: Emil Velikov \u003cemil.velikov@collabora.com\u003e\nDate: Mon May 27 09:17:39 2019 +0100\n\n drm/vgem: drop DRM_AUTH usage from the driver\n \n The authentication can be circumvented, by design, by using the render\n node.\n \n From the driver POV there is no distinction between primary and render\n nodes, thus we can drop the token.\n \n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190527081741.14235-11-emil.l.velikov@gmail.com\n\ncommit 7042a33deb1411523c88e98d8e79ac7399fdee4c\nAuthor: Emil Velikov \u003cemil.velikov@collabora.com\u003e\nDate: Mon May 27 09:17:37 2019 +0100\n\n drm/omap: drop DRM_AUTH from DRM_RENDER_ALLOW ioctls\n \n The authentication can be circumvented, by design, by using the render\n node.\n \n From the driver POV there is no distinction between primary and render\n nodes, thus we can drop the token.\n \n Note: the outstanding DRM_AUTH instance is:\n - (badly coped) legacy DRI1 ioctl, which is a noop\n \n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Acked-by: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190527081741.14235-9-emil.l.velikov@gmail.com\n\ncommit a305f6c5c8194fbe53e2f1c101476373d04538d0\nAuthor: Emil Velikov \u003cemil.velikov@collabora.com\u003e\nDate: Mon May 27 09:17:36 2019 +0100\n\n drm/nouveau: drop DRM_AUTH from DRM_RENDER_ALLOW ioctls\n \n The authentication can be circumvented, by design, by using the render\n node.\n \n From the driver POV there is no distinction between primary and render\n nodes, thus we can drop the token.\n \n Note: the outstanding DRM_AUTH instance is:\n - legacy DRI1 ioctl, which is already neutered\n \n Cc: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n Cc: nouveau@lists.freedesktop.org\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190527081741.14235-8-emil.l.velikov@gmail.com\n\ncommit 88209d2c5035737f96bcfc2fd73c0fd8d80e9bf1\nAuthor: Emil Velikov \u003cemil.velikov@collabora.com\u003e\nDate: Mon May 27 09:17:35 2019 +0100\n\n drm/msm: drop DRM_AUTH usage from the driver\n \n The authentication can be circumvented, by design, by using the render\n node.\n \n From the driver POV there is no distinction between primary and render\n nodes, thus we can drop the token.\n \n Cc: Rob Clark \u003crobdclark@gmail.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: freedreno@lists.freedesktop.org\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190527081741.14235-7-emil.l.velikov@gmail.com\n\ncommit 921d57323657eb5ae4a2ec460b137e62ee27c5c7\nAuthor: Emil Velikov \u003cemil.velikov@collabora.com\u003e\nDate: Mon May 27 09:17:34 2019 +0100\n\n drm/lima: drop DRM_AUTH usage from the driver\n \n The authentication can be circumvented, by design, by using the render\n node.\n \n From the driver POV there is no distinction between primary and render\n nodes, thus we can drop the token.\n \n Cc: Qiang Yu \u003cyuq825@gmail.com\u003e\n Cc: lima@lists.freedesktop.org\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Qiang Yu \u003cyuq825@gmail.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190527081741.14235-6-emil.l.velikov@gmail.com\n\ncommit 951ecc022c47bc8d0bc814de5eac1a3612393606\nAuthor: Emil Velikov \u003cemil.velikov@collabora.com\u003e\nDate: Mon May 27 09:17:32 2019 +0100\n\n drm/exynos: drop DRM_AUTH from DRM_RENDER_ALLOW ioctls\n \n The authentication can be circumvented, by design, by using the render\n node.\n \n From the driver POV there is no distinction between primary and render\n nodes, thus we can drop the token.\n \n Cc: Inki Dae \u003cinki.dae@samsung.com\u003e\n Cc: Joonyoung Shim \u003cjy0922.shim@samsung.com\u003e\n Cc: Seung-Woo Kim \u003csw0312.kim@samsung.com\u003e\n Cc: Kyungmin Park \u003ckyungmin.park@samsung.com\u003e\n Cc: Tobias Jakobi \u003ctjakobi@math.uni-bielefeld.de\u003e\n Signed-off-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Acked-by: Inki Dae \u003cinki.dae@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190527081741.14235-4-emil.l.velikov@gmail.com\n\ncommit b8602f9a65a6744060cde6fb153729e6a4099437\nAuthor: Emil Velikov \u003cemil.velikov@collabora.com\u003e\nDate: Mon May 27 09:17:31 2019 +0100\n\n drm/etnaviv: drop DRM_AUTH usage from the driver\n \n The authentication can be circumvented, by design, by using the render\n node.\n \n From the driver POV there is no distinction between primary and render\n nodes, thus we can drop the token.\n \n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Christian Gmeiner \u003cchristian.gmeiner@gmail.com\u003e\n Cc: etnaviv@lists.freedesktop.org\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Christian Gmeiner \u003cchristian.gmeiner@gmail.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190527081741.14235-3-emil.l.velikov@gmail.com\n\ncommit cbfbe47fc5391852bd426e07aad7f5cf026e94c5\nAuthor: Emil Velikov \u003cemil.velikov@collabora.com\u003e\nDate: Wed May 22 17:41:17 2019 +0100\n\n drm/vmwgfx: use core drm to extend/check vmw_execbuf_ioctl\n \n Currently vmw_execbuf_ioctl() open-codes the permission checking, size\n extending and copying that is already done in core drm.\n \n Kill all the duplication, adding a few comments for clarity.\n \n Cc: VMware Graphics \u003clinux-graphics-maintainer@vmware.com\u003e\n Cc: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Tested-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Reviewed-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190522164119.24139-3-emil.l.velikov@gmail.com\n\ncommit bcde7d345c9038a88b790aa1fd27d4430ce54876\nAuthor: Emil Velikov \u003cemil.velikov@collabora.com\u003e\nDate: Wed May 22 17:41:16 2019 +0100\n\n drm/vmgfx: kill off unused init_mutex\n \n According to the docs - prevents firstopen/lastclose races. Yet never\n used in practise.\n \n Cc: VMware Graphics \u003clinux-graphics-maintainer@vmware.com\u003e\n Cc: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190522164119.24139-2-emil.l.velikov@gmail.com\n\ncommit 3cd74023ea185fee2de49d20d6881252a89695fa\nAuthor: Emil Velikov \u003cemil.velikov@collabora.com\u003e\nDate: Wed May 22 17:41:15 2019 +0100\n\n vmwgfx: drop empty lastclose stub\n \n Core DRM is safe when the callback is NULL.\n \n Cc: VMware Graphics \u003clinux-graphics-maintainer@vmware.com\u003e\n Cc: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Thomas Hellstrom \u003cthellstrom@vmware.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190522164119.24139-1-emil.l.velikov@gmail.com\n\ncommit ccdae42575695ab442941310bd67c7ed1714e273\nAuthor: Emil Velikov \u003cemil.velikov@collabora.com\u003e\nDate: Wed May 22 16:02:19 2019 +0100\n\n drm/nouveau: remove open-coded drm_invalid_op()\n \n Cc: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n Cc: nouveau@lists.freedesktop.org\n Signed-off-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190522150219.13913-2-emil.l.velikov@gmail.com\n\ncommit b402375699ae84880bf2af56e455239b8807c3a6\nAuthor: Emil Velikov \u003cemil.velikov@collabora.com\u003e\nDate: Wed May 22 16:47:01 2019 +0100\n\n drm/i915: remove irrelevant DRM_UNLOCKED flag\n \n DRM_UNLOCKED doesn\u0027t do anything for non-legacy drivers. Remove it.\n \n Cc: intel-gfx@lists.freedesktop.org\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190522154702.16269-3-emil.l.velikov@gmail.com\n\ncommit d6891db2ad0a51f64aa535e3f1932f7ca66a45bf\nAuthor: Emil Velikov \u003cemil.velikov@collabora.com\u003e\nDate: Wed May 22 16:46:59 2019 +0100\n\n drm/tegra: remove irrelevant DRM_UNLOCKED flag\n \n DRM_UNLOCKED doesn\u0027t do anything for non-legacy drivers. Remove it.\n \n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n Cc: linux-tegra@vger.kernel.org\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190522154702.16269-1-emil.l.velikov@gmail.com\n\ncommit caa54ec03c7fb6ec4382f7c7e54ab27d977be510\nAuthor: Jerry Han \u003cjerry.han.hq@gmail.com\u003e\nDate: Thu Apr 25 11:18:42 2019 +0800\n\n dt-bindings: panel: Add Boe Himax8279d is 1200x1920, 4-lane MIPI-DSI LCD panel\n \n The Boe Himax8279d is a 8.0\" panel with a 1200x1920 resolution and\n connected to DSI using four lanes.\n \n V8:\n - Modify communication address\n \n V7:\n - Add the information of the reviewer\n \n V6:\n - Add the information of the reviewer\n \n V5:\n - Added changelog\n \n V4:\n - None\n \n V3:\n - None\n \n V2:\n - Add compatible device \"boe,himax8279d10p\" (Sam)\n - Add the necessary property descriptions (Sam)\n \n Signed-off-by: Jerry Han \u003cjerry.han.hq@gmail.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Derek Basehore \u003cdbasehore@chromium.org\u003e\n Cc: Jitao Shi \u003cjitao.shi@mediatek.com\u003e\n Cc: Rock wang \u003crock_wang@himax.com.cn\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190425031842.17202-1-jerry.han.hq@gmail.com\n\ncommit 77b893df0c587dc0958f0924f713c2b323b65a5e\nAuthor: Marek Vasut \u003cmarex@denx.de\u003e\nDate: Fri May 3 17:47:49 2019 +0200\n\n dt-bindings: display: Add ETM0700G0DH6 compatible string\n \n The ETM0700G0DH6 is currently documented as using edt,etm070080dh6\n compatible string, however the Linux kernel driver as well as a\n couple of DTs use edt,etm0700g0dh6 compatible string. Add it into\n the documentation.\n \n Signed-off-by: Marek Vasut \u003cmarex@denx.de\u003e\n Cc: Rob Herring \u003crobh+dt@kernel.org\u003e\n Cc: Jan Tuerk \u003cjan.tuerk@emtrion.com\u003e\n Cc: Thierry Reding \u003ctreding@nvidia.com\u003e\n Cc: devicetree@vger.kernel.org\n To: dri-devel@lists.freedesktop.org\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190503154749.5630-1-marex@denx.de\n\ncommit 1a14e0c256521ea9e74487000e265662fa82f692\nAuthor: Guido Günther \u003cagx@sigxcpu.org\u003e\nDate: Wed Jun 26 12:37:51 2019 +0200\n\n drm/panel: jh057n00900: Add regulator support\n \n Allow to specify regulators for vcc and iovcc. According to the data\n sheet the panel wants vcc (2.8V) and iovcc (1.8V) and there\u0027s no startup\n dependency between the two.\n \n Signed-off-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/f78611fb26329e50ec1533810fbb76562f2f4e48.1561542477.git.agx@sigxcpu.org\n\ncommit d4bd9a58d81beeb2fab89085b97ba10ed10f09bc\nAuthor: Guido Günther \u003cagx@sigxcpu.org\u003e\nDate: Wed Jun 26 12:37:50 2019 +0200\n\n dt-bindings: display/panel: jh057n00900: Document power supply properties\n \n Document the vcc-supply and iovcc-supply properties of the Rocktech\n jh057n00900 panel.\n \n Signed-off-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/ec5ee11786036dc8826a3f8555cbb1dce4dbca25.1561542477.git.agx@sigxcpu.org\n\ncommit 428733c72d03d7fca9e66bf7d44e9ac5eb4e12ea\nAuthor: Guido Günther \u003cagx@sigxcpu.org\u003e\nDate: Wed Jun 26 12:37:49 2019 +0200\n\n drm/panel: jh057n00900: Don\u0027t use magic constant\n \n 0xBF isn\u0027t in any ST7703 data sheet so mark it as unknown. This avoids\n confusion on whether there is a missing command in that\n dsi_generic_write_seq() call.\n \n Signed-off-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/7f50fa6de2ae029111f158f8ea7fd69a0903eb97.1561542477.git.agx@sigxcpu.org\n\ncommit 60ba34c6a5b399fb398e61cd6daa7753426165c4\nAuthor: Guido Günther \u003cagx@sigxcpu.org\u003e\nDate: Wed Jun 26 12:37:48 2019 +0200\n\n MAINTAINERS: Add Purism mail alias as reviewer for their devkit\u0027s panel\n \n Add a mail alias as reviewer for the rocktech jh057n00900 panel driver.\n \n Signed-off-by: Guido Günther \u003cagx@sigxcpu.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/a392758c914d436ec5e449980f619bc906c3054a.1561542477.git.agx@sigxcpu.org\n\ncommit f63df31d2dddf418a3c0bacaa292a38faaccdcd9\nAuthor: Robert Chiras \u003crobert.chiras@nxp.com\u003e\nDate: Wed Jun 26 15:08:27 2019 +0300\n\n drm/panel: Add support for Raydium RM67191 panel driver\n \n This patch adds Raydium RM67191 TFT LCD panel driver (MIPI-DSI\n protocol).\n \n Signed-off-by: Robert Chiras \u003crobert.chiras@nxp.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/1561550907-9733-3-git-send-email-robert.chiras@nxp.com\n\ncommit 059b39f30aa6c3c5cc63971751f3b2d47643a812\nAuthor: Robert Chiras \u003crobert.chiras@nxp.com\u003e\nDate: Wed Jun 26 15:08:26 2019 +0300\n\n dt-bindings: display: panel: Add support for Raydium RM67191 panel\n \n Add dt-bindings documentation for Raydium RM67191 DSI panel.\n \n Signed-off-by: Robert Chiras \u003crobert.chiras@nxp.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/1561550907-9733-2-git-send-email-robert.chiras@nxp.com\n\ncommit f1bd37f300a4ed1e0d9766d35d2a1360ea294591\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Mon Jun 3 17:31:20 2019 +0200\n\n drm/panel: simple: Add Sharp LS020B1DD01D panel support\n \n The Sharp LS020B1DD01D is a simple 2.0\" 240x160 16-bit TFT panel.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190603153120.23947-3-paul@crapouillou.net\n\ncommit f9a763d2746299cd4fd00067bd3df8e7e9899019\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Mon Jun 3 17:31:19 2019 +0200\n\n drm: Add bus flag for Sharp-specific signals\n \n Add the DRM_BUS_FLAG_SHARP_SIGNALS to the drm_bus_flags enum.\n \n This flags can be used when the display must be driven with the\n Sharp-specific signals SPL, CLS, REV, PS.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190603153120.23947-2-paul@crapouillou.net\n\ncommit a63f4612e1de3888edf0bdcccf6bb198ab30d39d\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Mon Jun 3 17:31:18 2019 +0200\n\n dt-bindings: display: Add Sharp LS020B1DD01D panel documentation\n \n The LS020B1DD01D is a 2.0\" 240x160 16-bit TFT LCD panel.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190603153120.23947-1-paul@crapouillou.net\n\ncommit ecdcbbb7ceb1188474f603fac0ec48171761c842\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Mon Jun 3 17:25:55 2019 +0200\n\n drm/panel: Add Novatek NT39016 panel support\n \n Add support for display panels built around the Novatek NT39016 display\n controller, as found on e.g. the King Display KD035G6-54NT 24-bit\n 320x240 3.5\" LCD panel which equips the GCW Zero open-source handheld\n gaming console.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Signed-off-by: Maarten ter Huurne \u003cmaarten@treewalker.org\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190603152555.23527-2-paul@crapouillou.net\n\ncommit c79e7a293733d6e9d589ba929ae0a87f6a333afd\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Mon Jun 3 17:25:54 2019 +0200\n\n dt-bindings: display: Add King Display KD035G6-54NT panel documentation\n \n The KD035G6-54NT is a 3.5\" 320x240 24-bit TFT LCD panel.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190603152555.23527-1-paul@crapouillou.net\n\ncommit 93d7a3b4be30c705c39377b9e75b24a184f6c4fe\nAuthor: Mika Kahola \u003cmika.kahola@intel.com\u003e\nDate: Mon Jun 17 11:24:13 2019 +0300\n\n drm/i915/icl: Add missing device ID\n \n We are missing PCI device ID for SKU ICLLP U GT 1.5F (0x8A54) as per BSPec.\n \n BSpec: 19092\n \n Signed-off-by: Mika Kahola \u003cmika.kahola@intel.com\u003e\n Reviewed-by: Clint Taylor \u003cClinton.A.Taylor@intel.com\u003e\n Signed-off-by: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190617082413.22549-1-mika.kahola@intel.com\n\ncommit 719bda751e183477e8fd21cb9b3942fd579cf5b3\nAuthor: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\nDate: Mon Jun 24 11:08:51 2019 +0200\n\n MAINTAINERS: Update Maintainers and Reviewers of DRM Bridge Drivers\n \n Add myself as co-maintainer of DRM Bridge Drivers then add Jonas Karlman\n and Jernej Škrabec as Reviewers of DRM Bridge Drivers.\n \n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: Jonas Karlman \u003cjonas@kwiboo.se\u003e\n Cc: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Cc: Jernej Škrabec \u003cjernej.skrabec@siol.net\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Signed-off-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Reviewed-by: Jernej Skrabec \u003cjernej.skrabec@siol.net\u003e\n Reviewed-by: Jonas Karlman \u003cjonas@kwiboo.se\u003e\n Acked-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190624090851.17859-1-narmstrong@baylibre.com\n\ncommit cba17e5d647b4dabb3c9d8bc1bf2636f1a3a9687\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jun 25 20:48:59 2019 +0100\n\n drm/i915/gt: Add some debug tracing for context pinning\n \n Add the context pin/unpin events to the trace for post-mortem debugging.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190625194859.28005-1-chris@chris-wilson.co.uk\n\ncommit 9c31dcb6dd94f4e7a6bb803aa935c9801e718d62\nAuthor: H. Nikolaus Schaller \u003chns@goldelico.com\u003e\nDate: Fri Jun 7 13:11:08 2019 +0200\n\n drm/panel: simple: Add Ortustech COM37H3M panel support\n \n The change adds support for the Ortustech COM37H3M05DTC/99DTC 3.7\" TFT LCD panel.\n \n Tested on Letux3704.\n \n Signed-off-by: H. Nikolaus Schaller \u003chns@goldelico.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/43b47034b618cff26cea0484591c6deafb7f0685.1559905870.git.hns@goldelico.com\n\ncommit dda0e4bdbeac54d5e383bd14f436d197ee9985b0\nAuthor: H. Nikolaus Schaller \u003chns@goldelico.com\u003e\nDate: Fri Jun 7 13:11:07 2019 +0200\n\n drm/panel: simple: Add Sharp LQ070Y3DG3B panel support\n \n The change adds support for the Sharp LQ070Y3DG3B 7.0\" TFT LCD panel.\n \n Tested on Letux7004.\n \n Signed-off-by: H. Nikolaus Schaller \u003chns@goldelico.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/d16aaa1ac93e4f15c13cd7d621de95836257676a.1559905870.git.hns@goldelico.com\n\ncommit a6e5a47f40edbe439fa103f44e40e37d4271934e\nAuthor: H. Nikolaus Schaller \u003chns@goldelico.com\u003e\nDate: Fri Jun 7 13:11:11 2019 +0200\n\n dt-bindings: drm/panel: simple: add sharp, lq070y3dg3b panel\n \n Signed-off-by: H. Nikolaus Schaller \u003chns@goldelico.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/ee90ed56d2c294ce8ac3b44bf2229c5ab9f85e91.1559905870.git.hns@goldelico.com\n\ncommit e24fd0cf73d183c616e2ac37b99082854ece9e88\nAuthor: H. Nikolaus Schaller \u003chns@goldelico.com\u003e\nDate: Fri Jun 7 13:11:10 2019 +0200\n\n dt-bindings: drm/panel: simple: add ortustech, com37h3m99dtc panel\n \n Signed-off-by: H. Nikolaus Schaller \u003chns@goldelico.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/3e0720fbddfd7b35ad8551440544411485d8ad9f.1559905870.git.hns@goldelico.com\n\ncommit 75909cb6a7537142576ef5160f7117f9bfc88d25\nAuthor: H. Nikolaus Schaller \u003chns@goldelico.com\u003e\nDate: Fri Jun 7 13:11:09 2019 +0200\n\n dt-bindings: drm/panel: simple: add ortustech, com37h3m05dtc panel\n \n Signed-off-by: H. Nikolaus Schaller \u003chns@goldelico.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/a650cb8df1bdf58ec3c8a182532692db16b77f70.1559905870.git.hns@goldelico.com\n\ncommit f0ca820cc0067ce4debd9893650e5c188c83b941\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jun 26 00:33:49 2019 +0100\n\n drm/i915/gt: Always call kref_init for the timeline\n \n Always initialise the refcount, even for the embedded timelines inside\n mock devices.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190625233349.32371-2-chris@chris-wilson.co.uk\n\ncommit b38565faded7da1f84b2b9a9a6d41ea1ebc48936\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jun 26 00:33:48 2019 +0100\n\n drm/i915/gt: Drop stale commentary for timeline density\n \n We no longer allocate a contiguous set of timeline ids for all engines\n upon creation, so we no longer should assume that the timelines are\n densely allocated within a context. Hopefully, the set of fences used\n within a workload are still dense enough for us to take advantage of\n the compressed radix tree used for the syncmap.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190625233349.32371-1-chris@chris-wilson.co.uk\n\ncommit c8d84778e52733cbbc05c1f3ea77635feb099642\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jun 25 14:01:22 2019 +0100\n\n drm/i915/selftests: Hold ref on request across waits\n \n As we wait upon the request, we should be sure to hold our own reference\n for our checks.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190625130128.11009-14-chris@chris-wilson.co.uk\n\ncommit 2c6574a9ed666d6e00b0deac66be4b793097b0b7\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Thu Jun 6 00:22:47 2019 +0200\n\n drm/panel: simple: Add GiantPlus GPM940B0 panel support\n \n The GiantPlus GPM940B0 is a simple 3.0\" 320x240 24-bit TFT panel.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Tested-by: Artur Rojek \u003ccontact@artur-rojek.eu\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190605222247.25657-3-paul@crapouillou.net\n\ncommit 0565ae92835cdde613149b32809a8d060f2815ff\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Thu Jun 6 00:22:46 2019 +0200\n\n media: uapi: Add MEDIA_BUS_FMT_RGB888_3X8 media bus format\n \n This patch adds MEDIA_BUS_FMT_RGB888_3X8, used for the GiantPlus\n GPM940B0 24-bit TFT panel, where the RGB components are transferred\n sequentially on a 8-bit bus.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Acked-by: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\n Acked-by: Sakari Ailus \u003csakari.ailus@linux.intel.com\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190605222247.25657-2-paul@crapouillou.net\n\ncommit d55bcb4c9a74a4e19a84b9a31cb20b44e607bff5\nAuthor: Paul Cercueil \u003cpaul@crapouillou.net\u003e\nDate: Thu Jun 6 00:22:45 2019 +0200\n\n dt-bindings: display: Add GiantPlus GPM940B0 panel documentation\n \n The GPM940B0 is a 3.0\" 320x240 24-bit TFT LCD panel.\n \n Signed-off-by: Paul Cercueil \u003cpaul@crapouillou.net\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190605222247.25657-1-paul@crapouillou.net\n\ncommit 5f22e5b3116ce42f0fdd38d645b001cddf5336d7\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jun 25 14:01:14 2019 +0100\n\n drm/i915: Rename intel_wakeref_[is]_active\n \n Our general rule is to use is/has as the verb for boolean functions,\n rename intel_wakeref_active to intel_wakeref_is_active so the question\n being asked is clear.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190625130128.11009-6-chris@chris-wilson.co.uk\n\ncommit 0c91621cad492e362c37330e1a0985bcdda00fda\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jun 25 14:01:10 2019 +0100\n\n drm/i915/gt: Pass intel_gt to pm routines\n \n Switch from passing the i915 container to newly named struct intel_gt.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190625130128.11009-2-chris@chris-wilson.co.uk\n\ncommit 8e9c2f621893ae2cc57c34fa96ece011eebd94b7\nAuthor: Robert M. Fosha \u003crobert.m.fosha@intel.com\u003e\nDate: Tue Jun 25 09:41:07 2019 -0700\n\n drm/i915/guc: Add debug capture of GuC exception\n \n Detect GuC firmware load failure due to an exception during execution\n in GuC firmware. Output the GuC EIP where exception occurred to dmesg\n for GuC debug information.\n \n v2: correct typos, change debug message and error code returned for\n GuC exception (Michal)\n \n Signed-off-by: Robert M. Fosha \u003crobert.m.fosha@intel.com\u003e\n Cc: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190625164107.21512-1-robert.m.fosha@intel.com\n\ncommit 07bfe6bf1052f074093cdea95d6041f48b994c4b\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jun 25 14:01:09 2019 +0100\n\n drm/i915/execlists: Convert recursive defer_request() into iterative\n \n As this engine owns the lock around rq-\u003esched.link (for those waiters\n submitted to this engine), we can use that link as an element in a local\n list. We can thus replace the recursive algorithm with an iterative walk\n over the ordered list of waiters.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190625130128.11009-1-chris@chris-wilson.co.uk\n\ncommit 9a6a644096a1066c2ef5c47db0db79ac72af7967\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Fri Jun 14 14:37:49 2019 -0700\n\n drm/i915/ehl: Add missing VECS engine\n \n EHL can have up to one VECS(video enhancement) engine, so add it to\n the device_info.\n \n BSpec: 29152\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Bob Paauwe \u003cbob.j.paauwe@intel.com\u003e\n Cc: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Cc: Clint Taylor \u003cClinton.A.Taylor@intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Reviewed-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614213749.15870-1-jose.souza@intel.com\n\ncommit 5ebeb02c8678cb4b242e4f11ba15839445621c54\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:36:01 2019 +0200\n\n drm/msm: Drop robj from msm_gem_new_impl\n \n Only user was the prime import, and drm_prime.c takes care of that\n now.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Acked-by: Rob Clark \u003crobdclark@gmail.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Rob Clark \u003crobdclark@gmail.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: linux-arm-msm@vger.kernel.org\n Cc: freedreno@lists.freedesktop.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-46-daniel.vetter@ffwll.ch\n\ncommit c6be8086182419f266013d505c089f1db377ddbd\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:58 2019 +0200\n\n drm/etnaviv: Drop resv argument from etnaviv_gem_new_impl\n \n Only user was the prime import, and drm_prime.c takes care of that\n now.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Russell King \u003clinux+etnaviv@armlinux.org.uk\u003e\n Cc: Christian Gmeiner \u003cchristian.gmeiner@gmail.com\u003e\n Cc: etnaviv@lists.freedesktop.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-43-daniel.vetter@ffwll.ch\n\ncommit fcca453c8884e119056eb19c48cca38ae43c5e93\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:36:14 2019 +0200\n\n drm/todo: Add new debugfs todo\n \n Greg is busy already, but maybe he won\u0027t do everything ...\n \n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Acked-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-59-daniel.vetter@ffwll.ch\n\ncommit 8de6ca2e0e9318d9d784a45112712beffb87727b\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:36:13 2019 +0200\n\n drm/todo: Update mmap todo\n \n Thanks to Noralf some good progress already.\n \n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-58-daniel.vetter@ffwll.ch\n\ncommit 3c745e0b5dea0397cb034486e67ef9ddfade617d\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:36:12 2019 +0200\n\n drm/todo: Update backlight todo\n \n Basic helpers have been extracted, now there\u0027s a pile more todo still\n across the entire tree.\n \n Reviewed-by: Daniel Thompson \u003cdaniel.thompson@linaro.org\u003e\n Acked-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-57-daniel.vetter@ffwll.ch\n\ncommit 9dfed7013083e9af2225f9fbd70ef3457aa64b03\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:36:11 2019 +0200\n\n drm/todo: remove gem_prime_import/export todo\n \n I\u0027ve done that.\n \n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-56-daniel.vetter@ffwll.ch\n\ncommit 3886666a2568007f0fa46a5c01eb178def22b3b5\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:36:03 2019 +0200\n\n drm/vc4: Don set gem_obj-\u003eresv in prime import anymore\n \n This is done in drm_prime.c now.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-48-daniel.vetter@ffwll.ch\n\ncommit 0ecaea14843e7b3d6aa53e86141d5e43136e71ca\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:36:02 2019 +0200\n\n drm/panfrost: don\u0027t set gem_obj-\u003eresv for prime import anymore\n \n This is now done in drm_prime.c\n \n Acked-by: Rob Herring \u003crobh@kernel.org\u003e\n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Rob Herring \u003crobh@kernel.org\u003e\n Cc: Tomeu Vizoso \u003ctomeu.vizoso@collabora.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-47-daniel.vetter@ffwll.ch\n\ncommit 2a11d881255ba71ed13602a979250bdad7465889\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:36:00 2019 +0200\n\n drm/mediatek: Use drm_atomic_helper_wait_for_fences\n \n If we use the gem fb helper as the prepare_fb hook, plus the\n drm_prime.c import helpers now automatically setting obj-\u003eresv, we can\n use the shared helpers to wait for fences instead of rolling our own.\n Note that this relies on mtk setting drm_fb-\u003eobj, which is already\n done in mtk_drm_framebuffer_init().\n \n Aside: Probably can use the default commit_tail with this again, but I\n didn\u0027t check for that.\n \n Reviewed-by: CK Hu \u003cck.hu@mediatek.com\u003e\n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: CK Hu \u003cck.hu@mediatek.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: linux-mediatek@lists.infradead.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-45-daniel.vetter@ffwll.ch\n\ncommit 0fbbd27e5ab18a0d2bd91e63ec9d16349d42d5bf\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:59 2019 +0200\n\n drm/lima: Drop resv argument from lima_bo_create_struct\n \n It was only used for prime import, which is now handled by\n drm_prime.c.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Qiang Yu \u003cyuq825@gmail.com\u003e\n Cc: lima@lists.freedesktop.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-44-daniel.vetter@ffwll.ch\n\ncommit fc25441c7b9d81dc18cce0a0fa5527370e386652\nAuthor: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\nDate: Fri Jun 21 08:18:47 2019 -0700\n\n drm/i915/ehl: Add one additional PCH ID to MCC\n \n There\u0027s one additional ID that we should treat as Mule Creek Canyon.\n \n Cc: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Signed-off-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621151847.31302-1-matthew.d.roper@intel.com\n Reviewed-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n\ncommit 8dcfdfb4501012a8d36d2157dc73925715f2befb\nAuthor: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\nDate: Mon Jun 10 11:19:14 2019 +0300\n\n drm/i915/perf: fix ICL perf register offsets\n \n We got the wrong offsets (could they have changed?). New values were\n computed off an error state by looking up the register offset in the\n context image as written by the HW.\n \n Signed-off-by: Lionel Landwerlin \u003clionel.g.landwerlin@intel.com\u003e\n Fixes: 1de401c08fa805 (\"drm/i915/perf: enable perf support on ICL\")\n Acked-by: Kenneth Graunke \u003ckenneth@whitecape.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190610081914.25428-1-lionel.g.landwerlin@intel.com\n\ncommit f9a393875d3af13cc3267477746608dadb7f17c1\nAuthor: Kenneth Graunke \u003ckenneth@whitecape.org\u003e\nDate: Tue Jun 25 10:06:55 2019 +0100\n\n drm/i915: Disable SAMPLER_STATE prefetching on all Gen11 steppings.\n \n The Demand Prefetch workaround (binding table prefetching) only applies\n to Icelake A0/B0. But the Sampler Prefetch workaround needs to be\n applied to all Gen11 steppings, according to a programming note in the\n SARCHKMD documentation.\n \n Using the Intel Gallium driver, I have seen intermittent failures in\n the dEQP-GLES31.functional.copy_image.non_compressed.* tests. After\n applying this workaround, the tests reliably pass.\n \n v2: Remove the overlap with a pre-production w/a\n \n BSpec: 9663\n Signed-off-by: Kenneth Graunke \u003ckenneth@whitecape.org\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: stable@vger.kernel.org\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190625090655.19220-1-chris@chris-wilson.co.uk\n\ncommit 975304100a700d24d1c09b288060a66b1d55c360\nAuthor: Rob Herring \u003crobh@kernel.org\u003e\nDate: Mon Jun 24 15:56:44 2019 -0600\n\n dt-bindings: display: Convert tpo,tpg110 panel to DT schema\n \n Convert the tpo,tpg110 panel binding to DT schema.\n \n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: dri-devel@lists.freedesktop.org\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190624215649.8939-11-robh@kernel.org\n\ncommit ca5be902a87ddccc88144f2dea21b5f0814391ef\nAuthor: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nDate: Tue Jun 18 13:52:45 2019 +0200\n\n drm/mcde: Fix uninitialized variable\n \n We need to handle the case when of_drm_find_bridge() returns\n NULL.\n \n Reported-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Acked-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Signed-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190618115245.13915-1-linus.walleij@linaro.org\n\ncommit 871918dffefc594e765cc7e885a36a7fd3f38da7\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Jun 24 15:16:30 2019 +0100\n\n drm/i915/gem: Clear read/write domains for GPU clear\n \n Update the domains for the write via the GPU so that we do not\n shortcircuit any set-domain clflush afterwards.\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d110978\n Fixes: b2dbf8d982a4 (\"drm/i915/blt: Remove recursive vma-\u003elock\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190624141630.11015-1-chris@chris-wilson.co.uk\n\ncommit 1e612a0f62febce70da35891f1b9a3cc270d4af1\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Tue Mar 26 19:38:10 2019 +0200\n\n drm/sun4i: Eliminate pointless on stack copy of drm_display_info\n \n Just use a pointer to the display_info rather than make a copy\n on stack.\n \n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190326173810.11979-1-ville.syrjala@linux.intel.com\n Acked-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n\ncommit 9f396ab4de152408c7007fce255a7144ae7d70a0\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Thu Jun 20 17:26:37 2019 +0300\n\n drm: WARN on illegal aspect ratio when converting a mode to umode\n \n WARN if the incoming drm_display_mode has an illegal aspect ratio\n when converting it to a user mode. This should never happen unless\n the driver made a mistake and put an invalid value into the aspect\n ratio.\n \n Cc: Ilia Mirkin \u003cimirkin@alum.mit.edu\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190620142639.17518-4-ville.syrjala@linux.intel.com\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 993a80671ed9a149d5136965d04d0a71f7b21c48\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Thu Jun 20 17:26:36 2019 +0300\n\n drm: Do not accept garbage mode aspect ratio flags\n \n Don\u0027t let userspace feed us any old garbage in the mode aspect ratio\n flags.\n \n Cc: Ilia Mirkin \u003cimirkin@alum.mit.edu\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190620142639.17518-3-ville.syrjala@linux.intel.com\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 700710e977fdaae5dcfc9a01d56eaaf0720b05f5\nAuthor: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\nDate: Thu Jun 20 17:26:35 2019 +0300\n\n drm: Do not use bitwise OR to set picure_aspect_ratio\n \n enum hdmi_picture_aspect is not a bitmask, so don\u0027t use bitwise OR\n to populate it.\n \n Cc: Ilia Mirkin \u003cimirkin@alum.mit.edu\u003e\n Signed-off-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190620142639.17518-2-ville.syrjala@linux.intel.com\n Reviewed-by: Alex Deucher \u003calexander.deucher@amd.com\u003e\n\ncommit 50b0946de4c6bccc9af78505f8b0cb33e87941ce\nAuthor: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\nDate: Mon Jun 24 10:40:16 2019 +0200\n\n drm/connector: Fix warning in debug message\n \n The commit 3aeeb13d8996 (\"drm/modes: Support modes names on the command\n line\") added name support to the DRM modes, and added that name to the\n debug message.\n \n However, that code tests for whether or not the name variable is NULL and\n only prints it if it\u0027s not. Except that that variable is an array, so it\n will never be NULL.\n \n The original intent was to print it only when the name has been specified.\n Just printing the array directly will achieve the same thing since the\n drm_cmdline_mode structure that holds it is itself contained in\n drm_connector, that is allocated with its whole content zero\u0027d.\n \n That means that if the name is not declared, the array will be all zeros,\n which will not print anything.\n \n Cc: Nick Desaulniers \u003cndesaulniers@google.com\u003e\n Reported-by: kbuild test robot \u003clkp@intel.com\u003e\n Reviewed-by: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Fixes: 3aeeb13d8996 (\"drm/modes: Support modes names on the command line\")\n Signed-off-by: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190624084016.12937-1-maxime.ripard@bootlin.com\n\ncommit 1effe5a3bb3ddf172649e06f7a5607317e077930\nAuthor: Matt Redfearn \u003cmatt.redfearn@thinci.com\u003e\nDate: Tue Apr 30 08:17:04 2019 +0000\n\n drm/bridge/synopsys: dsi: Allow VPG to be enabled via debugfs\n \n The Synopsys MIPI DSI IP contains a video test pattern generator which\n is helpful in debugging video timing with connected displays.\n Add a debugfs directory containing files which allow the VPG to be\n enabled and disabled, and its orientation to be changed.\n \n Signed-off-by: Matt Redfearn \u003cmatt.redfearn@thinci.com\u003e\n Tested-by: Yannick Fertré \u003cyannick.fertre@st.com\u003e\n Reviewed-by: Philippe Cornu \u003cphilippe.cornu@st.com\u003e\n Signed-off-by: Andrzej Hajda \u003ca.hajda@samsung.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190430081646.23845-1-matt.redfearn@thinci.com\n\ncommit 8db7933ee3d442627c716ee49a70fa3aef18640d\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Mon Jun 24 10:20:09 2019 +0100\n\n drm/i915/execlists: Always clear ring_pause if we do not submit\n \n In the unlikely case (thank you CI!), we may find ourselves wanting to\n issue a preemption but having no runnable requests left. In this case,\n we set the semaphore before computing the preemption and so must unset\n it before forgetting (or else we leave the machine busywaiting until the\n next request comes along and so likely hang).\n \n v2: Replace readback with only a wmb after asserting the semaphore\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190624092009.30189-1-chris@chris-wilson.co.uk\n\ncommit 0e343b086b05abcb9820c90b057a2f237ce8bc1c\nAuthor: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\nDate: Fri Jun 21 18:21:17 2019 +0200\n\n drm/lima: Reduce the amount of logs on deferred probe of clocks and reset controller\n \n There is no point to print deferred probe messages as errors. Adjust\n the printks for error paths of obtaining clocks and reset controller.\n This removes the error message of lima_clk_init() call in favor or\n specific failure messages inside.\n \n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Signed-off-by: Qiang Yu \u003cyuq825@gmail.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621162117.22533-4-krzk@kernel.org\n\ncommit 7b4e467d2f995702ad473cacd05a00a744a43cc5\nAuthor: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\nDate: Fri Jun 21 18:21:16 2019 +0200\n\n drm/lima: Reduce number of PTR_ERR() calls\n \n Store the PTR_ERR() result in local variable in clock init error path.\n This makes the code consistent with similar section in regulator init\n code.\n \n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Signed-off-by: Qiang Yu \u003cyuq825@gmail.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621162117.22533-3-krzk@kernel.org\n\ncommit 34e88f9e2fff2d46a206e4d94c9e8fdf45b85d9a\nAuthor: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\nDate: Fri Jun 21 18:21:15 2019 +0200\n\n drm/lima: Reduce the amount of logs on deferred probe\n \n There is no point to print deferred probe (and its failures to get\n resources) as an error. For example getting a regulator causes three\n unneeded error messages:\n \n lima 13000000.gpu: failed to get regulator: -517\n lima 13000000.gpu: regulator init fail -517\n lima 13000000.gpu: Fatal error during GPU init\n \n Also do not print clock rates before the initialization finishes\n because they will be duplicated after deferral. Each probe step already\n prints error so remove the final error message \"Fatal error during GPU\n init\".\n \n In case of multiple probe tries this would pollute the dmesg.\n \n Fixes: a1d2a6339961 (\"drm/lima: driver for ARM Mali4xx GPUs\")\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Signed-off-by: Qiang Yu \u003cyuq825@gmail.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621162117.22533-2-krzk@kernel.org\n\ncommit adbb48bcf4002fdf0901b936bd75db21bca8f514\nAuthor: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\nDate: Fri Jun 21 18:21:14 2019 +0200\n\n drm/lima: Mark 64-bit number as ULL\n \n Mark long numbers with ULL to silence the Smatch warning:\n \n drivers/gpu/drm/lima/lima_device.c:314:32: warning: constant 0x100000000 is so big it is long long\n \n Fixes: a1d2a6339961 (\"drm/lima: driver for ARM Mali4xx GPUs\")\n Signed-off-by: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Signed-off-by: Qiang Yu \u003cyuq825@gmail.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621162117.22533-1-krzk@kernel.org\n\ncommit b2dbf8d982a4f02a00261a5f8f75d2f0bf765de4\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jun 21 22:57:33 2019 +0100\n\n drm/i915/blt: Remove recursive vma-\u003elock\n \n As we have already plugged the w-\u003edma into the reservation_object, and\n have set ourselves up to automatically signal the request and w-\u003edma on\n completion, we do not need to export the rq-\u003efence directly and just use\n the w-\u003edma fence.\n \n This avoids having to take the reservation_lock inside the worker which\n cross-release lockdep would complain about. :)\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621215733.12070-1-chris@chris-wilson.co.uk\n\ncommit 4cdfb27ba80d29aa04bc340b2e4b59e9ae74004c\nAuthor: Juergen Gross \u003cjgross@suse.com\u003e\nDate: Fri Jun 14 07:46:04 2019 +0200\n\n xen/swiotlb: remember having called xen_create_contiguous_region()\n \n Instead of always calling xen_destroy_contiguous_region() in case the\n memory is DMA-able for the used device, do so only in case it has been\n made DMA-able via xen_create_contiguous_region() before.\n \n This will avoid a lot of xen_destroy_contiguous_region() calls for\n 64-bit capable devices.\n \n As the memory in question is owned by swiotlb-xen the PG_owner_priv_1\n flag of the first allocated page can be used for remembering.\n \n Signed-off-by: Juergen Gross \u003cjgross@suse.com\u003e\n Signed-off-by: Konrad Rzeszutek Wilk \u003ckonrad.wilk@oracle.com\u003e\n\ncommit f3a7bb9acba2a4ed44187bc5b050eb4aaec7fcf5\nAuthor: Juergen Gross \u003cjgross@suse.com\u003e\nDate: Fri Jun 14 07:46:03 2019 +0200\n\n xen/swiotlb: simplify range_straddles_page_boundary()\n \n range_straddles_page_boundary() is open coding several macros from\n include/xen/page.h. Use those instead. Additionally there is no need\n to have check_pages_physically_contiguous() as a separate function as\n it is used only once, so merge it into range_straddles_page_boundary().\n \n Signed-off-by: Juergen Gross \u003cjgross@suse.com\u003e\n Reviewed-by: Boris Ostrovsky \u003cboris.ostrovsky@oracle.com\u003e\n Signed-off-by: Konrad Rzeszutek Wilk \u003ckonrad.wilk@oracle.com\u003e\n\ncommit b0f24237c431a6e7353cbd389311eb581ffc2a67\nAuthor: Juergen Gross \u003cjgross@suse.com\u003e\nDate: Fri Jun 14 07:46:02 2019 +0200\n\n xen/swiotlb: fix condition for calling xen_destroy_contiguous_region()\n \n The condition in xen_swiotlb_free_coherent() for deciding whether to\n call xen_destroy_contiguous_region() is wrong: in case the region to\n be freed is not contiguous calling xen_destroy_contiguous_region() is\n the wrong thing to do: it would result in inconsistent mappings of\n multiple PFNs to the same MFN. This will lead to various strange\n crashes or data corruption.\n \n Instead of calling xen_destroy_contiguous_region() in that case a\n warning should be issued as that situation should never occur.\n \n Cc: stable@vger.kernel.org\n Signed-off-by: Juergen Gross \u003cjgross@suse.com\u003e\n Reviewed-by: Boris Ostrovsky \u003cboris.ostrovsky@oracle.com\u003e\n Reviewed-by: Jan Beulich \u003cjbeulich@suse.com\u003e\n Signed-off-by: Konrad Rzeszutek Wilk \u003ckonrad.wilk@oracle.com\u003e\n\ncommit d0b1e925e406813dc61357f1ef790d06c4ccc657\nAuthor: Arnd Bergmann \u003carnd@arndb.de\u003e\nDate: Mon Jun 17 15:28:43 2019 +0200\n\n swiotlb: fix phys_addr_t overflow warning\n \n On architectures that have a larger dma_addr_t than phys_addr_t,\n the swiotlb_tbl_map_single() function truncates its return code\n in the failure path, making it impossible to identify the error\n later, as we compare to the original value:\n \n kernel/dma/swiotlb.c:551:9: error: implicit conversion from \u0027dma_addr_t\u0027 (aka \u0027unsigned long long\u0027) to \u0027phys_addr_t\u0027 (aka \u0027unsigned int\u0027) changes value from 18446744073709551615 to 4294967295 [-Werror,-Wconstant-conversion]\n return DMA_MAPPING_ERROR;\n \n Use an explicit typecast here to convert it to the narrower type,\n and use the same expression in the error handling later.\n \n Fixes: b907e20508d0 (\"swiotlb: remove SWIOTLB_MAP_ERROR\")\n Acked-by: Stefano Stabellini \u003csstabellini@kernel.org\u003e\n Signed-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n Signed-off-by: Konrad Rzeszutek Wilk \u003ckonrad.wilk@oracle.com\u003e\n\ncommit 4d571204897e8a599eeafd92945f53ba4aa02d61\nAuthor: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\nDate: Tue Jun 11 10:58:25 2019 -0700\n\n swiotlb: Return consistent SWIOTLB segments/nr_tbl\n \n With a specifically contrived memory layout where there is no physical\n memory available to the kernel below the 4GB boundary, we will fail to\n perform the initial swiotlb_init() call and set no_iotlb_memory to true.\n \n There are drivers out there that call into swiotlb_nr_tbl() to determine\n whether they can use the SWIOTLB. With the right DMA_BIT_MASK() value\n for these drivers (say 64-bit), they won\u0027t ever need to hit\n swiotlb_tbl_map_single() so this can go unoticed and we would be\n possibly lying about those drivers.\n \n Signed-off-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Signed-off-by: Konrad Rzeszutek Wilk \u003ckonrad.wilk@oracle.com\u003e\n\ncommit 396bbe92b25a07b08882ff406ccf00db9310f32b\nAuthor: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\nDate: Tue Jun 11 10:58:24 2019 -0700\n\n swiotlb: Group identical cleanup in swiotlb_cleanup()\n \n Avoid repeating the zeroing of global swiotlb variables in two locations\n and introduce swiotlb_cleanup() to do that.\n \n Signed-off-by: Florian Fainelli \u003cf.fainelli@gmail.com\u003e\n Signed-off-by: Konrad Rzeszutek Wilk \u003ckonrad.wilk@oracle.com\u003e\n\ncommit 7b0cdf3e56b9b0bd05afa4904dc249c0f22dd30b\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:36:04 2019 +0200\n\n drm/vgem: Ditch attach trickery in the fence ioctl\n \n It looks like this was done purely to get a consistent place to look\n up the reservation object pointer. With the drm_prime.c helper code\n now also setting gem_object-\u003eresv for imported objects we can just use\n that pointer directly, instead of first ensuring a dma-buf exists.\n \n v2: Note that I screwed up the patch ordering, hence why this needed\n a Fixes: tag - CI spotted the broken intermediate state.\n \n Fixes: 5eab998741c8 (\"drm/vgem: Drop drm_gem_prime_export\")\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-49-daniel.vetter@ffwll.ch\n\ncommit 5f6ed9879a414636405a2bd77f122881695959e4\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:57 2019 +0200\n\n drm/prime: automatically set gem_obj-\u003eresv on import\n \n It\u0027s really the only reasonable thing to do, and it won\u0027t hurt drivers\n which don\u0027t (yet) use drm_gem_object-\u003eresv.\n \n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-42-daniel.vetter@ffwll.ch\n\ncommit fb993aa7a4049263fc1a139daf89fadc79a0ecc7\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jun 21 19:38:01 2019 +0100\n\n drm/i915: Local debug BUG_ON for intel_wakeref\n \n Avoid pulling in i915_gem.h just so that we can use a conditional BUG_ON\n for debugging.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621183801.23252-5-chris@chris-wilson.co.uk\n\ncommit 12c255b5dad115e87f81ea45708b5f82b9a55253\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jun 21 19:38:00 2019 +0100\n\n drm/i915: Provide an i915_active.acquire callback\n \n If we introduce a callback for i915_active that is only called the first\n time we use the i915_active and is symmetrically paired with the\n i915_active.retire callback, we can replace the open-coded and\n non-atomic implementations -- which will be very fragile (i.e. broken)\n upon removing the struct_mutex serialisation.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621183801.23252-4-chris@chris-wilson.co.uk\n\ncommit a93615f900bd19b59e74e04f7d8d4663ee5ea68f\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jun 21 19:37:59 2019 +0100\n\n drm/i915: Throw away the active object retirement complexity\n \n Remove the accumulated optimisations that we have for i915_vma_retire\n and reduce it to the bare essential of tracking the active object\n reference. This allows us to only use atomic operations, and so will be\n able to avoid the struct_mutex requirement.\n \n The principal loss here is the shrinker MRU bumping, so now if we have\n to shrink, we will do so in much more random order and more likely to\n try and shrink recently used objects. That is a nuisance, but shrinking\n active objects is a second step we try to avoid and will always be a\n system-wide performance issue.\n \n The other loss is here is in the automatic pruning of the\n reservation_object when idling. This is not as large an issue as upon\n reservation_object introduction as now adding new fences into the object\n replaces already signaled fences, keeping the array compact. But we do\n lose the auto-expiration of stale fences and unused arrays. That may be\n a noticeable problem for which we need to re-implement autopruning.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621183801.23252-3-chris@chris-wilson.co.uk\n\ncommit 5361db1a33c7e2d58af7df045d4d3ddd4c87ab56\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jun 21 19:37:58 2019 +0100\n\n drm/i915: Track i915_active using debugobjects\n \n Provide runtime asserts and tracking of i915_active via debugobjects.\n For example, this should allow us to check that the i915_active is only\n active when we expect it to be and is never freed too early.\n \n One consequence is that, for simplicity, we no longer allow i915_active\n to be on-stack which only affected the selftests.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621183801.23252-2-chris@chris-wilson.co.uk\n\ncommit 9e9539800dd44b1190128d48a116f4660f5d206f\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jun 21 19:37:57 2019 +0100\n\n drm/i915: Remove waiting \u0026 retiring from shrinker paths\n \n i915_gem_wait_for_idle() and i915_retire_requests() introduce a\n dependency on the timeline-\u003emutex. This is problematic as we want to\n later perform allocations underneath i915_active.mutex, forming a link\n between the shrinker, the timeline and active mutexes. Nip this cycle in\n the bud by removing the acquisition of the timeline mutex (i.e.\n retiring) from inside the shrinker.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621183801.23252-1-chris@chris-wilson.co.uk\n\ncommit 2ae70d28f27f6a5358541845c623760754d63ece\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Fri Jun 21 11:21:23 2019 -0700\n\n drm/i915/guc: handle GuC messages received with CTB disabled\n \n There is a very small chance of triggering a log flush event when\n enabling or disabling CT buffers. Events triggered while CT buffers\n are disabled are logged in the SCRATCH_15 register using the same bits\n used in the CT message payload. Since our communication channel with\n GuC is turned off, we can save the message and handle it after we turn\n it back on.\n GuC should be idle and not generate more events in the meantime because\n we\u0027re not talking to it.\n \n v2: clear the mmio register on stop_communication as well (Chris)\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621182123.31368-2-daniele.ceraolospurio@intel.com\n\ncommit e29cc1d7e820d662d2d73b35281b70f01a24610b\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Fri Jun 21 11:21:22 2019 -0700\n\n drm/i915/guc: reorder enable/disable communication steps\n \n Make sure we always have CT buffers enabled when the interrupts are\n enabled, so we can always handle interrupts from GuC. Also move the\n setting of the guc-\u003esend and guc-\u003ehandler functions to the GuC\n communication control functions for consistency.\n \n The reorder also fixes the onion unwinding of intel_uc_init_hw, because\n guc_enable_communication would\u0027ve left interrupts enabled when failing\n to enable CTB.\n \n v2: always retunr the result of ctch_enable() in\n intel_guc_ct_enable() (Michal)\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d110943\n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Reviewed-by: Matthew Brost \u003cmatthew.brost@intel.com\u003e\n Reviewed-by: Michal Wajdeczko \u003cmichal.wajdeczko@intel.com\u003e\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621182123.31368-1-daniele.ceraolospurio@intel.com\n\ncommit 634cdf7fd91a44f52a4d56fb9a94c376c322f82a\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:56 2019 +0200\n\n drm/vram-helper: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Acked-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Cc: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-41-daniel.vetter@ffwll.ch\n\ncommit 42169858d75c08558e288921c3d21d5fa1995072\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:55 2019 +0200\n\n drm/zte: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Acked-by: Shawn Guo \u003cshawnguo@kernel.org\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Shawn Guo \u003cshawnguo@kernel.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-40-daniel.vetter@ffwll.ch\n\ncommit f30dec2f82402c21b6ddf81b8f8350e72d49c90e\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:54 2019 +0200\n\n drm/xen: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Oleksandr Andrushchenko \u003coleksandr_andrushchenko@epam.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Oleksandr Andrushchenko \u003coleksandr_andrushchenko@epam.com\u003e\n Cc: xen-devel@lists.xenproject.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-39-daniel.vetter@ffwll.ch\n\ncommit aada617da7b657e625aa3bca4dab8830cd5c6401\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:53 2019 +0200\n\n drm/virtio: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Acked-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Cc: virtualization@lists.linux-foundation.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-38-daniel.vetter@ffwll.ch\n\ncommit 5eab998741c868fc1dc3c90b306ce0e807ef5fca\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:52 2019 +0200\n\n drm/vgem: Drop drm_gem_prime_export\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-37-daniel.vetter@ffwll.ch\n\ncommit 9030142e320685d21eca85db66b50ddf835bcbf1\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:51 2019 +0200\n\n drm/radeon: Drop drm_gem_prime_import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-36-daniel.vetter@ffwll.ch\n\ncommit ddad8b452e5f52bcfaf05a3bb250dcf087c98617\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:50 2019 +0200\n\n drm/vc3: Drop drm_gem_prime_import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-35-daniel.vetter@ffwll.ch\n\ncommit 8a015561479639b8a7a2c3214bd24dbb055d8d39\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:49 2019 +0200\n\n drm/vboxvideo: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Hans de Goede \u003chdegoede@redhat.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-34-daniel.vetter@ffwll.ch\n\ncommit cb822cab0537a8b65d68f2750534fd199e2ef84e\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:48 2019 +0200\n\n drm/tve2000: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-33-daniel.vetter@ffwll.ch\n\ncommit 52de698543c94807c753bdf40d9addbcff3dc2d8\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:47 2019 +0200\n\n drm/tilcdc: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Jyri Sarha \u003cjsarha@ti.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-32-daniel.vetter@ffwll.ch\n\ncommit b8764eaf8ad85615245e6039b7d2ca8a92940014\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:46 2019 +0200\n\n drm/stm: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Acked-by: Philippe Cornu \u003cphilippe.cornu@st.com\u003e\n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Yannick Fertre \u003cyannick.fertre@st.com\u003e\n Cc: Philippe Cornu \u003cphilippe.cornu@st.com\u003e\n Cc: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Cc: Vincent Abriou \u003cvincent.abriou@st.com\u003e\n Cc: Maxime Coquelin \u003cmcoquelin.stm32@gmail.com\u003e\n Cc: Alexandre Torgue \u003calexandre.torgue@st.com\u003e\n Cc: linux-stm32@st-md-mailman.stormreply.com\n Cc: linux-arm-kernel@lists.infradead.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-31-daniel.vetter@ffwll.ch\n\ncommit 03988e4fa15689da0ab954457c8578773c6e05b6\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:45 2019 +0200\n\n drm/sti: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Cc: Vincent Abriou \u003cvincent.abriou@st.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-30-daniel.vetter@ffwll.ch\n\ncommit cf64beca45f9dec6e87a2820a49a2872a846925a\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:44 2019 +0200\n\n drm/shmob: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Cc: linux-renesas-soc@vger.kernel.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-29-daniel.vetter@ffwll.ch\n\ncommit f59bcca620e84a9e7a21779d9acfeca373b41731\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:43 2019 +0200\n\n drm/rockchip: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Tested-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Reviewed-by: Heiko Stuebner \u003cheiko@sntech.de\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Sandy Huang \u003chjc@rock-chips.com\u003e\n Cc: \"Heiko Stübner\" \u003cheiko@sntech.de\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: linux-rockchip@lists.infradead.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-28-daniel.vetter@ffwll.ch\n\ncommit b392cb98dfff242d4edb6e045e52f155c12bcd8b\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:42 2019 +0200\n\n drm/rcar-du: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Laurent Pinchart \u003claurent.pinchart@ideasonboard.com\u003e\n Cc: Kieran Bingham \u003ckieran.bingham+renesas@ideasonboard.com\u003e\n Cc: linux-renesas-soc@vger.kernel.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-27-daniel.vetter@ffwll.ch\n\ncommit baa78332a7ca01f72390d8fe7ade9b278ab7f5d4\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:41 2019 +0200\n\n drm/qxl: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Acked-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Dave Airlie \u003cairlied@redhat.com\u003e\n Cc: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Cc: virtualization@lists.linux-foundation.org\n Cc: spice-devel@lists.freedesktop.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-26-daniel.vetter@ffwll.ch\n\ncommit a23916462848873018e2300a54dc195899223548\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:40 2019 +0200\n\n drm/pl111: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-25-daniel.vetter@ffwll.ch\n\ncommit 78fc89063f0c73aed9b9d35c84c81277b7115c6c\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:39 2019 +0200\n\n drm/nouveau: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Ben Skeggs \u003cbskeggs@redhat.com\u003e\n Cc: nouveau@lists.freedesktop.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-24-daniel.vetter@ffwll.ch\n\ncommit 4fdfffc8f31830aef2016bb3c2de050056f975a2\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:38 2019 +0200\n\n drm/mxsfb: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Acked-by: Stefan Agner \u003cstefan@agner.ch\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Marek Vasut \u003cmarex@denx.de\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Shawn Guo \u003cshawnguo@kernel.org\u003e\n Cc: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\n Cc: Pengutronix Kernel Team \u003ckernel@pengutronix.de\u003e\n Cc: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Cc: NXP Linux Team \u003clinux-imx@nxp.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-23-daniel.vetter@ffwll.ch\n\ncommit d229c592d12d1cf1c288adb03fe6f087c6b206e6\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:37 2019 +0200\n\n drm/msm: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Rob Clark \u003crobdclark@gmail.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: linux-arm-msm@vger.kernel.org\n Cc: freedreno@lists.freedesktop.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-22-daniel.vetter@ffwll.ch\n\ncommit 4bcc9543ad4ccbe88e2cfc2c432eec62da605064\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:36 2019 +0200\n\n drm/meson: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Acked-by: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Neil Armstrong \u003cnarmstrong@baylibre.com\u003e\n Cc: Kevin Hilman \u003ckhilman@baylibre.com\u003e\n Cc: linux-amlogic@lists.infradead.org\n Cc: linux-arm-kernel@lists.infradead.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-21-daniel.vetter@ffwll.ch\n\ncommit 3baeeb21983a5ecc16955cc3f1d986bee88939f2\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:35 2019 +0200\n\n drm/mtk: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: CK Hu \u003cck.hu@mediatek.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: CK Hu \u003cck.hu@mediatek.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Matthias Brugger \u003cmatthias.bgg@gmail.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: linux-mediatek@lists.infradead.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-20-daniel.vetter@ffwll.ch\n\ncommit bf6f1fa62e76f93d1030a9e22923881cebc770b3\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:34 2019 +0200\n\n drm/mcde: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-19-daniel.vetter@ffwll.ch\n\ncommit c5be0ddd500dd6201d168b4fdd1ffa90846cb437\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:33 2019 +0200\n\n drm/imx: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Philipp Zabel \u003cp.zabel@pengutronix.de\u003e\n Cc: Shawn Guo \u003cshawnguo@kernel.org\u003e\n Cc: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\n Cc: Pengutronix Kernel Team \u003ckernel@pengutronix.de\u003e\n Cc: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Cc: NXP Linux Team \u003clinux-imx@nxp.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-18-daniel.vetter@ffwll.ch\n\ncommit 8c6555d4a6fb18eb05a78173becc90e00333e6c6\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:32 2019 +0200\n\n drm/hisilicon: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Cc: Xinliang Liu \u003cz.liuxinliang@hisilicon.com\u003e\n Cc: \"Noralf Trønnes\" \u003cnoralf@tronnes.org\u003e\n Cc: CK Hu \u003cck.hu@mediatek.com\u003e\n Cc: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-17-daniel.vetter@ffwll.ch\n\ncommit 40e546c5f9ca0054087ce5ee04de96a4f28e9a97\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:31 2019 +0200\n\n drm/fsl-dcu: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Acked-by: Stefan Agner \u003cstefan@agner.ch\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Stefan Agner \u003cstefan@agner.ch\u003e\n Cc: Alison Wang \u003calison.wang@nxp.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-16-daniel.vetter@ffwll.ch\n\ncommit 13ea486fe1c6990a9a39f8e36dce8fde1def48bc\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:30 2019 +0200\n\n drm/exynos: Drop drm_gem_prime_export\n \n They\u0027re the default. We can\u0027t do the same on the import side, due to\n the exynos_drm-\u003edma_dev not necessarily matching the overall drm\n device.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Inki Dae \u003cinki.dae@samsung.com\u003e\n Cc: Joonyoung Shim \u003cjy0922.shim@samsung.com\u003e\n Cc: Seung-Woo Kim \u003csw0312.kim@samsung.com\u003e\n Cc: Kyungmin Park \u003ckyungmin.park@samsung.com\u003e\n Cc: Kukjin Kim \u003ckgene@kernel.org\u003e\n Cc: Krzysztof Kozlowski \u003ckrzk@kernel.org\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: linux-samsung-soc@vger.kernel.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-15-daniel.vetter@ffwll.ch\n\ncommit 2550416ccf19cf882f2f8c06b6aa97fa732c552f\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:29 2019 +0200\n\n drm/etnaviv: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Christian Gmeiner \u003cchristian.gmeiner@gmail.com\u003e\n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Lucas Stach \u003cl.stach@pengutronix.de\u003e\n Cc: Russell King \u003clinux+etnaviv@armlinux.org.uk\u003e\n Cc: Christian Gmeiner \u003cchristian.gmeiner@gmail.com\u003e\n Cc: etnaviv@lists.freedesktop.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-14-daniel.vetter@ffwll.ch\n\ncommit b82a6fd04ec371e1100984aabf7b93d28502649e\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:28 2019 +0200\n\n drm/atmel: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Boris Brezillon \u003cbbrezillon@kernel.org\u003e\n Cc: Nicolas Ferre \u003cnicolas.ferre@microchip.com\u003e\n Cc: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n Cc: Ludovic Desroches \u003cludovic.desroches@microchip.com\u003e\n Cc: linux-arm-kernel@lists.infradead.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-13-daniel.vetter@ffwll.ch\n\ncommit d3bc25f3bff30881051012bf949dc89f8cfcfd1f\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:27 2019 +0200\n\n drm/arm: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: James Qian Wang (Arm Technology China) \u003cjames.qian.wang@arm.com\u003e\n Acked-by: Liviu Dudau \u003cliviu.dudau@arm.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: \"James (Qian) Wang\" \u003cjames.qian.wang@arm.com\u003e\n Cc: Liviu Dudau \u003cliviu.dudau@arm.com\u003e\n Cc: Brian Starkey \u003cbrian.starkey@arm.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-12-daniel.vetter@ffwll.ch\n\ncommit b5031e86a9afd9f01104178faa24b3096bded907\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:26 2019 +0200\n\n drm/arc: Drop drm_gem_prime_export/import\n \n They\u0027re the default.\n \n Aside: Would be really nice to switch the others over to\n drm_gem_object_funcs.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Alexey Brodkin \u003cabrodkin@synopsys.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-11-daniel.vetter@ffwll.ch\n\ncommit 75426367cd377120a256cad0b35b02eec4b83591\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Wed Jun 5 14:08:35 2019 +0200\n\n drm/ioctl: Ditch DRM_UNLOCKED except for the legacy vblank ioctl\n \n This completes Emil\u0027s series of removing DRM_UNLOCKED from modern\n drivers. It\u0027s entirely cargo-culted since we ignore it on\n non-DRIVER_LEGACY drivers since:\n \n commit ea487835e8876abf7ad909636e308c801a2bcda6\n Author: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Date: Mon Sep 28 21:42:40 2015 +0200\n \n drm: Enforce unlocked ioctl operation for kms driver ioctls\n \n Now justifying why we can do this for legacy drives too (and hence\n close the source of all the bogus copypasting) is a bit more involved.\n DRM_UNLOCKED was introduced in:\n \n commit ed8b67040965e4fe695db333d5914e18ea5f146f\n Author: Arnd Bergmann \u003carnd@arndb.de\u003e\n Date: Wed Dec 16 22:17:09 2009 +0000\n \n drm: convert drm_ioctl to unlocked_ioctl\n \n As a immediate hack to keep i810 happy, which would have deadlocked\n without this trickery. The old BKL is automatically dropped in\n schedule(), and hence the i810 vs. mmap_sem deadlock didn\u0027t actually\n cause a real deadlock. But with a mutex it would. The solution was to\n annotate these as DRM_UNLOCKED and mark i810 unsafe on SMP machines.\n \n This conversion caused a regression, because unlike the BKL a mutex\n isn\u0027t dropped over schedule (that thing again), which caused a vblank\n wait in one thread to block the entire desktop and all its apps. Back\n then we did vblank scheduling by blocking in the client, awesome isn\u0027t\n it. This was fixed quickly in (ok not so quickly, took 2 years):\n \n commit 8f4ff2b06afcd6f151868474a432c603057eaf56\n Author: Ilija Hadzic \u003cihadzic@research.bell-labs.com\u003e\n Date: Mon Oct 31 17:46:18 2011 -0400\n \n drm: do not sleep on vblank while holding a mutex\n \n All the other DRM_UNLOCKED annotations for all the core ioctls was\n work to reach finer-grained locking for modern drivers. This took\n years, and culminated in:\n \n commit fdd5b877e9ebc2029e1373b4a3cd057329a9ab7a\n Author: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\n Date: Sat Dec 10 22:52:54 2016 +0100\n \n drm: Enforce BKL-less ioctls for modern drivers\n \n DRM_UNLOCKED was never required by any legacy drivers, except for the\n vblank_wait IOCTL. Therefore we will not regress these old drivers by\n going back to where we\u0027ve been in 2011. For all modern drivers nothing\n will change.\n \n To make this perfectly clear, also add a comment to DRM_UNLOCKED.\n \n v2: Don\u0027t forget about drm_ioc32.c (Michel).\n \n Cc: Michel Dänzer \u003cmichel@daenzer.net\u003e\n Cc: Emil Velikov \u003cemil.l.velikov@gmail.com\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Acked-by: Michel Dänzer \u003cmichel@daenzer.net\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190605120835.2798-1-daniel.vetter@ffwll.ch\n\ncommit e4fa8457b2197118538a1400b75c898f9faaf164\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:25 2019 +0200\n\n drm/prime: Align gem_prime_export with obj_funcs.export\n \n The idea is that gem_prime_export is deprecated in favor of\n obj_funcs.export. That\u0027s much easier to do if both have matching\n function signatures.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Acked-by: Christian König \u003cchristian.koenig@amd.com\u003e\n Acked-by: Thierry Reding \u003ctreding@nvidia.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Russell King \u003clinux@armlinux.org.uk\u003e\n Cc: Maarten Lankhorst \u003cmaarten.lankhorst@linux.intel.com\u003e\n Cc: Maxime Ripard \u003cmaxime.ripard@bootlin.com\u003e\n Cc: Sean Paul \u003csean@poorly.run\u003e\n Cc: David Airlie \u003cairlied@linux.ie\u003e\n Cc: Daniel Vetter \u003cdaniel@ffwll.ch\u003e\n Cc: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n Cc: Zhi Wang \u003czhi.a.wang@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@linux.intel.com\u003e\n Cc: Joonas Lahtinen \u003cjoonas.lahtinen@linux.intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Cc: Alex Deucher \u003calexander.deucher@amd.com\u003e\n Cc: \"Christian König\" \u003cchristian.koenig@amd.com\u003e\n Cc: \"David (ChunMing) Zhou\" \u003cDavid1.Zhou@amd.com\u003e\n Cc: Thierry Reding \u003cthierry.reding@gmail.com\u003e\n Cc: Jonathan Hunter \u003cjonathanh@nvidia.com\u003e\n Cc: Dave Airlie \u003cairlied@redhat.com\u003e\n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: \"Michel Dänzer\" \u003cmichel.daenzer@amd.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Huang Rui \u003cray.huang@amd.com\u003e\n Cc: Felix Kuehling \u003cFelix.Kuehling@amd.com\u003e\n Cc: Hawking Zhang \u003cHawking.Zhang@amd.com\u003e\n Cc: Feifei Xu \u003cFeifei.Xu@amd.com\u003e\n Cc: Jim Qu \u003cJim.Qu@amd.com\u003e\n Cc: Evan Quan \u003cevan.quan@amd.com\u003e\n Cc: Matthew Auld \u003cmatthew.auld@intel.com\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Cc: Thomas Zimmermann \u003ctdz@users.sourceforge.net\u003e\n Cc: Kate Stewart \u003ckstewart@linuxfoundation.org\u003e\n Cc: Sumit Semwal \u003csumit.semwal@linaro.org\u003e\n Cc: Jilayne Lovejoy \u003copensource@jilayne.com\u003e\n Cc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n Cc: Mikulas Patocka \u003cmpatocka@redhat.com\u003e\n Cc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n Cc: Junwei Zhang \u003cJerry.Zhang@amd.com\u003e\n Cc: intel-gvt-dev@lists.freedesktop.org\n Cc: intel-gfx@lists.freedesktop.org\n Cc: amd-gfx@lists.freedesktop.org\n Cc: linux-tegra@vger.kernel.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-10-daniel.vetter@ffwll.ch\n\ncommit 8b3026a74fa764d6ae42eb80fda5670876c6339a\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:24 2019 +0200\n\n drm/omapdrm: drop fb_debug_enter/leave\n \n This is a no-op on atomic drivers because with atomic it\u0027s simply too\n complicated to get all the locking and workers and nonblocking\n synchronization correct, from essentially an NMI context. Well, too\n complicated \u003d impossible. Also, omapdrm never implemented the\n mode_set_base_atomic hook, so I kinda wonder why this was ever added.\n \n Drop the hooks.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: Tomi Valkeinen \u003ctomi.valkeinen@ti.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-9-daniel.vetter@ffwll.ch\n\ncommit 055a12ff16dd8ac999b55278545183aeba59a280\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:23 2019 +0200\n\n drm/arm/komeda: Remove DRIVER_HAVE_IRQ\n \n Read the docs, komeda is not an old enough driver for this :-)\n \n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: James Qian Wang (Arm Technology China) \u003cjames.qian.wang@arm.com\u003e\n Acked-by: Liviu Dudau \u003cliviu.dudau@arm.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: \"James (Qian) Wang\" \u003cjames.qian.wang@arm.com\u003e\n Cc: Liviu Dudau \u003cliviu.dudau@arm.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-8-daniel.vetter@ffwll.ch\n\ncommit 0424fdaf883a689d5185c0d0665b265373945898\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Mon Jun 17 17:39:24 2019 +0200\n\n drm/prime: Actually remove DRIVER_PRIME everywhere\n \n Split out to make the functional changes stick out more.\n \n All places where DRIVER_PRIME was used have been removed in previous\n patches already.\n \n v2: amdgpu gained DRIVER_SYNCOBJ_TIMELINE.\n \n v3: amdgpu lost DRIVER_SYNCOBJ_TIMELINE.\n \n v4: Don\u0027t add a space in i915_drv.c (Sam)\n \n v5: Add note that previous patches removed all the DRIVER_PRIME users\n already (Emil).\n \n v6: Fixupe ingenic (new driver) while applying.\n \n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Cc: amd-gfx@lists.freedesktop.org\n Cc: etnaviv@lists.freedesktop.org\n Cc: freedreno@lists.freedesktop.org\n Cc: intel-gfx@lists.freedesktop.org\n Cc: lima@lists.freedesktop.org\n Cc: linux-amlogic@lists.infradead.org\n Cc: linux-arm-kernel@lists.infradead.org\n Cc: linux-arm-msm@vger.kernel.org\n Cc: linux-aspeed@lists.ozlabs.org\n Cc: linux-renesas-soc@vger.kernel.org\n Cc: linux-rockchip@lists.infradead.org\n Cc: linux-samsung-soc@vger.kernel.org\n Cc: linux-stm32@st-md-mailman.stormreply.com\n Cc: linux-tegra@vger.kernel.org\n Cc: nouveau@lists.freedesktop.org\n Cc: NXP Linux Team \u003clinux-imx@nxp.com\u003e\n Cc: spice-devel@lists.freedesktop.org\n Cc: virtualization@lists.linux-foundation.org\n Cc: VMware Graphics \u003clinux-graphics-maintainer@vmware.com\u003e\n Cc: xen-devel@lists.xenproject.org\n Link: https://patchwork.freedesktop.org/patch/msgid/20190617153924.414-1-daniel.vetter@ffwll.ch\n\ncommit 80fc1c1991cdaefb9d5c9733c6c5b85b92498eee\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jun 21 14:16:40 2019 +0100\n\n drm/i915/gt: Fixup kerneldoc parameters\n \n drivers/gpu/drm/i915/gt/intel_mocs.c:513: warning: Function parameter or member \u0027gt\u0027 not described in \u0027intel_mocs_init_l3cc_table\u0027\n drivers/gpu/drm/i915/gt/intel_mocs.c:513: warning: Excess function parameter \u0027dev_priv\u0027 description in \u0027intel_mocs_init_l3cc_table\u0027\n \n intel_vgt_balloon/deballoon, i915_ggtt_probe_hw intel_wopcm_init_hw need\n similar treatment\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621131640.28864-2-chris@chris-wilson.co.uk\n\ncommit c6fe28b0c27dfc8103af453ed9723907cd96e3f0\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jun 21 14:16:39 2019 +0100\n\n drm/i915/gt: Rename i915_gt_timelines\n \n Since the anonymous i915_gt became struct intel_gt and encloses\n struct i915_gt_timelines, rename i915_gt_timelines to intel_gt_timelines\n to match its parentage.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621131640.28864-1-chris@chris-wilson.co.uk\n\ncommit 5f6730a4689269240e911853403456e243513ae0\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jun 21 14:52:46 2019 +0100\n\n drm/i915: Prevent dereference of engine before NULL check in error capture\n \n smatch caught,\n drivers/gpu/drm/i915/i915_gpu_error.c:1418 gem_record_rings() warn: variable dereferenced before check \u0027engine\u0027 (see line 1413)\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621135246.20683-1-chris@chris-wilson.co.uk\n\ncommit db56f974941b75b2c96e577100e7abe64b03d9b9\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:08:11 2019 +0100\n\n drm/i915: Eliminate dual personality of i915_scratch_offset\n \n Scratch vma lives under gt but the API used to work on i915. Make this\n consistent by renaming the function to intel_gt_scratch_offset and make\n it take struct intel_gt.\n \n v2:\n * Move to intel_gt. (Chris)\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-33-tvrtko.ursulin@linux.intel.com\n\ncommit f0c02c1b91888ccac539388eacb0659bf263a557\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:08:10 2019 +0100\n\n drm/i915: Rename i915_timeline to intel_timeline and move under gt\n \n Move all timeline code under gt and rename to intel_gt prefix.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Suggested-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-32-tvrtko.ursulin@linux.intel.com\n\ncommit 4c6d51ea2a68699d0d47c2361f691cf9265371b6\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:08:09 2019 +0100\n\n drm/i915: Make timelines gt centric\n \n Our timelines are stored inside intel_gt so we can convert the interface\n to take exactly that and not i915.\n \n At the same time re-order the params to our more typical layout and\n replace the backpointer to the new containing structure.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-31-tvrtko.ursulin@linux.intel.com\n\ncommit ba4134a41931e0e25d29bb81f8b30a522e29f6f0\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:08:08 2019 +0100\n\n drm/i915: Save trip via top-level i915 in a few more places\n \n For gt related operations it makes more logical sense to stay in the realm\n of gt instead of dereferencing via driver i915.\n \n This patch handles a few of the easy ones with work requiring more\n refactoring still outstanding.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-30-tvrtko.ursulin@linux.intel.com\n\ncommit db45fb5bc1a09af37d2b0e6c4c3c258d1d732b89\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:08:07 2019 +0100\n\n drm/i915: Compartmentalize ring buffer creation\n \n Continuing on the theme of compartmentalizing the code better to make\n future split between gt and display in global i915 clearer.\n \n v2:\n * Pass in ggtt instead of gt. (Chris)\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-29-tvrtko.ursulin@linux.intel.com\n\ncommit d8a4424839eead6da07835e7d2f8b8f54ad6c6b0\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:08:06 2019 +0100\n\n drm/i915: Store ggtt pointer in intel_gt\n \n This will become useful in the following patch.\n \n v2:\n * Assign the pointer through a helper on the top level to work around\n the layering violation. (Chris)\n \n v3:\n * Handle selftests.\n \n v4:\n * Move call to intel_gt_init_hw into mock_init_ggtt. (Chris)\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-28-tvrtko.ursulin@linux.intel.com\n\ncommit 1d66377a76bd146bb5c16431549ed1d586fdfeb0\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:08:05 2019 +0100\n\n drm/i915: Compartmentalize i915_gem_init_ggtt\n \n Continuing on the theme of better logical organization of our code, make\n the first step towards making the ggtt code better isolated from wider\n struct drm_i915_private.\n \n v2:\n * Bring the ickle onion unwind back. (Chris)\n * Rename to i915_init_ggtt. (Chris)\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-27-tvrtko.ursulin@linux.intel.com\n\ncommit 3b8966287fa91403337742fa6b954880131bb1c0\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:08:04 2019 +0100\n\n drm/i915: Compartmentalize i915_ggtt_cleanup_hw\n \n Continuing on the theme of better logical organization of our code, make\n the first step towards making the ggtt code better isolated from wider\n struct drm_i915_private.\n \n v2:\n * Cleanup of mm.wc_stash does not need struct_mutex. (Chris)\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-26-tvrtko.ursulin@linux.intel.com\n\ncommit 390c82055b74e14d36a1a18ec772af1e032bde71\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:08:03 2019 +0100\n\n drm/i915: Compartmentalize timeline_init/park/fini\n \n Continuing on the theme of better logical organization of our code, make\n the first step towards making the timeline code better isolated from wider\n struct drm_i915_private.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-25-tvrtko.ursulin@linux.intel.com\n\ncommit baea429dc5ed90cf9afbdf326fa527581745c8f1\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:08:02 2019 +0100\n\n drm/i915: Move i915_gem_chipset_flush to intel_gt\n \n This aligns better with the rest of restructuring.\n \n v2:\n * Move call out of line. (Chris)\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Suggested-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-24-tvrtko.ursulin@linux.intel.com\n\ncommit a1c8a09e0c7c39ee079fc7755cd665efefd9ef73\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:08:01 2019 +0100\n\n drm/i915: Convert i915_gem_flush_ggtt_writes to intel_gt\n \n Having introduced struct intel_gt (named the anonymous structure in i915)\n we can start using it to compartmentalize our code better. It makes more\n sense logically to have the code internally like this and it will also\n help with future split between gt and display in i915.\n \n v2:\n * Keep ggtt flush before fb obj flush. (Chris)\n \n v3:\n * Fix refactoring fail.\n * Always flush ggtt writes. (Chris)\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-23-tvrtko.ursulin@linux.intel.com\n\ncommit 68c754b850840eb8258211306371e9fa3d706667\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:08:00 2019 +0100\n\n drm/i915: Compartmentalize i915_gem_suspend/restore_gtt_mappings\n \n Having made start to better code compartmentalization by introducing\n struct intel_gt, continue the theme elsewhere in code by making functions\n take parameters take what logically makes most sense for them instead of\n the global struct drm_i915_private.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-22-tvrtko.ursulin@linux.intel.com\n\ncommit 763c1e63125ef18561b36aafcf983230aeda4212\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:59 2019 +0100\n\n drm/i915: Store intel_gt backpointer in vm\n \n This will come useful in the following patch.\n \n v2:\n * Handle mock ggtt.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-21-tvrtko.ursulin@linux.intel.com\n\ncommit 759e4a74f354c51509d2aae93d32381879f11344\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:58 2019 +0100\n\n drm/i915: Make ggtt invalidation work on ggtt\n \n It is more logical for ggtt invalidation to take ggtt as input parameter.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-20-tvrtko.ursulin@linux.intel.com\n\ncommit 8b5342f58d3cea762280f34eb68eb0b640393997\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:57 2019 +0100\n\n drm/i915: Compartmentalize i915_ggtt_init_hw\n \n Having made start to better code compartmentalization by introducing\n struct intel_gt, continue the theme elsewhere in code by making functions\n take parameters take what logically makes most sense for them instead of\n the global struct drm_i915_private.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-19-tvrtko.ursulin@linux.intel.com\n\ncommit ee1de7dd7d33bb753725178dae3bc652eb7611a4\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:56 2019 +0100\n\n drm/i915: Compartmentalize i915_ggtt_probe_hw\n \n Having made start to better code compartmentalization by introducing\n struct intel_gt, continue the theme elsewhere in code by making functions\n take parameters take what logically makes most sense for them instead of\n the global struct drm_i915_private.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-18-tvrtko.ursulin@linux.intel.com\n\ncommit 6b0a8dfdf27e6d6a180598b6ff6b205d9055d975\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:55 2019 +0100\n\n drm/i915: Stop using I915_READ/WRITE in intel_wopcm_init_hw\n \n More legacy mmio accessor removal. We pass in intel_gt explicitly allowing\n code to use new intel_uncore_read/write helpers.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-17-tvrtko.ursulin@linux.intel.com\n\ncommit 8649187a9520f61b742c85f5d313ad89617d86fe\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:54 2019 +0100\n\n drm/i915: Move intel_engines_resume into common init\n \n Since this part still operates on i915 and not intel_gt, move it to the\n common (top-level) function.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-16-tvrtko.ursulin@linux.intel.com\n\ncommit abc584f9aa042ff75b7ff1749c221957b46a8c99\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:53 2019 +0100\n\n drm/i915: Convert i915_gem_init_hw to intel_gt\n \n More removal of implicit dev_priv from using old mmio accessors.\n \n Actually the top level function remains but is split into a part which\n writes to i915 and part which operates on intel_gt in order to initialize\n the hardware.\n \n GuC and engines are the only odd ones out remaining.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-15-tvrtko.ursulin@linux.intel.com\n\ncommit 28a1f7894f6b518bbc4f5d23f7bc7b2edc1d6c3f\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:52 2019 +0100\n\n drm/i915: Consolidate some open coded mmio rmw\n \n Replace some gen6/7 open coded rmw with intel_uncore_rmw.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-14-tvrtko.ursulin@linux.intel.com\n\ncommit acb56d97d907de6b853f888128755116ee5ff3a9\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:51 2019 +0100\n\n drm/i915: Convert i915_ppgtt_init_hw to intel_gt\n \n More removal of implicit dev_priv from using old mmio accessors.\n \n v2:\n * Rebase for uncore_to_i915 removal.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-13-tvrtko.ursulin@linux.intel.com\n\ncommit 20a7f2fc4d7a4eb1811595d18ffa32b0811ede0b\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:50 2019 +0100\n\n drm/i915: Convert intel_mocs_init_l3cc_table to intel_gt\n \n More removal of implicit dev_priv from using old mmio accessors.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-12-tvrtko.ursulin@linux.intel.com\n\ncommit f937f5613b6f108b2f93470f357c0880c1b57ffb\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:49 2019 +0100\n\n drm/i915: Store backpointer to intel_gt in the engine\n \n It will come useful in the next patch.\n \n v2:\n * Do mock_engine as well.\n \n v3:\n * And the virtual engine...\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-11-tvrtko.ursulin@linux.intel.com\n\ncommit d10cfee4d81ba3483a09ea70ddd78889bbca0384\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:48 2019 +0100\n\n drm/i915: Convert gt workarounds to intel_gt\n \n More conversion of i915_gem_init_hw to uncore.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-10-tvrtko.ursulin@linux.intel.com\n\ncommit cf6844b234e5832c3ab4d8ad76fee433905a1de6\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:47 2019 +0100\n\n drm/i915: Convert init_unused_rings to intel_gt\n \n More removal of implicit dev_priv from using old mmio accessors.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-9-tvrtko.ursulin@linux.intel.com\n\ncommit f88709bd1c5e2bc9fb62b1b3151fae1ef4f7a2c5\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:46 2019 +0100\n\n drm/i915: Use intel_uncore_rmw in intel_gt_init_swizzling\n \n Two easy opportunities to compact the code by using the existing helper.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Suggested-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-8-tvrtko.ursulin@linux.intel.com\n\ncommit 500bfa380e112f744f8ae316c88c0e849833d804\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:45 2019 +0100\n\n drm/i915: Convert i915_gem_init_swizzling to intel_gt\n \n Start using the newly introduced struct intel_gt to fuse together correct\n logical init flow with uncore for more removal of implicit dev_priv in\n mmio access.\n \n v2:\n * Move code to i915_gem_fence_reg. (Chris)\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-7-tvrtko.ursulin@linux.intel.com\n\ncommit eaf522f62b7454c4c040c4672c36ad0f4cb21fcd\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:44 2019 +0100\n\n drm/i915: Make i915_check_and_clear_faults take intel_gt\n \n Continuing the conversion and elimination of implicit dev_priv.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Suggested-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-6-tvrtko.ursulin@linux.intel.com\n\ncommit 99f2eb966749eb3e7523323a2f09b2f677d3a2d0\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:43 2019 +0100\n\n drm/i915: Move intel_gt_pm_init under intel_gt_init_early\n \n And also rename to intel_gt_pm_init_early and make it operate on gt.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Suggested-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-5-tvrtko.ursulin@linux.intel.com\n\ncommit 724e9564c5acd40c422db2ccb57b1f66433021aa\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:42 2019 +0100\n\n drm/i915: Store some backpointers in struct intel_gt\n \n We need an easy way to get back to i915 and uncore.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-4-tvrtko.ursulin@linux.intel.com\n\ncommit 24635c5152af5a6b79eec17d721685277f8d4cfc\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:41 2019 +0100\n\n drm/i915: Move intel_gt initialization to a separate file\n \n As it will grow in a following patch make a new home for it.\n \n v2:\n * Convert mock_gem_device as well. (Chris)\n \n v3:\n * Rename to intel_gt_init_early and move call site to i915_drv.c. (Chris)\n \n v4:\n * Adjust SPDX tags.\n * No need to gt/ path when including intel_gt_types.h. (Chris)\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-3-tvrtko.ursulin@linux.intel.com\n\ncommit e5be5c7a8ffa8471fb31554cf56f3efd11f7e454\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:40 2019 +0100\n\n drm/i915: Introduce struct intel_gt as replacement for anonymous i915-\u003egt\n \n We have long been slighlty annoyed by the anonymous i915-\u003egt.\n \n Promote it to a separate structure and give it its own header.\n \n This is a first step towards cleaning up the separation between\n i915 and gt.\n \n v2:\n * Adjust SPDX header.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-2-tvrtko.ursulin@linux.intel.com\n\ncommit 3cb4ce0024f1e24f7da96ed1d02430c407de13fa\nAuthor: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\nDate: Fri Jun 21 08:07:39 2019 +0100\n\n drm/i915: Convert intel_vgt_(de)balloon to uncore\n \n More removal of implicit dev_priv from using old mmio accessors.\n \n Furthermore these calls really operate on ggtt so it logically makes sense\n if they take it as parameter.\n \n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621070811.7006-1-tvrtko.ursulin@linux.intel.com\n\ncommit ee8375d5dc5bbb50b03bedfb0020d3e1c27ceacb\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:21 2019 +0200\n\n drm/prime: Make DRIVER_PRIME a no-op\n \n Drivers must fill out the handle_to_fd and fd_to_handle hooks to\n enable export/import prime functionality already. The additional\n DRIVER_PRIME flag doesn\u0027t serve any real purpose, since the overall\n flag doesn\u0027t even tell you whether import or export or maybe even both\n is supported.\n \n Ditch it.\n \n This patch just makes it defunct, subsequent patches will remove it\n from all the drivers.\n \n Note this changes the userspace visible errno in some situations from\n EOPNOTSUPP to ENOSYS. Userspace already needs to treat both as \"no\n prime support\", so this should break anything.\n \n v2: Improve commit message as suggested by Emil\n \n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-6-daniel.vetter@ffwll.ch\n\ncommit ae75f836804267f28048e389f6036a7044872112\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Fri Jun 14 22:35:20 2019 +0200\n\n drm/prime: Unconditionally set up the prime file private\n \n It\u0027s tiny, already embedded, and setup/teardown cost is trivial.\n \n Reviewed-by: Eric Anholt \u003ceric@anholt.net\u003e\n Reviewed-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-5-daniel.vetter@ffwll.ch\n\ncommit 805dc614d58a8fb069ed079005e591247df85246\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Thu Jun 20 14:46:15 2019 +0200\n\n drm/prime: Update docs\n \n Yes this is a bit a big patch, but since it\u0027s essentially a complete\n rewrite of all the prime docs I didn\u0027t see how to better split it up.\n \n Changes:\n - Consistently point to drm_gem_object_funcs as the preferred hooks,\n where applicable.\n \n - Document all the hooks in \u0026drm_driver that lacked kerneldoc.\n \n - Completely new overview section, which now also includes the cleaned\n up lifetime/reference counting subchapter. I also mentioned the weak\n references in there due to the lookup caches.\n \n - Completely rewritten helper intro section, highlight the\n import/export related functionality.\n \n - Polish for all the functions and more cross references.\n \n I also sprinkled a bunch of todos all over.\n \n Most important: 0 code changes in here. The cleanup motivated by\n reading and improving all this will follow later on.\n \n v2: Actually update the prime helper docs. Plus add a few FIXMEs that\n I won\u0027t address right away in subsequent cleanup patches.\n \n v3:\n - Split out the function moving. This patch is now exclusively\n documentation changes.\n - Typos and nits (Sam).\n \n v4: Polish suggestions from Noralf.\n \n Acked-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Acked-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Cc: Thomas Zimmermann \u003ctzimmermann@suse.de\u003e\n Cc: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Cc: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: Emil Velikov \u003cemil.l.velikov@gmail.com\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190620124615.24434-1-daniel.vetter@ffwll.ch\n\ncommit b283e92a2315f9368dda010c9633183147fe87e0\nAuthor: Daniel Vetter \u003cdaniel.vetter@ffwll.ch\u003e\nDate: Tue Jun 18 11:20:37 2019 +0200\n\n drm/prime: Shuffle functions.\n \n Reorder all the functions in drm_prime.[hc] into three groups: core,\n export helpers, import helpers.\n \n Not other changes beyond moving the functions and their unchanged\n kerneldoc around in here.\n \n Cc: Sam Ravnborg \u003csam@ravnborg.org\u003e\n Cc: Eric Anholt \u003ceric@anholt.net\u003e\n Cc: Emil Velikov \u003cemil.l.velikov@gmail.com\u003e\n Acked-by: Gerd Hoffmann \u003ckraxel@redhat.com\u003e\n Acked-by: Emil Velikov \u003cemil.velikov@collabora.com\u003e\n Acked-by: Noralf Trønnes \u003cnoralf@tronnes.org\u003e\n Signed-off-by: Daniel Vetter \u003cdaniel.vetter@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190618092038.17929-1-daniel.vetter@ffwll.ch\n\ncommit b8142e3a63dad1e2b7a71a539d73c02d4d0c1353\nAuthor: YueHaibing \u003cyuehaibing@huawei.com\u003e\nDate: Thu Jun 20 15:03:06 2019 +0000\n\n drm/sti: Remove duplicated include from sti_drv.c\n \n Remove duplicated include.\n \n Signed-off-by: YueHaibing \u003cyuehaibing@huawei.com\u003e\n Signed-off-by: Benjamin Gaignard \u003cbenjamin.gaignard@linaro.org\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190620150306.124839-1-yuehaibing@huawei.com\n\ncommit 12fdaf19e05b69992502eb9c97fc3c180f797564\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Fri Jun 21 09:07:29 2019 +0100\n\n drm/i915/execlists: Keep virtual context alive until after we kick\n \n The call to kick_siblings() dereferences the rq-\u003econtext, so we should\n not drop our local reference until afterwards!\n \n v2: Stick to setting ce.inflight\u003dNULL before kicking as this is what the\n other threads will check to see if the context is ready for takeover.\n \n Fixes: 22b7a426bbe1 (\"drm/i915/execlists: Preempt-to-busy\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190621080729.2652-1-chris@chris-wilson.co.uk\n\ncommit b32fa811156328aea5a3c2ff05cc096490382456\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Jun 20 19:37:05 2019 +0100\n\n drm/i915/gtt: Defer address space cleanup to an RCU worker\n \n Enable RCU protection of i915_address_space and its ppgtt superclasses,\n and defer its cleanup into a worker executed after an RCU grace period.\n \n In the future we will be able to use the RCU protection to reduce the\n locking around VM lookups, but the immediate benefit is being able to\n defer the release into a kworker (process context). This is required as\n we may need to sleep to reap the WC pages stashed away inside the ppgtt.\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d110934\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190620183705.31006-1-chris@chris-wilson.co.uk\n\ncommit 683d672c425aa29c0e74583ed28a0e011cc0bb43\nAuthor: José Roberto de Souza \u003cjose.souza@intel.com\u003e\nDate: Wed Jun 19 16:31:34 2019 -0700\n\n drm/i915/ehl/dsi: Enable AFE over PPI strap\n \n The other additional step in the DSI sequence for EHL.\n \n v2:\n - Using REG_BIT()(Matt)\n - Fixed commit message typo(Vandita)\n \n BSpec: 20597\n Cc: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Reviewed-by: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\n Reviewed-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619233134.20009-2-jose.souza@intel.com\n\ncommit 6a7bafe8fdb6019101ec90a5f5ddeea6f58d1158\nAuthor: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\nDate: Wed Jun 19 16:31:33 2019 -0700\n\n drm/i915/ehl/dsi: Set lane latency optimization for DW1\n \n EHL has 2 additional steps in the DSI sequence, this is one of then\n the lane latency optimization for DW1.\n \n BSpec: 20597\n Cc: Uma Shankar \u003cuma.shankar@intel.com\u003e\n Cc: Rodrigo Vivi \u003crodrigo.vivi@intel.com\u003e\n Cc: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Reviewed-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Signed-off-by: Vandita Kulkarni \u003cvandita.kulkarni@intel.com\u003e\n Signed-off-by: José Roberto de Souza \u003cjose.souza@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619233134.20009-1-jose.souza@intel.com\n\ncommit 60a0933bff57af52f99a17cbe9f0bc5d27771236\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Jun 20 11:24:32 2019 +0100\n\n drm/i915/selftests: Use request managed wakerefs\n \n Since commit 79ffac8599c4 (\"drm/i915: Invert the GEM wakeref\n hierarchy\"), the request creation itself took responsibility for\n managing the engine/GT wakerefs and so we can remove the redundant grabs\n in our selftests.\n \n References: 79ffac8599c4 (\"drm/i915: Invert the GEM wakeref hierarchy\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190620102432.31580-1-chris@chris-wilson.co.uk\n\ncommit cffa1eb8a22392a02dbc6a372094262f888fd3ed\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jun 19 21:35:04 2019 +0100\n\n drm/i915: Rings are always flushed\n \n Our intel_rings are always flushed as they are continually used to submit\n commands to the GPU, and so do not need to be flushed on unpinning. This\n avoids pulling in the flush_ggtt_writes locking into our context\n unpin, which we want to allow from atomic context (for simplicity).\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619203504.4220-1-chris@chris-wilson.co.uk\n\ncommit 136db994852a9b405ac1074de0e7a1c4c840b8ee\nAuthor: John Johansen \u003cjohn.johansen@canonical.com\u003e\nDate: Fri May 31 06:54:54 2019 -0700\n\n apparmor: increase left match history buffer size\n \n There have been cases reported where a history buffer size of 8 was\n not enough to resolve conflict overlaps. Increase the buffer to and\n get rid of the size element which is currently just storing the\n constant WB_HISTORY_SIZE.\n \n Signed-off-by: John Johansen \u003cjohn.johansen@canonical.com\u003e\n\ncommit 8ac2ca328ec9356f56d0dad3aa350d9600db951a\nAuthor: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\nDate: Fri Apr 5 15:34:58 2019 +0200\n\n apparmor: Switch to GFP_KERNEL where possible\n \n After removing preempt_disable() from get_buffers() it is possible to\n replace a few GFP_ATOMIC allocations with GFP_KERNEL.\n \n Replace GFP_ATOMIC allocations with GFP_KERNEL where the context looks\n to bee preepmtible.\n \n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: John Johansen \u003cjohn.johansen@canonical.com\u003e\n\ncommit df323337e507a0009d3db1ea25948d4c7f320d62\nAuthor: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\nDate: Fri May 3 16:12:21 2019 +0200\n\n apparmor: Use a memory pool instead per-CPU caches\n \n The get_buffers() macro may provide one or two buffers to the caller.\n Those buffers are pre-allocated on init for each CPU. By default it\n allocates\n 2* 2 * MAX_PATH * POSSIBLE_CPU\n \n which equals 64KiB on a system with 4 CPUs or 1MiB with 64 CPUs and so\n on.\n \n Replace the per-CPU buffers with a common memory pool which is shared\n across all CPUs. The pool grows on demand and never shrinks. The pool\n starts with two (UP) or four (SMP) elements. By using this pool it is\n possible to request a buffer and keeping preemption enabled which avoids\n the hack in profile_transition().\n \n It has been pointed out by Tetsuo Handa that GFP_KERNEL allocations for\n small amount of memory do not fail. In order not to have an endless\n retry, __GFP_RETRY_MAYFAIL is passed (so the memory allocation is not\n repeated until success) and retried once hoping that in the meantime a\n buffer has been returned to the pool. Since now NULL is possible all\n allocation paths check the buffer pointer and return -ENOMEM on failure.\n \n Signed-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\n Signed-off-by: John Johansen \u003cjohn.johansen@canonical.com\u003e\n\ncommit bf1d2ee7bc6215dd92427625a4c707227457a5db\nAuthor: Bharath Vedartham \u003clinux.bhar@gmail.com\u003e\nDate: Tue Apr 23 22:23:00 2019 +0530\n\n apparmor: Force type-casting of current-\u003ereal_cred\n \n This patch fixes the sparse warning:\n warning: cast removes address space \u0027\u003casn:4\u003e\u0027 of expression.\n \n Signed-off-by: Bharath Vedartham \u003clinux.bhar@gmail.com\u003e\n Signed-off-by: John Johansen \u003cjohn.johansen@canonical.com\u003e\n\ncommit 8ee36e048c98d4015804a23f884be2576f778a93\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Jun 20 15:20:52 2019 +0100\n\n drm/i915/execlists: Minimalistic timeslicing\n \n If we have multiple contexts of equal priority pending execution,\n activate a timer to demote the currently executing context in favour of\n the next in the queue when that timeslice expires. This enforces\n fairness between contexts (so long as they allow preemption -- forced\n preemption, in the future, will kick those who do not obey) and allows\n us to avoid userspace blocking forward progress with e.g. unbounded\n MI_SEMAPHORE_WAIT.\n \n For the starting point here, we use the jiffie as our timeslice so that\n we should be reasonably efficient wrt frequent CPU wakeups.\n \n Testcase: igt/gem_exec_scheduler/semaphore-resolve\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190620142052.19311-2-chris@chris-wilson.co.uk\n\ncommit 22b7a426bbe1ebe1520f92da4cd1617d1e1b5fc4\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Thu Jun 20 15:20:51 2019 +0100\n\n drm/i915/execlists: Preempt-to-busy\n \n When using a global seqno, we required a precise stop-the-workd event to\n handle preemption and unwind the global seqno counter. To accomplish\n this, we would preempt to a special out-of-band context and wait for the\n machine to report that it was idle. Given an idle machine, we could very\n precisely see which requests had completed and which we needed to feed\n back into the run queue.\n \n However, now that we have scrapped the global seqno, we no longer need\n to precisely unwind the global counter and only track requests by their\n per-context seqno. This allows us to loosely unwind inflight requests\n while scheduling a preemption, with the enormous caveat that the\n requests we put back on the run queue are still _inflight_ (until the\n preemption request is complete). This makes request tracking much more\n messy, as at any point then we can see a completed request that we\n believe is not currently scheduled for execution. We also have to be\n careful not to rewind RING_TAIL past RING_HEAD on preempting to the\n running context, and for this we use a semaphore to prevent completion\n of the request before continuing.\n \n To accomplish this feat, we change how we track requests scheduled to\n the HW. Instead of appending our requests onto a single list as we\n submit, we track each submission to ELSP as its own block. Then upon\n receiving the CS preemption event, we promote the pending block to the\n inflight block (discarding what was previously being tracked). As normal\n CS completion events arrive, we then remove stale entries from the\n inflight tracker.\n \n v2: Be a tinge paranoid and ensure we flush the write into the HWS page\n for the GPU semaphore to pick in a timely fashion.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190620142052.19311-1-chris@chris-wilson.co.uk\n\ncommit 9e138ea1bdb1d1c1f1c0b74f022f886196abf625\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Wed Jun 19 18:00:21 2019 -0700\n\n drm/i915/gvt: decouple check_vgpu() from uncore_init()\n \n With multiple uncore to initialize (GT vs Display), it makes little\n sense to have the vgpu_check inside uncore_init(). We also have\n a catch-22 scenario where the uncore is required to read the vgpu\n capabilities while the vgpu capabilities are required to decide if\n we need to initialize forcewake support. To remove this circular\n dependency, we can perform the required MMIO access by mmapping just\n the vgtif shared page in mmio space and use raw accessors.\n \n v2: rename check_vgpu to detect_vgpu (Chris)\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Zhenyu Wang \u003czhenyuw@linux.intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190620010021.20637-7-daniele.ceraolospurio@intel.com\n\ncommit f833cdb06bf7cea6827d67b73cb24da335b70266\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Wed Jun 19 18:00:20 2019 -0700\n\n drm/i915: dynamically allocate forcewake domains\n \n We\u0027d like to introduce a display uncore with no forcewake domains, so\n let\u0027s avoid wasting memory and be ready to allocate only what we need.\n Even without multiple uncore, we still don\u0027t need all the domains on all\n gens.\n \n v2: avoid hidden control flow, improve checks (Tvrtko), fix IVB special\n case, add failure injection point\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190620010021.20637-6-daniele.ceraolospurio@intel.com\n\ncommit 2e81bc61d4e3ee4592b2a1d341696fe4fe3fc5eb\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Wed Jun 19 18:00:19 2019 -0700\n\n drm/i915: skip forcewake actions on forcewake-less uncore\n \n We always call some of the setup/cleanup functions for forcewake, even\n if the feature is not actually available. Skipping these operations if\n forcewake is not available saves us some operations on older gens and\n prepares us for having a forcewake-less display uncore.\n \n v2: do not make suspend/resume functions forcewake-specific (Chris,\n Tvrtko), use GEM_BUG_ON in internal forcewake-only functions (Tvrtko)\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@linux.intel.com\u003e\n Reviewed-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190620010021.20637-5-daniele.ceraolospurio@intel.com\n\ncommit 013857580bbece05a89d6a323799960ffaefdb8c\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Wed Jun 19 18:00:18 2019 -0700\n\n drm/i915: kill uncore_to_i915\n \n Let\u0027s get rid of it before it proliferates, since with split GT/Display\n uncores the container_of won\u0027t work anymore.\n \n I\u0027ve kept the rpm pointer as well to minimize the pointer chasing in the\n MMIO accessors.\n \n v2: swap parameter order for intel_uncore_init_early (Tvrtko)\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190620010021.20637-4-daniele.ceraolospurio@intel.com\n\ncommit 19e0a8d454c3e581a877071a87b349a8bf6b9a7c\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Wed Jun 19 18:00:17 2019 -0700\n\n drm/i915: kill uncore_sanitize\n \n uncore_sanitize performs no action on the uncore structure and just\n calls intel_sanitize_gt_powersave, so we can just call the latter\n directly.\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Reviewed-by: Jani Nikula \u003cjani.nikula@intel.com\u003e\n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190620010021.20637-3-daniele.ceraolospurio@intel.com\n\ncommit ccb2aceaaa5f9267ef7b485b41ae9be3f04b50d3\nAuthor: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\nDate: Wed Jun 19 18:00:16 2019 -0700\n\n drm/i915: use vfuncs for reg_read/write_fw_domains\n \n Instead of going through the if-else chain every time, let\u0027s save the\n function in the uncore structure. Note that the new functions are\n purposely not used from the reg read/write functions to keep the\n inlining there.\n \n While at it, use the new macro to call the old ones to clean the code a\n bit.\n \n v2: Rename macros for no-forcewake function assignment (Tvrtko)\n \n Signed-off-by: Daniele Ceraolo Spurio \u003cdaniele.ceraolospurio@intel.com\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@linux.intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Signed-off-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190620010021.20637-2-daniele.ceraolospurio@intel.com\n\ncommit 09c5ab384f6fb30f834a5777888b4486dd7f015d\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jun 19 18:01:35 2019 +0100\n\n drm/i915: Keep rings pinned while the context is active\n \n Remember to keep the rings pinned as well as the context image until the\n GPU is no longer active.\n \n v2: Introduce a ring-\u003epin_count primarily to hide the\n mock_ring that doesn\u0027t fit into the normal GGTT vma picture.\n \n v3: Order is important in teardown, ringbuffer submission needs to drop\n the pin count on the engine-\u003ekernel_context before it can gleefully free\n its ring.\n \n Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id\u003d110946\n Fixes: ce476c80b8bf (\"drm/i915: Keep contexts pinned until after the next kernel context switch\")\n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619170135.15281-1-chris@chris-wilson.co.uk\n\ncommit bdeb18dbcf8227e8f4b46db3f53a4155409e68cd\nAuthor: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\nDate: Tue Jun 18 10:51:31 2019 -0700\n\n drm/i915/ehl: Allow combo PHY A to drive a third external display\n \n EHL has a mux on combo PHY A that allows it to be driven either by an\n internal display (DDI-A or DSI DPHY) or by an external display (DDI-D).\n This is a motherboard design decision that can not be changed on the\n fly. Unfortunately there are no strap registers that allow us to detect\n the board configuration directly, so let\u0027s use the VBT to try to figure\n it out and program the mux accordingly.\n \n For now if we run across a broken VBT that tries to claim that PHY A\n is attached to both internal and external displays at the same time,\n we\u0027ll resolve the conflict in favor of the internal display. To help\n debug these kind of bad VBT\u0027s, let\u0027s also add a quick DRM_DEBUG message\n during child device parsing so that it\u0027s easier to understand these\n cases if they show up in bug reports.\n \n v2:\n - Confirmed that VBT\u0027s dvo port refers to the DDI and not the PHY.\n Thus we can check more explicitly for (ddi_d \u0026\u0026 !(ddi_a || dsi)). If\n a bad VBT contradicts itself, let internal display win. (Ville)\n \n v3:\n - Switch condition from !IS_ICELAKE to IS_ELKHARTLAKE. Although the\n convention is usually to assume that future platforms will inherit\n all current platform behavior, this feels more like a one-platform\n quirk. (Ville)\n - Update commit message to describe what we do if/when we encounter\n broken VBT\u0027s, and note that the new debug print during child device\n parsing is intentional.\n \n Cc: Clint Taylor \u003cClinton.A.Taylor@intel.com\u003e\n Cc: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Signed-off-by: Matt Roper \u003cmatthew.d.roper@intel.com\u003e\n Reviewed-by: Ville Syrjälä \u003cville.syrjala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190618175131.9139-1-matthew.d.roper@intel.com\n\ncommit b87b6c0dfce7d6c9df7f2c4c0ebacfc71bf58b89\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Tue Jun 18 08:41:31 2019 +0100\n\n drm/i915: Flush the execution-callbacks on retiring\n \n In the unlikely case the request completes while we regard it as not even\n executing on the GPU (see the next patch!), we have to flush any pending\n execution callbacks at retirement and ensure that we do not add any\n more.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Reviewed-by: Mika Kuoppala \u003cmika.kuoppala@linux.intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190618074153.16055-4-chris@chris-wilson.co.uk\n\ncommit ce94bef9354e15b99a329e7c5a451ea0ffd59fb1\nAuthor: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\nDate: Wed Jun 19 12:23:37 2019 +0100\n\n drm/i915: Signal fence completion from i915_request_wait\n \n With the upcoming change to automanaged i915_active, the intent is that\n whenever we wait on the set of active fences, they are signaled and\n collected. The requirement is that all successful returns from\n i915_request_wait() signal the fence, so fixup the one remaining path\n where we may return before the interrupt has been run.\n \n Signed-off-by: Chris Wilson \u003cchris@chris-wilson.co.uk\u003e\n Cc: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Reviewed-by: Tvrtko Ursulin \u003ctvrtko.ursulin@intel.com\u003e\n Link: https://patchwork.freedesktop.org/patch/msgid/20190619112341.9082-3-chris@chris-wilson.co.uk\n\ncommit 3bda7f0ae0f786b202c0c5124c208be067acca30\nAuthor: Mark Cave-Ayland \u003cmark.cave-ayland@ilande.co.uk\u003e\nDate: Fri Jun 14 19:57:45 2019 +0100\n\n KVM: PPC: Book3S PR: Fix software breakpoints\n \n QEMU\u0027s kvm_handle_debug() function identifies software breakpoints by checking\n for a value of 0 in kvm_debug_exit_arch\u0027s status field. Since this field isn\u0027t\n explicitly set to 0 when the software breakpoint instruction is detected, any\n previous non-zero value present causes a hang in QEMU as it tries to process\n the breakpoint instruction incorrectly as a hardware breakpoint.\n \n Ensure that the kvm_debug_exit_arch status field is set to 0 when the software\n breakpoint instruction is detected (similar to the existing logic in booke.c\n and e500_emulate.c) to restore software breakpoint functionality under Book3S\n PR.\n \n Signed-off-by: Mark Cave-Ayland \u003cmark.cave-ayland@ilande.co.uk\u003e\n Reviewed-by: Fabiano Rosas \u003cfarosas@linux.ibm.com\u003e\n Signed-off-by: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\n\ncommit 0e84ec96b561d298bc9a22d58e7a4730bc408aca\nAuthor: Fabiano Rosas \u003cfarosas@linux.ibm.com\u003e\nDate: Thu May 30 14:10:14 2019 -0300\n\n KVM: PPC: Remove leftover comment from emulate_loadstore.c\n \n Commit 4eeb85568e56 (\"KVM: PPC: Remove mmio_vsx_tx_sx_enabled in KVM\n MMIO emulation\") removed the mmio_vsx_tx_sx_enabled field, but its\n documentation was left behind. Remove the superfluous comment.\n \n Signed-off-by: Fabiano Rosas \u003cfarosas@linux.ibm.com\u003e\n Signed-off-by: Paul Mackerras \u003cpaulus@ozlabs.org\u003e\n\ncommit 16009db47c51d4ef61270f22a2f1bb385233c34f\nAuthor: Peter Chen \u003cpeter.chen@nxp.com\u003e\nDate: Tue May 21 17:23:50 2019 +0800\n\n usb: chipidea: udc: workaround for endpoint conflict issue\n \n An endpoint conflict occurs when the USB is working in device mode\n during an isochronous communication. When the endpointA IN direction\n is an isochronous IN endpoint, and the host sends an IN token to\n endpointA on another device, then the OUT transaction may be missed\n regardless the OUT endpoint number. Generally, this occurs when the\n device is connected to the host through a hub and other devices are\n connected to the same hub.\n \n The affected OUT endpoint can be either control, bulk, isochronous, or\n an interrupt endpoint. After the OUT endpoint is primed, if an IN token\n to the same endpoint number on another device is received, then the OUT\n endpoint may be unprimed (cannot be detected by software), which causes\n this endpoint to no longer respond to the host OUT token, and thus, no\n corresponding interrupt occurs.\n \n There is no good workaround for this issue, the only thing the software\n could do is numbering isochronous IN from the highest endpoint since we\n have observed most of device number endpoint from the lowest.\n \n Cc: \u003cstable@vger.kernel.org\u003e #v3.14+\n Cc: Fabio Estevam \u003cfestevam@gmail.com\u003e\n Cc: Greg KH \u003cgregkh@linuxfoundation.org\u003e\n Cc: Sergei Shtylyov \u003csergei.shtylyov@cogentembedded.com\u003e\n Cc: Jun Li \u003cjun.li@nxp.com\u003e\n Signed-off-by: Peter Chen \u003cpeter.chen@nxp.com\u003e\n\ncommit 21560067fb1f5e87abedc3ecec5d46f67ac0c019\nAuthor: Rasmus Villemoes \u003crasmus.villemoes@prevas.dk\u003e\nDate: Mon May 13 11:15:01 2019 +0000\n\n soc: fsl: qe: fold qe_get_num_of_snums into qe_snums_init\n \n The comment \"No QE ever has fewer than 28 SNUMs\" is false; e.g. the\n MPC8309 has 14. The code path returning -EINVAL is also a recipe for\n instant disaster, since the caller (qe_snums_init) uncritically\n assigns the return value to the unsigned qe_num_of_snum, and would\n thus proceed to attempt to copy 4GB from snum_init_46[] to the snum[]\n array.\n \n So fold the handling of the legacy fsl,qe-num-snums into\n qe_snums_init, and make sure we do not end up using the snum_init_46\n array in cases other than the two where we know it makes sense.\n \n Reviewed-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Reviewed-by: Qiang Zhao \u003cqiang.zhao@nxp.com\u003e\n Signed-off-by: Rasmus Villemoes \u003crasmus.villemoes@prevas.dk\u003e\n Signed-off-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n\ncommit 5cfca8911ddd59150ba53cff16956eabe1b3dcac\nAuthor: Rasmus Villemoes \u003crasmus.villemoes@prevas.dk\u003e\nDate: Mon May 13 11:15:00 2019 +0000\n\n soc: fsl: qe: support fsl,qe-snums property\n \n Add driver support for the newly introduced fsl,qe-snums property.\n \n Conveniently, of_property_read_variable_u8_array does exactly what we\n need: If the property fsl,qe-snums is found (and has an allowed size),\n the array of values get copied to snums, and the return value is the\n number of snums - we cannot assign directly to num_of_snums, since we\n need to check whether the return value is negative.\n \n Reviewed-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Reviewed-by: Qiang Zhao \u003cqiang.zhao@nxp.com\u003e\n Signed-off-by: Rasmus Villemoes \u003crasmus.villemoes@prevas.dk\u003e\n Signed-off-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n\ncommit 8f762fe5819f8561343cf02e234789744ed68a98\nAuthor: Rasmus Villemoes \u003crasmus.villemoes@prevas.dk\u003e\nDate: Mon May 13 11:14:58 2019 +0000\n\n dt-bindings: soc: fsl: qe: document new fsl,qe-snums binding\n \n Reading table 4-30, and its footnotes, of the QUICC Engine Block\n Reference Manual shows that the set of snum _values_ is not\n necessarily just a function of the _number_ of snums, as given in the\n fsl,qe-num-snums property.\n \n As an alternative, to make it easier to add support for other variants\n of the QUICC engine IP, this introduces a new binding fsl,qe-snums,\n which automatically encodes both the number of snums and the actual\n values to use.\n \n Signed-off-by: Rasmus Villemoes \u003crasmus.villemoes@prevas.dk\u003e\n Reviewed-by: Rob Herring \u003crobh@kernel.org\u003e\n Signed-off-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n\ncommit d7fc5963664198d25ca525f120f72ff053ba334f\nAuthor: Rasmus Villemoes \u003crasmus.villemoes@prevas.dk\u003e\nDate: Mon May 13 11:14:57 2019 +0000\n\n soc: fsl: qe: introduce qe_get_device_node helper\n \n The \u0027try of_find_compatible_node(NULL, NULL, \"fsl,qe\"), fall back to\n of_find_node_by_type(NULL, \"qe\")\u0027 pattern is repeated five\n times. Factor it into a common helper.\n \n Reviewed-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Reviewed-by: Qiang Zhao \u003cqiang.zhao@nxp.com\u003e\n Signed-off-by: Rasmus Villemoes \u003crasmus.villemoes@prevas.dk\u003e\n Signed-off-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n\ncommit 875f2aabfcfa321dd39298849e3833b44ee54b4d\nAuthor: Rasmus Villemoes \u003crasmus.villemoes@prevas.dk\u003e\nDate: Mon May 13 11:14:56 2019 +0000\n\n soc: fsl: qe: reduce static memory footprint by 1.7K\n \n The current array of struct qe_snum use 256*4 bytes for just keeping\n track of the free/used state of each index, and the struct layout\n means there\u0027s another 768 bytes of padding. If we just unzip that\n structure, the array of snum values just use 256 bytes, while the\n free/inuse state can be tracked in a 32 byte bitmap.\n \n So this reduces the .data footprint by 1760 bytes. It also serves as\n preparation for introducing another DT binding for specifying the snum\n values.\n \n Reviewed-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Reviewed-by: Qiang Zhao \u003cqiang.zhao@nxp.com\u003e\n Signed-off-by: Rasmus Villemoes \u003crasmus.villemoes@prevas.dk\u003e\n Signed-off-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n\ncommit f03de74d92aca6a5399c088074550a164ae44eec\nAuthor: Rasmus Villemoes \u003crasmus.villemoes@prevas.dk\u003e\nDate: Mon May 13 11:14:54 2019 +0000\n\n soc: fsl: qe: drop useless static qualifier\n \n The local variable snum_init has no reason to have static storage duration.\n \n Reviewed-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Reviewed-by: Qiang Zhao \u003cqiang.zhao@nxp.com\u003e\n Signed-off-by: Rasmus Villemoes \u003crasmus.villemoes@prevas.dk\u003e\n Signed-off-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n\ncommit 4cb9f043447ebb3948aad78b5f9d12a0a7fac6ae\nAuthor: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nDate: Wed May 15 12:34:21 2019 +0300\n\n thermal: thermal_mmio: remove some dead code\n \n The platform_get_resource() function doesn\u0027t return error pointers, it\n returns NULL. The way this is normally done, is that we pass the NULL\n resource to devm_ioremap_resource() and then check for errors from that.\n See the comment in front of devm_ioremap_resource() for more details.\n \n Signed-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\n Acked-by: Talel Shenhar \u003ctalel@amazon.com\u003e\n Signed-off-by: Eduardo Valentin \u003cedubezval@gmail.com\u003e\n\ncommit 5677eab3c86d47a40105295719e24d419f8314c1\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Tue May 21 09:56:24 2019 +0100\n\n soc: fsl: fix spelling mistake \"Firmaware\" -\u003e \"Firmware\"\n \n There is a spelling mistake in a pr_err message. Fix it.\n \n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Li Yang \u003cleoyang.li@nxp.com\u003e\n\ncommit 1c41f624e7582311f7ed24d0c7852dcd24f7bb8e\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Mon May 6 16:47:41 2019 -0700\n\n xtensa: implement initialize_cacheattr for MPU cores\n \n Use CONFIG_MEMMAP_CACHEATTR to initialize MPU as described in the Xtensa\n LSP RM document. Coalesce adjacent regions with the same cacheattr.\n Update Kconfig help text.\n \n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit a444d1aabe6b0cf5bc0814f3029001c0089efd41\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Thu Dec 20 17:18:12 2018 -0800\n\n xtensa: add exclusive atomics support\n \n Implement atomic primitives using exclusive access opcodes available in\n the recent xtensa cores.\n Since l32ex/s32ex don\u0027t have any memory ordering guarantees don\u0027t define\n __smp_mb__before_atomic/__smp_mb__after_atomic to make them use memw.\n \n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit 2656d1569d93d12945cd23f40d7885df9ac13c73\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Mon May 6 12:02:26 2019 -0700\n\n xtensa: clean up inline assembly in futex.h\n \n Replace numeric inline assembly parameters with named parameters.\n Drop unused parameters from the futex_atomic_cmpxchg_inatomic.\n Use new temporary variable to hold target address in the fixup code in\n futex_atomic_cmpxchg_inatomic.\n \n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit 2061da36848d0372eb61d4612ed2ee91e1d6bbd4\nAuthor: Chengguang Xu \u003ccgxu519@gmail.com\u003e\nDate: Sun May 5 22:26:07 2019 +0800\n\n xtensa: set proper error code for simdisk_setup()\n \n Change error code to -ENOMEM from -EIO because\n the error is related to memory allocating.\n \n Signed-off-by: Chengguang Xu \u003ccgxu519@gmail.com\u003e\n Message-Id: \u003c1557066367-4783-2-git-send-email-cgxu519@gmail.com\u003e\n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit 1e644f4b0b4ff8236c63d827e8b849df1dc4b3a8\nAuthor: Chengguang Xu \u003ccgxu519@gmail.com\u003e\nDate: Sun May 5 22:26:06 2019 +0800\n\n xtensa: fix incorrect fd close in error case of simdisk_setup()\n \n dev-\u003efd is opened in attach operation, so should not\n close dev-\u003efd in error case of simdisk_setup().\n \n Signed-off-by: Chengguang Xu \u003ccgxu519@gmail.com\u003e\n Message-Id: \u003c1557066367-4783-1-git-send-email-cgxu519@gmail.com\u003e\n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit 17796b9ea445a2c6f1903c03261d2bab10e2bef3\nAuthor: Pankaj Bansal \u003cpankaj.bansal@nxp.com\u003e\nDate: Thu Feb 28 14:07:39 2019 +0000\n\n mux: mmio: add generic regmap bitfield-based multiplexer\n \n Generic register bitfield-based multiplexer that controls the multiplexer\n producer defined under a parent node.\n The driver corresponding to parent node provides register read/write\n capabilities.\n \n Signed-off-by: Pankaj Bansal \u003cpankaj.bansal@nxp.com\u003e\n Signed-off-by: Peter Rosin \u003cpeda@axentia.se\u003e\n\ncommit 8f445d9e8d6f62f66bf4aa4a5febc949c402ea00\nAuthor: Pankaj Bansal \u003cpankaj.bansal@nxp.com\u003e\nDate: Thu Feb 28 14:07:38 2019 +0000\n\n dt-bindings: add register based devices\u0027 mux controller DT bindings\n \n This adds device tree binding documentation for generic register based\n multiplexer controlled by a bitfields in a parent device\u0027s register range.\n \n since MMIO mux is a special case of generic register based mux, the\n MMIO mux bindings have been subsumed in these bindings.\n \n Signed-off-by: Pankaj Bansal \u003cpankaj.bansal@nxp.com\u003e\n Signed-off-by: Peter Rosin \u003cpeda@axentia.se\u003e\n\ncommit 058c4f342582362c75dd5e162dc4ff73a392ffad\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Tue Apr 16 15:42:18 2019 +0100\n\n apparmor: fix spelling mistake \"immutible\" -\u003e \"immutable\"\n \n There is a spelling mistake in an information message string, fix it.\n \n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: John Johansen \u003cjohn.johansen@canonical.com\u003e\n\ncommit 145a0ef21c8e944957f58e2c8ffcd8a10f46266a\nAuthor: John Johansen \u003cjohn.johansen@canonical.com\u003e\nDate: Sat Mar 9 16:58:10 2019 -0800\n\n apparmor: fix blob compression when ns is forced on a policy load\n \n When blob compression is turned on, if the policy namespace is forced\n onto a policy load, the policy load will fail as the namespace name\n being referenced is inside the compressed policy blob, resulting in\n invalid or names that are too long. So duplicate the name before the\n blob is compressed.\n \n Fixes: 876dd866c084 (\"apparmor: Initial implementation of raw policy blob compression\")\n Signed-off-by: John Johansen \u003cjohn.johansen@canonical.com\u003e\n\ncommit fe166a9f2868839a1e2f7bd950164d05e86eb154\nAuthor: John Johansen \u003cjohn.johansen@canonical.com\u003e\nDate: Mon Feb 11 21:56:46 2019 -0800\n\n apparmor: fix missing ZLIB defines\n \n On configs where ZLIB is not already selected we are getting\n \n undefined reference to `zlib_deflateInit2\u0027\n undefined reference to `zlib_deflate\u0027\n undefined reference to `zlib_deflateEnd\u0027\n \n For now just select the necessary ZLIB configs.\n \n Fixes: 876dd866c084 (\"apparmor: Initial implementation of raw policy blob compression\")\n Signed-off-by: John Johansen \u003cjohn.johansen@canonical.com\u003e\n\ncommit 6a59d9243d349ae598e8ea74c36aa8e31705071a\nAuthor: John Johansen \u003cjohn.johansen@canonical.com\u003e\nDate: Fri Feb 8 17:14:35 2019 -0800\n\n apparmor: fix blob compression build failure on ppc\n \n security/apparmor/policy_unpack.c: In function \u0027deflate_compress\u0027:\n security/apparmor/policy_unpack.c:1064:4: error: implicit declaration of function \u0027vfree\u0027; did you mean \u0027kfree\u0027? [-Werror\u003dimplicit-function-declaration]\n vfree(stgbuf);\n ^~~~~\n kfree\n \n Fixes: 876dd866c084 (\"apparmor: Initial implementation of raw policy blob compression\")\n Signed-off-by: John Johansen \u003cjohn.johansen@canonical.com\u003e\n\ncommit 63c16c3a7608558a8e5ced96b8b6b06c490fd513\nAuthor: Chris Coulson \u003cchris.coulson@canonical.com\u003e\nDate: Wed Jan 23 19:17:09 2019 +0000\n\n apparmor: Initial implementation of raw policy blob compression\n \n This adds an initial implementation of raw policy blob compression,\n using deflate. Compression level can be controlled via a new sysctl,\n \"apparmor.rawdata_compression_level\", which can be set to a value\n between 0 (no compression) and 9 (highest compression).\n \n Signed-off-by: Chris Coulson \u003cchris.coulson@canonical.com\u003e\n Signed-off-by: John Johansen \u003cjohn.johansen@canonical.com\u003e\n\ncommit 812cd88749e098ad976b6a1c149984dfd6c0f195\nAuthor: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\nDate: Mon Feb 11 12:03:56 2019 -0600\n\n firewire: mark expected switch fall-throughs\n \n In preparation to enabling -Wimplicit-fallthrough, mark switch\n cases where we are expecting to fall through.\n \n This patch fixes the following warnings:\n \n drivers/firewire/core-device.c: In function ‘set_broadcast_channel’:\n drivers/firewire/core-device.c:969:7: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n if (data \u0026 cpu_to_be32(1 \u003c\u003c 31)) {\n ^\n drivers/firewire/core-device.c:974:3: note: here\n case RCODE_ADDRESS_ERROR:\n ^~~~\n drivers/firewire/core-iso.c: In function ‘manage_channel’:\n drivers/firewire/core-iso.c:308:7: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n if ((data[0] \u0026 bit) \u003d\u003d (data[1] \u0026 bit))\n ^\n drivers/firewire/core-iso.c:312:3: note: here\n default:\n ^~~~~~~\n drivers/firewire/core-topology.c: In function ‘count_ports’:\n drivers/firewire/core-topology.c:69:23: warning: this statement may fall through [-Wimplicit-fallthrough\u003d]\n (*child_port_count)++;\n ~~~~~~~~~~~~~~~~~~~^~\n drivers/firewire/core-topology.c:70:3: note: here\n case SELFID_PORT_PARENT:\n ^~~~\n \n Warning level 3 was used: -Wimplicit-fallthrough\u003d3\n \n Notice that in some cases, the code comment is modified in\n accordance with what GCC is expecting to find.\n \n This patch is part of the ongoing efforts to enable\n -Wimplicit-fallthrough.\n \n Signed-off-by: Gustavo A. R. Silva \u003cgustavo@embeddedor.com\u003e\n Cc: Kees Cook \u003ckeescook@chromium.org\u003e\n Cc: Mathieu Malaterre \u003cmalat@debian.org\u003e\n Signed-off-by: Stefan Richter \u003cstefanr@s5r6.in-berlin.de\u003e (reworded a comment)\n\ncommit 34a783a07a6f6c4ac64a72aff8f68b952c569269\nAuthor: Corentin Labbe \u003cclabbe@baylibre.com\u003e\nDate: Wed Jan 23 09:49:18 2019 +0000\n\n xtensa: rename BUILTIN_DTB to BUILTIN_DTB_SOURCE\n \n When building some xtensa config, I hit the following warning:\n drivers/staging/mt7621-dts/Kconfig:4:warning: \u0027BUILTIN_DTB\u0027 has wrong type.\n \u0027select\u0027 only accept arguments of bool and tristate type\n \n It is due to some arch use BUILTIN_DTB as a flag for the need to builtin\n dtb but xtensa use it as a string for which dtb to bulltin.\n But for this (which dtb to build), it is better to use\n BUILTIN_DTB_SOURCE like other arch do.\n \n Signed-off-by: Corentin Labbe \u003cclabbe@baylibre.com\u003e\n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit 16be6fd6a52b6638f7d8ff2dfc663cb9d60688d6\nAuthor: Corentin Labbe \u003cclabbe@baylibre.com\u003e\nDate: Fri Jan 18 13:45:27 2019 +0000\n\n xtensa: Fix typo use space\u003d\u003euser space\n \n This patch fix a simple typo.\n \n Signed-off-by: Corentin Labbe \u003cclabbe@baylibre.com\u003e\n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit 8cd4105fd234d91659a1242e4fa6641d723a18f7\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Thu Jan 24 14:06:49 2019 -0800\n\n drivers/irqchip: xtensa-mx: fix mask and unmask\n \n xtensa_irq_mask and xtensa_irq_unmask don\u0027t do the right thing when\n called for the first two external IRQs. Treat these IRQs as per-CPU\n IRQs.\n \n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit f2b9dea93c0218dc2da0f048a9c455784b7d272b\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Thu Jan 24 16:51:28 2019 -0800\n\n drivers/irqchip: xtensa: add warning to irq_retrigger\n \n XEA2 and MX PIC can only retrigger software interrupts. Issue a warning\n if an interrupt of any other type is retriggered.\n \n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit 18031f71ea014b6b32499fd9f390006dfe601063\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Sat Jan 19 00:26:48 2019 -0800\n\n xtensa: SMP: mark each possible CPU as present\n \n Otherwise it is impossible to enable CPUs after booting with \u0027maxcpus\u0027\n parameter.\n \n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit a169c7463e5171f23e88a3d146bf5b0377aede9d\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Thu Jan 24 17:16:11 2019 -0800\n\n xtensa: smp_lx200_defconfig: fix vectors clash\n \n Secondary CPU reset vector overlaps part of the double exception handler\n code, resulting in weird crashes and hangups when running user code.\n Move exception vectors one page up so that they don\u0027t clash with the\n secondary CPU reset vector.\n \n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit da26d363189aefec04360d20af7d58e54f5b81a3\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Fri Dec 21 08:26:20 2018 -0800\n\n xtensa: SMP: fix secondary CPU initialization\n \n - add missing memory barriers to the secondary CPU synchronization spin\n loop; add comment to the matching memory barrier in the boot_secondary\n function;\n - use READ_ONCE/WRITE_ONCE to access cpu_start_ccount instead of\n reading/writing it directly;\n - re-initialize cpu_running every time before starting secondary CPU to\n flush possible previous CPU startup results.\n \n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit d81a6df30db24afb4252f9010c86a087185bc667\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Mon Nov 26 16:30:51 2018 -0800\n\n xtensa: don\u0027t clear cpenable unconditionally on release\n \n Clearing cpenable special register for a task without changing\n coprocessor owner for the coprocessors that were enabled will result in\n coprocessor context flush and immediate reload at the next attempt to\n access this coprocessor if it happens before the context switch.\n Avoid it by only clearing cpenable special register if\n coprocessor_release_all is called for the current task.\n \n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit e836b0528f187660feaa786a22928e7c4433adb3\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Sun Nov 25 23:32:28 2018 -0800\n\n xtensa: simplify coprocessor.S\n \n Use addresses instead of offsets and drop unneeded offset -\u003e address\n calculations. Don\u0027t generate any code for undefined coprocessors.\n \n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit e22864e5a665b5c43facd85350bb433a1d27c86b\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Tue Nov 27 16:27:47 2018 -0800\n\n xtensa: clean up WSR*/RSR*/get_sr/set_sr\n \n WSR and RSR are too generic and collide with other macro definitions in\n the kernel causing warnings in allmodconfig builds. Drop WSR and RSR\n macros and WSR_* and RSR_* variants. Change get_sr and set_sr to\n xtensa_get_sr and xtensa_set_sr. Fix up users.\n \n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit 7c17225f2059b48d0678ad278580b8247b7d6ba8\nAuthor: Firoz Khan \u003cfiroz.khan@linaro.org\u003e\nDate: Tue Nov 13 15:49:30 2018 +0530\n\n xtensa: generate uapi header and syscall table header files\n \n System call table generation script must be run to gener-\n ate unistd_32.h and syscall_table.h files. This patch will\n have changes which will invokes the script.\n \n This patch will generate unistd_32.h and syscall_table.h\n files by the syscall table generation script invoked by\n xtensa/Makefile and the generated files against the removed\n files must be identical.\n \n The generated uapi header file will be included in uapi/-\n asm/unistd.h and generated system call table header file\n will be included by kernel/syscall.c file.\n \n Signed-off-by: Firoz Khan \u003cfiroz.khan@linaro.org\u003e\n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit b4130fdef9a76387e8a525157b90f0154b14cfa4\nAuthor: Firoz Khan \u003cfiroz.khan@linaro.org\u003e\nDate: Tue Nov 13 15:49:29 2018 +0530\n\n xtensa: add system call table generation support\n \n The system call tables are in different format in all\n architecture and it will be difficult to manually add,\n modify or delete the syscall table entries in the res-\n pective files. To make it easy by keeping a script and\n which will generate the uapi header and syscall table\n file. This change will also help to unify the implemen-\n tation across all architectures.\n \n The system call table generation script is added in\n kernel/syscalls directory which contain the scripts to\n generate both uapi header file and system call table\n files. The syscall.tbl will be input for the scripts.\n \n syscall.tbl contains the list of available system calls\n along with system call number and corresponding entry\n point. Add a new system call in this architecture will\n be possible by adding new entry in the syscall.tbl file.\n \n Adding a new table entry consisting of:\n - System call number.\n - ABI.\n - System call name.\n - Entry point name.\n \n syscallhdr.sh and syscalltbl.sh will generate uapi header\n unistd_32.h and syscall_table.h files respectively. Both\n .sh files will parse the content syscall.tbl to generate\n the header and table files. unistd_32.h will be included\n by uapi/asm/unistd.h and syscall_table.h is included by\n kernel/syscall.c - the real system call table.\n \n ARM, s390 and x86 architecuture does have similar support.\n I leverage their implementation to come up with a generic\n solution.\n \n Signed-off-by: Firoz Khan \u003cfiroz.khan@linaro.org\u003e\n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit 3db1569a17dde05be65365e0c6b2a74c063ddd60\nAuthor: Firoz Khan \u003cfiroz.khan@linaro.org\u003e\nDate: Tue Nov 13 15:49:28 2018 +0530\n\n xtensa: add __NR_syscalls along with __NR_syscall_count\n \n __NR_syscall_count macro holds the number of system call\n exist in xtensa architecture. We have to change the value\n of __NR_syscall_count, if we add or delete a system call.\n \n One of the patch in this patch series has a script which\n will generate a uapi header based on syscall.tbl file.\n The syscall.tbl file contains the total number of system\n calls information. So we have two option to update __NR-\n _syscall_count value.\n \n 1. Update __NR_syscall_count in asm/unistd.h manually by\n counting the no.of system calls. No need to update __NR-\n _syscall_count until we either add a new system call or\n delete existing system call.\n \n 2. We can keep this feature it above mentioned script,\n that will count the number of syscalls and keep it in\n a generated file. In this case we don\u0027t need to expli-\n citly update __NR_syscall_count in asm/unistd.h file.\n \n The 2nd option will be the recommended one. For that, I\n added the __NR_syscalls macro in uapi/asm/unistd.h.\n The macro __NR_syscalls also added for making the name\n convention same across all architecture. While __NR_syscalls\n isn\u0027t strictly part of the uapi, having it as part of the\n generated header to simplifies the implementation. We also\n need to enclose this macro with #ifdef __KERNEL__ to avoid\n side effects.\n \n Signed-off-by: Firoz Khan \u003cfiroz.khan@linaro.org\u003e\n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n [Max: Drop __NR_syscall_count completely, use __NR_syscalls instead]\n\ncommit 91eaf0f3c78a55189b2daac7c5a31b19280fe3f0\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Mon Nov 5 15:52:25 2018 -0800\n\n Revert \"xtensa: xtfpga: implement platform DMA address helpers\"\n \n This reverts commit 21ec04090f59647e69fb775ae9d447fcc778701e.\n \n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit 3bd67b37e35069b43cc5de8d138727360c503e8f\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Thu Sep 27 17:17:57 2018 +0000\n\n kdb: print real address of pointers instead of hashed addresses\n \n Since commit ad67b74d2469 (\"printk: hash addresses printed with %p\"),\n all pointers printed with %p are printed with hashed addresses\n instead of real addresses in order to avoid leaking addresses in\n dmesg and syslog. But this applies to kdb too, with is unfortunate:\n \n Entering kdb (current\u003d0x(ptrval), pid 329) due to Keyboard Entry\n kdb\u003e ps\n 15 sleeping system daemon (state M) processes suppressed,\n use \u0027ps A\u0027 to see all.\n Task Addr Pid Parent [*] cpu State Thread Command\n 0x(ptrval) 329 328 1 0 R 0x(ptrval) *sh\n \n 0x(ptrval) 1 0 0 0 S 0x(ptrval) init\n 0x(ptrval) 3 2 0 0 D 0x(ptrval) rcu_gp\n 0x(ptrval) 4 2 0 0 D 0x(ptrval) rcu_par_gp\n 0x(ptrval) 5 2 0 0 D 0x(ptrval) kworker/0:0\n 0x(ptrval) 6 2 0 0 D 0x(ptrval) kworker/0:0H\n 0x(ptrval) 7 2 0 0 D 0x(ptrval) kworker/u2:0\n 0x(ptrval) 8 2 0 0 D 0x(ptrval) mm_percpu_wq\n 0x(ptrval) 10 2 0 0 D 0x(ptrval) rcu_preempt\n \n The whole purpose of kdb is to debug, and for debugging real addresses\n need to be known. In addition, data displayed by kdb doesn\u0027t go into\n dmesg.\n \n This patch replaces all %p by %px in kdb in order to display real\n addresses.\n \n Fixes: ad67b74d2469 (\"printk: hash addresses printed with %p\")\n Cc: \u003cstable@vger.kernel.org\u003e\n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Signed-off-by: Jason Wessel \u003cjason.wessel@windriver.com\u003e\n\ncommit a0ca72c2d1ac83d0853a23ffde8f3624648b1ee8\nAuthor: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nDate: Fri Sep 14 12:35:44 2018 +0000\n\n kdb: use correct pointer when \u0027btc\u0027 calls \u0027btt\u0027\n \n On a powerpc 8xx, \u0027btc\u0027 fails as follows:\n \n Entering kdb (current\u003d0x(ptrval), pid 282) due to Keyboard Entry\n kdb\u003e btc\n btc: cpu status: Currently on cpu 0\n Available cpus: 0\n kdb_getarea: Bad address 0x0\n \n when booting the kernel with \u0027debug_boot_weak_hash\u0027, it fails as well\n \n Entering kdb (current\u003d0xba99ad80, pid 284) due to Keyboard Entry\n kdb\u003e btc\n btc: cpu status: Currently on cpu 0\n Available cpus: 0\n kdb_getarea: Bad address 0xba99ad80\n \n On other platforms, Oopses have been observed too, see\n https://github.com/linuxppc/linux/issues/139\n \n This is due to btc calling \u0027btt\u0027 with %p pointer as an argument.\n \n This patch replaces %p by %px to get the real pointer value as\n expected by \u0027btt\u0027\n \n Fixes: ad67b74d2469d9b82 (\"printk: hash addresses printed with %p\")\n Signed-off-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\n Cc: \u003cstable@vger.kernel.org\u003e # 4.15+\n Reviewed-by: Daniel Thompson \u003cdaniel.thompson@linaro.org\u003e\n Signed-off-by: Jason Wessel \u003cjason.wessel@windriver.com\u003e\n\ncommit f0f84837c6dadf4c5518892524a57ae089511d52\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Sun Aug 12 06:01:40 2018 -0700\n\n xtensa: rework noMMU cache attributes initialization\n \n Marking default memory region as cached is not always sufficient and is\n not flexible. Allow specifying cache attributes for the whole memory\n address space with new config entry MEMMAP_CACHEATTR. Apply it after\n cache initialization.\n \n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit 2cf848801042612933d52d4d9f931c670d914934\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Fri Aug 10 22:21:22 2018 -0700\n\n xtensa: increase ranges in ___invalidate_{i,d}cache_all\n \n Cache invalidation macros use cache line size to iterate over\n invalidated cache lines, assuming that all cache ways are invalidated by\n single instruction, but xtensa ISA recommends to not assume that for\n future compatibility:\n In some implementations all ways at index Addry-1..z are invalidated\n regardless of the specified way, but for future compatibility this\n behavior should not be assumed.\n \n Iterate over all cache ways in ___invalidate_icache_all and\n ___invalidate_dcache_all.\n \n Cc: stable@vger.kernel.org\n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit 30bd09dcbfb7eef9e4b6a9455988552de9349327\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Fri Aug 10 20:43:48 2018 -0700\n\n xtensa: limit offsets in __loop_cache_{all,page}\n \n When building kernel for xtensa cores with big cache lines (e.g. 128\n bytes or more) __loop_cache_all and __loop_cache_page may generate\n assembly instructions with immediate fields that are too big. This\n results in the following build errors:\n \n arch/xtensa/mm/misc.S: Assembler messages:\n arch/xtensa/mm/misc.S:464: Error: operand 2 of \u0027diwbi\u0027 has invalid value \u0027256\u0027\n arch/xtensa/mm/misc.S:464: Error: operand 2 of \u0027diwbi\u0027 has invalid value \u0027384\u0027\n arch/xtensa/kernel/head.S: Assembler messages:\n arch/xtensa/kernel/head.S:172: Error: operand 2 of \u0027diu\u0027 has invalid value \u0027256\u0027\n arch/xtensa/kernel/head.S:172: Error: operand 2 of \u0027diu\u0027 has invalid value \u0027384\u0027\n arch/xtensa/kernel/head.S:176: Error: operand 2 of \u0027iiu\u0027 has invalid value \u0027256\u0027\n arch/xtensa/kernel/head.S:176: Error: operand 2 of \u0027iiu\u0027 has invalid value \u0027384\u0027\n arch/xtensa/kernel/head.S:255: Error: operand 2 of \u0027diwb\u0027 has invalid value \u0027256\u0027\n arch/xtensa/kernel/head.S:255: Error: operand 2 of \u0027diwb\u0027 has invalid value \u0027384\u0027\n \n Add parameter max_immed to these macros and use it to limit values of\n immediate operands. Extract common code of these macros into the new\n macro __loop_cache_unroll.\n \n Cc: stable@vger.kernel.org\n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit 21ec04090f59647e69fb775ae9d447fcc778701e\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Wed Jul 11 15:51:46 2018 -0700\n\n xtensa: xtfpga: implement platform DMA address helpers\n \n XTFPGA platform has two views of the system memory when used with noMMU\n cores: 256M at 0x60000000 and the same 256M at 0x90000000.\n Mapping at 0x60000000 is meant to be cached, mapping at 0x90000000 is\n meant to be uncached.\n Implement platform_vaddr_* functions accordingly.\n \n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit 1f008cfec5d529b30ca8da1a1f5fbbd457c10382\nAuthor: Amit Pundir \u003camit.pundir@linaro.org\u003e\nDate: Tue Jun 12 19:44:10 2018 +0530\n\n NFC: fdp: Fix unused variable warnings\n \n Fix unused variable warnings reported on next-20180612.\n \n Fixes: 7579d009c4a1 (\"NFC: fdp: Remove __func__ from dev_dbg()\")\n Signed-off-by: Amit Pundir \u003camit.pundir@linaro.org\u003e\n Signed-off-by: Samuel Ortiz \u003csameo@linux.intel.com\u003e\n\ncommit 7579d009c4a1e01d0c21e72ee9c4dbf0832f83f3\nAuthor: Amit Pundir \u003camit.pundir@linaro.org\u003e\nDate: Fri May 4 00:08:56 2018 +0530\n\n NFC: fdp: Remove __func__ from dev_dbg()\n \n Remove redundant __func__ parameter from dev_dgb() calls and\n delete empty dev_dbg() trace calls, which are redundant if\n function tracer is enabled.\n \n Signed-off-by: Amit Pundir \u003camit.pundir@linaro.org\u003e\n Reviewed-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Samuel Ortiz \u003csameo@linux.intel.com\u003e\n\ncommit 430c62fa23d4806488cf7d4a226a4638f1945363\nAuthor: Suren Baghdasaryan \u003csurenb@google.com\u003e\nDate: Fri May 4 00:08:55 2018 +0530\n\n NFC: fdp: Fix possible buffer overflow in WCS4000 NFC driver\n \n Possible buffer overflow when reading next_read_size bytes into\n tmp buffer after next_read_size was extracted from a previous packet.\n \n cc: Stable \u003cstable@vger.kernel.org\u003e\n Signed-off-by: Suren Baghdasaryan \u003csurenb@google.com\u003e\n Signed-off-by: Amit Pundir \u003camit.pundir@linaro.org\u003e\n Reviewed-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Samuel Ortiz \u003csameo@linux.intel.com\u003e\n\ncommit 761e37b71e26de04de744794ef7155d41a78a8dd\nAuthor: Suren Baghdasaryan \u003csurenb@google.com\u003e\nDate: Fri May 4 00:08:54 2018 +0530\n\n NFC: Fix possible memory corruption when handling SHDLC I-Frame commands\n \n When handling SHDLC I-Frame commands \"pipe\" field used for indexing\n into an array should be checked before usage. If left unchecked it\n might access memory outside of the array of size NFC_HCI_MAX_PIPES(127).\n \n cc: Stable \u003cstable@vger.kernel.org\u003e\n Signed-off-by: Suren Baghdasaryan \u003csurenb@google.com\u003e\n Signed-off-by: Amit Pundir \u003camit.pundir@linaro.org\u003e\n Reviewed-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Samuel Ortiz \u003csameo@linux.intel.com\u003e\n\ncommit 48badd41fbfd803331dc8796cbb12b826a6446b9\nAuthor: Suren Baghdasaryan \u003csurenb@google.com\u003e\nDate: Fri May 4 00:08:53 2018 +0530\n\n NFC: st21nfca: Fix out of bounds kernel access when handling ATR_REQ\n \n Out of bounds kernel accesses in st21nfca\u0027s NFC HCI layer\n might happen when handling ATR_REQ events if user-specified\n atr_req-\u003elength is bigger than the buffer size. In\n that case memcpy() inside st21nfca_tm_send_atr_res() will\n read extra bytes resulting in OOB read from the kernel heap.\n \n cc: Stable \u003cstable@vger.kernel.org\u003e\n Signed-off-by: Suren Baghdasaryan \u003csurenb@google.com\u003e\n Signed-off-by: Amit Pundir \u003camit.pundir@linaro.org\u003e\n Reviewed-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\n Signed-off-by: Samuel Ortiz \u003csameo@linux.intel.com\u003e\n\ncommit c99f996b2ba49572d228075a5517455b8ecf6585\nAuthor: Daniel Mack \u003cdaniel@zonque.org\u003e\nDate: Wed May 16 15:32:47 2018 +0200\n\n NFC: st95hf: drop illegal kfree_skb()\n \n The skb that is passed in to -\u003ein_send_cmd() is freed by the core when the\n function returns. Calling kfree_skb() on it from the driver callback will\n hence lead to a double-free.\n \n Signed-off-by: Daniel Mack \u003cdaniel@zonque.org\u003e\n Signed-off-by: Samuel Ortiz \u003csameo@linux.intel.com\u003e\n\ncommit 85ef4f34cc64ecd13099817bbcab0b46564f7413\nAuthor: Daniel Mack \u003cdaniel@zonque.org\u003e\nDate: Wed May 16 15:32:46 2018 +0200\n\n NFC: st95hf: initialize semaphore and mutex earlier\n \n \u0027rm_lock\u0027 and \u0027exchange_lock\u0027 need to be ready before the IRQ handler has a\n chance to fire.\n \n This fixes the oops below.\n \n [ 1.040255] Internal error: Oops: 96000004 [#1] PREEMPT SMP\n [...]\n [ 1.181564] Call trace:\n [ 1.188591] Exception stack(0xffff00000a473c40 to 0xffff00000a473d80)\n [ 1.190943] 3c40: ffff80003673b118 0000000000000000 ffff800036374380 0000000000000000\n [ 1.197542] 3c60: 0000000000000000 0000000000000000 00000000044b2ac5 0000000000000001\n [ 1.205354] 3c80: ffff800036374d60 ffff00000a473d70 0000000000000980 0000000000000000\n [ 1.213166] 3ca0: 0000000000000001 0000000000000000 000000000000004c 0000000000000033\n [ 1.217590] st95hf spi2.0: err: por seq failed for st95hf\n [ 1.228788] 3cc0: 0000000000000019 0000000000000001 0000000000000007 ffff80003673b118\n [ 1.234175] 3ce0: ffff800009f27000 0000000000000000 ffff80003673b1c8 ffff80003673b1b0\n [ 1.241986] 3d00: ffff0000080f0000 ffff800009f716a4 ffff00000894bb40 0000000000000000\n [ 1.249800] 3d20: 0000000000000000 ffff00000a473d80 ffff0000084268c0 ffff00000a473d80\n [ 1.257611] 3d40: ffff0000084268c4 0000000040000005 ffff00000a473d60 ffff0000080e5688\n [ 1.265424] 3d60: ffffffffffffffff ffff0000084268a4 ffff00000a473d80 ffff0000084268c4\n [ 1.273239] [\u003cffff0000084268c4\u003e] st95hf_irq_thread_handler+0x44/0x3a0\n [ 1.281048] [\u003cffff0000080f00b0\u003e] irq_thread_fn+0x28/0x68\n [ 1.287468] [\u003cffff0000080f035c\u003e] irq_thread+0x10c/0x1a0\n [ 1.292850] [\u003cffff0000080c50bc\u003e] kthread+0x12c/0x130\n [ 1.297799] [\u003cffff000008084b88\u003e] ret_from_fork+0x10/0x18\n [ 1.303008] Code: aa1603e0 f9403675 940d010f aa1303e0 (f94066a1)\n [ 1.308307] ---[ end trace d058c1b88aad74d8 ]---\n \n Signed-off-by: Daniel Mack \u003cdaniel@zonque.org\u003e\n Signed-off-by: Samuel Ortiz \u003csameo@linux.intel.com\u003e\n\ncommit 00afd4c21f2f2e97a1c8eed9d54253439ff9e902\nAuthor: Andrey Konovalov \u003candreyknvl@google.com\u003e\nDate: Thu Apr 12 18:56:56 2018 +0200\n\n NFC: fix attrs checks in netlink interface\n \n nfc_genl_deactivate_target() relies on the NFC_ATTR_TARGET_INDEX\n attribute being present, but doesn\u0027t check whether it is actually\n provided by the user. Same goes for nfc_genl_fw_download() and\n NFC_ATTR_FIRMWARE_NAME.\n \n This patch adds appropriate checks.\n \n Found with syzkaller.\n \n Signed-off-by: Andrey Konovalov \u003candreyknvl@google.com\u003e\n Signed-off-by: Samuel Ortiz \u003csameo@linux.intel.com\u003e\n\ncommit 4e4e8232f0ae36573a66f40c765198baa75f327b\nAuthor: Fabio Estevam \u003cfabio.estevam@nxp.com\u003e\nDate: Sat Mar 24 10:44:59 2018 -0300\n\n nfc: st21nfca: Remove unnecessary devm_kzalloc() cast\n \n There is no need to use cast for the returned value\n from memory allocation functions, so remove the unnecessary cast.\n \n Detected via Coccinelle script:\n scripts/coccinelle/api/alloc/alloc_cast.cocci.\n \n Signed-off-by: Fabio Estevam \u003cfabio.estevam@nxp.com\u003e\n Signed-off-by: Samuel Ortiz \u003csameo@linux.intel.com\u003e\n\ncommit 6a893401a728236d9bec0575d73527992ca7929c\nAuthor: Fabio Estevam \u003cfabio.estevam@nxp.com\u003e\nDate: Sat Mar 24 10:44:58 2018 -0300\n\n nfc: st21nfca: Check for devm_kzalloc() failure\n \n devm_kzalloc() may fail, so we should better check for error and\n propagate the error in the case of allocation failure.\n \n This avoids a potential NULL pointer dereference later on.\n \n Signed-off-by: Fabio Estevam \u003cfabio.estevam@nxp.com\u003e\n Signed-off-by: Samuel Ortiz \u003csameo@linux.intel.com\u003e\n\ncommit e530d198bdbfe59b5c99ee48287e81f893326731\nAuthor: Colin Ian King \u003ccolin.king@canonical.com\u003e\nDate: Tue Jan 30 15:33:59 2018 +0000\n\n NFC: fdp: fix signed less or equal zero check in u8 max_size\n \n The u8 variable max_size is being assigned a return value from the\n call to nci_conn_max_data_pkt_payload_size that can return a -ve\n error return. Since max_size is a u8, the -ve check for the error will\n always be false. Fix this by making max_size an int type.\n \n Detected using Coccinelle:\n drivers/nfc/fdp/fdp.c:208:5-13: WARNING: Unsigned expression compared\n with zero: max_size \u003c\u003d 0\n \n Fixes: a06347c04c13 (\"NFC: Add Intel Fields Peak NFC solution driver\")\n Signed-off-by: Colin Ian King \u003ccolin.king@canonical.com\u003e\n Signed-off-by: Samuel Ortiz \u003csameo@linux.intel.com\u003e\n\ncommit ede179d96eb840ee20487f10ea18e1e24409f0e1\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Wed Feb 14 16:12:54 2018 -0800\n\n xtensa: support DMA_ATTR_NO_KERNEL_MAPPING attribute\n \n An application that doesn\u0027t need kernel mapping of the DMA memory it\n allocates may specify DMA_ATTR_NO_KERNEL_MAPPING attribute to allocation\n function. Support this attribute and return address of the first struct\n page that covers the allocation.\n \n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit 9e0031ddb60803279e0b4a39e49e04d362c712c8\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Sun Feb 11 01:07:54 2018 -0800\n\n xtensa: fix build with KASAN\n \n The commit 917538e212a2 (\"kasan: clean up KASAN_SHADOW_SCALE_SHIFT\n usage\") removed KASAN_SHADOW_SCALE_SHIFT definition from\n include/linux/kasan.h and added it to architecture-specific headers,\n except for xtensa. This broke the xtensa build with KASAN enabled.\n Define KASAN_SHADOW_SCALE_SHIFT in arch/xtensa/include/asm/kasan.h\n \n Reported by: kbuild test robot \u003cfengguang.wu@intel.com\u003e\n Fixes: 917538e212a2 (\"kasan: clean up KASAN_SHADOW_SCALE_SHIFT usage\")\n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit 73b37067623c682edd4153640d09728a7c581a33\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Fri Jan 5 14:27:58 2018 -0800\n\n xtensa: fix futex_atomic_cmpxchg_inatomic\n \n Return 0 if the operation was successful, not the userspace memory\n value. Check that userspace value equals passed oldval, not itself.\n Don\u0027t update *uval if the value wasn\u0027t read from userspace memory.\n \n This fixes process hang due to infinite loop in futex_lock_pi.\n It also fixes a bunch of glibc tests nptl/tst-mutexpi*.\n \n Cc: stable@vger.kernel.org\n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit 5874c700308f11a8fe7126e185e4cebb7d71efa2\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Wed Mar 29 15:44:47 2017 -0700\n\n xtensa: make __pa work with uncached KSEG addresses\n \n When __pa is applied to virtual address in uncached KSEG region the\n result is incorrect. Fix it by checking if the original address is in\n the uncached KSEG and adjusting the result. It looks better than masking\n off bits because pfn_valid would correctly work with new __pa results\n and it may be made working in noMMU case, once we get definition for\n uncached memory view.\n \n This is required for the dma_common_mmap and other functions calling __pa\n on coherent DMA memory addresses.\n \n Cc: stable@vger.kernel.org\n Tested-by: Boris Brezillon \u003cboris.brezillon@free-electrons.com\u003e\n Reviewed-by: Boris Brezillon \u003cboris.brezillon@free-electrons.com\u003e\n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\n\ncommit 6e72293ab0e9687648fab1e9f6841ebb36cd2404\nAuthor: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nDate: Tue Jan 31 18:35:37 2017 -0800\n\n xtensa: fix noMMU build on cores with MMU\n \n Commit bf15f86b343ed8 (\"xtensa: initialize MMU before jumping to reset\n vector\") calls MMU management functions even when CONFIG_MMU is not\n selected. That breaks noMMU build on cores with MMU.\n \n Don\u0027t manage MMU when CONFIG_MMU is not selected.\n \n Cc: stable@vger.kernel.org\n Signed-off-by: Max Filippov \u003cjcmvbkbc@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "3ca91a98962397cad578261559e2f1fdccdadfa9", "tree": "c55dddbf247cd1870fca215a29daec978d142c57", "parents": [ "d57689352b80046449f1ab97c6d683de71a386c3" ], "author": { "name": "Lu Shuaibing", "email": "shuaibinglu@126.com", "time": "Wed Sep 04 05:02:28 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:28 2019 +0000" }, "message": "ipc/msg.c: consolidate all xxxctl_down() functions\n\nA use of uninitialized memory in msgctl_down() because msqid64 in\nksys_msgctl hasn\u0027t been initialized. The local | msqid64 | is created in\nksys_msgctl() and then passed into msgctl_down(). Along the way msqid64\nis never initialized before msgctl_down() checks msqid64-\u003emsg_qbytes.\n\nKUMSAN(KernelUninitializedMemorySantizer, a new error detection tool)\nreports:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nBUG: KUMSAN: use of uninitialized memory in msgctl_down+0x94/0x300\nRead of size 8 at addr ffff88806bb97eb8 by task syz-executor707/2022\n\nCPU: 0 PID: 2022 Comm: syz-executor707 Not tainted 5.2.0-rc4+ #63\nHardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Ubuntu-1.8.2-1ubuntu1 04/01/2014\nCall Trace:\n dump_stack+0x75/0xae\n __kumsan_report+0x17c/0x3e6\n kumsan_report+0xe/0x20\n msgctl_down+0x94/0x300\n ksys_msgctl.constprop.14+0xef/0x260\n do_syscall_64+0x7e/0x1f0\n entry_SYSCALL_64_after_hwframe+0x44/0xa9\nRIP: 0033:0x4400e9\nCode: 18 89 d0 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 \u003c48\u003e 3d 01 f0 ff ff 0f 83 fb 13 fc ff c3 66 2e 0f 1f 84 00 00 00 00\nRSP: 002b:00007ffd869e0598 EFLAGS: 00000246 ORIG_RAX: 0000000000000047\nRAX: ffffffffffffffda RBX: 00000000004002c8 RCX: 00000000004400e9\nRDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000\nRBP: 00000000006ca018 R08: 0000000000000000 R09: 0000000000000000\nR10: 00000000ffffffff R11: 0000000000000246 R12: 0000000000401970\nR13: 0000000000401a00 R14: 0000000000000000 R15: 0000000000000000\n\nThe buggy address belongs to the page:\npage:ffffea0001aee5c0 refcount:0 mapcount:0 mapping:0000000000000000 index:0x0\nflags: 0x100000000000000()\nraw: 0100000000000000 0000000000000000 ffffffff01ae0101 0000000000000000\nraw: 0000000000000000 0000000000000000 00000000ffffffff 0000000000000000\npage dumped because: kumsan: bad access detected\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSyzkaller reproducer:\nmsgctl$IPC_RMID(0x0, 0x0)\n\nC reproducer:\n// autogenerated by syzkaller (https://github.com/google/syzkaller)\n\nint main(void)\n{\n syscall(__NR_mmap, 0x20000000, 0x1000000, 3, 0x32, -1, 0);\n syscall(__NR_msgctl, 0, 0, 0);\n return 0;\n}\n\nLink: http://lkml.kernel.org/r/20190613014044.24234-1-shuaibinglu@126.com\nSigned-off-by: Lu Shuaibing \u003cshuaibinglu@126.com\u003e\nSuggested-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nCc: Davidlohr Bueso \u003cdave@stgolabs.net\u003e\nCc: Manfred Spraul \u003cmanfred@colorfullife.com\u003e\nCc: NeilBrown \u003cneilb@suse.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "d57689352b80046449f1ab97c6d683de71a386c3", "tree": "1d91c71529c09cfaacf5b51590e253a84f1e1338", "parents": [ "e54f4b1e018e0cb4058bcb93e04dd76afe33e2d4" ], "author": { "name": "Markus Elfring", "email": "elfring@users.sourceforge.net", "time": "Wed Sep 04 05:02:28 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:28 2019 +0000" }, "message": "ipc/mqueue: improve exception handling in do_mq_notify()\n\nNull pointers were assigned to local variables in a few cases as exception\nhandling. The jump target “out” was used where no meaningful data\nprocessing actions should eventually be performed by branches of an if\nstatement then. Use an additional jump target for calling dev_kfree_skb()\ndirectly.\n\nReturn also directly after error conditions were detected when no extra\nclean-up is needed by this function implementation.\n\nLink: http://lkml.kernel.org/r/592ef10e-0b69-72d0-9789-fc48f638fdfd@web.de\nSigned-off-by: Markus Elfring \u003celfring@users.sourceforge.net\u003e\nCc: Davidlohr Bueso \u003cdave@stgolabs.net\u003e\nCc: Manfred Spraul \u003cmanfred@colorfullife.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "e54f4b1e018e0cb4058bcb93e04dd76afe33e2d4", "tree": "0a76e39e1f3e8431954db87b38bbeb99577ee7e6", "parents": [ "4355e7f0a37ec623658bde9359f933e553d84ac1" ], "author": { "name": "Markus Elfring", "email": "elfring@users.sourceforge.net", "time": "Wed Sep 04 05:02:28 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:28 2019 +0000" }, "message": "ipc/mqueue.c: delete an unnecessary check before the macro call dev_kfree_skb()\n\ndev_kfree_skb() input parameter validation, thus the test around the call\nis not needed.\n\nThis issue was detected by using the Coccinelle software.\n\nLink: http://lkml.kernel.org/r/07477187-63e5-cc80-34c1-32dd16b38e12@web.de\nSigned-off-by: Markus Elfring \u003celfring@users.sourceforge.net\u003e\nCc: Davidlohr Bueso \u003cdave@stgolabs.net\u003e\nCc: Manfred Spraul \u003cmanfred@colorfullife.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "4355e7f0a37ec623658bde9359f933e553d84ac1", "tree": "02fe8e45925374df2070bfba9d1f09fcd6086b08", "parents": [ "448602037f1b69f78484248891483f8da12c83b1" ], "author": { "name": "Kees Cook", "email": "keescook@chromium.org", "time": "Wed Sep 04 05:02:28 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:28 2019 +0000" }, "message": "bug: move WARN_ON() \"cut here\" into exception handler\n\nThe original clean up of \"cut here\" missed the WARN_ON() case (that does\nnot have a printk message), which was fixed recently by adding an explicit\nprintk of \"cut here\". This had the downside of adding a printk() to every\nWARN_ON() caller, which reduces the utility of using an instruction\nexception to streamline the resulting code. By making this a new BUGFLAG,\nall of these can be removed and \"cut here\" can be handled by the exception\nhandler.\n\nThis was very pronounced on PowerPC, but the effect can be seen on x86 as\nwell. The resulting text size of a defconfig build shows some small\nsavings from this patch:\n\n text data bss dec hex filename\n19691167 5134320 1646664 26472151 193eed7 vmlinux.before\n19676362 5134260 1663048 26473670 193f4c6 vmlinux.after\n\nThis change also opens the door for creating something like BUG_MSG(),\nwhere a custom printk() before issuing BUG(), without confusing the \"cut\nhere\" line.\n\nLink: http://lkml.kernel.org/r/201908200943.601DD59DCE@keescook\nFixes: 6b15f678fb7d (\"include/asm-generic/bug.h: fix \"cut here\" for WARN_ON for __WARN_TAINT architectures\")\nSigned-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\nReported-by: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nCc: Drew Davenport \u003cddavenport@chromium.org\u003e\nCc: Arnd Bergmann \u003carnd@arndb.de\u003e\nCc: \"Steven Rostedt (VMware)\" \u003crostedt@goodmis.org\u003e\nCc: Feng Tang \u003cfeng.tang@intel.com\u003e\nCc: Petr Mladek \u003cpmladek@suse.com\u003e\nCc: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nCc: Borislav Petkov \u003cbp@suse.de\u003e\nCc: YueHaibing \u003cyuehaibing@huawei.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "448602037f1b69f78484248891483f8da12c83b1", "tree": "8f3323375c2985ca05e9a9d2ba60e1b8d247e3be", "parents": [ "256dbe6c2783ed4fabb15aae66fac564c3cc45f8" ], "author": { "name": "Kees Cook", "email": "keescook@chromium.org", "time": "Wed Sep 04 05:02:27 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:27 2019 +0000" }, "message": "bug: consolidate __WARN_FLAGS usage\n\nInstead of having separate tests for __WARN_FLAGS, merge the two #ifdef\nblocks and replace the synonym WANT_WARN_ON_SLOWPATH macro.\n\nLink: http://lkml.kernel.org/r/20190819234111.9019-7-keescook@chromium.org\nSigned-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: Arnd Bergmann \u003carnd@arndb.de\u003e\nCc: Borislav Petkov \u003cbp@suse.de\u003e\nCc: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nCc: Drew Davenport \u003cddavenport@chromium.org\u003e\nCc: Feng Tang \u003cfeng.tang@intel.com\u003e\nCc: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Petr Mladek \u003cpmladek@suse.com\u003e\nCc: \"Steven Rostedt (VMware)\" \u003crostedt@goodmis.org\u003e\nCc: YueHaibing \u003cyuehaibing@huawei.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "256dbe6c2783ed4fabb15aae66fac564c3cc45f8", "tree": "b7bdfd8e1408e5e167ef37406aa88ad083935579", "parents": [ "1c47e1b67f475d2c42a79e19e86e955f66fad36a" ], "author": { "name": "Kees Cook", "email": "keescook@chromium.org", "time": "Wed Sep 04 05:02:27 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:27 2019 +0000" }, "message": "bug: clean up helper macros to remove __WARN_TAINT()\n\nIn preparation for cleaning up \"cut here\" even more, this removes the\n__WARN_*TAINT() helpers, as they limit the ability to add new BUGFLAG_*\nflags to call sites. They are removed by expanding them into full\n__WARN_FLAGS() calls.\n\nLink: http://lkml.kernel.org/r/20190819234111.9019-6-keescook@chromium.org\nSigned-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: Arnd Bergmann \u003carnd@arndb.de\u003e\nCc: Borislav Petkov \u003cbp@suse.de\u003e\nCc: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nCc: Drew Davenport \u003cddavenport@chromium.org\u003e\nCc: Feng Tang \u003cfeng.tang@intel.com\u003e\nCc: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Petr Mladek \u003cpmladek@suse.com\u003e\nCc: \"Steven Rostedt (VMware)\" \u003crostedt@goodmis.org\u003e\nCc: YueHaibing \u003cyuehaibing@huawei.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "1c47e1b67f475d2c42a79e19e86e955f66fad36a", "tree": "d7665a370bb733d8211691133b1bd8fe073062a9", "parents": [ "68cc91c3ffc6617f788a953d0187e3977e47ec93" ], "author": { "name": "Kees Cook", "email": "keescook@chromium.org", "time": "Wed Sep 04 05:02:27 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:27 2019 +0000" }, "message": "bug: lift \"cut here\" out of __warn()\n\nIn preparation for cleaning up \"cut here\", move the \"cut here\" logic up\nout of __warn() and into callers that pass non-NULL args. For anyone\nlooking closely, there are two callers that pass NULL args: one already\nexplicitly prints \"cut here\". The remaining case is covered by how a WARN\nis built, which will be cleaned up in the next patch.\n\nLink: http://lkml.kernel.org/r/20190819234111.9019-5-keescook@chromium.org\nSigned-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: Arnd Bergmann \u003carnd@arndb.de\u003e\nCc: Borislav Petkov \u003cbp@suse.de\u003e\nCc: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nCc: Drew Davenport \u003cddavenport@chromium.org\u003e\nCc: Feng Tang \u003cfeng.tang@intel.com\u003e\nCc: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Petr Mladek \u003cpmladek@suse.com\u003e\nCc: \"Steven Rostedt (VMware)\" \u003crostedt@goodmis.org\u003e\nCc: YueHaibing \u003cyuehaibing@huawei.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "68cc91c3ffc6617f788a953d0187e3977e47ec93", "tree": "23ae8199fae70b96cee9f3644fd28846e97bd707", "parents": [ "221827855d537f2f621fb3a786b9ec5b2590e977" ], "author": { "name": "Kees Cook", "email": "keescook@chromium.org", "time": "Wed Sep 04 05:02:27 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:27 2019 +0000" }, "message": "bug: consolidate warn_slowpath_fmt() usage\n\nInstead of having a separate helper for no printk output, just consolidate\nthe logic into warn_slowpath_fmt().\n\nLink: http://lkml.kernel.org/r/20190819234111.9019-4-keescook@chromium.org\nSigned-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: Arnd Bergmann \u003carnd@arndb.de\u003e\nCc: Borislav Petkov \u003cbp@suse.de\u003e\nCc: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nCc: Drew Davenport \u003cddavenport@chromium.org\u003e\nCc: Feng Tang \u003cfeng.tang@intel.com\u003e\nCc: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Petr Mladek \u003cpmladek@suse.com\u003e\nCc: \"Steven Rostedt (VMware)\" \u003crostedt@goodmis.org\u003e\nCc: YueHaibing \u003cyuehaibing@huawei.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "221827855d537f2f621fb3a786b9ec5b2590e977", "tree": "281eb6c6e75f073af4b900d68b8738758ac4ce30", "parents": [ "5f132023b22609b5a7a9feb186acd58105a5c655" ], "author": { "name": "Kees Cook", "email": "keescook@chromium.org", "time": "Wed Sep 04 05:02:26 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:26 2019 +0000" }, "message": "bug: rename __WARN_printf_taint() to __WARN_printf()\n\nThis just renames the helper to improve readability.\n\nLink: http://lkml.kernel.org/r/20190819234111.9019-3-keescook@chromium.org\nSigned-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: Arnd Bergmann \u003carnd@arndb.de\u003e\nCc: Borislav Petkov \u003cbp@suse.de\u003e\nCc: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nCc: Drew Davenport \u003cddavenport@chromium.org\u003e\nCc: Feng Tang \u003cfeng.tang@intel.com\u003e\nCc: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Petr Mladek \u003cpmladek@suse.com\u003e\nCc: \"Steven Rostedt (VMware)\" \u003crostedt@goodmis.org\u003e\nCc: YueHaibing \u003cyuehaibing@huawei.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "5f132023b22609b5a7a9feb186acd58105a5c655", "tree": "30b36e9ec72ead00330aac4c834c29bd29d4b525", "parents": [ "6305ac2d32bd220113b6f261bce06724b3cbb948" ], "author": { "name": "Kees Cook", "email": "keescook@chromium.org", "time": "Wed Sep 04 05:02:26 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:26 2019 +0000" }, "message": "bug: refactor away warn_slowpath_fmt_taint()\n\nPatch series \"Clean up WARN() \"cut here\" handling\", v2.\n\nChristophe Leroy noticed that the fix for missing \"cut here\" in the WARN()\ncase was adding explicit printk() calls instead of teaching the exception\nhandler to add it. This refactors the bug/warn infrastructure to pass\nthis information as a new BUGFLAG.\n\nLonger details repeated from the last patch in the series:\n\n\nbug: move WARN_ON() \"cut here\" into exception handler\n\nThe original cleanup of \"cut here\" missed the WARN_ON() case (that does\nnot have a printk message), which was fixed recently by adding an explicit\nprintk of \"cut here\". This had the downside of adding a printk() to every\nWARN_ON() caller, which reduces the utility of using an instruction\nexception to streamline the resulting code. By making this a new BUGFLAG,\nall of these can be removed and \"cut here\" can be handled by the exception\nhandler.\n\nThis was very pronounced on PowerPC, but the effect can be seen on x86 as\nwell. The resulting text size of a defconfig build shows some small\nsavings from this patch:\n\n text data bss dec hex filename\n19691167 5134320 1646664 26472151 193eed7 vmlinux.before\n19676362 5134260 1663048 26473670 193f4c6 vmlinux.after\n\nThis change also opens the door for creating something like BUG_MSG(),\nwhere a custom printk() before issuing BUG(), without confusing the \"cut\nhere\" line.\n\n\nThis patch (of 7):\n\nThere\u0027s no reason to have specialized helpers for passing the warn taint\ndown to __warn(). Consolidate and refactor helper macros, removing\n__WARN_printf() and warn_slowpath_fmt_taint().\n\nLink: http://lkml.kernel.org/r/20190819234111.9019-2-keescook@chromium.org\nSigned-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Christophe Leroy \u003cchristophe.leroy@c-s.fr\u003e\nCc: Drew Davenport \u003cddavenport@chromium.org\u003e\nCc: Arnd Bergmann \u003carnd@arndb.de\u003e\nCc: \"Steven Rostedt (VMware)\" \u003crostedt@goodmis.org\u003e\nCc: Feng Tang \u003cfeng.tang@intel.com\u003e\nCc: Petr Mladek \u003cpmladek@suse.com\u003e\nCc: Mauro Carvalho Chehab \u003cmchehab+samsung@kernel.org\u003e\nCc: Borislav Petkov \u003cbp@suse.de\u003e\nCc: YueHaibing \u003cyuehaibing@huawei.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "6305ac2d32bd220113b6f261bce06724b3cbb948", "tree": "5902fd65fab1d8948a469fb077ecc5bb6dafb503", "parents": [ "39ce2265aeb437892b1fe46cd83ded75ebe9e8a6" ], "author": { "name": "Douglas Anderson", "email": "dianders@chromium.org", "time": "Wed Sep 04 05:02:26 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:26 2019 +0000" }, "message": "scripts/gdb: handle split debug\n\nSome systems (like Chrome OS) may use \"split debug\" for kernel modules. \nThat means that the debug symbols are in a different file than the main\nelf file. Let\u0027s handle that by also searching for debug symbols that end\nin \".ko.debug\".\n\nThis is a packaging topic. You can take a normal elf file and split the\ndebug out of it using objcopy. Try \"man objcopy\" and then take a look at\nthe \"--only-keep-debug\" option. It\u0027ll give you a whole recipe for doing\nsplitdebug. The suffix used for the debug symbols is arbitrary. If\npeople have other another suffix besides \".ko.debug\" then we could\npresumably support that too...\n\nFor portage (which is the packaging system used by Chrome OS) split debug\nis supported by default (and the suffix is .ko.debug). ...and so in\nChrome OS we always get the installed elf files stripped and then the\nsymbols stashed away.\n\nAt the moment we don\u0027t actually use the normal portage magic to do this\nfor the kernel though since it affects our ability to get good stack dumps\nin the kernel. We instead pass a script as \"strip\" [1].\n\n[1] https://chromium.googlesource.com/chromiumos/overlays/chromiumos-overlay/+/refs/heads/master/eclass/cros-kernel/strip_splitdebug\n\nLink: http://lkml.kernel.org/r/20190730234052.148744-1-dianders@chromium.org\nSigned-off-by: Douglas Anderson \u003cdianders@chromium.org\u003e\nReviewed-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\nReviewed-by: Jan Kiszka \u003cjan.kiszka@siemens.com\u003e\nCc: Kieran Bingham \u003ckbingham@kernel.org\u003e\nCc: Jason Wessel \u003cjason.wessel@windriver.com\u003e\nCc: Daniel Thompson \u003cdaniel.thompson@linaro.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "39ce2265aeb437892b1fe46cd83ded75ebe9e8a6", "tree": "2a2c8dd133af66ef46075d916e5a4e0bf337a7a8", "parents": [ "11008d2aed55b6ba817e51278fc52f4cb5d109c2" ], "author": { "name": "Douglas Anderson", "email": "dianders@chromium.org", "time": "Wed Sep 04 05:02:26 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:26 2019 +0000" }, "message": "kgdb: don\u0027t use a notifier to enter kgdb at panic; call directly\n\nRight now kgdb/kdb hooks up to debug panics by registering for the panic\nnotifier. This works OK except that it means that kgdb/kdb gets called\n_after_ the CPUs in the system are taken offline. That means that if\nanything important was happening on those CPUs (like something that might\nhave contributed to the panic) you can\u0027t debug them.\n\nSpecifically I ran into a case where I got a panic because a task was\n\"blocked for more than 120 seconds\" which was detected on CPU 2. I nicely\ngot shown stack traces in the kernel log for all CPUs including CPU 0,\nwhich was running \u0027PID: 111 Comm: kworker/0:1H\u0027 and was in the middle of\n__mmc_switch().\n\nI then ended up at the kdb prompt where switched over to kgdb to try to\nlook at local variables of the process on CPU 0. I found that I couldn\u0027t.\nDigging more, I found that I had no info on any tasks running on CPUs\nother than CPU 2 and that asking kdb for help showed me \"Error: no saved\ndata for this cpu\". This was because all the CPUs were offline.\n\nLet\u0027s move the entry of kdb/kgdb to a direct call from panic() and stop\nusing the generic notifier. Putting a direct call in allows us to order\nthings more properly and it also doesn\u0027t seem like we\u0027re breaking any\nabstractions by calling into the debugger from the panic function.\n\nDaniel said:\n\n: This patch changes the way kdump and kgdb interact with each other. \n: However it would seem rather odd to have both tools simultaneously armed\n: and, even if they were, the user still has the option to use panic_timeout\n: to force a kdump to happen. Thus I think the change of order is\n: acceptable.\n\nLink: http://lkml.kernel.org/r/20190703170354.217312-1-dianders@chromium.org\nSigned-off-by: Douglas Anderson \u003cdianders@chromium.org\u003e\nReviewed-by: Daniel Thompson \u003cdaniel.thompson@linaro.org\u003e\nCc: Jason Wessel \u003cjason.wessel@windriver.com\u003e\nCc: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: Borislav Petkov \u003cbp@suse.de\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: Feng Tang \u003cfeng.tang@intel.com\u003e\nCc: YueHaibing \u003cyuehaibing@huawei.com\u003e\nCc: Sergey Senozhatsky \u003csergey.senozhatsky.work@gmail.com\u003e\nCc: \"Steven Rostedt (VMware)\" \u003crostedt@goodmis.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "11008d2aed55b6ba817e51278fc52f4cb5d109c2", "tree": "b9a04c3e2dcbd940bf205345d48d3192026f3c78", "parents": [ "82383e56b3406f5a55e9972a5b344b44d8261924" ], "author": { "name": "Masahiro Yamada", "email": "yamada.masahiro@socionext.com", "time": "Wed Sep 04 05:02:25 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:25 2019 +0000" }, "message": "compiler: enable CONFIG_OPTIMIZE_INLINING forcibly\n\nCommit 9012d011660e (\"compiler: allow all arches to enable\nCONFIG_OPTIMIZE_INLINING\") allowed all architectures to enable this\noption. A couple of build errors were reported by randconfig, but all of\nthem have been ironed out.\n\nTowards the goal of removing CONFIG_OPTIMIZE_INLINING entirely (and it\nwill simplify the \u0027inline\u0027 macro in compiler_types.h), this commit changes\nit to always-on option. Going forward, the compiler will always be\nallowed to not inline functions marked \u0027inline\u0027.\n\nThis is not a problem for x86 since it has been long used by\narch/x86/configs/{x86_64,i386}_defconfig.\n\nI am keeping the config option just in case any problem crops up for other\narchitectures.\n\nThe code clean-up will be done after confirming this is solid.\n\nLink: http://lkml.kernel.org/r/20190830034304.24259-1-yamada.masahiro@socionext.com\nSigned-off-by: Masahiro Yamada \u003cyamada.masahiro@socionext.com\u003e\nAcked-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nCc: Ingo Molnar \u003cmingo@redhat.com\u003e\nCc: Borislav Petkov \u003cbp@alien8.de\u003e\nCc: Miguel Ojeda \u003cmiguel.ojeda.sandonis@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "82383e56b3406f5a55e9972a5b344b44d8261924", "tree": "187c96d04e8b057ef020e959eaa30d6f3836aa51", "parents": [ "a46dc6d4840b1c8eb3a266d6c0c0d9f956ec7ffb" ], "author": { "name": "Oleg Nesterov", "email": "oleg@redhat.com", "time": "Wed Sep 04 05:02:25 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:25 2019 +0000" }, "message": "aio: simplify read_events()\n\nChange wait_event_hrtimeout() to not call __wait_event_hrtimeout() if\ntimeout \u003d\u003d 0, this matches other _timeout() helpers in wait.h.\n\nThis allows to simplify its only user, read_events(), it no longer needs\nto optimize the \"until \u003d\u003d 0\" case by hand.\n\nNote: this patch doesn\u0027t use ___wait_cond_timeout because _hrtimeout()\nalso differs in that it returns 0 if succeeds and -ETIME on timeout. \nPerhaps we should change this to make it fully compatible with other\nhelpers.\n\nLink: http://lkml.kernel.org/r/20190607175413.GA29187@redhat.com\nSigned-off-by: Oleg Nesterov \u003coleg@redhat.com\u003e\nReviewed-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Benjamin LaHaise \u003cbcrl@kvack.org\u003e\nCc: Arnd Bergmann \u003carnd@arndb.de\u003e\nCc: David Laight \u003cDavid.Laight@ACULAB.COM\u003e\nCc: Deepa Dinamani \u003cdeepa.kernel@gmail.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nCc: Eric Wong \u003ce@80x24.org\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "a46dc6d4840b1c8eb3a266d6c0c0d9f956ec7ffb", "tree": "f894c3c9ab8555085992cb53a3e17a53a901e2cc", "parents": [ "e06f7ad527181aa992b86bcf995ec59d5880395b" ], "author": { "name": "Kees Cook", "email": "keescook@chromium.org", "time": "Wed Sep 04 05:02:25 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:25 2019 +0000" }, "message": "uaccess: add missing __must_check attributes\n\nThe usercopy implementation comments describe that callers of the\ncopy_*_user() family of functions must always have their return values\nchecked. This can be enforced at compile time with __must_check, so add\nit where needed.\n\nLink: http://lkml.kernel.org/r/201908251609.ADAD5CAAC1@keescook\nSigned-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: Alexander Viro \u003cviro@zeniv.linux.org.uk\u003e\nCc: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "e06f7ad527181aa992b86bcf995ec59d5880395b", "tree": "62d8bce45b2a633c6778a12ea2bbcb5ade15fdfa", "parents": [ "070bfc3cbeb3de6b04c288c8f696f95a1dac539e" ], "author": { "name": "Vasily Gorbik", "email": "gor@linux.ibm.com", "time": "Wed Sep 04 05:02:25 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:25 2019 +0000" }, "message": "kexec: restore arch_kexec_kernel_image_probe declaration\n\narch_kexec_kernel_image_probe function declaration has been removed by\ncommit 9ec4ecef0af7 (\"kexec_file,x86,powerpc: factor out kexec_file_ops\nfunctions\"). Still this function is overridden by couple of architectures\nand proper prototype declaration is therefore important, so bring it back.\nThis fixes the following sparse warning on s390:\narch/s390/kernel/machine_kexec_file.c:333:5: warning: symbol\n\u0027arch_kexec_kernel_image_probe\u0027 was not declared. Should it be static?\n\nLink: http://lkml.kernel.org/r/patch.git-ff1c9045ebdc.your-ad-here.call-01564402297-ext-5690@work.hours\nSigned-off-by: Vasily Gorbik \u003cgor@linux.ibm.com\u003e\nAcked-by: Dave Young \u003cdyoung@redhat.com\u003e\nReviewed-by: Bhupesh Sharma \u003cbhsharma@redhat.com\u003e\nCc: Eric Biederman \u003cebiederm@xmission.com\u003e\nCc: AKASHI Takahiro \u003ctakahiro.akashi@linaro.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "070bfc3cbeb3de6b04c288c8f696f95a1dac539e", "tree": "c49d2afcc63d3d4de3d6b36e79b06bf3fd807ca1", "parents": [ "2ea2259c026adcc225b6f7635a39584d654c517f" ], "author": { "name": "Tetsuo Handa", "email": "penguin-kernel@I-love.SAKURA.ne.jp", "time": "Wed Sep 04 05:02:24 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:24 2019 +0000" }, "message": "kexec: bail out upon SIGKILL when allocating memory.\n\nsyzbot found that a thread can stall for minutes inside kexec_load() after\nthat thread was killed by SIGKILL [1]. It turned out that the reproducer\nwas trying to allocate 2408MB of memory using kimage_alloc_page() from\nkimage_load_normal_segment(). Let\u0027s check for SIGKILL before doing memory\nallocation.\n\n[1] https://syzkaller.appspot.com/bug?id\u003da0e3436829698d5824231251fad9d8e998f94f5e\n\nLink: http://lkml.kernel.org/r/993c9185-d324-2640-d061-bed2dd18b1f7@I-love.SAKURA.ne.jp\nSigned-off-by: Tetsuo Handa \u003cpenguin-kernel@I-love.SAKURA.ne.jp\u003e\nReported-by: syzbot \u003csyzbot+8ab2d0f39fb79fe6ca40@syzkaller.appspotmail.com\u003e\nCc: Eric Biederman \u003cebiederm@xmission.com\u003e\nReviewed-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "2ea2259c026adcc225b6f7635a39584d654c517f", "tree": "9a4516da4a7d60230d3045cce1575b1f18b73f64", "parents": [ "9801876c423a93b8d2522d9e0005ef813d8f5943" ], "author": { "name": "Alexey Dobriyan", "email": "adobriyan@gmail.com", "time": "Wed Sep 04 05:02:23 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:23 2019 +0000" }, "message": "cpumask: nicer for_each_cpumask_and() signature\n\nMask arguments can be swapped without changing anything. Make arguments\nnames reflect that:\n\n\t#define for_each_cpu_and(cpu, mask1, mask2)\n\nLink: http://lkml.kernel.org/r/20190724183350.GA15041@avx2\nSigned-off-by: Alexey Dobriyan \u003cadobriyan@gmail.com\u003e\nReviewed-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "9801876c423a93b8d2522d9e0005ef813d8f5943", "tree": "65ddeb118fc20d23fc6a896b4738dd0b8915c82d", "parents": [ "90efc9b244ac28721e40d259ae7a1590e5f0c601" ], "author": { "name": "Sai Praneeth Prakhya", "email": "sai.praneeth.prakhya@intel.com", "time": "Wed Sep 04 05:02:23 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:23 2019 +0000" }, "message": "fork: improve error message for corrupted page tables\n\nWhen a user process exits, the kernel cleans up the mm_struct of the user\nprocess and during cleanup, check_mm() checks the page tables of the user\nprocess for corruption (E.g: unexpected page flags set/cleared). For\ncorrupted page tables, the error message printed by check_mm() isn\u0027t very\nclear as it prints the loop index instead of page table type (E.g:\nResident file mapping pages vs Resident shared memory pages). The loop\nindex in check_mm() is used to index rss_stat[] which represents\nindividual memory type stats. Hence, instead of printing index, print\nmemory type, thereby improving error message.\n\nWithout patch:\n--------------\n[ 204.836425] mm/pgtable-generic.c:29: bad p4d 0000000089eb4e92(800000025f941467)\n[ 204.836544] BUG: Bad rss-counter state mm:00000000f75895ea idx:0 val:2\n[ 204.836615] BUG: Bad rss-counter state mm:00000000f75895ea idx:1 val:5\n[ 204.836685] BUG: non-zero pgtables_bytes on freeing mm: 20480\n\nWith patch:\n-----------\n[ 69.815453] mm/pgtable-generic.c:29: bad p4d 0000000084653642(800000025ca37467)\n[ 69.815872] BUG: Bad rss-counter state mm:00000000014a6c03 type:MM_FILEPAGES val:2\n[ 69.815962] BUG: Bad rss-counter state mm:00000000014a6c03 type:MM_ANONPAGES val:5\n[ 69.816050] BUG: non-zero pgtables_bytes on freeing mm: 20480\n\nAlso, change print function (from printk(KERN_ALERT, ..) to pr_alert()) so\nthat it matches the other print statement.\n\nLink: http://lkml.kernel.org/r/da75b5153f617f4c5739c08ee6ebeb3d19db0fbc.1565123758.git.sai.praneeth.prakhya@intel.com\nSigned-off-by: Sai Praneeth Prakhya \u003csai.praneeth.prakhya@intel.com\u003e\nReviewed-by: Anshuman Khandual \u003canshuman.khandual@arm.com\u003e\nSuggested-by: Dave Hansen \u003cdave.hansen@intel.com\u003e\nAcked-by: Michal Hocko \u003cmhocko@suse.com\u003e\nAcked-by: Vlastimil Babka \u003cvbabka@suse.cz\u003e\nAcked-by: Dave Hansen \u003cdave.hansen@intel.com\u003e\nCc: Ingo Molnar \u003cmingo@kernel.org\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "90efc9b244ac28721e40d259ae7a1590e5f0c601", "tree": "95974bd6f8039e43f492f581ad926d9d7d79cd6d", "parents": [ "e1156442f01d9c1cac79baa4410a1e7b990fa0b7" ], "author": { "name": "OGAWA Hirofumi", "email": "hirofumi@mail.parknet.co.jp", "time": "Wed Sep 04 05:02:23 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:23 2019 +0000" }, "message": "fat: add nobarrier to workaround the strange behavior of device\n\nThere was a report of strange behavior of device with recent\nblkdev_issue_flush() position change.\n\nThe following is simplified usbmon trace:\n\n 4203 9.160230 host -\u003e 1.25.1 USBMS 95 SCSI: Synchronize Cache(10) LUN: 0x00 (LBA: 0x00000000, Len: 0)\n 4206 9.164911 1.25.1 -\u003e host USBMS 77 SCSI: Response LUN: 0x00 (Synchronize Cache(10)) (Good)\n 4207 9.323927 host -\u003e 1.25.1 USBMS 95 SCSI: Read(10) LUN: 0x00 (LBA: 0x00279950, Len: 240)\n 4212 9.327138 1.25.1 -\u003e host USBMS 77 SCSI: Response LUN: 0x00 (Read(10)) (Good)\n\n[...]\n\n 7323 10.202167 host -\u003e 1.25.1 USBMS 95 SCSI: Synchronize Cache(10) LUN: 0x00 (LBA: 0x00000000, Len: 0)\n 7326 10.432266 1.25.1 -\u003e host USBMS 77 SCSI: Response LUN: 0x00 (Synchronize Cache(10)) (Good)\n 7327 10.769092 host -\u003e 1.25.1 USBMS 95 SCSI: Test Unit Ready LUN: 0x00 \n 7330 10.769192 1.25.1 -\u003e host USBMS 77 SCSI: Response LUN: 0x00 (Test Unit Ready) (Good)\n 7335 12.849093 host -\u003e 1.25.1 USBMS 95 SCSI: Test Unit Ready LUN: 0x00 \n 7338 12.849206 1.25.1 -\u003e host USBMS 77 SCSI: Response LUN: 0x00 (Test Unit Ready) (Check Condition)\n 7339 12.849209 host -\u003e 1.25.1 USBMS 95 SCSI: Request Sense LUN: 0x00\n\nIf \"Synchronize Cache\" command issued then there is idle time, the device\nstop to process further commands, and behave as like no media. (it\nreturns NOT_READY [MEDIUM NOT PRESENT] for SENSE command, and this\nhappened on Kindle) [just a guess, the device is trying to detect the\n\"safe-unplug\" operation of Windows or such?]\n\nTo work around those devices and provide flexibility, this adds\n\"barrier\"/\"nobarrier\" mount options to fat driver.\n\nLink: http://lkml.kernel.org/r/87woh5pyqh.fsf@mail.parknet.co.jp\nSigned-off-by: OGAWA Hirofumi \u003chirofumi@mail.parknet.co.jp\u003e\nCc: Christoph Hellwig \u003chch@lst.de\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "e1156442f01d9c1cac79baa4410a1e7b990fa0b7", "tree": "b37a71d2e1f48343a7ceec9b74738a9681fde75d", "parents": [ "a60e5aecc73a50fe998380dff2bfd66f310b66af" ], "author": { "name": "Jason Yan", "email": "yanaijie@huawei.com", "time": "Wed Sep 04 05:02:23 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:23 2019 +0000" }, "message": "fs/reiserfs/do_balan.c: remove set but not used variable\n\nFix the following gcc warning:\n\nfs/reiserfs/do_balan.c: In function balance_leaf_insert_right:\nfs/reiserfs/do_balan.c:629:6: warning: variable ret set but not used\n[-Wunused-but-set-variable]\n\nLink: http://lkml.kernel.org/r/20190827032932.46622-2-yanaijie@huawei.com\nSigned-off-by: Jason Yan \u003cyanaijie@huawei.com\u003e\nCc: zhengbin \u003czhengbin13@huawei.com\u003e\nCc: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "a60e5aecc73a50fe998380dff2bfd66f310b66af", "tree": "6787a28502d80a659d312df41db7fad8af8ade8b", "parents": [ "085e4dd148124bed4b6bfeb44298cfa2a0d3054e" ], "author": { "name": "Jason Yan", "email": "yanaijie@huawei.com", "time": "Wed Sep 04 05:02:22 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:22 2019 +0000" }, "message": "fs/reiserfs/journal.c: remove set but not used variable\n\nFix the following gcc warning:\n\nfs/reiserfs/journal.c: In function flush_used_journal_lists:\nfs/reiserfs/journal.c:1791:6: warning: variable ret set but not used\n[-Wunused-but-set-variable]\n\nLink: http://lkml.kernel.org/r/20190827032932.46622-1-yanaijie@huawei.com\nSigned-off-by: Jason Yan \u003cyanaijie@huawei.com\u003e\nCc: zhengbin \u003czhengbin13@huawei.com\u003e\nCc: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "085e4dd148124bed4b6bfeb44298cfa2a0d3054e", "tree": "510d140922724b07ea9c9315956b7548e7ba0585", "parents": [ "30a8b8479099b056b4a543b50c94344410fc6570" ], "author": { "name": "zhengbin", "email": "zhengbin13@huawei.com", "time": "Wed Sep 04 05:02:22 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:22 2019 +0000" }, "message": "fs/reiserfs/do_balan.c: remove set but not used variables\n\nfs/reiserfs/do_balan.c: In function balance_leaf_when_delete:\nfs/reiserfs/do_balan.c:245:20: warning: variable ih set but not used [-Wunused-but-set-variable]\nfs/reiserfs/do_balan.c: In function balance_leaf_insert_left:\nfs/reiserfs/do_balan.c:301:7: warning: variable version set but not used [-Wunused-but-set-variable]\nfs/reiserfs/do_balan.c: In function balance_leaf_insert_right:\nfs/reiserfs/do_balan.c:649:7: warning: variable version set but not used [-Wunused-but-set-variable]\nfs/reiserfs/do_balan.c: In function balance_leaf_new_nodes_insert:\nfs/reiserfs/do_balan.c:953:7: warning: variable version set but not used [-Wunused-but-set-variable]\n\nLink: http://lkml.kernel.org/r/1566379929-118398-8-git-send-email-zhengbin13@huawei.com\nSigned-off-by: zhengbin \u003czhengbin13@huawei.com\u003e\nReported-by: Hulk Robot \u003chulkci@huawei.com\u003e\nCc: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "30a8b8479099b056b4a543b50c94344410fc6570", "tree": "a374ae019273e4d151b17b6f5f0db149e3e130e1", "parents": [ "44e7f336a92873dcb45802dcf8ee652d9d31afdd" ], "author": { "name": "zhengbin", "email": "zhengbin13@huawei.com", "time": "Wed Sep 04 05:02:22 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:22 2019 +0000" }, "message": "fs/reiserfs/fix_node.c: remove set but not used variables\n\nfs/reiserfs/fix_node.c: In function get_num_ver:\nfs/reiserfs/fix_node.c:379:6: warning: variable cur_free set but not used [-Wunused-but-set-variable]\nfs/reiserfs/fix_node.c: In function dc_check_balance_internal:\nfs/reiserfs/fix_node.c:1737:6: warning: variable maxsize set but not used [-Wunused-but-set-variable]\n\nLink: http://lkml.kernel.org/r/1566379929-118398-7-git-send-email-zhengbin13@huawei.com\nSigned-off-by: zhengbin \u003czhengbin13@huawei.com\u003e\nReported-by: Hulk Robot \u003chulkci@huawei.com\u003e\nCc: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "44e7f336a92873dcb45802dcf8ee652d9d31afdd", "tree": "f607aca16e2935709612173eb41d6966d4822961", "parents": [ "2283f2731dd7315898486946f349d650739db472" ], "author": { "name": "zhengbin", "email": "zhengbin13@huawei.com", "time": "Wed Sep 04 05:02:22 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:22 2019 +0000" }, "message": "fs/reiserfs/prints.c: remove set but not used variables\n\nFixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n\nfs/reiserfs/prints.c: In function check_internal_block_head:\nfs/reiserfs/prints.c:749:21: warning: variable blkh set but not used [-Wunused-but-set-variable]\n\nLink: http://lkml.kernel.org/r/1566379929-118398-6-git-send-email-zhengbin13@huawei.com\nSigned-off-by: zhengbin \u003czhengbin13@huawei.com\u003e\nReported-by: Hulk Robot \u003chulkci@huawei.com\u003e\nCc: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "2283f2731dd7315898486946f349d650739db472", "tree": "4a5b63bca06b76cfec826f04c2b81a48e843205c", "parents": [ "9f97e8f4bf6a7839de09da69722dcdf052a4676f" ], "author": { "name": "zhengbin", "email": "zhengbin13@huawei.com", "time": "Wed Sep 04 05:02:21 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:21 2019 +0000" }, "message": "fs/reiserfs/objectid.c: remove set but not used variables\n\nFixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n\nfs/reiserfs/objectid.c: In function reiserfs_convert_objectid_map_v1:\nfs/reiserfs/objectid.c:186:25: warning: variable new_objectid_map set but not used [-Wunused-but-set-variable]\n\nLink: http://lkml.kernel.org/r/1566379929-118398-5-git-send-email-zhengbin13@huawei.com\nSigned-off-by: zhengbin \u003czhengbin13@huawei.com\u003e\nReported-by: Hulk Robot \u003chulkci@huawei.com\u003e\nCc: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "9f97e8f4bf6a7839de09da69722dcdf052a4676f", "tree": "57b6a892496e92a53b0dc2c9229caccc350acdd3", "parents": [ "cdb38e763c68148c11601570bca0644ac0337466" ], "author": { "name": "zhengbin", "email": "zhengbin13@huawei.com", "time": "Wed Sep 04 05:02:21 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:21 2019 +0000" }, "message": "fs/reiserfs/lbalance.c: remove set but not used variables\n\nFixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n\nfs/reiserfs/lbalance.c: In function leaf_paste_entries:\nfs/reiserfs/lbalance.c:1325:9: warning: variable old_entry_num set but not used [-Wunused-but-set-variable]\n\nLink: http://lkml.kernel.org/r/1566379929-118398-4-git-send-email-zhengbin13@huawei.com\nSigned-off-by: zhengbin \u003czhengbin13@huawei.com\u003e\nReported-by: Hulk Robot \u003chulkci@huawei.com\u003e\nCc: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "cdb38e763c68148c11601570bca0644ac0337466", "tree": "bf8bf0fa2d8e70911ccd6e9a41d73248d6dd5aa5", "parents": [ "8d0a2fb1157e35b7f0c2c21e6fb9424182cd919d" ], "author": { "name": "zhengbin", "email": "zhengbin13@huawei.com", "time": "Wed Sep 04 05:02:21 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:21 2019 +0000" }, "message": "fs/reiserfs/stree.c: remove set but not used variables\n\nFixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n\nfs/reiserfs/stree.c: In function search_by_key:\nfs/reiserfs/stree.c:596:6: warning: variable right_neighbor_of_leaf_node set but not used [-Wunused-but-set-variable]\n\nLink: http://lkml.kernel.org/r/1566379929-118398-3-git-send-email-zhengbin13@huawei.com\nSigned-off-by: zhengbin \u003czhengbin13@huawei.com\u003e\nReported-by: Hulk Robot \u003chulkci@huawei.com\u003e\nCc: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "8d0a2fb1157e35b7f0c2c21e6fb9424182cd919d", "tree": "1bf0997c54a545c5648233d4416faa2076ba0415", "parents": [ "8a246e6b274d2bac0ef0acbf293685a562033f8c" ], "author": { "name": "zhengbin", "email": "zhengbin13@huawei.com", "time": "Wed Sep 04 05:02:21 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:21 2019 +0000" }, "message": "fs/reiserfs/journal.c: remove set but not used variables\n\nFixes gcc \u0027-Wunused-but-set-variable\u0027 warning:\n\nfs/reiserfs/journal.c: In function flush_older_commits:\nfs/reiserfs/journal.c:894:15: warning: variable first_trans_id set but not used [-Wunused-but-set-variable]\nfs/reiserfs/journal.c: In function flush_journal_list:\nfs/reiserfs/journal.c:1354:38: warning: variable last set but not used [-Wunused-but-set-variable]\nfs/reiserfs/journal.c: In function do_journal_release:\nfs/reiserfs/journal.c:1916:6: warning: variable flushed set but not used [-Wunused-but-set-variable]\nfs/reiserfs/journal.c: In function do_journal_end:\nfs/reiserfs/journal.c:3993:6: warning: variable old_start set but not used [-Wunused-but-set-variable]\n\nLink: http://lkml.kernel.org/r/1566379929-118398-2-git-send-email-zhengbin13@huawei.com\nSigned-off-by: zhengbin \u003czhengbin13@huawei.com\u003e\nReported-by: Hulk Robot \u003chulkci@huawei.com\u003e\nCc: Jan Kara \u003cjack@suse.cz\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "8a246e6b274d2bac0ef0acbf293685a562033f8c", "tree": "17124df5088bcb1e87370ba975c0a988c71a6710", "parents": [ "c1c6dc85a787fb33f8eac0249c909a65bca19b52" ], "author": { "name": "Jia-Ju Bai", "email": "baijiaju1990@gmail.com", "time": "Wed Sep 04 05:02:20 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:20 2019 +0000" }, "message": "fs: reiserfs: remove unnecessary check of bh in remove_from_transaction()\n\nOn lines 3430-3434, bh has been assured to be non-null:\n cn \u003d get_journal_hash_dev(sb, journal-\u003ej_hash_table, blocknr);\n if (!cn || !cn-\u003ebh) {\n return ret;\n }\n bh \u003d cn-\u003ebh;\n\nThus, the check of bh on line 3447 is unnecessary and can be removed.\nThank Andrew Morton for good advice.\n\nLink: http://lkml.kernel.org/r/20190727084019.11307-1-baijiaju1990@gmail.com\nSigned-off-by: Jia-Ju Bai \u003cbaijiaju1990@gmail.com\u003e\nReviewed-by: Jan Kara \u003cjack@suse.cz\u003e\nCc: Arnd Bergmann \u003carnd@arndb.de\u003e\nCc: Hariprasad Kelam \u003chariprasad.kelam@gmail.com\u003e\nCc: Bharath Vedartham \u003clinux.bhar@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "c1c6dc85a787fb33f8eac0249c909a65bca19b52", "tree": "ac88d6af3bd3c971c2efc072e1d1b0d2cd755940", "parents": [ "7ced633f800f4377010d2d7e983e2cbe734eafa6" ], "author": { "name": "Joe Perches", "email": "joe@perches.com", "time": "Wed Sep 04 05:02:20 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:20 2019 +0000" }, "message": "checkpatch: make git output use LANGUAGE\u003den_US.utf8\n\ngit output parsing depends on the language being en_US english.\n\nMake the backtick execution of all `git \u003cfoo\u003e` commands set the\nLANGUAGE of the process to en_US.utf8 before executing the actual\ncommand using `export LANGUAGE\u003den_US.utf8; git \u003cfoo\u003e`.\n\nBecause the command is executed in a child process, the parent\nLANGUAGE is unchanged.\n\nLink: http://lkml.kernel.org/r/bb9f29988f3258281956680ff39c3e19e37dc0b8.camel@perches.com\nSigned-off-by: Joe Perches \u003cjoe@perches.com\u003e\nReported-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\nReviewed-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\nTested-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "7ced633f800f4377010d2d7e983e2cbe734eafa6", "tree": "4f0ef845eeec1d9ddd2ba8acce1132300ff61581", "parents": [ "061b49e1ead5347da0a7cd5dbc20d37d14a3ccb9" ], "author": { "name": "Sean Christopherson", "email": "sean.j.christopherson@intel.com", "time": "Wed Sep 04 05:02:20 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:20 2019 +0000" }, "message": "checkpatch: remove obsolete period from \"ambiguous SHA1\" query\n\nGit dropped the period from its \"ambiguous SHA1\" error message in commit\n0c99171ad2 (\"get_short_sha1: mark ambiguity error for translation\"), circa\n2016. Drop the period from checkpatch\u0027s associated query so as to match\nboth the old and new error messages.\n\nLink: http://lkml.kernel.org/r/20190830163103.15914-1-sean.j.christopherson@intel.com\nSigned-off-by: Sean Christopherson \u003csean.j.christopherson@intel.com\u003e\nAcked-by: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "061b49e1ead5347da0a7cd5dbc20d37d14a3ccb9", "tree": "22d26346f41db56ca9e2b3cd2efb34b91890fa32", "parents": [ "ac39928d086bdf9267d7191d94a06bcbc4594289" ], "author": { "name": "Joe Perches", "email": "joe@perches.com", "time": "Wed Sep 04 05:02:20 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:20 2019 +0000" }, "message": "checkpatch: allow consecutive close braces\n\ncheckpatch allows consecutive open braces, so it should also allow\nconsecutive close braces.\n\nLink: http://lkml.kernel.org/r/bfdb49ae2c3fa7b52fa168769e38b48f959880e2.camel@perches.com\nSigned-off-by: Joe Perches \u003cjoe@perches.com\u003e\nAcked-by: Jeff Kirsher \u003cjeffrey.t.kirsher@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "ac39928d086bdf9267d7191d94a06bcbc4594289", "tree": "f50195010a549eae9ced702c4b36b6019e87c6ba", "parents": [ "1ca91357cd83dada3589b382f3ef7d056050885b" ], "author": { "name": "Joe Perches", "email": "joe@perches.com", "time": "Wed Sep 04 05:02:19 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:19 2019 +0000" }, "message": "checkpatch: prefer __section over __attribute__((section(...)))\n\nAdd another test for __attribute__((section(\"foo\"))) uses that should be\n__section(foo)\n\nLink: http://lkml.kernel.org/r/2f374c3c27054b7f978115270d587c624d9962fc.camel@perches.com\nSuggested-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nSigned-off-by: Joe Perches \u003cjoe@perches.com\u003e\nTested-by: Nick Desaulniers \u003cndesaulniers@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "1ca91357cd83dada3589b382f3ef7d056050885b", "tree": "b5fa965a15c763843ce852de777b4324412c52b2", "parents": [ "8d01dea7e8cf622d09374eb58cb5fbbd4128aece" ], "author": { "name": "Brendan Jackman", "email": "brendan.jackman@bluwireless.co.uk", "time": "Wed Sep 04 05:02:19 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:19 2019 +0000" }, "message": "checkpatch: exclude sizeof sub-expressions from MACRO_ARG_REUSE\n\nThe arguments of sizeof are not evaluated so arguments are safe to re-use\nin that context. Excluding sizeof subexpressions means macros like\nARRAY_SIZE can pass checkpatch.\n\nLink: http://lkml.kernel.org/r/20190806070833.24423-1-brendan.jackman@bluwireless.co.uk\nSigned-off-by: Brendan Jackman \u003cbrendan.jackman@bluwireless.co.uk\u003e\nAcked-by: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "8d01dea7e8cf622d09374eb58cb5fbbd4128aece", "tree": "3dd89e6e9f00c68d9fe8267b313063f737a1f341", "parents": [ "4e4f24e996fab7fe0aeb08a0a92edb750a9ac5c6" ], "author": { "name": "Matteo Croce", "email": "mcroce@redhat.com", "time": "Wed Sep 04 05:02:19 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:19 2019 +0000" }, "message": "checkpatch.pl: warn on invalid commit id\n\nIt can happen that a commit message refers to an invalid commit id,\nbecause the referenced hash changed following a rebase, or simply by\nmistake. Add a check in checkpatch.pl which checks that an hash\nreferenced by a Fixes tag, or just cited in the commit message, is a valid\ncommit id.\n\n $ scripts/checkpatch.pl \u003c\u003c\u0027EOF\u0027\n Subject: [PATCH] test commit\n\n Sample test commit to test checkpatch.pl\n Commit 1da177e4c3f4 (\"Linux-2.6.12-rc2\") really exists,\n commit 0bba044c4ce7 (\"tree\") is valid but not a commit,\n while commit b4cc0b1c0cca (\"unknown\") is invalid.\n\n Fixes: f0cacc14cade (\"unknown\")\n Fixes: 1da177e4c3f4 (\"Linux-2.6.12-rc2\")\n EOF\n WARNING: Unknown commit id \u00270bba044c4ce7\u0027, maybe rebased or not pulled?\n #8:\n commit 0bba044c4ce7 (\"tree\") is valid but not a commit,\n\n WARNING: Unknown commit id \u0027b4cc0b1c0cca\u0027, maybe rebased or not pulled?\n #9:\n while commit b4cc0b1c0cca (\"unknown\") is invalid.\n\n WARNING: Unknown commit id \u0027f0cacc14cade\u0027, maybe rebased or not pulled?\n #11:\n Fixes: f0cacc14cade (\"unknown\")\n\n total: 0 errors, 3 warnings, 4 lines checked\n\nLink: http://lkml.kernel.org/r/20190711001640.13398-1-mcroce@redhat.com\nSigned-off-by: Matteo Croce \u003cmcroce@redhat.com\u003e\nCc: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "4e4f24e996fab7fe0aeb08a0a92edb750a9ac5c6", "tree": "a1e5e3acf66756fd01ac8198c0021d102680bf85", "parents": [ "f7151dc37b171cc0a036a6caf5e63db45118d49f" ], "author": { "name": "Joe Perches", "email": "joe@perches.com", "time": "Wed Sep 04 05:02:19 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:19 2019 +0000" }, "message": "checkpatch: improve SPDX license checking\n\nUse perl\u0027s m@\u003cmatch\u003e@ match and not /\u003cmatch\u003e/ comparisons to avoid\nan error using c90\u0027s // comment style.\n\nMiscellanea:\n\no Use normal tab indentation and alignment\n\nLink: http://lkml.kernel.org/r/5e4a8fa7901148fbcd77ab391e6dd0e6bf95777f.camel@perches.com\n\nLink: http://lkml.kernel.org/r/f08eb62458407a145cfedf959d1091af151cd665.1563575364.git.joe@perches.com\nSigned-off-by: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "f7151dc37b171cc0a036a6caf5e63db45118d49f", "tree": "cb8db6331e08cb6b401a3b664bfde9bcd8a89bef", "parents": [ "5f45eed44acf51f907364efa5548d6d754ed8889" ], "author": { "name": "Joe Perches", "email": "joe@perches.com", "time": "Wed Sep 04 05:02:18 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:18 2019 +0000" }, "message": "checkpatch: don\u0027t interpret stack dumps as commit IDs\n\nAdd more types of lines that appear to be stack dumps that also include\nhex lines that might otherwise be interpreted as commit IDs.\n\nLink: http://lkml.kernel.org/r/ff00208289224f0ca4eaf4ff7c9c6e087dad0a63.camel@perches.com\n\nLink: http://lkml.kernel.org/r/f7dc9727795db3802809a24162abe0b67e14123b.1563575364.git.joe@perches.com\nSigned-off-by: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "5f45eed44acf51f907364efa5548d6d754ed8889", "tree": "ec4eaeb59b8537c810594069183dda2e7f84ddf3", "parents": [ "3728236c536812ce41db217fed4ab7b718bf2b67" ], "author": { "name": "Trent Piepho", "email": "tpiepho@gmail.com", "time": "Wed Sep 04 05:02:18 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:18 2019 +0000" }, "message": "lib/math/rational.c: fix possible incorrect result from rational fractions helper\n\nIn some cases the previous algorithm would not return the closest\napproximation. This would happen when a semi-convergent was the closest,\nas the previous algorithm would only consider convergents.\n\nAs an example, consider an initial value of 5/4, and trying to find the\nclosest approximation with a maximum of 4 for numerator and denominator. \nThe previous algorithm would return 1/1 as the closest approximation,\nwhile this version will return the correct answer of 4/3.\n\nTo do this, the main loop performs effectively the same operations as it\ndid before. It must now keep track of the last three approximations,\nn2/d2 .. n0/d0, while before it only needed the last two.\n\nIf an exact answer is not found, the algorithm will now calculate the best\nsemi-convergent term, t, which is a single expression with two divisions:\n\n min((max_numerator - n0) / n1, (max_denominator - d0) / d1)\n\nThis will be used if it is better than previous convergent. The test for\nthis is generally a simple comparison, 2*t \u003e a. But in an edge case,\nwhere the convergent\u0027s final term is even and the best allowable\nsemi-convergent has a final term of exactly half the convergent\u0027s final\nterm, the more complex comparison (d0*dp \u003e d1*d) is used.\n\nI also wrote some comments explaining the code. While one still needs to\nlook up the math elsewhere, they should help a lot to follow how the code\nrelates to that math.\n\nThis routine is used in two places in the video4linux code, but in\nthose cases it is only used to reduce a fraction to lowest terms, which\nthe existing code will do correctly. This could be done more\nefficiently with a different library routine but it would still be the\nEuclidean alogrithm at its heart. So no change.\n\nThe remain users are places where a fractional PLL divider is\nprogrammed. What would happen is something asked for a clock of X MHz\nbut instead gets Y MHz, where Y is close to X but not exactly due to\nthe hardware limitations. After this change they might, in some cases,\nget Y\u0027 MHz, where Y\u0027 is a little closer to X then Y was.\n\nUsers like this are: Three UARTs, in 8250_mid, 8250_lpss, and imx. One\nGPU in vp4_hdmi. And three clock drivers, clk-cdce706, clk-si5351, and\nclk-fractional-divider. The last is a generic clock driver and so\nwould have more users referenced via device tree entries.\n\nI think there\u0027s a bug in that one, it\u0027s limiting an N bit field that is\noffset-by-1 to the range 0 .. (1\u003c\u003cN)-2, when it should be (1\u003c\u003cN)-1 as\nthe upper limit.\n\nI have an IMX system, one of the UARTs using this, so I can provide a\nreal example. If I request a custom baud rate of 1499978, the driver\nwill program the PLL to produce a baud rate of 1500000. After this\nchange, the fractional divider in the UART is programmed to a ratio of\n65535/65536, which produces a baud rate of 1499977.0625. Closer to the\nrequested value.\n\nLink: http://lkml.kernel.org/r/20190330205855.19396-1-tpiepho@gmail.com\nSigned-off-by: Trent Piepho \u003ctpiepho@gmail.com\u003e\nCc: Oskar Schirmer \u003coskar@scara.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "3728236c536812ce41db217fed4ab7b718bf2b67", "tree": "bf4f4369052a1a990ca6f4a26d4e797fca32308c", "parents": [ "497acd9866701a6a56268e3e5a481daeafe72570" ], "author": { "name": "Stephen Boyd", "email": "swboyd@chromium.org", "time": "Wed Sep 04 05:02:18 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:18 2019 +0000" }, "message": "lib/hexdump: make print_hex_dump_bytes() a nop on !DEBUG builds\n\nI\u0027m seeing a bunch of debug prints from a user of print_hex_dump_bytes()\nin my kernel logs, but I don\u0027t have CONFIG_DYNAMIC_DEBUG enabled nor do I\nhave DEBUG defined in my build. The problem is that\nprint_hex_dump_bytes() calls a wrapper function in lib/hexdump.c that\ncalls print_hex_dump() with KERN_DEBUG level. There are three cases to\nconsider here\n\n 1. CONFIG_DYNAMIC_DEBUG\u003dy --\u003e call dynamic_hex_dum()\n 2. CONFIG_DYNAMIC_DEBUG\u003dn \u0026\u0026 DEBUG --\u003e call print_hex_dump()\n 3. CONFIG_DYNAMIC_DEBUG\u003dn \u0026\u0026 !DEBUG --\u003e stub it out\n\nRight now, that last case isn\u0027t detected and we still call\nprint_hex_dump() from the stub wrapper.\n\nLet\u0027s make print_hex_dump_bytes() only call print_hex_dump_debug() so that\nit works properly in all cases.\n\nCase #1, print_hex_dump_debug() calls dynamic_hex_dump() and we get same\nbehavior. Case #2, print_hex_dump_debug() calls print_hex_dump() with\nKERN_DEBUG and we get the same behavior. Case #3, print_hex_dump_debug()\nis a nop, changing behavior to what we want, i.e. print nothing.\n\nLink: http://lkml.kernel.org/r/20190816235624.115280-1-swboyd@chromium.org\nSigned-off-by: Stephen Boyd \u003cswboyd@chromium.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "497acd9866701a6a56268e3e5a481daeafe72570", "tree": "8cd56917e9d3eaeee7d6bb1cc2399ecf1d83f273", "parents": [ "84ada9611011c005c5fbe03c8cb5527429776a9c" ], "author": { "name": "Valdis Kletnieks", "email": "valdis.kletnieks@vt.edu", "time": "Wed Sep 04 05:02:18 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:18 2019 +0000" }, "message": "lib/extable.c: add missing prototypes\n\nWhen building with W\u003d1, a number of warnings are issued:\n\n CC lib/extable.o\nlib/extable.c:63:6: warning: no previous prototype for \u0027sort_extable\u0027 [-Wmissing-prototypes]\n 63 | void sort_extable(struct exception_table_entry *start,\n | ^~~~~~~~~~~~\nlib/extable.c:75:6: warning: no previous prototype for \u0027trim_init_extable\u0027 [-Wmissing-prototypes]\n 75 | void trim_init_extable(struct module *m)\n | ^~~~~~~~~~~~~~~~~\nlib/extable.c:115:1: warning: no previous prototype for \u0027search_extable\u0027 [-Wmissing-prototypes]\n 115 | search_extable(const struct exception_table_entry *base,\n | ^~~~~~~~~~~~~~\n\nAdd the missing #include for the prototypes.\n\nLink: http://lkml.kernel.org/r/45574.1565235784@turing-police\nSigned-off-by: Valdis Kletnieks \u003cvaldis.kletnieks@vt.edu\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "84ada9611011c005c5fbe03c8cb5527429776a9c", "tree": "766c856cf26a4bce93b650f8a30c06c994a9445b", "parents": [ "8764e35b4eeb0e7d44a260788d75e3a4ddc92d2c" ], "author": { "name": "Valdis Kletnieks", "email": "valdis.kletnieks@vt.edu", "time": "Wed Sep 04 05:02:17 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:17 2019 +0000" }, "message": "lib/generic-radix-tree.c: make 2 functions static inline\n\nWhen building with W\u003d1, we get some warnings:\n\nl CC lib/generic-radix-tree.o\nlib/generic-radix-tree.c:39:10: warning: no previous prototype for \u0027genradix_root_to_depth\u0027 [-Wmissing-prototypes]\n 39 | unsigned genradix_root_to_depth(struct genradix_root *r)\n | ^~~~~~~~~~~~~~~~~~~~~~\nlib/generic-radix-tree.c:44:23: warning: no previous prototype for \u0027genradix_root_to_node\u0027 [-Wmissing-prototypes]\n 44 | struct genradix_node *genradix_root_to_node(struct genradix_root *r)\n | ^~~~~~~~~~~~~~~~~~~~~\n\nThey\u0027re not used anywhere else, so make them static inline.\n\nLink: http://lkml.kernel.org/r/46923.1565236485@turing-police\nSigned-off-by: Valdis Kletnieks \u003cvaldis.kletnieks@vt.edu\u003e\nCc: Kent Overstreet \u003ckent.overstreet@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "8764e35b4eeb0e7d44a260788d75e3a4ddc92d2c", "tree": "bec14fa509190cdf9b6db49aa9de7d5c0c8812f9", "parents": [ "69cb2c8d5d302a80232b5acb01a755f7139018f8" ], "author": { "name": "Kees Cook", "email": "keescook@chromium.org", "time": "Wed Sep 04 05:02:17 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:17 2019 +0000" }, "message": "strscpy: reject buffer sizes larger than INT_MAX\n\nAs already done for snprintf(), add a check in strscpy() for giant (i.e. \nlikely negative and/or miscalculated) copy sizes, WARN, and error out.\n\nLink: http://lkml.kernel.org/r/201907260928.23DE35406@keescook\nSigned-off-by: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: Joe Perches \u003cjoe@perches.com\u003e\nCc: Rasmus Villemoes \u003clinux@rasmusvillemoes.dk\u003e\nCc: Yann Droneaud \u003cydroneaud@opteya.com\u003e\nCc: David Laight \u003cDavid.Laight@aculab.com\u003e\nCc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\nCc: Stephen Kitt \u003csteve@sk2.org\u003e\nCc: Jann Horn \u003cjannh@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "69cb2c8d5d302a80232b5acb01a755f7139018f8", "tree": "b4daa36665f30a7d46e4a8dd037960fb31b90092", "parents": [ "00098525ccff4937d9169716011ae9142bf7e869" ], "author": { "name": "Qian Cai", "email": "cai@lca.pw", "time": "Wed Sep 04 05:02:17 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:17 2019 +0000" }, "message": "include/trace/events/writeback.h: fix -Wstringop-truncation warnings\n\nThere are many of those warnings.\n\nIn file included from ./arch/powerpc/include/asm/paca.h:15,\n from ./arch/powerpc/include/asm/current.h:13,\n from ./include/linux/thread_info.h:21,\n from ./include/asm-generic/preempt.h:5,\n from ./arch/powerpc/include/generated/asm/preempt.h:1,\n from ./include/linux/preempt.h:78,\n from ./include/linux/spinlock.h:51,\n from fs/fs-writeback.c:19:\nIn function \u0027strncpy\u0027,\n inlined from \u0027perf_trace_writeback_page_template\u0027 at\n./include/trace/events/writeback.h:56:1:\n./include/linux/string.h:260:9: warning: \u0027__builtin_strncpy\u0027 specified\nbound 32 equals destination size [-Wstringop-truncation]\n return __builtin_strncpy(p, q, size);\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n\nFix it by using the new strscpy_pad() which was introduced in \"lib/string:\nAdd strscpy_pad() function\" and will always be NUL-terminated instead of\nstrncpy(). Also, change strlcpy() to use strscpy_pad() in this file for\nconsistency.\n\nLink: http://lkml.kernel.org/r/1564075099-27750-1-git-send-email-cai@lca.pw\nFixes: 455b2864686d (\"writeback: Initial tracing support\")\nFixes: 028c2dd184c0 (\"writeback: Add tracing to balance_dirty_pages\")\nFixes: e84d0a4f8e39 (\"writeback: trace event writeback_queue_io\")\nFixes: b48c104d2211 (\"writeback: trace event bdi_dirty_ratelimit\")\nFixes: cc1676d917f3 (\"writeback: Move requeueing when I_SYNC set to writeback_sb_inodes()\")\nFixes: 9fb0a7da0c52 (\"writeback: add more tracepoints\")\nSigned-off-by: Qian Cai \u003ccai@lca.pw\u003e\nReviewed-by: Jan Kara \u003cjack@suse.cz\u003e\nCc: Tobin C. Harding \u003ctobin@kernel.org\u003e\nCc: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\nCc: Ingo Molnar \u003cmingo@redhat.com\u003e\nCc: Tejun Heo \u003ctj@kernel.org\u003e\nCc: Dave Chinner \u003cdchinner@redhat.com\u003e\nCc: Fengguang Wu \u003cfengguang.wu@intel.com\u003e\nCc: Jens Axboe \u003caxboe@kernel.dk\u003e\nCc: Joe Perches \u003cjoe@perches.com\u003e\nCc: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: Jann Horn \u003cjannh@google.com\u003e\nCc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\nCc: Nitin Gote \u003cnitin.r.gote@intel.com\u003e\nCc: Rasmus Villemoes \u003crasmus.villemoes@prevas.dk\u003e\nCc: Stephen Kitt \u003csteve@sk2.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "00098525ccff4937d9169716011ae9142bf7e869", "tree": "f319ae94a3f493188f40a7ada7ee929a2cb44d08", "parents": [ "68c52457ca2a78639908fa92ff9b2a154d307868" ], "author": { "name": "Joe Perches", "email": "joe@perches.com", "time": "Wed Sep 04 05:02:17 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:17 2019 +0000" }, "message": "kernel-doc: core-api: Include string.h into core-api\n\ncorrect return of -E2BIG descriptions\n\nLink: http://lkml.kernel.org/r/29f998b4c1a9d69fbeae70500ba0daa4b340c546.1563889130.git.joe@perches.com\nSigned-off-by: Joe Perches \u003cjoe@perches.com\u003e\nCc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\nCc: Stephen Kitt \u003csteve@sk2.org\u003e\nCc: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: Nitin Gote \u003cnitin.r.gote@intel.com\u003e\nCc: Rasmus Villemoes \u003crasmus.villemoes@prevas.dk\u003e\nCc: Jann Horn \u003cjannh@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "68c52457ca2a78639908fa92ff9b2a154d307868", "tree": "392aef4ebc507d2edbcd816bd80196cb98d9e141", "parents": [ "4efe79e1ac4933f5a38a7d3f76bf97b56ca25d21" ], "author": { "name": "Joe Perches", "email": "joe@perches.com", "time": "Wed Sep 04 05:02:16 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:16 2019 +0000" }, "message": "kernel-doc: core-api: include string.h into core-api\n\ncore-api should show all the various string functions including the newly\nadded stracpy and stracpy_pad.\n\nMiscellanea:\n\no Update the Returns: value for strscpy\no fix a defect with %NUL)\n\nLink: http://lkml.kernel.org/r/224a6ebf39955f4107c0c376d66155d970e46733.1563841972.git.joe@perches.com\nSigned-off-by: Joe Perches \u003cjoe@perches.com\u003e\nReviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\nCc: Stephen Kitt \u003csteve@sk2.org\u003e\nCc: Nitin Gote \u003cnitin.r.gote@intel.com\u003e\nCc: Rasmus Villemoes \u003crasmus.villemoes@prevas.dk\u003e\nCc: Jann Horn \u003cjannh@google.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "4efe79e1ac4933f5a38a7d3f76bf97b56ca25d21", "tree": "e326acdb180d5aa227b23d6de58d1b9fb8df5e6d", "parents": [ "813d6bac90aacdc274b29a5ddfe90daef7288260" ], "author": { "name": "Nitin Gote", "email": "nitin.r.gote@intel.com", "time": "Wed Sep 04 05:02:16 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:16 2019 +0000" }, "message": "Documentation/checkpatch: prefer stracpy/strscpy over strcpy/strlcpy/strncpy.\n\nAdd check in checkpatch.pl to deprecate strcpy(), strlcpy() and strncpy()\nin favor of stracpy() or strscpy().\n\nUpdate Documentation/process/deprecated.rst for stracpy() and strscpy().\n\nLink: http://lkml.kernel.org/r/20190728160608.3119-1-nitin.r.gote@intel.com\nSigned-off-by: Nitin Gote \u003cnitin.r.gote@intel.com\u003e\nCc: Joe Perches \u003cjoe@perches.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "813d6bac90aacdc274b29a5ddfe90daef7288260", "tree": "ad9580766476909d97586b4e2d98a4c42f75823e", "parents": [ "c34026e4920b87aeaee60b7680028425948ee1d3" ], "author": { "name": "Joe Perches", "email": "joe@perches.com", "time": "Wed Sep 04 05:02:16 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:16 2019 +0000" }, "message": "string: add stracpy and stracpy_pad mechanisms\n\nSeveral uses of strlcpy and strscpy have had defects because the last\nargument of each function is misused or typoed.\n\nAdd macro mechanisms to avoid this defect.\n\nstracpy (copy a string to a string array) must have a string array as the\nfirst argument (dest) and uses sizeof(dest) as the count of bytes to copy.\n\nThese mechanisms verify that the dest argument is an array of char or\nother compatible types like u8 or s8 or equivalent.\n\nA BUILD_BUG is emitted when the type of dest is not compatible.\n\nLink: http://lkml.kernel.org/r/ed4611a4a96057bf8076856560bfbf9b5e95d390.1563889130.git.joe@perches.com\nSigned-off-by: Joe Perches \u003cjoe@perches.com\u003e\nReviewed-by: Kees Cook \u003ckeescook@chromium.org\u003e\nCc: Jann Horn \u003cjannh@google.com\u003e\nCc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\nCc: Nitin Gote \u003cnitin.r.gote@intel.com\u003e\nCc: Rasmus Villemoes \u003crasmus.villemoes@prevas.dk\u003e\nCc: Stephen Kitt \u003csteve@sk2.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "c34026e4920b87aeaee60b7680028425948ee1d3", "tree": "f98edf9bb4328c5a05d1fd75e933ef63ef9bdf72", "parents": [ "902aa1cda22097ccb96b20e929a18fa7b61817e9" ], "author": { "name": "Huang Shijie", "email": "sjhuang@iluvatar.ai", "time": "Wed Sep 04 05:02:15 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:15 2019 +0000" }, "message": "Documentation: rename addr_in_gen_pool to gen_pool_has_addr\n\nUpdate the document, since we have renamed addr_in_gen_pool to\ngen_pool_has_addr.\n\nLink: http://lkml.kernel.org/r/20181229015914.5573-1-sjhuang@iluvatar.ai\nSigned-off-by: Huang Shijie \u003csjhuang@iluvatar.ai\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "902aa1cda22097ccb96b20e929a18fa7b61817e9", "tree": "5dba225499b797b26495a18e617dfde2355cfa93", "parents": [ "a91b60922305d0838987f05f840c152bc2b4e8a4" ], "author": { "name": "Huang Shijie", "email": "sjhuang@iluvatar.ai", "time": "Wed Sep 04 05:02:15 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:15 2019 +0000" }, "message": "lib/genalloc.c: rename addr_in_gen_pool to gen_pool_has_addr\n\nFollow the kernel conventions, rename addr_in_gen_pool to\ngen_pool_has_addr.\n\nLink: http://lkml.kernel.org/r/20181228083950.20398-1-sjhuang@iluvatar.ai\nSigned-off-by: Huang Shijie \u003csjhuang@iluvatar.ai\u003e\nReviewed-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Russell King \u003clinux@armlinux.org.uk\u003e\nCc: Arnd Bergmann \u003carnd@arndb.de\u003e\nCc: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nCc: Christoph Hellwig \u003chch@lst.de\u003e\nCc: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\nCc: Robin Murphy \u003crobin.murphy@arm.com\u003e\nCc: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "a91b60922305d0838987f05f840c152bc2b4e8a4", "tree": "9dd5a83a1c824fba1e5de66c3d3f42246a66da02", "parents": [ "94ea5c41983e7f925f309149f9a64d7e23230121" ], "author": { "name": "Huang Shijie", "email": "sjhuang@iluvatar.ai", "time": "Wed Sep 04 05:02:15 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:15 2019 +0000" }, "message": "lib/genalloc.c: export symbol addr_in_gen_pool\n\nWe use addr_in_gen_pool() in a driver module. So export it.\n\nLink: http://lkml.kernel.org/r/20181224070622.22197-2-sjhuang@iluvatar.ai\nSigned-off-by: Huang Shijie \u003csjhuang@iluvatar.ai\u003e\nReviewed-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Alexey Skidanov \u003calexey.skidanov@intel.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "94ea5c41983e7f925f309149f9a64d7e23230121", "tree": "210779a9d68aafea71f79a65189bcce8498bb88d", "parents": [ "1466d84031f8cf22db90c0b0b9cf3ef1a5f1f299" ], "author": { "name": "Michel Lespinasse", "email": "walken@google.com", "time": "Wed Sep 04 05:02:15 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:15 2019 +0000" }, "message": "augmented rbtree: rework the RB_DECLARE_CALLBACKS macro definition\n\nChange the definition of the RBCOMPUTE function. The propagate callback\nrepeatedly calls RBCOMPUTE as it moves from leaf to root. it wants to\nstop recomputing once the augmented subtree information doesn\u0027t change. \nThis was previously checked using the \u003d\u003d operator, but that only works\nwhen the augmented subtree information is a scalar field. This commit\nmodifies the RBCOMPUTE function so that it now sets the augmented subtree\ninformation instead of returning it, and returns a boolean value\nindicating if the propagate callback should stop.\n\nThe motivation for this change is that I want to introduce augmented\nrbtree uses where the augmented data for the subtree is a struct instead\nof a scalar.\n\nLink: http://lkml.kernel.org/r/20190703040156.56953-4-walken@google.com\nSigned-off-by: Michel Lespinasse \u003cwalken@google.com\u003e\nAcked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\nCc: David Howells \u003cdhowells@redhat.com\u003e\nCc: Davidlohr Bueso \u003cdbueso@suse.de\u003e\nCc: Uladzislau Rezki \u003curezki@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "1466d84031f8cf22db90c0b0b9cf3ef1a5f1f299", "tree": "fae6840ce2893c73a9b5111d3b268fd878fa5556", "parents": [ "54e8a71969e794ebdd995a1173388b84ba6865f8" ], "author": { "name": "Michel Lespinasse", "email": "walken@google.com", "time": "Wed Sep 04 05:02:15 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:15 2019 +0000" }, "message": "augmented-rbtree-add-new-rb_declare_callbacks_max-macro-fix-3\n\nAfter introducing RB_DECLARE_CALLBACKS_MAX, we do not need the\naugment_recompute function to recompute node-\u003eaugmented during\nrbtree rebalancing callbacks. However, this function was also\nused in check_augmented() to verify that node-\u003eaugmented was\ncorrectly set, so we need to reintroduce the code for that check.\n\nLink: http://lkml.kernel.org/r/20190727022027.GA86863@google.com\nSigned-off-by: Michel Lespinasse \u003cwalken@google.com\u003e\nCc: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\nCc: David Howells \u003cdhowells@redhat.com\u003e\nCc: Davidlohr Bueso \u003cdbueso@suse.de\u003e\nCc: Uladzislau Rezki \u003curezki@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "54e8a71969e794ebdd995a1173388b84ba6865f8", "tree": "7f8ff711cabc35a5acd314aff9877a313f1b6474", "parents": [ "6a0a12fd0777972ffec46e1111c233b69ee95340" ], "author": { "name": "Michel Lespinasse", "email": "walken@google.com", "time": "Wed Sep 04 05:02:14 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:14 2019 +0000" }, "message": "augmented-rbtree-add-new-rb_declare_callbacks_max-macro-fix\n\nSyncing up with v5.2, I see that there is a new use for augmented\nrbtrees in mm/vmalloc.c which does not compile after applying my\npatchset.\n\nRB_DECLARE_CALLBACKS is replaced by RB_DECLARE_CALLBACKS_MAX and the\ncompute_subtree_max_size argument is replaced by va_size.\n\nLink: http://lkml.kernel.org/r/CANN689FXgK13wDYNh1zKxdipeTuALG4eKvKpsdZqKFJ-rvtGiQ@mail.gmail.com\nCc: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\nCc: David Howells \u003cdhowells@redhat.com\u003e\nCc: Davidlohr Bueso \u003cdbueso@suse.de\u003e\nCc: Uladzislau Rezki \u003curezki@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "6a0a12fd0777972ffec46e1111c233b69ee95340", "tree": "bc04a0ba3b27976f706f3210e9faa2fcc0f34856", "parents": [ "638f0caf61a5545a511b34885133849fc1c90531" ], "author": { "name": "Michel Lespinasse", "email": "walken@google.com", "time": "Wed Sep 04 05:02:14 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:14 2019 +0000" }, "message": "augmented rbtree: add new RB_DECLARE_CALLBACKS_MAX macro\n\nAdd RB_DECLARE_CALLBACKS_MAX, which generates augmented rbtree callbacks\nfor the case where the augmented value is a scalar whose definition\nfollows a max(f(node)) pattern. This actually covers all present uses of\nRB_DECLARE_CALLBACKS, and saves some (source) code duplication in the\nvarious RBCOMPUTE function definitions.\n\nLink: http://lkml.kernel.org/r/20190703040156.56953-3-walken@google.com\nSigned-off-by: Michel Lespinasse \u003cwalken@google.com\u003e\nAcked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\nCc: David Howells \u003cdhowells@redhat.com\u003e\nCc: Davidlohr Bueso \u003cdbueso@suse.de\u003e\nCc: Uladzislau Rezki \u003curezki@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "638f0caf61a5545a511b34885133849fc1c90531", "tree": "6e55ecc59e7f8e940c77fc5013b9f9f7c601d6a8", "parents": [ "1921a39fd18ef8cd17065a634ffede2bb9b6a66a" ], "author": { "name": "Michel Lespinasse", "email": "walken@google.com", "time": "Wed Sep 04 05:02:14 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:14 2019 +0000" }, "message": "augmented rbtree: add comments for RB_DECLARE_CALLBACKS macro\n\nPatch series \"make RB_DECLARE_CALLBACKS more generic\", v3.\n\nThese changes are intended to make the RB_DECLARE_CALLBACKS macro more\ngeneric (allowing the aubmented subtree information to be a struct instead\nof a scalar).\n\nI have verified the compiled lib/interval_tree.o and mm/mmap.o files to\ncheck that they didn\u0027t change. This held as expected for interval_tree.o;\nmmap.o did have some changes which could be reverted by marking\n__vma_link_rb as noinline. I did not add such a change to the patchset; I\nfelt it was reasonable enough to leave the inlining decision up to the\ncompiler.\n\n\nThis patch (of 3):\n\nAdd a short comment summarizing the arguments to RB_DECLARE_CALLBACKS. \nThe arguments are also now capitalized. This copies the style of the\nINTERVAL_TREE_DEFINE macro.\n\nNo functional changes in this commit, only comments and capitalization.\n\nLink: http://lkml.kernel.org/r/20190703040156.56953-2-walken@google.com\nSigned-off-by: Michel Lespinasse \u003cwalken@google.com\u003e\nAcked-by: Davidlohr Bueso \u003cdbueso@suse.de\u003e\nAcked-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\nCc: David Howells \u003cdhowells@redhat.com\u003e\nCc: Uladzislau Rezki \u003curezki@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "1921a39fd18ef8cd17065a634ffede2bb9b6a66a", "tree": "a5945d90ec15ab4332ab733a7299fb0b30bc03c1", "parents": [ "ac9dd4759a96a38bf7636aab8806da0ed0aba133" ], "author": { "name": "Michel Lespinasse", "email": "walken@google.com", "time": "Wed Sep 04 05:02:14 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:14 2019 +0000" }, "message": "rbtree: avoid generating code twice for the cached versions (tools copy)\n\nAs was already noted in rbtree.h, the logic to cache rb_first (or\nrb_last) can easily be implemented externally to the core rbtree api.\n\nThis commit takes the changes applied to the include/linux/ and lib/\nrbtree files in 9f973cb38088 (\"lib/rbtree: avoid generating code twice\nfor the cached versions\"), and applies these to the\ntools/include/linux/ and tools/lib/ files as well to keep them\nsynchronized.\n\nLink: http://lkml.kernel.org/r/20190703034812.53002-1-walken@google.com\nSigned-off-by: Michel Lespinasse \u003cwalken@google.com\u003e\nCc: David Howells \u003cdhowells@redhat.com\u003e\nCc: Davidlohr Bueso \u003cdbueso@suse.de\u003e\nCc: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" }, { "commit": "ac9dd4759a96a38bf7636aab8806da0ed0aba133", "tree": "7a484a6899907d35646e5dbee64a302d9302d843", "parents": [ "801c7184867e5397237598ce82b0f9247d713b3e" ], "author": { "name": "Dmitry Safonov", "email": "dima@arista.com", "time": "Wed Sep 04 05:02:13 2019 +0000" }, "committer": { "name": "Johannes Weiner", "email": "hannes@cmpxchg.org", "time": "Wed Sep 04 05:02:13 2019 +0000" }, "message": "hung_task: allow printing warnings every check interval\n\nHung task detector has one timeout and has two associated actions on it:\n\n- issuing warnings with names and stacks of blocked tasks\n- panic()\n\nWe want switches to panic (and reboot) if there\u0027s a task in\nuninterruptible sleep for some minutes - at that moment something ugly has\nhappened and the box needs a reboot. But we also want to detect\nconditions that are \"out of range\" or approaching the point of failure. \nUnder such conditions we want to issue an \"early warning\" of an impending\nfailure, minutes before the switch is going to panic.\n\nThose \"early warnings\" serve a purpose while monitoring the network\ninfrastructure. Those are also valuable on post-mortem analysis, when the\nlogs from userspace applications aren\u0027t enough. Furthermore, we have a\ntest pool of long-running duts that are constantly under close to\nreal-world load for weeks. And such early warnings allowed to figure out\nsome bottle necks without much engineer work intervention.\n\nThere are also not yet upstream patches for other kinds of \"early\nwarnings\" as prints whenever a mutex/semaphore is released after being\nheld for long time, but those patches are much more intricate and have\ntheir runtime cost.\n\nIt seems rather easy to add printing tasks and their stacks for\nnotification and debugging purposes into hung task detector without\ncomplicating the code or major cost (prints are with KERN_INFO loglevel\nand so don\u0027t go on console, only into dmesg log).\n\nSince a2e514453861 (\"kernel/hung_task.c: allow to set checking interval\nseparately from timeout\") it\u0027s possible to set checking interval for hung\ntask detector with `hung_task_check_interval_secs`.\n\nProvide `hung_task_interval_warnings` sysctl that allows printing hung\ntasks every detection interval. It\u0027s not ratelimited, so the root should\nbe cautious configuring it.\n\nLink: http://lkml.kernel.org/r/20190724170249.9644-1-dima@arista.com\nSigned-off-by: Dmitry Safonov \u003cdima@arista.com\u003e\nCc: Dmitry Vyukov \u003cdvyukov@google.com\u003e\nCc: Ingo Molnar \u003cmingo@kernel.org\u003e\nCc: Jonathan Corbet \u003ccorbet@lwn.net\u003e\nCc: Tetsuo Handa \u003cpenguin-kernel@I-love.SAKURA.ne.jp\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: \"Peter Zijlstra (Intel)\" \u003cpeterz@infradead.org\u003e\nCc: Vasiliy Khoruzhick \u003cvasilykh@arista.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\n" } ], "next": "801c7184867e5397237598ce82b0f9247d713b3e" }